INDUSTRY AND FINANCE SERIES VOLUME 14 IAF-14

Interest Rate Policiesin Selected Developing Countries, 1970-82

James A. Hanson and Craig R. Neal

FiLE COPY

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Pub

lic D

iscl

osur

e A

utho

rized

Interest Rate Policies in Selected Developiing Countries,1970-82

- ii -

Industry and Finance Series

Volume 14

This series is produced by the Industry Department of the WorldBank to disseminate ongoing work done by the department and to stimulatefurther discussions on the issues. The series will include reports onindividual sectors in industry, as well as studies on global aspects ofworld industry, problems of industrial strategy and policy, and issues inindustrial finance and financial development.

Already published are the following:

*Volume 1. Structural Changes in World Industry: A Quantitative Analysisof Recent Developments

*Volume 2. Energy Efficiency and Fuel Substitution in the Cement Industrywith Emphasis on Developing Countries

*Volume 3. Industrial Restructuring: Issues and Experiences in SelectedDeveloped Economies

*Volume 4. Energy Efficiency in the Steel Industry with Emphasis onDeveloping Countries

*Volume 5. World Sulphur Survey

*Volume 6. Industrialization in Sub-Saharan Africa: Strategies andPerformance

*Volume 7. Small Enterprise Development: Economic Issues from AfricanExperience

*Volume 8. World Refinery Industry: Need for Restructuring

*Volume 9. Guidelines for Calculating Financial and Economic Rates ofReturn for DFC Projects (also in French and Spanish)

Volume 10. A Framework for Export Policy and Administration: Lessonsfrom the East Asian Experience (also in Spanish)

Volume 11. Fertilizer Producer Pricing in Developing Countries: Issuesand Approaches

Volume 12. Iron Ore: Global Prospects for the Industry, 1985-95

Volume 13. Tax Policy and Tax Reform in Semi-Industrial Countries

* Published as World Bank Technical Papers.

INDUSTRY AND FINANCE SERIES VOLUME 14

Interest Rate Policiesin Selected Developing Countries,, 1970-82

James A. Hanson and Craig R. Neal

The World BankWashington, D.C., U.S.A.

Copyright ©) 1986The International Bank for Reconstructionand Development/THE WORLD BANK

1818 H Street, N.W.Washington, D.C. 20433, U.S.A.

All rights reservedManufactured in the United States of AmericaFirst printing June 1986

This is a document published informally by the World Bank. In order that theinformation contained in it can be presented with the least possible delay, thetypescript has not been prepared in accordance with the procedures appropriate toformal printed texts, and the World Bank accepts no responsibility for errors. Thepublication is supplied at a token charge to defray part of the cost of manufacture anddistribution.

The World Bank does not accept responsibility for the views expressed herein, whichare those of the author(s) and should not be attributed to the World Bank or to itsaffiliated organizations. The findings, interpretations, and conclusions are the resultsof research supported by the Bank; they do not necessarily represent official policy ofthe Bank. The designations employed, the presentation of material, and any maps usedin this document are solely for the convenience of the reader and do not imply theexpression of any opinion whatsoever on the part of the World Bank or its affiliatesconcerning the legal status of any country, territory, city, area, or of its authorities, orconcerning the delimitation of its boundaries or national affiliation.

The most recent World Bank publications are described in the annual spring and falllists; the continuing research program is described in the annual Abstracts of CurrentStudies. The latest edition of each is available free of charge from the Publications SalesUnit, Department T, The World Bank, 1818 H Street, N.W, Washington, D.C. 20433,U.S.A., or from the European Office of the Bank, 66 avenue d'1ena, 75116 Paris, France.

James A. Hanson is senior financial economist and Craig R. Neal a researcher in theIndustry Department of the World Bank.

Library of Congress Cataloging-in-Publication Data

Hanson, James A.Interest rate policies in selected developing

countries, 1970-82.

(Industry and finance series, ISSN 0256-2235 : v. 14)Bibliography: p.1, Interest rates--Government policy--Developing

countries. I. Neal, Craig R., 1954- . II. Title.HG1623.D44H36 1986 332.8'2'091724 86-13301ISBN 0-8213-0782-7

- v -

Abstract

This paper examines the level and structure of interest rates for1970-82 in ten developing countries: Bangladesh, Kenya, Republic of Korea,Morocco, Nigeria, Pakistan, Peru, Thailand, Turkey and Uruguay. Inaddition, the paper briefly examines the countries' broader financial andeconomic policies.

In most cases nominal interest rates were controlled by thegovernments and varied substantially less than inflation during thisperiod. As a result, real interest rates rose (or fell) with declines (orincreases) in inflation; hence, highly negative real rates were a directreflection of high inflation. A simple grouping of ninE! of the ten samplecountries into two categories demonstrates this clearly: the first group(six countries) had low inflation and mildly negative real rates, whichaveraged within a few percentage points of the corresponding averages forthe industrialized countries; and the second group (three countries) hadhigh inflation and highly negative real rates, which averaged ten totwenty-five percentage points below rates in the industrialized countries.The tenth country, Uruguay, switched from the high-inflation, highly nega-tive real rate group to a high-inflation, highly positive and volatile realrate category about half way through the period, due to a major liberaliza-tion, and thus did not fall into either category.

Aside from controlling interest rates, governments in the samplecountries intervened extensively in financial markets to affect the alloca-tion of resources. Besides direct and indirect interest rate controls,other forms of intervention included large directed credit programs, publicownership of financial institutions, and sizable public sector borrowing.

The paper's principal conclusion is that the narrow concern forraising real interest rates to positive levels, so oftea voiced in develop-ment policy dialogues, should give way to broader concerns for improvementsin the financial system as a whole and for greater consistency betweenfinancial and macroeconomic policies. There are three basic reasons forthis shift in emphasis: (1) unless real rates are considerably out ofline, i.e., highly negative and well below those prevailing in inter-national capital markets, the scope for, and thus the impact of, raisingreal rates would be fairly limited; (2) even if real rates are considerablyout of line, the focus of attention should be on the problem of highinflation and the economic imbalances driving it, since otherwise it willbe difficult to achieve genuine progress on real interest rates; and (3) inand of themselves, positive real interest rates are no assurance that thefinancial system is efficiently allocating resources. Intervention infinancial markets commonly fragments markets, distorts incentives withexplicit and implicit subsidies, and burdens the system with counterproduc-tive taxes. Moroever, economic and financial rates of return often divergedue to macroeconomic and pricing policies. Accordingly, interest ratepolicies must be part of a consistent economic policy iramework, and thelevel of the real rates should not be considered in isolation.

