report on financial stability · report on financial stability 3 contents overview 5 1...

REPORT ON FINANCIAL STABILITY

DECEMBER 2003

Prepared by the Financial Stability Departmentand Economics Department of

the Magyar Nemzeti Bank

Dr. Tamás Kálmán, Managing Directorand István Hamecz, Managing Director

Published by the Magyar Nemzeti BankPublisher in charge: Krisztina Antalffy,Head of Communication Department

8–9 Szabadság tér, Budapest 1850

www.mnb.hu

ISSN 1419-2926

http://www.mnb.hu

3REPORT ON FINANCIAL STABILITY

CONTENTS

OVERVIEW 5

1 MACROECONOMIC INDICATORS 11

1.1 THE GLOBAL BUSINESS CYCLE AND RISK PERCEPTION 131.2 DOMESTIC FINANCIAL MARKETS 151.3 GROWTH AND INFLATION 211.4 EXTERNAL EQUILIBRIUM 26

2 STABILITY OF THE BANKING SYSTEM 35

2.1 RISKS ASSOCIATED WITH NON-FINANCIAL CORPORATIONS 392.2 DOMESTIC CORPORATE CREDIT RISK 422.3 HOUSEHOLD SECTOR 462.4 PORTFOLIO QUALITY 512.5 DERIVATIVES ACTIVITIES AND MARKET RISKS OF BANKS 542.6 BANKING SECTOR LIQUIDITY 592.7 FINANCIAL POSITION AND CAPITAL ADEQUACY 612.8 PROFITABILITY 65

3 CURRENT TOPICS RELATED TO STABILITY 69

3.1 STRESS TEST RESULTS 713.2 CORPORATE SECTOR PROFITABILITY AND STABILITY 74

4 ARTICLES 85

4.1 CSABA CSÁVÁS AND GERGELY KÓCZÁN: DEVELOPMENT OF THE HUNGARIAN DERIVATIVESMARKET AND ITS EFFECT ON FINANCIAL STABILITY 87

4.2 GYÖRGY SZALAY–GYULA TÓTH: THE FINANCE OF HOME PURCHASE AND CONSTRUCTION,THE RISKS INVOLVED AND THEIR MANAGEMENT IN THE HUNGARIAN BANKING SYSTEM 100

4.3 DR. JUDIT GELEGONYA: THE ROLE OF FOREIGN-OWNED BANKS IN HUNGARY 116

INDEX OF CHARTS AND TABLES 127

CONTENTS


OVERVIEW

Of the major regions of the global economy, the outlook for growth in Japan andthe US has been improving since the summer of 2003. Euro area economic growth,however, has been modest and may not reach its potential rate before the secondhalf of 2004.

Despite the improving global outlook, the risks to domestic macroeconomic stabili-ty have mounted in the past six months: the current account deficit has soared,exchange rate and yield volatility have increased considerably and the risk premiumon forint-denominated investments has risen.

Uncertainty about the future course of convergence is the main factor responsiblefor the higher volatility affecting Hungarian financial markets and the deterioratingrisk perception. Market perception of economic fundamentals has worsened, withthe country’s external equilibrium position being the primary cause of concern formarket participants. As a result of market expectations of rising inflation and a slow-er-than-planned reduction in the fiscal deficit, the market anticipates increasing diffi-culties in meeting the convergence criteria, required for the adoption of the euro in2008. The devaluation of the central parity of the forint on 4 June considerablyaltered the market’s expectations about the future path of the exchange rate and thecentral parity to be adopted within ERM II, which in turn has contributed to highervolatility. As a result of the general uncertainties surrounding the outlook for thedomestic economy, the vulnerability and exposure of the forint to financial conta-gion have increased. Compared to earlier periods, investors have reacted more sen-sitively to regional developments, in particular to recent events in Poland.

While uncertainties have increased in Hungary and the Central and EasternEuropean region, perception of the risks facing emerging markets was extremelypositive in 2003 in comparison with earlier years. All this is explained by the histor-ically low dollar and euro yields. Nevertheless, the turnaround in the interest ratecycle in developed countries, anticipated to occur in the second half of 2004, mayalter the currently stimulating environment for investing in emerging countries.

While persistent uncertainty surrounding the forint exchange rate may slow the fur-ther pick-up in Hungarian firms’ investment activity which is already underway, thiseffect is seen as modest as the majority of firms producing for export markets enjoya natural hedging position due to the high import content of their output. The morethan 200 basis point increase in the one-year ex ante real interest rate over the pastsix months does not carry risks to stability, given the return of the real interest ratefrom the very low levels experienced in the aftermath of the speculative attack tothe 3–4% range characterising past years.

Depreciation of the exchange rate relative to 2002 has contributed to disinflationcoming to a halt. Early in the summer inflation expectations began to increase. In2004, the increase in indirect taxes will likely add nearly 2% to the general pricelevel. In order to prevent the inflation shock caused by the one-off rise in the gener-al price level from leading to a lasting increase in inflation expectations, it is impor-tant that economic agents remain confident regarding the continuation of disinfla-

OVERVIEW

Worsening domestic riskperception

Causes of high exchangerate and yield volatility:

(i) worsening marketassessment offundamentals

(ii) expectations of slowerconvergence

(iii) changed exchangerate expectations

Turnaround in the interestrate cycle may underminethe attraction of emergingmarkets

For the moment, highervolatility does not carryrisk to growth; however, it may affect inflationnegatively

6 MAGYAR NEMZETI BANK

OVERVIEW

tion. Owing to this, the uncertainty surrounding future exchange rate developmentscarries risks to the longer-term path of inflation.

In the Bank’s forecast, the economy passes its cyclical trough in 2003, with growththen picking up slightly in 2004. The recovery of the economies of Europe from stag-nation is a prerequisite for a continued upswing, which in turn may ensurefavourable conditions for export-oriented growth. Consumption is expected to slowin 2004, contributing to a shift towards more balanced growth. In the Bank’s view,the corporate sector only has room for modest wage increases, due to the widen-ing gap between wage growth and performance in recent years and real exchangerate appreciation, while the increase in indirect taxes in 2004 will reduce the pur-chasing power of incomes.

There has been significant capacity enlargement in the services sector producing forthe domestic market, in addition to the rise in household demand in the past twoyears. The investment rate has been high, employment has increased continuouslyand the rate of wage growth has been higher than in other quarters of the privatesector. In these areas, a slowdown in consumption and a potential decline in hous-ing investment may lead to excess capacity and profitability problems.

As a result of the demand-driven growth characteristic of the Hungarian economyin the past two years, the current account deficit will likely rise significantly in 2003.The current high external borrowing requirement is due primarily to the change inhouseholds’ saving behaviour. In the past decade, as a rule households have con-tributed to the financing of other sectors; in 2003, however, the subsidised loan facil-ities and the rapid increase in income over recent years resulted in household finan-cial savings declining to almost zero as a proportion of GDP. Partly offsetting house-holds’ deteriorating position, the low propensity of the corporate sector to investhas considerably reduced foreign borrowing. However, the corporate sector’s bor-rowing requirement is expected to grow simultaneously with the recovery in eco-nomic activity, while households’ financial position is only expected to improveslowly. A reduction in the general government borrowing requirement may be theonly way to improve the external equilibrium position. A more pronounced shifttowards lower risks can only be achieved if household savings increase more strong-ly than forecast, or, if the increase in outstanding liabilities of the household sectortapers off.

The fiscal budget approved for 2004, EU regulations which also apply to Hungaryfrom next year, and the Government’s commitment to the adoption of the euro in2008 are all indicative of lower borrowing by general government abroad. In theBank’s forecast, the current account deficit declines by a total of 1%–1.2% of GDPin 2005–2006. However, taking into account the effect of transfers from theEuropean Union, the external borrowing requirement is likely to fall more stronglythan the decline in the current account deficit.

Despite a favourable projection in terms of stability, the risk of a higher-than-expect-ed external deficit is considerable. One of the underlying reasons for uncertainty isthat households’ reaction to a rapid fall in real income and changes in the subsidisedhousing loan system is unpredictable. If households regard next year’s changes astemporary and smooth their consumption to a considerable extent, then the currentaccount deficit may be higher than projected, owing to the worsening net financingcapacity of the sector. A rapid recovery in the European economy may lead to a fur-ther increase in the current account deficit as a proportion of GDP, as a vigorousrise in external demand would encourage companies to implement postponedinvestment and rapidly replenish their reduced stocks. This, in turn, would increasethe borrowing requirement of both the sector and the national economy. Finally,another risk factor is that the general government deficit may develop lessfavourably than expected.

Precondition for an upturnis a recovery in Europeaneconomic activity

The services sector mayface high credit risks dueto the change in thecomposition of growth

Adjustment in the generalgovernment sectorreduces the high currentaccount deficit

Adjustment by householdsto the stagnation in realincome may exacerbaterisks to the currentaccount


OVERVIEW

Developments in the general government borrowing requirement are critical notonly in terms of the amount of the current account deficit, but also in terms of itsfinancing. Over recent years, government securities purchases by foreign investorshave played a major role in financing the current account deficit. This has been thenatural consequence of the fact that it is mainly the general government that hasrequired financing, and this does not generate either direct investment import, typi-cal of the corporate sector, or funding in the form of corporate borrowing. If thereduction in the general government deficit falls below what has been declared andpriced in by the market, confidence in the convergence process could be shaken,which may entail a further rise in risk premia.

Recently, the average duration of government securities held by non-residents hasshortened. This, however, has been caused by increased demand for medium-term(2- to 4-year) rather than short-term securities.

Another risk related to non-residents’ government securities investments may arise ifinvestors decide to change the risk profile of their current investments. In the past, alarge majority of government securities investors opted not to hedge their exchangerate risk. Higher exchange rate volatility has, however, led to higher demand for hedg-ing exchange rate risks, typically through a combination of spot and swap transac-tions. The impact of such hedging transactions on supply and demand in the foreigncurrency market is identical to that of government securities sales.

Effective from 2004, the methodology of compiling the balance of payments willchange. In accordance with the international practice, foreign-owned enterprises’retained earnings will be recorded according to a different method. Retained earn-ings will be accounted for as an outflow among transfers of current income, with anequivalent value also being recorded within inflows of direct investment capital; thusthey will automatically finance the higher deficit. The deficit according to the newmethodology will likely be 2% higher as a percentage of GDP than stated earlier. Itis important to note that this higher amount will not constitute a change in the actu-al equilibrium position of the Hungarian economy.

Bank intermediation strengthened, as the balance sheet total of the Hungarian bank-ing sector grew rapidly, well in excess of GDP growth. The balance sheet total grewby 12.7% in 2003 H1 relative to end-2002 and by 25.6% compared to the sameperiod a year earlier. The rise in the balance sheet total was accompanied by stronglending to households. Increased corporate spending, often perceived as a sign ofeconomic recovery, triggered enhanced demand for loans by non-financial corpora-tions. In addition, households’ credit demand also continued to grow, which wasboosted further in Q2 by an upsurge in lending brought forward as a result of thestricter conditions imposed on the housing loan subsidy scheme in June 2003.Demand for loans by non-financial corporations also continued to grow.

Reflecting the brighter economic outlook, corporate sector demand for credit roseyet again. In respect of the producing sectors, lending to manufacturing gatheredpace. The increase in the number of lending transactions was mainly due to a risein short-maturity foreign currency loans typically extended for funding purchases ofcurrent assets, and, to a lesser degree, a rise in long-maturity foreign currency loansprimarily for fixed investment. The fact that the improving outlook for profitabilitymay reduce credit risks is seen as a favourable development. Banks’ project lendingcontinued to rise, which mostly meant an increase in lending for real estate devel-opment. Although, in recognition of their increased risks, banks require a 30% to50% pre-occupancy rate in the case of office building construction prior to financ-ing real estate development projects, the Bank continues to believe that banks’ risk-taking remains too high. Following a period of constant, vigorous increases, the pro-portion of SME loans in the entire corporate loan portfolio temporarily declined inthe period under review. In the longer term, however, increasingly sharp competi-

Fiscal consolidation andstabilisation of exchangerate expectations arepreconditions for safecurrent account financing

Changes to themethodology of compilingthe balance of payments

In 2003 H1, financialintermediation deepened

For cyclical reasons,increasing indebtedness inthe corporate sector doesnot add to risks


OVERVIEW

tion in the banking sector and saturation of the corporate market are likely to leadto a further rise in SME loans.

The increased volatility of official interest rates makes it more difficult to analysedevelopments in the corporate lending spread over the past one year. Given the cur-rent level of risks, the approximately 1 percentage point spread is considered low,even by international standards.

Despite a slowdown in household income growth rate, demand for consumer cred-it and housing loans continued to grow considerably. The underlying reasons forsuch exceptionally strong growth in indebtedness include the extension of the hous-ing loan subsidy scheme and demand for housing loans brought forward, anticipat-ing the introduction of stricter conditions for housing loans. Even the changes in theconditions of the housing loan subsidy scheme in June 2003 proved insufficient toslow growth in housing loans to any significant degree. Consequently, further sub-stantial increase in the level of indebtedness could bear serious risks. Changes in theconditions of the housing loan subsidy scheme are feeding through to a lower inter-est margin on house loans, which in turn will lead banks to tighten their lendingrequirements and standards somewhat. However, even the new conditions of lend-ing are unlikely to dampen demand markedly. Risks stemming from a price bubbleare not considered to be too high, as vigorous credit expansion has not be accom-panied by unrealistic growth in real estate prices. In the Bank’s view, the loan-to-value ratio in larger towns represents a safe escape value. However, in regions ofthe country where the real estate market is not very liquid, the possibility of sellingreal estate collateral at a fair market value is questionable. In addition to widespreaddemand for housing loans and a sharp increase in non-bank loans, demand for con-sumer credit and other loans has also been expanding briskly. The declining interestmargin on home loans is likely to lead banks to increasingly focus on consumer cred-it and other loans. Given the increase in average interest rates on new consumercredit and other loans, it is safe to assume that banks may want to maintain or evenincrease the growth in lending by relaxing their lending requirements and standards.

The overwhelming majority of household loans extended by financial corpora-tions owned by a bank are foreign currency based. Exchange rate volatilityleads to increased credit risks. Although such increase in credit risks would notpose a risk to the banking sector, it may prove to be a challenge at certainbanks.

Vigorous lending has led to portfolio improvement. The ratio of non-performingloans has also fallen in the corporate segment, putting an end to an earlier trend ofdeterioration. The ratio of non-performing household loans has also declined overthe past year. However, this improvement is likely to be temporary. If growth in thecredit portfolio slows, and as the vast portfolio of new loans enters into a moremature stage of its life cycle, the rate of non-performing loans will likely increase.

Fluctuation in the forint exchange rate was unprecedented in 2003. Due to the factthat banks remained exchange rate risk averse, the increased volatility of theexchange rate was not a source of considerable loss affecting them directly.However, the considerable weakening of the forint has increased banks’ indirectexposure to credit risk, owing to the FX exposure of certain groups of customers.Considering that the proportion of debtors’ credit portfolio affected adversely by theweakening of the forint is small within the entire portfolio, the risks attached to sucha portfolio are not significant at the level of the banking sector. On the other hand,the tendency of banks to rely heavily on forward transactions conducted with com-panies in hedging non-residents’ open forward positions is a major risk factor. In theBank’s opinion, it is questionable whether there are enough companies involved inhedge transactions, to which banks can consistently pass on the effects of sharp fluc-tuations in non-residents’ positions.

Household indebtednesshas grown significantly.Continuation of this trendmay result in serious risks

Foreign currency loansextended to households byleasing companiestransform exchange raterisks into credit risks

Quality of bank portfoliosimproved mainly as aresult of the large numberof new loans

Banks remained exchangerate risk averse


OVERVIEW

Money market rates and long-term yields have exhibited considerably higher volatil-ity in 2003. This increased interest rate volatility may contribute to fluctuations inbanks’ interest income, due to the significant gap between the general repricingperiods of assets and liabilities. A potential massive rise in government securitiesyields, similar to that seen in June, may cause significant losses to banks which holdlarge portfolios; however, the relatively short duration of the banking sector’s gov-ernment securities holdings may mitigate the exposure to interest rate risks.

As in earlier years, growth in deposits was much slower relative to the very rapid risein lending. Accordingly, the banking sector’s loans-to-deposits ratio increased signif-icantly, rising to nearly 100% by end-June. The substantial increase in the loans-to-deposits ratio over the last 18 months can be attributed mainly to the enormous risein housing loans and households’ low propensity to make deposits. Contributing tothis was the increase in the corporate sector borrowing requirement in 2003. Inview of the rapid increase in the loans-to-deposits ratio, the liquidity risks facing thesector are judged as increasing. The high loans-to-deposits ratio, coupled with adecline in the ratio of liquid assets, indicates that banks have increasingly less bufferto manage potential liquidity crisis situations.

The banking sector’s capital adequacy ratio has deteriorated somewhat over thepast 18 months, caused mainly by the increase in banks’ activity exceeding that inregulatory capital. On aggregate, the sector’s capital strength is adequate. This mod-erate deterioration of capital adequacy is expected to continue in the future.

Despite a further slowdown in economic growth, the sector’s profitability improvedconsiderably, even in relation to the previous year’s outstanding performance – atHUF 104 billion, after tax profits were 46% higher than in the base period. The per-ceptible improvement in banks’ profitability over the past 18 months has mainlybeen driven by the wave of government-subsidised housing loans, which has earnedbanks additional profits through the interest margin. In addition to the vast improve-ment in banks’ profits, the increase in differences in profits earned individually in2003, explained by the considerable advantage of the most profitable banks overthe average, also deserves mention. Owing to the rapid expansion of housing loans,net interest income grew more strongly than in the previous year, which in turnplayed an important role in the sector’s outstanding performance; and the robustincrease in fee income continued. The operating costs of the sector rose more quick-ly than inflation. This rise, however, was considerably lower than the increase in rev-enues and was slightly weaker than the growth in the balance sheet total. On thewhole, the sector’s cost efficiency indicators improved.

Interest rate volatility mayadd to the fluctuation inbanks’ interest income

Banks have less and lessbuffer to manage potentialliquidity crisis situations

Capital adequacycontinues to beacceptable

Banks’ profitabilityimproved strongly,however, differences interms of profitabilitywidened

1 MACROECONOMIC INDICATORS



1

THE GLOBAL BUSINESS CYCLE

Of the major regions of the global economy, the out-look for growth in Japan and the US has been improv-ing since the summer, as illustrated by the exceptional-ly strong growth rate seen in the US economy in Q3(above 7%), which considerably exceeded expecta-tions. To a great extent, the brighter prospects forgrowth in the US economy can be attributed to improv-ing profitability indicators in the corporate sector.Cautious optimism may be justified as the corporatesector has managed to cut excess capacity and reducethe debts that it accumulated during the previousupswing in the business cycle. This optimistic outlook isovershadowed by the fact that corporate restructuringmeasures have kept unemployment at high levels, andthe recovery in investment activity has been subdued.

In contrast to the improvement in business activity over-seas, in 2003 Europe failed to recover rapidly from the2001–2002 economic slowdown: euro area economicactivity rose by just one-half of a percentage point in2003 H1. Although expectations related to the businesscycle have been revised down since the spring, theimprovement in economic prospects in the autumnmonths gives some grounds for optimism. The moder-ate growth in Europe has mainly been sustained byincreases in government deficits and stockbuilding,whereas the development of private consumption,investment and net exports has been detrimental togrowth. Weak demand for exports from Europe, influ-enced by the recent appreciation of the euro, has beenthe major barrier to growth, in addition to weak con-sumer confidence.

The euro area economy may not return to its potentialgrowth rate before the second half of 2004. This pro-longed period of anaemic growth exposes economicagents to considerable financial strain, and thus carriesrisks in terms of financial stability as well.

INTERNATIONAL RISK PERCEPTION

As a result of weak economic performance in devel-oped countries, yields fell to extremely low levels in theeuro and dollar markets by the spring of 2003. In linewith the improvement in the outlook for growth in theUS, there was a correction in long-term yields in thesummer, but dollar and euro yields continue to be lowby historical standards.

Investors’ risk appetite for emerging-country assets con-tinued to increase in 2003, with a decline in risk indica-tors. The credit rating of several emerging countries (forexample, Brazil and Russia) has improved. Furthermore,the low yields in developed countries enhanced interna-tional investors’ interest in the region.

The EMBI, a risk indicator calculated from the riskspreads on dollar-denominated bonds issued by emerg-ing countries, has recently been at very low levels com-pared to earlier years. In international capital markets,there is a clear relationship between the level of yieldsseen in advanced economies and risk premia. Due tolow dollar and euro yields, investors have been moreinterested in higher-yielding investments, for example,bonds and shares offering higher returns, which hasexerted downward pressure on risk premia. In additionto the market of developing-country sovereign bonds,this trend has also been witnessed in the market of high-er-risk corporate bonds, both in European andAmerican markets.

A prolonged rise in dollar and euro yields could carrypotential risks to stability over the longer term. Areversal in the interest rate cycle which is expected tobegin in advanced economies in 2004 H2 (althoughthis development remains uncertain at the time being),may alter the investment climate which is currentlybenign for emerging economies. Under such circum-stances, investors will likely hold higher-risk assets onlyat a higher yield differential, making it more expensive

1.1 THE GLOBAL BUSINESS CYCLE AND RISK PERCEPTION

Global and regional growth rates

2001 2002 2003* 2004*

April Sep. April Sep.

Global economy 2.4 3.0 3.2 3.2 4.1 4.1USA 0.3 2.4 2.2 2.6 3.6 3.9Euro area 1.5 0.9 1.1 0.5 2.3 1.9CEECs 3.0 3.0 3.4 3.4 4.3 4.1

Source: IMF (2003): World Economic Outlook, April, September.* Forecast.

Table 1-1



1

for emerging countries to raise finance in the capitalmarkets.

REGIONAL RISK PERCEPTION

Despite the global improvement in the risk perceptionof emerging economies, however, the risk perception ofCentral and Eastern Europe has deteriorated. Thisincreased regional uncertainty can mainly be ascribedto the fact that over the past few years governmentdeficits in some EU Acceding Countries, which weregenerally already higher levels than the 3% thresholdspecified in the Maastricht convergence criterion, havecontinued to rise even higher. In investors’ view, thishas increased the uncertainties surrounding the date ofmeeting the convergence criteria and the commitmentof governments to adopting the euro. All this, coupledwith sluggish growth attributed in part to theunfavourable developments in the European business

cycle, has added to the risks facing the region and theuncertainties related to exchange rates.

In the Czech Republic, the general government deficitis likely to be more than 7% of GDP in 2003, accordingto the government’s forecast. Admittedly though, theCzech government has declared that it would not has-ten the country’s entry into the euro area.Consequently, the deficit forecast has not raised asmuch concern as in Poland and Hungary.

Poland, as Hungary, has publicly committed itself torapid EMU entry. However, the convergence processremains fairly fragile, due to the risks represented by thegovernment deficit. Market participants are also awareof these risks, and thus do not believe in a fast track toEMU membership. Serious doubts were raised inSeptember in connection with the Polish government’sbudget plans, and the zloty depreciated massively, bymore than 5%, against the euro by the end of themonth. The Polish currency continued to weaken up toend-October and long-term yields rose sharply, as eventhe government’s medium-term budget plan, publishedin October, was not sufficient to bolster confidence in arapid adoption of the euro.

Greater regional risks have also been reflected in thecredit ratings by international rating agencies. In June,Standard & Poor’s downgraded the outlook for Poland’screditworthiness to negative. In November, it down-graded the country’s zloty-denominated sovereigndebt. The rating agency Fitch also changed the ratingoutlook for Hungary’s forint and foreign currency-denominated debt to negative.

Russia has had an ambiguous effect on the risk percep-tion of the Central and Eastern European region. Whileinternational rating agencies upgraded the country’sdebt in the summer, the tensions related to the oil com-pany Yukos in October had an opposite effect.

Global indicators of risk

Chart 1-1

250350450550650750850950

1050

Jan. 02

Mar

. 02

May

02

July

02

Sep

. 02

Nov. 02

Jan. 03

Mar

. 03

May

03

July

03

Sep

. 03

Nov. 03

EMBI spread

S&P Speculative Grade Credit Spread

MAGGIE High Yield

Basis points Basis points

250350450550650750850950

1050

Source: S&P, J.P.Morgan.



1

Since May 2003, the Hungarian financial markets havebeen characterised by greater volatility and uncertainty.This increased volatility was expressed primarily in forintdepreciation and a massive rise in yields, mirroring thesignificant increase in the required risk premium onforint-denominated assets. These developments canmainly be attributed to country-specific factors, in par-ticular uncertainties about the convergence process.Market evaluation of Hungary’s macroeconomic funda-mentals has deteriorated; in particular the country’sexternal equilibrium position has raised market partici-pants’ concern. Fiscal policy and developments in infla-tion have raised serious doubts as to whether the coun-try will be able to meet the convergence criteriarequired for the scheduled entry into the euro area in2008. As a result of the general uncertainty surroundingdomestic macroeconomic developments, investorshave reacted more sensitively to regional events, partic-ularly in Poland, as well as to macroeconomic datawhich do not contain relevant information for longer-term economic developments. Hence, on the whole theforint’s vulnerability and its exposure to financial conta-gion have increased.

Although exchange rate expectations have in part beenguided by communications by the Government andthe MNB in relation to the expected date of entry intoERM II, the uncertainties surrounding future exchangerate movements continue to be considerable. Furtherloss of investors’ confidence in the convergenceprocess carries the greatest risk factor, which in turnmay cause an additional increase in required yields onforint-denominated assets and may lead to forint depre-ciation and even higher volatility. The importance offoreign investors’ assessment of the Hungarian econo-my has increased in past years, as the role of govern-ment securities purchases by non-residents in financingthe current account deficit has become ever moreimportant, given the decline in the inflow of directinvestment capital.

The level of the current account deficit may cause adeterioration in the risk perception of forint-denominat-ed assets, which in turn may raise the possibility of anexchange rate correction on the part of investors. If thereduction in the government deficit proves slower or

inflation develops less favourably than expected, thismay also lead to a worsening of risk assessment. Suchnegative developments may cause a postponement ofthe expected date of Hungary’s entry into the euro area.

EXCHANGE RATE AND YIELD DEVELOPMENTS

The past 18 months can be broken down into three dif-ferent periods in terms of exchange rate and yieldsdevelopments.

The period May–July was characterised by unusuallyhigh volatility in the exchange rate of the forint and itspersistent weakness, which has not been experiencedsince the exchange rate band was widened in May2001 (see Chart 1-2). The forint exchange rate, whichhad been relatively stable in the months following thespeculative attack in January, began to decline as aneffect of a number of factors. Investment bank analystshad been concerned about the rising current accountand government budget deficits ever since the begin-ning of the year. On 26 May, the MNB announced thatit would end its operations conducted in the foreignexchange market in the months following the January

1.2 DOMESTIC FINANCIAL MARKETS

HUF/EUR exchange rate

Chart 1-2

230

235

240

245

250

255

260

265

270

Jan

. 0

2

Feb

. 0

2

Ap

r. 0

2

May

02

July

02

Au

g.

02

Sep

. 0

2

No

v.

02

Dec

. 0

2

Feb

. 0

3

Mar

. 0

3

May

03

Jun

e 0

3

July

03

Sep

. 0

3

Oct

. 0

3

HUF/EUR – Inverted scale

The stronger edge of the intervention band

Forint/euro

Source: MNB.



1

speculative attack.1 The exchange rate weakened toEUR/HUF 255 by the end of the month, and then weak-ened further significantly in response to the Governmentand the Bank devaluing the central parity of the forint’sintervention band slightly, by 2.26%, on 4 June. Marketparticipants could not understand this move in terms ofthe Bank’s previous monetary policy and objectives. Theexchange rate weakened more than 5% in the followingtwo days, and the indicators reflecting exchange rateuncertainty increased considerably (see Chart 1-3).

Simultaneously with this exchange rate depreciation,there was a massive rise in yields (see Chart 1-4). As anattempt to offset the increase in required yields, in Junethe Bank raised its major policy rate in two steps by atotal 300 basis points, in order to defend the currency,which resulted in a similar jump in short-term marketyields. Long-term yields also rose considerably, althoughless strongly than short-term yields. Consequently, theyield curve, which was practically flat throughout H1,became strongly negatively sloping, as seen in earlieryears.

The two months following the band shift were charac-terised by uncertainties surrounding the country’s eco-nomic policy objectives, deterioration in the outlookand a loss of confidence in the convergence process. Asa result, exchange rate and yield movements werevolatile. Variations in long-term yields were closelyaligned with those in the forint exchange rate in thisperiod: in most cases, rises in yields were coupled withepisodes of exchange rate weakening. All this indicatesthat changes in yields were predominantly shaped bychanges in the exchange rate risk premium required bynon-resident investors and/or changes in depreciationexpectations. These two factors are difficult to distin-guish between on the basis of the information available.The Reuters survey suggests that, simultaneously withthe shift in the level of exchange rate expectations, theycontinued to follow an appreciating trend. This impliesthat the increase in uncertainty was the decisive factorin the rise in required yields, in addition to a rise in long-term euro yields, starting in mid-June.

The second phase which lasted from end-July until mid-October featured much more stable exchange ratemovements, slow appreciation and a gradual decline inlong-term yields relative to the volatile June–July period.From end-July, there was a perceptible reduction in mar-ket uncertainty, as indicated by the drop in impliedvolatility derived from options prices. Communicationsby the Bank which channelled expectations towards the250 to 260 range must have played an important rolein the reduction of the uncertainties related to theexchange rate.

However, underlying the fragility of the appreciationprocess, the indicators of market uncertainty began ris-ing again in September and, in October the forint beganto weaken. The period since mid-October has beencharacterised by volatile exchange rate movements aswell, with the forint losing some 2% of its value relativeto end-September.

Presumably, a number of factors contributed to the inter-ruption of the appreciation trend. First, concerns over thecountry’s external equilibrium position and the fragility ofcurrent account deficit financing intensified following therelease of the much worse-than-expected balance of pay-ments data for August, which was also reflected in theanalyses by a number of investment banks. Of the eventsrelated to the region, developments in the zloty’sexchange rate must have contributed to the resumptionof the increase in exchange rate uncertainty and to theweakening of the forint exchange rate.

Regional effects clearly played a dominant role in thetransient exchange rate depreciation experienced after30 October – the Hungarian and Polish governmentsecurities markets saw extraordinary rises in yields,which also affected the forint exchange rate. This yieldrise mainly affected medium-term yields: three- and five-year benchmark yields soared by more than 100 basispoints in one day, and yields at short end of the curverose by 50–90 basis points. According to informationavailable to the Bank, this rise in yields may have beendriven by sales of government securities related to one-off, regional causes. Due to the market events in Russiaand Poland, a few market participants were forced to

Changes in implied volatilities relating to theforint exchange rate*

Chart 1-3

0

5

10

15

20

25

May

03

June

03

July

03

Aug. 03

Sep

. 03

Oct

. 03

Nov. 03

Per cent

1 week 12 month

* Derived from option prices.Source: Reuters.

1 Following the January 2003 speculative attack, the central bank triggered an outflow of speculative capital by purchasing forints on the FX market, thus contributingto market conditions returning to normal and improving exchange rate stability. For further details, see Report on Financial Stability, 2003/1.



1

rearrange their portfolios, which resulted in a massiveincrease in yields in a market characterised by low liq-uidity, causing others to close their positions. However,it cannot be ruled out that the longer-term increase inthe risk premium on forint-denominated assets mayhave played a role in the rise in yields.

The Bank attempted to ease the disturbance of the mar-ket by conducting open market operations. The aim ofthe intervention was not to permanently influence thelevel of yields, but to prevent excessive rise in yieldsdue to temporary drying up of liquidity. The interven-tion proved successful: yields have been falling as atrend since early November, although in mid-Novemberthey were still 50–100 basis points higher than at end-October. Taking account of the fact that a similar rise inyields overnight on such a scale has not been seen foryears, a permanent liquidity risk premium componentmay have been built into yields by the market.

EXPECTATIONS RELATED TO EMU ENTRY

Since April the Reuters poll has indicated a clear post-ponement of the expected date of EMU entry.Uncertainties surrounding the course of economic poli-cy in the wake of the band shift and next year’s budgetplans contributed to this development (see Chart 1-5).Whereas in April more than half of the respondents indi-cated 2008 as the most probable date, in Novembermost analysts considered 2009 as the most likely date ofEMU entry. Hence, analysts do not believe that the con-vergence criteria can be met on the appropriate time,despite the Government and the Bank both having spec-ified 2008 as the expected year of entry into EMU.

While analysts see little likelihood of 2008 as the yearof adopting the euro, the orientation effect of commu-nications by the Government and the Bank can bedemonstrated in the expectations related to entry into

ERM II. Most analysts expect Hungary to enter ERM IIin 2004-2005, in line with the timetable specified by theGovernment. Expectations of the central parity of theforint have been fluctuating between HUF/EUR250–260 since August. In August, the MonetaryCouncil specified this range as supporting both pricestability and competitiveness.

REAL INTEREST RATES

In response to the nearly 300-basis-point rise in yieldssince May, real interest rates have increased, althoughmuch more modestly than nominal rates, given theupward shift in inflation expectations. The one-year exante real interest rate has risen by more than 200 basispoints in the past six months (see Chart 1-6). In theBank’s view, this does not carry a risk to stability, as thereal interest rate has returned from the very low levelsseen in the aftermath of the speculative attack to the3%–4% range characteristic of recent years.

Benchmark yields on government securities andthe MNB’s major policy rate

Chart 1-4

4.5

5.5

6.5

7.5

8.5

9.5

10.5

Jan

. 0

2

Mar

. 0

2

May

02

July

02

Sep

. 0

2

No

v.

02

Jan

. 0

3

Mar

. 0

3

May

03

July

03

Sep

. 0

3

No

v.

03

Per cent Per cent

4.5

5.5

6.5

7.5

8.5

9.5

10.5

3-month 3-year

10-year 2-week deposit

Source: MNB, Debt Management Agency (DMA).

Chart 1-5

0

20

40

60

80

2007

2008

2009

2010

Percentage of answers

January April June October

Source: Reuters.

Distribution of most likely dates of EMU entry (Reuters poll)

One-year real interest rates

Chart 1-6

–1

0

1

2

3

4

5

6

7

8

1-year ex ante real interest rate

(Reuters poll inflation expectations)

1-year contemporenous real interest rate

(actual YoY inflation)

–1

0

1

2

4

5

6

7

8Per cent Per cent

Jan

. 9

7

July

97

Jan

. 9

8

July

98

Jan

. 9

9

July

99

Jan

. 0

0

July

00

Jan

. 0

1

July

01

Jan

. 0

2

July

02

Jan

. 0

3

July

03

3

Source: MNB, Reuters.



1

NON-RESIDENTS’ GOVERNMENT SECURITIES PURCHASES

Purchases of government debt securities by non-resi-dents have played a dominant role in current accountdeficit financing in the past 1–2 years. Foreign investors’demand for Hungarian government securities anddevelopments in the factors shaping such demand arethus of utmost importance from the perspective offinancial stability.

Since May 2003, non-resident holdings of governmentpaper increased substantially, by some HUF 250 billion,although this represented a much slower growth ratecompared to earlier years (see Chart 1-7). Holdingswere volatile in the period under review: there wereseveral episodes of sell-offs. The first wave of sales mayhave been related to the confidence crisis following theband shift, when existing holdings declined by aroundHUF 100 billion, although only temporarily. Sales ofgovernment securities on 30–31 October resulted in amuch smaller drop of HUF 40 billion in holdings, which,however, was accompanied by an enormous rise inyields.

Considering the government securities market develop-ments of the past six months, the question can be raisedas to what extent this form of portfolio investments is astable source of deficit financing, and whether thisimplies speculative inflows which could entail massivewithdrawals of capital in the event of a change in for-eign investors’ demand for government securities.Below is an analysis, carried out on the basis of foreigninvestors’ behaviour and the maturity profile of non-res-ident holdings of government securities, of the risks a

sudden change in foreign investors’ demand for govern-ment securities may entail for stability.

Non-residents currently hold 38% of the total stock ofgovernment securities. This high proportion and therapid increase in non-residents’ holdings raise the ques-tion of whether there may be a critical level, in excessof which a decline in foreign demand can be presumed.International experience, on the other hand, seem toindicate that non-residents’ share of forint bonds is notexceptionally high, and thus the current ratio should notrepresent a barrier to growth in its own right.

In past years, the increase in holdings of governmentsecurities has only been interrupted in cases when theeconomic policy credibility has been shaken, andexpectations of the likely continuation of the conver-gence process have deteriorated considerably (forexample, following the band shift in June 2003). Basedon the experience of past years, changes in expecta-tions related to monetary policy over the short term andin investors’ risk appetite influence demand for govern-ment securities to a much smaller extent. In the Bank’sview, therefore, the drying-up of foreign demand forforint-denominated government paper can be expectedto be of longer duration if investors’ confidence in theconvergence declines considerably, which in turn maylead to a significant, lasting increase in the risk premium.

Due to the size limits, large amounts of governmentpaper can only be sold in the market with a considerableimpact on prices – hence, there is little likelihood of asudden reduction in holdings in the event of a confi-dence shock. However, large transactions individuallycould lead to a significant change in yields, due to therelatively low liquidity, causing unjustified yield volatility.

These liquidity considerations may have played a role inthe strong rise in yields at end-October, as outstandingholdings only fell by HUF 40 billion relative to the endof October. There are, however, uncertainties in respectof the extent to which foreign investors’ gloomierexpectations related to Hungary’s economic conditionsplayed a role in the sell-off. Consequently, whether alasting decline in demand should be expected in thefuture remains to be seen.

In past years, foreign investors usually carried out gov-ernment bond purchases without hedging exchangerate risk, consequently taking an exchange rate posi-tion as well as an interest rate position. However, itshould be taken into account that investors canreduce their exchange rate positions using swaps,without actually selling government securities.2 Thishappened in June, at the time of the confidence crisisfollowing the band shift, when there was an increase

Non-resident holdings and average maturity ofgovernment securities

Chart 1-7

10001150130014501600175019002050220023502500

2.02.22.42.62.83.03.23.43.63.84.0

Years HUF billions

Jan.

02

Feb.

02

Mar

. 02

May

02

June

02

Aug

. 02

Sep.

02

Oct

. 02

Dec

. 02

Jan.

03

Mar

. 03

Apr

. 03

June

03

July

03

Sep.

03

Oct

. 03

Outstanding stock (rhs) Average maturity (lhs)

Source: DMA.

2 For details, see Csávás–Kóczán: “Development of the derivative HUF market and its impacts on financial stability” in: Report on Financial Stability, 2003/2.



1

in outstanding swap contracts, simultaneously with adecline in non-resident holdings of Hungarian govern-ment securities. The increase in outstanding swapssuggests that, due to the increased exchange rateuncertainty, some investors reduced their exposure toexchange rate risks by concluding swaps, instead ofselling government paper. In respect of supply anddemand in the foreign exchange market, this strategyis equivalent to selling government securities.Reducing exposure to exchange rate risks in this man-ner and financing government securities purchasesusing swaps, however, is an expensive method, as thenet return on government paper is offset by theimplied return of the swap, which in turn is paid by thebuyer. Any loss or gain derives from the different dura-tions on the assets and liabilities sides in the event ofa shift in the yield curve, given that swaps are general-ly concluded for a short term. The decline in outstand-ing swaps in July–August, for example, can beascribed to the fact that financing by swaps (i.e. bor-rowing in forints) became considerably more expen-sive as a consequence of the official interest rateincreases.

In respect of exposure to fluctuations in capital flows,the maturity profile of non-resident holdings of govern-ment securities contains important information. Theusual target of speculation on short-term, high interestrate differentials or exchange rate changes is the marketshort-dated paper. However, only 16% of non-residentinvestors’ government securities holdings are in shortmaturities and, moreover, such paper as a proportion oftotal holdings has clearly been on a downward trend inrecent years (see Chart 1-8). Whereas foreign investorshold 40–60% of outstanding long-term Hungarian gov-ernment securities, their share of the market of short-

dated paper is less than 20% (see Chart 1-9). This ratiois even smaller in the case of discount treasury bills orig-inally purchased as short-term investments. Here, non-resident holdings have never even reached 10%. Fromthis it can be inferred that the vast majority of foreigninvestors participating in the Hungarian governmentsecurities market hold forint-denominated governmentpaper for the longer term, maintaining their confidencein the convergence process, rather than for the purpos-es of short-term speculation.

Nonetheless, the upward trend in the duration of non-resident holdings of government paper seen since 2001broke off in June, with the indicator falling from 3.9 yearsat end-May to 3.6 years at end-October (see Chart 1-10).Looking at cumulative changes in existing holdings ofnon-residents calculated from end-May, however, it wasnot the increase in demand for short-dated paper butrather for medium-term paper (2 to 4 years) that causedthe duration to shorten, which is not an unfavourabledevelopment from a stability perspective.

EQUITIES MARKET

In the past six months, prices have remained on anupward trend which started in the spring of 2003, inter-rupted by small, transient declines. The major stockindices have risen by 15%–30% since May (see Chart1-11). This increase in market optimism is due mainly tothe improvement in the outlook for global economicactivity, and firms’ rising profitability in particular, aswell as to corporate restructuring contributing to moreefficient operations. Supporting the demand for shareswas the rise in long-term yields, which could also berelated to improving economic prospects.

Maturity profile of non-residents’ governmentsecurities holdings

Chart 1-8

3327

20 29

2713 14 22

35 3138 30

3152

4242

3 4 6 1120 15

26 20

0

10

20

30

40

50

60

70

80

90

100Per cent

May

00

Nov. 00

May

01

Nov. 01

May

02

Nov. 02

May

03

Nov. 03

2938 35 30

22 21 18 16

0–1 year 1–2 year 2–5 year 5 year–

Source: DMA.

Non-residents’ share of the government securitiesmarket in the various maturity brackets

Chart 1-9

17

43

54

40

17

42

5448

16

48

59

47

0

10

20

30

40

50

60

70

0–1 years 1–2 years 2–5 years 5 years–

Per cent

Nov. 2002 May 2003 Oct. 2003

Source: DMA.



1

The CESI stock index for the Central and EasternEuropean region broadly followed international trends(see Chart 1-12). However, at end-October, followingthe scandal at the Russian oil firm Yukos, share prices

in the region started to decline. The Yukos affair hasno direct impact on the economies of Central andEastern Europe. Various events in other emergingcountries have had an increasingly less significantimpact on the risk perception of the EU AccedingCountries in past years. Rather, the regional uncertain-ty characterising Central and Eastern Europe showsthat, due to the general uncertainty surrounding theregion, investors have become more sensitive to infor-mation which do not directly affect the AccedingCountries.

The BUX rose by 17% to the end of October. Thiswas broadly in line with the rise in the major interna-tional indices. However, developments in the indica-tor show a different time pattern, due to country-spe-cific factors. In June, the BUX weakened by more than5%, in contrast to the international trend, reflectingthe uncertainties related to economic policy and theconvergence process which emerged in the aftermathof the band shift. Simultaneously uncertainty sub-sided, the Hungarian stock index resumed rising inJuly. Similarly to the CESI, the BUX also declined inthe wake of the regional uncertainties, although byless.

Stock market activity increased considerably in H1, par-ticularly in Q2. At HUF 8.5 billion, daily average turn-over was high relative to previous years.

Changes in government securities holdingsaccording to maturity betweenJanuary–November

Chart 1-10

–400

–200

0

200

400

600

800

HUF billions HUF billions

–400

–200

0

200

400

600

800

With maturity until Dec. 31, 2005 (inc. T-bills)

With maturity between Jan. 1 2006 and Dec. 31, 2008.

With maturity expiring after Jan. 1, 2009.

Jan. 03

Feb

. 03

Mar

. 03

Apr.

03

May

03

June

03

July

03

Aug. 03

Sep

. 03

Oct

. 03

Nov. 03

Source: DMA

Major stock exchange indices(31 December 2002 = 100%)

Chart 1-11

70

Dec

. 02

Jan.

03

Feb.

03

Mar

. 03

Apr

. 03

May

03

June

03

July

03

Aug

. 03

Sep.

03

Oct

. 03

Nov

. 03

90

110

130

150Per cent Per cent

70

90

110

130

150

Nasdaq comp S&P 500 DAX

Source: Reuters.

Chart 1-12

7080

90100

110120

130140

150

7080

90100

110120

130140

150

BUX HUF BUX USD CESI USD

Per cent Per cent

Dec

. 02

Jan.

03

Feb.

03

Mar

. 03

Apr

. 03

May

03

June

03

July

03

Aug

. 03

Sep.

03

Oct

. 03

Source: Reuters.

Central and Eastern European stock exchangeindices (31 December 2000 = 100%)



1

PROSPECTS FOR GROWTH

The rate of GDP growth has declined gradually inrecent years in Hungary, with economic growth drop-ping to 2.5% in 2003 H1 (see Chart 1-13). From a stabil-ity perspective, this longer period of decelerating eco-nomic growth entails risks, as the unfavourable develop-ment of business activity influences corporate sectorprofits, firms’ prospects and, indirectly, the financial sec-tor providing finance for firms.

In terms of the main components of economic growth,the slowdown in exports was principal factor behind thefall in GDP growth, in line with languid European busi-ness activity. Whereas import growth has remaineduninterrupted since 2001, due to strong domesticdemand, export growth has fallen by more than 6 per-centage points. Consequently, net exports, which hasbecome strongly negative, detracted from economicgrowth in 2003 H1.

According to the latest data, the Hungarian economypassed its cyclical trough in 2003 H1, and a modestrecovery is expected for the period ahead, in line withthe recovery in economic activity in the euro area. As asmall, open economy, Hungary conducts the majorityof its foreign trade with members states of the euro

area, thus events in Europe have a profound impact oneconomic cycles. Nevertheless, in past years the behav-iour of economic agents and the very robust increase indomestic demand have diverted the path of the econo-my from that justified by the external economic environ-ment.

Hungarian export dynamics has closely followed thebusiness cycle of the country’s trading partners. Importvolume of Hungary’s trading partners had beendecreasing since the end of 2000 and will likely onlyreach the 2000 level at the end of 2003. Despite theunfavourable business climate the Hungarian economywas able to continuously increase its market share inthe EU, following a temporary decline in 2000 (seeChart 1-14). Although this trend was broken in 2003Q1, it was soon corrected: for the whole year of 2003the market share of Hungarian companies will increasefurther in the EU.