- vii -

Table of Contents

Summary and Conclusions** ............... .......... .... ix

I. INTRODUCTION .................... ... ... ....... , 1

II. THE SAMPLE COUNTRIES, DATA SOURCES AND SOME METHODOLOGICALCONSIDERATIONS ... *...................*..*...* ............ 2

III. THE AVERAGE LEVEL OF REAL INTEREST RATES: THE INTERESTRATE REGIME .......... , 5

IV. THE DETERMINATION OF INTEREST RATES: NOMINAL ANT) REAL ...... 18

V. FINANCIAL SECTOR LIBERALIZATION: THE "NEW" VIEW ............ 32

VI. FINANCIAL SECTOR REFORMS: THE RECORD .......... .*..... . ... 36

ANNEX I SUMMARY AND COUNTRY TABLES AND FIGURES ............. 45ANNEX 2 COUNTRY STUDY: BANGLADESH ........................ 83ANNEX 3 COUNTRY STUDY: KENYA ............... . ..... too. 89ANNEX 4 COUNTRY STUDY: NIGERIA .......................... 97ANNEX 5 COUNTRY STUDY: PERU . .............................. 103ANNEX 6 COUNTRY STUDY: THAILAND ......................... 113ANNEX 7 COUNTRY STUDY: TURKEY ........................... 121ANNEX 8 COUNTRY STUDY: URUGUAY ............................ 129

SOURCES AND REFERENCES ... 137

- viii -

List of Text Tables and Figures

Table 1 Average Real Interest Rates, 1970-82 .... ........... 7

Figure 1 Nominal Term Deposit Rates versus Inflation ........ 8

Table 2 Major Policy Changes in Nominal Deposit Rates....... 14

Table 3 Average Real Interest Rates, 1970-76 and1977-82 16

Table 4 Average Real Interest Rates, 1970-80 and1981-82 ........................... 17

Figure 2 Ex-Post Real Term Deposit Rates versus Inflation ... 22

Figure 3 Nominal Term Deposit Rates versus DevaluationAdjusted U.S. Treasury Bill Rate ................ 26

Figure 4 Nominal Rates on Domestic Foreign Currency,US$ Term Deposits: Uruguay, 1978-82 ............ 30

Table 5 Nominal Interest Rate Differentials, 1982 .......... 40

- ix -

Summary and Conclusions

A. Purpose and Organization

i. This paper examines the financial sector policies in arepresentative sample of developing countries over the period 1970 to 1982,focusing on the evolution of interest rate policies. The objective of thisstudy is to assess the policies pursued by these countries in relation tothe standard policy advice to maintain positive real rates and therecommendations made by a growing number of proponents of greatermarket-orientation in the financial sector.

ii. The financial sector policy advice most often given developingcountries is to maintain positive real interest rates, i.e., nominalinterest rates in excess of inflation. This recommendation is nowbeginning to give way to concerns for increased market-orientation in thefull range of financial sector policies. As opposed to simply a mechanicalinsistence on positive real interest rates, the market oriented perspectivestresses, among things, the need to reduce the size of subsidies passedthrough the financial sector and to increase the reliance on interest ratesfor the mobilization and allocation of resources, paying attention not onlyto the real levels of rates, but to the need for differentials whichreflect differences in risk, maturity and cost.

iii. Reflecting this shifting policy perspective, the paper isorganized into two parts: first, an examination of the development of realinterest rates in a sample of ten developing countries (Bangladesh, Kenya,Republic of Korea, Morocco, Nigeria, Pakistan, Peru, Thailand, Turkey andUruguay); and second, an investigation of the patterns of financial marketintervention and liberalization in seven of the ten ccuntries (Bangladesh,Kenya, Nigeria, Peru, Thailand, Turkey and Uruguay). The first half of thepaper focusses on the questions: "Have interest rates been positive inreal terms?" and, "Was there movement towards higher real interest ratesover time?" The second half of the paper examines a broader set offinancial sector policies, focussing on the questions: "What are the formsand extent of government intervention?" and "Is there evidence of a declinein the scope and degree of intervention, suggesting growing acceptance ofthe newer, market-oriented policy perspective?"

B. Principal Conclusions

iv. In terms of interest rate policies, the principal conclusions ofthis study are:

x

(1) The observed nominal interest rates were predominantly set bygovernments, although there was some latitude for financialinstitutions to adjust effective interest rates through suchpractices as compensating balances on loans and more frequentcompounding on deposits.

(2) Administered interest rates were very "sticky," There was verylittle change in nominal rates over time, particularly incomparison to fluctuations in inflation rates.

(3) The "stickiness" of the nominal rates meant that year-to-yearvariations in the real rates were primarily determined byvariations in the inflation rates.

(4) Real deposit rates were predominantly negative throughout the1970-82 period. However, it should be noted that slightlynegative real deposit rates also prevailed in most of theindustrialized market economies, including the U.S., during thisperiod.

(5) Real lending rates on general credits were between 2 to 7percentage points higher than deposit rates in the samplecountries. These rates were also generally negative during the1970-82 period, although in six of the ten sample countriesaverage real rates ranged between -1.3% and +2.9%.

(6) The sample countries' interest rate policies fell into threedistinct groups: six countries had low average inflation ratesand real rates that were not very negative on average, threecountries had relatively high inflation rates and quite lowaverage real rates, and one country (Uruguay) had high inflationbut, in the middle of the sample period, moved to marketdetermined interest rates, which swung from low to high reallevels.

(7) Countries with relatively high average inflation rates alsogenerally had relatively low average real rates, indicating thatthe governments were unwilling or unable to set nominal rates inline with high rates of inflation. The exception is Uruguayafter its reforms.

(8) Uruguay was not the only country to undertake reforms. In mostof the sample countries, nominal rates in the second half of theperiod tended to be somewhat higher and there was generally someincrease in the average level of real interest rates. Thissuggests that the authorities took some measures to raise realrates--in accordance with "traditional" Bank recommendations.

(9) However, most of the improvement in the average real rates in the1976-82 period was attributable to a decline in inflation ratesin the sample countries, especially during 1981-82, and, only toa lesser extent, the upward shift in the nominal rates.

- xi -

(10) Real rates in all the countries (including the U.S.) generallyrose together in 1981 or 1982 and declined together during 1973or 1974 and again in 1979 or 1980. These contemporaneousmovements suggest similar factors were at work in the samplecountries. In fact, the common element is the contemporaneousmovements in inflation.

(11) The hypothesis that interest rates are determined in the shortrun by flows of international capital, and thus are tied toindustrialized country interest rates adjusted for devaluation,is not supported by the record of the ten dieveloping countriesstudied.

v. The principal conclusions from the investigation of financialsector intervention and liberalization, in the seven countries are:

(12) All the countries maintained substantial directed creditprograms. Three types of instruments typically were used:(a) regulations on the portfolio composition of intermediaries;e.g., requirements to devote a certain portion of lending tospecific activities; (b) Central Bank rediscounting- of credits topriority sectors, usually at subsidized rates; and (c) control offinancial intermediaries through direct ownership.

(13) Public sector borrowing to finance the fiscal deficit and publicenterprises was an important factor in the sample countries.Regulations often existed which forced the financial system tohold low-interest government debt, thereby imposing substantialcost on the rest of the system.

(14) The directed credit programs and the public sector borrowingscombined to reduce the supply of credit available tonon-preferred borrowers--the well known "crowding out"phenomenon. Increased competition for the remaining credit droveup interest rates when these were free, in some cases producingvery high rates on non-preferential credits. When interest rateswere constrained, crowding out greatly complicated the overallallocation of credit.

(15) All of the sampled countries also maintained some form ofadministered interest rate regime, with this exception of Uruguayafter 1978. The principal differences between the countries werein the relative size of the directed credit programs and in theinterest rate differentials between preferiential andnon-preferential credits.

(16) All countries exercised control over the allocation of credit.Among the three low real interest countries the fraction ofcredit affected by directed allocation in 1982 rarnged from almost100% in Nigeria to nearly 55% in Peru. In Uruguay, which

- xii -

underwent sweeping reforms during the 1974-1979 period, 37% ofcredit was still extended by the two government banks in 1982.Even among the higher real rate-low inflation countries there wassubstantial government control of credit allocation, from as muchas 70% to figures in the neighborhood of 33%.

(17) In general terms, at the end of 1982 the differentials betweenrates on deposits, general lending and preferential lending rateswere fairly small in the low inflation-high real rate countries,with the exception of one or two programs. In the three highinflation-low real rate countries, the differentials were quitelarge. This difference can probably be explained by the simplefact that low inflation and low nominal interest rateseffectively place a ceiling on the absolute size of thesubsidies. High inflation provides more room for subsidiesthrough large differentials, and because subsidizes can be largerthere is a greater incentive to demand them. Also preferentialrates were often lower than deposits rates in the highinflation-low real rate countries, an interest rate structurewhich encourages diversion of directed credit.

(18) In all but one of the seven sample countries there were definitemoves towards greater market-oriented financial sector policies,from major reform movements to more limited realignments in thestructure of nominal interest rates and directed creditprograms. This movement could well be interpreted as a growingacceptance of value of financial sector liberalization, inaccordance with the new market-oriented perspective. However,the scope of market determination still remains quite constrainedby many factors, including a wide array of governmentregulations.

C. Sample Country Experience

vi. A closer look at the evidence on the development of the realinterest rates in the ten sample countries shows: Over the period 1970-82,the countries' policies toward real interest rates fell into three distinctgroups. In the first six countries--Pakistan, Morocco, Thailand, Korea,Bangladesh and Kenya (in descending order of their average real depositrates)-the average real interest rates on deposits (defined as the ex-postreal rate) were somewhat negative during the period 1970-82, ranging fromroughly zero to negative 7 percent. Average non-preferential loan rateswere just negative or slightly positive in real terms. Preferential ratestypically were above or just below deposit rates. While not exactlyconforming to the standard policy advice, these deposit and loan rates weresimilar to the comparable average U.S. real rates during the same period,which were negative one percent and positive two percent, respectively.

vii. A second set of three countries--Nigeria, Turkey andPeru--maintained real deposit rates which were, on average, quite negativeover the period, ranging from -16.5% to -18.6%. Their non-preferential

- xiii -

lending rates also tended to be highly negative, althcough various financialdevices such as discounting and compensating balances kept these ratesabove deposit rates. Preferential loan rates were often substantiallybelow deposit rates. The last country, Uruguay, undertook a significantfinancial reform starting in 1974, switching from highly negative realdeposit and lending rates to a regime in which interest rates weredetermined in an open capital market by 1978.

viii. Uruguay was not the only country in the sample to reform itsinterest rate policy. Turkey, after 1980, increased the permissiblenominal level for many interest rates and permitted market determination ofsome others. In combination with a sharp decline in Lnflation, thisresulted in unprecedented high real rates, rates that reached +34% on shortterm credits at the end of 1982. Bangladesh, Pakistan, and even Nigeriaachieved higher real rates during the second half of the period, butprimarily though a reduction in the average rate of inflation. Peruincreased the permissible ceilings on nominal interest rates beginning in1978, however this increase was insufficient to compensate for the sharpescalation in inflation, producing lower average real rates in the secondhalf of the period than the first.

ix. The Uruguayan reforms included an opening up of the domesticfinancial markets to the international capital markets. However, thespread between nominal rates on domestic currency and foreign currencydeposits remained relatively constant, despite the (preannounced) declinein the rate of devaluation. These results are at variance with thestandard theories that suggest convergence between world and domesticinterest rates (adjusted for devaluation) once the capital market isopened. Either the public's expectations were for a constant rate ofdevaluation, rather than the actual rate (which was preannounced duringmost of the period), or a growing risk premium was required on pesodeposits to cover the growing threat of a maxi-devaluation, a devaluationwhich finally occurred in November 1982. Thailand also provided a test forthe interest parity model, since it has a long history of an open financialmarket and a fully convertible currency. In the Thai case the model worksreasonably well as the nominal term deposit rate remains fairly close tothe devaluation adjusted U.S. Treasury Bill Rate, and some of the smalldifference in the two rates probably could be attributed to the differencesin the two instruments. In all the other countries, however, neitherexpected devaluation nor expected inflation seemed to have much to do withthe setting of interest rates. Thus real rates were largely determined byinflation.

x. A country-by-country look at the wider question of financialsector policy reforms, in the seven country sub-samp:Le shows:

(a) Bangladesh raised most ceilings on interesi: rates in late 1980.The government also loosened its institution-specific ceiling oncredit expansion, relaxing some of the restrictions which hadprevented the transfer of funds to banks constrained by theircredit ceiling with a surplus of investment opportunities, frombanks below their credit ceiling but without good investmentopportunities.

- xiv -

(b) In Kenya, the minimum deposit rates and the maximum lending rateswere increased between 1980 and 1982. Beyond thesereadjustments, Kenya did not undertake any sweeping reforms. Inlarge part, this lack of reforms was attributable to thecountry's history of relative price stability and relativelymarket-oriented economic policies, which kept pressures frombuilding to a level that would require major reforms.

(c) Due to Thailand's long history as an open and relatively pricestable economy, its policy makers were not forced to undertakemassive reforms. However, the government's overriding policycommitment to a freely convertible currency did require theauthorities to make some macroeconomic policy changes to managedevelopments which threatened convertibility. In particular,after roughly five years of increasing protectionism and in theface of sharply rising world interest rates, the authoritiesinstituted a set of reforms in 1980, which included an upwardadjustment in the interest rate structure.

(d) Since 1978, Nigeria has been raising its interest rate structureslowly, though rates have remained far below historicalinflation. Credit allocation remained highly controlled by thegovernment, through a complex and sometimes inconsistent set ofregulations. Up until 1981, when the countries oil revenuescollapsed, the Nigerian authorities apparently felt little needto promote the domestic financial sector, as most of thecountry's investments were undertaken directly by the governmentwith the oil earnings. The softening of the long term outlookfor oil is likely to increase pressures for greater domesticresource mobilization, and improved financial intermediation ingeneral.

(e) In Turkey the financial sector was highly controlled by theauthorities. Among the important government policy instrumentswere large directed credit programs and a highly complex set ofinterest rate controls. In 1980 some these interest rates wereliberalized, including term deposit rates and the ceiling rateson general credits. In addition, the size of the directed creditprograms were scaled back somewhat. By 1982, the newlyliberalized interest rates reached extraordinarily highlevels--up to +22.1% on term deposits and +34.3% on generalcredits--causing some serious difficulties for the heavilyindebted firms in the corporate sector. These high real ratesare due, in part, to the rapid decline in inflation and probablyreflect subsidies required to cover the cost of large share oftotal credit which is still extended at unliberalized rates. Inaddition, fears of a devaluation may have pushed up rates inlocal currency. However, the authorities have made it a matterof announced policy to continue reducing the scope and complexityof preferential credit schemes. This should help rationalize thecredit allocation process, reduce the amount of thecross-subsidization, and lower intermediation costs.