The stimulation of domestic demand over the last twoyears, closely related to the parliamentary cycle, hasmitigated the risks arising from the weakness of theEuropean economy, as it has contributed to sustainingeconomic growth and diminished the fluctuations of theHungarian economic cycle. However, such a composi-tion of economic growth cannot be sustained over the

1.3 GROWTH AND INFLATION

GDP growth(Annualised quarter-on-quarter growth rates)

Chart 1-13

0.0

1.0

2.0

3.0

4.0

5.0

6.0

7.0

Per cent

1998 Q

1

1998 Q

3

1999 Q

1

1999 Q

3

2000 Q

1

2000 Q

3

2001 Q

1

2001 Q

3

2002 Q

1

2002 Q

3

2003 Q

1

Hungary’s market share in the EU*

Chart 1-14

1.0

1.5

2.0

2.5

3.0

3.5

1995

1996

1997

1998

1999

2000

2001

2002

2003

Per cent

* Share in total external import of the EU.Source: MNB calculations based on Eurostat.

Source: CSO, MNB.



1

longer term. Consequently, Hungarian economic policyhas shifted towards a gradual reduction of the govern-ment deficit. However, the necessary contraction ofdomestic demand will likely have different effects in thevarious sectors, and these effects may have implicationsfor stability, primarily in the case households.

According to forecasts, this unsustainable pattern ofgrowth will be followed by very significant adjustmentin 2004, affecting households in particular, which willallow for more balanced growth to unfold in the follow-ing years. Household real income increased by morethan 20% in 2002–2003 period, but it will remain prac-tically flat in 2004. The changing behaviour of thehousehold sector, caused by the stagnation in realincomes, exposes to a number of stability risks. The sec-tor’s expectations related to its income may deterioratefurther, due to unfavourable labour market develop-ments, which in turn may add to risks. First, despite theupswing in business activity, a significant increase in pri-vate sector demand for labour is not expected, due tothe recent substantial rise in real wages. Second, theannounced large-scale staff reduction in the govern-ment sector has caused a negative turn in incomeexpectations.

The strong decline in consumption growth bears signif-icant risks, the effect of which extends beyond thehousehold sector. One factor that may add to risks at

the macroeconomic level is that the services sector hasembarked on a large investment programme, in thebelief that consumption expenditure would grow con-tinuously. According to forecasts, consumption growthwill only slightly exceed the modest increase in house-hold income in the coming years. The expected slowgrowth in consumption in 2004 raises doubts in respectof capacity expansion in the services sector, particularlyif such is based on previous years’ rapid consumptiongrowth. Any increase in household sector demand forconsumer credit for the purposes of consumptionsmoothing may add to the risks facing the financial sec-tor, as a result of which the quality of outstanding con-sumer credit, accounting for an increasing share of thebanking systems’ balance sheet, may deteriorate signifi-cantly. In addition to new consumer credit, the repay-ment of existing loans may cause difficulties, given theexpected stagnation in income.

In the housing market, risks may appear through numer-ous channels, in connection with the extremely strongrise in housing loans in recent years. The rapid rise inhousing loans could still carry risks in its own right, whileadditional risks may also emerge. The overwhelmingmajority of existing debtors incurred their loan liabilitiesat a time when real incomes were rising rapidly.Consequently, stagnating income may become a sourceof risk, due to the burden of repayment. This is furtheraggravated by the partial repeal of tax preferences linkedwith housing loans. In addition to these factors, the dete-rioration in income expectations may slow the appear-ance of new demand in the housing market.

The banking sector is not directly affected by a poten-tial default in the repayment of existing housing loans,as, very prudently by international comparison, mort-gage loans account for only 60%–70% of the value ofthe property. Nonetheless, if a substantial number ofdebtors were unable to continue servicing their debtsamid deteriorating income conditions, and the homesof those in default were to be sold en masse, this wouldentail significant social costs. The tightening of PIT pref-erences as a result of government measures furtherreduces the disposable income of households affected.

Annual growth rate of GDP and its components(Percentage changes on a year earlier)

Actual Forecast**

2001 2002 2003 2003 2004 2005H1

Householdconsumption 5.3 9.4 7.9 7.2 1.7 2.5

Household finalconsumptionexpenditure 5.7 10.5 8.7 7.9 2.3 2.6Social trans-fers in kind 3.8 4.9 4.4 4.0 –1.0 2.0

Public consumption 4.9 5.0 5.7 4.0 1.5 2.0Gross fixed capital formation 3.5 7.2 0.5 3.6 3.9 4.5‘Final domesticsales’* 4.8 8.4 6.2 6.0 2.2 2.9Domestic absorption 1.9 5.4 6.6 6.6 2.3 3.0Exports 8.8 3.8 2.4 3.4 7.5 8.1Imports 6.1 6.1 7.5 8.0 6.0 7.0

GDP 3.8 3.5 2.5 2.9 3.2 3.6

* Final domestic sales = household consumption + public consump-tion + gross fixed capital formation.

** Forecasts based on the Inflation Report (November 2003).

Table 1-2

Household consumption, savings and fixedinvestment (Annualised growth rates, in percent)

Household Real Real valuereal net consumption of fixed capitalincome expenditure formation

2002 Actual/Estimate 12.3 10.5 20 – 30

2003 7.9 7.8 0 – 102004 Forecast* 0.8 2.3 (–10) – 02005 3.1 2.6 (–5) – 5

* Forecasts based on the Inflation Report (November 2003).

Table 1-3



1

In the case of the average loan (HUF 6.5 million) and anincome above HUF 4 million, the abolition of the taxincentive, on an accruals basis, equals the effect of anapproximately 5-7% hike in interest rates. This tighten-ing does not only affect new loans, but also adds to theburdens of those who borrowed in earlier years.

A change in demand in the housing market may lead toincreased volatility of house prices, as supply isextremely inflexible over the short term, due to the timerequirement of construction. If, as a consequence ofstagnating income and the unfavourable outlook for thelabour market, income expectations fall considerablyand households markedly reduce their housing invest-ment and thus demand for housing loans out of caution,then this may cause increased volatility of house prices,on the one hand. Falling demand may cause difficultiesfor investors, adding to the risks facing their financinginstitutions, on the other hand.

In terms of the structure of corporate investment, onecan see that the growth rate of corporate investment atthe aggregate level has exceeded manufacturing invest-

ment. The gap in growth rates can be attributed to thestrong investment activity of market service providersbased on domestic demand.

In the corporate sector, the prospects for external eco-nomic activity are a source of uncertainty for manufac-turing investment, in addition to the risks facing marketservice providers caused by the change in householdconsumption. The economic outlook for manufacturingfirms producing mainly for export has improved recent-ly, and capacity utilisation and the ratio of firms judgingtheir capacity as low relative to future orders for pro-duction have both increased. Taking into account thatthese firms have based their investment programmes ona rapid upswing in economic performance in the euroarea, a slow recovery in export markets may causestrains on the financing of firms.

Overall, in the Bank’s forecast the economy passes itscyclical trough in 2003, with growth then gaining pacein 2004. The economy may also see a shift toward ahealthier structure, due to the very strong slowdown inhousehold consumption growth and the anticipated risein corporate investment, which will allow for more bal-anced economic growth in the years ahead.

The anticipated changes in households’ consumptionand savings behaviour may be major sources of risks tomacroeconomic stability. A slower-than-expected recov-ery in the euro area economy, which in turn may causecorporate sector profitability to improve more slowly,represents another possible source of risk.

INFLATION

Since the beginning of 2003, inflation has remainedbelow 5% and core inflation has continued to decline(see Chart 1-16). But despite this year’s favourableresults in brining inflation down, the general price levelwill pick up considerably in 2004 as a one-off impact ofthe increases in indirect taxes, which adds to uncertain-ties. Rising inflation carries the risk that inflation may

Corporate investment

Chart 1-15

Manufacturing investment Corporate investment

–20

–15

–10

–5

0

5

10

–20

–15

–10

–5

0

5

10

1515

2000 Q

1

2000 Q

2

2000 Q

3

2000 Q

4

2001 Q

1

2001 Q

2

2001 Q

3

2001 Q

4

2002 Q

1

2002 Q

2

2002 Q

3

2002 Q

4

2003 Q

1

2003 Q

2

Fixed investment (Annual percentage changes)

Weights 2001 2002 2003 2004 2005%

Estimates Forecast**

Corporate sector 57 1.0 (–2.1) 2 – 6 3 – 7 4 – 8General government* 19 (–6.9) 29.5 2 – 5 0 – 8 0 – 6Households 24 21.4 18 0 – 5 (–5) – 2 (–5) – 0

Whole-economyfixed investment 100 3.2 7.2 2.0 – 4.0 2.0 – 6.0 2.5 – 6.5

Investment data which may differ from those on gross fixed capital formation.* Government spending on motorway construction is included in general government data.

** Forecasts based on the Inflation Report (November 2003).

Table 1-4

Source: CSO.



1

become more volatile and that the uncertainties sur-rounding inflation may increase. Evidence from surveysof corporate managers conducted by Reuters andTÁRKI shows that, although one-year ex ante inflationexpectations (standard deviation of expectations) hasincreased, the uncertainty surrounding the expectationshas not exceeded the standard level, despite the infla-tion shock anticipated for 2004. This suggests that theexpected rise in inflation was derived from factors thatcould be calculated with relative ease.

Despite market analysts’ aptitude for anticipating infla-tion developments, in the last two years actual inflationhas been lower than their forecasts looking 12 monthsahead. The margin of error (1–2%), however, was sonarrow that we cannot consider any surprise disinflationas having implications for stability (see Chart 1-17).

In respect of inflation expectations, the fact that corpo-rate managers’ inflation expectations have persistentlyand significantly remained above actual inflation in theyears since the widening of the intervention band car-ries risks to stability. Nevertheless, corporate managers’expectations have been consistent with their observa-tion of inflation, as the difference between expectedand observed inflation is less significant.

The reaction of corporate pricing behaviour to the increasein VAT may be an additional source of risk to inflation, asthis changes the information content of prices in marketeconomies. This issue is relevant for firms which producegoods that will be affected by the tax increase. Assumingthat their costs rise in line with inflation, these firms practi-cally face two opportunities. First, they may increase pro-ducer prices, in line with inflation. However, tax increaseswill add to the consumer prices, with an ultimate increasein real prices of goods, which translates into lower salesand profits. In this manner, firms risk pricing themselves outof the market. If, however, firms raise producer prices byless than the rate of inflation, then the real prices of goodswill not increase excessively, despite the tax increases, andthus the tax increases will have less of an impact on theirsales. In this case, firms’ profits will also fall, as their costswill rise more rapidly than their prices. Cost adjustmentmay offer a solution to the problem, an important elementof which could be curtailing wage costs.

According to the survey conducted by TÁRKI, corpo-rate managers’ wage expectations for the next 12months do not increase, unlike their inflation expecta-tions, which is indicative of adjustment on the costs side(see Chart 1-18).

In contrast with last year’s practice, in October theNational Interest Reconciliation Council made a pro-posal for gross national wage increase in 2004 ratherthan for an increase in real wages, which may help co-ordinate inflation expectations.

The inflation shock caused by the one-off jump in pricesmay lead to a lasting increase in inflation expectations,which carries a significant risk to longer-term inflationdevelopments. According to the Reuters poll, analystsexpect disinflation to continue in 2005 and hence donot expect inflation to get stuck (see Chart 1-19).However, key labour market participants have not yetformulated their own inflation expectations for 2005.Consequently, the effect of next year’s price increaseson expectations continues to be uncertain.

CPI and core inflation(Percentage changes on a year earlier)

Chart 1-16

3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

7.0

7.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

6.5

7.0

7.5

Core inflationConsumer price index

Jan. 02

Mar

. 02

May

02

July

02

Sep

. 02

Nov. 02

Jan. 03

Mar

. 03

May

03

July

03

Sep

. 03

Per cent Per cent

Actual inflation and expected inflation rate for12 months ahead

Chart 1-17

0

5

10

15

0

5

10

15

20 20

Corporate leader's expectations

Reuters expectations

Corporate leader's observed inflation

CPI

Jan. 97

June

97

Nov. 97

Apr.

98

Sep

. 98

Feb

. 99

July

99

Dec

. 99

May

00

Oct

. 00

Mar

. 01

Aug. 01

Jan. 02

June

02

Nov. 02

Apr.

03

Sep

. 03

Per cent Per cent

Source: CSO, Reuters, TÁRKI.

Source: CSO.



1

Corporate managers’ inflation and wage expectations in the TÁRKI survey

Chart 1-18

6.0

7.0

8.0

9.0

10.0

11.0

6.0

7.0

8.0

9.0

10.0

11.0

Feb.

02

Apr

. 02

July

02

Oct

. 02

Jan.

03

Apr

. 03

July

03

Oct

. 03

Expected wage increase for 12 months ahead

Expected inflation rate for 12 months ahead

Per cent Per cent

Inflation expectations based on the Reuters poll

Chart 1-19

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

December 2003. December 2004.

December 2005.

Jan. 02

Mar

. 02

May

02

July

02

Sep

. 02

Nov. 02

Jan. 03

Mar

. 03

May

03

July

03

Sep

. 03

Nov. 03

Per centPer cent

Source: TÁRKI.Source: Reuters.



1

In the course of 2002–2003, the saving/investment bal-ance of the individual sectors of the national economyunderwent major restructuring. As the currently appliedmethod disregards reinvested earnings, the GDP-pro-portionate current account deficit (which was 3.9% in2002) is expected to reach 6.4% in 2003 (see Chart 1-20). If reinvested earnings are also considered, in ourestimation the current account deficit (and the inflow ofFDI and portfolio capital) grows by more than 2% ofGDP. As a methodological change will be introducednext year, the external economic equilibrium, and thepositions of the economic sectors are analysed hereaccording to the new method, i.e. including reinvestedearnings. In 2003, the current account deficit is likely to

increase as a result of an expected 4% drop in the pri-vate sector’s GDP-proportionate net financial savings.Thus, more foreign resources are required to financethe general government deficit.

Based on the Bank’s projections, in coming years thecurrent account deficit is expected to remain nominallyunchanged but decrease slightly as a proportion ofGDP. Nevertheless, the composition of the saving-investment position according economic sectors isexpected to take a favourable turn, the net borrowingrequirement of the general government to decreaseand that of the corporate sector to increase.

Saving/investment balance of institutional sectors

Household lending capacity dropped to historic lowsin 2003. In the first six months of the year this sectorwas a net borrower, and according to the balance

1.4 EXTERNAL EQUILIBRIUM

Methodological changes in the balance ofpayments statistics

In 2003 and 2004, the methodology used forHungarian balance of payments statistics isbeing changed significantly in two stages.From 2003, calculation of the trade in goodsbalance is based on customs data in order toalign the system of national accounts with bal-ance of payments statistics, in contrast to theearlier methodology based on the cash account-ing approach, which relied on reporting bybanks.

A further change will involve the recording ofreinvested earnings from 2004. This affects threeitems and will lead to a deterioration ofHungary’s international income balance in thebalance of payments, while increasing the inflowof FDI and portfolio investment on the financingside. According to the Bank’s calculations, themethodological change may increase the cur-rent account deficit by more than 2% of GDP.The Magyar Nemzeti Bank will release the datarevised for the period between 1995–2003 forthe first time in April 2004.3

Box 1-1

Net lending of sectors and current accountdeficit as a proportion of GDP as per current and2004 methodology

Chart 1-20

–10

–8

–6

–4

–2

0

2

4

6

1999 2000 2001 2002 2003 2004–10

–8

–6

–4

–2

0

2

4

6

8

Forecast

8

General government

Household sector

Corporate sector (with reinvested earnings)

Corporate sector (without reinvested earnings)

Current account (with reinvested earnings)

Current account (without reinvested earnings)

Per cent Per cent

3 For more details on methodology changes, see Inflation Report November 2003. For details on the deficiencies of the new methodology with respect to previousperiods see Sisak-Fekete, Zsuzsa–Vadas, Gábor: A new approach to external balance (MNB Background Study, currently in publication).

Source: MNB, MNB estimation.



1

sheets of commercial banks, this phenomenon contin-ued in 2003 Q3. Due to the seasonal nature of sav-ings, the central bank projects slightly positive netlending capacity.

The 2003 drop in household lending capacity was due tostructural as well as one-off factors. Households’ financialassets rose by nearly 14% in 2003 Q1, but their debtsincreased much faster: the level of debt grew by approx-imately 70% due to a steep rise in consumer and hous-ing loans. Growth in the debt portfolio is due in part to adevelopment in the system of financial intermediation,because the evolution of household lending allows forthe advancement and funding of consumption and capi-tal formation expenditures originally planned to be spentat a later date. However, the introduction of state subsi-dies for housing loans gave sudden momentum to thestructural development of financial intermediation. Theupswing in household lending and the advancement ofconsumption and capital formation expenditures is likelyto continue. As a result, households’ financing capacity isexpected to remain low over the medium term.

In 2003, one-off factors also led to an acceleration ofthe sector’s long-term structural indebtedness. Althoughboth households’ real income and consumptionincreased by 7.8% and the gross savings rate4 did notchange on a year earlier, restructuring of savings result-ed in a lower financial savings rate. The reason for thisis that, by taking housing loans, households increasedtheir housing investment at the expense of their finan-cial savings, in order to avoid the consequences of theannounced regulatory changes directed at reducingstate subsidies.5

As a result of the 2004 public sector staff reductions,decelerating wage growth and a simultaneous rise ininflation, real growth of disposable household incomewill decrease considerably. As the sector is expected tocontinue smoothing its consumption, a drop is forecastin the gross savings rate (see Chart 1-21). However,changes in the housing loan subsidy scheme are likelyto dampen mortgage loans, and consequently, theexpansion of household investment activity. Thus, thecomposition of gross savings is expected to change.Contrary to 2003, household capital formation expendi-tures will probably decrease, while financial savings willincrease. If in adjustment to the current macroeconom-ic trends, households moderate their expectationsregarding the future course of their available incomeand increase the amount of prudential savings, the rateof financial savings could increase. Based on the above

factors, household financing capacity is expected toincrease slightly in 2004.

In 2002, the corporate sector was a net saver.6 Thus, asector that is characteristically a net borrower reducedthe external borrowing requirement of the nationaleconomy. Owing to flagging economic growth anddeclining external demand, companies postponedinvestment projects and reduced stocks. As a result,their borrowing requirement dropped drastically.However, this should be a temporary phenomenon, aswith an economic recovery investment demand willboom, and companies will have to replenish stocksagain. The fact that due to an improvement in cyclicalexpectations, in 2003 corporate demand for fixedinvestments grew while their financing capacity fell isindicative of the aforementioned process. Based on theBank’s projections, in 2004 as well as over the mediumterm, the increase in net household savings will proba-bly be unable to offset growth in the corporate sector’sborrowing requirement. For this reason, only a decliningborrowing requirement of the general government cancounterbalance the resulting impact on the currentaccount deficit.

In the Bank’s estimation, the borrowing requirement ofthe general government may decrease by one percent-age point next year. This is expected to offset the pri-vate sector’s declining net financial savings, thus in2004 the GDP-proportionate current account deficit islikely to decrease.

4 Gross household savings are that part of the available income which has not been consumed, i.e. the amount used for accumulation and financial savings.5 As early as the first half of 2003, the government announced its intention to change the system of state subsidies. Nevertheless, the specific content of the changes

remained uncertain for a longer period of time. The current regulation may affect the housing market adversely for several reasons: cuts in tax allowances increaseeffective interest rates charged on housing loans, and as VAT is to be imposed on construction lots, property prices will increase.

6 Disregarding reinvested earnings, the sector is expected to remain a net lender in 2003 as well as 2004.

Gross savings of households (and their impacton financial savings and fixed investment) as aproportion of disposable income

Chart 1-21

0

2

4

6

8

10

12

14

16

18

1997 1998 1999 2000 2001 2002 2003

Per cent Per cent

0

2

4

6

8

10

12

14

16

18

20 20

Gross saving rate Accumulation rate

Net financial saving rate

Source: MNB.



1

CURRENT ACCOUNT FINANCING

In addition to the deterioration in external equilibrium,the composition of capital flows financing the currentaccount deficit also underwent a major change (seeChart 1-22). Simultaneously with the declining invest-ment activity, net FDI inflow dropped considerably,falling behind the data recorded in other Central andEastern European countries. The short-fall in net FDI isfunded primarily from debt-generating items, morespecifically, the role of HUF-denominated governmentsecurities purchased by foreigners grew considerably. Asa result of these trends, the GDP-proportionate grossexternal debt of the national economy is expected torise by 5 percentage points to reach 22% of GDP by theend of 2003. For cyclical and structural reasons a furtherrise in this indicator is likely over the next few years.

The decline in net FDI inflow is due, inter alia, to inter-sector restructuring of the savings-investment balance(see Chart 1-23).

In contrast to previous years, the external borrowingrequirement of the national economy is being generat-ed by a consistently high borrowing requirement ofthe general government and not the corporate sector.

To a considerable extent, the corporate sectorfinances its investment from FDI, which has ensured asignificant amount of capital inflow in recent years.However, due to the slump in the growth of externaldemand and investment, over the last two years theborrowing requirement of the sector has fallen consid-erably, leading to a decline in the inflow of gross for-eign direct investment as well. A further factor reduc-ing net FDI inflow was the fact that the regional expan-sion of domestic companies generated an outflow ofFDI. The general government finances its deficit prima-rily by issuing bonds and (on a provisional basis and toa minor extent) privatisation. Hence, the nationaleconomy’s external borrowing requirement is beinggenerated by a sector, which is unable to absorb FDI.Consequently, net FDI inflow has naturally decreasedwhile the role of debt generating items has increased,in terms of external resources financing the currentaccount deficit. The banking system obtains externalresources required for rechannelling by taking long-term FX loans. However, the overwhelming majority ofgeneral government deficit is financed directly by non-residents: in 2002 and 2003, more than 50% of theissued net HUF-denominated bonds were purchasedby non-residents.

Although the FDI balance is currently determined bythe position of the general government and the cyclicalsituation, long-term structural trends are indicative of adeclining significance of FDI in financing the currentaccount deficit. Corporate capital structure dependsequally on profit and expected volatility. The rising prof-

Current account balance and structure of financing as a percentage of GDP

Chart 1-22

–6

–4

–2

0

2

4

6

8

1999 2000 2001 2002 2003

Per cent–8

–6

–4

–2

0

2

4

6

Per cent

The net change of banks external debt

The net growth of forint deposits

and government securities owned by non-residents

Net purchase of shares of non-residents

The net change of foreign debt of the corporate sector

Net FDI balance (with reinvested earnings)

Current account balance with reinvested earnings (right axis)

Hungary’s FDI balance

Chart 1-23

FDI abroad

FDI in Hungary without reinvested earnings

FDI in Hungary with reinvested earnings

Net FDI in Hungary without reinvested earnings

Net FDI in Hungary with reinvested earnings

–2

–1

0

1

2

3

4

5

1999 2000 2001 2002 2003 2004 2005–2

–1

0

1

2

3

4

5

Forecast

EUR billions EUR billions

Source: MNB.

Source: MNB, MNB estimation.



1

itability of small and medium-sized enterprises improvestheir creditworthiness, as simultaneous saturation of thecorporate market forces the banking system to increas-ingly turn towards clients with higher risks, and as themarket economy develops further, the indebtedness ofSMEs may rise. Another factor increasing capital gearingmay be the fact that in the event of a lower expectedvolatility of profits, in terms of optimum funding struc-ture the significance of debts grows while the role ofequity decreases.7 This is because in addition to the spe-cific features of individual companies and sectors,macroeconomic and regulatory risks also influence theexpectable volatility of profits. In the case of convergingcountries the latter two factors taper off with develop-ment, thus the optimum corporate capital gearing8 grad-ually increases. This is the reason that in developedcountries the corporate sector generally involves exter-nal resources through debt generating items. Althoughthe Hungarian corporate sector’s indebtedness roseconsiderably in the second half of the 1990s, by inter-national comparison it cannot be deemed as high.Therefore, a cyclical boom may result in a further rise.As a result of the this structural trend, Hungarian com-panies finance their FDI outflow-generating regionalexpansion by borrowing.

Various analyses often voice the concern that in thepast two years the net FDI inflow into Hungary has fall-en to a level considered extremely low even by region-al standards. However, a simple regional comparisonhas many pitfalls. The analyses usually compare the FDIbalances of the CEE counties, but ignore the fact thatthe methodology behind balance of payment statisticsare different. In contrast to the other countries of theregion, Hungary currently applies a methodology thatdisregards reinvested earnings in the balance of foreigndirect investment, and this amounts to more than 2%of GDP.

Moreover, in regional comparisons it must be takeninto consideration that in contrast to the other coun-tries of the region, Hungary is already over the first twowaves of FDI influx, namely privatisation and the relo-cation of foreign companies and subsidiaries.Consequently, in Hungary’s case the FDI balance isdetermined primarily by cyclical expectations and thestructural trend of indebtedness. Whereas in countriesopening up at a later date, these trends are still sup-pressed by the inflow of capital for privatisation pur-poses. For this reason, it is more expedient to comparethe individual countries of the region on the basis of

the per capita FDI stock or the FDI balance excludingprivatisation and reinvested earnings. Based on theabove indicators, Hungary’s situation is by no meansunfavourable.

7 The reason is that with more volatile income and the same debt, the probability of bankruptcy is higher and characteristically the owners’ residual property left afterwinding up is charged with its costs. The ideal size of own equity is further increased by information asymmetry and higher bankruptcy costs (e.g. if the company hasa great deal of assets that are difficult to sell, such intangible assets). Due to the above, small and medium-sized enterprises as well as rapidly growing large compa-nies and the corporations active in high-risk sectors typically finance themselves from their own equity. This is the reason that in the 1990s, despite a highly devel-oped financial system in the United States, the current account deficit was funded from foreign direct investment. As the IT sector runs high risks, grows fast and hasa high rate of intangible assets, it had to form its own equity to meet its borrowing requirements.

8 See Világi, Balázs: The relationship between current account deficit and financial stability, in: Report on Financial Stability, 2002/2.

Inflow of investment capital to Hungary ina regional comparison

Since the early 1990s, FDI has played in increas-ingly significant role in modernisation processesin CEECs. The conditions for a vast inflow of FDIwere created the most rapidly in Hungary. Thiswas facilitated by the privatisation strategiesadopted by the successive governments inpower after the regime change and by legal reg-ulations, which were uniquely transparent in theregion. As a result, by the mid-1990s in terms ofits FDI attraction capacity, Hungary had becomethe most important participant in Central andEastern Europe, on the basis of both absolute(currency unit-denominated) and relative (GDP-and population-proportionate) indicators.

The initial advantages that Hungary enjoyed dis-appeared in the late 1990s. The reason for this isthat investor-friendly legislation was enacted inrapid succession in other countries in the region,and that in the privatisation process, the relevantgovernments offered a wealth of increasinglyhigher value to investors. Meanwhile, privatisa-tion was drawing to a close in Hungary, with thestock of skilled labour depleted. As a result,Hungary ceded its leading position to otherCEECs, first in terms of absolute figures and thenin relative FDI indicators. In fact, FDI inflow intoHungary seemed to be slowing down. This wasespecially the case in the very first years of the21st century, though a downturn in the globalbusiness cycle, which curtailed FDI inflow con-siderably, was also a major contributing factor.

This issue is all the more delicate as FDI inflow,as the most desirable financing component ofthe current account deficit (inasmuch as it doesnot add to the country’s indebtedness), has car-ried considerable weight from the outset inHungary. Since Hungarian firms, which havealso gained financial strength over the past peri-

Box 1-2



1

od, have also emerged as direct investorsabroad, net FDI (the difference between internaland external FDI), which actually finances thecurrent account deficit, has fallen sharply.

Owing to the developments that balance of pay-ments statistics reveal, recent comparisons withother CEECs, in particular the Czech Republicand Poland, have often portrayed Hungary dis-advantageously.

The chart shows that since 1995 Hungary hasattracted a steadily declining amount of FDI asa proportion of GDP. By contrast, in Polandand especially in the Czech Republic FDIinflow has accelerated. This comparison is,however, inaccurate for reasons of content, asbalance of payments statistics in Hungary donot include what is called reinvested earnings,which denotes the amount of the profit earnedby the FDI in the country during any given yearwhich is reinvested in the same country. Thiscomponent has been included in the balanceof payments statistics in both Czech Republicand Poland, and will also be in Hungary’s statis-tics from 2004 (including data for the past). Inthe interest of the comparability of content,reinvested earnings have been estimated. Suchestimates have been employed to correct FDIinflow into Hungary.

As indicated above, the countries in the regionimplemented privatisation at very differenttimes; nevertheless, the amount of assets thatcan be sold is limited in every country. In otherwords, over the long run this constituent has nosignificance in the FDI inflow. For this reason,the Bank has attempted to estimate the share of

privatisation-related FDI in these three countries.After revenues from privatisation had been cal-culated, these were deducted from the figures ofthe balance of payments statistics. In this man-ner, it is possible to compare the three countriesin terms of long-term FDI trends, with the tempo-rary effects stemming from the time differencesin privatisation filtered out.

Compared to the balance of payments data, asignificantly different picture is given of FDIinflow after this double adjustment.

As is clear from the chart, a considerable amountof the annual profit on FDI received in Hungaryhad been reinvested after 1995. This is to say, upto 2001 the FDI inflow remained permanentlyhigh, in excess of 6%, as a proportion of GDPand no unambiguous decline can be established.Similarly to Hungary, in the other two countriesthe FDI inflow as a percentage of GDP does notincrease after 2000 if privatisation revenues areremoved from the balance, although Czech FDIinflow may be deemed as outstanding in theregion. At the same time, it cannot be deniedthat an extremely significant drop in HungarianFDI inflow was seen in 2002, which continues in2003 according to available data.

Having completed the adjustments that providefor the content-related comparability of the FDIinflow data of the countries of the region, it canbe concluded that up to 2001 Hungary’s FDIattraction was not impaired. An undoubtedly sig-nificant fall took place only in 2002. Although toa minor extent, such a drop can also beobserved in the other two countries, and mayreflect a global decline in FDI flows.

FDI inflow as a percentage of GDP basedon balance of payments statistics

Chart 1-24

0.0

2.5

5.0

7.5

10.0

12.5

15.0

1994

1995

1996

1997

1998

1999

2000

2001

2002

Per cent

Poland Czech R. Hungary

Chart FDI inflow as a percentage of GDPwith comparable content

Chart 1-25

0.0

2.0

4.0

6.0

8.0

10.0

12.0

1994

1995

1996

1997

1998

1999

2000

2001

2002

Per cent

Poland Czech R. Hungary



1

RISKS TO THE EXPECTED DEVELOPMENTS INEXTERNAL EQUILIBRIUM

In countries where the per capita physical and humancapital fall short of the corresponding figures of devel-oped countries and the yields on such are consequent-ly higher, a long-term current account deficit is a naturaland beneficial phenomenon. If it is not excessively high,the external borrowing requirement of the corporatesector’s productivity generating investment usually doesnot pose a risk to stability. However, Hungary’s deficitadjusted for reinvested earnings is extremely high evenin a regional comparison, although it does not exceedthe corresponding figure recorded in 2000.Nevertheless, the sector composition of the nationalsavings-investment balance is currently far lessfavourable than it was three years ago, and hence sta-bility is more at a risk.

The current deficit of the current account and its struc-ture will result in a rapid increase in external debt andcannot be maintained for a longer period of time. Theabove analysis reveals that in line with a cyclical recov-ery, from 2004 the corporate sector’s borrowingrequirement will continue to grow, and no change isexpected in the household sector’s position either. Thismeans that the private sector’s increasing borrowingrequirement must be offset by an improvement of thegeneral government position.

Though according to the Bank’s projection the aboverequirements will be met over the medium term, andwith the gradual decline of the GDP-proportionate cur-rent account deficit the sector structure of the externalborrowing requirement will become more healthy, therisks of a higher-than-expected external deficit are signifi-cant. The most serious concern is the higher-than-expect-ed general government deficit, as the inflexibility of theprivate sector position would result in a rise in the GDP-proportionate deficit of the current account balance.

Uncertainty is also high in respect of household netfinancial savings. Although the 2003 changes imple-mented in state subsidies to housing led to a drop in thesector’s capital formation and a rise in its financingcapacity, risks remain high. If consumption or mortgagelending (along with household investment into housing)grows in excess of the forecast rate, then the currentaccount deficit may also be higher than projected. Littlechance is seen, however, for households to becomepermanent net borrowers.

A rapid recovery of the European economy may alsogenerate further increase the GDP-proportionate cur-rent account deficit. This is because a sharp rise in exter-nal demand would urge companies to implement post-poned investment projects and make up for depletedstocks, which would, in turn, suddenly increase the bor-

rowing requirement in each sector as well as in thenational economy. This, in itself, would not entail a riskon stability, as foreign ownership is predominant in theHungarian corporate sector, and investment projectsfinance the development of market resource allocationas well as productivity generating capacities. But as thecurrent account deficit, closely monitored by investorsin terms of vulnerability indicators, is one of the mostessential factors, a deficit exceeding a certain level mayincrease the exchange rate regime’s vulnerability andlead to a deterioration in the country’s risk assessmentirrespective of its reasons.

By international standards, the national economy’s GDP-proportionate external debt cannot be considered con-spicuously high, even if the cumulative FDI is included inthe total external debt. Household and corporate indebt-edness is insignificant and their exposure to FX risks ismoderate. Barely one-fourth of the government deficit isdenominated in foreign currency, with continuouslylengthening terms, and this reduces the relevant risks ofrenewal. For this reason the risk of insolvency and theevolution of a debt crisis is moderate. The Hungarianbank sector’s direct and indirect exposure to FX risks isalso small. The loans extended to the private sector andthe loan structure would prevent bank crises even in thecase of a major weakening of exchange rates. An analy-sis of the risk factors reveals that an extensive and deep-ening FX crisis is moderately likely to evolve.

Macroeconomic indicators, also reflected in the currentstructure of capital flows, indicate that exchange ratevulnerability has grown and is manifested the increasingvolatility of the prices of HUF instruments. More than50% of the HUF-denominated government bondsfinancing the government deficit were purchased bynon-residents trusting Hungary’s entry in the EMU, andthose bonds which are uncovered by swaps serve thepurposes of financing the external borrowing require-ment of the national economy and the foreignexchange market balance. If the general governmentdeficit decreases more slowly than announced and thanhas been incorporated into prices by the markets, con-fidence in convergence may be undermined and pur-chase of HUF-denominated government securities maystop. Experiences in 2003 indicate that a loss of confi-dence may lead to a considerable weakening of theforint and drive up yields.

Depreciation forced by market powers or initiated byeconomic policy makers would render implementationof the most significant economic policy objectivesimpossible. In Hungary, exchange rate pass through isfast, so a weakening of the forint is expressed in pricesvery fast, thus nominal depreciation could improvecompetitiveness only temporarily, as the increasinginflation difference would soon use up the initial depre-ciation of real exchange rates. Although the inflation



1

generated by a weakening forint would improve thegovernment balance, sticky inflation expectations andthe declining credibility of the economic policy regimewould significantly raise the costs of future disinflation.For these reasons, exchange-rate-based stabilisationwould impede the timely performance of the inflationtarget and the country’s entry to EMU in 2008.

Foreign exchange reserves

Emerging and EU Acceding Countries may not rule outthe possibility that the exchange rate of their nationalcurrencies may easily suffer from depreciation in theevent of an epidemic crisis. This would be a case whenthe central bank would be forced to protect theexchange rate regime even at the expense of foreigncurrency reserves. If the foreign currency reserves arebelieved to be insufficient in a crisis situations byinvestors, this alone could easily undermine the credibil-ity of the exchange rate regime. In such cases, the inter-est premium may increase even if it is fundamentallyunfounded, and in extremely severe cases the nationalcurrency may be devalued.

The Monetary Council establishes the optimum level ofinternational reserves several years in advance, with aview to the specific features of the exchange rateregime as well as the monetary policy objectives. Whenthe exchange rate regime was modified in 2001, as aresult of practically continuous, strong intervention atthe edge of the band, reserves significantly exceededthe optimum level, and thus it became possible toreduce the stock of international reserves. As the gov-ernment renewed most of its foreign exchange loansthat matured in 2001 and 2002 in HUF, reservesdecreased and the actual amount of reserves becameconsistent with the amount considered as optimum.

In order to assess the vulnerability of a given currency,investors examine numerous indicators. These usuallyinclude more reliable variables of crisis-forecast modelssuch as volatility of reserves, or certain key economicindicators. Other indicators analyse whether the foreignexchange reserves would suffice to protect theexchange rate in the event of a particular financialshock by a simple crisis simulation. Such indicatorsinclude the reserves/monetary base, reserves/M3, theforeign active debts of the national economy expiringwithin a year or the reserves/monthly import indicator.Because of the significance of self-fulfilling expectations,when the Council elaborates the appropriate reservepolicies, it always takes into consideration the changesin the aforementioned investor-preferred indicators inaddition to the indicators adapted to the peculiarities ofthe Hungarian economy.

In the wake of the January speculative attack and thesubsequent intervention of the central bank, the volatil-ity of foreign exchange reserves increased considerablyin 2003. At the beginning of the year the central bankpurchased more than EUR 5 billion in the foreignexchange market in order to protect the exchange rateregime. This was followed by an intervention in theopposite direction, because foreign exchange demandby non-residents closing their positions would have ledto an unjustified weakening of the forint.

The intervention altered the amount of optimal curren-cy reserves, required to maintain the credibility of theexchange rate regime, and by a similar amount itchanged the real level of the reserves. By selling forintsin January the central bank increased foreigners’ short-term and easily liquidated assets (“hot money”), there-fore the amount of the required reserves increasedequally. Later, when the central bank sold foreign cur-rency, the amount of hot money circulating in the finan-cial system also decreased by the closing of speculativepositions of non-residents, resulting a parallel decreasein the optimal and the real amount of reserves.

The reserves/monetary base indicator specifies the cov-erage rate of the central bank money by the foreignexchange reserves. More tangible information can beobtained by expressing the indicator in terms of figureswhen the sterilised stock is filtered out of the monetarybase (see Chart 1-27). If, for example, the internationalreserves exceed the volume of the monetary baseadjusted for the sterilised stock, theoretically the centralbank has the option to introduce the institution of a“currency board” or convert to the singular use of abenchmark currency (dollarisation). Adjustment is nec-essary because if the central bank’s main instrument isa deposit-type instrument, then during conversion thereserves must also cover the sterilised stock depositedat the central bank. If, however, the main instrument ofmonetary policy is loans extended to commercial

International reserves

Chart 1-26

International reserves

6

7

8

9

10

11

12

13

14

6

7

8

9

10

11

12

13

14

15 15

Jan. 99

Apr.

99

July

99

Oct

. 99

Jan. 00

Apr.

00

July

00

Oct

. 00

Jan. 01

Apr.

01

July

01

Oct

. 01

Jan. 02

Apr.

02

July

02

Oct

. 02

Jan. 03

Apr.

03

July

03

EUR billionsEUR billions

Source: MNB.



1

banks, then the rate of required reserves is lower thanthe monetary base by the amount of the credit receivedby commercial banks from the central bank.

The reserves/M2 or reserves/M3 indicators are alsooften used in different analyses to compare the levels ofnational reserves in particular countries. At the sametime, this figure may vary greatly from country to coun-try: its value is usually very low in more developedfinancial systems, whereas it is typically high in lessdeveloped countries, where the credibility of theexchange rate regime is weaker.

Since the Asian crisis, the indicator comparing the shortterm external debt of the national economy to the stockof foreign exchange reserves (Guidotti rule) hasenjoyed increasing popularity. The indicator presumes acrisis situation where, due to an extensive liquidityshock, none of the economic participants are in a posi-tion to assume new borrowing, and gives us informa-tion on how long the FX reserves would be able tofinance the repayments of external debt. Although inHungary it is difficult to imagine that this scenario couldever come to pass because of the typically high foreign

interest in the private sector, nonetheless, the centralbank continues to monitor the changes in the indicatorbecause of its inherent importance. The value of thequotient has been steadily declining over the last threeyears because of the conscious reduction of the reservelevels. Nevertheless, it is still in excess of the short termexternal debt of the national economy (see Chart 1-28).

One of the traditionally monitored indicators is the“import rule”. This indicator provides information onhow many times the monthly foreign exchange require-ment to cover one month of a country’s commodityand service imports goes into the FX reserves. Althoughwith the development of credit and money markets, thefigure has completely lost relevance in economics, it stillappears in many investment bank analyses. In Hungary,the indicator has been steadily declining partly becauseof the reduction of superfluous reserves, and partlybecause of the increasing openness of the country andthe dynamic growth of re-exported manufacturingimports.

International reserves as a percentage of whole-economy short-term debt and one-monthimports

Chart 1-28

0

1

2

3

4

5

0

1

2

3

4

5

66

International reserves/One month of imports

International reserves/Liabilities up to one year*

Jan.

99

Apr

. 99

July

99

Oct

. 99

Jan.

00

Apr

. 00

July

00

Oct

. 00

Jan.

01

Apr

. 01

July

01

Oct

. 01

Jan.

02

Apr

. 02

July

02

Oct

. 02

Jan.

03

Apr

. 03

July

03

* Based on balance of payments statistics and according to originalmaturity.

Source: MNB.

International reserves compared to various monetary aggregates

Chart 1-27

0.5

1.0

1.5

2.0

2.5

3.0

0.1

0.2

0.3

0.4

0.5

0.6

International reserves / M0

International reserves / M0corrected with the stock of sterilization instruments

International reserves / M3 (right axis)

Jan.

99

Apr

. 99

July

99

Oct

. 99

Jan.

00

Apr

. 00

July

00

Oct

. 00

Jan.

01

Apr

. 01

July

01

Oct

. 01

Jan.

02

Apr

. 02

July

02

Oct

. 02

Jan.

03

Apr

. 03

July

03

Source: MNB.

2 STABILITY OF THE BANKING SYSTEM



2

INTRODUCTION

The first six months of 2003 in Hungary were charac-terised by deepening banking intermediation, as growthin the banking sector’s balance sheet total and totallending accelerated and substantially exceeded theGDP growth rate. Over the first six months, the balancesheet total rose 12.7% (7.2% in real terms) compared toend-2002, and by 25.6% (15.1% in real terms) on thesame period last year (Chart 2-1).

In the first half of 2003, total domestic borrowing byhouseholds and non-financial corporations rose by 18%(12.2% in real terms), and by 24.7% over a year earlier(representing a 14.3% increase in real terms). Thisupsurge was borne largely by the growing demand forloans by households, which was boosted further in Q2by an upsurge in lending brought forward as a result ofthe stricter conditions imposed on the housing loan sub-sidy scheme in June 2003. At the same time, increasedcorporate spending, often perceived as a sign of eco-nomic recovery, triggered stronger demand for loans by

non-financial corporations, which mainly fed into therise in foreign currency loans extended for real estatedevelopment projects, manufacturing and commercialactivities. Some two-thirds of the growth in total lendingto non-financial corporations stemmed from the trans-action effect, with the remaining one-third stemmingfrom revaluation following the weakening of the forintexchange rate.

Similarly, there was also a sharp increase in bank loansextended to non-banking financial firms. The first sixmonths of 2003 saw a dramatic upswing of 35.9%,resulting in an annual growth rate of 70.3%. In addition,as a general tendency, banks continued to finance pri-vate individuals and non-financial corporations indirect-ly through leasing services and loans for motor vehiclesoffered primarily by financial enterprises within bankinggroups.

Vigorous growth in private sector lending was financedby foreign funds, the issue of bank securities, mostlymortgage bonds, and a rise in the deposits of institution-al investors as well as own earnings. Within liabilities, adrop in households’ propensity to save resulted in alower ratio of household deposits; while at the sametime, increased corporate spending weakened the roleof non-financial corporate deposits.

It should be noted that the speculative attack in Januarybriefly provided domestic banks with a substantialexcess of short-term foreign funds. However, the vastmajority of this excess liquidity was not allocated tofinance growth in foreign currency loans (it was insteadused in swap transactions), since, despite the sharpdecline in short-term foreign funds, short-term and long-term loans for non-financial corporations continued onan upward trend in 2003 H1, after the Bank managed toward off speculation regarding appreciation of the forintexchange rate. Banks’ additional need for foreign financ-ing is clearly indicated by the fact that in May 2003short-term foreign funds began to rise again, accompa-nied with a steady rise in long-term foreign funds overthe course of the first six months (see Chart 2-2).

Although growth in risk-weighed balance sheet items(14.7%) was still higher than that of the balance sheet

Real growth* in balance sheet total and lending

Chart 2-1

Balance sheet total

Balance sheet total (seasonally adjusted data)

Non-financial corporations and household loans

Non-financial corporations and household loans

(seasonally adjusted data)

Per cent

28

24

20

16

12

8

4

0

–4

–8

Mar

. 98

June

98

Sep

. 98

Dec

. 98

Mar

. 99

June

99

Sep

. 99

Dec

. 99

Mar

. 00

June

00

Sep

. 00

Dec

. 00

Mar

. 01

June

01

Sep

. 01

Dec

.01

Mar

. 02

June

02

Sep

. 02

Dec

. 02

Mar

. 03

June

03

* The GDP deflator was used for calculating real growth.



2

total (12.7%) in 2003 H1, the narrowing gap betweenthese rates may be assessed as a favourable trend driv-en by the fact that, in contrast to previous years, highergrowth in lending was financed by banks’ more pro-nounced reliance on additional foreign liabilities, due tothe diminishing possibilities for restructuring on theasset side.

Analysing the market share of the five leading banksand the Herfindhal-Hirschman Index (HHI), the Bankconcludes that concentration in the banking systemcontinued to decline in 2003 H1. Yet this picture maybe slightly misleading: if one regards mortgage banksas part of the corresponding parent bank, it becomesclear that market concentration is in effect stagnating(see Chart 2-3).

Changes in foreign currency loans of non-financial corporations and foreign liabilities

Chart 2-2

6

5

4

3

2

1

0

EUR billions

Dec

. 0

1

Jan

. 0

2

Feb

. 0

2

Mar

. 0

2

Ap

r. 0

2

May

02

Jun

e 0

2

July

02

Au

g. 0

2

Sep

. 0

2

Oct

. 0

2

No

v. 0

2

Dec

. 0

2

Jan

. 0

3

Feb

. 0

3

Mar

. 0

3

Ap

r. 0

3

May

03

Jun

e 0

3

July

03

Au

g. 0

3

Short term foreign currency loans of non-financial corporations

Long term foreign currency loans of non-financial corporations

Short term foreign liabilities

Long term foreign liabilities

Balance sheet concentration in the banking system [Herfindhal-Hirschman Index (HHI)]

Chart 2-3

1200

1150

1100

1050

1000

950

900

850

800

Mar

. 97

June

97

Sep

. 97

Dec

. 97

Mar

. 98

June

98

Sep

. 98

Dec

. 98

Mar

. 99

June

99

Sep

. 99

Dec

. 99

Mar

. 00

June

00

Sep

. 00

Dec

. 00

Mar

. 01

June

01

Sep

. 01

Dec

.01

Mar

. 02

June

02

Sep

. 02

Dec

. 02

Mar

. 03

June

03

Individual banks separately

Investigating the mortgage banks

together with the parent banks



2

INCOME POSITION

Last year, non-financial corporations were net lenders.By contrast, in the first half of 2003, which was charac-terised by a brighter economic outlook, they once againbecame net borrowers. As growth in the sector’s out-standing debts was stronger than that in financial assets,non-financial corporations’ net financing needincreased to 4% of GDP.

It is highly likely that foreign currency loans, which wereresponsible for the upsurge in outstanding debts, wereused to finance inventory-building and fixed capital for-mation. Hence, there was a renewed rise in the sector’sinvestment spending, which by the end of 2003 H1increased to nearly 14% of GDP.

Based on the financial and capital formation positions ofnon-financial corporations, the Bank has concluded that

the sector’s income position in the period remainedunchanged (see Chart 2-4). Nevertheless, it is importantto point out that data excluding the income transfereffect of the government’s lending in December 2002to state-owned corporations9 show a moderateimprovement in non-financial corporations’ incomeposition in the first half of this year.

Regardless of the government’s tight fiscal policy, thegradual increase in external demand and higher com-petitiveness will have the overall effect of boostingdomestic companies’ output in 2003–2004. Growth incorporate sales, a less vigorous rise in unit labour costsand higher inflation may all contribute to improve non-financial corporations’ income position. The deprecia-tion of the forint, the higher value of foreign currencyloans and higher interest rates may, on the other hand,reduce profits. Nevertheless, this will be hardly notice-able due to the high ratio of companies with naturalcover in the form of export revenues and the increasingnumber of firms resorting to hedging instruments aswell as the fact that the overwhelming majority of for-eign currency loans have long-term maturity. Based onthese observations, the Bank’s view is that10 the incomeposition of the corporate sector is likely to take a grad-ual growth path over the next few years.

FINANCIAL ASSETS

In 2003 H1, the ratio of financial assets within totalassets remained level at 45%. However, growingdemand by non-financial corporations for financingaltered the structure of financial assets in the sector.Having examined non-financial corporations’ financialstatements, the Bank has concluded that weak growthin deposits was accompanied by a major upswing inlending. Growth in the stock of loans was driven by thesharp upward trend in long-term lending. As a result, theratio of liquid assets (cash, deposits, short-term securi-ties and loans) within total assets declined slightly from48% to 45%. However, due to the effects of the busi-ness cycle, this trend is not regarded as an unwelcomedevelopment.

Financial position of non-financial corporationsas a percentage of GDP

Chart 2-4

20

16

12

8

4

0

5

1

–3

–7

–11

–15

Per cent Per cent

Mar

. 98

June

98

Sep

. 98

Dec

. 98

Mar

. 99

June

99

Sep

. 99

Dec

. 99

Mar

. 00

June

00

Sep

. 00

Dec

. 00

Mar

. 01

June

01

Sep

. 01

Dec

. 01

Mar

. 02

June

02

Sep

. 02

Dec

. 02

Mar

. 03

June

03

Disposible income (left scale)

Gross capital formation (left scale)

Net financial need (right scale)

2.1 RISKS ASSOCIATED WITH NON-FINANCIAL CORPORATIONS

9 In December 2002 the central budget assumed a total of HUF 252 billion of debt accumulated by Hungarian Motorway Ltd. (Magyar Autópálya Rt.),Hungarian State Railway (MÁV) and Budapest Transport Company (BKV).

10 For more details, consult the study in the Inflation Report published in November.



2

COMMERCIAL PROPERTY MARKET

In respect of the commercial property market, 2003 H1saw a general cyclical contraction. In the period underreview growth continued to slow in terms of the size ofthe office and warehouse-logistics markets, as well asthe number of retail outlets.

Over the first six months of this year the office rentalmarket was characterised by an increasingly rapidadjustment of supply to demand. During this period theoffice market in Budapest expanded only to a minordegree, while the number of new leases stayed at lastyear’s level. As a result, vacancy rates were down from22% last year to 19.4%. However, despite lower vacan-cies, rental fees declined from EUR 14–16 last year toEUR 13–15, while purchase prices remained at thesame level.11

Real estate agents estimate that by the end of the yearthe office market may reach 1.35 million square metres(sqm). Forecasts for 2003 H2 suggest that 70,000-80,000 sqm of new floor space will be built, and 60,000sqm of floor space will be let, leaving vacancy rates ataround 20% and stabilising rental fees at EUR 13–15per square meter (see Chart 2-5).

According to market participants, Hungary’s upcomingaccession to the EU will not have a major impact on thedomestic office rental market.12 In contrast to previousopinions, most experts now believe that accession willnot trigger another boom on the office property market.According to estimates, the current stock of office

rentals will be sufficient to satisfy the continuous rise indemand expected over the coming years.

Over the longer run, however, four main developmentscan be expected.13 From the investors’ point of view,one negative trend is that returns on office buildings areexpected to continue to drop following accession. Withthe decline in risks in the investment environment andthe cyclical slowdown on the office space market,returns have now fallen to 8.5%, down from the typicallevels of over 10% seen in the 1998–2000 period.Following accession, returns can be expected to con-verge to the EU average of 6–7%. On the other hand,one positive development is that domestic rental feesfor office space are expected to increase, also gradual-ly converging with the higher levels seen in other EUcountries. Another positive trend which may occur isthat large foreign real estate investment funds areexpected to appear on the market, resulting in addition-al demand for high-value, marketable office buildings.Finally, it is also possible that the further developmentand strengthening of the domestic corporate sector willalso generate additional demand for more, high-qualityoffice rental space.

All things considered, the Bank feels that the financingof the office rental market bears significant risks forbanks, despite the currently decline of over-supply onthe market, as the prevailing vacancy rate (roughly20%) is relatively high compared to past years, andrental fees and property collateral are relatively low.

Due to the cyclical downturn and the on-going processof market consolidation (reduction of regional oversup-ply and satisfaction of regional demand), the number ofretail outlets fell by 0.3% over the first six months of theyear. The rise in the number of outlets observed in thesecond quarter (mainly food-related, consumerdurables, apparel and filling stations) was insufficient tocounter the decline recorded in the first quarter. Whilethe reduction in the number of certain kinds of outletsmay indicate saturation, it remains difficult to assess thelevel of risks due to the structural changes taking placeon the retail property market.

INDEBTEDNESS

The capital gearing of non-financial corporations hasdeveloped in line with the improvement in the econom-ic outlook over the past six months. Due to increasinginvestment spending by the corporate sector, the debt-to-equity ratio rose from 80% to 85%. The strong rise inforeign currency loans from Hungary and abroad wasthe primary factor behind this increase in indebtedness.

Developments in the Budapest office market

Chart 2-5

Size of the first-class office

Vacancy rate (right scale)

Rent of the first-class office

Squarmetre millions Per cent

1998 1999 2000 2001 2002 2003*

1.5

1.2

0.9

0.6

0.3

0.0

25

20

15

10

5

0

Source: DTZ Hungary, Budapest Real Estate Consultants’ ConsultationForum.

11 Source: Calm Office Market in Budapest. Ingatlanbefektetés, vol. 16, 7 August 2003.12 Source: Cyclical developments on the Real Estate Market. 2003 Q2, Economic Research Institute (GKI) and Wallis Ingatlan Rt.13 Source: ‘The Property Market following Accession’, Ingatlanbefektetés, Issue 11, 15 May 2003.



2

In forint terms, total outstanding loans from credit institu-tions and inter-company loans grew by 18.7% in 2003 H1,representing an 8.9% increase after adjustment for theexchange rate effect. Sub-average growth was observed inboth short-term and long-term forint loans, while growth inlong-term domestic and foreign loans and inter-companyloans from abroad was average. The increase in short-termforeign currency loans from Hungary and abroad showedabove-average growth. This trend can be explained by thefact that, in line with the improving economic outlook,companies which have natural cover in the form of exportrevenues (mainly multinational companies) are likelyfinancing their purchases for inventory using short-termloans, and turning to long-term foreign currency loans tofinance their growing investments.

Viewed from the perspective of the depth of bank inter-mediation, however, it is a disappointing developmentthat large companies are continuing to meet their for-eign currency financing needs through non-residentbanks for the most part. In 2003 H1, for example,Hungarian companies borrowed more than two timesas much from abroad as they did from domesticsources (see Chart 2-6).

In terms of assessing bank risk, the increase in the cap-ital leverage of Hungarian firms is not considered torepresent a serious risk, as the rise in indebtedness is

Non-financial corporations’ bank and inter-company loans as a percentage of GDP

Chart 2-6

70

60

50

40

30

20

10

0

Per cent

Mar

. 96

June

96

Sep

. 96

Dec

. 96

Mar

. 97

June

97

Sep

. 97

Dec

. 97

Mar

. 98

June

98

Sep

. 98

Dec

. 98

Mar

. 99

June

99

Sep

. 99

Dec

. 99

Mar

. 00

June

00

Sep

. 00

Dec

. 00

Mar

. 01

June

01

Sep

. 01

Dec

. 01

Mar

. 02

June

02

Sep

. 02

Dec

. 02

Mar

. 03

June

03

Forint credit

Foreign currency denominated cross-border loans

Foreign currency denominated credit from resident banks

Foreign intercompany loans

clearly linked to cyclical factors and the level of capi-tal leverage is still substantially lower than the EU aver-age.



2

In the first half of 2003, total lending by domestic banksto non-financial corporations grew moderately strongerthan the balance sheet total, increasing by 13.8% com-pared to end-2002. This sharp increase in the growthrate was the result of an upswing in corporate demandfor credit as well as the significant depreciation of theforint. Adjusting the changes in total lending for theexchange rate effect results in a 9.3% rise in the volumeof lending to non-financial corporations. In line with theincreasingly bright economic outlook, banks are onceagain focussing on extending foreign currency loans tolarge companies. There was a significant rise in projectfinancing activity, and lending also rose to sectorswhich are cyclically sensitive.

The rise in both long-term loans (up 14.3%) and short-term loans (up 12.7%) was spectacular, due mainly toincreased lending for operating assets and investment,as well as the revaluation caused by depreciation of theforint’s exchange rate. Disregarding the impact of theexchange rate, the rates of growth for both short- andlong-terms loans were still quite high, amounting to9.7% and 8.9%, respectively.

In terms of the denominational composition of corpo-rate loans there were major changes, both as a result ofthe transaction and the exchange rate effect. Corporateforint-denominated loans grew 5.4%, while foreign cur-rency loans increased by 28.1%. Consequently, theshare of foreign currency loans in total lending contin-ued to rise. At the end of the first half of 2003, HUF-denominated corporate loans accounted for 60% oftotal corporate lending, with foreign currency loans cov-ering the remaining 40%. Of the growth in foreign cur-rency loans, 16% was the result of the transactioneffect, 13.9% came from the exchange rate effect and–1.8% from the impact of cross exchange rates. As thegrowth rate of short-term loans was considerably higherthan that of long-term loans (43.8% vs. 21.5%), theaverage maturity of foreign currency loans declined (seeChart 2-7).

In the first half of 2003, demand for credit by large com-panies, which is extremely sensitive to the businesscycle, began to pick up once again, in line with theupturn in economic prospects among Hungary’s trading

2.2 DOMESTIC CORPORATE CREDIT RISK

Distribution of non-financial corporate lending bycompany size

Chart 2-8

Loans granted to the micro sized enterprises

Loans granted to the small sized enterprises

Loans granted to the middle sized enterprises

Loans granted to the large sized enterprises

Loans granted to SMEs/Total corporate loans (right scale)

Loans granted to LEs/Total corporate loans (right scale)

4000

3500

3000

2500

2000

1500

1000

500

0

HUF billions

80

70

60

50

40

30

20

10

0

Per cent

Dec

. 9

9

Jun

e 0

0

Dec

. 0

0

Jun

e 0

1

Dec

. 0

1

Jun

e 0

2

Dec

. 0

2

Jun

e 0

2

Components of changes in lending to non-financial corporations

Chart 2-7

HUF billions

Effect of cross exchange rate movements

Effect of appreciation/depreciation of HUF

Transaction effect

250

200

150

100

50

0

–50

–100

–150

2000 H

1

2000 H

2

2001 H

1

2001 H

2

2002 H

1

2002 H

2

2003 H

1



2

partners. Despite the fierce competition and low inter-est margins, large corporate lending is still one of banks’main areas of activity, as a strong banking relationshipwith large corporate customers can offer long-termincreases in profitability (commissions, account man-agement fees, etc.). Over the first six months of theyear, total lending to large companies increased by19%, while lending to SMEs grew by 7.6%. Lending tomicro-enterprises was up 4.1% in this period, with lend-ing to small enterprises and medium-sized enterprisesup by 7.8% and 10.3%, respectively. Consequently, theshare of large companies rose from 54% to 56.5% with-in total corporate lending, while the share of SMEsdropped from 46% to 43.5% (see Chart 2-8).

Due to cyclical reasons, the share of SME loans in totalcorporate lending temporarily dropped in the first halfof 2003, following a long period of consistent growth.Over the longer term, lending to SMEs should continueto gain ground as competition between banks increas-es and the market for large companies becomes evermore saturated.

Project financing by banks continued to grow during2003 H1, backed mainly by an increase in lending forreal estate-related projects, and to a less extent, by lend-ing for projects in the fields of energy and telecommu-nications. Despite the downturn on the commercial realestate market, over the first six months lending for officebuilding and shopping mall construction grew sharply,at a rate of 26.6%. Total lending of this nature, which ismainly denominated in foreign currency, increased by14.8% adjusted for the exchange rate effect; thus its

share in the total balance sheet and in total corporatelending now stands at 1.9% and 6%, respectively (seeChart 2-9). While this level can still be viewed as low inabsolute terms, one unfavourable development is thatthe share of loans extended since the end of theupswing on the office space market in 2000 and thebeginning of the downturn on this market has nearlydoubled within total corporate lending.

In terms of evaluating banks’ risks, the picture is evengloomier due to the fact that in contrast to their previ-ous opinion, real estate experts now no longer forecastthat Hungary’s accession to the EU next year will triggeranother boom on the real estate market. They nowbelieve that the impact of accession will be felt gradual-ly over a longer period of time. In addition to this,another source of serious risk is that 80% of the loansextended for office building and shopping mall con-struction are concentrated at three large banks. At thesebanks, the ratio of these loans to the total corporateloan portfolio is more than twice as high as the averagevalue for the banking sector in general. The risk of con-centration is also well reflected in the HHI figure forcommercial lending, which – although it has been on adownward trend since 1998 – is currently still above2400, a very high figure indeed, compared to the figureof less than HHI 1000 for the overall corporate sector.While the banks financing these developments haverecognised the increase in risks and expect their cus-tomers to ensure 30–50% pre-leasing for office buildingconstruction projects, the Bank still believes that therisks assumed by banks on this market are too great.

Analysis of the sectoral breakdown of non-financial cor-porate lending indicates that at the end of the first halfof 2003 lending to the production sector accounted for43% of total corporate lending, with services and othersectors accounting for 52% and 5%, respectively.Within the production sector, lending to manufacturingincreased, whereas lending to agriculture and miningstagnated. In respect of manufacturing branches, themost dynamic growth in lending was registered inchemicals, mechanical engineering and light industry.The growth in transactions was borne mainly by a risein short-term, foreign currency loans for operatingassets, and to a lesser extent by a rise in long-term for-eign currency loans for investment purposes.

Within the service sector, banks mainly financed realestate development and commercial activities. While itwas primarily long-term foreign currency loans whichincreased in respect of real estate development, in thecommercial sector short-term forint and foreign curren-cy loans both saw growth. In one of the main servicesectors, the construction industry, financing tapered off,whereas there was a moderate rise in lending to trans-port, logistics and telecommunications.

Lending to finance office building and shoppingmall construction as a share of total corporatelending and its concentration

Chart 2-9

Commercial property financing loans

as percentage of total corporate sector loans

Concentration of commercial property loans

(HHI) (right scale)

8

7

6

5

4

3

2

1

0

Per cent

4000

3500

3000

2500

2000

1500

1000

500

0

Dec

. 98

June

99

Dec

. 99

June

00

Dec

. 00

June

01

Dec

. 01

June

02

Dec

. 02

June

03



2

The structure of the corporate lending market continuesto be characterised by a low level of concentration (seeChart 2-10). In terms of the HHI level, the large corpo-rate lending market can be considered a competitivemarket (at 960), whereas concentration on the SMElending market is at a medium level, reflected by HHIvalue of over 1000 (currently 1210). One interestingpoint to note is that the market for corporate depositsshows the lowest level of concentration, and that thismarket has the lowest HHI (830) in the corporate bank-ing fields.

Analysis of the development of the interest rate marginon corporate loans over the recent years is renderedconsiderably more difficult by the volatility of the cen-tral bank interest rate. In 2002 H2, the interest margindropped significantly, as a result of the increases in thecentral bank interest rate in May and July 2002.Following this, however, the interest rate marginremained low for a long period through to the end of2002, at around 1 percentage point, despite theincrease in credit risks and the reduction in interestrates in November and December. A correction onlyoccurred in January 2003 when the central bank maderepeated, deep to cuts in interest rates. Nevertheless,due to the gradual adjustment of loan rates, the interestmargin subsequently began to decline and fell below 1percentage point again in July and August, in responseto the sharp rises in the central bank interest rate in June(see Chart 2-11). From these developments, it appearsthat, despite the increased credit risk due to the cyclicalslowdown, the adjustment of interest rates on corpo-rate loans was quicker to respond to rises in the interestrate, than it was to respond to reductions in the interestrate. An explanation for this may be that banks mayhave viewed the increase in the reference interest rate

(BUBOR) as a temporary rise, and that the sharp com-petition countered a rapid upward correction in lendingrates.

During 2003 H1, the average interest margin was 1.46percentage point, which was almost twice as high asduring 2002 H2, when it was 0.74 percentage point. If,however, one compares the figures for the end of thefirst half of 2003 and the end of 2002, it becomes clearthat that the interest rate margin stagnated at around 1percentage point. Indeed, in June and August of 2003,the interest rate margin fell to one-half of its level at theend of 2002 H1. Consequently, the Bank believes thatover the course of the past year the interest rate marginhas been driven mainly by asymmetric adjustment tothe interest rate moves of the central bank and is less sothe result of conscious risk awareness and managementby banks.

Over the last five years, the corporate interest rate mar-gin generally fluctuated in a range of 1–1.5 percentagepoint. With due consideration of the extent of risks andfrom an international perspective (the correspondingmargin was between 2–2.5 percentage point in 2003H1 in the euro area), this figure of 1 percentage pointcan be viewed as low. Nevertheless, it should also benoted that the period under review can be consideredextreme in terms of volatility, and that, despite the keencompetition and low interest rate margin, on the wholebanks are able to achieve suitable profitability in thefield of corporate banking through the prices chargedon corporate liabilities (current account interest rate,interest on time deposits, etc.) and by increasing non-interest revenues (standby charges for lines of credit,account fees, etc.).

Interest rate margin on forint loans to non-financial corporations

Chart 2-11

Spread between corporate lending rate

and the 3 month BUBOR rate

Smoothed time series (Hodrick-Prescott filter)

2.0

1.5

1.0

0.5

0.0

Percentage point

Jan

. 9

8

May

98

Sep

. 9

8

Jan

. 9

9

May

99

Sep

. 9

9

Jan

. 0

0

May

00

Sep

. 0

0

Jan

. 0

1

May

01

Sep

. 0

1

Jan

. 0

2

May

02

Sep

. 0

2

Jan

. 0

3

May

03

July

03

Concentration of non-financial corporations’loans and deposits (HHI)

Chart 2-10

1300

1200

1100

1000

900

800

700

600

Total corporate loan

Total corporate deposit

Loans to LEs

Loans to SMEs

Dec

. 98

June

99

Dec

. 99

June

00

Dec

. 00

June

01

Dec

. 01

June

02

Dec

. 02

June

03



2Contingent liabilities

Chart 2-12

Contingent liabilities at contract value

Contingent liabilities weighted by transaction risk

Contingent liabilities weighted

by total risk/total assets (right scale)

4000

3500

3000

2500

2000

1500

1000

500

0

16

14

12

10

8

6

4

2

0

HUF billions Per cent

June

Dec

.

June

Dec

.

J

une

Dec

.

June

Dec

.

June 9

9

99

00

00

01

01

02

02

03

CONTINGENT LIABILITIES

The contractual value of contingent liabilities increasedby 8.8% during the first half of 2003, rising some 4 per-centage points more slowly than growth in the balancesheet total. The risk rating of contingent liabilitiesimproved, which the Bank views as a positive develop-ment in terms of system stability. The transaction riskadjusted value of contingent liabilities rose by 5.5%,while the customer and transaction risk adjusted valueincreased by 8.5%. As a result, the ratio of contingentliabilities adjusted for total risk to the balance sheet totaldeclined to around 10% (see Chart 2-12).

In contrast to the past and in light of the improved glob-al economic prospects and the developments seen inthe field of Hungary’s economic policy, the Bank nowfeels that it is primarily the better economic outlook, thedepreciation of the real exchange rate and the forecastgrowth in unit labour costs (expected to be lower thaninflation) which will support further growth in demandfor credit by non-financial corporations. The supply ofcredit may also increase due to the keen competition forthe business of large companies and the increasinglystrong competition in the field of SMEs.

The Bank still does not view the risks of non-financialcompanies as being excessive. There is only one area,

namely the financing of real estate development proj-ects, where the Bank believes that the level of risk ishigh. On the other hand, one benign trend is that theimproving income outlook for the manufacturing sectormay reduce the credit risks associated with this sector.



2

INCOME POSITION, CONSUMPTION, INVESTMENT, NET SAVINGS/BORROWINGS

The annual rate of growth in household income can beexpected to decline as the effects of the income-boost-ing measures implemented last year taper off. In con-trast to the slower growth rate in households’ real netincome in 2003 and the expected stagnation forecastfor 2004, household consumption expenditure and finalconsumption have continued to increase at a robustpace, resulting in further strong growth in consumerborrowing. Due to the strong base effect and theunfavourable weather,14 household investment expendi-ture grew at a much slower rate than previously, where-as borrowing for home construction continued to growsignificantly. The combination of declining growth ratein household income and the unbroken strong rise inconsumption and housing related borrowing led to thehousehold sector to be a net borrower in 2003 Q2,which is an extremely unusual and fundamentallyunfavourable development (see Chart 2-13). Due to thechanges made to the housing loan subsidy scheme inJune, the market was expecting a decline in demand forhousing loans starting from August, which would haveput an end to the deteriorating trend in households’ netfinancing capacity (indeed, reversing the developmentof a financing requirement). Nevertheless, based on thelatest data, September growth in total housing loansexceeded all previous records. Part of this can presum-ably be ascribed to demand being brought forward inexpectation of other planned changes (such as theimpact of introducing VAT on building lots).Furthermore, as households smooth their consumptionno decline in consumer-related borrowing is expected.

INDEBTEDNESS

The majority of the unbroken, dynamic growth in house-hold borrowing (27% in six months) continues to resultfrom the strong rise in housing loans and, to a lesserdegree, a substantial rise in non-bank lending, whichamounted to 41% and 26%, respectively (see Chart 2-14).

2.3 HOUSEHOLD SECTOR

Net financing capacity/requirement of households

Chart 2-13

1998 Q

2

1998 Q

4

1999 Q

2

1999 Q

4

2000 Q

2

2000 Q

4

2001 Q

2

2001 Q

4

2002 Q

2

2002 Q

4

2003 Q

2

500

400

300

200

100

0

–100

500

400

300

200

100

0

–100

HUF billions HUF billions

Operational financial savings

Operational borrowing

Net financing capacity(+)/need(–)

14 Due to the long, hard winter construction projects were started later than usual.15 Liquid assets are classified as cash, deposits, non-equity securities with the exception of compensation vouchers, listed shares and investment fund certificates.

Lending to households as a percentage of GDP increasedfrom 7.5% to 11.3% in the space of single year. Even so,this level can still be viewed as low by international stan-dards. Nevertheless, due to the weak income earningcapacity of Hungarian households, the level of indebted-ness, measured as the ratio of financial obligations tofinancial assets (which rose from 12.6% to 18.7% in justone year), is not seen as low by international standards.The financial obligations of households now amount toroughly one-third of households’ liquid financial assets,15

and this ratio is expected to continue rising due to the fur-ther strong growth in financial obligations and the declin-ing trend in the share of liquid assets as a result of increas-ing popularity of institutional investors (see Chart 2-15).Furthermore, due to the weaker income earning capaci-ty of households in Hungary and the high interest rateson consumer loans and other borrowing, the debt serv-ice burden on households is also not particularly low byinternational comparison.



2

Based on the aforementioned, it can be seen that seri-ous risks could arise if, over the longer term, the level ofindebtedness continues to rise as quickly as it has in therecent past. Fundamentally speaking, the expansion ofthe housing loan subsidy scheme in March 2002 trig-gered this extremely sharp increase in the level ofindebtedness. One additional aspect was however thatdemand was being brought forward due to the expec-tations of restrictions being placed on the subsidy sys-tem. It does not appear that the changes made to thesubsidy system in June have slowed down the growthrate in housing-related lending considerably; conse-quently, further growth in the level of indebtedness canbe expected. A further strong rise in indebtedness inconjunction with the expected stagnation of realincome may lead to a considerable rise in the debt serv-ice burden on households. On the other hand, as theshare of long-term, subsidised loans with low interestrates for borrowers is growing within overall lending,the debt service burden will likely grow more slowlythan total lending.

One factor that increases credit risk is the fact that con-sumer lending has a track record of just a few briefyears in Hungary. Consequently, for lack of experience,many banks presumably do not have at their disposalhighly refined risk management practices, with whichcustomers’ repayment difficulties could be temporarilyresolved. Moreover, companies and organisations spe-cialising in providing credit-related counselling to house-holds have not been established (in contrast to the USA,

for example), and households’ level of financial knowl-edge is low.

In total, 97% of households’ bank loans are denominatedin forints. Accordingly, there is essentially no rise in bankdirect lending risk due to the exchange rate volatility. Asubstantial ratio of household financial obligations, how-ever, are vis-à-vis non-bank entities (amounting to 22.3%in 2003 H1). Of this, roughly one-half16 is accounted forby loans from financial enterprises which are in bankownership. The vast majority of household loans extend-ed by bank-owned financial enterprises is denominatedin foreign currency (88%17), and thus any increase inrepayments due to depreciation of the forint may resultin higher credit risk. At the end of 2002, foreign currencylending to households by bank-owned financial enterpris-es amounted to roughly 10%18 of households’ total liabil-ities; consequently, any increase in credit risk fromchanges in the exchange rate would not have a drasticimpact on the banking sector as a whole. Nevertheless,at the level of individual banks this is not necessarily thecase, as the ratio of foreign currency lending in the con-solidated household loan portfolio is extremely high atseveral banks. In respect of financial enterprises whichare not bank-owned, such companies rely to a greatdegree on equity (18%), in addition to funds from creditinstitutions (66%), i.e. their capital leverage is relativelylow. Accordingly, the credit risk assumed by the banksproviding loans for financial enterprises that are not bank-owned is not considered to be particularly great.

Households’ financial obligations as a percentage of liquid assets

Chart 2-15

Liquid assets / total financial assets

Liabilities / liquid assets

80

70

60

50

40

30

20

10

0

Per cent

Dec

. 9

8

Mar

. 9

9

Jun

e 9

9

Sep

. 9

9

Dec

. 9

9

Mar

. 0

0

Jun

e 0

0

Sep

. 0

0

Dec

. 0

0

Mar

. 0

1

Jun

e 0

1

Sep

. 0

1

Dec

. 0

1

Mar

. 0

2

Jun

e 0

2

Sep

. 0

2

Dec

. 0

2

Mar

. 0

3

Jun

e 0

3

Lending to households

Chart 2-14

Housing loans Non-bank loans

Consumer and other credit

3000

2500

2000

1500

1000

500

0

HUF billions

Dec

. 98

Mar

. 99

June

99

Sep

. 99

Dec

. 99

Mar

. 00

June

00

Sep

. 00

Dec

. 00

Mar

. 01

June

01

Sep

. 01

Dec

. 01

Mar

. 02

June

02

Sep

. 02

Dec

. 02

Mar

. 03

June

03

16 Data as of end-2002.17 Data as of end-2002.18 Data as of end-2002.



2

HOUSING LOANS

Since July 2002 the stock of outstanding housing loanshas increased by a monthly average of HUF 55 billion,as a result of the subsidies being extended to includethe purchase of used homes (see Chart 2-16).Compared to the figure of 5.8% for the previous year,by end-September 2003 the share of housing loans inbank assets had risen to 10.2%, and is thus growingcloser to the average 15% ratio found in the euro area.

As a result of changes in the subsidy system made inJune 2003, banks’ margin on newly granted housingloans has declined. The possibility of increasing theinterest rates paid by borrowers is quite restricted, dueto the interest rate ceilings prescribed as a preconditionfor the subsidies, and banks will thus likely make theircredit conditions and standards somewhat more strict.The changes in the subsidy system, however, have notled to a significant decline in demand. First, in respectof loans with supplementary interest subsidy for the pur-chase of new homes, the maximum loan amount wasincreased. Second, in respect of loans subsidised on theliability side, while the maximum loan amount wasreduced by 50%, this amount is still three times higherthan the average loan amount. Third, there are stillexceptions to the restriction on obtaining subsidisedloans for the second and third times, etc. (as a co-sign-er on subsidised loans for purchasing homes for grown-up children and grandchildren). The very low interestcharged to borrowers continues to make these sub-sidised loans very attractive. Finally, demand may bebrought forward in expectations of further changes andserious restrictions in the subsidy system, as well as the

famed price-boosting impact of the introduction of VATon building lots. Based on all of the aforementioned fac-tors, it is still felt that subsidised loans are still availablewidely even after the changes; hence, subsidised loanscontinue to crowd-out housing loans granted at marketrates. Opinions differ on the impact of the changes tothe tax benefits related to housing loans.

According to the Bank’s experience, a number of fac-tors with a bearing on borrowers creditworthiness, suchas income, sociological and demographic aspects, arehardly taken into account in the evaluation of loanapplications, which is why the ratio of the loan amountto the collateral property (Loan To Value, or LTV) isespecially important. In respect of loans subsidised onthe liabilities side, which account for the majority ofaggregate housing loans, banks extend loans up to amaximum of 60% of the credit collateral value (which is80-90% of the value stated by the property appraiser).In the Bank’s opinion, the domestic average LTV valuesin Hungary represent a safe escape value in Budapest,the county seats and in larger cities in the country. Inparts of the country, however, where there is no liquidreal estate market, the question remains as to whetherthe collateral property can be sold at a suitable price.Considering the continuing strong demand and falling

Cumulative values of newly granted subsidisedhousing loans

Chart 2-16

Loans with supplementary interest subsidy

Loans subsidised on the liabilities side

600

500

400

300

200

100

0

HUF billions

Jan.

00

Mar

. 00

May

00

July

00

Sep.

00

Nov

. 00

Jan.

01

Mar

. 01

May

01

July

01

Sep.

01

Nov

. 01

Jan.

02

Mar

. 02

May

02

July

02

Sep.

02

Nov

. 02

Jan.

03

Mar

. 03

May

03

Source: Ministry of Economic Affairs and Transportation.

It is possible that the tightening of the conditionsfor tax benefits related to housing loans willincrease the level of credit risk somewhat, butthe degree of this increase is not considered tobe too great. In the first year of repaymenthouseholds must earn the entire amount that isto be repaid despite the tax benefit, as the taxrefund is received afterwards. The repaymentcapacity of borrowers with better income earn-ing capacity will either not be affected or only beaffected to a negligible degree by the ceasing ofthe tax benefit, which results in a maximum addi-tional burden of HUF 20,000 per month. On theother hand, borrowers with less pronouncedearning capacity will likely take out smallerloans. It is also possible that the tax burden ontheir income is relatively low, and thus amountof lost income due to the reduction in the maxi-mum amount of the tax benefit plays a minor orinsignificant role in many cases. Taking the aver-age amount of loans subsidised on the liabilityside and loans with supplementary interest ratesubsidy with maturity of 20 years as a basis, theBank’s estimates indicate that at an annual levelthe 50% reduction in the maximum tax benefitwill result in income loss amounting to roughly1.2 and 2.1 months of repayment per year, witha delay of one year.

Box 2-1



2

margins, it is not likely that any significant decline in theLTV ratio will be seen in the near future.

The exceptionally strong growth in housing loans hasnot yet led to a sharp rise in property prices. One of thereasons for this is that the home construction market iskeeping up with the increase in demand (see Chart 2-17). Another reason may be the modest income earningpower and debt servicing capacity of households. Withdue consideration of the moderate rise in prices and thefact that the majority of mortgage loans are for homepurchase, the Bank does not feel that the possible risksfrom a price bubble are too serious, despite theextremely robust growth rate in lending. There are,however, differing opinions as to how much of animpact the introduction of VAT for building sites willhave on home prices.

The housing loan subsidy system is having a majorimpact on the acquisition of market share by the varioustypes of institutions. Loans subsidised on the liabilityside are experiencing the greatest demand. Extension ofsuch loans is associated with the issue of mortgagebonds, a privilege reserved for mortgage banks (see

Chart 2-18)19. As long as the current logic underlying thesubsidy system is followed, there is little chance of a sig-nificant increase in the issuance of long-term bonds bybanks, in the interests of mitigating the liquidity risksstemming from the dynamic growth of housing loans.This is because, according to this logic, loans subsidisedon the liability side are linked to mortgage bonds.

Despite the extraordinarily strong growth rate, marketconcentration on the housing loans market has stagnat-ed at a high level, as indicated by the fact that theHerfindahl index remained above 3500 in the first halfof 2003, with mortgage banks considered together with

Number of building permits issued quarterly*

Chart 2-17

Pieces

1997 Q

11997 Q

21997 Q

31997 Q

41998 Q

11998 Q

21998 Q

31998 Q

41999 Q

11999 Q

21999 Q

31999 Q

42000 Q

12000 Q

22000 Q

32000 Q

42001 Q

12001 Q

22001 Q

32001 Q

42002 Q

12002 Q

22002 Q

32002 Q

42003 Q

12003 Q

2

16,000

14,000

12,000

10,000

8,000

6,000

4,000

2,000

0

Budapest Other towns Villages Total

* Seasonally adjusted figures.Source: CSO.

Market share of the individual institution-types inhousing finance

Chart 2-18

Commercial banks Contract saving institutions

Mortgage banks Saving co-operatives

1400

1200

1000

800

600

400

200

0

HUF billions

1997 1998 1999 2000 2001 2002 Aug. 2003

In the view of the Bank, introduction of VAT onbuilding sites will not lead to a significant rise inhome prices over the short term. In terms ofdirect impact, only newly-built homes will beaffected. As most building sites are owned bynatural persons and natural persons do not fallunder the scope of VAT, in most cases VAT onthe building sites will not play a role in the con-struction of a family house. In respect of newcondominiums with just a few units, introductionof VAT may significantly increase home prices inareas where site prices are relatively high. Inthese cases the price of the building lot itselfaccounts for a fairly high proportion of the totalpurchase price. On the other hand, in parts ofthe country where the price of the lot itself is rel-atively low compared to the price of construc-tion, and in respect of condominiums withnumerous units (regardless of where they arelocated in the country), introduction of VAT forlots will have a less pronounced effect on homeprices, as in these cases the lot price accountsfor a relatively smaller share of the total price ofthe unit. As for the indirect effects, the extent towhich price increases will be passed on to usedhomes and lots owned by private individuals isquestionable and will be felt with a delay.

Box 2-2

19 For details on the institutional structure, problems and risks of housing loans, see Report on Financial Stability, 2002/2.



2

parent banks. The shares of the 3 largest and the 5largest participants in the housing loans market essen-tially remained unchanged at 75% and 83%, respective-ly, in 2003 H1.

CONSUMER CREDIT AND OTHER LOANS

In the first half of 2003, outstanding consumer creditand other bank loans increased by 9.1%, with themajority of growth seen in the second quarter. Despitethe increase in interest rates on loans, considerablegrowth was seen on the market of consumer and otherbank loans in the third quarter as well, along with strongdemand for housing loans and concurrently stronggrowth in non-bank loans. If new home-owners pur-chase their new home furnishings in part also on credit,this could lead to a concentration of loans, an increasein the indebtedness of households which have takenout loans and thus consequently result in an increase inbanks’ credit risks. Nevertheless, for lack of a creditbureau with a positive list of debtors, this risk cannot bequantified.

Presumably, lower-income households without signifi-cant savings tend to use small-amount loans (e.g. loansfor consumer goods), since it would be rather unreason-able for households with savings to finance their pur-chases with loans, due to the large difference betweenthe high level of current interest rates on loans and therelatively low yields on savings. Small-amount loans aregenerally unsecured, thus, the risk associated with suchloans is higher. An increase in loan-loss ratios can beexpected in light of the deterioration in householdincome expectations.

The further rise in the already high average interest rateson loans after the central bank’s interest rate increase inJune can be explained in part by the fact that concentra-tion on certain segments of the consumer and other bankloans market is rather high, with three or four banks dom-inating the market for certain loan types (see Chart 2-19).

The growth rate is expected to remain robust in the fieldof consumer and other bank loans. As a result of thechanges to the subsidy system, the interest margin onnewly granted housing loans has declined significantly,and thus banks are expected to turn again to the mar-ket of consumer and other loans. In light of the increaseseen in average interest rates on newly granted con-

sumer and other loans, it could be presumed that bankswish to maintain or even increase in the growth in lend-ing by loosening the credit standards and conditions forgranting loans.

The combination of declining growth rate in householdincome and the continued strong rise in consumptionand housing related borrowing led to the household sec-tor being a net borrower at the end of the period underreview, which is an extremely unusual and fundamental-ly unfavourable development. The sharp increase in thelevel of indebtedness was essentially caused by theexpansion of the housing subsidy system, and bydemand brought forward due to persistent expectationsof restrictions in subsidies. It does not appear that thechanges made to the subsidy system in June have sloweddown considerably the growth rate in housing loans;consequently, further significant growth in the level ofindebtedness can be expected, which the Bank views asbearing serious risk.

Interest rates on banks’ non-housing loans*

Chart 2-19

30

25

20

15

10

5

0

Per cent

May

01

July

01

Sep

. 0

1

No

v. 0

1

Jan

. 0

2

Mar

. 0

2

May

02

July

02

Sep

. 0

2

No

v. 0

2

Jan

. 0

3

Mar

. 0

3

May

03

July

03

Nominal interest rate Real interest rate

Euro-zone real interest rate

3 moths treasury bill interest rate spread

Total cost of loans

* Excluding overdraft credits and other loans; hence the data differfrom those published in the previous issues of the Report. Throughend-2002 it includes personal, consumer goods and automobile loans,from 2003 loans classified as consumer loans by the data provider areincluded in this category.



2

In the first half of 2003, the ratio of classified portfolio con-tinued to improve, both in terms of the total portfolio andclassified balance sheet items (see Chart 2-20). The consid-erably more dynamic improvement in the indicator for thetotal portfolio was mainly caused by the boom in deriva-tive transactions, which are rated as 100% problem-free.

In respect of on-balance-sheet items, the ratio of non-per-forming loans declined to 3.3%. The main reason for thisimprovement was the strong increase in lending. Classifiedreceivables increased by 17.7%, while non-performingclaims only rose by 5%. Compared to these items, thesmall stock of written-off and sold claims does not result inany changes in the developments described above.

In line with the strong pick-up in lending, net provisioningfor own claims (Table 2-1), an item reducing profit, alsorose strongly, to almost two times the level recorded forJune 2002. The rate of stock of provision compared to

the gross value of classified assets categories and the totaloutstanding lending declined over the last 12 months(from 17.3% to 15.0% and from 1.81% to 1.51%, respec-tively); the loan loss rate declined primarily in respect ofhousehold lending and investments, while the rateincrease marginally in respect of corporate lending.

QUALITY OF THE CORPORATE PORTFOLIO

Economic activity continued to slow in the last half-year,while the economic outlook improved. As a consequenceof this improvement in prospects, fixed investment spend-ing by non-financial companies increased, leading to a13.8% jump in their borrowings. By contrast, the impactof the slower economic growth was reflected in a rathera moderate rise in non-performing loans (up 3.4%). Thenew loans had a beneficial impact on portfolio quality,leading to a reversal of the previous deterioration (seeChart 2-21). If the economic situation continues toimprove and there is further growth in lending, a reneweddeterioration of portfolio quality is not expected.

Over the last year, stock of provisions as a percentageof gross value increased in all of the classified categoriesof corporate loans, which indicates prudent behaviour(see Table 2-2).

CREDIT QUALITY OF HOUSEHOLD LENDING

The share of housing loans within the total portfolio ofhousehold lending increased further, due to the excep-

Development of classified asset categories in thevarious portfolios

Chart 2-20

1996

1997

1998

1999

2000

2001

2002

2002

2003

June

June

16

15

14

13

12

11

10

9

8

7

6

5

4

3

Per cent

Classified assets as a proportion of the total portfolio

Non-performing loans as a proportion

of balance sheet items to be classified

Classified assets as a proportion

of balance sheet items to be qualified

Special watch loans as a proportion

of balance sheet items to be classified

2.4 PORTFOLIO QUALITY

Table 2-1

Net provisioning (Change in value adjustments)

Cumulated, Dec. June Dec. JuneHUF million 2001. 2002. 2002. 2003.

Net provisioning 29,809 5,265 24,319 12,392of which:Provisioning for own claims 18,042 6,443 24,945 12,243Provisioning forpurchased claims –127 199 143 363Provisioning for securitiesfor investment purpose 2,288 701 650 0Provisioning for participating interests 9,606 –2,078 –1,419 –214



2

tionally strong growth seen in this market segment. Thisshift in ratios towards a credit segment with lower riskswill result in an improvement of portfolio quality overthe longer term. Moreover, the significant growth in theamount of total lending also leads to an improvement inportfolio quality. This short-term impact is enhanced bythe fact that that non-performance problems occurmuch later with housing loans than with consumer loans.

Once again, it was primarily the rise in lending which wasbehind the decline in the share of recorded loss in value

in total gross household lending, this indicator falling byalmost one-half in the course of 12 months (from 3.3.%to 1.7%). This development strengthened a trend thathad already been observed: the share of recorded loss invalue in all of the classified categories has been consis-tently declining (see Table 2-2). This latter process canmore than likely be ascribed to the slow increase in hous-ing loans backed by mortgages (which exhibit lower lossrates) within overall classified household lending.Housing loans secured with mortgages have become thecentre of attention, due the exceptionally strong growthseen in this type of loan. It is difficult to quantify the riskassociated with this type of loan, as past experience isonly relevant to a very low level of total outstanding lend-ing and has thus lost a great deal of significance.Unfortunately, no data on the portfolio quality of housingloans is available. The available data relates to the ratio ofrecorded loss to gross value, which fell considerably overthe last year, from 1.5% to 0.5%, while a less pronounceddecline was observed in respect of non-housing loans,where the ratio dropped from 4.8% to 4.0%. In additionto the growth in total lending, the drop in the amount oftotal recorded losses also played a role in the declineseen in these ratios. In respect of housing loans, totalrecorded losses fell from HUF 6.1 billion to 5.1 billion,while for other household loans total recorded lossesdropped from HUF 23.1 billion to HUF 21.2 billion.Assuming that there were no changes in writes off andwork-out policy, on the one hand, this reflects that theincome position of households has improved consider-ably in this period, and on the other hand, that banksprobably did not provision for growing losses – mainly inhousing loans segment – in the future.

Changes in the ratio of non-performing corporateloans

Chart 2-21

1.6

1.4

1.2

1.0

0.8

0.6

Per cent

June

00

Dec

. 00

June

01

Dec

. 02

June

02

Dec

. 02

June

03

Substandard Doubtful Bad

Growth rate of loans and changes in the ratios ofspecial watch and non-performing loans

Chart 2-22

Growth rate of loans (6-month growth rate) – RHS

Special watch (LHS)

Non performing (LHS)

10.0

8.8

7.5

6.3

5.0

3.8

2.5

Per cent Per cent

40

35

30

25

20

15

10

0

June

00

Dec

. 00

June

01

Dec

. 01

June

02

Dec

. 02

June

03

Recorded losses in value in the various classifiedasset categories as a proportion of the grossvalue of balance sheet items

Recorded losses Special Sub- Doubt- Badin value as a percentage of watch stan- fulgross value – loans to dardnon-financial firms

June 2001 2.4% 18.2% 50.3% 90.5%December 2001 2.0% 23.6% 47.9% 88.9%June 2002 2.0% 17.6% 44.9% 87.7%December 2002 1.8% 19.3% 45.1% 88.9%June 2003 2.0% 20.2% 46.4% 89.1%

Recorded losses in valueas a percentage of gross value – loans to households

June 2001 6.8% 20.7% 47.6% 96.3%December 2001 2.0% 17.5% 43.6% 94.3%June 2002 1.8% 16.2% 41.8% 90.9%December 2002 1.6% 16.7% 41.9% 87.5%June 2003 1.5% 14.5% 41.2% 82.2%

Table 2-2



2

The quality of the portfolio has improved as a result of thestrong upswing in lending. In the corporate segment, if theeconomic situation continues to improve and there is fur-ther growth in lending, a renewed deterioration of portfo-lio quality is not expected. In respect of the household sec-

tor, in the event that growth in lending levels off in thefuture and the exceptionally large new stock of loans pass-es into a more mature stage of the credit cycle, the shareof non-performing loans can be expected to rise, which willbe accompanied by an increase in net recorded losses.



2

BANKS’ ACTIVITIES IN THE DERIVATIVES MARKET

The robust increase in Hungarian banks’ activities in thederivatives market continued in 2003 H1. The outstand-ing total of the banking system’s derivative contractsrose by 53% in the period under review, exceedingHUF 12,000 billion by end-June. Foreign exchangederivatives played a key role in this brisk increase.