- xv -

(f) Peru initiated a shift in interest rate policy in 1978, whichcontinued through 1982. Despite the massive upscaling of theinterest rate structure, the adjustments in the ceiling rateslagged well behind the increases in inflation, so real interestrates on deposits remained quite negative, with the exception of1981. The adjustments in the interest rates were also part a setof major reforms expressly aimed at giving ma.rket forces a largerrole in the economy, which included opening cf both the goods andcapital markets, the legalization of dollar t;ransactions such asdeposits and loans, a reduction in the vast rLumber of controlledinterest rates through a simplification of the preferentiallending categories. Reserve requirements on deposits in localcurrency were also reduced and interest was paid on the remainingpart. This led to a reduction in the inflation tax base, therebyincreasing the inflation resulting from a given increase incentral bank credit. In view this potentially explosivesituation and the lack of progress towards hi'gher real interestrates, the Peruvian reforms must be judged to have been of ratherlimited success.

(g) Uruguay, much like Peru and Turkey, had a history of extremelyhigh inflation rates, though its experience began much earlier.The combination of a stagnating economy, balance of paymentsproblems and high inflation prompted the authorities to begin aliberalization process in 1974 that was almost without precedentin its dimensions. By the end of 1978 virtually all interestrate controls had been dismantled, the fiscaL deficit was closed,sectoral credit guidelines and reserve requirements on domesticcurrency deposits were eliminated, and foreign currencytransactions were legalized, including bank deposits and dollarloans. With the important exceptions of the operations of thepublicly owned Banco de la Republica and the Banco F[ipotecariodel Uruguay and the authorities' control over the exchange rate,control over the financial sector resided primarily in the handsof the private sector in 1982.

xi. It is very important to note that, based on the evidence,financial sector liberalization is neither a simple matter, nor a painlessone. Specifically, the success of the financial sector liberalizationprocess is highly dependent on the mix of domestic fiscal, monetary,exchange, commercial and trade policies, particularly when the domesticfinancial market is open to international capital flows. For example,inconsistencies between fiscal and monetary policies, on the one hand, andexchange rate policies, on the other, can cause major disruptiLons offinancial markets through inflows and outflows of capital. In countrieswhere significant protection exists, financial liberalization may have theundesirable effect of providing additional credit to economicallyinefficient firms or, as in Uruguay, a boom-bust cycle based on credit.

xii. On this point, the record shows that both Peru and lJruguay mademajor changes in their financial sector policies in the latter part of theseventies which, among things, opened the financial sector to international

- xvi -

markets. However, when the exchange rate policy became inconsistent withdomestic monetary, fiscal, and wage policies, a common pattern emerged ofeither high real interest rates in the local currencies, and/or"dollarization" and capital flight. This experience can be contrasted tothat of Thailand, which also began to experience large capital out flowsthrough its very open financial system at the end of the seventies.However, Thailand was able to escape the full impact of the crisis whichstruck Peru, Uruguay and much of the rest of Latin America, because itsinitial macro-policy framework was much more consistent, and because itadjusted its monetary, fiscal and trade policies more rapidly to thechanging conditions. While a full exploration of these two experienceslies beyond the scope of this paper, there is general agreement that one ofthe crucial lessons of the financial liberalization process in LatinAmerican was the necessity of a consistent macroeconomic policy packagewhen the capital account is "open." Thus, both policy makers contemplatingfinancial sector liberalizations and proponents of such reforms must takeserious note of this lesson.

D. The Call for Greater Market-Oriented Financial Sector Policies

xiii. The growing consensus among development policy analysts forgreater market-orientation of financial sector policies has received itsimpetus from a number of sources including: (1) an increasing appreciationof the social costs imposed by large scale government intervention infinancial markets, (2) a greater sense of the importance of thewell-functioning financial markets for development, and (3) a growing doubtas to the effectiveness of many forms of intervention in achieving theprofessed policy goals. In particular, government policies such asinterest rate controls and large directed credit programs severely distortthe pattern of incentives in the economy and thus affect the allocation ofresources, often with questionable efficiency and distributionalimplications. The very administration of such programs also absorbs agreat deal of scarce resources, particularly skilled labor. Furthermore,the fungibility of financial resources and the politicized nature of the"non-price" rationing mechanisms result in substantial leakages ofresources away from the targetted sectors or groups, making it difficultfor credit programs to achieve their stated objectives.

xiv. Briefly, the market-oriented perspective suggests that thereshould be three broad objectives of financial sector policies: (1) themobilization of adequate amounts of resources, both domestically andinternationally; (2) the allocation of credit to its socially mostproductive uses; and (3) the promotion of stability in the financialsystem, which will encourage development. In terms of interest rates,these objectives can best be achieved by allowing rates to vary accordingto the state of the business cycle, pressures on the foreign exchangemarket, and the risk, maturity and administrative cost of the individualtransaction. Also subsidies and complex systems of directed credit shouldbe avoided. This "new perspective" does, however, recognize the need forbetter financial market supervision to control fraud and mismanagement andfor regulations which promote greater competitiveness in the financial

- xvii -

system. The new perspective also emphasizes the importsLnce of narrowingthe differences between economic and financial rates of return, by reducingprice distortions through a more consistent fiscal, monetary, exchange,commercial and trade policy mix.

E. Implications for Development Financial Policy

xv. Financial sector policy should shift its emphasis from amechanical insistence on positive real rates to promotion of broad basedimprovements in the functioning of the financial system and greaterconsistency in the macroeconomic policy framework. While real interestrates are important, they are only an imperfect indicator of the generalhealth of the financial system. The concept of the "real" interest ratesimply does not permit any but the broadest judgment about the functioningof the financial system. Slightly negative real interent rates in a givenyear, or even over a period of time, can be explained by the composition ofthe country's price index, by taxation, by differences between nominal andeffective interest rates, or by expectations regarding the future.Moreover, even tying local interest rates to world rates; does not guaranteepositive real rates. The evidence presented in this paper suggests that,even in developing countries with open capital markets, real interest ratesmay vary sharply from year-to-year, and may even be negative in real termsfor a number of years.

xvi. Despite the foregoing qualifications, it is possible to identifysituations when interest rates are excessively negative in real terms.However, the evidence of the paper indicates that negatLve real rates aregenerally not a problem, in and of themselves, but a symnptom of much largerproblems: that the whole macroeconomic framework--monetary, fiscal,exchange rate and tariff policies, as well as the interest rate-are out ofline. Moreover, abnormally high real rates, as well as abnormally low realrates, may be indicators of these problems. For example, as shown in thepaper, very low real rates typically reflect high inflation, which in turnreflects the government's use of inflationary finance to cover large publicsector deficits. The paper suggests that liberalizing financial markets,without closing the deficit, could actually produce higher inflation andhigher nominal interest rates if the government continues to resort toinflationary finance. Alternatively, if the public sector tries to financeits deficit with debts at market rates, or if the government tries toshield some sectors from the effects of higher interest rates by providinglow-interest, directed credit, then interest rates in the rest of themarket may become abnormally high, as in the Turkish case. Judging fromthe experience of Uruguay, and perhaps Turkey, abnormally high realinterest rates may also develop in open capital markets due to expectationsof a maxi-devaluation, which in turn reflect a perceived inconsistencybetween the exchange rate policy and the other macroeconomic policies. Theabove cases suggest that liberalization of real rates, without payingsufficient attention to the underlying imbalances in the economy, couldwell destabilize the corporate and banking sectors, by creating significantcash flow and liquidity problems.