After stagnating during the first five months of the yearand soaring in June, Hungarian banks’ open foreignexchange derivatives contracts rose by 67% in nominalterms, exceeding HUF 9,000 billion towards the end ofthe period. Swaps with non-residents and forwardsmainly with firms continued to account for the bulk ofthe increase in foreign exchange derivatives (see Chart2-23). Swaps and forwards, respectively, accounted for61% and 26% of the outstanding total at the end of H1.There was also a vigorous increase in the amount ofoption contracts, although only a few players have beenactive in the market. At the end of June, almost 90% ofthe total outstanding option contracts was concentrat-ed at three banks.20

The outstanding total of interest rate derivatives rose ata much more modest rate – the stock of open contractsexceeded HUF 3,000 billion by the end of H1. Whereasin 2002 the stock of FRAs and interest rate swaps bothincreased robustly, the increase in 2003 H1 wasaccounted for by swap transactions. At end-June, theshare of interest rate swaps and forwards within interestrate derivatives was 53% and 45%, respectively. Itshould be noted that the degree to which interest rateswaps are concentrated is extremely high in the bank-ing sector, as one bank accounts for nearly 80% of thetotal.

EXCHANGE RATE EXPOSURE

Looking at the denominational structure of the sector’sbalance sheet, foreign exchange items stopped declin-ing further in 2003 H1 (see Chart 2-24). However, theslight increase in the ratios of foreign exchange assetsand liabilities was ascribable mainly to the depreciation

of the forint. Examining the various balance sheet items,foreign currency loans extended to firms and non-bankfinancial intermediaries as well as foreign currency liabil-ities from non-resident banks rose sharply even aftereliminating the exchange rate effect. Except for corpo-rate sector foreign currency deposits, items denominat-ed in euro continued to increase as a proportion ofloans and deposits broken down by major currencies.The depreciation of the dollar slightly contributed to thisdevelopment. The denominational structure of corpo-rate foreign currency loans experience the largest shift,as a result of which the percentage share of euro-denominated loans rose to 82%.

The speculative attack in January 2003 triggered a shiftin the sector’s on-balance sheet and forward position,the scale of which has not been experienced before,while the total open position remained unchanged at alow level (see Chart 2-25). In the attack, non-residentsbuilt synthetic forint forward positions, intermediatedby the banking system, mainly by buying forints in thespot market in combination with related swap transac-

Foreign exchange derivatives of the banking system (Open contracts)

Chart 2-23

Swap Forward Others (options. futures)

6000

5500

5000

4500

4000

3500

3000

2500

2000

1500

1000

500

0

HUF billions

June

02

July

02

Aug.

02

Sep

. 02

Oct

. 02

Nov.

02

Dec

. 02

Jan.

03

Feb

. 03

Mar

. 03

Apr.

03

May

03

June

03

2.5 DERIVATIVES ACTIVITIES AND MARKET RISKS OF BANKS

20 For a comparison, in the FX swap market, the three banks with the largest share of the total outstanding contracts accounted for 49% together.



2

tions.21 The resulting long foreign exchange forwardpositions of banks were closed down as a consequenceof interventions conducted by the MNB at the upperlimit of the intervention band, leaving the sector’s totalopen position broadly unchanged. The measures takenby the Bank to defend the band led to a weakening ofthe forint exchange rate. After the speculative attackwas rebuffed, non-residents gradually wound downtheir forint positions up to end-May, and simultaneous-ly the banking system’s on-balance sheet and forwardopen position returned to the level characterising theperiod preceding the speculative attack.

The consolidation period, after the speculative attackwas repulsed, ended with non-residents opening posi-tions against the forint at end-May. At the Government’sinitiative, on 4 June the Bank and the Governmentagreed to shift the central parity of the currency down-wards by 2.26%, leaving the ±15% intervention bandunchanged. After the shift in the intervention band, non-residents increased their positions against the forint,which resulted in a further significant weakening in theexchange rate.22 Besides the opening of speculativepositions, hedging transactions by holders of govern-ment securities with open foreign exchange positionsmay have played a significant role in the massive(around HUF 500 billion) increase in non-residents’short forint forward positions. Banks’ long open forwardforint positions vis-à-vis non-residents were mainly cov-ered by forward foreign currency sales of domestic non-banks (principally non-financial corporations). Con-

sequently, the banking sector’s total open position re-mained insignificant.

Banks’ forward positions vis-à-vis both non-residentsand domestic firms opened up to a large extent in theperiod surrounding the shift of the intervention band,which has not been experienced in the past. This raisesthe question as to what extent a sudden, large increasein non-residents’ short forint forward position can behedged by inverse forward transactions with domesticfirms. There are no data available on the degree of con-centration of banks’ derivative transactions with domes-tic companies and also on the nature of these forwards(i.e. hedging or speculative purpose).23 In the absence ofsuch information it is difficult to make a judgementabout the extent of potential counterparty risk stem-ming from forward transactions with corporations. It isseen as a further risk factor, that the number of compa-nies which are active in the derivative markets is sup-posed to be relatively limited. Thus, there is no guaran-tee that a sudden, large shift in non-residents’ shortforint positions can always be offset by forward transac-tions with the domestic corporate sector.

Due to the Bank raising interest rates twice by a total300 basis points, non-residents’ forward position againstthe forint stopped increasing by the end of June, fol-lowed by a strengthening of the forint up the end ofSeptember, accompanied by slight fluctuations. A large-value transaction caused a significant change in theopen position of the banking system from July 2003onwards, as a result of which the total open positionvaried in the range of HUF 40–60 billion. Apart fromthis one-off effect, the sector’s total position has re-mained near-neutral.

Although banks’ direct exposure to exchange rate riskscontinued to be low in 2003, the considerableexchange rate volatility and forint depreciation seen inmost recently may expose banks to additional risksindirectly, through their customers’ foreign exchangeexposure. Foreign currency lending to the corporatesector and non-bank financial intermediaries, andbank-owned financial enterprises in particular, wasrobust in 2003 H1. In the case of non-financial corpo-rations, the majority of firms with large outstandingforeign currency borrowings have long foreign curren-cy positions (for example, exporters) or are able toprotect themselves against the negative consequencesof the forint depreciation, due to the nature of theiractivities (for example, firms selling imported goods).However, for a smaller part of firms with short foreignexchange positions, which are not ”naturally” hedgedand do not use derivatives, the negative influences of

Foreign exchange assets and liabilities as a proportion of the balance sheet total

Chart 2-24

FX assets/balance sheet totals

FX liabilities/balance sheet total

40

35

30

25

20

15

10

5

0

Per cent

Dec

. 00

June

01

Dec

. 01

June

02

Dec

. 02

June

03

21 A forward forint position is the result of two transactions, a forint purchase in the spot market and a related swap transaction.22 As seen in the January speculative attack, non-residents opened their position using spot-swap pairs in the opposite direction.23 Anecdotal information from the banks suggests that the dominance of exporters, which use forwards for hedging purposes, is the most likely.

the depreciation of the forint in January and June mayhave added to default risk.

Bank-owned financial enterprises, for example, leasingfirms, do not undertake direct exchange rate risk, simi-larly to banks, as they finance their foreign currency-based leasing and lending transactions mainly from for-eign currency loans provided by their parent bank, andthus pass the exchange rate risk on to debtors.Households, and small and medium-sized enterprisesaccount for a dominant share of financial enterprises’customer base, which in turn do not hedge theirexchange rate exposure with derivatives. Consequently,the strong increase in foreign currency loans to non-bank financial intermediaries may have added to creditrisks at the group level.

In sum, exchange rate volatility has been quite high in2003, a phenomenon not experienced in earlier periods.Banks, however, have continued to refrain from takingexchange rate risks. Consequently, the massive deprecia-tion of the forint does not represent a significant source ofloss for banks. Nevertheless, the observed strong expan-sion of foreign currency lending in certain sectors and theconsiderable forint depreciation has increased banks’exposure indirectly. Taking account of the fact that the

ratio of outstanding loans of debtors negatively affectedby the forint depreciation to the total portfolio is relative-ly low, the resulting increase in risks remains insignificant.

The tendency of banks to rely primarily on forwardtransactions with non-financial corporations in hedgingpositions opened by non-residents, can be consideredas a more serious risk factor. It seems questionable,whether the scope of firms interested in derivatives witha hedging purpose is broad enough to repeatedly ”passon” the large fluctuations in non-residents’ positions.

INTEREST RATE RISK EXPOSURE

The volatility of money market yields and bank interestrates increased in 2003. In order to reign in speculationon forint appreciation in January, the MNB lowered offi-cial interest rates by 200 basis points in two days,24 andthen, in June, raised rates by a total 300 basis points, inan attempt to prevent the forint from depreciating fur-ther. Bank interest rates followed movements in bench-mark money market rates relatively swiftly and to alarge extent (see Chart 2-26). However, banks pricedcorporate loans and household deposits asymmetricallyin response to the interest rate changes in differentdirections – banks reacted more flexibly to the interestrate reductions in both market segments than to theinterest rate increases.25 In the period under review, cor-porate lending and deposit rates adjusted more flexibly,as seen in earlier episodes, than household depositrates. As a consequence, corporate lending and corpo-rate deposit margins changed only very slightly (fallingby 10–15 basis points), while household deposit marginsrose by 50 basis points during the first nine months ofthe year.26 It should be noted, however, that, presum-ably as a result of competition for household fundsbecoming increasingly more intense, household depositrates became more flexible in adjusting to money mar-ket rates in 2003.

Analysing banks’ repricing gaps, the sector’s exposureto interest rate risks moderated in 2003 H1, as the gapnarrowed in absolute and relative terms compared toend-200227 (see Chart 2-27). The three-month cumula-tive forint repricing gap narrowed by more than HUF100 billion. As a result, the ratio of the gap to the bal-ance sheet total was –8.2% at end-June (see Table 2-3).The increase in mortgage lending triggered significantchanges in the repricing profile of the banking system’s



2

24 In the period from the central bank measures taken to check the speculative attack until the restoration of its policy instruments at end-February, the overnight inter-est rate became the Bank’s main policy instrument, instead of the base rate, so effectively there was a decrease by 500 basis points. However, banks’ repricing activ-ity suggests that they considered only the 200 basis point reduction of the base rate as permanent.

25 In part, this asymmetric reaction can be explained by different competitive conditions in the market of corporate loans and household deposits. Presumably, fiercecompetition in corporate lending prevented banks from incorporating the rise in money market rates in full. In the household market, banks incorporated less of theinterest rate increase in household deposit rates, utilising their market power.

26 The three-month BUBOR has been used as a reference rate to calculate the margin between short-term lending and deposit rates.27 It is important to note that much remains to be done in order to improve the reliability of data related to forint and foreign exchange repricing gaps. Consequently,

the Bank has used a number of estimates in compiling the time series for repricing gaps, which, therefore warrants caution when drawing conclusions.

Banks’ total foreign currency position

Chart 2-25

Total open FX position*

On-balance-sheet open FX position*

Off-balance-sheet open FX position*

–500–400–300–200–100

0100200300400

HUF billions500

July

01

Sep.

01

Nov

. 01

Jan.

02

Mar

. 02

May

02

July

02

Sep.

02

Nov

. 02

Jan.

03

Mar

. 03

May

03

July

03

Sep.

03

Nov

. 03

* Positive value: long FX position.

assets and liabilities. The ratio of assets with a repricingperiod of over one year has risen from 8% to 20% andthat of liabilities with a repricing period of over one yearfrom 3% to 11% in the past 18 months.28 The degree towhich the banking system’s gap is concentrated amongindividual banks continued to be high and evenincreased further in 2003 H1.

It should be noted that the analysis of the repricing gapprovides relatively little information about the actual sizeof interest rate risk exposure. For example, the size ofthe negative gap is overestimated by the fact that morethan one-third of forint deposits are overnight and cur-rent account deposits, which, although they are classi-

fied into the shortest (i.e. 0–30 day) category, actuallyare repriced much less frequently and to a much lesserextent relative to market rates. In addition, there havebeen significant changes in banks’ asset profiles (forexample, the increase in the ratio of loans to total assets)and the shape of the yield curve. These factors, in turn,significantly reduce the applicability of the gap analysis.

The strong volatility of yields in 2003 has had a signifi-cant influence on the value of banks’ government secu-rities holdings and the profit and loss realised on gov-ernment securities. Over the first five months of theyear, and in January–February in particular, banksrealised massive gains on their existing holdings, duemainly to the decline in short-term yields. The yieldcurve shifted upwards considerably in June, as a resultof which banks incurred a nearly HUF 3 billion loss onsecurities held for trading purposes. Compared with therise in yields (200–300 basis points), this loss was mod-est, also explained by the relatively short duration of thesector’s total government securities holdings. At end-June, banks held a total HUF 1,434 billion in govern-ment securities. This accounted for 12% of the balancesheet total. The duration of total bank government secu-rities holdings can be estimated at 1.2 years which isbelow the duration of the total outstanding governmentpaper stock. Exposure to interest rate risk is significant-ly reduced by the fact that treasury bills and variable-rate consolidation bonds together account for morethan 40% of total bank government securities holdings(20% and 21%, respectively). Indicating the relativelymodest exposure of government securities holdings tointerest rate risk, the capital requirement of tradingbook position risk accounts for only 1.5% of the bank-ing system’s regulatory capital.

28 It should be noted that these ratios contain duplications on both the assets and liabilities sides, due to the purchases by credit institutions of a part of mortgage bondsand the refinancing of a part of housing loans by mortgage institutions.

90-day cumulative forint repricing gaps of thebanking system

Chart 2-27

Per cent

0

–14

–12

–10

–8

–6

–4

–2

June

98

Dec

. 98

June

99

Dec

. 99

June

00

Dec

. 00

June

01

Dec

. 01

June

02

Dec

. 02

June

03



2

Three-month BUBOR and banks’ interest rates

Chart 2-26

0

2

4

6

8

10

12

Per cent14

Jan

. 0

1

Mar

. 0

1

May

01

July

01

Sep

. 0

1

No

v. 0

1

Jan

. 0

2

Mar

. 0

2

May

02

July

02

Sep

. 0

2

No

v. 0

2

Jan

. 0

3

Mar

. 0

3

May

03

July

03

Sep

. 0

3

3-month BUBOR

Short-term corporate loans

Short-term corporate deposit

Short-term household loans

Major indicators of banks’ interest rate riskexposure

2002 2003 H1

90-day cumulative HUFrepricing gap (HUF billions) –1060 –93790-day cumulative EURrepricing gap (HUF billions) –38 1690-day cumulative USDrepricing gap (HUF billions) –112 –8690-day cumulative HUF repricing gap/ –10.4% –8.2%balance sheet total90-day cumulative EUR repricing gap/balance sheet total –0.4% 0.1%90-day cumulative USD repricing gap/balance sheet total –1.1% –0.8%

Table 2-3



2

Taken together, the volatility of money market interestrates and long-term yields has increased considerablyin 2003. Given the substantial difference betweenassets’ and liabilities’ average repricing period,increased interest rate volatility may strengthen thefluctuations in banks’ net interest income. Sharp rises

in bond yields, as experienced in June, could potential-ly cause serious losses for banks with larger portfolios.However, from the perspective of interest rate riskexposure, the relatively short duration of the bankingsystem’s government securities portfolio can be seenas a mitigating factor.



2

Continuing the previous years’ trend, deposits grew at aconsiderably more modest rate than outstanding loansin 2003 H1. Whereas total outstanding loans grew by33% in one year, deposits and securities (on the liabili-ties side) only continued to grow at a rate of 16%.29

Accordingly, the loan-to-deposit ratio30 continued torise, to reach nearly 100% by the end of the period (seeChart 2-28). This very strong increase in the loan-to-deposit ratio can be attributed to the extremely robustrise in housing loans and households’ low propensity tomake deposits. Contributing to this was the increase inthe corporate sector’s net financing requirement in2003 H1. To a smaller extent, the increase in the loan-to-deposit ratio was also explained by the fact that thedepreciation of the forint in June contributed more tooutstanding loans than to deposits, due to the higherratio of foreign exchange items.

The banking sector’s liquid asset ratio continued to fallin 2003 H1, apart from the transient sharp rise relatedto the speculative attack in January (see Chart 2-29).However, it was not the rundown of liquid assets thatserved as an additional source of funding the expansion

of lending, unlike in 2002 Q1–Q3, as the value of liquidassets remained around the level at end-2002 through-out H1, apart from the jump in January. This was madepossible mainly by the strong inflow of foreign funds.After eliminating the exchange rate effect, foreignexchange liabilities from abroad increased by around30% in the period, short and long-term liabilities havingnearly equal shares in this growth. As a result of the con-siderable rise in short-term liabilities with non-residentbanks, domestic banks’ money market exposureincreased somewhat in 2003 H1 (see Chart 2-30).

Both long-term assets and long-term liabilities increasedconsiderably as a proportion of the total in the periodunder review, while maturity transformation in the bank-ing sector increased only marginally, in contrast to theprevious year (see Chart 2-31). The shift towards long

29 Here, deposits alone increased by 13%.30 Loans and deposits not only include corporate and household sector data, but all loans provided to credit institutions as well as deposits of non-banks and securi-

ties liabilities. Securities purchases by other domestic credit institutions have been eliminated from the data on securities liabilities.

Loan-to-deposit ratio of the banking sector

Chart 2-28

50

60

70

80

90

100

Per cent110

Dec

. 98

Mar

. 99

June

99

Sep

. 99

Dec

. 99

Mar

. 00

June

00

Sep

. 00

Dec

. 00

Mar

. 01

June

01

Sep

. 01

Dec

. 01

Mar

. 02

June

02

Sep

. 02

Dec

. 02

Mar

. 03

June

03

Sep

. 03

Liquid asset ratio

Chart 2-29

Liquid assets* (left-hand scale)

Liquid assets as a percentage of total assets (right-hand scale)

0

500

1000

1500

2000

2500

HUF billions

0

5

10

15

20

25

30

Per cent3000 35

June

01

Sep

. 01

Dec

. 01

Mar

. 02

June

02

Sep

. 02

Dec

. 02

Mar

. 03

June

03

2.6 BANKING SECTOR LIQUIDITY

* Liquid assets: cash and settlement accounts, treasury bill and govern-ment bond holdings (excluding consolidation bonds), securitiesissued by the central bank, short-term deposits at the central bankand short-term claims on foreign banks.



2

maturities within the asset profile in H1 was due to thestrong rise in mortgage lending and in long-term loansto the corporate sector. The sharp jump in foreign bankliabilities was one of the dominant factors responsiblefor the increase in the ratio of long-term liabilities to thetotal, in addition to the rise in the issuance of mortgagebonds.

Overall, the liquidity risk of the banking systemincreased in H1, due mainly to the rapid increase in theloan-to-deposit ratio. The sector’s high loan-to-depositratio, coupled with the decline in liquid assets as a pro-portion of the total, indicates that banks have anincreasingly smaller buffer to manage potential liquiditycrisis situations.

Money market exposure

Chart 2-30

Money market funds*/total labilities

Money market funds(excl. foreign interbank labilities) /total labilities

0

2

4

6

8

10

Per cent12

Dec

. 99

Mar

. 00

June

00

Sep.

00

Dec

. 00

Mar

. 01

June

01

Sep.

01

Dec

. 01

Mar

. 02

June

02

Sep.

02

Dec

. 02

Mar

. 03

June

03

Long-term assets and liabilities of the bankingsector as a proportion of the balance sheet total

Chart 2-31

05

10152025303540455055

Per cent

0

5

10

15

20

25

30

35

Per cent60 40

June

01

Sep.

01

Dec

. 01

Mar

. 02

June

02

Sep.

02

Dec

. 02

Mar

. 03

June

03

Long-term assets/BS totals

Long-term liabilities/BS totals

(Long-term assets–Long-term liabilities)/BS totals (right-hand scale)

* Money market funds: short term domestic and foreign interbank lia-bilities + central bank repo.



2

As of 1 January 2003, several modifications were made inthe method of calculating credit institutions’ capital ade-quacy and regulatory capital. As they affect the indicatorsused in this Report, a summary of the modifications is pre-sented in the box text. Similar to the immediately preced-ing period, capital adequacy ratios have dropped: CARhas fallen to 11.6% and stress CAR to 7.9%. Moreover,the regulatory change reduced the stress CAR by 0.3%percentage points. After adjustment for reinvested earn-ings,31 indicators are but slightly lower than the correspon-ding year-end figures.32 Consequently, the capital strengthof the sector as a whole has slightly declined but remainssatisfactory all in all. Disregarding expected reinvestedearnings, all banks comply with the statutory 8% mini-mum for solvency ratio. However, if the capital require-ment of the trading book is included, there are two banksunable to meet the 8% CAR (see Chart 2-32).

The CAR figures registered by the five largest banksremain below the sector average. Although the differ-

ence has shrunk to less than 3 percentage points, this isa merely temporary phenomenon as it is due to the factthat one of the large banks has been preparing its equi-ty side for the purchase of a subsidiary (see Table 2-4).

The ten largest banks’ resilience to stress has declined(see Chart 2-33). If the maximum loss is incurred, nomore than three banks would have a Tier 1 capital-based stress CAR in excess of 8% (four of them, withadjustment for reinvested earnings). As a result of theregulatory change affecting the Tier 1 capital, one ofthese banks, which used to be far above the limit, hasfallen below the limit.

Over the past six months the risk-adjusted balancesheet total rose by 15%, hardly exceeding the growthpace of the balance sheet total (13%). At end-2002,the trend seen over the last several years that the risk-adjusted balance sheet total grew significantly fasterthan the balance sheet total came to a halt. The major-ity of banks cannot afford any longer to finance creditexpansion via a restructuring the balance sheet. Banksmust increasingly resort to external funding (see Table2-5).

Capital adequacy ratio (CAR), stress CAR andTier 1 CAR

Chart 2-32

CAR Tier 1 CAR* Stress CAR**

7

8

9

10

11

12

13

14

15

1998

1999

2000

2001

2002 Q

2

Q2

2002

2003

16Per cent

2.7 FINANCIAL POSITION AND CAPITAL ADEQUACY

31 Regulatory capital is increased by “banks’ aggregate after-tax profit x (1 – dividend pay-out ratio)”. Based on a cautious estimate, a 30% indicator was used for thedividend pay-out ratio. General risk provisions can be disregarded, as they increase the regulatory capital and reduce the after-tax profit and, consequently, the reg-ulatory capital through retained earnings. Thus, due to reduced tax-payment obligations, their overall positive impact on the regulatory capital is insignificant.

32 This would make CAR 12.6% and the stress CAR 8.9% as against 13% and 9.1%, respectively, at the end of last year.

Capital adequacy of the five largest banks andthe sector

June Dec. June Dec. June2001 2001 2002 2002 2003

Share of the five largestbanks of the bankingsector’s assets 54.8% 60.4% 59.6% 58.7% 59.6%Average* CAR of the fivelargest banks 12.1% 11.9% 10.5% 11.7% 10.5%Banking sector CAR*,excluding the five largestbanks 14.8% 16.7% 15.5% 14.9% 13.3%Average CAR* of thebanking sector 13.5% 13.9% 12.5% 13.0% 11.6%

* Weighted.

Table 2-4

* Tier 1 capital/risk adjusted total assets.** (tier 1 capital – net value of non-performing claims)/(risk adjusted

total assets – net value of non-performing claims).



2

The sector’s internal ability to accumulate capitalremains strong, nearly equivalent with the expansion ofactivity. Primarily as a result of an extremely high rate ofreinvested capital due to last year’s low dividend pay-out ratio, regulatory capital before deductions rose by19% over the past 12 months (and by 23% if adjustedfor the reinvested earnings during the first six months of2003), and by 5% in the past six months (or 12% withthe same adjustment). Despite regulatory changesaffecting the regulatory capital, the ratio of Tier 2 capi-tal has seen little growth,33 with its ratio remaining thesame even if adjusted for reinvested earnings. As aresult of investments and subordinated loan capitalextended to financial institutions, insurance companiesand investment companies, capital deductions contin-ued to increase, rising by nearly 10% to HUF 73.5 bil-lion.

Changes implemented in the methodologyof calculating credit institutions’ capitaladequacy and regulatory capital

The capital adequacy ratio (CAR) used so far hasbeen replaced by the solvency ratio as a result oflast year’s amendment of Act CXII of 1996 onCredit Institutions and Financial Enterprises (here-inafter: Hpt.). This is more than just a change inname, as it also involves major changes to thecontent. The numerator of the new index con-tains the capital constituents that can be used forthe fulfilment of the capital requirement for trad-ing book, exchange rate and commodity risks. Inrespect of the CAR numerator, these items arededucted. Hence, the solvency ratio of institu-tions keeping trade books will always exceedtheir CAR. From now on, the measures specifiedin Hpt. in relation to CAR (i.e. maintenance ofminimum capital) will be assigned to the solven-cy ratio. This means that the capital requirementfor trading book, exchange rate and commodityrisks is no longer subject to punitive sanctions.Thus, the minimum capital requirement has effec-tively decreased. In the Bank’s opinion, from aprudential perspective CAR is a more appropri-ate indicator, as it includes capital requirementsfor more risks. For this reason, the Bank will con-tinue to use CAR in its analyses.

Annex 5 of the Hpt. on the components and cal-culation of regulatory capital has also beenchanged as follows:

– Cumulative preferential shares (also paying div-idends from previous years in the profitable

Box 2-2

33 The ratio of supplementary capital within regulatory capital before all deductions rose by less than one percentage point on a year earlier, and reached 12.8%.

Financial position of the ten largest banks, theiraverage, and sector average and maximum losses incurred as a result of non-performingassets, 30 June 2002 and 2003

Chart 2-33

4

5

6

7

8

9

10

11

12

13

14

0 1 2 3 4 5 6 7Net value of non-performing claims/risk adjusted total assets (per cent)

Tier 1 capital/risk adjusted total assets (per cent)

Banking sector average, 30 June 2002

Banking sector average, 30 June 2003

Average of the ten largest banks, 30 June 2002

Average of the ten largest banks, 30 June 2003

15

Weighted averages.

Assets at risk-adjusted values (percentage) Dec. June Dec. June Dec. June June 2003/2000. 2001. 2001. 2002. 2002. 2003. June 2002*

20 per cent weight 4.6 5.0 3.4 4.0 3.9 1.4550 per cent weight 2.0 2.4 3.5 5.2 6.0 2.15100 per cent weight 73.7 73.6 72.2 70.6 71.0 1.23Sum of adjusted balance sheet items 80.3 80.9 79.1 79.8 81.0 1.28Weighted value of contingent and otherfuture liabilities 18.9 18.2 20.0 19.2 18.0 1.13Weighted value of forward claims 0.8 0.8 0.9 1.0 1.0 1.46Risk-adjusted balance sheet total(HUF billions) = 100% 4,698 5,031 5,363 5,955 6,508 7,468 1.25Balance sheet total (HUF billions) 8,130 8,495 9,053 9,142 10,185 11,480 1.26

* The index numbers have been derived from the increases in the underlying data.

Components of the risk-adjusted balance sheet total

Table 2-5



2

The concentration of excesses over limits has notchanged: the majority remain related to three banks.The amount of deductions have, however, increased farslower than activities (by 3.7%), and a simultaneousshift has been seen in proportions: capital deductionsresulting from limit excesses over large exposures relat-ed to affiliated companies and investment limits havedeclined, while other large exposure limit excesses havegrown (see Chart 2-34).

Within six months the amount of capital requirementrelated to country risk exposure increased by roughly90% to HUF 16.2 billion, but the total amount as wellas the strong fluctuation may be linked to a singlebank. The total amount of limit excesses and countryrisk exposure to be covered by capital has grown fromHUF 60 billion34 last year to HUF 69.6 billion (seeChart 2-35).

The capital requirement for trading book risks continuesto grow at an exceptional pace: over the past six monthsit rose by a massive 67% (to HUF 25.3 billion). Its growthrate far exceeded the 13% increase in capital requirementfor exchange rate risks (which rose to HUF 3.5 billion).

Over the past six months the sector CAR has deterioratedby a total of 1.4 percentage points. 1.7 percentage pointsof this change was due to the expansion of activities, withthe increase in deductions affecting regulatory capital con-

Excesses over limits pursuant to the CreditInstitutions Act

Chart 2-34

Limit overruns at parent and subsidiary companies*

Excesses of investment limit

Excesses of large exposures limit

80

70

60

50

40

30

20

10

0

HUF billions

1998

1999

2000

2001

2002 Q

2

Q2

2002

2003

Regulatory capital and its components

Chart 2-35

Deductions: amount of limits excesses and country risk

to be covered by capital

Trading book capital requirement

Tier 2 capital

Tier 1 capital*

Regulatory capital

–200

0

200

400

600

800

1 000HUF billions

1998

1999

2000

2001

2002 Q

2

Q2

2002

2003

year are deducted from the Tier 1 capital andcomprise part of the Tier 2 capital. This changebring the Hungarian category of core capital inalignment with the category “Tier 1” specifiedin Directive 2000/12/EC of the EuropeanUnion. According to the audited data of 31December 2002, this change decreases Tier 1capital by 1.4% but increases Tier 2 capital by11.1%. At the sector level, dividend preferenceshares are insignificant: only one bank usesthem.

– Henceforth, that part of Tier 2 capital whichcannot be used because of limits related tothe ratios of Tier 2 capital to Tier 1 capitaland the subordinated loan capital within Tier2 capital can be considered in Tier 3 capital.Currently, the change affects practically onlyone bank, and its impact is negligible at a sec-tor level.

– Henceforth the commodity risk of all activi-ties – not only trading book, but also bankingitems – must be covered by regulatory capi-tal. As of 30 June 2003, its amount wasinsignificant at the level of both individualbanks and the sector as a whole (the latterbeing a mere HUF 23 million).

34 The current figure is nearly HUF 18 billion less than the value specified in the June 2003 Stability Report (HUF 77.9 billion). This is due to the fact that the auditeddata of one of the banks contained HUF 15 billion less limit excesses than indicated in the preliminary data.

* Lowered by the amount of deductions due to investments in otherfinancial intermediaries.

* According to the Hungarian regulations limit excesses should becovered by capital.



2

tributing a mere 0.4 percentage points. The rise in regula-tory capital, on the other hand, improved the CAR by 0.7percentage points. If the above computation is adjustedfor estimated reinvested earnings, then the increase in reg-

ulatory capital almost completely offsets the growth in fac-tors reducing CAR. On aggregate, the sector’s capitalstrength is adequate. In the future, a moderate deteriora-tion of capital adequacy is expected to continue.



2

Despite further slow-down in economic growth, bankingsector profitability improved significantly in the first halfof 2003, even in comparison to last year’s extremelypleasing performance: the sector’s HUF 104 billion after-tax profit was 46% up on the base period. The return onasset and return on equity indicators indicate spectacularimprovement on a year earlier: ROA rose to 1.58% (from1.33% in the first half of 2002), while ROE advanced to17.4%, from 14.6% in 2002 H135 (see Chart 2-36).

The rapid increase in subsidised housing loans played anoutstanding role in the improvement of banking systemprofitability over the last one and a half years, as it provid-ed banks with significant extra profit through high interestmargins.36 According to our estimates, profit realised onsubsidised housing loans may have accounted for 10-13% of the banking sector’s total net profit in 2003 H1.37

In addition to the significant improvement in bankingsector results, it is important to emphasise that prof-itability differences widened in 2003: the advantages ofthe most profitable banks increased significantly in com-parison to the ”average bank”. This can be explained bythe high concentration of mortgage lending. In terms of

banks operating in both 2002 H1 and the first half of2003, the number of loss-making banks rose from sevento nine.38 Despite a moderate increase, the aggregatemarket share of loss-making banks remains low (8%).From the point of view of systemic stability it is afavourable development that only smaller institutionscan be found among banks not making a profit.

In contrast to 2002, the growth in net interest income –in absolute terms – exceeded the change in net fee andcommission income. Furthermore, the rapid rise in netfee and commission income also played an importantrole in the favourable development of banking sectorprofitability. Although the rise in operating costs gath-ered considerable speed in comparison to last year, itlagged far behind growth in revenues. In comparison tothe first half of 2002, the negative effect of the netincrease in value adjustments and provisions was some-what more pronounced (see Table 2-6).

To a major extent, the banking sector’s outstanding prof-it in the first half of 2003 was due to an increase of netinterest income far in excess of the previous year’s cor-responding figure. For the most part this is the result ofthe favourable volume effect fuelled by strengtheninglending activity and the continuing rise in the proportionof housing loans which offer a high interest margin. Theincreasing role of household lending in profits is clearlyindicated by the fact that interest income earned onhousing and consumer loans amounted to nearly one-quarter of total interest income on lending in the first halfof 2003, while in 1999 this ratio was less than 10%.

Despite the increasing significance of household lending,the spread between banks’ average return on assets andthe cost of funds fell by 25 basis points in the first half of2003 compared to a year earlier (see Chart 2-37). In thewake of measures adopted by the central bank to man-age the speculative attack, at the beginning of the yearthere was a temporary upsurge in the share of banks’ lowinterest instruments (central bank O/N deposit), which

2.8 PROFITABILITY

35 Data for ROA and ROE always refer to the last two half-years.36 Prior to the modification of the subsidy conditions in June 2003, banks could realise a 7-9% interest margin on state-subsidised housing loans.37 In this category, we refer to the housing loans granted under the subsidy scheme introduced in 2001. ”Old” type subsidized housing loans (from OTP) have not

been taken into account.38 In addition, one bank commencing operations in 2003 recorded a loss in H1.

Banking sector ROE

Chart 2-36

0

5

10

15

20

Per cent

25

Dec

. 99

June

00

Dec

. 00

June

01

Dec

. 01

June

02

Dec

. 02

June

03



2

tightened the spread. Moreover, as demand and currentaccount deposits yield very low interest, only part of theearly-2003 major decline in interest rates could beapplied. In the second half of 2003, no further decreaseis expected in the spread, as the inflexible pricing ofdemand and current account deposits (amounting to37% of the total stock of forint deposits) following theJune interest rate rises is expected to widen the depositmargin. Although the significant decline in the margin onnewly disbursed housing loans will tend to tighten thespread in the second half of the year, in the case of theoverwhelming majority of the loans, banks are stillexpected to realise the high interest margin appliedbefore modification of the subsidy system.

Net fee and commission income realised by the bank-ing sector exceeded the figure registered in the baseperiod by 32%. As a result, the significance of fee andcommission income has further enhanced in profit gen-eration.39 As fee and commission income is among themost stable sources of income, this change in theincome structure is deemed as favourable. The positiveimage is somewhat overshadowed by the fact that ifbanks are considered individually, there are extremelysignificant differences as to the weight of net fee andcommission income, and these differences increasedfurther in 2003. Consequently, fee and commissionincome is highly concentrated within the banking sec-tor: 65% of the aggregate fee and commission incomeearned by all banks was concentrated at three banks inthe first half of 2003. By comparison, the three bankswith the largest net interest income have an aggregate

share of 43% of the total and the three largest bankshave roughly equivalent market shares in the sector’sbalance sheet total.

Non-interest income related to housing loans consti-tutes one of the most important sources of the rapidgrowth in net fee and commission income.40 In thewake of a 25% increase in turnover with bank cardsissued by domestic banks, revenues earned from thebank card business also contributed considerably to therise in fee and commission income. In addition, thebanks’ fee income from guarantees doubled (reachingHUF 8.1 billion). A further factor pushing up fee andcommission income was the considerable increase ofcommissions and fees charged for payment services in2003 Q1.

Net profit on financial operations rose by 7.5% in thefirst six months of 2003 on a year earlier. As a result ofa considerable increase in exchange rate volatility, boththe gains realised on foreign exchange assets and liabil-ities as well as the valuation results, and the losses suf-fered on foreign exchange derivatives have increasedgreatly. Consequently, the net profit on foreignexchange transactions and exchange rate changes,accounting for the largest portion of profit on financialoperations, exceeded the corresponding figure from thebase period by roughly 3%. As a consequence of a sig-nificant intensification of interest rate volatility, morefavourable aggregate gains were earned on securitiesbought for trading than in the base period. In the first

Components of spread

Chart 2-37

–15

–10

–5

0

5

10

15

20Per cent

0.00.51.01.52.02.5

3.03.54.04.5

Per cent5.0

Interest income/interest-bearing assets

Interest expenses/interest-bearing liabilities

Spread (right-hand scale)

Dec

.

June

Dec

.

June

Dec

.

June

Dec

.

June

99

00

00

01

01

02

02

03

39 The proportion of net commission and fee income rose from 22.9% in 2002 to 24.4% in 2003 within gross operating income.40 Pursuant to the relevant accounting regulations, commission income qualifying as interest like, such as service charges, disbursement fees or availability fees should

not be included in this category.

Banking sector profits

2002 2003 ChangeH1 H1

HUF Indexbillions

Net interest income 179.9 209.6 29.7 116.5%Dividend income 1.3 11.3 10.0 893.7%Net fee and commissionincome 60.4 79.6 19.2 131.8%Net profit on financialoperations 20.6 28.1 7.5 136.4%Other income/loss –9.6 –13.3 –3.7 138.6%Operating costs 164.2 181.2 17.0 110.4%Change in value adjust-ments and provisions –5.9 –13.5 –7.7 230.6%Profit on ordinaryactivities 80.3 118.3 38.0 147.4%Extraordinary profit 0.2 –1.0 –1.2Pre-tax profit 80.5 117.3 36.8 145.7%After-tax profit 70.8 103.6 32.8 146.3%

Table 2-6



2

half of 2003 as a whole, banks realised nearly HUF 4 bil-lion in profits, despite the fact that in June the sectorbooked a loss of approximately HUF 3 billion on secu-rities purchased for trading purposes.41

In the first half of 2003, the proportion of non-interestincome rose from 32% in the previous year to 36% ofgross operating income (see Chart 2-38). In addition toa robust rise in net fee and commission income, a sud-den jump in dividend income – linked primarily to onelarge bank – had a major impact on this trend. Half ofthe spectacular rise in the proportion of non-interestincome can be attributed to this latter one-off effect.

In 2003 H1, the net increase in value adjustments andspecific provisions was nearly twice as high as in thebase period. The major part of the rise in loan losses isrelated to corporate loans, which resulted in a moder-ate increase in the ratio of provisions/gross value ofloans for the corporate loan portfolio on a year earlier.Net changes in provisions for equity interests had prac-tically no impact on 2003 H1 results, as compared tothe base period when this item improved the balancemoderately.

In 2003 H1, operating costs rose far more quickly thaninflation (10.4%) in the banking sector. As a result of5.5% growth in average staff and a 3.6% increase in percapita expenditure, staff costs rose by 9.3%. Growth inthe number of employees can be associated primarilywith a boom in housing loans. The 26% growth in mar-keting costs can also mainly explained by the increasedrole of household lending, whereas IT costs fell by 10%after rising quite rapidly in 2002.

As banks’ gross operating income has grown by 26%,despite a considerable increase in the real value ofcosts, the cost-to-income ratio improved markedly: it fellfrom 63% last year to 56%. Although more moderately,the cost-to-balance sheet total ratio also improved: itdropped from 3.8% to 3.6% in a year42 (see Chart 2-39).The improvement in cost efficiency ratios, and the cost-to-income ratio in particular, should be assessed withcaution since a substantial part of the increase in totalassets and income can be attributable to the boom insubsidised housing loans. Thus, the favourable changein efficiency ratios may provide a somewhat skewedpicture of banks’ cost efficiency.

Net interest and non-interest income as a proportion of gross operating income

Chart 2-38

Net interest income Non-interest income

100

90

80

70

60

50

40

30

20

10

0

Per cent

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003 H

1

Operating costs as a percentage of total assetsand the cost/income ratio

Chart 2-39

Per cent Per cent

Dec

. 96

June

97

Dec

. 97

June

98

Dec

. 98

June

99

Dec

. 99

June

00

Dec

. 00

June

01

Dec

. 01

June

02

Dec

. 02

June

03

5.0

4.5

4.0

3.5

3.0

2.5

2.0

90858075706560555045403530

Costs/total assets (left-hand scale)

Costs/income (right-hand scale)

41 In 2002 H1, banks did not make profit on the price changes related to securities purchased for trading purposes.42 The data related to the costs-to-balance sheet total ratio always refer to the last two half-years.

3 CURRENT TOPICS RELATED TO STABILITY



3

In keeping with the Bank’s practice of recent years, theimpact of a few relevant extreme adverse events areestimated in order to contribute to the assessment ofsystemic risk in the banking sector. The losses incurredby the individual banks are aggregated and expressedas a percentage of the sector’s core capital. Note thatthe adopted approach differs markedly from that fol-lowed in other parts of the Stability Report. One of themajor differences is that here we focus on changes inasset value rather than changes in income. As in somecases losses in asset values are not immediatelyrealised, and it is possible to avoid some of those loss-es, our results should be considered as very rough esti-mates.

The tests’ results are affected by banks’ capital strength,the exposure to certain risk factors and the shocks con-sidered. In respect of capital strength, as of 31December, 2002 the sector’s core capital43 was 16.3%up on a year earlier.44

MARKET RISK45

The range of shocks investigated did not change.46 In cal-culating extreme events we heavily rely on the method-ological recommendations of the BIS and a few largeinternational banks’ practices. It must be noted, that dueto the recent strong foreign exchange and interest ratevolatility, historic shocks approached the hypotheticalevents closer than ever. A 500-bp increase and a 300-bpdecrease in domestic interest rates are analysed first. Asfor exchange rates, 40% appreciation and depreciationare considered. Losses due to foreign interest ratesfalling or rising by 200 basis points are also estimated.

The exposure is the largest to a potential increase indomestic interest rates. In addition to the level of this

exposure, its increase seen in the past one and a halfyears is also extremely high. Nevertheless, a major partof this increase can be attributed to a single bank, whichsaw significant growth in its maturity transformation dueto the boom in household mortgage lending.

No definite shift is seen in exposure to exchange raterisk and to large moves of foreign interest rates in com-parison to previous years. However, quarterly data indi-cate significant fluctuation.

The relative size of losses generated by rises in domesticinterest rate grew continuously during the period underreview, while the potential losses caused by exchangerate moves47 rose in 2003. Losses due to an increase indomestic interest rate also became more concentrated:50% of the total loss would be incurred by a single bank(68.7% at the end of 2003 Q2). This loss would, howev-er, account for only 23% (or 37%, respectively) of thecore capital of the bank affected. Although losses due toexchange rate changes are less concentrated, severalbanks would suffer losses in excess of 50%, in one caselosses would even exceed 100%. A drop in domesticinterest rates and changes in foreign interest rateswould, however, have insignificant impact.

Based on the market stress tests there are two phenome-na that should be emphasised. On the one hand, the

3.1 STRESS TEST RESULTS

43 The end-2002 core capital figures are used for the purposes of computation for each quarter’s data. Naturally, upon availability the analysis will rely on appropriate-ly adjusted quarterly core capital data.

44 This analysis covers banks active at the end of December 2002. In contrast to the practice followed in stress tests in earlier years, but in line with the chapter on thebanking sector in the Stability Report, Eximbank and MFB have been removed from the analysis. Results for earlier years have been modified accordingly.

45 Two of the changes related to data reporting deserve mentioning. The breakdown of repricing balance sheets has been modified (the three-months – one-year cat-egory has been replaced by the categories 3–9 month and 9 month – one-year). In addition, foreign exchange denominated items are now reported by currenciesinstead of the aggregated FX position (EUR- and USD-denominated items are analysed here). As a result, more accurate information is available. These changes dohave an impact on the tests’ result, but it is not likely to be significant.

46 See Report on Financial Stability, 2002/2.47 In the case of FX risks, the EUR and USD losses on net assets are simply added up. This way an upper limit is generated and the result slightly exceeds the losses

calculated by netting between currencies. However, the difference is insignificant.

Market risk shocks

Market risk Domestic Foreign Exchange rateinterest rate interest rate (per cent)

Shock 1 +500 bp +200 bp 40%Shock 2 –300 bp –200 bp –40%

Table 3-1



3

growing maturity transformation of a single bank entails asignificant interest rate risk. However, it must be takeninto consideration that losses materialise only over thelonger term (they are realised slowly or not at all), and thustheoretically there is sufficient time for adjustment. On theother hand, the effects of interest rate rises on liabilitiescannot be felt because of the pricing practice of banks.That is, a considerable portion of short-term liabilities isless sensitive to interest rate changes. The losses incurredby the individual banks and the sector, as a whole, due tochanges in the exchange rate would slightly exceed thecorresponding figures in previous years, although both theexposure and the losses fluctuate widely.

CREDIT RISK

The trends in banks’ portfolio that commenced earlierhave continued: the share of non-performing loans(hereinafter “NPL”)48 and risk-free assets have decreased,along with the volatility of non-performing claims.

Although our method of defining credit shocks49 has notchanged, the positive changes in the portfolio impliedless severe shocks. The four shocks considered are thefollowing: 1) half of the risk-free investments are turnedinto loans; 2) the NPL of each portfolio increases by 2standard deviations, estimated from the data for the last7 years; 3) and from the data for the last 6 years; and 4)the amount of NPL doubles.

The results indicate an improvement in the bankingsector’s resilience to shocks. The consolidation of

large banks which were previously loss-making hascontributed significantly to the improving trendobserved during recent years. Another issue of con-cern is that the boom in housing loans has positiveeffects on the results of the stress tests through two

Losses due to market risk shocks, as a percentage of tier1 capital

Chart 3-2

Foreign int. rate+ Foreign int. rate–

Domestic int. rate+

Foreign exchange rate+

Domestic int. rate–

Foreign exchange rate–

–20

–15

–10

–5

0

–25Per cent

Q4

Q4

Q1

Q2

Q3

Q4

Q1

Q2

2000

2001

2002

2002

2002

2002

2003

2003

48 NPL defined as the sum of doubtful and bad claims.49 See Report on Financial Stability, 2002/2.

Market risk exposure (Duration-weighted sum

of discounted net positions – positive and negative)

Chart 3-1

FX2+ FX1+

Ft– FX2– FX1–

Ft+

–1,000,000

–500,000

0

500,000

1,000,000

1,500,000

2,000,000

2,500,000

Q4

Q4

Q1

Q2

Q3

Q4

Q1

Q2

Q3

Q4

Q1

Q2

1999

2000

2001

2001

2001

2001

2002

2002

2002

2002

2003

2003

Characteristics of the aggregate portfolio at theend of 2002

Year NPL Risk- Volatility of NPLfree (HUF million and per cent)

Share Change assets(per Last 7 years Last 6 yearscent)

1999 2.6% 106.7% 21.10%2000 1.7% 77.5% 18.20% 61,557 (31.4%) 48,531 (26.8%)2001 1.5% 105.1% 15.20% 44,822 (24.7%) 45,372 (25.7%)2002 1.1% 101.0% 13.10% 35,958 (22.8%) 38,709 (24.6%)

Table 3-2

Losses caused by credit shocks (As percentage of Tier1 capital)

Shock 1 Shock 2 Shock 3 Shock 4

2000 3.2% 31.1% 25.0% 17.3%2001 2.0% 19.2% 16.3% 14.2%2002 1.4% 13.9% 13.1% 12.1%

Table 3-3



3

channels over the short run: improving profitabilityand the quality of portfolio. However, these effects aremerely temporary.

The market share of banks with losses exceeding theircore capital is low in terms of the total asset of the bank-ing sector and has dropped further in comparison to theprevious years. However, their share in the total

amount of losses stagnated. The concentration of lossesremained practically unchanged.

Share of banks with losses in excess of 100%

Shock 2 Shock 3

2000 2001 2002 2000 2001 2002

Assets 22.85% 5.54% 3.92% 5.51% 5.54% 3.92%Losses 80.22% 35.23% 37.22% 37.20% 41.07% 40.89%

Table 3-4 Concentration of losses

Shock 1 Shock 2

2001 2002 2001 2002

Largest 39.5% 35.0% 30.1% 37.2%Largest 5 90.2% 83.7% 77.4% 71.1%

Shock 3 Shock 4

2001 2002 2001 2002

Largest 37.1% 40.9% 24.2% 20.3%Largest 5 73.2% 71.1% 69.8% 68.3%

Table 3-5



3

The following provides an analysis of the current situa-tion in the non-financial corporate sector from the per-spective of profitability, indebtedness and liquidity, all ofwhich are factors key to financial stability. The analysisis based on balance sheet and profit and loss accountdata in corporate tax returns.50

SUMMARY

In terms of stability, no material change occurred in thenon-financial corporate sector in 2002. As regards prof-itability, the entire sector performed worse than in2001. However, compared to earlier years, the changewas not substantial.

Changes in the economic environment in 2001 and2002 affected the individual sub-sectors of the corpo-rate sector differently. Business conditions in those sub-sectors facing competition from the non-resident sectorfailed to improve during those two years, which in turnaffected the profitability of exporters adversely. Afavourable development is that, relative to the steepdecline experienced in 2001, profitability in these tradi-tionally highly profitable sub-sectors deteriorated onlyslightly in 2002, despite the downturn in the businesscycle and the appreciation of the forint exchange rate.Moreover, manufacturing profitability improved in2002. Profitability in the services sector hardly changedin 2002. The underlying reason for the profitability levelbeing still high is that consumption grew strongly in2001–2002, with consumption expenditure on servicesrising more rapidly than aggregate consumption. Twomajor impacts are expected to affect future develop-ments in aggregate profitability. An anticipated upturnin the global business cycle and hence in economicgrowth in Europe in 2003 and 2004 is expected to

boost external demand for Hungarian tradables, andthus enhance the profitability of exporting companies.At the same time, a considerable slowdown in con-sumption growth is expected to occur, which may havean adverse impact on the services sector.

Similar to profitability, there were no changes to thecapital structure of companies, which could have posedupside risks to the stability of the sector as a whole.However, one unfavourable development was that thesituation in the corporate sub-sectors with the greatestrisks to financial stability failed to improve in 2002. Theweight of these sub-sectors has hardly changed and isstill significant.

PROFITABILITY

In order to measure corporate profitability, the indicatorsused in last year’s Report were employed. The first natu-ral indicator is the ratio of loss-making companies to totalcompanies. In 2002, 29% of all companies incurred oper-ational losses. This does not represent a substantialchange compared to 2001,51 and another favourabledevelopment was that in 2002 the ratio of aggregate lossfell by 3 percentage points from 44% to 41%.

A detailed analysis of the indicators calculated from bal-ance sheet and profit and loss account data in corporatetax returns provides for the possibility of a more thoroughinvestigation. As a first step, ROA and profit margin wereused to analyse non-financial corporations.

ROA (Return on Assets) = operational income/balancesheet total.52

Profit margin = operational income/net sales revenue.

3.2 CORPORATE SECTOR PROFITABILITY AND STABILITY

50 The results of the analysis cannot always be compared directly with those in the Report on Financial Stability published in November 2002, since, relative to the sam-ple analysed last year, the number of the companies included in this analysis has changed. Similarly to last year’s analysis, this year’s analysis also excludes financial,budgetary and non-profit companies. What is new is that those companies where the number of employees was fewer than two or fewer than five for three succes-sive years have been excluded. Thus, the number of observations has fallen by 66% to 429,000.

51 The indicator was 28% in 2001.52 In the case of the ROA indicator, the year-end value of the stock variable was used in the denominator instead of the annual average. Indicators calculated from

stock and flow data cannot be regarded as authoritative. The analysis focussed on annual changes. It should be noted that as balance sheet and profit and lossaccount data reflect a situation at a point of time and during a given period respectively, ROA-type indicators may be biased during both period of high and declin-ing inflation. The reason for this is that flow data reflect average annual inflation, whereas stock data show year-end inflation. As this analysis is focussed on recentdevelopments, it does not seek to filter out such effects.



3

Corporate profitability between 1993 and 2002 wasexamined on an aggregate level and in a sectoral break-down.

AGGREGATE PROFITABILITY

In order to present an accurate picture of the changesin the profitability of the sector, operational profitabilitywas divided into a component indicating input efficien-cy and another indicating output efficiency.

ROA = (operational income/sales revenue)*(sales rev-enue/balance sheet total)

Operational income as a proportion of sales revenuerepresents the operational profit margin, which canfunction as an indicator of corporate cost efficiency.It measures the extent to which corporate output istranslated into profit, i.e. to what extent companiescan adjust on the cost side to changes in the eco-nomic environment. The ratio of sales revenue toassets functions as an indicator of capacity utilisa-tion, which denotes corporate adjustment in terms ofoutput.

Chart 3-353 shows ROA and its two components. Nearlythroughout the period under review, ROA was moresensitive to developments in the profit margin, i.e. inputefficiency. This was also the case in 2001, when deteri-orating cost efficiency resulted in deteriorating corpo-rate profitability. By contrast, the aggregate corporateprofit margin remained unchanged in 2002. The slightdecline in operating profitability can be attributed to amoderate decrease in sales revenues.

The deterioration in profitability in 2002, however,was not significant, and did not exceed the slightdecrease registered in 2001. Currently, the earningcapacity of the corporate sector is still close to thehigh levels recorded in both 1997 and 1999. Thestrong profitability in the period between 1997 and1999 was due to the global economic recovery and inparticular to foreign direct investment, which wasespecially significant in the mid-1990s. The latterimproved export profitability particularly rapidly. Aseparate section analyses the profitability of the exportsector.

In addition to aggregate profitability, a factor which isof relevance to stability is whether or not the profitabil-ity of the least profitable companies deteriorated to alarger than average degree. In order for this to beassessed, the distribution of the individual corporateindicators must be examined (see Chart 3-4). In order

to analyse the distribution, for firms’ ROA we selectedthe lower and upper quintiles, i.e. the limit of the lowerand upper 20% of companies, and the median and theaverage. These indicators strongly suggest improvingcorporate profitability until 1998. Both mean valuesand quintiles increased prior to 1998, i.e. the distribu-tion was not skewed. The narrowing of the distributionin 1999–2000, however, points to the fact that thedeterioration of aggregate profitability was mainlyattributable to the poorer performance of the mostprofitable companies.

Developments in the position of least profitable compa-nies relative to the mean is relevant for stability, since itindicates whether these firms are catching up or loosing

Corporate sector profitability indicators (1993–2002)

Chart 3-3

Profit margin excl. MOLProfit margin

ROA ROA excl. MOL

Sales revenue to assets, right scale

0

1

2

3

4

5

6

7

8

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Per cent

0.0

0.2

0.4

0.6

0.8

1.0

1.2

1.4

1.6

53 In 2001, due to the losses suffered by its gas division, MOL’s operating income was negative, which is an unusual development in the period as a whole. The chartalso includes the profitability indicators excluding MOL, lest the one-off effect of this significant anomaly distort the picture of the sector as a whole. In the analysiswe rely on these latter numbers.

Distribution of ROA (1993–2002)

Chart 3-4

Lower quintile Upper quintile

Median ROA Mean ROA

–15

–10

–5

0

5

10

15

20

25

Per cent

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002



3

ground. The difference between the bottom quintileand the median shows just what we are investigating,namely to what extent the position of the least prof-itable companies, relative to the average, has improvedor deteriorated. In terms of stability, the mostfavourable scenario is when both the mean value of thedistribution and the bottom quintile increase, with thelatter growing more vigorously. This, i.e. improvementamong the least profitable companies, was discerniblebefore 2000 (see Chart 3-5). The only exception is 1999,when the entire distribution shifted left. Even so, the rel-ative position of the least profitable companies contin-ued to improve, since the median value decreasedmore significantly than the bottom quintile.

In 2001, this improvement came to a halt, and the leastprofitable companies drifted away from the median. In2002, the difference between the median and the bot-tom quintile further increased, i.e. differentiation, whichstarted a year earlier, continued. This, however, does notmean that the risks that the least profitable companiespose to stability are significantly higher, for the profitabil-ity of such companies is still close to their 2000 high.

SECTORAL PROFITABILITY

Detailed data show that in the period under review thetwo largest sub-sectors,54 i.e. manufacturing and services,are key to the dynamics of the corporate sector as awhole (see Chart 3-6). As manufacturing plays a decisiverole, the issue of its profitability is dealt with separately.

The profitability of the services sector improved steadilyuntil 1999, but the declined slightly in 2000. However inboth 2001 and 2002, it remained close to its 1999 high.

The high earning capacity of the services sector over thelast two years can be attributed to the fact that the prof-itability of Trade and repairs,55 the most important sectorin terms of operational income, continued to improve.This is very likely to have been closely related to a vigor-ous increase in household income and consumption.After a period of massive deterioration in their profitabil-ity in 2002, companies in the other major services sub-sector, i.e. Transport, storage and telecommunications,were still unable to improve profitability in 2001. In fact,in 2002, their profitability declined. One reason for thismight be that world market prices of crude oil spikedagain in 2002, directly boosting transport costs.

Profitability of the Construction sub-sector remainedbroadly flat in 2002, close to its high in 2001. The rea-sons for this are likely to include the Government’shousing policy-related measures and infrastructureinvestment. Profitability in Energy & water declined.Nevertheless, it was still close to its peak values record-ed in the second half of the 1990s.56

Manufacturing profitability, which deteriorated signifi-cantly in 2001, stopped declining in 2002. As this sub-sector contributes to the profit of the entire corporatesector considerably, we will discuss its developments ina separate section.

Real appreciation of the forint in 2001 and its strengthin 2002 were two major macroeconomic phenomena

Changes in mean and median ROA (1993–2002)

Chart 3-5

–1.5–1.0–0.5

0.00.51.01.52.02.53.03.5

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Ppoint

Change of median Change of lower quintile

Profitability (ROA) in individual sub-sectors(1993–2002)

Chart 3-6

Manufacturing Services

Construction Energy & water

Corporate sector

–2

0

2

4

6

8

10

12

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Per cent

54 In the period under review, within the corporate sector as a whole, manufacturing and the services sector contributed to the operational income by 50% and 33%,respectively. Agriculture and mining, excluded from the chart, accounted for a mere 5% of the entire non-financial corporate sector.

55 Trade and repairs as well as Transport, storage and telecommunications account for nearly 50% and 20%, respectively, of the operational income of the entire serv-ices sector.

56 In assessing the performance of this sub-sector, it should be noted that prices here are mostly regulated. Therefore, decisions by the government distort the the effectsof market developments.



3

during these two years. Therefore, the profitability ofboth export-driven companies57 and firms operating onthe domestic market were examined, in order to decidewhether differentiation in the corporate sector can beattributed to the varying degrees of corporate forint-exposure.

Profitability of exporters and non-exporters

Chart 3-7 reveals that, throughout the period underreview, exporters were more profitable than non-exporters. Export profitability improved rapidly in thefirst half of the 1990s and in 1997, its culmination coin-ciding with accelerating growth in EU member statesand declining oil prices. In addition to an upturn in theglobal business cycle, foreign direct investment, whichpeaked during the period between 1997 and 1998, alsoboosted profitability.

The downward trend in profitability that followed is like-ly to have been closely related to the prolonged pass-through effects of the Russian crisis, as exports to Russiawere still considerable at that time. The differencebetween exporters and the companies targeting domes-tic markets diminished markedly in 2001, when theforint appreciated strongly. Since the forint exchangerate appreciation in 2001 coincided with declining eco-nomic growth in EU member states and resultant slug-gish demand for Hungarian tradables, the impact ofthese two factors on the profitability of exporters can-not be distinguished separately.

The fact that in 2002, despite a strong forint exchangerate, export profitability decreased to a lesser degreethan when the forint appreciated, was a favourabledevelopment.

An examination of the various types of costs and oper-ational income, or rather changes in the various types ofcosts as a proportion of sales revenues and the opera-tional profit margin (see Table 3-6 and Table 3-7) isinstrumental in identifying the differences between thetwo groups of companies. The 2002 data reveal thatcompanies targeting domestic markets continued toimprove their profit margin. This improvement occurreddespite the fact that labour costs rose even moredynamically than sales revenues. Wage increases aloneresulted in a 1-percentage-point reduction in the opera-tional profit margin, which was, however, offset byfavourable developments in material-related expenses.

If it is postulated that a consistently strong forint exchangerate had a major adverse impact on export profitability,this should be reflected in an increase in the proportion offorint-denominated cost categories. Of these compo-nents, the most important are labour costs, material-relat-ed services and the value of subcontractors’ performance.At the same time, however, due to the high import con-tent of the materials used in manufacturing, the FX-denominated component (costs of material) of material-related expenses may also be significant. Hence, a strongexchange rate may even have a beneficial impact on prof-itability through this cost component.

Table 3-6 shows that one of the causes behind the slightdeterioration in export profitability was a decline in salesrevenues. The setback in sales reflects the impact onprofitability of a longer-term downturn in the global busi-ness cycle. Developments in labour-related expenseswere another cause. Although the aggregate wage billremained unchanged in 2002, this does not mean thatwage dynamics in 2001 and 2002 failed to have animpact on exporters. On the contrary, it was fast wagegrowth that forced exporters to cut staff as early as2001. The number of layoffs continued to rise in 2002(see Chart 3-8). As a result, this cost component reducedthe profit margin by one percentage point.

Material-related expenses, another major cost compo-nent, however, affected export profitability beneficiallyin 2002. Material-related expenses fell more rapidlythan sales revenues, which added 2 percentage pointsto exporters’ profit margin. This might reflect the posi-tive effect the strong forint exchange rate had on mate-rial costs of exporters with a high import content. Theconflicting effects of these two cost categories leftexporters’ operating profit margin, i.e. operating costefficiency unchanged in 2002.

Profitability in manufacturing

In the next step of the analysis, we investigated howprofitability developed in the individual sectors in man-

57 Export-driven companies include ones where more than 20% of sales stems from exports.

Profitability (ROA) of exporters and non-exporters (1993–2002)

Chart 3-7

0

2

4

6

8

10

12

14

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Per cent

Exporters Non-exporters



3

ufacturing, and to what extent changes in profitabilitycan be attributed to changes in different cost cate-

gories. To investigate this, manufacturing sales revenuesand operational profits were compared. Sectoral datasuggest that both sales and export revenues fell afterthe appreciation of the forint exchange rate in 2001(see Table 3-8). In turn, the sector was unable to achievean increase in sales revenue last year. The same holdsfor export revenues.58

According to detailed data (Table 3-9), the nominalvalue of sales revenue in manufacturing was 5% lowerin 2002 than in 2001. This can be attributed mainly tothe fact that sales revenues in Machinery and equipment(representing the greatest weight on the basis of opera-tional income) fell significantly. As 2002 saw a down-turn in the business cycle, this decline in sales revenuesin Machinery and equipment came as no surprise, sincemotor vehicle manufacturing is considered to be one ofthe most cyclical industries in the world.

In contrast to the decrease in sales revenues indicat-ing adverse developments from the output side, cor-porate income from operational income did not dropoff in manufacturing. Operational income in manufac-turing, with MOL excluded, increased, albeit onlymoderately in 2002 following a decrease in 2001.This moderate growth rate was the combined resultof different performances at the sub-sectoral level.

In order for the causes of improving profitability in theface of falling sales revenues in 2002 to be identified,developments in the profit margin of the individual sub-sectors were examined and the contribution of the var-ious cost categories to changes in the profit margin in2002 was quantified (see Table 3-10).

Reviewing the data on profit components comparedwith data on profitability clearly shows that staff-relatedexpenses, one of the major components, adverselyaffected the profitability of almost every sub-sector.

Changes in real wages* and average number ofemployees at export companies

Chart 3-8

–15

–10

–5

0

5

10

15

20

1999

2000

20

01

20

02

Per cent

Real wage Employees

* Calculated on the basis of PPI.

58 Consolidated data somewhat modify our view on the developments in manufacturing in 2001. Relative to what was stated in the Report in December 2002, the cur-rent report finds a minor change affecting the textile industry, where, based on final data, both sales revenues (3%) and export revenues (8%) grew less dynamical-ly in 2001. (Last year’s data reveal growth in excess of 10%.) Contrary to an anticipated slight increase, operational income actually fell by 3% (it increased by 7%.).Last year’s assessment for the chemical industry has also been modified. It claimed that the operational income of the chemical industry fell markedly in 2001.Contrary to this, the profit calculations relating to the chemical industry, excluding MOL, reflected a reassuring growth rate of 9% in 2001. Accordingly, the overallprofit in manufacturing only fell by 15%, rather than by 25%.

59 The 2001 operating MOL profit, unusually low and negative, and also considered as abnormal, led to unusually low profits in the chemical industry in 2001. In 2002,profits both at MOL and in the chemical industry were back to ‘normal’. As a result, the rate of growth in the chemical industry including MOL was quite dynamic(28%) in 2002.

Changes in exporters and non-exporters’ profitsand cost categories

Changes Operational Net sales Material-re-(2002/2001) profit lated expenses

Exporters 0.90 0.90 0.88Non-exporters 1.23 1.07 1.05

Staff-related Other Depreciationexpenses expenses

Exporters 0.99 1.06 1.10Non-exporters 1.13 1.03 1.13

Table 3-6

Contribution of the individual cost categories tochanges in the profit margin*

Changes Opera- Material- Staff- Other Depre-(in percentage tional related related expenses ciation

points) profit expenses expenses

Exporters 0 2 –1 –1 –1Non-exporers 1 2 –1 0 0

* Due to rounding, the sum of the changes in items as a proportion ofsales revenues does not necessarily add up to changes in operationalprofit as a proportion of sales revenues.

Table 3-7

Operational income and growth rates of salesand export revenues59 excluding MOL (percent)

Net sales Export sales Operationalrevenue revenue income

2000 24 27 232001 16 22 –152002 –5 –11 3

Table 3-8



3

Wage growth was felt most acutely in the textile indus-try, which is hardly surprising, as, compared to othersub-sectors, wage costs represent the highest propor-tion (30%) in sales revenues there (see Table 3-11). This

adverse effect was amplified by the appreciation of theforint exchange rate and the protracted sluggishness ofthe global business cycle, since, based on its export rev-enues, the textile industry is one of the most export-driv-en sub-sectors (see Chart 3-9).

Representing the greatest weight among the cost com-ponents, material-related expenses, or rather, theirratio to net sales revenues improved the profitability ofevery sub-sector but the textile industry. The largestimpact was felt in machinery and equipment, whichrecorded the highest export revenues. Although bothexport and domestic sales revenues fell more thanaverage in this sub-sector, its profits increased. Thisincrease can unequivocally be attributed to afavourable, 4 percentage-point change in material-related expenses.

In both machinery and equipment and the food indus-try, the decrease in material-related expenses as aproportion of revenues can be attributed tofavourable changes in material costs and the cost ofgoods sold. These two types of costs contain thoseinput costs that can be favourably affected by astronger forint exchange rate, i.e. import-related mate-rial costs. As both of these industries use importedraw material to a large extent, the beneficial effects ofthe appreciation of the forint exchange rate cannotbe ruled out.

Of the remaining components, only the value of capi-talised own performance as a proportion of sales rev-enues can have an economic interpretation. The reasonfor that is that it reflects changes in self-manufactured

Sales revenues and operational income in manufacturing sub-sectors (Change, 2002/2001)

Net sales Export Opera- Weightrevenue sales tional in manu-

revenue income facturingprofit

(per cent,average1992–2002)

Food industry 1.04 0.98 1.20 15Textile industry 0.93 0.89 0.67 3Manufacture of paperand paper products 0.99 0.89 1.00 5Chemical industry 1.01 1.04 2.69 17Manufacture of non-metallic mineral products 1.07 1.01 1.03 5Metallurgy, manufactureof basic metals andfabricated metalproducts 1.01 0.99 0.73 13Machinery andequipment 0.88 0.84 1.13 40Other manufacturing 0.95 0.79 1.10 2Manufacturing 0.96 0.89 1.28Chemical industry excluding MOL 1.03 1.06 0.96Manufacturingexcluding MOL 0.95 0.89 1.03

Table 3-9

Contribution of the individual components of operating income to changes in the profit margin(Change 2002–2001, in percentage points)*

Change Material- Staff-related Value of Other Depreciation Other Profit(2002–2001) related expenses activated expenses revenues margin

expenses ownperformance

Food industry 3 –1 –1 –1 0 1 1Textile industry 0 –1 –1 0 0 2 –1Manufacture of paper and paper products 1 –1 0 0 0 0 0Chemical industry 4 0 –1 2 0 0 4Manufacture of non-metallicmineral products 1 0 –1 1 0 –1 0Metallurgy, manufacture ofbasic metals and fabricatedmetal products 1 –1 –1 0 0 0 –1Machinery and equipment 4 –1 0 –1 –1 1 1Other manufacturing 1 –1 0 1 0 –1 1Chemical industry excluding MOL 1 –1 0 1 0 –1 –1Manufacturing excluding MOL 3 –1 0 –1 0 1 0

* Due to rounding, the sum of the changes in items as a proportion of sales revenues does not necessarily add up to changes in the profit margin,i.e. operational profit as a proportion of sales revenues.

Table 3-10



3

stocks. Available data show that changes in the value ofcertain sub-sectors’ capitalised own performancereduced the profits of those sub-sectors, suggesting thatcompanies adjusted to the bleaker growth outlook. Thisadjustment took place through a reduction of self-man-ufactured stocks. However, this response was confinedto sub-sectors with a smaller weight, and thus this cate-gory had a neutral impact on the profitability of manu-facturing in 2002.

INDEBTEDNESS

Leverage, i.e. the debt-to-assets ratio, was used toanalyse developments in the indebtedness of the non-financial corporate sector.

In principle, the most obvious source of an increase inleverage is enhanced corporate investment activity.Generally, companies rely on external sources of funds,mainly bank loans, to fund their investment. The reasonsfor this are that, on the one hand, funding through shareissuance is costlier, and that, on the other hand, their owndisposable funds usually fail to cover the necessary invest-ment costs. The increase in leverage in the entire sectorwas most rapid between 1997 and 1999 (see Chart 3-10).As this was a period of general economic recovery, suchan increase in leverage was no surprise. The upturn in eco-nomic activity also affected the entire corporate sectorbeneficially: corporate investment grew the most vigor-ously in these years. Sectoral data corroborate this as well.Chart 3-11 shows clearly that leverage increased in eachsub-sector between 1997 and 1999.

Leverage in the corporate sector has stopped increasingover the last two years. In fact, in 2002 indebtednesseven declined slightly. The lengthy downturn in thebusiness cycle offers an explanation for this develop-ment: companies decided not to launch new invest-ment projects and not to apply for further loans undersuch circumstances. As a result, the entire sector took anet saving position in 2002.

Detailed data reveal that indebtedness changed differ-ently across sub-sectors in 2002. A slight decrease inaggregate leverage can mainly be attributed to manu-facturing. This suggests that the majority of manufactur-

60 The ratio of the other cost categories was 3 to 7% in the period under review.

Ratio of export sales revenues of manufacturingsub-sectors to net sales revenues (1993–2002)

Chart 3-9

Food industry Textile industry

Manufacture of paper and paper products

Chemical industry Manufacture of non-metallic

mineral products

Metallurgy, manufacture of basic metals

and fabricated metal products

Machinery and equipmentOther manufacturing

0

10

20

30

40

50

60

70

80

90

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Per cent

Leverage of non-financial corporations (Debt-to-assets ratio, 1994–2002)

Chart 3-10

0

10

20

30

40

50

60

70

1994

1995

1996

1997

1998

1999

2000

2001

2002

Per cent

Average ratio of material- and staff-relatedexpenses to sales revenues60 (1992–2002) (Per cent)

Material- Staff-relatedrelated expenses

expenses

Food industry 79 11Textile industry 60 30Manufacture of paperand paper products 68 16Chemical industry 67 10Manufacture of non-metallic mineral products 62 19Metallurgy, manufacture ofbasic metals and fabricatedmetal products 72 15Machinery and equipment 74 14Other manufacturing 70 20

Table 3-11



3

ing companies, in an effort to adjust to the slowdown inthe economy, modified their financial structure to beable to continue paying off their debts even from theirreduced income. At first sight, the slight decrease in theleverage of the construction industry may look strange.Although construction activity has been extremely buoy-ant over the past two years, its indebtedness did notgrow. This, however, does not mean that the industrydecided not to rely on external sources of funds. Thishas more to do with the fact that the growing earningsof these companies over the last two years have result-ed in a simultaneous increase in their capital strength.

The indebtedness of the services sub-sectors increasedsomewhat in 2001 and 2002. The underlying reason forsuch increase was that consumption expenditure onservices grew faster than aggregate consumption.Consumption was boosted by wage dynamics, as wageincrease in both 2001 and 2002 reached proportionsunprecedented since 1991. Despite upward trends inindebtedness, services providers’ indebtedness is by nomeans high compared to average indebtedness overthe whole period. However, consumption is expectedto slow down considerably in 2003 and 2004, whichmay lead to lower earnings in the services sector. If serv-ices providers fail to adjust their financial structure tothis downturn, a further increase in leverage may easilyresult in stability problems in the future.

In terms of stability, developments in the leverage of lowprofitability companies provides further food for thought.The average leverage of the least profitable companies(the bottom 20% based on ROA) is much higher thanthat of the entire corporate sector (Chart 3-12).

Higher leverage does not, however, mean that thesecompanies rely on external sources of funds more heav-ily. As described in the section on aggregate profitabili-ty, these companies generated losses throughout theperiod under review. Consequently, higher leverage ismore likely to suggest that these companies, more oftenthan not, finance their losses from external funds. This,however, by no means suggests that any company inHungary can remain in business in the face of marketcompetition for years without generating profit. Thechanging composition of the sample reveals that quitea few companies went out of business during the peri-od under review. On average, 10 to 15% of the compa-nies went out of business every year between 1992 and2002. At the same time, however, in 1998, when corpo-rate profitability was at its peak, even the least profitablecompanies increased their equity considerably. The bot-tom 20% of the companies account for approximately10 to 15% of the overall sectoral liabilities, which sug-gests that small companies are dominant among theleast profitable firms.

The flow equivalent of the debt-to-assets ratio is theinterest coverage ratio.

Interest coverage ratio = ordinary operatingprofit61/interest paid and interest-related disbursements

The interest coverage ratio denotes the extent towhich earnings generated through ordinary businessactivities can cover interest liabilities. In principle, prof-itability, corporate borrowing rates and changes inleverage exert a combined effect on changes in inter-est coverage.

61 Ordinary operating profit = Profit before tax and interest payments.

Sub-sectoral leverage indicators (1994–2002)

Chart 3-11

Energy & waterConstructions

Manufacturing Services

Agriculture & mining

0

10

20

30

40

50

60

70

80

90

19

94

19

95

19

96

19

97

19

98

19

99

20

00

20

01

20

02

Per cent

Leverage of the least profitable companies andthe weight of their overall liabilities in the overallliabilities of the entire sector (1993–2002)

Chart 3-12

LeverageLiability ratio

0

10

20

30

40

50

60

70

80

90

19

93

19

94

19

95

19

96

19

97

19

98

19

99

20

00

20

01

20

02

Per cent



3

Of these factors, profitability and borrowing rates werethe most decisive for Hungarian companies. Interestcoverage grew steadily between 1992 and 2002 (seeChart 3-13). Prior to 1998, improved aggregate prof-itability affected interest coverage beneficially, whichwas further boosted by falling interest rates after 1995(see Chart 3-14). Since then, however, declining interestrates have been at the forefront, as the profitability ofthe entire corporate sector has been deteriorating sincethe late 1990s.

Detailed data suggest that the dominance of both prof-itability and interest rates varies somewhat across sub-sectors (see Chart 3-14 and Table 3-12) and was differ-ent between 2001 and 2002. Interest coverage in theservices sector, currently the engine of economicgrowth, was affected beneficially by both profitabilityand interest rates in both years. Likewise, interest cover-age in the construction industry has been strengtheningfor years despite the fact that it is the most heavilyindebted sub-sector. This is explained by the steadyupward trend in the profitability of construction compa-nies. The earning capacity of this sub-sector was at his-torical highs in 2001 and 2002. In 2002, companies inthe energy and water sector suffered from falling prof-itability and a simultaneous increase in leverage, which

led to a deterioration in interest coverage. By contrast,in 2001, the beneficial effect of falling interest rates inmanufacturing was cancelled out by deteriorating prof-itability. Improved performance and repeated decreas-es in the aggregate manufacturing debt burdenundoubtedly played a major role in further improve-ment of the aggregate situation in 2002.

From the perspective of financial stability, leverage posesthe greatest danger to companies whose cash flow fromordinary business activity fails to provide sufficient cover-age for their interest expenses. The reason for this is thatthese companies would be able to meet their interest lia-bilities only if they further relied on external credit, which,in the long run, is unsustainable and leads to insolvency.Therefore, close attention should be paid to such compa-nies whose interest coverage indicator is below one.Hereafter, we refer to these as unsafe companies.

Aggregate interest coverage (1993–2002)

Chart 3-13

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Weighted average corporate borrowing rates withmaturity over one year (1993–2002)

Chart 3-14

0

5

10

15

20

25

30

35

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Per cent

Sub-sectoral interest coverage indicators (1993–2002)

Chart 3-15

ConstructionEnergy & water

Agriculture & mining Manufacturing

Services

01

234

56

78

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Year-on-year changes in the components ofinterest coverage in the individual sub-sectors(2001 and 2002, changes)

Change Agricultu- Manufac- Energy Construc- Services(year-on- ture & turing tionyear) mining

Ordinary operational profit

2001 1.29 0.92 0.79 1.65 1.232002 1.01 1.18 0.93 1.12 1.16

Interest-related disbursements

2001 0.93 1.03 0.93 1.18 1.052002 0.95 0.98 1.10 0.99 1.00

Overall effect

2001 1.39 0.90 0.84 1.40 1.172002 1.07 1.20 0.85 1.13 1.16

Table 3-12



3

In respect of these companies, their leverage grewfaster than aggregate leverage over the period underreview. As a result, their indebtedness in 2001 and2002 considerably exceeded the corporate average(see Chart 3-16) Such companies invariably generatedlosses.62 Their weight within the entire sectorremained around 15% in the second half of the1990s.

On average, every year roughly 20% of unsafe compa-nies are not present in the sample in the consecutiveyear. This, in turn, suggests that approximately 3% ofthe overall corporate sector ceases to be competitivedue to interest payment-related financial difficulties. Onaverage, 40% of the unsafe remain under threat foranother year. This points to the lack of significant per-sistence in this group of companies, since 40% of theunsafe manage to overcome financial difficulties fromone year to the next.

From a stability perspective, the fact that a remarkablenumber of these companies can take control of theirtemporary financial problems is a favourable develop-ment. However, the difficulties of adjustment are amplyproved by the following: as within the unsafe group theproportion of the companies that face permanent diffi-culties remains permanently stable, around 40%, theproportion of those that have experienced no financialdifficulties beforehand is also considerable every year.Putting all this together, this means that despite of sta-ble proportions, the composition of the group of com-panies under threat varies considerably from year toyear.

LIQUIDITY

When a company is heavily indebted and the value of itsassets falls rapidly, its liquid assets may prove insufficient tocover the current liabilities, even if the proportion of suchassets is high. Yet, if the company manages to increase theproportion of its most liquid assets on its balance sheet,this reduces the risk of insolvency significantly.

In order for company liquidity to be analysed, liquidity andcash ratios were used. These ratios denote the extent towhich liquid assets can cover short-term liabilities.

Liquidity ratio = (Receivables + Securities + Cash) /Short-term liabilities.63

Money ratio = (Securities + Cash) / Short-term liabilities.

Aggregate liquidity ratio reveals that the liquidity posi-tion of the entire sector improved further in 2002,though the most liquid assets only covered an approxi-mately 90% of short-term liabilities (see Chart 3-17).

In contrast with what was outlined in connection withthe aggregate situation, the fact that the liquidity posi-tion of the least liquid companies failed to improve in2002 was an unfavourable development (see Chart 3-18). Should such companies face bankruptcy, theywould only be able to meet approximately one-third oftheir current liabilities, using their most liquid assets.

A similar conclusion arises when the numerator onlyconsists of liquid assets and securities (cash ratio). Thecash ratio reveals that the cash and securities stock of atypical Hungarian company together can cover 20 to

62 As a result, their interest coverage indicator is negative.63 Short-term liabilities: short-term loans, short-term credit, liabilities arising from services and transport of goods, inter-company loans and other short-term liabilities.

Weight and leverage of companies with the low-est interest coverage* (1993–2002)

Chart 3-16

0

5

10

15

20

25

30

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Per cent Per cent

0

10

20

30

40

50

60

70

80

Ratio of companies with low interest coverage, left scale

Leverage, right scale

Liquidity ratio in the corporate sector (1992–2002)

Chart 3-17

78

80

82

84

86

88

90

92

94

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Per cent

* Where the interest coverage ratio is below one.



3

Median cash ratio (1992–2002)

Chart 3-19

0

5

10

15

20

25

30

35

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Per cent

The lower quintile in the distribution of theliquidity ratio (1992–2002)

Chart 3-18

0

5

10

15

20

25

30

35

40

45

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Per cent

25% of short-term liabilities, which is by no means lowby international standards.64

64 The cash ratio was around 20% in the United Kingdom in the second half of the 1990s. (Bank of England, Financial Stability Review, June 2000, page 88, Chart 6).

4 ARTICLES


4 ARTICLES

4

INTRODUCTION

Forint band widening on 4 May 2001 and subsequentforeign exchange liberalisation have led to significantchanges in the Hungarian foreign exchange market. Theeffect of these changes on exchange rate movementsand on the behaviour of market participants has beenparticularly observable in the past one year: both theoutstanding total of, and turnover in FX derivatives65 forshort-term speculation or risk management haveincreased robustly, as well as their influence on theforint market.

Up to June 2001, Hungarian regulations on forint/forexderivatives transactions conducted by resident and non-resident market participants were extremely tight.66

With foreign exchange liberalisation in June 2001, how-ever, these restrictions were lifted, meeting the require-ments associated with the wide exchange rate band. Asa consequence, the markets under review experiencedan immediate upsurge. FX swap transactions accountedfor the vast bulk of this marked rise. The size of theforint swap market grew continuously. Today, this is themost liquid segment of the forint/FX market, with aver-age daily turnover several times higher than turnover inthe spot market.

Band widening has also triggered an upswing in anoth-er market segment, which may be less significant interms of trading volume, but is of key importance formonetary policy by virtue of its information content.Forint/foreign currency options provide market partici-pants with the opportunity to bet on and trade in devel-opments in exchange rate volatility (in addition to tak-ing open exchange rate positions). Developments inthis partial market carry information for monetary poli-cy decision-makers which places options in the centreof attention, despite the lower liquidity.

One common feature of FX swaps and options is thattheir turnover increased significantly immediately before

and after episodes of turbulence in the forint market inthe past 18 months. Based on this observation and othermarket information, these two derivative products havebecome the major tools of speculation on the forintexchange rate and its expected volatility by foreign par-ticipants who dominate the domestic forint market.

The primary purpose of this article is to analyse (i)whether the use of derivatives carries risks to the domes-tic banking sector; and (ii) the extent to which therecording of derivative products off balance sheet influ-ences banks’ open forex positions and, through this,banking sector’s stability. In addition, the article presentsthe important role of, and the underlying reasons forusing, these two segments of the foreign exchange mar-ket in the transformation of the forint market structure.

First, we present FX swaps and options from a technicalviewpoint, then analyse the structural changes in theforint market since the band widening and liberalisation,concentrating on the importance of derivatives transac-tions, and provide a description of their role played dur-ing turbulent periods in the forint market. Then we dis-cuss how monetary policy can use information deriv-able from these markets and how interest rate policyinfluences prices in these markets. Finally, we deal withthe implications for financial stability and point outopen positions taken by banks and their impact onexchange rate volatility.

FX SWAPS AND OPTIONS

FX swaps

Swaps are an exchange of current and future cash flows.Currency swaps, in particular, are conversions betweentwo counterparties of a cash flow denominated in two dif-ferent currencies. The simplest form of these is an FX swap.

An FX swap is a conversion agreement consisting oftwo legs. It involves the exchange of principal denomi-

4.1 CSABA CSÁVÁS AND GERGELY KÓCZÁN: DEVELOPMENT OF THE

HUNGARIAN DERIVATIVES MARKET AND ITS EFFECT ON FINANCIAL

STABILITY

65 Derivative: a transaction the value of which depends on changes in the price of the underlying product (in this context foreign exchange) of the derivative transac-tion.

66 Except stocks and stock index derivatives, residents were not permitted to deal in derivatives with non-resident investors, which made it difficult to settle their deliv-erable forint derivative transactions with each other as well. Consequently, residents’ off-balance sheet activities were confined to forward transactions with thedomestic banking sector, while non-residents could use the non-deliverable forward market centred in London for their transactions with each other.


4 ARTICLES

4

nated in two different currencies by two counterpartiesand an agreement to swap back principal at a pre-deter-mined price and a future date (maturity). An FX swaptherefore is equivalent to a spot forex transaction cou-pled with a forward forex agreement. This implies thatthe parties to the swap transaction simultaneously buyand sell the currencies in question, and so they do notopen forex positions and, consequently, they do notexpose themselves to a situation in which a potentialshift in the exchange rate of the two currencies couldcause profit or loss to them.

In the literature, FX swaps are categorised into thebroader class of forwards; however, they are distin-guished from outright forwards for two main reasons: (i)they involve cash flow in the present (on the settlementday of the spot leg); and (ii) counterparties do not openforex positions (the direction of the spot leg is opposite,and equal in amount to, the direction of the forwardleg). In another approach, a swap is a forex repo, i.e. aloan received/granted against foreign currency as collat-eral, the market of which is generally highly liquid,67 andtherefore is a major financing tool available for banktreasury departments active in the forex market.68

Major money market strategies related to FX swaps

Acquiring financial assets financed by an FX swap

This is a foreign currency investment with an automatichedge, whereby the buyer enters into a swap agreement,on the spot leg of which he receives the currency in ques-tion and invests the amount in a financial asset (e.g. gov-ernment paper). The forward leg of the swap provides(automatic) hedge against the exchange rate risk of theinvestment and so an FX position is not opened. The netreturn (interest rate differential) on the acquired asset isoffset by the implied rate of the swap, which is paid bythe buyer and is equivalent to the interest rate differential.In this case, one reason for entering into the agreementmay be opening an interest rate position. If the swap andthe acquired asset do not have matching maturities, aprofit or loss may result from the interest rate exposurecaused by the different duration of the asset and liabilitysides in the event of a shift in the yield curve.

Taking a spot forex position financed by an FX swap (synthetic forward position)

By establishing a spot forex position financed by an FXswap, the market participant (usually a speculator) usesthe more liquid spot market instead of the outright for-

ward segment to open a position. However, given thatthe entity does not want to acquire an asset denominat-ed in the currency in question (does not want to see achange in its balance sheet), it enters into a swap trans-action financing the position, whose spot leg is oppositeto the original spot transaction. Thus, the market entity’snet position is limited to the forward leg of the swap,which therefore has the same direction as the underlyingspot transaction (i.e. the desired direction in which theposition is opened), while there is no cash flow on thespot leg and the balance sheet remains unchanged.Practically, this means entering into a synthetic forwardposition. Compared with an outright forward, this tech-nique has the advantage that the swap market is muchmore liquid and more flexible in respect of maturities.Closing the position requires either entering into an iden-tical transaction pair (spot + swap) but in the oppositedirection, or simply conducting a spot deal opposite tothe original position and letting the underlying swapexpire.69 The advantage of this technique lies not only inhigher liquidity and protecting the balance sheet, butalso in that the maturity of the synthetic position, i.e. thesize of interest rate exposure undertaken using the forexposition, can be ‘tuned in’ at discretion. The position canthus be rolled over or closed at will at the desiredmoment, due to the very high liquidity of the market.70

Currency options

A currency option is the right to purchase or sell a for-eign currency in the future at an exchange rate (strikeprice or exercise price) pre-determined by the contract-ing parties. For this right, the holder of the option has topay the option premium at the time of entering into theagreement. In the case of call options, the stronger thecurrency relative to the strike price the higher the prof-it; however, any loss is limited to the option premiumalready paid. In the case of put options, an oppositemove in the exchange rate earns a profit, but any poten-tial loss is also limited for the holder of the option.

In over-the-counter (OTC) markets, quotation of optionsis standardised. Their strike prices may be of any size;however, so-called ATM options are used the most fre-quently. They are different from other options in thattheir strike price is equal to the forward exchange ratewith an identical date of expiry, and so on the expirydate the options will be at the money if the spotexchange rate equals the strike price.71

It often happens that, instead of the option premium,only implied volatilities are quoted for within-year

67 For example, in Hungary it is much more liquid than the market of unsecured loans.68 For a stylised diagram of the effects of swaps on and off balance sheet, see the Appendix.69 In principle, it may happen that the transaction is closed by entering into an outright forward instead of a spot deal, but this occurs much less frequently.70 The most obvious method of using this tactic is the tom/next rollover, which creates an overnight synthetic forward position, whose interest rate exposure is mini-

mal, only representing exchange rate exposure (foreign currency position). On the next day, by entering into another swap, the spot leg of the transaction will financethe forward leg of the first. By renewing a tom/next swap every day, the synthetic overnight forward can be rolled over to the desired maturity.

71 Actually, these options are called at-the-money forward (ATMF) options, however, in the following the name ATM option is used for the sake of simplicity.


4 ARTICLES

4

expiry dates, which express the expected annualisedstandard deviation of the exchange rate move. Theonly factor influencing the option premium which isunknown at the time of undertaking the transaction isfuture volatility of the exchange rate. Therefore, this isthe variable which depends on market judgement andbrings supply and demand into equilibrium in theoptions market.

One of the major assumptions used in pricing ATMoptions is that, upon expiry of the option, the expectedprice will be equal to the forward exchange rate. Inother words, the expected value of the exchange ratemove equals the interest rate differential. Implied volatil-ity measures the distribution of returns (exchange ratechanges) against the forward exchange rate, assuming asymmetrical distribution.

One characteristic feature of these products is that, rel-ative to options with different strike prices, they havethe largest convexity value.72 This means that the valueof such options is the most sensitive to changes involatility. With the passage of time, ATM options tendto behave differently from other options – as the expirydate draws near, these options offer the greatest proba-bility to exercise them. Owing to these characteristics,ATM options play a dominant role in options markets.

One way in which the effect of options markets onexchange rates is manifested is the dynamic hedging ofoptions. Market-making banks both buy and sell curren-cy options and, if the resulting open position is otherthan zero, they must hedge it. If, for example, a largenumber of participants take long forint positions in themarket, the bank writing the option may hedge them bypurchasing forints in the spot market. This influencemay appear simultaneously with entering into theoption contracts, but, as the value of options reacts lesssensitively to a given exchange rate change than theprice itself (the average delta is lower than 1), it gener-ates lower trading volumes in the spot market relativeto the total notional value of option contracts. The typeof hedge provided against exchange rate risk, known asdelta hedge, may have a considerable effect on the spotexchange rate, particularly as the expiry date approach-es. The build-up of option positions in large amounts,provided that they have the same expiry dates andstrike prices, may have the consequence that the strikeprices of the options become technical (support orresistance) levels prior to expiry. If, for example, theforint strengthens close to the strike price of a calloption on the expiry date and, moreover, crosses it,then the appreciation may accelerate as a consequenceof hedging the position, similarly to the breach of animportant resistance level by the exchange rate.

Major currency option strategies

Straddle, strangle: Using at-the-money (ATM) options,positions can be built which are suitable for speculat-ing on future increases or reductions in exchange ratevolatility. By establishing a straddle composed of acall option and a put option with the same strikeprices, an entity locks in a profit from either the weak-ening or strengthening of the exchange rate; if, how-ever, the exchange rate remains near the strike price,associated with low volatility, the options will expirewithout the holder exercising them.73 Another advan-tage of such an option strategy is that an increase involatility adds to the value of both options, and so itmay even earn a profit for the holder before expiry.By contrast, investment banks usually offer to selloptions of this type when many market participantsexpect exchange rate volatility to remain low and,consequently, this strategy may contribute to stabilis-ing exchange rates.

Risk reversal (RR): a complex option strategy whereby along call for a currency is combined with a buying obli-gation (short put). The expiry dates of the two optionsare the same, with their nominal values also being iden-tical. Option premiums are also equal, so the strategyhas zero cost, but the strike prices are different.Providing that the investor holds his position untilexpiry, he will profit if the euro exchange rate strength-ens towards the higher strike price of the call option.Conversely, he will incur a loss if the euro weakensbelow the lower price of the put option – if it remainsbetween the two strike prices, then his position will beneutral (see Chart 4-1).

A risk reversal position mainly resembles a forwardagreement. It is suitable for hedging against exchangerate risk and it only costs the difference between thecall price and the put price included in the option pre-miums. At times of wide exchange rate oscillations,exposure to exchange rate risk is eliminated, with anunhedged position only remaining in the face of narrowexchange rate movements. Hence, the other name ofthe position – range forward.

One characteristic feature of the position is that differentimplied volatilities are used to calculate the premiums ofthe options underlying it. The difference between theimplied volatilities of the two options, expressed in per-centage points, is called the risk reversal spread, andprices are quoted for this spread. When the volatility ofeuro call options is higher (‘euro call options arefavoured’), this means that the market attaches greaterlikelihood to forint depreciation than to forint apprecia-tion (relative to the forward exchange rate).

72 The convexity value is the part of the option premium which is determined by implied volatility.73 In the case of a strangle, the strike prices are different; however, the earlier remarks are true for this position as well.


4 ARTICLES

4

The risk reversal spread can also be interpreted as thepremium on a call option vis-à-vis the put option – if theprobability of euro appreciation is greater than that ofeuro depreciation, as investors expect it, then the callwill be worth more for them than the put. This expecta-tion of the market can be described by the asymmetric,skewed probability distribution of the exchange rate(change). For example, in the case of skewed probabili-ty distribution towards a stronger euro, the cumulativeprobability that the euro exchange rate will be higherthan the forward rate is more than 50 per cent.