- xviii -

xvii. The scope and scale of low interest, directed credit programsshould be reduced, in order to improve the functioning of the financialsystem. The sources surveyed for this paper indicate a number of problemsassociated with large amounts directed credit. In particular, the fungi-bility of financial resources and the politicized nature of the creditallocation process often lead to substantial leakages of resources from thetarget groups or sectors. Attempts to prevent such leakages are alsoproblematic, since they necessitate imposing a degree of fragmentation onfinancial markets that thwarts the markets' role as efficient, decentra-lized allocators of resources. The evidence also suggests that thedistortions produced by the directed credit programs often have question-able efficiency and distributional consequences, particularly when accountis taken for the ultimate bearers of the implicit and explicit taxesinherent in such programs. Finally, the administration of these programsand the efforts taken to circumvent their regulations tend to absorb largeamounts of scarce human resources without generating any social benefit.

xviii. Two final points are worth noting: First, the urgency of suchpolicy reforms is directly related to both the size of the programs and thesize of the interest differentials. Small interest differentials and smallprograms simply cannot create large distortions. Second, even if financialdistortions are small, the financial system may be efficiently allocatingresources to economically inefficient projects if the price and profita-bility signals that it receives do not reflect economic efficiency. Thisproblem once again highlights the need for a link between macroeconomicpolicy, trade and pricing policy, and financial sector policy in order toachieve development goals.

I. INTRODUCTION

1.01 Broadly speaking, the recommendations for finarncial sectorpolicies in developing countries have evolved along the following lines:The orthodoxy of the fifties and sixties was that interest rates should bekept low to stimulate investment. Rising world inflatiorL in the latesixties and the seventies focussed attention on the detrimental effects ofnegative real interest rates on the allocation of resources, the distribu-tion of income, and the mobilization of savings. As a result, policyrecommendations often emphasized that lending and deposit: rates should bepositive in real terms and intermediation spreads should provide an ade-quate return to financial institutions. In recent years, a growing consen-sus has developed regarding the need for greater reliance on market forcesin the determination of interest rates and in the financial sector ingeneral.

1.02 Several factors have contributed to the growing recognition ofthe need for more market-oriented financial sector policles. First, it hasbecome exceedingly difficult either to prescribe a fixed structure ofinterest rates, or to expect that governments could adjust interest rateswith sufficient speed to maintain stability in the financial markets andavoid massive flows of international capital, given the sharp increases inthe level and variability of local inflation and of world interest rates.Second, there has been an increasing perception that welL-functioningfinancial markets are an important factor in development., As evidence hasaccumulated on the functioning of financial markets in developing coun-tries, it has become increasingly clear that the distortLons introduced bymany government financial sector policies, particularly Interest ratecontrols, were substantial. Moreover, the effectiveness of many of thesepolicies in attaining the intended policy objectives was also cast intodoubt.

1.03 Another growing perception in the development 'Literature is thatfinancial markets and interest rates cannot be considered in isolation;they are greatly influenced by what goes on in the rest of the economy. Inparticular, the importance of factors such as openness and distortions inthe real side of the economy have been rather rudely brought to light inthe last couple of years. For example, in some developing countriesunprecedented high real interest rates have appeared, producing a crisis inthe corporate sector by sharply increasing debt service burdens. To someextent these high real rates simply reflect the higher real rates in theworld financial markets and capital market openness. However, the highreal interest rate problem has been most acute in historically high infla-tion countries which have undertaken interest rate and exchange ratereforms. In some cases these high real rates appear to be related toinconsistencies between fiscal and monetary policy, exchange rate and tradepolicy, and expectations; in other cases they seem to be related todistress borrowing and/or the poor performance of a substantial portion of

- 2 -

outstanding credits, which in turn are largely a reflection of poorperformance in the real economy. No clear consensus has formed on eitherthe explanations for these high rates, or the optimum policy response.Nonetheless, the whole problem underscores the fact that formulating an"appropriate" interest rate policy is an exceedingly difficult task andinvolves far more than the real rate of interest.

1.04 This paper examines interest rate policies in developingcountries by asking the following two questions: "Have interest rates beenpositive in real terms?" and, in light of the growing consensus over theneed for more market-oriented policies, "Have market forces been importantfactors in the financial sectors of the developing countries?"

1.05 To answer these questions the paper proceeds as follows: Section2 discusses the selection of the countries used in the investigation, thetime frame, the interest rate series collected and the data sources.Section 2 also discusses two important methodological issues: (1) whetherto focus on lending or deposit rates, and (2) the calculation of the "real"interest rate, i.e., either "ex-ante" using lagged inflation or "ex-post"using the actual inflation rate. Section 3 addresses the first question,using evidence on the real and nominal interest rates in the samplecountries. Section 4 examines whether interest rates were set according totwo well-known models: the Fisherian Model and the Uncovered Interest RateParity Model. Section 5 begins the discussion of the paper's secondquestion--the degree of market orientation of financial sector policies--with a brief statement of the arguments for greater reliance on marketforces. Section 6 summarizes the evidence relating to the changing role ofthe market in the financial sectors of developing countries.

II. THE SAMPLE COUNTRIES, DATA SOURCES ANDSOME METHODOLOGICAL CONSIDERATIONS

2.01 This paper examines the financial sector policies pursued by asample of ten developing countries over the period 1970 to 1982. Thesecountries are: Bangladesh, Kenya, Republic of Korea, Morocco, Nigeria,Pakistan, Peru, Thailand, Turkey and Uruguay. To investigate the questionof positive real interest rates, evidence is presented on nominal and realinterest rates on term deposits, general short term credits, and apreferential lending category in each of the ten countries. The secondissue investigated in this paper--the extent of financial marketliberalization--is treated through a broad, qualitative examination offinancial sector policies in a subset of seven of the ten countries. Theseanalyses appear as Annexes 2 through 8 and are summarized in Section 7,which focuses on government intervention in financial markets, financialsector reforms, the structure of lending and deposit rates, and therelation between interest rate and exchange rate policies.

2.02 The ten countries were selected on the basis of: (1) diversity,both geographical and developmental; (2) the availability of interest rateseries covering the 1970-82 period; and (3) the availability of data on theholdings of financial assets, specifically demand, savings and termaccounts, plus other forms of widely held near-monies. The sample contains

-3-

countries from East Asia, South Asia, Sub-Saharan Africa, North Africa,Latin America and an OECD member. It is fairly representative of develop-ing countries in general, as it contains countries at very low income, lowincome and middle income levels. There are countries with inward lookingeconomic policies and countries pursuing export-oriented growth strategies,and there is a major oil-exporter.