The asymmetric distribution of the exchange rate derivesfrom the observation that, in the case of small currencies(e.g. the forint), large depreciation occurs more frequent-ly than appreciation of the same extent. Consequently,even under normal market conditions an asymmetrytowards a weaker forint is more characteristic of the dis-tribution of the forint exchange rate, the forward-lookingindicator of which is the risk reversal spread.

DEVELOPMENT OF THE DERIVATIVES MARKETS

The FX swap market

Since band widening, the structure of the forint markethas come considerably closer to that of developed for-eign exchange markets. This process has taken place ina similar fashion in the other important foreignexchange markets of the region (for example, inPoland), and has entailed a very fast and massive rise inturnover in FX swaps. As a result, the swap market has

become the largest segment of the foreign exchangemarket in terms of volume, similarly to the situation indeveloped countries (see Chart 4-2).74

The size of the FX swap market has expanded continu-ously since foreign exchange liberalisation (June 2001),and in 2002 it registered a higher trading volume than thespot market (see Chart 4-2). Currently, the swap segmentis 2.5–3 times the size of the spot segment, based ongross turnover data. Looking at the market in a break-down by sector, Hungarian banks conduct 90 per cent ofswap transactions with non-residents (and exclusivelywith banks). In addition, there is also some turnover inthe domestic interbank market; however, turnover withdomestic non-banks is insignificant (see Chart 4-3).

According to market rumours, the forint swap market ismainly concentrated in Budapest. Consequently, in thevast majority of cases at least one of the counterpartiesis a Hungarian bank. This means that a relatively precisepicture of total market activity can be derived fromreports by commercial banks.75 It is mainly Hungarianbanks with a parent abroad, also dominating the othersegments, that are the most active in the swap market.Unlike in the case of spot transactions, the US dollar isthe major currency of the market, accounting for 90 percent of transactions. The maturity profile of the marketis dominated by maturities of less than one week, andby the tom/next maturity in particular.

Market participants’ behaviour in the forint swap market

Foreign participants occupied the swap market immedi-ately after liberalisation, as reflected by developments in

Profit and loss profiles of forward and risk reversal positions (EUR/HUF)

Chart 4-1

–20

–15

–10

–5

0

5

10

15

235 240 245 250 255 260 265 270 275

HUF HUF20

–20

–15

–10

–5

0

5

10

15

20

Forward position

RR at maturity RR before maturity

Gross turnover in the forint/FX market by segment (5-day moving average)

Chart 4-2

0100

200300400500

600700800

900

HUF billions

0100

200300400500

600700800

9001000

HUF billions1000

Jan. 01

Mar

. 01

May

01

Aug. 01

Oct

. 01

Jan. 02

Mar

. 02

May

02

Aug. 02

Oct

. 02

Jan. 03

Mar

. 03

May

03

Aug. 03

Spot Forward Swap

74 The turnover and stock data referred to in this section have been derived from commercial banks’ daily reports.75 No quantitative information is available on transactions of non-residents with each other and other domestic participants, as neither party reports the transaction to

the Bank.


4 ARTICLES

4

gross turnover. In addition to this indicator, the behav-iour of foreign participants dominating the swap marketcan also be characterised by their net outstandingforint/swap transactions76 (see Chart 4-4), which devel-oped very unevenly in the period under review.

The behaviour of foreign participants can be bestdescribed by the erratic changes in the net total ofswaps. Analysing the relevant times series, both majorstrategies are used in the forint market.

Spot + FX swap (synthetic forward strategy)

In the case of major episodes of turbulence in the forintmarket, accompanied by forint depreciation and anupsurge in spot market turnover, net outstanding swapsgenerally soar. There have been three major turbulentperiods since band widening: (i) the forint depreciationof July 2002; (ii) the speculative attack on the upperlimit of the intervention band in January 2003; and (iii)the exchange rate weakening linked to the devaluationof the forint band in June 2003.

During all three episodes, opening (and closing) substan-tial positions in favour of or against the forint (this beingthe cause of the exchange rate move) by foreign partic-ipants of the market was observable; however, theirforint assets did not change by the same measure,despite the fact that 90 per cent of the positions wereopened in the spot market. Explanation for this is that the

spot sales and purchases observed actually did not standalone, but constituted part of a spot + swap strategy.

The opposite developments in the spot and swap timeseries clearly indicate the active use of synthetic for-ward strategies (see Chart 4-4). This was particularly vis-ible in participants taking large positions, for example inthe period between January–early July 2003.

In January, the amount of forints purchased by foreignspeculators was so high that it was impossible to investin the government securities market simultaneously.Actually, speculators’ objective was not investing butopening long forint positions betting on the furtherstrengthening of the exchange rate. Therefore, in orderto avoid acquiring forint-denominated assets ‘out ofpressure’ on the settlement day, they reduced their netoutstanding swap positions to the extent whichabsorbed some half of all forint amounts purchased.Practically, foreigners established synthetic long forintforward positions. By doing so, foreign speculatorsmounted an attack on the upper limit of the band in halfby establishing synthetic forward transactions. Theyplaced the other half of forints purchased during theattack in short-term deposits with Hungarian banks.

At the time of the band shift in June, foreigners sudden-ly opened large-amount short forint positions, whichalso did not involve a similar change in their forint-

76 The time series for net swaps plotted on the Chart has been produced using the daily turnover data (D01) reported by Hungarian commercial banks to the MNBand the Supervisory Authority by cumulating, relying on the fact that, prior to liberalisation, non-residents were not allowed to transact in currency swaps. The timeseries, therefore, is made up of estimates rather than accurate stock data.

Gross FX swap turnover by sector (5-day moving average)

Chart 4-3

050

100150200250300350400450500

HUF billions

050

100150200250300350400450500550

HUF billions550

Jan. 01

Mar

. 01

May

01

Aug. 01

Oct

. 01

Jan. 02

Mar

. 02

May

02

Aug. 02

Oct

. 02

Jan. 03

Mar

. 03

May

03

Aug. 03

Non-residents Domestic banks

Other domestic banks

Non-residents’ net FX swap and spot transactions vis-à-vis Hungarian banks (Cumulative values since band widening; 4 May 2001 4 = 0)*

Chart 4-4

–200

0

200

400

600

800

1000

May

01

July

01

Sep

. 0

1

Dec

. 0

1

Feb

. 0

2

Ap

r. 0

2

July

02

Sep

. 0

2

No

v. 0

2

Jan

. 0

3

Ap

r. 0

3

Jun

e 0

3

Au

g. 0

3

HUF billions

Swap position Spot position

1200

* An increase in non-residents’ net forint swaps means that foreign par-ticipants buy forints spot using a swap conducted with the bankingsector and simultaneously sell them forward.


4 ARTICLES

4

denominated asset holdings (e.g. sales of governmentpaper). Through spot deals, they sold large amounts offorints; however, they did not want (or were unable) tosell their forint-denominated government securities inthe same amount, so they acquired the forints to settlethe sales by establishing FX swaps. Thus, in this casethey opened very short-term synthetic short forint for-ward positions (most of the underlying swaps were forthe tom/next maturity) which they could roll over to thedesired maturity.

Acquiring a financial asset financed by a forex swap

As was pointed out earlier, application of this strategy isuseful if speculation concentrates on changing theshape of the yield curve (interest rate speculation),instead of anticipating a particular move in theexchange rate. A good example of this is the very livelypurchases by non-residents of government paper inautumn 2002, which were accompanied by an upsurgein their net outstanding swaps. Presumably due toincreased expectations of an official interest rate cut,their forint exposure did not open to the extent towhich they built up their government securities hold-ings. The part of positions which could not be financedby spot purchases was financed by an increase inswaps, which in turn allowed for certain participants totake only interest rate positions, instead of undertakingexchange rate risks and to finance the purchase of along-dated government paper by establishing androlling over a short-term swap transaction.

The financed asset may not only be a governmentpaper, but also a reverse FX swap with a different matu-rity. The objective of this is also speculating on a changein the slope of the yield curve. This strategy can bereflected in the opposite shifts in non-residents’ shortand long-term swaps, the best example of it was theperiod following the devaluation of the interventionband in June 2003. Financing short-term swaps withlonger-term swap transactions suggested participants’mounting expectations of an official interest rateincrease (see Chart 4-5).

The forint options market

The domestic currency options market picked up some-what later than the swap market, as a genuine increasein turnover only started in mid-2002. Higher exchangerate volatility in July 2002 was a factor contributing tothis, in addition to liberalisation.

Turnover in options has been erratic in the past oneyear, with trading picking up particularly at times of sig-nificant exchange rate moves (see Chart 4-6). In 2003,daily average turnover in the options market hoveredaround EUR 25 million (HUF 6 billion) in the forint/forexsegment. At 70 per cent, banks conduct the larger part

of turnover with foreign participants. Turnover with resi-dents accounts for one-quarter of the total.

It may be useful to analyse the role options play in theforeign exchange market in an international compari-son. Options turnover in other Central and EasternEuropean currencies is higher than in the forint (dailyEUR 50 million in the Czech crown market and EUR100 million in the zloty market); however, the size ofthe spot segment in these markets is proportionatelyhigher. Options turnover accounts for 5–10 per cent ofturnover in the spot segments, which, although it is low,

Cumulative changes in non-residents’ net swapsat the less than 1-week maturity and over 1 week (Cumulative values from January 2003)*

Chart 4-5

Longer than 1 week Shorter than 1 week

–400

–200

0

200

400

600

800

1000

HUF billions

–400

–200

0

200

400

600

800

1000

1200HUF billions

1200

Jan.

03

Feb.

03

Mar

. 03

Apr

. 03

May

03

June

03

July

03

Aug

. 03

Sep.

03

Oct

. 03

* The sum of the two time series does not add up to the change in netswaps in 2003, as it does not include swaps conducted prior to 1January and maturing in 2003.

Gross options turnover by sector (5-day moving average)

Chart 4-6

0

5

10

15

20

25

30

35

Jan.

02

May

02

July

02

Sep.

02

Nov

. 02

Feb.

03

Apr

. 03

June

03

Aug

. 03

Oct

. 03

HUF billions

0

5

10

15

20

25

30

35

HUF billions40 40

Non-residents

Other domestic banks

Domestic banks


4 ARTICLES

4

is only marginally less than the proportions accountedfor by advanced options markets – the ratio of optionsto spot transactions is 10–15 per cent in the euro/dol-lar pair. Hence, the structure of the domestic optionsmarket is increasingly similar to developed foreignexchange markets.

The market is fairly concentrated, with four banks con-ducting three-quarters of total turnover. As in the swapsegment, banks with large foreign owners are active inoptions. The currency composition of turnover is similarto the spot segment: the vast majority of transactionsare conducted in the forint/euro pair.

As regards the maturity profile of the options market,one-third of the contracts entered into expire within amonth, although the ratio of those with maturities ofmore than three months is relatively high (45 per cent).This suggests that deals struck in the domestic optionsmarket are not exclusively speculative. One advantagein terms of financial stability is that maturities are notconcentrated around one particular day, so the expiryof options as well as their exercise does not lead tomajor disruptions in the foreign exchange market.

Since foreign exchange liberalisation, foreign investorsalso have had the opportunity to enter into option agree-ments for forint, not only with domestic banks but witheach other as well. Consequently, in contrast to theswap market, a foreign forint options market has evolvedwith a centre in London. Only limited information isavailable on turnover in the London market, however,the size of the market can be estimated on the basis ofinformation by foreign traders. This suggests that foreignturnover in options is 2–3 times higher than turnover inthe domestic market – average daily turnover in theLondon forint options market may amount to EUR50–100 million, although it may be substantially higherat times of strong exchange rate movements.

Firm conclusions can only be drawn regarding one seg-ment of the foreign market – options between domes-tic banks and non-residents. Domestic non-banks caneven transact in options directly with foreign banks,avoiding the domestic banking system. However, as thedomestic and foreign markets are not segmented, thedirect effect of the pick-up in turnover in the foreignoptions market is also reflected in the data reported byHungarian banks.

Market participants’ behaviour in the options market

It is mainly characteristic of options transactions bydomestic banks that they only intermediate deals andthat they close their open positions vis-à-vis their foreign

parent bank, by conducting options in the oppositedirection (so-called option pairs). As a result of this, res-idents and non-residents’ open positions arising fromoption contracts move in opposite directions (see Chart4-7). The overwhelming majority of domestic non-bankparticipants buy options which would earn a profit inthe case of forint appreciation, so presumably it isexporting firms’ hedging activity against exchange raterisk that lies in the background.77 Most of these transac-tions are initiated by domestic participants whosedemand is channelled through to non-residents bybanks. An explanation for this may be that banks activein the larger foreign options markets are able to hedgetheir positions more effectively.

Another feature of the options market is banks’ higherpositions vis-à-vis non-residents than residents, suggest-ing that Hungarian banks’ long open forint positions arefaced with a lower portion of non-resident short posi-tions in the forint (as expressed by the differencebetween the two charts).78 Banks must hedge againstthe resulting exchange rate risk in the spot market,which they can do conveniently, given the much higherturnover in the spot market than in the options market.

From spring 2002, domestic non-bank participants con-tinuously took long forint positions, in anticipation offorint appreciation. During the events of October–

Stock of foreign currency options by sector(Cumulated from May 2001)*

Chart 4-7

–250–200–150–100

–500

50100150200

Jan.

02

Feb.

02

Apr

. 02

June

02

Aug

. 02

Oct

. 02

Dec

. 02

Feb.

03

Apr

. 03

June

03

Aug

. 03

Oct

. 03

–250–200–150–100–50

050

100150200

HUF billionsHUF billions250 250

Non-residents Domestic non-banks

Domestic participants

* Net option positions, taking account of maturities. Positive valuesdenote long forint positions. The time series have been produced bycumulating daily turnover data, on the assumption that, prior to lib-eralisation, non-residents did not conduct forint options. As the dataalso include those for include American options as well, the total ofopen contracts may be lower than the values on the Chart.Consequently, the time series should be considered as estimates.

77 There may be domestic participants betting on forint appreciation in the background of transactions; however, the longer expiry of options suggests that hedgingagainst exchange rate risk may rather be in the background.

78 The motivating factor of non-residents’ behaviour may be that, in certain periods, they do not only hedge their forint assets with swaps but with options as well.


4 ARTICLES

4

November 2002, outstanding positions fluctuated, butto a considerably smaller extent than in the swap mar-ket. The trend continued after speculation against theintervention band ended, which may suggest hedgingactivities against exchange rate risk by residents.However, this trend turned around in the immediateaftermath of the band shift in June 2003, with neithernon-residents nor residents taking significant positionsamidst the more uncertain market conditions character-ising the period June–October.

FINANCIAL DERIVATIVES AND MONETARY POLICY

By establishing FX swaps, participants in the foreignexchange market have the opportunity to detach theirinterest rate position from their foreign currency posi-tion using the strategies discussed above, i.e. they areable to create positions whose profit is only determinedby exchange rate moves but not influenced by yieldmovements, and vice versa.

Currency options give participants the opportunity to ‘beton’ changes in expected volatility without opening anexchange rate position and to build complex positionswhich are variably sensitive to shifts in the exchange rate,yields and volatility using combined options products.

The role of derivatives in the transmission mechanism

Using financial derivatives, market participants mayshape their actual foreign currency positions independ-ently from their balance sheets, and thus the exchangerate may not only be determined by demand for bal-ance sheet items, for example, bank deposits or govern-ment securities denominated in the given currency.However, this does not weaken the central bank’s influ-ence on the exchange rate. Derivative positions arebuilt mainly for the short term, and thus the central bankhas the ability to influence directly the costs of, or prof-its on, holding such positions by influencing short-termyields (see Chart 4-8).

The return on a long position opened in the domesticcurrency is the carry, in addition to a potential move inthe exchange rate in the favourable direction. If thecarry is positive and high enough (in November 7.5 percent), it makes a long foreign currency position openedby the derivative transaction attractive (the carry isincluded in the settlement price). A positive carry, how-ever, has a cost for the participant opening a short posi-tion in the domestic currency. Moreover, the higher thecarry the more expensive it is to take a foreign curren-cy position against the domestic currency (for any typeof derivative transaction). For example, an official rate

increase makes it considerably more difficult to take ashort position in the domestic currency and easier totake a long position in the domestic currency, and so ittends to strengthen the exchange rate. This relationshipis particularly true for foreign currency positions estab-lished by using any type of derivatives.79

Economic theory and research have emphasised as oneof the positive effects of derivative positions that theycontribute to the transmission of short-term interestrates through the exchange rate channel as well, i.e.they strengthen the relationship between interest ratesand the exchange rate.80

If, however, market participants judge that the centralbank’s monetary policy has low credibility, then bybuilding large derivative positions, they can more eas-ily carry out an exchange rate correction against thecentral bank’s policy which they consider as desirable.This may in turn lead to declining central bank influ-ence.

Information content of derivatives markets

The market of financial derivatives not only improvestransmission between the various market segments, butcan also assist monetary policy in refining the picture ofmarket participants’ behaviour and expectations. Theactivities and yields evolving in the swap market maypresent a picture of the participants’ expectations par-ticularly in respect of short-term fluctuations in theexchange rate and the yield curve.

Two-week (major policy) rate of the MNB, andyields on three-month FX swaps and governmentsecurities

Chart 4-8

6

7

8

9

Mar

. 0

3

Ap

r. 0

3

May

03

Jun

e 0

3

July

03

Au

g. 0

3

Sep

. 0

3

Oct

. 0

3

Per cent

6

7

8

9

Per cent

Base rate 3-month FX swap

3-month benchmark

10 10

79 Naturally, this reasoning is only valid on the ceteris paribus principle – the effect of the positive interest rate differential may be offset by a number of other, mainlyendogenous, factors (e.g. well-defined exchange rate expectations, high anticipated volatility, etc.). In the case of currency options, this effect is more indirect; andthe central bank mainly influences developments in option prices with credible exchange rate policy, thereby reducing exchange rate volatility.

80 Vrolijk, Coenraad [1997].


4 ARTICLES

4

In addition, price developments in currency optionsmay reveal the market’s expectations, and thus theinformation content of option prices may be of particu-lar importance, despite the relatively poor liquidity ofthe market. In the following, we give a short descriptionof the opportunities for using currency options whichthe Bank currently monitors, in particular the variousinterpretations of implied volatilities deriving from ATMoptions.

Uncertainties in exchange rate movements

Volatility curves may be fitted to the implied volatilitiesrelating to various maturities, which can be used todraw conclusions about the future expected standarddeviation of the exchange rate (see Chart 4-9).

One tool used to illustrate volatility curves is the so-called option volatility cone which plots historical maxi-mum and minimum implied volatilities at different matu-rity ranges. Using this method, conclusions can bedrawn as to the relative amplitude of implied volatilities.

The highest implied volatilities were observed in June2003, which reflected the high uncertainty in the wakeof the band shift and official interest rate increases. Anearly horizontal volatility curve expresses the market’sexpectation of unchanged volatility in the period ahead.The lowest-level volatility curve in turn suggests veryfavourable, optimistic market sentiment.

Another method of revealing expectations related toexchange rate move uncertainties is to estimate: (i) theprobability that the exchange rate will exceed a certainlevel; and (ii) the probability of the exchange rate will bein a certain band when the options expire. A wider fluc-

tuation band pertains to higher implied volatilities. Forexample, during the period of stable exchange ratemovements in spring 2003, implied volatilities of theforint indicated that a month later the exchange ratewould be moving within a 1–2 percentage point bandwith a 50 per cent probability.

Credibility of the exchange rate regime

In an exchange rate system with a band, implied volatil-ities may supply information on the credibility of theexchange rate policy and the intervention band. If theexchange rate fluctuates near the limit of the band, thisindicates that the market attaches greater likelihood toa band shift or a large depreciation of the currency.Consequently, implied volatility may also indicate theextent to which the market considers the interventionband as credible. In the exchange rate system of theEuropean Union (Exchange Rate Mechanism, ERM I),increases in implied volatilities often reflected expecta-tions of realignments in central parities. In the case ofthe Italian lira and the pound sterling, for example,changes in option prices anticipated the currency crisesof 1992 several months before the crises actuallyoccurred. Based on information available to the Bank,options markets failed to predict the speculative attackagainst the forint band of January 2003 – volatilitiesonly increased simultaneously with FX intervention con-ducted at the band’s limit. Moreover, implied volatilitieswere at their lowest level to date during the days imme-diately preceding the intervention (on 12 January).

Estimating the risk premium

The risk premium on forint assets required by non-resi-dents is composed of a number of factors. Default riskis reflected well by the difference between the ratings ofgovernment securities and the yield spread on govern-ment bonds denominated in foreign currency. Liquidityrisk is low, due to significant turnover in the governmentsecurities and foreign exchange markets. The most vari-able component of the risk premium is exchange raterisk, which can be captured by the standard deviation ofthe exchange rate. However, historical volatility is notthe best indicator of exchange rate risk, due to its back-ward-looking nature.

By contrast, implied volatilities express expected futurefluctuation in the exchange rate or, in other words, themarket’s expectations related to exchange rate risk. Asonly actual interest rate differentials are observable inthe market, it is important that there be an indicatorwhich reflects expectations. Analyses have shown thatthe relationship between quoted volatilities and theforint exchange rate is very close – with an unchangedinterest premium, implied volatility (and the increase inexchange rate risk) generally leads to exchange ratedepreciation (see Chart 4-10).

Forint/euro volatility curve*

Chart 4-9

2

5

8

11

14

17

0 1 2 3 4 5 6 7 8 9 10 11 12

Maturity (month)

Per cent

2

5

8

11

14

17

Per cent20 20

12 Jan. 2003 27 June 2003 19 Aug. 2003

* Data for the period 12 January and 27 June denote the lowest andhighest values for 2003.


4 ARTICLES

4

DERIVATIVE TRANSACTIONS FROM FINANCIALSTABILITY PERSPECTIVE

One of the most important components of the stabilityof the financial intermediary system is the extent ofexposure to exchange rate risk. The extent of exchangerate risk depends in part on the size of open positionsin the various currencies and in part on the volatility ofthe forint vis-à-vis the various currencies. If the bankingsector has significant open foreign currency positionseither directly or indirectly, then strong oscillations inthe exchange rate may cause large losses to banks (seeChart 4-11).

During the episodes of market turbulence in 2003, bothhistoric and implied volatilities of the forint exchangerate soared. Nonetheless, the forint market stabilisedrelatively quickly, with implied volatilities approachingtheir earlier levels within a couple of months (see Chart4-12). The average volatility of around 8 per cent is notoutstandingly high in comparison with the more devel-oped currency markets – volatility of the forint/euroexchange rate is generally lower than the annualiseddeviation of the zloty/euro or US dollar/euro pairs(more than 10 per cent), and is only slightly higher thanthat of the Czech crown (6–7 per cent).

The episodes of sudden exchange rate depreciation inthe last 18 months, using derivative transactions, havebeen associated with large shifts in the position of thevarious sectors vis-à-vis banks. However, the domesticbanking sectors’ total (on and off-balance sheet) openposition has always remained at a minimum level (see

Chart 4-12). This has been required by the strict regula-tions on banks’ assumption of exchange rate risks. Inthe past, a regulation in force since 1992 set the upperlimit of exchange rate risk that banks were allowed totake at 30 per cent their regulatory capital.81 Althoughcompliance with the regulation is controlled on a dailybasis and sanctions could easily be imposed on partieswhich fail to comply, the regulation only restricted theinterest rate risk taken by credit institutions directly: itseffect did not extend to other financial intermediaries,such as investment firms owned by banks. This mayhave been the reason why, in the period preceding the

EUR/HUF implied volatilities for different maturities*

Chart 4-11

1-week 1-month 3-months 1-year

Jan

. 0

3

Feb

. 0

3

Mar

. 0

3

Ap

r. 0

3

May

03

Jun

e 0

3

July

03

Au

g. 0

3

Sep

. 0

3

Oct

. 0

3

Per cent Per cent

2

4

6

8

10

12

14

16

18

20

2

4

6

8

10

12

14

16

18

20

* Implied volatilities of ATM options are quotes from Reuters.

Forint open positions of the domestic bankingsector, domestic non-banks and non-residentssince band widening*

Chart 4-12

Domestic non-banks Domestic banks

Non-residents

–600–400–200

0200400600800

HUF billions

–600–400–200

0200400600800

HUF billions1000 1000

May

01

July

01

Oct

. 01

Jan.

02

Apr

. 02

July

02

Sep.

02

Dec

. 02

Mar

. 03

June

03

Aug

. 03

Nov

. 03

* Turnover data cumulated from 4 May 2001; positive values denotelong forint positions.

81 Ministry of Finance Decree 41/1996 ceased to have effect on 1 November 2003.

EUR/HUF implied volatility, the interest rate differential and the exchange rate

Chart 4-10

HUF/EUR exchange rate (right hand scale)

3-month implied volatility

3-month interest rate premium

6.5

7.5

8.5

9.5

10.5

11.5

Per cent

248

252

256

260

264

268

EUR/HUF

June

03

July

03

July

03

Aug

. 03

Sep.

03

Oct

. 03

Oct

. 03

27212.5


4 ARTICLES

4

Russian financial crisis, banks, while formally complyingwith the regulation on excessive position limits, held sig-nificant long open forint positions through their invest-ment firms in order to realise the interest premium ofthe forint. During the forint deprecation in the aftermathof the Russian crisis, the investment firms participatingin this speculation, and the parent banks indirectly,incurred heavy losses on those positions.

Later, the regulations on financial institutions’ open for-eign currency limits were widened. The GovernmentDecree on the trading book, in effect since mid-2001,82

introduced a new regulation which equally applied tobanks and investment firms – to cover their exchangerate risk, these financial institutions must allocate 8 percent capital for the part of their aggregate open foreigncurrency positions in excess of 2 per cent of the regula-tory capital. It should be noted that the regulation ontrading books applies to open positions both on and offthe balance sheet (i.e. it refers to the total open posi-tion) and that it controls exposure arising from specialderivative transactions in a modern way.

As the regulation on trading books also covers invest-ment firms, there remains no opportunity for banks tocircumvent the rules on open positions via their securi-ties brokers. However, banks own other financial enter-prises (for example, leasing companies) to which thetrading book regulations do not apply and so, in princi-ple, they could be used by banks to artificially reduceexposure to exchange rate risk recorded in their books.Credit institutions must ensure prudent operationsincluding the risks taken by their enterprises belongingto the holding – they must comply with the regulationspertaining to risk taking and capital adequacy on a con-solidated level as well.83 Although actually theHungarian Financial Supervisory Authority will monitorcompliance with these rules on a continuous basis onlyafter the introduction of consolidated supervision in2004, the surveys conducted up to now84 and theauthority’s data for end-June 2003 suggest that, typical-ly, financial enterprises do not take open foreign curren-cy positions either on or off balance sheet. For example,leasing companies, being bank-owned financial enter-prises with the largest assets, have their foreign curren-cy-based loans refinanced by their parent bank, andtheir forward forex activities are insignificant.

The Bank has little information on the degree of concen-tration of banks’ forward transactions with the corpo-rate sector and on the breakdown of firms’ forwardsinto hedging and speculative positions. The deprecia-tion of the forint exchange rate may cause losses for

domestic participants establishing forwards with banks,temporarily for exporting firms and permanently forspeculators. Although information mainly indicates thedominance of exporting firms, in cannot be excluded ineither case that banks may face counterparty riskslinked to exchange rate weakening.

Nevertheless, banks mitigate their counterparty risk byrequiring maintenance margin from their counterpar-ties, similar to the case of financial futures, againstwhich they can settle accumulated losses. If, as a resultof the losses, the maintenance margin falls below a cer-tain level, then they call on the counterparty to replen-ish his margin account. Due to the competitionbetween financial futures and over-the-counter deriva-tives, the size of the maintenance margin is influencedby the basic margin required for transactions enteredinto on the Budapest Stock Exchange and the BudapestCommodity Exchange, which is HUF 7,500 for one con-tract in the case of euro futures (the size of the contractis the forint equivalent of EUR 1,000), i.e. approximate-ly 2–3 per cent of the nominal value of the contract.Occasionally, banks require margins amounting up to10 per cent primarily from counterparties establishingspeculative positions.

In the case of hedging transactions, with the passage oftime the transitory loss is offset by the increased forintvalue of the exporting firm’s sales revenue. It is charac-teristic of banks that, in the case of transactions con-ducted by a permanent client with good creditworthi-ness, they do not require a margin at all up to a certainlimit, because the client’s bank account provides coverfor counterparty risk and the margin could be a sourceof liquidity risk for the client. (During the forint depreci-ation in June, many clients with hedging transactionsfaced liquidity problems, due to the replenishment ofaccounts.) However, in the case of a counterparty witha speculative position, banks usually require a mainte-nance margin, or change the transaction to a financialfutures.

Based on the above, the regulation on open foreign cur-rency positions adequately limits the financial system’staking direct and indirect exchange rate risk, and hencevolatility of the forint exchange rate may not cause loss-es to banks to an extent which would endanger theirstability. However, it may happen that, under extremeevents, forward transactions entered into with domesticparticipants, particularly speculators, transform the orig-inal exchange rate risk into counterparty risk. But banksprotect themselves against these risks by requiring mar-gin accounts and regular settlement of losses incurred.

82 Government Decree 244/2000.83 The Credit Institutions Act and the Capital Markets Act contain provisions in this respect (Act CXI of 1996., Act CXX of 2001, Tpt.).84 For example, Éva Fischer and Lívia Sánta [2003].


4 ARTICLES

4

APPENDIX

Effect of an FX swap on and off the balancesheet

Chart 4-13

Spot leg

Forward leg

Ft

€

Ft

€

Ft

€

Ft

€

A B

Spot + Swap strategy (synthetic forward)

Chart 4-14

Ft

€

Ft

€

A B

I.

Spot leg

Forward leg

Ft

€

Ft

€

Ft

€

Ft

€

A B

II.

Ft

€

Ft

€

A BIII.

I. Spot transaction

II. Opposite FX swap

III. Result of the two transactions


4 ARTICLES

4

BIBLIOGRAPHY

Ministry of Finance Decree No. 41/1996 (XII. 28.) onThe calculation of foreign currency open posi-tions.

Government Decree No. 244/2000 (XII. 24.) on Therules of establishing capital requirements tocover positions recorded in the trading book,risk undertakings, exchange rate risk and largeexposures and the detailed rules of keeping thetrading book.

Act XCIII of 2001 on The lifting of foreign exchangecontrols and the modification of certain acts.

Adão, Bernardino–Cassola, Nuno–Barros, Luís Jorge[1998]: Informação sobre Expectativas deConverge^ncia do Escudo Contida nas Volatilida-des Implícitas das Opções Cambiais. Banco dePortugal, September 1998

Bank for International Settlements [2001]: TriennialCentral Bank Survey of Foreign Exchange andDerivatives Market Activity, 2001

Bank for International Settlements [2003]: The GlobalOTC Derivatives Market at end-December 2002.May 2003

Campa, José M.–Chang, Kevin [1994]: RealignmentRisk in the ERM: Evidence from Pound-MarkCross Rate Options. Working Paper, New YorkUniversity 1994

Chan-Lau, Jorge A.–Morales, R. Armando [2003]:Testing the Informational Efficiency of OTCOptions on Emerging Market Currencies. IMFWorking Paper, January 2003

Cooper, Neil–Talbot, James [1999]: The Yen/DollarExchange Rate in 1998: Views from OptionMarkets. Bank of England Quarterly Bulletin,February 1999

Dunis, Christian–Lequeux, Pierre [2001]: TheInformation Content of Risk Reversals. Centre forInternational Banking, Economics and Finance,Working Paper April 2001

Eitrheim, Øyvind–Espen Frøyland–Øistein Røisland[1999]: Can the Price of Currency OptionsProvide an Indication of Market Perceptions ofthe Uncertainty Attached to the Krone ExchangeRate? Norges Bank, Economic Bulletin

Fischer, Éva–Sánta, Lívia [2003]: Risk management atbanking groups in the Hungarian banking system.MNB Report on Financial Stability, June 2003

Gereben, Áron [2002]: Extracting market expectationsfrom option prices: an application to over-the-counter New Zealand dollar options. ReserveBank of New Zealand Discussion Paper Series,April 2002

László, Flóra–Móré, Csaba–Vígh, M. Szabolcs–Wolf,Zoltán [2002]: Effects of the band widening andforeign exchange liberalisation. MNB Report onFinancial Stability, June 2002

Száz, János [1999]: Stock exchange options for buyingand selling. Tanszék Kft, 1999, Budapest (onlyavailable in Hungarian)

Vrolijk, Coenraad [1997]: Derivatives Effect onMonetary Policy Transmission. IMF WorkingPaper 1997


4 ARTICLES

4

Lending to the household sector has taken priority overother banking activities in the last three years. The explo-sive growth in banks’ housing loan portfolio can beascribed primarily to a strong upswing in housing loans,triggered by the introduction of a system of subsidisedinterest rates on housing loans in early 2001. This explosivegrowth in the housing loan portfolio materialised only afterthe extension of the subsidised housing loan system toused homes in March 2002. Financing facilities based onthe co-operation between commercial banks and mort-gage banks thrust mortgage banks to the forefront offinancing home purchases and construction. As a result,banks actively involved in lending sought co-operationwith three mortgage banks. This study focuses on how thisextremely strong growth in financing home purchases andconstruction is influencing market participants’ exposure tolending, market and operational risks. In addition to identi-fying the risks, it also addresses the issue of the profitabili-ty of this business line. The data for this study have beenprovided, on a case by case basis, by the three Hungarianmortgage banks and nine commercial banks, each withmajor market share. These banks account for a combinedshare of nearly 87% of banks’ total housing loan portfolio.

DEVELOPMENTS IN LENDING TO THE HOUSEHOLD SECTOR

In the first half of the 1990s, the dismantling of the ear-lier system of subsidised housing loans, high inflationand the bleak economic outlook led to a fall in house-hold loans. High interest rates and a limited selection ofbank products made housing loans unaffordable forhouseholds. Both home purchases and home construc-tion were financed from households’ own resources.The upswing in lending to the household sector can beattributed to the introduction of the subsidised housingloan system in early 2001, which has led to steadygrowth in the proportion (+9.9%) of lending to thehousehold sector within the assets of credit institutionsover the past three years.

Banks’ activity in housing finance has fundamentallyaffected their respective market shares in lending to thehousehold sector. Banks which have been active in thehome lending market have been able to increase theirrespective market shares considerably. By contrast,credit institutions that play no active role in this businessline seen a significant fall in their respective marketshares.86

Almost all of the growth in the housing loan portfolio ofthe market participants that are most actively involvedin the provision of housing loans can be attributed tosubsidised loans. There has been no substantial shift inhousing loans offered at market rates.

Factors contributing to growth in the housing loanportfolio

Real estate trends

The value of housing property assets depreciated grad-ually in Hungary between 1990 and 1997. This trendwas followed by an explosion in market prices between1998 and 2001, the underlying reasons for which werethe following: unfavourable developments in the stockmarket and a gradual decrease in interest rates turnedforeign and domestic investors’ attention to the realestate market; a new generation of investors appeared,owing to the saturation of the real estate market in EUmember states. As most households could not afford ahome because of increasing real estate prices, a systemof subsidised housing loans was needed that wouldencourage credit institutions to change their rather pas-sive attitude to financing home purchases and construc-tion, and make homes more affordable.

Government decree on subsidised housing loans

Under the subsidised loan system (Government Decree12/2001), interest rates on housing loans are well

4.2 GYÖRGY SZALAY–GYULA TÓTH: THE FINANCE OF HOME PURCHASE

AND CONSTRUCTION, THE RISKS INVOLVED AND THEIR MANAGEMENT

IN THE HUNGARIAN BANKING SYSTEM85

85 For more information on the lending practices of banks and mortgage banks, please see the additional information on the MNB’s website.86 The market share in the overall loan portfolio of the ten market participants that have been the most actively involved in lending to the household sector has been

growing steadily. Their overall market share (87%) in late September 2003 was 15.1% higher than at end-2000, prior to the introduction of the system of subsidisedhousing loans. There was a jump in the market share of the banks under review in 2002, when the system of subsidised housing loans was also extended to the pur-chase of used homes. As the following sections of this study focus on the lending practices adopted by the banks most actively involved housing finance, the issueof the developments affecting co-operative banks and building societies is not addressed.


4 ARTICLES

4

below market rates, and thus an increasing number ofhouseholds can afford such loans. The interest rates andservice charges88 to be paid by customers have beenfixed, with government subsidies compensating banksfor related losses in the income. Amendments of theoriginal Decree (prior to its modification in June 2003)broadened the availability and purposes of such loans.In addition, the value of actual subsidies (i.e. capital,interest rates and taxes) also increased. From a regula-tory aspect, the greatest impetus was the extension ofsubsidies to the purchase of used homes in March2002. Furthermore, an amendment in June 2001 to ActXXX of 1997 on Mortgage Banks and Mortgage Bondscreated the operational framework for refinancing hous-ing loans originated by commercial banks through mort-gage banks and access to subsidised interest recordedon the liabilities side.

Business policy-related and other structural factors

Macro-economic and other structural factors:

– As the economy stabilised, households’ willingnessto assume indebtedness grew considerably.

– Real wages have increased significantly over the lastthree years owing to falling inflation and expecta-tions of high raises in nominal wages.

– Most homes are privately owned and are still unen-cumbered with mortgages in Hungary, representinga large amount of eligible collateral.

– The cost of obtaining data needed for creditworthi-ness appraisal and other service charges have fallenmarkedly.

– Demand for housing loans has been brought forwardin anticipation of restrictions on the subsidy system.89

87 The loan portfolio of mortgage banks includes loans originated by the institutions themselves and syndicated loans. Refinancing loans based on the purchase of inde-pendent liens are presented in the relevant section on the originator of such loans, since they are recorded in the originator’s balance sheet.

88 No other fees (e.g. creditworthiness appraisal fees, appraisal fees, disbursement commission, etc) charged by banks have been fixed.89 See the June 2003 modification of the decree.

Loan portfolio of the household sector betweenDecember 2000 and September 2003

In HUF Dec. Dec. Dec. Sep. Changes in

billions 2000. 2001. 2002. 2003. the portfolio

between

Dec. 2000

and Sep. 2003

In HUF Indexbillions

Banking system 462.0 700.5 1190.8 1788.3 1326.3 387.1%

– housing loans 149.5 263.3 695.1 1217.7 1068.2 814.5%– consumer,

personal, hirepurchaseloans 284.3 403.5 446.7 504.6 220.3 177.5%

– other(overdraft,Lombard) 28.2 33.7 49.0 66.0 37.8 234.0%

Savings andloans co-operatives 132.2 168.6 212.7 260.2 128.0 196.8%

– housing loans 38.5 61.0 83.1 102.9 64.4 267.3%– consumer,

personal, hirepurchaseloans 86.7 101.7 123.0 149.5 62.8 172.4%

– other(overdraft,Lombard) 7.0 5.9 6.6 7.8 0.8 111.4%

Creditinstitutions 594.2 869.1 1403.5 2048.5 1454.3 344.7%

Share of creditinstitutions’total assets 7.3% 9.6% 13.8% 17.2%

Table 4-1

Performance and market share of major banks involved in housing finance87

Dec. 2000 Dec. 2001 Dec. 2002 Sep. 2003 Change Dec. 2000–Sep. 2003

Loan port- Housing Market Housing Market Housing Market Housing Market In HUF Ratio infolio in HUF loan share loan share loan share loan share billions percentagebillions portfolio portfolio portfolio portfolio pointsMarket shareper cent

Banks underreview total 134.9 71.9 240.0 74.0 655.8 84.3 1154.0 87.0 1019.1 15.1

Table 4-2


4 ARTICLES

4

Business incentives for the banking sector:

– The market of lending to the corporate sector hadbecome so saturated by 2000 that a growing num-ber of banks looked to the household sector tointensify their involvement.

– Subsidies guarantee larger-than-average interestmargins over a long period of time.

– As households’ indebtedness is low by internationalstandards, the household lending market offers sig-nificant growth potential.

– Credit risks can be kept at an appropriately lowlevel, owing to the inclusion of eligible real estatecollateral combined with other guarantees.

Main products available in the market of housing loansand their profitability

There are two types of loans secured with liens: mort-gage loans and mortgage-based loans. Credit institu-tions issue mortgage bonds to finance mortgage loans,and use traditional liabilities for mortgage-based loans.Further criteria for the classification of housing loansinclude subsidies and the various economic purposes ofhousing loans, as stipulated in the decree on subsidies.

Subsidised mortgage loans to finance the purchase ofnew and used homes

The types and availability criteria of credit facilities forfinancing the purchase of new and used homes are pri-marily determined by the decree on subsidies and theamendements thereunto. Prior to the June 2003 modi-fication of the decree, banks originated the majority ofhousing loans within the framework of the followingtwo basic facilities:90

– Loans subsidised on the liabilities side: loans at avariable 6% interest rate, which can be changedevery five years, service charges added; to be usedfor either the purchase or construction of newhomes, or the purchase of used homes; availableto residents; the credit limit is HUF 30 million. (Inthis case, the interest on mortgage bonds, issued tofinance this type of loans, is subsidised. As inHungary only mortgage loan companies have theright to issue mortgage bonds, subsidies are chan-

nelled into the banking system through such com-panies.)91

– Loans subsidised on both the asset and liabilitiessides (double subsidy loans): loans at an interest rateof 4.5% fixed for five years, with fixed service chargesadded, to finance either the purchase or constructionof new homes; the price of the home to be pur-chased or constructed should be under HUF 30 mil-lion; applicants are either married couples or singleparents with custody of their children; the credit limitis HUF 10 million. (In contrast to subsidy on the liabil-ities side, the amount of the subsidy on such loansrecorded on the asset side, based on the actualamount of the relevant loan, is recorded in the booksof the credit institution originating the loan.)

The above facilities, if they were originated prior to themodification of the Decree, are to earn an approximate-ly 7–9% interest margin92 in five years from their origi-nation and refinancing.93 As commercial banks can onlyacquire subsidies related to mortgage bonds throughmortgage banks, they may well earn lower margins. Partof the relevant subsidy, the proportion of which is deter-mined either in a refinancing or syndicate agreement, isrecorded in the books of the mortgage bank. Since theinterest margins to be earned were well in excess of thebanking sector average, profitability proved to be anexcellent incentive for credit institutions.

Modification of the decree on subsidies has led tochanges in the amount of and stricter eligibility criteriafor loans.

– The credit limit for loans subsidised on the liabilitiesside fell from HUF 30 to 15 million. However, themaximum amount of subsidised loans to financeeither the purchase or construction of new homeshas risen from HUF 10 to 15 million. In contrast tothe old system, under the new regulations, a house-hold is only eligible for one type of subsidised loansat a time. Though there has been no substantialchange in the extent of subsidies, the modificationof the decree has resulted in fewer differencesbetween loans subsidised on the liabilities side,which were invariably used earlier to finance thepurchase of used homes, and loans with double sub-sidy. As a result, a large number of banks no longeroffer double subsidy loans, originated earlier only tomarried couples or single parents with custody of

90 As loans at variable rates, i.e. with an interest rate changed every five years, account for over 95% of the overall loan portfolio in the entire banking sector, this studyfocuses on the profitability of such products. Since the size of the portfolio of subsidised loans recorded on the asset side and their profitability are insignificant, thisstudy does not discuss them.

91 For a more detailed treatment of this facility, see later sections.92 Both interest-like fees and subsidies included, the interest margin to be earned on a facility denotes the interest rate differential. The way it is calculated is different

from the one applied in the case of special loan facilities, which grant a grace period before repayment starts.93 In order for the costs incurred (e.g. registration fee at the Land Registry Office, loss of earnings, etc) to be compensated, banks were originated a one-off reimburse-

ment of 2% by the government.


4 ARTICLES

4

their children. Consequently, subsidised loansrecorded on the liabilities side have taken the lead.

At the level of the entire banking system, the interestmargin on loans originated and refinanced under thenew terms and conditions has fallen to 3.4–5.4%,owing to the reduction in subsidies.94 We believe thatthe current interest margin, although it is 3–3.5 per-centage points lower than previously, still offers astrong incentive for banks, as it is guaranteed for along period of time in a market where average inter-est margins95 are falling.

General purpose, market-rate loans

General-purpose loans are market-rate loans collater-alised with a lien. Subsidised loans, however, are to beused for either home purchase or home construction.The low interest rates on subsidised loans has resultedin a lower share of market-rate loans collateralised witha lien in the loan portfolio in the past two years. In con-trast to subsidised loans, interest on and the profitabil-ity of market-rate loans depend on client risks and thecost of liabilities. Profit on market-rate loans may beearned from both interest rates and other fees.

Incurrence of liabilities to finance lending

Dynamic growth in assets has been accompanied withsimilarly vigorous growth in liabilities. By nature, mort-gage lending, in keeping with international practice,relies on such techniques of incurrence of liabilities thatare different from those of traditional lending. Statutoryand regulatory stipulations have also been instrumentalin the development of the necessary structures: subsi-dies on the liabilities side reduce the cost of liabilitiesincurred by credit institutions through interest subsidieslinked with mortgage bonds. Under Hungarian law,mortgage bonds may only be issued by mortgage banksthat comply with certain specific statutory stipulations.

In terms of the financing techniques adopted, the prac-tice and procedure of mortgage lending varies from onecountry to another, and may be different even withinone country. In EU Member States there are two waysof financing of mortgage loans through mortgagebonds:96

– Direct financing (the one-tier model): banks originat-ing mortgage loans group loans themselves andissue mortgage bonds to incur liabilities. This is wide-spread practice in Germany and Denmark.

– Indirect financing (the two-tier model): banks origi-nating mortgage loans sell their loans to a third com-pany; the latter then puts them into a ‘pool’, andmortgage backed securities (MBS). This is wide-spread practice mainly in Anglo-Saxon countries.

Currently, there are three types of refinancing tech-niques in Hungary. One is the purchase of independentlien, a special form of indirect finance; the other isbased on syndicated agreements; and the third is loansoriginated by mortgage loan companies on their own,the latter two correspond to the concept of the one-tiermodel.

Refinancing model based on the purchase of independent lien

The underlying idea of this refinancing facility is thatmortgage banks that issue mortgage bonds refinancecommercial banks originating mortgage loans underindependent lien sales contracts. Banks repurchaseindependent lien at the pace loans are repaid.Mortgage bonds are collateralised through the refinanc-ing of loans with independent lien. As mortgage loanscontinue to be recorded in the balance sheet of theoriginator credit institutions, both credit and prepay-ment risks are borne by the bank that originates theloan and, indirectly, by the mortgage bank involved.