2.03 For inclusion in the seven-country sub-sample, an additionalcriteria was used: the availability of a reasonably up-to-date World Bankstudy of the financial sector, permitting a characterization of the broaderpolicy setting. These seven countries are: Bangladesh, Kenya, Nigeria,Peru, Thailand, Turkey and Uruguay.

2.04 In addition to the World Bank financial sector studies, thesources for the data used in this study include the bulletins of thecentral banks for the interest rate and asset data, and the IMF'sInternational Financial Statistics for the consumer prices, exchange ratesand national income data. The interest rates quoted in the central bankpublications typically are the rates established by governments, e.g.,minimum or maximum deposit rates and maximum lending rates. The publishedfigures usually do not reflect various common financial practices such ascompounding, compensating balances required of borrowers, or free servicesrendered to depositors; nor do they reflect the impact of taxes on interestrates. Whenever the information was available, effective interest rateswere used or computed: for example, rates were adjusted for compounding ordiscounting. Unfortunately, for most countries only unadjusted rates wereavailable. Thus the figures presented here should be used primarily toidentify the range of actual interest rates and should not be taken asoverly precise.

2.05 Bearing this in mind, the first important methodological questionis whether to cast the discussion in terms of lending rates or depositrates. Some technical considerations suggest focusing on deposit rates.First, it is relatively easy to identify a "standard" deposit rate that iscomparable across countries, but there is no similar "sitandard" lendingrate; in a free market lending rates vary substantially by risk and matur-ity. Moreover, in the controlled markets of developing countries varyingdegrees of credit rationing occur, making direct comparLsons of lendingrates difficult. Second, the available data predominantly refer toofficial rates and in practice, compensating balances, commissions, graceperiods and taxes allow effective lending rates to diverge further from theofficial lending rates than compounding and taxes allow effective depositrates to differ from official deposit rates. Again thia makes it prefer-able to compare deposit rates. Third and finally, the standard depositrate provides a reasonable amount of information about the loan rates.This is because the standard deposit rate and a comparable loan rate shoulddiffer by a fairly constant margin within a country, reflecting country

-4-

specific intermediation costs and reserve requirements.l/ Acrosscountries the differences in these margins are also likely to be small,relative to differences in the levels of interest rates. For these reasonsmost of the discussion of real rates will be focus on deposit rates,although lending rates also will be mentioned.

2.06 The second *ethodological issue is the appropriate calculation ofthe "real" rate. While a wide range of policy discussions express concernthat interest rates be positive in real terms, i.e., in excess of the rateof change in prices, little is said about what price index should be usedin comparison. In the conventional model of the demands for financialassets or credit, agents are assumed to make their decisions based on acomparison of interest rates with their expectations about future prices(including possibly the price of foreign exchange). However, such expecta-tions are not observable, nor is there general agreement regarding the"relevant" array of prices or interest rates. The typical proxy for the"real" rate involves a single interest rate adjusted either by some averageof past changes in the consumer price index (and, for the exchange rate,past changes in the official exchange rate) or by some projection ofdeclining inflation. However, in periods of rising inflation (or majordevaluations) such as the 1970-80, such procedures systematically under-state investors' expected returns on credit and systematically overstatedepositors' expected returns on financial assets, assuming the publicunderstands that inflation is rising. In effect, using lagged inflation tocalculate a real rate implicitly assumes that economic agents are system-atically fooled by rising inflation.

2.07 Such an assumption is highly unsatisfactory, particularly sincemany of the important conclusions to be drawn from a study of interestrates relate to the allocation of financial resources and the mobilizationof financial savings, both of which depend heavily on the public's expecta-tions. In an attempt to provide a better proxy for the expected realinterest rates, this paper concentrates its analysis on the "ex-post"realized rate, i.e., the nominal interest rate deflated by the actual pricechange over the relevant period. For example, the real interest rate for1982 was calculated by deflating the nominal interest rate in effect inDecember 1982 by the annualized rate of change in consumer prices (or theexchange rate, when applicable) from December 1982 to June 1983. While notimputing perfect foresight, the authors feel that this method provides abetter proxy for the expectations of depositors and borrowers than theusual method of averaging past inflation, particularly in periods of rising

1/ For a profit maximizing bank, the interest rate on a standard loan, R,should cover the marginal interest cost of deposits used to make theloan, D, multiplied by an adjustment for the reserve requirement, RR,plus intermediation costs expressed as a percentage of the loan, C,i.e., R - C + [ D/(1 - RR) ].

- 5 -

inflation.2 / A six month time horizon was utilized in these calculationsto reflect the short maturities of portfolios which characterize the finan-cial markets of the developing countries. However, in view of all thequestions regarding both the interest rates and the appropriate choice ofproxy for expectations it is worth repeating the warnirLg that the figuresin this study should not be interpreted as a precise estimate of the realrate but only as a general indicator of the range of real interestrates.3/

III. THE AVERAGE LEVEL OF REAL INTEREST RATES:THE INTEREST RATE REGIME

3.01 The real interest rates policy regimes of the ten countries fallinto three distinct groups, as measured by the average ex-post realinterest rates for the period 1970-82. This division shows up in Table 1,which presents averages 4/ of year-end real interest rates in localcurrency on term deposits, general short term credits and preferential

2/ The ex-post real rate has become a standard approach for studying theextent to which inflationary expectations are incorporated into theinterest rate. See for example E. Fama, "Short Term Interest Rates asPredictors of Inflation," American Economic Review, June 1975, pp.269-82 and R. Saracoglu, "Expectations of Inflation and Interest RateDetermination," IMF Staff Papers, March 1984, pp. 141-78 and the workscited there.

3/ It should also be noted that the analysis is affected very little bythe choice of lagged or future inflation as the deflator for thecalculation of the real rates. While the choice cdoes affect the realrates on a year to year basis, it has little impact on the broadpicture over a longer period of time. First, nominal interest rateswere fairly constant over time in the sample, thuE real rates wereprimarily determined by the inflation rates. Second, inflation tendedto move contemporaneously across the countries, thus the choice oflagged or future inflation primarily shifted the point in time at whichreal interest rates rose or fell, but made little impact on therelative levels of average real rates across count:ries. On this point,the interested reader is encouraged to compare Figure 1 in Section 3,which plot the nominal deposit rates against the Lnflation rates sixmonths forward from December of the current year, to Figure IA inAnnex 1, which plot the nominal deposit rates agaLnst the rates ofinflation over the twelve months preceding Decembar of the currentyear. In addition, see Footnote 16 which also relates to this issue.

4/ The average is defined as the average compound real interest rate overthe period. Thus it is the average growth of the real purchasing powerof a one unit deposit over the period, or the average real cost of oneunit of credit over the period. For the exact formula see Table 1,Note 1.

- 6 -

credits for each country for the period 1970 to 1982.5/ The groupings alsoshows up in Figure 1, which charts the end-year nominal term deposit ratesand inflation rates over the same period.6/ The ex-post real term depositrates can be read off the charts as roughly the difference between the twolines. The real rate is negative when the inflation rate lies above thedeposit rate and positive when the opposite is true.

3.02 The first group of countries--Pakistan, Morocco, Korea, Thailand,Bangladesh and Kenya, in descending order of their average real depositrates--maintained real rates which were only slightly negative whenaveraged over the whole period. These rates ranged from roughly zero inPakistan to -6.8%, in Kenya. Real lending rates for short term creditswere 2 to 7 percentage points above these rates, depending on the country,and ranged from about +3% to -1%. Preferential lending rates generallyfell between the average deposit and the average lending rate.

3.03 Even though real deposit rates were, on average, not positive inthese six countries, they did approximate the average real deposit ratesprevailing in developed countries. In particular, the average real depositrates for these countries ranged from rough equivalence to six percentagepoints below the comparable U.S. real deposit rate of -0.7%, which isdisplayed in the last line of Table 1.7/ Moreover, one would not expectthe real rates to be exactly the same, even if financial markets wereintegrated or if both individual preferences and investment opportunitieswere the same across countries, because of cross-country differences in the

5/ As shown in the Annexes, Uruguay and Peru recently allowed deposits andcredits denominated in foreign currency. These deposits and creditsrepresented a substantial portion of total financial assets andliabilities, and rates on them approximated international levels.

6/ Annex 1 presents the raw data for Tables 1 through 4 and Figures 1 and2. As shown in Figure 1 in Annex 1, the countries fall into the samethree groups if past inflation is used to calculate real interestrates.

7/ Using a more complicated methodology, Saracoglu estimated that realinterest rates in the Federal Republic of Germany, France, Japan andthe U.K. were also negative during most of the 1970's. See Saracogluop. cit.

- 7 -

Table 1

Average Real Interest Rates a/

(percent per annum)

Deposit Lending RepresentativeCountry Rate Rate Preferential Rate e/

Pakistan 1970-82 0.1 2.2 1.5Morocco 1974-82 -1.7 2.7 -1.8Korea 1970-82 -4.1 -1.1 -8.3Thailand 1970-82 -4.1 2.9 -4.6Bangladesh 1971-82 -5.4 -1.3 -3.4Kenya 1970-82 -6.8 -1.0 b/ 4.4

Uruguay 1970-82 -11.1 14.7 c/ #N/A

Turkey 1974-82 -16.2 -13.6 -20.5Nigeria 1970-82 -16.5 -11.8 -7.8 d/Peru 1970-82 -18.6 -9.1 -24.8

U.S. 1970-82 -0.7 1.9 #N/A

Sources: Annex 1 tables. U.S. figures are based on F'ederal ReserveBulletin nominal rates and IMF CPI data.

a! Average refers to the average compound real interest rate over thethe period, i.e.,

Antilog([Ft( ln(l+rt) - ln(l+pt)) 1/n ) - 1

where rt - nominal rate of interest at end of year t,Pt - annualized inflation rate from end of yeart to the following June,n - number of years.

b/ 1977-82; the corresponding average real deposit rate was -4.7X andthe average preferential lending rate was -3.8%.

c/ 1976-82; the corresponding average real deposit rate was 0.9% and theprime lending rate was 4.6%.

d/ 1978-82; the corresponding average real deposit rate was -10.8% andthe average real general lending rate was -6.0%

e/ Specifically, the preferential rates were as follows: loans againstjute, jute goods and tea, Bangladesh; loans from the AgriculturalFinance Corp., Kenya; preferred sector maximum, NLgeria; rediscountsfrom the Banco Agrario, Peru; export credits, ThaLland; andagricultural credits, Turkey.

- 8 -

Figure 1

Nominal Term Deposit Rate versus Inflation

- NOMINAL INTEREST RATE ON TERM DEPOSITS (6 MONTH)

--. INFLATION RATE (DECEMBER TO JUNE, ANNUALIZED)

80-

4' :

X 20-

10- "',." " ', ,

-10- BANGLADESH

48-

308

50-

KENYA

408

30-

X 20-

0- KOREA

-18- , I I I I I I I I I I ,1797 1972 1874 1978 1978 1880 1982

-9-

Figure 1 (continued)

NOMINAL INTEREST RATE ON TERM DEPOSITS (6 MONTH)

------ INFLATION RATE (DECEMBER TO JUNE, ANNUALIZED)

20-

X15- ,, t h\ ~~~~~............ "'

5- MOROCCO

410- l

20

IS ~~~~~~~~' '' ''

X * . I a

0

-20

30-

20-3 0-_XE=v,,,

Is-

-5- PAKISTAN

1970 1972 1974 1976 1978 1989 1982

- 10 -

Figure 1 (continued)

NOMINAL INTEREST RATE ON TERM DEPOSITS (6 MONTH)

----- INFLATION RATE (DECEMBER TO JUNE, ANNUALIZED)

1208

3 - ........ .. . ...go

PERU

30-

2089r l

0-.

30,

0- TURKEY

-970 - 187 1Q74 1 1Q7 1 1

150-~~97 .97 ,:6196ies1

- 11 -

Figure 1 (continued)

NOMINAL INTEREST RATE ON TERM DEIPOSITS (6 MONTH)

--.-- INFLATION RATE (DECEMBER TO JUNE, ANNUALIZED)

408-

a q URUGUAY

-20-

25-

20-

0 -

_5_ UNITED STATES

-10 I. I I I I I I-74 7 I I I I 821075 1972 1974 1970 1S176 ices 192

- 12 -

composition of the price index.8/ Also, differences in the tax treatmentof deposit interest affect any comparison between U.S. and developingcountry interest rates. For example, a pre-tax real rate of -1% isequivalent to about -3.4% after taxes, given a nominal interest rate of 8%and a tax rate of 30%.9/ Since effective tax rates were probably muchhigher in the U.S.--because interest is often legally free of taxation indeveloping countries and, in any case, is often unreported to the taxofficials--a comparison of pre-tax real returns exaggerates the relativeattractiveness of U.S. deposits.

3.04 The last line of Table 1 also displays the average real U.S.Prime Rate, which can be used to roughly estimate an intermediation spreadfor the U.S. A comparison of the differences between lending and depositrates in the first six developing countries and the corresponding spreadfor the U.S. indicates that spreads in these six developing country werereasonable--roughly one to three percentage points greater than the U.S.margin. Also, the preferential lending rates in the six countries were,by-and-large, above the deposit rates. In sum, the evidence suggests thatin these six countries interest rate policy was similar to policies indeveloped countries.

3.05 The second group of countries--Turkey, Peru, and Nigeria--clearlypursued policies of low interest rates. The average real deposit ratesover the period were -16.2%, -18.6% and -16.5% respectively, substantiallynegative in real terms. While the lending rates shown in Table 1 aresomewhat above the deposit rates, the true spreads probably were muchlarger, since the full effects of commissions, compensating balances anddiscounting could not be accurately factored into the ceiling rates.Finally, in two of the three cases the reported preferential lending ratesactually fell below the deposit rates. Thus, over much the 1970-82 periodthese three countries followed an interest rate policy directly opposed tostandard policy advice to maintain positive real rates. However, it mustbe pointed out that under the pressures of serious economic crisis bothTurkey and Peru recently attempted interest rate liberalizations.10 /

8/ For example, the rapid escalation in petroleum and transport priceswould produce higher measured inflation rates in countries where thesegoods and services have larger weights in the consumer price index andwhere these prices changes were more fully reflected in higher pricesto consumers.