94 For a detailed description of the calculation, see the section on product profitability.95 In the banking sector the average interest margin was 3.9% at the end of 2003 H1. It is, however, also exposed to the proportion of interest-bearing instruments in

the balance sheet of assets and liabilities. In the case of mortgage loans, by contrast, this proportion is nearly identical on both the asset and liabilities sides.96 For a detailed treatment of the issue, see Judit Vincze, ‘The market position of and trends in the development of mortgage bonds’, Hitelintézeti Szemle 2002/3.

Product composition of the loan portfolio held by banks with a dominant market share, 30 September(In HUF billions)

Housing mortgage loans Mortgage-based loans

Subsidised Doubly- Market-rate Total Asset side Market-rate* Totalrecorded as subsidised subsidised

liabilities

Banks underreview: 683.6 235.3 2.0 920.9 10.3 222.7 233.0

* Other subsidised loans have been included in market rate loans.

Table 4-3


4 ARTICLES

4

Mortgage bonds investors only take low risks, sincemortgage loans are repaid to the mortgage bank if thepartner bank goes bankrupt.

Financing model based on syndicated loan agreements

Under such agreements, commercial banks and lendersto mortgage banks form a syndicate and, as such, they dis-tribute loans via their distribution network. When a loan isoriginated, a mortgage bank purchases it, so it is no longerrecorded in the balance sheet of the relevant commercialbank. As a result, lending risks are taken by mortgagebanks. Members of the syndicate earn income in propor-tion to the tasks and risks taken. Mortgage banks issuemortgage bonds collateralised with the mortgage loanspurchased.

Financing of loans by mortgage banks

Mortgage banks finance the loans sold via their ownnetwork of distribution or agents by issuing mortgagebonds. Both loans and mortgage bonds are recorded onthe balance sheet of mortgage banks.

Mortgage banks play a pivotal role in all three modelsand determine the structure of the housing loan mar-ket. Currently, there are three mortgage banks operat-ing in the banking system in Hungary. No new marketparticipants are expected to emerge even over thelonger term. Földhitel és Jelzálogbank (FHB),97 still instate ownership at the end of September 2003, hasentered into a refinancing agreement with nine com-mercial banks, and into a syndicate agreement withsixty-two co-operative banks. In addition to these com-mercial banks and co-operative banks the FHB alsooriginates housing loans through agents and via itsown distribution network. HVB Jelzálogbank, 100%-owned by its parent bank, has entered into a refinanc-ing agreement with four commercial banks.Furthermore, it also originates a limited selection ofhousing loans. OTP Jelzálogbank, which is also 100%-owned by its parent bank, only originates housingloans under a syndicate agreement with its parentbank.

In late September 2003, the value of the mortgagebonds issued by these three mortgage banks was HUF752.8 billion, of which a package in the amount of HUF420.2 billion98 was sold through a public offering. Theyield on mortgage bonds placed through the publicoffering was 60 to 150 basis points higher than thebenchmark (government bond yield).

RISKS IMPLIED IN HOUSING LOANS

The chart below shows the practice adopted by banksand mortgage banks in originating housing loans andthe factors that, by the very nature of the procedure forsuch lending, add to the risks involved. Except for pric-ing, there is no real difference between subsidised andmarket-rate loans.

The following sections describe how, based on the lend-ing procedures outlined above, the extremely rapidgrowth in housing loans has affected market partici-pants’ exposure to lending, market and operationalrisks99 and the profitability of products.

Chart 4-15

Assets Liabilities

Mortgage loan (1)

Mortgage loan (2)

Ordinary resources

External resources

Refinancing loan

Commercial bank

Sale of

independent lienRefinancing

Level 1.

Level 2.

Paying

Repayment

Bor

row

ers

Assets Liabilities

Refinancing loan Mortgage bond

Mortgage bank

Chart 4-16

Assets Liabilities

Mortgage loan (1)

External resources

Ordinary resources

Mortgage loan (2)

Commercial bank

Sale of loans

Level 1.

Level 2.

Paying out

Repayment

Bor

row

ers

Assets Liabilities

Mortgage loan Mortgage bond

Mortgage bank

Chart 4-17

Assets Liabilities

Mortgage loan (1)

External resources

Ordinary resources

Mortgage loan (2)

Mortgage bank

Mortgage bond

BorrowerPaying out

Repayment

97 The owner decided to privatise a 50%-1 ownership stake in the mortgage bank. Privatisation took place in November 2003.98 In effect, the mortgage bonds offered publicly by OTP Jelzálogbank in the amount of HUF 222.2 billion qualified as privately offered bonds because of the subscrip-

tion preference that the parent bank enjoyed.99 Conclusions on the risk exposure of credit institutions are based on data as at end-2002, and on the results of questionnaire surveys.


4 ARTICLES

4

ations of risk management should enjoy priorityat this level.

Lending policy should include the bank’s short-term and medium-term strategic outlook and theexpectations for the credit portfolio, the qualityof the portfolio and its profitability. At the levelof lending policy and product regulations, port-folio-level limits and exclusive and/or limitingconditions should be imposed and laid downrespectively. The pricing of subsidised loans isdependent on statutory regulations to a largeextent. In the case of loans at market rates, inter-est rates and fees should be proportionate withcustomer risks. In the longer run, the pricing ofsubsidised loans and that of market-rate loansshould converge to an increasing degree.

Preliminary credit appraisal is anticipated totake some burden off middle office areas, whichare already rather overloaded. The appraisal ofthe real estate to serve as collateral, scoring andexamination of customers’ financial capabilitiesshould all form the basis for credit appraisal.Because of a considerable increase in demandand inexperience of households in paying backtheir loans, any reduction in the weight attachedto scoring and income-related deliberations mayadd to risks. The involvement of risk managementareas in credit appraisal is of vital importance.

Banks should obtain long-term liabilities as soonas a lien is registered and loans are disbursed.The reasons are that this reduces the repricinggap, and they can only record subsidies subse-quent to the registration of lien and the disburse-ment of loans.100 In order for mortgage banks tobe refinanced and loans to be provided, theloans assumed and refinanced respectivelyshould qualify as eligible collateral as soon asrealistically possible. Once this has taken place,the issue of mortgage bonds can commence.The sales potential of mortgage bonds and yieldspread depend heavily on the functioning of thedomestic securities market.

As for overall risk management, risk manage-ment related to housing loans involves credit,market and operational risks, while taking intoconsideration the special characteristics of hous-ing loans.

Although, as a result of a current upswing inhousing loans, portfolio quality, which deter-

100 Subsidies recorded on the liabilities side, reimbursement of finance-related costs.

Composition of the mortgage bond market byissuers as of 30 September 2003

Chart 4-18

Private issue Public issue

0

100

200

300

400

500

OTP Jelzálogbank FHB HVB Jelzálogbank

HUF billions600

Flow chart of the provision of housing loans

Chart 4-19

Risk management

Background of regulation and organisation

Lending-financing

Credit risk

Marketing, sale

Acceptance of loan applications

Preliminary classifications

Valuation of coverages

Credit approval

Registration of mortgage

Paying out

Credit and pricing policy

Issuance of mortgage bonds

Refinancing/sale of credits

Market risks

Operation risk

The following preliminary expectations regarding lending practice have been drafted:

Both internal regulations and organisationalstructure provide framework conditions for theorigination of housing loans. In keeping with theexpectations, both internal regulations andorganisational structure should take into accountthe special characteristics of the provision ofhousing loans to the greatest possible extent,and should be centralised. The various consider-

Box 4-1


4 ARTICLES

4

The rating of the housing loan portfolio is better than thequality of the entire bank portfolio, within which the qual-ity of the portfolio comprising market-rate loan lags wellbehind that of subsidised loans. The credit institutionsunder review recognised impairment in the amount ofHUF 1.17 billion on subsidised loans prior to end-2002,compared to HUF 1.67 billion on market-rate loans.Impairment of market-rate and subsidised loans as a pro-portion of the portfolio was 1.2% and 0.2%, respectively.

According to the Ministry of Finance Decree, banksshall, through CEO orders, stipulate procedures for therating of their housing loans-related liabilities and riskprovisioning. Liabilities are rated monthly and impair-ment is established quarterly. Overdue payment formsthe basis for credit rating.

At end-2002, payment of 4.3% of subsidised loans onaverage was overdue, compared to 9% of market-rateloans. Most had been overdue for less than 30 days,suggesting lax repayment discipline rather than habitualfailure to pay. The reasons why the rate of overdue pay-ment is higher in the case of market-rate loans includelenient regulations governing sequestration and an age-ing portfolio. In the case of subsidised loans, borrowers’discipline to repay their debts in a timely manner wasrather remarkable (payment had been overdue for over30 days in a mere 0.9% of all loans).

As nearly 70% of the housing loan portfolio was origi-nated within a year, overdue payment and the quality ofthe portfolio may change considerably during the matu-rity of the loans. Duration varies between 5 to 25 yearsand is increasingly steadily. The loans originated so farare extremely young. Compared to a highly successful2003, growth rate is expected to slow down in theyears to come, with the average age of the portfoliogrowing higher.

The average amount of refinanced housing loans isHUF 4 to 5 million,103 the monthly instalment of which,

Credit risks

The portfolio quality of housing loans depends on bor-rowers’ ability to pay and expected changes in thevalue of the collateral portfolio. Borrowers’ paymentattitude depends primarily on their income position. Anynegative change in the value of the real estate serving ascollateral depends on developments in global and localreal estate markets. In addition, changes in borrowers’ability to pay are also affected by amendments of theGovernment Decree on subsidised housing loans,101 andadjustment of bank strategies to meet market demand(e.g. higher degree of collateralisation, calling for a mora-torium on the payment of principal, etc).102

mines credit risks, is far better than the bankaverage, regulations governing the appraisal ofreceivables and the recognition of impairmentshould be clearly defined. Monitoring requires agreat deal of information, which helps assessfuture developments in portfolio quality.

The explosive growth in housing loans affectsthe interest rate and liquidity risks that banksand, in particular, mortgage banks face. As themajority of positions are assumed by mortgagebanks, such companies are expected to adjusttheir interest rate and liquidity risk managementso that it complies with the characteristics ofhousing loans. The technical aspects of refinanc-ing or sale of loans may not lead to an unjustifiedrise in interest rate and liquidity risks.

In terms of the entire portfolio, the operationalrisks implied in housing loans (e.g. abuses relat-ed to collateral appraisal, the use of loans forpurposes other than specified in the loan agree-ment, losses arising from IT system malfunctions,etc) are not expected to be significant.

101 For instance, restrictions on favourable tax treatment regarding housing loans.102 For a detailed treatment of the income position of households and changes in the value of real estate, see the special topics in the MNB’s Report on Financial Stability

published in December 2002.103 As no reliable data on the average volume of loans was available, the value of HUF 4 to 5 million disclosed by mortgage banks was used as a reference value in this study.

Distribution of the housing loan portfolio at banks under review by rating categories

Problem-free Qualified Total

31 December 2002 gross ratio net gross/recorded ratio net gross/recordedHUF millions HUF millions HUF millions HUF millions HUF millions

Subsidised loans 509,508 98.9% 4,246 5,418 1.1% 513,754 514,926Market-rate loans 135,097 95.9% 4,152 5,826 4.1% 139,249 140,922

Total 644,605 98.3% 8,398 11,243 1.7% 653,003 655,848

Table 4-4


4 ARTICLES

4

at an interest rate of 6% and with maturity of 15 years,is HUF 37,973. Based on average monthly net wages inthe national economy in 2002 (NSO) and assuming afamily model with two active earners, this represents24.4% of the net wages of an average family. As, in alllikelihood, it was higher income earners that applied forloans at the initial stage of lending, the average month-ly instalment is likely to place less burden on them.

Credit risk exposure on the basis of the collateral included

Current practice suggests that the maximum amount ofhousing loans is 60 to 70% of the value of the collat-eral (LTV),104 which represents only 55 to 60% of theactual market value of the housing property in ques-tion.105 Some banks may set higher LTV’s in the case ofloans not financed through mortgage bonds. Evenstricter limits (30 to 50%) may be established bybanks in their product regulations.

Under bank regulations governing appraisal, only inde-pendently marketable unencumbered real estate locat-ed in the territory of Hungary qualifies as eligible col-lateral.

In terms of the territorial distribution of collateral,supply is better in Western Hungary. Because of thehigher earning potential of the households in thatregion, there is stronger demand for new homes. InEastern Hungary, by contrast, it is used homes thatcontinue to be in demand, for the construction of newhomes is far and few between. The majority of thetransactions involving real estate, and as a result, relat-ed lending, are carried out in Budapest and a few larg-er towns in the provinces. In the settlements neartowns, residential property is practically unmarketable

104 Under the Act on Mortgage Banks, a maximum of 60% of the value of the relevant collateral shall qualify as eligible collateral for mortgage securities. A further 10%shall qualify as supplementary collateral. Under this arrangement, banks find a maximum 70% LTV ratio acceptable for loans financed through mortgage bonds.

105 For expected development in real estate prices, see Gergely Kiss, ‘The housing market and financial stability in the light of EU accession’ (MNB’s Report on FinancialStability, December 2002).

106 The high proportion of market-rate loans with an LTV value below 20% reflects loans, originated a number of years ago, the majority of which have been repaidby now.

Distribution of collateral by real estate type

Chart 4-20

Detached house

49.4%

Block of flats

29.2%

Housing estate

6.3%

Other flats

0.2%

Other properties

14.9%

Distribution of the housing loan portfolio at banksunder review on the basis of overdue payment

Market-rate loans Subsidised loans31 December

2002 HUF Distribution HUF Distributionmillions (per cent) millions (per cent)

No overduepayment 128,281 91.0 492,627 95.7< 30 days 6,751 4.8 17,427 3.4

30–60 days 2,118 1.5 2,124 0.460–90 days 692 0.5 595 0.190–180 days 987 0.7 911 0.2180–360 days 1,074 0.8 730 0.1

>360 days 1,019 0.7 512 0.1

Total 140,922 100.0 514,926 100.0

Table 4-5

Age and maturity composition of housing loans

Age of housing loans Distribution Maturity Distribution31 December (%) 31 December (%)

2002 2002

Less than 1 year 69.7 Less than 5 years 2.4Between 1 and 2 years 15.7 5 to 10 years 25.8Between 2 and 3 years 7.1 10 to 15 years 22.1Between 3 and 4 years 0.6 15 to 20 years 31.64 years and over 6.9 Over 20 years 18.1

Total 100.0 Total 100.0

Table 4-6

Average distribution of loans according to theLTV ratio (In percentage)106

31 December Market-rate Subsidised2002 loans loans

Below 20% 65.6 18.1Between 20 and 30% 9.0 6.7Between 30 and 40% 7.1 11.3Between 40 and 50% 8.7 15.6Between 50 and 60% 5.3 32.2Between 60 and 70% 2.0 15.2Over 70% 2.3 0.8

Total 100.0 100.0

Table 4-7


4 ARTICLES

4

at most places, which can be attributed to a rise in theconcentration of urban population.

As, in the case of default on loans, the sale of residen-tial property at smaller settlements in the provinces mayrun into difficulties, lending policy should set up a limitsystem regulating the geographical exposure of theloan portfolio.

Factors adding to the risks arising from lending and riskmanagement practice

Shortcomings in lending policies

Both the content and level of detail of the lending poli-cies adopted by the banks under review vary in quali-ty. Five banks have an elaborate lending policy thattackles the issues of expected volume and profitabilityas well as risk tolerance in a detailed manner. Threebanks set general expectations about target groups,directions of development and objectives. In effect,they seek to increase the size and maintain the currentquality of their portfolio. They fail, however, to laydown requirements (e.g. client portfolio, products, IT-related development, etc.) needed for their doing so.It follows that major short- and medium-term objec-tives reflect qualitative (e.g. increasing market share)rather than risk considerations. The lending policy ofthe majority of the banks under review fails to specifyquantified limits of housing loans. Only two banks setportfolio-level risk limits in a detailed manner. Onebank specifies product-level limits.107

Shortcomings in internal regulations

Only few banks have adopted regulations governingthe entire lending procedure. A further shortcoming isthat banks do not have special regulations regarding the

market risks (e.g. interest rate and liquidity risks) gener-ated by the provision of housing loans.

Supervision by the same person of the lending and riskmanagement areas on the level of top management

Responses to the survey suggest that all banks haveindependent organisational units responsible for theindividual stages of the lending procedure. Both lend-ing and risk management are separate areas on divi-sional and management level. It is, however, worryingthat, at some banks, the same deputy CEO and manag-ing director is in charge of the two areas.

Unused capacity of the credit bureau system

An efficient use of the CB system can help identify uncred-itworthy clients upon the placement of their loan applica-tions. Despite the obvious advantages to be enjoyed,banks rely on CB databases to a varying degree.Responses to the survey reveal that two banks do not relyon them at all in preparing their preliminary creditworthi-ness appraisal. Another bank does, but not as it should,because it only relies on them in its final decision-making.Only four banks use the CB system efficiently.

Biased collateral appraisal

As the result of collateral appraisal has a fundamentalimpact on loan approval and the amount of the loan,there have been to attempts to bias either the appraisalprocedure or its results. In order for prudent lendingpolicy to be implemented, banks have a vested interestin drawing up a contract with appraisal companies,stipulating that the agreement with such companiesshall be terminated in the event of negligence, suspi-cion of fraudulent appraisal and even unintentionalmalpractice. Some mortgage banks even apply a sys-tem of professional rotation to appraisal companies,which means that the contract with the companywhose performance is the worst in a given year is auto-matically terminated.

Shortcomings in the various scoring systems

As the scoring systems that most banks use are unso-phisticated, the conditions for the risk-related classifica-tion of clients and the review of such classification canonly be created over a longer period of time, owing tothe short time series available. The scoring systems cur-rently used do not carry out a deliberate check on‘behaviour-type’ (the applicant’s credit history). Twobanks do not use the scoring system and do not exam-ine the client’s credit capacity in preparing their credit-worthiness appraisal. Instead, they only check whether

107 The internal regulations of the remaining banks set forth the criteria of exclusion pertaining to the individual segments and limits to be applied on a case basis.

Average distribution of loans by geographicalregions

Distribution31 December

2002 In HUF billions Per cent

Budapest + Pest County 342.4 52.2Southern Great Plain 67.6 10.3Mid-Transdanubia 63.0 9.6Western Transdanubia 52.5 8.0Eastern Hungary 49.2 7.5Southern Transdanubia 42.6 6.5Northern Hungary 38.7 5.9

Total 655.8 100.0

Table 4-8


4 ARTICLES

4

the minimum requirements laid down in their respectiveproduct regulations are met. Although, due to theirexperience in the household sector and a clear-cut setof minimum criteria, the lending practice of two banksdoes not pose egregiously high risks, their rating systemshould be upgraded. A further shortcoming in the scor-ing system and the creditworthiness appraisal proce-dure is that they do not examine the income positionproperly in some cases. Income position carries lessweight in loan approvals than does either scoring orcreditworthiness appraisal.

Limited use of monitoring

During an upswing in lending, banks’ priorities are salesand origination, with monitoring – mapping of anypotential risk that may arise during maturity – taking alow profile. As the average maturity of housing loansexceeds 10 years, the fact that after their origination,most of the banks under review fail to carry out a regu-lar check on debtors’ income position increases therisks considerably. They primarily focus on monitoringclients’ willingness to pay (overdue payment). In thecase of refinanced loans and syndicated loans, mort-gage banks’ main priority is checking the eligibility ofthe collateral in question. This, however, does notincrease their risks as they are only in refinancing posi-tion. Due to the lack of sophisticated scoring andscorecards, debtors’ income position and other fac-tors are less thoroughly monitored than they shouldbe, which may pose risks especially to banks withclients whose income position is weaker.

Limited marketability of collateral

How long collateral can be sold depends on how co-operative the relevant debtor is, the prevailing conditionson the real estate market and the geographical locationof the housing property in question. In the case of sub-sidised loans, a compulsory preparation of a notarial doc-ument of loan and mortgage agreements allows for thepossibility that enforcement can be carried out of court.Thus, it only takes the public auction of the housing prop-erty in question 6 to 10 months to occur after banks ini-tiate a repossession procedure. A number of banks haveno subsidised housing loans-related experience, as theyhave had no default loans that have reached the stageof the completion of the enforcement process. Thosebanks that do have any experience reported favourableresults. The reasons for this include stringent rules govern-ing collateral appraisal, with resultant reliable values ofcollateral and related lending regulations. In the case ofmarket-rate loans, where no compulsory preparation of anotarial document of the relevant loan agreement is stip-

ulated, public auctions may take 2 to 3 years. In this case,moratoria interest adds to receivables significantly, andowing to the length of time, no full repayment is likely tobe effected.

Market risks

Interest risk

As home financing essentially involves long-term loans(with maturities between 5–20 years), the alignment ofasset and liability interest rate structure is of primaryconsequence from the point of view of interest risk.Considering that in home financing the leading productis the variable rate loan rolling over once every fiveyears,108 it is expedient to adjust interest rate structureof liabilities to this schedule. During the periodreviewed, on a sector level there was no perceptibletrend in the ratio of 90-day cumulative HUF gap to thebalance sheet total, which suggests that housing loansdid not have a significant effect on the interest riskexposure of the banking sector. The primary reason forthis is that the banking sector has financed the sub-sidised loan portfolio practically completely by issuinglong-term fixed interest rate mortgage bonds, while itrelied on short-term FRN liabilities only on a temporarybasis.

Although no sector level trend was traceable in expo-sure to interest risk, in this respect financed commercialbanks and mortgage banks differ significantly.

Financed commercial banks

Mortgage banks generally quote refinancing interestrates for refinanced commercial banks once a month.This rate depends on the cost of funds and the govern-ment subsidy109 (to be shared between the bank and themortgage bank). Thus, the interest rate structure ofloans and liabilities are nearly identical for housing loansextended by commercial banks and the funds used forfinancing them. Consequently, this activity has no signif-icant impact on banks’ overall interest rate risk expo-sure. The loans originated in the framework of syndicat-ed agreements are transferred to the balance sheet ofthe mortgage bank, and therefore the originator isexposed to interest risk only in the period between orig-inating and credit selling (usually 30 days), but its extentis insignificant.

Mortgage banks

The reason for that mortgage banks’ HUF-gap exceedsthe average and theoretically continuously increases is

108 Although the relevant decree allows fixing interest rates for a period longer than 5 years, the banks have originated only very few such loans up till now. For thetime being banks extend loans with longer rollover periods only in a very small number.

109 See the chapter on the profitability of products.


4 ARTICLES

4

that there is a difference between the duration of annu-ity-based loans and loans redeemed at the end of theterm. Thus, due to intensifying activity, the 90-day HUFgap/balance sheet as well as the interest risk increaseconstantly. The adverse impact could be mitigated bythe application of interest derivatives. Alignment of theasset and liability interest rate structure was alreadyproblematic when these credit institutions were started,when they only originated their own loans.110 At thattime, the gap between origination and issuance of mort-gage bonds was longer and issues were more irregular.However, in recent years mortgage bond issues havefollowed the more regular and larger volume of loansmore closely.

Interest risk generated by prepayment

The fact that the provisions of the Hungarian Civil Codedo not allow commercial banks and mortgage banks topreclude prepayment for their customers has a signifi-cant influence on interest risk. Although banks cancharge a prepayment fee to make up for their losses,their interest risk exposure may rise considerably if pre-payment becomes widespread.

Prepayment can move the gap in a positive direction, asin this case the prepaid amount (asset) becomes a short-term loan (within 30 days). The gap increase resultingfrom the prepayment of housing loans raises the inter-est risk for commercial banks only if they fail to pass iton to the financing mortgage banks (i.e. they do notprepay as well).111 Due to their relatively low share inthe total assets, a prompt prepayment of 50% of allhousing loans would result in only a 1–5 percentagepoint gap increase on average.

However, if mortgage banks’ receivables (directly orig-inated housing loans + refinancing loans) are prepaid ina large scale, the HUF gap would move significantly inthe positive direction. In this case, a 50% prepaymentas mentioned above may push the gap up by even 30-40 percentage points. But as mortgage banks mustmeet the collateral requirements at all times, prepaidamounts must be either re-originated or invested in gov-ernment securities (and thus classify among additionalcollateral112), and the interest risk exposure is automati-cally reduced.

Risk factors of the lending and risk management practice

Fluctuation beside upward trend in mortgage banks’ gap

In addition to an increasing trend of interest risk, sub-sequent to the issue of mortgage bonds, the HUF gap

opens in positive direction, while upon acceptance ofloan packages it closes. Thus, measuring and projectinginterest risk may be confronted with considerable diffi-culties in the case of mortgage banks.

Fluctuation in commercial banks’ gap

During the time between originating and refinancing/loan sales (usually 30 days), the banks’ HUF gap mayopen, although low portfolio levels and continuouslyloan extension it does not have considerable impact onthe total HUF gap of banks.

The priority of profitability over interest risk

In the past few months, several mortgage banks issuedshort-term mortgage bonds for speculative purposes inorder to generate a potential increase in interest subsi-dies (capitalising on the difference between the formerand the new subsidy decrees) and have disregardedgrowth in the interest risk.

Preferential mortgage bond subscription within groups

As up until now the parent bank had a subscription pri-ority to the mortgage bonds issued by OTPJelzálogbank, part of the interest risk of housing loanswas ultimately undertaken by the parent bank.

Shortcomings in the management of prepayments

Market participants do not have sufficient experiencein the increase of interest risk generated by prepay-ment. Mass prepayment may considerably boost interestrisk primarily in the case of mortgage banks and due tothe prevailing collateral requirements, it may evenimpede smooth business. Such an extreme case may,however, occur only in case loans are fully and promptlyprepaid in the amount of HUF 150–200 billion, but thispossibility can be ruled out for all intents and purposes.

Risk management deficiencies at mortgage banks

Risk management deficiencies are found first of all withmortgage banks. In their case, exposure to interest risk isstill being measured and the instruments of risk manage-ment have not yet been fully created. For the time beingthe Asset-Liability Committee can intervene only on theasset side in the case of excessive interest risk, and morespecifically in the volume and timing of issues. The appli-cation of interest derivatives is ad-hoc and rudimentary.The quality of risk management performed by mort-gage banks is not in line with the extent of risks. The“traffic lights” rule of the German mortgage loan act may

110 In addition, within 3 years no less than 80% of the issued mortgage bonds had to be covered by eligible collateral with a term exceeding 5 years.111 Pursuant to statutory authorisation (Mortgage Loan Act – “Jht.”), mortgage banks may exclude the possibility of prepayment for refinanced commercial banks.

However, the Hungarian Civil Code prevents them from excluding prepayment by retail customers.112 The rate of additional collateral (securities) within the total collateral covering mortgage bonds may not exceed 20%.


4 ARTICLES

4

be taken as a guiding principle both for mortgage banksand for the regulatory bodies.113

Liquidity risk

In terms of housing loans and home financing, liquidityrisks may stem from an alignment of the maturity struc-ture of assets (housing loans and refinancing loans) andliabilities (mortgage bonds and refinancing loans).

Financed commercial banks

As housing loans are repaid on an annuity basis and therefinancing loans of mortgage banks are repaid by refi-nanced commercial banks simultaneously with cus-tomer repayment, theoretically no liquidity risk arisesfor commercial banks. Mortgage banks acting in theframework of an independent lien generally make refi-nancing loans available on a monthly basis, thereforecommercial banks are required to rely on short-termmoney market resources merely for a temporary 30-dayperiod. Due to the relatively small size and the continu-ity of the monthly loans, this may not generate increasein the liquidity risk. The commercial banks acting in theframework of syndicate agreements also rely on moneymarket instruments on a temporary basis.

Mortgage banks

The inter-bank positions of mortgage banks are subjectto the periodic movements of continuous lending andmortgage bond issues. Over the short term, continuousloan redemption generally results in excess liquidity. Inthe case of mortgage banks, liquidity risk increases if thecash-flow structure of assets and liabilities do notmatch: mortgage bonds are repaid in full upon maturitywhile refinancing loans are repaid continuously. The liq-uidity gap generated by the mismatch in the cash-flowstructure and the time difference between loan purchas-es and mortgage bond issues results in fluctuatingmoney market exposure and dependence, which aremitigated by open credit lines.

Risk factors in the lending and risk management practice

Liquidity risks of the re-issue of mortgage bonds

As the relevant decree on subsidies limits the averageterm to 15–20 years, and the average term of mortgagebonds is 5 years, mortgage bonds must be re-issuedtwice or three times under the current terms. The liq-uidity gap generated between the final repayment ofmortgage bonds and their re-issue makes mortgagebanks temporarily and highly dependent on the

money market. The internal regulation of such situa-tions is deficient. The issuing banks could reduce thisliquidity risk by further increasing the maturity of mort-gage bonds.

Preferential mortgage bond subscription within groups

As up until now the parent bank had a subscription pri-ority to the mortgage bonds issued by OTPJelzálogbank, similarly to the interest risk, part of theliquidity risk was ultimately undertaken by the parentbank.

Financing delay in the case of newly built homes

In case of newly built homes, the long-term fundsbecome available to the commercial bank and the mort-gage bank only subsequently to the registration of lien.Due to continuous lending and refinancing, this doesnot generate outstanding increase in the liquidity risk.

Risk management deficiencies

Similarly to interest risks, deficiencies in the manage-ment of liquidity risks are found primarily with mortgagebanks, which mitigate fluctuations in liquidity only byopen credit lines. The liquidity risk generated by a mis-match between the cash-flow structures of assets andliabilities increases in line with the intensification ofactivity. Mortgage banks’ liquidity risk management sys-tems (limit system, internal limits) are not always suffi-ciently prepared to handle this risk.

Exchange rate risk

Due to the weak absorption capacity of the domesticbond market, the mortgage bonds issued by mortgagebanks may appear on the international capital marketsas well.114 FX-denominated bonds generates exchangerate risk in the housing loan market. So far, only FHB hasissued euro-denominated bonds. The bonds were soldby an organisation established specifically for the pur-poses of securitisation (SPV), ensuring the required forintfunds in this manner. The actual appearance of FX fundsin the balance sheets of mortgage banks and eventuallycommercial banks would be perceivable only if thescope of the home loan subsidies decree extended tothe interest rates of mortgage bonds issued in euro.

Operating risks

In connection with housing loans, operating risks maystem from inappropriate organisational background ofthe lending process, underdeveloped management

113 Pursuant to the rule, German mortgage banks prepare interest risk reports on a daily basis, simulating the impacts of yield curve shifts by 1, 10 and 100 base points.In case this effect reaches certain specific rates of the equity capital (10, 20 and 30%, respectively), the Supervisory Board may order special measures to be takenand special inspections to be conducted.

114 Vincze, Judit: “A jelzáloglevelek piaci helyzete és fejlôdési irányai” [The market situation and development trends of mortgage bonds] (Hitelintézeti Szemle, 2002/3).


4 ARTICLES

4

information systems supporting executive decisions(MIS), and a limited IT background in the credit evalua-tion process. Based on the responses given to our ques-tionnaire, operational risk can be demonstrated in rela-tion to the current state of MIS systems. With regard tofurther risk components it can be established that allbanks measure and analyse exposure to the individualmarket risk segments in an organisational unit separat-ed from their business organisation. Despite significantdifferences between the levels of MIS systems, only onebank mentioned the IT background as a bottleneck inthe smooth transaction of the lending procedure.

Profitability in housing loans

The profitability of housing loans results from the differ-ence between the interest rates charged without gov-ernment subsidies and the cost of funds. The interestrates charged for customers are the sum of the follow-ing factors:

+ cost of funds+ expected loss on credit risk+ the annualised value of a single disbursement fee

and other fees115

+ (value of the prepayment option – prepayment fee)+ gross income of banks

interest rates and fees charged to customers

Based on the experiences of the Member States of theEuropean Union,116 the expected loss on credit risk, thenet prepayment fees and the profitability of banks gen-erate an interest premium of approximately 1.0–1.5 per-centage points. In Hungary calculating a 5% probabilityof default and 20% loss given default (LGD), a 1 per-centage point credit risk premium would result; and infunction of the fees charged, the net prepayment optionwould come to somewhere between 0–0.5 percentagepoints. Thus, the gross income content of the subsidiesextended by the government is 1.0–1.5 percentagepoints less than the gross interest margin achievablefor the banking sector as a whole.117

Due to the high rate of subsidised loans, the profitabili-ty of home financing is solely dependent on interestmargin ensured by the government. Due to the maxi-mum loan interest specified in the relevant statutoryregulation, however, subsidised loan products arepriced only on the basis of risks generated by prepay-ment and the one-off fees charged upon origination.

Interest margin on subsidised housing loans

In general, it can be established that until the Decree wasamended in June 2003, the margin achievable for thebanking sector grew, but significantly less margins couldbe realised on loans extended and financed subsequently.

Up to June 2003, a 7–9% margin was realisable on theleading products119 in the first 5 years following exten-sion and refinancing.120 The maximum margin wasachieved in the case of the 8% interest/yield on mort-gage bonds, while in the cases of any other interestrates profitability dropped. As commercial banks canobtain access to the subsidies conditional upon mort-gage bonds only through mortgage banks, they havelower margins as they share liabilities-side subsidies withthe financing mortgage banks.

In the case of loans extended and financed with the helpof mortgage bonds following the June 2003 amendmentof the Decree on subsidies, instead of the earlier 10%maximum, liabilities-side subsidies were modified to either105% of the yield on government securities of the sameterm issued in the immediately preceding three months orthe interest/yield of the mortgage bonds, whichever islower. In the case of subsidised loans with a term of noless than 5 years, the aggregate value of loan interest andservice charges on loans was maximised at 5% for newhomes and 6% for used homes. Further loss of incomeresults from the fact that the previous 2% one-off reim-bursement fee on financing was reduced to 0.5%.

Composition of the gross interest margin of housing loans in April/May 2003118

Chart 4-21

Credit risk Option of prepayment

Net interest income

0.000.501.001.502.002.503.003.504.004.50

LGD Credit risk10% 0.5%20% 1.0%30% 1.5%40% 2.0%50% 2.5%

Den

mar

k

Fra

nce

Ger

man

y

Ital

y

Net

her

lands

Port

ugal

Spai

n

Gre

at B

rita

in

Hungar

y

5.00Per cent

Source: Mercer–Oliver–Wyman

115 Subsequent calculations disregard the annualised value of single feed charged because of the average loan term of 10–15 years.116 Mercer–Oliver–Wyman: Study on the financial integration of European mortgage markets (European Mortgage Federation, study, October 2003).117 In Hungary’s case the approximately 4.7–4.8% banking sector interest margin ensured upon amendment of the decree on subsidies is indicated in the chart.118 In Hungary’s case the period to follow the amendment of the decree on 16 June 2003.119 This margin calculation does not refer to the special products in which loan repayment commences only following a certain grace period.120 In order to recover their costs incurred in relation to the sales of loans or the independent lien to mortgage banks (registration at the Land Registry, profit missed

etc.) banks received a single 2% reimbursement from the government.


4 ARTICLES

4

Pursuant to the new Decree on subsidies, mortgagebanks and their partner commercial banks are no longerinterested in the stabilisation of the interest/yield ofmortgage bonds at around 8%, but rather in pushingdown the yield premium as much as possible. Based on

experiences gained in the issue of mortgage bonds sofar the margin may be expected to come to around4.5–4.8%. In contrast to the earlier Decree, however,the margin can also be guaranteed for periods exceed-ing 5 years provided that the customers choose thenewly applicable 10-year fixed interest period and themortgage bank issues mortgage bonds of similar terms.

The 3.0–3.5 percentage point drop in the gross interestmargin will have its impact fully felt in the banking sec-tor with a delay (3–4 years), as it affects only thoseloans that are originated after the amendment of theDecree and earlier originated loans that are financed byre-issuing mortgage bonds during the whole loan term.

The income impact of prepayment

Prepayments may also affect the profitability of housingloans. Pursuant to the relevant statutory regulation,commercial banks are not allowed to preclude loan pre-payment for their customers, but most of them chargesignificant fees. The prepayment fee changes in func-tion of the date of prepayment: if it is made on the inter-est rate fixing day or any other day (1–4%), the incomeimpacts of which is given in the following formula:

Income risk = single prepayment fee – PV (interest mar-gin lost until the interest rate fixing day)

The revenues missed because of prepayment decreasewith time approaching the interest rate fixing day. Thepossibility of prepayment may be understood as a putoption provided for the customers. Western Europeanexperience indicates that the annual value of this optionis between 0.38–0.65%.122 If the 1–4% prepayment feesapplied in the Hungarian banking sector are divided

121 In this case the yield premium means the difference between mortgage bond yield and the average yield of a government security of equal term in the immediate-ly preceding three months.

122 Based on the calculations of Merrill Lynch Conditional Prepayment Rates (CPR) (in: Mercer–Oliver–Wyman).

Banking sector interest margin on subsidisedhousing loans before the June 2003 amendmentof the Decree on subsidies

Revenues Expenditures

Loans subsidised on the liabilities side

Customer interest rate: max. (6%) (7–9%) Yield on mortgage bondsSubsidies on the liabilities side:

min{(interest on mortgagebond + 2%);10%}

Loans subsidised on the assets-liabilities side

Customer interest rate:max. (4,5%) (7–9%) Yield on mortgage bonds

Subsidies on the asset side:(benchmark yield* –3,5%)

Subsidies on the liabilities side:min{(interest on mortgage

bonds – 1%);7%}

Total banking sector margin: ca. 7–9%

* Average yield of 5-year government securities shaped up at auctionsin the immediately preceding six months.

Table 4-9

Banking sector interest margin on subsidisedhousing loans after the June 2003 amendmentof the decree on subsidies

Revenues Expenditures

Loans subsidised on the liabilities side for purchasing used homes

Customer interest rate: (6%) (7–9%) Yield on mortgage bondsSubsidies on the liabilities side:min{(benchmark yield*105%);

mort. bond interest}–1%

Loans subsidised on the liabilities side for the purchase/construc-tion of new homes

Customer interest rate: (5%) (7–9%) Yield on mortgage bondsSubsidies on the liabilities side:min{(benchmark yield*105%);

mort. bond interest}

Total banking sector margin: ca. 3.4–5.4%

* Benchmark yield – average of the yields on government bonds of aterm identical with the mortgage bond in the immediately precedingthree months.

Table 4-10

Distribution of yield premiums121 on publiclyissued mortgage bonds by the end of 2003 H1and the margin on loans subsidised on the liabilities side after amendment of the Decree,as a function of the yield premium

Chart 4-22

0

1

2

3

4

5

6

7

3.0

3.5

4.0

4.5

5.0

5.5

Number of issues Marge in pct. point

Spread on gov. bonds in pct. point

Interest marge

Distribution of issues

by spread on gov. bonds

8 6.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

1.9

2.0


4 ARTICLES

4

into the average 5-year interest fixing periods, an annu-al income rate of 0.2–0.8% is the result. Thus, the aver-age 2% prepayment fee applied in the Hungarianbanking system112233 is theoretically in harmony with thevalue of the prepayment option.

Factors affecting the product profitability arising fromthe Decree on subsidies and the characteristic featuresof lending process

The income risk generated by the amendments of theDecree on subsidised housing loans

At the current interest rates and market yields, the prof-itability of subsidised loans is ensured exclusively bygovernment subsidies and their maintenance.Tightening of the Decree on subsidised housing loans(June 2003) reduces the profitability of products andthus also the entire bank system, along with its growthrate.112244

Deficiencies in pricing policies

Over the longer term, the significance of government sub-sidies will automatically drop simultaneously with theoverall decline of interest rates. As a consequence, the dif-ference between the interest rates and demand for mar-ket priced loans and subsidised loans may also decrease.However, loans bearing market interest rates are pricedwith a minimum of regard for the gradual decline of differ-ences between market and subsidised loans.

Income risks arising from extreme yield fluctuations

Subsequent to the amendment of the Decree on sub-sidised housing loans, the difference between mortgagebond yields and yields on government securities under-lying subsidies affects the product profitability. Due tothe 3-month averaging and delay of subsidies, suddenjumps in yield alter the product profitability in eitherway. A steep rise in market interest rates pushed theyield premiums of mortgage bonds up. As subsidies arecomputed from the average market yields of the pre-ceding 2–4 months, the cost of fund grows, while prof-itability decreases. If the trend turns, and yields sudden-ly drop, the opposite effect may be triggered. A slightyield fluctuation may offset these effects.

Delay in access to subsidies

Commercial banks can receive government subsidieswith a delay if 5–6 weeks, in general. This is primarilydue to a technical time requirement resulting from the

settlement of accounts. In theory, delay is 8–10 weeks,which is reduced only by the Finance Ministry’sadvance payment. The prime reason for delay lies inthe rigorous procedural rules of issuing mortgagebonds and, partly, the monthly timing of mortgagebond issues, due to capital market expectations. Bankscan recover their losses suffered as a result of delay bycharging prepayment fees, thus it does not alwaysinvolve profit missed.

SUMMARY

With a view to the banking practice and risk exposureof home financing, the following assumptions can bemade:

Credit risk is incurred primarily by the disbursing banks.At the end of 2002, the portfolio quality was significant-ly better than the quality of banks’ overall loan portfo-lio. Problem loans account for a very small share of mar-ket-priced and subsidised loans. In respect of the cover-age of housing loans, subsidised loans are sufficientlycovered by the maximised LTV ratio on evidence of theknown real estate market trends. In the coming twoyears, the proportion of late-paying customers is expect-ed to increase. The rate of problem loans within sub-sidised loans is expected to remain below 5% in thelong term.

Market risks are incurred primarily by mortgage banksand in some cases commercial banks. The reason is thatthe entry of long-term mortgage bonds into the balancesheets of commercial banks shifts banks’ interest riskconsiderably. Interest risk management is under devel-opment at mortgage banks, but measurement and man-agement has not yet been resolved. Extreme liquidityfluctuation is also insufficiently managed. The quality ofmanaging market risks is therefore not in line with theextent of these risks.

The most important potential operating risk is relatedto the condition of MIS systems. Due to the high rate ofsubsidised loans, the profitability of the home lendingbusiness is dependent primarily on the interest marginensured by the government. In our opinion, the 7–9%margin achieved earlier at a sector level, which declinedto 4.5–4.8% subsequent to the amendment of theDecree, provides sufficient coverage for potential risks.As a result of the expected deterioration in loan portfo-lio quality, a gradual decline of the margins and a likelyslowdown in lending growth in 2003, over the next 3–5years the profitability of the business line is expected togradually decrease.

123 Due to maximised loan interest rates, the prepayment fee is a strong competition factor.124 For instance, assuming the current portfolio amounting to approximately HUF 1,000 billion, a single decrease in the interest margin by 3.5 percentage points would

result in a nearly 10% drop in the banking sector’s total net interest income. The June amendment of the decree is less drastic, as drop in the margin will take placegradually in the coming 3–4 years.


4 ARTICLES

4

BIBLIOGRAPHY

Árvai–Dávid–Vincze: “Hitelinformációs rendszerek”[Loan information systems] (Hitelintézeti Szemle,October 2002).

Aspinwall, Richard C. & Eisenbeis, Robert A.:Handbook for banking strategy.

Bozsik, Sándor: “A lakáshitelezés és egyes makro-ökonómiai változók kapcsolata nemzetköziösszehasonlításban” [Relations between hous-ing loans and certain macroeconomic variablesin an international comparison], HitelintézetiSzemle 2002/3.

Goedecke, Wolfgang–Kerl, Volkher–Scholtz, Helmut:German Mortgage Banks, 1998.

Gombás–Marsi–Szalay: “A hitelkockázat kezelése amagyar bankrendszerben” [Credit risk manage-ment in the Hungarian banking sector] (MNBinternal study, December 2002).

Hoyland, Chris: Data-driven decisions for consumerlending.

Kaufman, George G.: Asset price bubbles: Implicationsfor monetary and regulatory policies.

Kiss, Gergely: The Housing Market and FinancialStability in the Light of EU Accession (MNBReport on Financial Stability, December 2002).

Mercer–Oliver–Wyman: Study on the FinancialIntegration of European Mortgage Markets(European Mortgage Federation, October 2003).

Nagy Vas, Erzsébet: The Risks and InstitutionalStructure of Housing Finance (MNB Report onFinancial Stability, December 2002).

Sirmans, C. F.–Benjamin, John D.: Pricing fixed ratemortgages: Some empirical evidence.

Thomas, Lyn C.–Edelman, David B.–Crook, JonathanN.: Credit scoring and its applications, 2002.

Valkovszky, Sándor: “The Situation of HungarianHousing Market, MNB Working Papers 3/2000.

Varga, Péter: “Hitelkockázat és kezelése II.” [Creditrisk and its management], Bankszemle 2000/9.

Vincze, Judit: “A jelzáloglevelek piaci helyzete ésfejlôdési irányai” [The market situation anddevelopment trends of mortgage bonds](Hitelintézeti Szemle, 2002/3).

Miscellaneous press materials published in economicdailies and weeklies.

CONDITIONS FOR THE MARKET ENTRY OF FOREIGN-OWNED BANKS

There were thirty commercial banks, three mortgagebanks, three building societies and two specialisedbanks with state responsibilities in the Hungarian bank-ing system in late 2002.126 Of the banks included in thisstudy, twenty-six were foreign-owned (hereinafter: for-eign) and seven domestic banks. Of the domesticbanks, three were 100%-owned subsidiaries of foreign-owned Hungarian banks. Thus, their foreign ownersalready had larger influence in 2002 than data on directownership suggested. Two of the remaining four weresold to strategic investors in the autumn of 2003. As tothe third, its privatisation via the stock market of over50% of its shares is currently being prepared.Transformation of the fourth into a mortgage bankfailed. Finally, in February 2003, it was transformed intoa financial enterprise. Owing to the situation outlinedabove, the importance of comparing the performanceof foreign-owned banks with that of domestically-owned banks has been taking an increasingly low pro-file in analyses of the Hungarian banking system. 2002was a turning point. It was then that foreign ownersacquired majority interest in the subscribed capital ofthe largest Hungarian bank.127

Foreign capital entered the Hungarian banking systemin four stages. The first stage goes back to the period ofthe single-tier banking system, prior to the politicalregime change, when three banks were established asjoint-venture banks with the involvement of the MagyarNemzeti Bank. The second stage was marked by theestablishment of the two-tier banking system in 1987and the entry into effect of Act XXIV of 1988 onEnterprises of Foreigners in Hungary. This was the timeof the liberalisation of banking and policies bolstering

the entry of foreign capital by granting favourable treat-ment. It was then that most of the foreign banks, still inbusiness, or their legal predecessors entered theHungarian market either through greenfield investment,or, in certain cases, partial privatisation or, in the case ofsmaller banks, full privatisation, via acquisition of major-ity ownership. At the time, i.e. during the first years oftransition, neither demand for the acquisition of majorbanks, nor supply of such banks on sale presented itself.This period, which ended in 1994, also saw both the cri-sis and the consolidation of the Hungarian banking sys-tem. Consolidation was crucial to the creation of bothdemand and supply. Debt consolidation, which, in fact,had three stages and the costs of capitalisation were amajor consideration when new privatisation strategieswere being devised. While a large amount of govern-ment stake in the wake of consolidation constituted thesupply needed for the privatisation of banks, portfoliocleaning and capitalisation aroused strategic investors’interest. The third stage was marked by the launch ofthe state privatisation programme in 1994 followingbank consolidation. During this period, until 1997, mostof the major Hungarian banks were privatised relativelyrapidly. Except for OTP, which was privatised throughstock exchange listing, all were sold to strategicinvestors. It was then that the government stake in thejoint-venture banks that had been among the first to be


4 ARTICLES

4

4.3 DR. JUDIT GELEGONYA: THE ROLE OF FOREIGN-OWNED BANKS

IN HUNGARY125

125 This study is based on a country study prepared as part of a background study for CEC5 Governors’ Meeting held in September 2003. A revised version of the back-ground study is available on the MNB’s website: Katalin Mérô and Mariann Endrész Valentinyi, ‘The Role of Foreign Banks in Five Central and Eastern EuropeanCountries’, MNB 2003.

126 This study relies on data on commercial banks and mortgage banks. The latter five institutions, including building societies and two specialised banks that have stateresponsibilities, have not been included in this study owing to the special characteristics of their business operation.

127 This turning point is highly likely to have occurred before 2002. The privatisation of the bank took place through stock exchange listing. As most of its shares andthe depository receipts thereof are traded abroad, the follow-up of the scattered ownership structure of the bank from the shareholders’ register has only becomemore reliable since the dematerialization of the bank’s shares.

128 5% participation in subscribed capital or below.129 Participation over 5%, but below 50%.130 Participation over 50%.

Number of commercial banks in Hungary

Designation 1996 1997 1998 1999 2000 2001 2002

Domestic bank128 11 6 8 7 5 5 7Domestic bankwith foreignparticipation129 2 4 3 1 1 1 0Foreign bank130 27 30 27 29 30 29 26

Table 4-11


4 ARTICLES

4

established was sold. The fourth stage started in 1998,and, in effect, continues to the present. However, it canno longer be described as a single, uniform process. Inthis stage, foreign banks were entering the market andleaving it as well. Although the sale of the two state-owned banks and one mortgage bank as the last act ofprivatisation may not necessarily lead to direct foreigncontrol, it will result in substantial foreign controlthrough indirect ownership.131 Aside from specialisedbanks with state responsibilities, there will be only onedomestically-owned bank among domestic credit insti-tutions in late 2003. Even so, privatisation through thestock exchange may result in a considerable rise in thenumber of foreign owners.

Regulatory environment

Immediately after the two-tier banking system was estab-lished, with the market opened and liberalised, foreigncapital was encouraged to enter the Hungarian bankingmarket with, among other things, favourable tax treat-ment. Both foreign and joint-venture banks foundedprior to 1990 were unconditionally eligible for substan-tial, long-term tax relief.132 As a result, positive discrimina-tion gave foreign and joint venture banks a competitiveadvantage over state-owned banks, which were alreadyfacing difficulties. Initially, tax relief was a major consid-eration in market entry.133 In some cases, the fact thatmultinational companies, especially those from the samecountries as banks, and entering the market roughlysimultaneously, could become the customers of thesebanks was another factor that encouraged the founda-tion of banks. Market entry, however, was not complete-ly without restrictions: prior to 30 June 1996, pursuant tothe Act on Financial Institutions, in order for a financialinstitution with a foreign stake amounting to 10% of thesubscribed capital to be founded, or in order for foreignparticipation to be acquired, preliminary governmentauthorisation, approved by the president of the MNB,had to be obtained.

First, the Act of 1991 on Financial Institutions, and thenthe Act of 1996 on Financial Institutions, and even moreimportantly the 1998 amendment of the latter, were sig-nificant steps towards creating a regulatory environ-ment that was in line with EU Directives. Nevertheless,there were a few rules and regulations, either new onesor existing ones, which limited the scope of services

that foreign-owned institutions were allowed to pro-vide.134 With Act XCIII of 2001 entering into effect on 1January 2002, foreign exchange restrictions were lifted.However, the required amount of dotation capital need-ed for opening a branch, the amount of which is equalto the required minimum amount of the initial capital ofcredit institutions, and the requirements for manage-ment composition are still effective.135 The remainingrestrictions will be removed in 2004, when Hungaryjoins the EU.

Consolidation of the banking system and privatisation

In the first half of the transition period, the transforma-tion crisis led to technical insolvency in the Hungarianbanking system as well. It became obvious as early as1992 that the banking system in its own right wasunable to improve the situation, and that the state con-solidation of the banking sector could no longer bedelayed. The true proportions of problems only becameclear in late 1993 and early 1994. Consensus on themeasures to be taken so that such problems could besolved also took a long time to emerge.136 Eventually,two types of consolidation materialised in severalstages. First, centralised credit consolidation and portfo-lio cleaning took place in 1992 and 1993, the result ofwhich, a substantial improvement of banks’ portfolio,proved transient. Gradual capitalisation put an end tothe chronic undercapitalisation of banks. As a result ofsuch capitalisation affecting eight banks, through pri-mary capital increases, direct state ownership in themajority of Hungarian-owned banks grew to 75 to 95%,and amounted to nearly 70% in the entire banking sys-tem in late 1993. Of the four fully capitalised largebanks, one was privatised with state assistance in late1995. As to the other three banks, the negative amountof their 1994 accumulated profit reserves was dissolvedby lowering the amount of their subscribed capital to arealistic level in 1995. With this, the process of stateconsolidation was completed. The reduction of capitalleft the ownership ratio unchanged. Consolidation cre-ated supply for bank privatisation through a large massof government stake. Portfolio cleaning and recapitalisa-tion were preconditions for arousing strategic investors’interest and generate demand.

In order to stabilise and strengthen the banking system,the government approved the privatisation programme

131 Of the two banks sold in the autumn of 2003, one passed into direct, and the other into indirect 100% foreign ownership.132 See Várhegyi (1995), Várhegyi (2002) and Majnoni et al. (2003).133 Foreign banks entering the market after the Act on Financial Institutions took effect on 1 December 1991 could no longer enjoy such advantages. Pursuant to the

Act on Financial Institutions, both favourable tax treatment and tax exemptions that the various authorities had granted to partly or wholly foreign-owned financialinstitutions were repealed on 31 December 1995.

134 For a more detailed treatment of the issue, see Király et al. (2000).135 The board of directors shall include at least two board members, who are of Hungarian nationality, qualify as FX residents under FX regulations and have been per-

manent residents of Hungary for at least 2 years.136 For a detailed description and contemporary criticism of the process of bank consolidation, see Balassa Ákos (1996); for a short summary and assessment in English,

see György Szapáry (2001) and Éva Várhegyi (2002).

of state-owned banks. Under this programme, sevenbanks, including the largest one, were privatised over aspan of three years.137 Six banks were sold to strategicinvestors, mostly through the sale of minority sharepackages. The government continued to have minorityownership in each bank. The EBRD also acquired minor-ity interest in three banks through privatisation. In theyears following the acquisitions, the strategic owners ofthe banks concerned bought out government stakesand, on a number of occasions, the share packages oftheir partners in privatisation. OTP, the largestHungarian bank with a dominant market share, was notsold to strategic investors. Rather, it was privatisedthrough the stock exchange. In that manner, the govern-ment managed to achieve a double target: the bankremained, at least in the very first years, in the hands ofHungarian owners, and the development of theHungarian capital market received a great impetus.

Ownership structure of foreign and domestic banks

In the fourth stage of market entry (from 1998 to 2001),along with the banks already in foreign majority owner-ship, OTP was the only other bank left and it was char-acterised by foreign participation over 5% with a steadyincrease. Strategic investors gradually bought out theshares that domestic shareholders still had in banks. Atend-2002, the foreign stake in 22 banks amounted to orexceeded 99%. In 2001, state ownership in foreign-owned banks was terminated. In 2002, the foreign own-ership in the subscribed capital of domestic banks dis-appeared due to the largest bank changed its groupaffiliation. Through their combined 87% participation inthe subscribed capital of Hungarian banks, foreign-owned companies controlled over 90% of the balancesheet total of the Hungarian banking system in 2002.

In 2002, based on their subscribed capital, the majorityof foreign owners (78%) were from nine EU countriesand 12% from the USA. In terms of overall foreign own-ership, the largest European owners include Austria(24%), the Netherlands (14%), Germany (13%),Belgium (12%) and Luxembourg (9%). The nationalityof a few owners of two listed companies, holding 4% ofthe shares, is unknown. Owners of various nationalitieshold the remaining shares.

A banking system nearly completely under foreign con-trol poses considerable risks to the host country if thegroup of owners is highly concentrated both by compa-ny and by country. The ownership structure of theHungarian banking system is currently adequately diver-sified. In fact, it is better diversified than in other coun-tries in the region. Even the largest amount of the sharesheld by one shareholder is below 5% of the totalamount of capital, and that of the shares held by one

shareholder per country is below 20%. The respectiveportfolios of the majority of strategic investors inHungary are adequately diversified by country andregion alike, although investments by some are concen-trated in the Central and Eastern European region.

While no opportunity has arisen in CEECs so far for theresponse of foreign banks to a crisis situation to beobserved, strategic investors’ rapid exit from the marketis unlikely, since such exit would be prolonged process


4 ARTICLES

4

Distribution of foreign ownership by countries in subscribed capital at end-2002

Chart 4-23

Austria

24%

Germany

13%

Netherlands

14%

USA

12%

Belgium

12%

Luxemburg

9%

France

4%

Others

12%

137 The top bank of building societies, which passed into majority state ownership through consolidation, was sold to foreign investors in 1997.

Ownership structure of foreign and domestic bankson the basis of subscribed capital in percentage

Ownership structure of domestically-owned banks

Designation 1998 1999 2000 2001 2002

– state 30.6 34.6 40.5 35.8 49.7– other domestic 52.6 46.3 38.2 40.3 49.5– foreign 14.2 16.2 18.4 21.2 0.0– shareholders of

preferred shares 1.5 1.7 1.9 0.2 0.0– own shares

bought back 1.1 1.2 1.1 2.5 0.7

Ownership structure of foreign-owned banks

Designation 1998 1999 2000 2001 2002

– state 5.9 4.9 1.7 0.0 0.0– other domestic 3.3 4.6 4.4 3.2 4.1– foreign 87.5 87.7 90.9 92.3 90.7– shareholders of

preferred shares 2.2 2.5 3.0 4.5 4.7– own shares

bought back 1.0 0.3 0.0 0.0 0.5

Table 4-12


4 ARTICLES

4

and result in losses of extraordinary proportions.Responses from strategic investors so far have beenfavourable in Hungary. All capital crunches faced bytheir subsidiaries have been tackled through eitherdirect capital injection or the provision of subordinatedloan capital.

The largest Hungarian bank is in the majority ownershipof foreign portfolio investors.138 This is inherently charac-teristic of foreign presence in a country in a regionwhere the capital market lags considerably behind thatof developed countries. Such characteristic also carriesparticular risks. The sale by portfolio investors of listedshares, despite their losses on stock exchange prices, isfar easier to make, and is less time-consuming. As inother countries in the regions, the volatility of stockexchange prices is high, plummeting prices in the capi-tal market may generate a huge wave of sales by for-eign owners. This is exactly what happened in theautumn of 1998. Today, however, it is highly question-able whether there would be a sufficient number ofinvestors, be they domestic or foreign, with a sufficientamount of capital, to buy up this bank’s shares, withoutthe state having to intervene, if there were a similarlysignificant drop in stock exchange prices with the result-ing massive sales by non-residents. Despite growth andexpanding foreign participation, the bank’s stockexchange capitalisation is relatively low, so it may wellbe a potential target of buyouts. Although potential buy-ers include primarily financial investors, a hostile buyoutcannot be ruled out either. The latter, however, wouldtake a considerable length of time because of thebank’s rather dispersed ownership structure.

Today the risks indicated above do not seem to be con-siderable, and will be further reduced by Hungary’saccession to the EU in 2004.

MARKET ENTRY STRATEGIES

The underlying reasons for entry by foreign banks intothe Hungarian banking market varied from one phaseto the next, and were often different even during thevarious the phases.139 In each phase, there was at leastone example of a bank following its customers toHungary, or preparing the ground for entry by new cus-tomers from the home country. The ultimate incentive,however, was the opportunity of reaping additionalprofits as a consequence of the relative underdevelop-ment of the financial intermediary system in the lessdeveloped markets and the growth potential offered bythe deepening of intermediation, which, compared withother transition or emerging regions, offered itself for

those interested under acceptable conditions in termsof regulations and infrastructure.

In line with their strategies, the majority of foreign banksinvesting in Hungary followed strategic objectives andsimultaneously acquired equity interests in the bankingmarkets of other countries in the Central and EasternEuropean region. Banks entering in the first and secondphases with greenfield investment started to build theirnetworks on their own, and increased their marketshares through corporate mergers, bank take-overs andacquisitions of lines of business. The banks participatingin privatisation acquired substantial shares of the marketby purchasing markets. These banks continue to beamong the largest.

A wide variety of factors were among the underlyingreasons for the majority of other small foreign-ownedbanks commencing business in Hungary. Some bankswere able to take root, while others failed and havesince withdrawn from the market. Based on the experi-ence of past years, in the first three phases all banksattempted to provide universal services to the market,apart from a few exceptions, with more or less success.But in the fourth phase only banks targeting narrow seg-ments or niches of the market succeeded.

The majority of Hungarian subsidiaries have met theowners’ requirements, and the strategic investors haveundertaken substantial capital increases and technolog-ical development. From the onset, these banks offeredthe full range of traditional bank services to their cus-tomers and, and they have taken the opportunity tomake advances in the area of universal banking whichencompass investment services as well. However, somehave been successful by narrowing the range of servic-es they offer. The majority of banks performing modest-ly in the areas of efficiency and profitability have contin-ued to operate as traditional banking institutions, with-out making a shift either towards universal banking orniche market strategy.

EFFECTS OF FOREIGN BANKS’ MARKET ENTRY

Competition

New market entrants, including an increasing numberof foreign banks, shattered the previous monopolistic oroligopolistic structures. As a result, by the end of the1990s the banking market had indeed become less con-centrated. From a level of 1,460 in 1993140 theHirschman-Herfindahl Index (HHI) fell considerably: in1996 it was in the lower range of the moderately con-

138 The majority of the shares of the largest Hungarian mortgage bank are to be sold to portfolio investors via the stock exchange. As far as the other, still public, com-pany is concerned, its strategic owner wishes to exit from the public market through buying up the company’s shares.

139 For more details, see Éva Várhegyi (2001), and Majnoni et al. (2003).140 Source: Várhegyi (2001).

centrated band, while in 1999 and 2000 it reached therange considered competitive. Due to a series of merg-ers between large banks, in 2001 concentration intensi-fied just to drop near the competitive band once againin 2002. A similar trend is seen in the market shares ofthe five largest and the ten largest banks. In 2001, thefive largest banks had a lower share in Hungary (61.2%)than in Belgium (78%), Denmark (68%), Finland (80%),Greece (66%), the Netherlands (82%) or Sweden(88%) in the same period based on data for all creditinstitutions.141

Structural changes clearly reflect the increasing pres-ence of foreign ownership. While under the prevailingcircumstances, only small and medium-sized foreignbanks were allowed to have majority ownership in thefirst two stages, following privatisation and in the wakeof the mergers over the past few years, the number ofnon-resident owners has steadily increased among thelargest banks. As in 2002 ownership title to the largestHungarian bank was changed, each of the five and nineof the ten largest banks passed into foreign hands. Thenine largest banks consist of five privatised banks, all thethree banks established as joint ventures prior to theexchange of regimes and purchased later by strategicowners, and a bank originally established as a greenfieldinvestment and later on merged several times.

Market

Naturally, in the first two stages of market entry, foreignbanks gained a foothold in corporate lending. Theirretail banking activity remained scant, with the clientelecoming primarily from among the employees of foreigncompanies who had moved to Hungary.

In terms of market segments, by 2002 the corporatemarket had become competitive in terms of both lend-

ing and deposit collection. Currently, there is no moregrowth potential in lending to large companies, andsteady growth in SME lending is still further ahead. Thelargest change can be seen in the retail market, whereOTP used to have a monopoly with 98.4% of retailloans and 93.2% of retail deposits in the beginning ofthe 1990s,142 whereas in 2002 it accounted for corre-sponding shares of 27.8%143 and 46.0%.

Following privatisation, the four privatised large banksengaged in intense competition for the relatively cheapretail deposits.144 They rapidly increased their shares atthe expense of OTP, and later on also Postabank. Lately,a few banks that were considered as small or medium-sized in the mid-1990s have successfully joined thisaggressive market penetration. They obviously wish toincrease their market shares, which are relatively lowconsidered on the basis of their balance sheet totals, inthe retail market at an above-average pace and insteadof allocating the acquired funds in other markets, theyhave a high loan-to-deposit ratio within the same marketsegment. Although every foreign bank expanded rapid-ly in the retail market in terms of both lending anddeposit collection, penetration into the two partial mar-kets is asymmetric: a higher concentration is seen in thedeposit market. The reason for this is that in 2002 onlytwo of the six domestic-owned banks collecteddeposits, three of the remaining banks are mortgagebanks, while one of them is specialised in motor vehicleloans and has an insignificant amount of retail deposits.

Despite drastic changes, OTP has managed to retain itsleading position in the Hungarian banking sector;Furthermore, for several years professional circles ratedit as the best Hungarian bank and it has proven attrac-tive on Hungarian and foreign stock exchanges alike.Uniquely in the Central and Eastern European region,under the control of completely Hungarian leadership,


4 ARTICLES

4

141 The Swedish data refer to the year 2000. Source: ECB (2003).142 Source: Bonin–Abel (2000).143 This picture is somewhat misleading as the second place is occupied by one of the bank’s subsidiaries, the mortgage bank leading the home purchase financing

market. Their joint market share was 42.9% in 2002.144 For more details see Bonin, J.–Abel, I. (2000).

Market concentration indicators on banks’ total assets

1996 1997 1998 1999 2000 2001 2002

Share of the five largest banks (in %) 59.1 54.9 56.2 56.1 55.8 61.2 59.3Of which: share of foreign banks (in %) 13.7 22.1 24.0 31.0 31.8 37.4 59.3

number of foreign banks 2 3 3 4 4 4 5Share of the ten largest banks (in %) 76.5 73.4 75.9 76.0 76.3 81.0 79.2Of which: share of foreign banks (in %) 31.0 40.6 43.8 46.3 48.2 53.0 75.2

number of foreign banks 7 8 8 8 8 8 9HHI – market 1,156 1,020 1,039 999 965 1,048 1,016HHI – foreign banks 185 277 315 348 371 463 1,008Total number of banks 40 40 38 37 36 35 33Total number of foreign banks 27 30 27 29 30 29 26

Table 4-13


4 ARTICLES

4

high profitability has enabled the bank to expand acrossthe borders and purchase foreign banks, despite itsrestricted capital formation capacity resulting from therelatively high dividend payment obligation. Whether ornot the bank’s ambitious attempts at undertaking amajor role in the region will yield similar results as inHungary cannot yet be predicted at the moment.

The relevant literature gives several reasons for OTP’ssuccess. In their analysis of the retail market, Bonin andAbel145 highlight competition as a factor threatening OTPin the largest measure on the part of foreign banks. Intheir opinion, loss of market suffered as a result of foreign

banks’ penetration prompted the bank to modernise itsproducts and services. Based on studies by Bikker et al.,and Molyneux, Várhegyi146 stresses contestability as athreat which is further augmented by a low threshold toentry, deregulation, scarce market segmentation by theregulatory regime and the technical development.

In addition to the above-mentioned factors, the privati-sation technique itself, namely privatisation through thestock exchange, also contributed to success. OTP’s his-tory after the mid-1990s clearly proves that with thehelp of the regulatory force of the market, daily stockexchange measures of strength, and the disciplinarypower of frequently applicable and strict disclosurerequirements, in a transition economy a bank in long-term domestic majority ownership can neverthelesssuccessfully compete with tough rivals having advan-tages right at the start, as they benefit from the power-ful control of strategic owners and the direct import ofhigh-tech solutions and know-how.

FOREIGN BANKS’ PERFORMANCES

Portfolio

Portfolios in foreign banks have always been of signifi-cantly higher quality than those held by domestic banks.In the first two stages of market entry, such a differencewas just natural, as foreign greenfield banks were notburdened by inherited bad loans, had a high-rated clien-tele (initially they recruited new customers from amongtheir compatriots in Hungary, and later on cherry-picked

Corporate and retail market concentration

Corporate Retail

Deposit Loan Deposit Loan

Share of the five largestbanks (in %) 56.4 62.6 76.5 62.3Of which: share of banks in foreign majority ownership (in %) 56.4 62.6 69.9 62.3

number of banks in foreign majority ownership (in %) 5 5 4 5

Share of the ten largestbanks (in %) 81.5 85.2 93.3 81.4Of which: share of banks in foreign majority ownership (in %) 81.5 82.5 86.6 62.8

number of banks in foreign majority ownership (in %) 10 9 9 9

HHI – market 832 962 2,471 1,244

Table 4-14

Quality of bank portfolios

1998 1999 2000 2001 2002

Foreign Domestic Foreign Domestic Foreign Domestic Foreign Domestic Foreign Domesticbanks banks banks banks banks banks banks banks banks banks

Totalportfolio

Share ofclassified

assets 8.4 13.5 6.9 14.6 6.8 11.8 8.9 10.6 6.8 13.7

Share ofnon-performing

assets 2.4 9.6 2.4 7.7 1.6 6.5 1.8 4.8 1.8 6.1

Loans Share ofclassified

loans 17.0 22.8 13.0 17.6 11.8 11.9 15.1 10.3 13.6 14.5

Share ofnon-performing

loans 4.5 15.3 3.8 6.5 2.4 6.1 2.4 4.4 3.2 5.8

Table 4-15

145 Bonin, J.–Abel, I. (2000).146 Várhegyi, Éva (2001).

the best Hungarian companies) and due to the parentbanks, in terms of risk management routines, they arefar ahead of Hungarian banks. Presumably, in their inter-nal rules their parent banks set up the same require-ments as applied in their home countries, and so theshortages of contemporary Hungarian regulation andsupervision notwithstanding, they did not becomeembroiled in irresponsible lending practices.

In the fourth stage following privatisation, the quality ofportfolios held by foreign banks grew better than thoseheld by domestic banks each year, in each classificationcategory and portfolio segment, despite the fact that inthe period reviewed all of the successor banks of theformer one-tier banks system had already gone into for-eign ownership. Until 2000, portfolio quality improvedin both groups. The deterioration observed in the lastyear in the domestic group was primarily due to struc-tural reasons: the largest bank, which had a remarkablygood portfolio among domestic banks, changedgroups. In the group of foreign banks its performance isrelatively low (above all in respect of non-lending bal-ance sheet items). By virtue of its size, the bank offsetimprovement in other banks, and thus in 2002 thegroup indicator did not fall. The deterioration in domes-tic banks’ portfolios was caused on the one hand by thedeparture of a bank with a good portfolio, and on theother, further impairment of the remaining banks’ port-folios, which affected nearly every segment.

Within the group of non-performing portfolios (substan-dard, doubtful and bad), the average rate of bad assetsdoes not differ significantly between the two groups.However, in the case of domestic banks, doubtful assetsas a proportion of the overall portfolio is on average near-

ly twice as high as the corresponding figure in foreignbanks. The difference is even greater in the quality of loanportfolios. Disregarding the outlying 1998 data, indomestic banks the average rate of bad loans within non-performing loans was nearly double the amount held byforeign banks between 1999 and 2002, while the propor-tion of doubtful loans amounted to more than two and ahalf times as much as those held by foreign banks.

As of 2001 H2, deteriorating economic conditionshave led to slower business activity in the MemberStates of the European Union as well. According to anECB analysis147 of nearly every credit institution withinthe EU, the above fact, coupled with a loss of confi-dence in the reliability of corporate data in the wake ofEnron’s scandal and other cases, lead to a revaluationof portfolio quality. This, in turn, resulted in a moderateportfolio deterioration and an increase in lower-qualityelements and reserves as early as late 2001. InHungary, however, no trace of this trend was dis-cernible in foreign banks’ rating and provisioning prac-tice even in 2002. There are three fundamental reasonsfor this. First, economic growth was stronger inHungary than in the EU. Second, in the election year aswell as in the preceding one, additional subsidies andincome was made available to smaller companies andhouseholds, and this reduced default risk in the groupsconcerned. Third, less risky housing mortgage loans,classified as problem-free upon disbursement,boomed, improving the portfolio.

Profitability

Banks’ performance indicators give an extremely variedpicture for the five years under review.


4 ARTICLES

4

147 ECB (2003).

Share of non-performing loans in the loan portfolio

Chart 4-24

Foreign banks

0

2

4

6

8

10

12

14

16

18

1998

1999

2000

2001

2002

Bad Doubtful Sub-standard

Share of non-performing loans in the loan portfolio

Chart 4-25

Domestic banks

0

2

4

6

8

10

12

14

16

18

1998

1999

2000

2001

2002

Bad Doubtful Sub-standard

With the exception of 1998, the spread and margin – thefundamental indicators of income structure – tended to besignificantly higher in domestic banks than in foreign ones.

From 1999, the spread increased year by year in bothgroups (in the case of domestic banks up to 2001), asdue to the relatively underdeveloped financial interme-diary system, both groups can cut interest rates soonerand to a larger degree in the deposit business thanthrough loan repricing. Growth may be explained bythe application of risk premiums stemming from portfo-lio deterioration. In the case of foreign banks, the datarelevant to the deterioration of the loan portfolio con-firmed this as of 2000, however, in the case of domes-tic banks the data do not.

Throughout the period under review, domestic banksregistered a higher interest margin. Up to 2001, this wasdue primarily to the outstanding achievement of OTP,and to a minor extent, in terms of volume, the high indi-cators of new mortgage banks. Difference between thetwo groups were more muted after OTP transferredinto the other group. Foreign banks’ interest margin isconsistently higher, whereas the spread was lower thanthe average of EU credit institutions as of 1999. With aview to the total margin, the difference is smaller. Thismeans that the ratio of interest and interest revenues toprofit earned on commission is higher in domesticbanks’ results than at foreign banks.

Banks’ cost efficiency has not improved in the requiredmeasure. There is no significant difference between foreign

and domestic banks in terms of the costs to total assetsratio, which considerably exceeded the average of EUcredit institutions, up from 1.5% in 1998 to 1.7% in 2001.

Over the past few years rates of return on assets andequity have gradually improved in the majority of banks,despite the fact that no improvement is reflected in theoverall development of the ROA and ROE indicators. In1998, domestic banks, while in 1999 foreign banks sawparticularly low profits, attributable to losses made bylarge banks as a result of certain one-off reasons. Startingfrom 1999, OTP’s results dominated the domesticgroup’s performance. When OTP was reclassified as abank in foreign ownership in 2002, the domestic group’sprofitability drastically fell, even though no major orextraordinary losses was incurred in the group.

In 2002, there were altogether six loss-making banks:four of them were the smallest foreign banks and twodomestic ones. Dispersion dimensions were extremelywide. ROA moved between –4.5% and 5.8%, whileROE was between –13.2% and 61.2% in foreign banks.Extreme values were recorded by relatively new smallbanks. Dispersion was less extensive in the case ofdomestic banks: ROA was between –0.5% and 3.1%,and ROE between –4.6% and 26.1%.

In a regional comparison, Hungary has a longer andmore varied experience in the analysis of the role of for-eign-owned banks in transition economies, as therestructuring and privatisation of the banking sectorstarted earlier than in the other countries. The time thathas elapsed from the beginning to the present can beconsidered sufficient for the purposes of empiricalanalyses, the relevant data are available in good quality,and as the most essential measures of the bank reformhad been taken before the period under survey, thecomplicating factor of simultaneous bank reforms canbe regarded as having a minimal impact. The results ofan empirical survey based on the analysis of data fromindividual banks active in Hungary148 between 1995-2000 provided econometric support of the statisticalobservation that Hungary’s banks in foreign ownership


4 ARTICLES

4

148 Majnoni, G., Shankar R., and Várhegyi, Éva (2003): “The Dynamics of Foreign Bank Ownership: Evidence from Hungary”, World Bank Policy Research WorkingPaper 3114, August 2003 The World Bank. The sample contained data for 26 commercial banks active between 1995–2000. The authors disregarded the 1998 datafor Postabank in the analysis of profitability.

Profitability indicators for foreign and domesticbanks

1998 1999 2000 2001 2002

Foreign Interest rate spread 3.9 3.2 3.2 3.4 3.9banks Net interest margin 4.4 3.6 3.6 3.8 4.1

Total margin* 6.0 5.4 5.5 6.0 6.2

Total costs**over total assets 3.9 4.0 3.7 3.7 3.6

Domestic Interest rate spread 5.1 4.8 4.9 5.3 4.9banks Net interest margin 5.0 5.0 4.8 5.1 5.0

Total margin* 3.2 6.5 6.5 6.7 5.8

Total costs**over total assets 4.3 4.1 4.0 4.1 4.8

* Net profit on financial and investment services excluding provisions(value adjustments) and other income from operations/total assets(chronological average).

** Operating costs/total assets (chronological average).

Table 4-16

Profitability of foreign and domestic banks

Description 1998 1999 2000 2001 2002

Foreign banks ROA* 0.7 0.0 0.9 1.4 1.5ROE* 7.7 –0.2 10.8 16.9 18.1

Domestic banks ROA* –5.9 1.1 1.4 1.5 0.2ROE* –96.2 17.8 18.9 20.2 1.7

* Chronological average calculated from profit after tax.

Table 4-17

are significantly more profitable than domestically-owned banks.149

The analysis supplemented the results of earlier studieson the subject in three fields: (1) it took explicit accountof the time elapsed since market entry, during whichthe ownership change has (or may have) made itseffects felt in the performance of banks; (2) it examinedthe interrelationship between acquisition, the manner ofmarket entry (greenfield investment vs. corporate take-over) and efficiency/profitability; and (3) it examinedthe effects of managerial methods and confidence inforeign management.

In the field of cost efficiency, the only significant resultwas the fact that foreign-owned banks employing localmanagement were more efficient in cost saving. In thecase of greenfield investments, the regression analysisjustified the empirical conclusion that both operatingand staff costs were lower than in the case of bankstaken over by foreign owners. In the latter case inherit-ed factors, an ineffective branch network, underdevel-oped IT and low skilled labour force pushed up costintensiveness.

An analysis of profitability revealed that irrespective ofthe management style applied, foreign banks are farmore profitable than domestic ones, and this offsets thefact that in terms of cost level and sometimes even coststructure, they are very similar to local banks. This sug-gests that higher revenues realised by foreign banks areattributable rather to a wider variety of services than thedistribution of traditional products, as local managershave a comparative advantage in the latter. Regardingprofitability, the banks that were established by way ofgreenfield investments unambiguously exceed not onlydomestic banks but also banks acquired by foreigners inthe framework of take-over. Results related to lendingspread justify the assumptions for the reasons of foreignbanks’ higher profitability. Narrowing spread is consis-tent with the overall decrease in fees accompanying for-eign banks’ entry, and highlights the fact that higherprofitability is attributable to a wide variety of servicesand a higher quality loan portfolio rather than to highermargins.

The volume of lending, tested last, was the single fieldwhere greenfield banks did not differ substantially fromother foreign banks. The results are insufficient to justi-

fy the wide-spread opinion that foreign banks broadenaccess to credit markets. In any case they are sufficientto confirm than in a country like Hungary, where thepresence of foreigners is comprehensive and affectsbanks of all sizes, foreign-owned banks do not decreaselending activity.

Capitalisation

Starting from 1998, regulatory capital grew continuous-ly in both groups; the capital adequacy ratio fluctuatedin line with risk-taking behaviour, but it remained appro-priate at all times. Disregarding the impacts of OTP’sreclassification, it is clear that the lending boom in 2002was reflected in a decrease of CAR in both groups,which resulted from the fact that the risk-adjusted totalassets increased much faster than regulatory capital. Amajor expansion of domestic banks is due to the homeloan activity of the new mortgage bank entering themarket, as in the very first year of its operation it forgedahead to the second place in household lending.

In respect of stability it should be noted that all thebanks met the minimum regulatory requirement of 8%capital adequacy ratio. Five foreign and two domesticbanks failed to meet at least the 10% ratio consideredas safe. Similarly to profitability, the CAR figures showwide dispersion. Foreign banks’ CARs varied between8.1% and 98.4%, the highest extreme was recorder fora relatively new small bank and the low value barelyexceeding the minimum requirement was reported by alarge bank undergoing strong expansion. Domesticbanks had CARs between 9.2% and 39.6%, if the bankwithdrawing from the market in 2003 is disregarded.

In 2002, Hungary had a reasonably stable and soundbanking system. This was to a large extent due to therole of foreigners who appeared in the region as strate-gic investors and provided for their input in terms ofexpertise and know-how.


4 ARTICLES

4

149 Based on the results of studies published on the subject, no consistent picture can be formed of foreign banks’ performance in Hungary and, in a wider context, intransition economies and other emerging countries, as the frequently contradictory conclusions are drawn from analyses based on different models applied for dif-ferent periods and samples. For further details see Mérô–Endrész Valentinyi (2003).

Banks’ capital adequacy (Capital adequacy ratio)

1998 1999 2000 2001 2002

Foreign banks 15.8% 13.8% 13.0% 13.7% 13.0%Domestic banks 13.7% 15.2% 16.1% 13.9% 12.1%

Table 4-18


4 ARTICLES

4

Majnoni, G., Shankar, R. and Várhegyi, É. (2003): “TheDynamics of Foreign Bank Ownership: Evidence fromHungary”. World Bank Policy Research Working Paper3114, August 2003 The World Bank.

Mérô, Katalin and Endrész Valentinyi, Mariann(2003): “The Role of Foreign Banks in Five Central andEastern European Countries”. MNB 2003.

Molyneux, P. (1999): “Increasing concentration andcompetition in European banking: The end of anti-trust?” EIB Papers, Vol. 4 No. 1 1999.

Szapáry, György (2001): “Banking Sector Reform inHungary: Lessons Learned, Current Trends andProspects”. MNB Working Papers 2001/5.

Éva, Várhegyi (1995): “Bankok versenyben” (Banks incompetition). Pénzügykutató Rt. Budapest.

Várhegyi, Éva (2001): “Foreign ownership inHungarian Banking Sector”, Közgazdasági Szemle,Vol. XLVIII. No. 7–8 2002 (in Hungarian).

Várhegyi, Éva (2002), “Hungary’s banking sector: Achi-evements and challenges” EIB Papers, Vol. 7 No. 1 2002.

REFERENCES

Balassa, Ákos (1996): “Restructuring and RecentSituation of the Hungarian banking Sector”. MNBOccasional Papers 10.

Bikker, J. A. and Groeneveld, J. M. (1998):“Competition and Concentration in the EU BankingIndustry”. De Nederlandsche Bank, Research SeriesSupervision (8) In: Kredit und Kapital, Heft 1/2000.

Bonin, J.–Abel, I. (2000): “Retail Banking in Hungary:A Foreign Affair?” (Prepared as a background paper forthe World Bank, World Development Report 2002:Institutions for Markets).

“EU Banking Sector Stability”. ECB, February 2003.

Király, Júlia–Májer, Bea–Mátyás, László–Öcsi,Béla–Sugár, András–Várhegyi, Éva (2000): “Ex-perience with Internationalization of FinancialSector Providers–Case Study: Hungary”, in: Claessens,S. and M. Jansen (eds) (2000) The Internationalizationof Financial Services, The World Bank and WorldTrade Organisations. Kluwer Law International407–435.

Index of charts and tables

Charts


Chart 1-1 Global indicators of risk 14Chart 1-2 HUF/EUR exchange rate 15Chart 1-3 Changes in implied volatilities relating to the forint exchange rate 16Chart 1-4 Benchmark yields on government securities and the MNB’s major policy rate 17Chart 1-5 Distribution of most likely dates of EMU entry 17Chart 1-6 One-year real interest rates 17Chart 1-7 Non-resident holdings and average maturity of government securities 18Chart 1-8 Maturity profile of non-residents’ government securities holdings 19Chart 1-9 Non-residents’ share of the government securities market in the various maturity brackets 19Chart 1-10 Changes in government securities holdings according to maturity between January–November 20Chart 1-11 Major stock exchange indices 20Chart 1-12 Central and Eastern European stock exchange indices 20Chart 1-13 GDP growth 21Chart 1-14 Hungary’s market share in the EU 21Chart 1-15 Corporate investment 23Chart 1-16 CPI and core inflation 24Chart 1-17 Actual inflation and expected inflation rate for 12 months ahead 24Chart 1-18 Corporate managers’ inflation and wage expectations in the TÁRKI survey 25Chart 1-19 Inflation expectations based on the Reuters poll 25Chart 1-20 Net lending of sectors and current account deficit as a proportion of GDP as per current

and 2004 methodology 26Chart 1-21 Gross savings of households (and their impact on financial savings and fixed investment)

as a proportion of disposable income 27Chart 1-22 Current account balance and structure of financing as a percentage of GDP 28Chart 1-23 Hungary’s FDI balance 28Chart 1-24 FDI inflow as a percentage of GDP based on balance of payments statistics 30Chart 1-25 Chart FDI inflow as a percentage of GDP with comparable content 30Chart 1-26 International reserves 32Chart 1-27 International reserves compared to various monetary aggregates 33Chart 1-28 International reserves as a percentage of whole-economy short-term debt and one-month imports 33


Chart 2-1 Real growth in balance sheet total and lending 37Chart 2-2 Changes in foreign currency loans of non-financial corporations and foreign liabilities 38Chart 2-3 Balance sheet concentration in the banking system 38Chart 2-4 Financial position of non-financial corporations as a percentage of GDP 39Chart 2-5 Developments in the Budapest office market 40Chart 2-6 Non-financial corporations’ bank and inter-company loans as a percentage of GDP 41Chart 2-7 Components of changes in lending to non-financial corporations 42Chart 2-8 Distribution of non-financial corporate lending by company size 42Chart 2-9 Lending to finance office building and shopping mall construction as a share of total corporate

lending and its concentration 43Chart 2-10 Concentration of non-financial corporations’ loans and deposits 44Chart 2-11 Interest rate margin on forint loans to non-financial corporations 44Chart 2-12 Contingent liabilities 45Chart 2-13 Net financing capacity/requirement of households 46Chart 2-14 Lending to households 47

INDEX OF CHARTS AND TABLES

REPORT ON FINANCIAL STABILITY 127



Chart 2-15 Households’ financial obligations as a percentage of liquid assets 47Chart 2-16 Cumulative values of newly granted subsidised housing loans 48Chart 2-17 Number of building permits issued quarterly 49Chart 2-18 Market share of the individual institution-types in housing finance 49Chart 2-19 Interest rates on banks’ non-housing loans 50Chart 2-20 Development of classified asset categories in the various portfolios 51Chart 2-21 Changes in the ratio of non-performing corporate loans 52Chart 2-22 Growth rate of loans and changes in the ratios of special watch and non-performing loans 52Chart 2-23 Foreign exchange derivatives of the banking system 54Chart 2-24 Foreign exchange assets and liabilities as a proportion of the balance sheet total 55Chart 2-25 Banks’ total foreign currency position 56Chart 2-26 Three-month BUBOR and banks’ interest rates 57Chart 2-27 90-day cumulative forint repricing gaps of the banking system 57Chart 2-28 Loan-to-deposit ratio of the banking sector 59Chart 2-29 Liquid asset ratio 59Chart 2-30 Money market exposure 60Chart 2-31 Long-term assets and liabilities of the banking sector as a proportion of the balance sheet total 60Chart 2-32 Capital adequacy ratio (CAR), stress CAR and Tier 1 CAR 61Chart 2-33 Financial position of the ten largest banks, their average, and sector average and maximum losses

incurred as a result of non-performing assets, 30 June 2002 and 2003 62Chart 2-34 Excesses over limits pursuant to the Credit Institutions Act 63Chart 2-35 Regulatory capital and its components 63Chart 2-36 Banking sector ROE 65Chart 2-37 Components of spread 66Chart 2-38 Net interest and non-interest income as a proportion of gross operating income 67Chart 2-39 Operating costs as a percentage of total assets and the cost/income ratio 67


Chart 3-1 Market risk exposure 72Chart 3-2 Losses due to market risk shocks, as a percentage of tier1 capital 72Chart 3-3 Corporate sector profitability indicators 75Chart 3-4 Distribution of ROA 75Chart 3-5 Changes in mean and median ROA 76Chart 3-6 Profitability (ROA) in individual sub-sectors 76Chart 3-7 Profitability (ROA) of exporters and non-exporters 77Chart 3-8 Changes in real wages and average number of employees at export companies 78Chart 3-9 Ratio of export sales revenues of manufacturing sub-sectors to net sales revenues 80Chart 3-10 Leverage of non-financial corporations 80Chart 3-11 Sub-sectoral leverage indicators 81Chart 3-12 Leverage of the least profitable companies and the weight of their overall liabilities in the overall

liabilities of the entire sector 81Chart 3-13 Aggregate interest coverage 82Chart 3-14 Weighted average corporate borrowing rates with maturity over one year 82Chart 3-15 Sub-sectoral interest coverage indicators 82Chart 3-16 Weight and leverage of companies with the lowest interest coverage 83Chart 3-17 Liquidity ratio in the corporate sector 83Chart 3-18 The lower quintile in the distribution of the liquidity ratio 84Chart 3-19 Median cash ratio 84

4 ARTICLES

Chart 4-1 Profit and loss profiles of forward and risk reversal positions 90Chart 4-2 Gross turnover in the forint/FX market by segment 90Chart 4-3 Gross FX swap turnover by sector 91Chart 4-4 Non-residents’ net FX swap and spot transactions vis-à-vis Hungarian banks 91



Chart 4-5 Cumulative changes in non-residents’ net swaps at the less than 1-week maturity and over 1 week 92Chart 4-6 Gross options turnover by sector 92Chart 4-7 Stock of foreign currency options by sector 93Chart 4-8 Two-week (major policy) rate of the MNB, and yields on three-month FX swaps

and government securities 94Chart 4-9 Forint/euro volatility curve 95Chart 4-10 EUR/HUF implied volatility, the interest rate differential and the exchange rate 96Chart 4-11 EUR/HUF implied volatilities for different maturities 96Chart 4-12 Forint open positions of the domestic banking sector, domestic non-banks

and non-residents since band widening 96Chart 4-13 Effect of an FX swap on and off the balance sheet 98Chart 4-14 Spot + Swap strategy 98Chart 4-15 104Chart 4-16 104Chart 4-17 104Chart 4-18 Composition of the mortgage bond market by issuers as of 30 September 2003 105Chart 4-19 Flow chart of the provision of housing loans 105Chart 4-20 Distribution of collateral by real estate type 107Chart 4-21 Composition of the gross interest margin of housing loans in April/May 2003 112Chart 4-22 Distribution of yield premiums on publicly issued mortgage bonds by the end of 2003 H1

and the margin on loans subsidised on the liabilities side after amendment of the Decree, as a function of the yield premium 113

Chart 4-23 Distribution of foreign ownership by countries in subscribed capital at end-2002 118Chart 4-24 Share of non-performing loans in the loan portfolio (Foreign banks) 122Chart 4-25 Share of non-performing loans in the loan portfolio (Domestic banks) 122

Tables


Table 1-1 Global and regional growth rates 13Table 1-2 Annual growth rate of GDP and its components 22Table 1-3 Household consumption, savings and fixed investment 22Table 1-4 Fixed investment 23


Table 2-1 Net provisioning 51Table 2-2 Recorded losses in value in the various classified asset categories as a proportion

of the gross value of balance sheet items 52Table 2-3 Major indicators of banks’ interest rate risk exposure 57Table 2-4 Capital adequacy of the five largest banks and the sector 61Table 2-5 Components of the risk-adjusted balance sheet total 62Table 2-6 Banking sector profits 66


Table 3-1 Market risk shocks 71Table 3-2 Characteristics of the aggregate portfolio at the end of 2002 72Table 3-3 Losses caused by credit shocks 72Table 3-4 Share of banks with losses in excess of 100% 73Table 3-5 Concentration of losses 73Table 3-6 Changes in exporters and non-exporters’ profits and cost categories 78Table 3-7 Contribution of the individual cost categories to changes in the profit margin 78



Table 3-8 Operational income and growth rates of sales and export revenues excluding MOL 78Table 3-9 Sales revenues and operational income in manufacturing sub-sectors 79Table 3-10 Contribution of the individual components of operating income to changes in the profit margin 79Table 3-11 Average ratio of material- and staff-related expenses to sales revenues 80Table 3-12 Year-on-year changes in the components of interest coverage in the individual sub-sectors 82

4 ARTICLES

Table 4-1 Loan portfolio of the household sector between December 2000 and September 2003 101Table 4-2 Performance and market share of major banks involved in housing finance 101Table 4-3 Product composition of the loan portfolio held by banks with a dominant market share,

30 September 103Table 4-4 Distribution of the housing loan portfolio at banks under review by rating categories 106Table 4-5 Distribution of the housing loan portfolio at banks under review on the basis of overdue payment 107Table 4-6 Age and maturity composition of housing loans 107Table 4-7 Average distribution of loans according to the LTV ratio 107Table 4-8 Average distribution of loans by geographical regions 108Table 4-9 Banking sector interest margin on subsidised housing loans before the June 2003

amendment of the Decree on subsidies 113Table 4-10 Banking sector interest margin on subsidised housing loans after the June 2003 amendment

of the decree on subsidies 113Table 4-11 Number of commercial banks in Hungary 116Table 4-12 Ownership structure of foreign and domestic banks on the basis of subscribed capital

in percentage 118Table 4-13 Market concentration indicators on banks’ total assets 120Table 4-14 Corporate and retail market concentration 121Table 4-15 Quality of bank portfolios 121Table 4-16 Profitability indicators for foreign and domestic banks 123Table 4-17 Profitability of foreign and domestic banks 123Table 4-18 Banks’ capital adequacy 124

Report on Financial Stability

December 2003

DTP-works: Iconos Advertising

H-1022 Budapest, Bogár u. 29.

Printing: Print-X

report on financial stability · report on financial stability 3 contents overview 5 1...

Documents