9/ The real interest rate after taxes is (1 + (1-tx)*r/(l+p), where tx isthe tax rate, r is the nominal rate, and p is the rate of inflation.This is approximately (1-tx)*r-p or using the text numbers (1.0 -0.3)*0.08 - 0.09 - -3.4%. For a further discussion of the impact oftaxes see M. Darby, "The Financial and Tax Effects of Monetary Policyon Interest Rates," Economic Inquiry, June 1975, pp. 266-276 and V.Tanzi, "Inflationary Expectations, Economic Activity, Taxes andInterest Rates," American Economic Review, March 1980, pp. 12-28.

10/ See Section 6 and Annexes 5 and 7 for more complete descriptions.

- 13 -

3.06 The remaining country--UrugIuay--is set apart from the other twogroups, first because it had an average real deposit rate over the periodthat fell between the two groups, and second, but most importantly, becauseit undertook a major interest rate policy shift in 1975, about the middleof the sample period. This shift is obscured in the average figures butappears clearly in Figure 1. Between 1971 and 1973, Uruguayan realinterest rates were among the lowest recorded of the ter. countrLes. How-ever, beginning in 1974 the Uruguayan capital market was opened andinterest rate controls were lifted gradually, so that by 1978 most interestrates were determined by market forces and closely linked to internationalrates. One rather interesting observation should be maele here: The policyof market determination of interest rates did not imply positive realdeposit rates in Uruguay in every year. Real deposit rates shifted backand forth between positive and negative levels after the! reforms, as shownin Figure 1. Some of the factors behind this outcome will be discussed inSections 4 and 6.

3.07 Aside from Uruguay, a number of the other sample countries alsochanged their interest rate policies, as shown in Figure 1. Table 2summarizes the major changes in the nominal deposit rates.11/ Bangladeshraised nominal rates between 1974 and 1976, and again in 1980. Kenya andThailand raised nominal rates in 1981 and 1980, respectively. Pakistan andMorocco raised rates gradually. In contrast to the other five countries,Korea adjusted its interest rates frequently throughout the period, withnominal term deposit rates ranging from 8% to 17%. In the high inflation-low real rate group, all three countries made some changes in the level ofnominal rates: Turkey embarked on a policy of greater :reliance on marketforces for determining rates on deposits and some credit:s in 1980; Perushifted to a policy of higher nominal rates between 1978 and 1982; andNigeria raised rates a few percentage points between 1978 and 1982. As inUruguay, these changes generally followed periods of substantially negativereal rates: e.g. Bangladesh (1972-73 and, to a much lesser extent,1976-79); Kenya (1972-80); Thailand (1978-79); Pakistan (1971-75); Turkey(1973-78); Peru (1972-77); and Nigeria (1970-77).

3.08 The upward adjustments in the nominal rates were in apparentresponse to the problems created by the low real rates and in line withstandard policy advice. However, the results of the adjustments weremixed, either in terms of achieving positive real rates, or even higherreal rates. Generally speaking, the nominal rates were not raised enoughto even exceed past inflation. Thus, it would be hard to argue that theinterest rates were reset to produce positive real rates ex-ante. Insteadthe reforms relied heavily, and not always successfully, on falling infla-tion for achieving subsequent positive real rates, ex-post. In Bangladesh

11/ More detailed discussions of these reforms are contained in Section 6and in the Country Annexes. In each case, these changes wereaccompanied by similar changes in the nominal rates for generalcredits.

- 14 -

(1974-75, and 1981), Morocco (1979 and 1982), Pakistan (1975-78) andThailand (1981-82) real rates became positive after the changes in thenominal rates largely due to declines in inflation, as shown in Figure 1.In Korea the role of declining inflation is even clearer; nominal rateswere lowered slightly in 1981 and sharply in 1982, but inflation dropped soswiftly that real rates shifted from slightly negative to fairly positivelevels. Among the low inflation-high real rate group, only in Kenya (1982)was the rise in the nominal rates the principal factor in achievingpositive real rates.

Table 2

Major Policy Changes in Nominal Deposit Rates a/(percent per annum)

Reform Pre-reform Post-reformCountry Period Interest Rate b/ Interest Rate c/

Bangladesh 1974-76 4.8 7.5Bangladesh 1980 7.5 13.0Kenya 1980-82 5.4 13.2Pakistan 1973-75 5.6 8.9Morocco 1978-82 4.5 8.5Thailand 1980 7.0 10.0

Uruguay 1974-79 18.0 50.6

Turkey 1980-82 12.0 50.0Nigeria 1978 3.0 5.0Nigeria 1982 6.0 8.5Peru 1978-82 14.0 71.2

Sources: Annex 1, Summary Table 4.

a/ Each of these increases in nominal deposit rates was accompanied by asimilar increase in the rates for general credits. See Annex 1,Summary Table 5.

b/ The "pre-reform' rates refer to the rates prevailing at the end of theyear preceding the reform period.

c/ The "post-reform" rates refer to the rates prevailing at the end of thelast year of the reform period.

3.09 Among the high inflation-low real rate group, the experience withinterest rate reforms was even more mixed. In Nigeria, as in the firstgroup, the decline in the rate of inflation was more important than the

- 15 -

increase in the nominal rates in raising the level of real rates. InTurkey the increase in the nominal rates was much larger than in Nigeria,and correspondingly was a mDre important factor in the appearance ofhigher real rates, ex-post. Together with the rapid decline in inflationduring 1980-82, the Turkish interest rate reforms resulted in markedlyhigher real rates in 1980-82, rates that reached as high as +22.1% forterm deposits and +34.3% for general credits at the end of 1982. Thesehigh rate were in striking contrast to the extraordinarily negative realrates--about -60.0 percent--that prevailed at the end of 1978. However, itmust be pointed out that these liberalized rates applied only to a rela-tively small fraction of all financial transactions, particularly on thelending side. Thus, to some degree, the high real rates on general creditsreflected the large amount of low interest, directed credit which the bankswere required to extend. In addition, fears of devaluation may have pushedup rates in local currency. Finally, Peru also switched to a policy ofsubstantially higher nominal rates during the 1978-82 period, but surginginflation generally kept real rates negative, rendering the reforms fairlyineffective in raising real rates.

3.10 In sum, over most or all of the 1970-82 period, six of the tencountries studied followed reasonable interest rate policies and twoothers--Uruguay and Turkey--switched to policies of the type of policieswhich are typically recommended. Thus, overall, there was a fair degree ofconcordance with the standard interest rate policy advice in the tensampled countries.

3.11 It is also worth inquiring whether acceptance of this policyperspective has increased over the period, i.e., whether real rates were onaverage higher during the second half of the period (1977-82) than thefirst half (1970-76). As shown in Table 2, the major changes in nominalrates generally occurred after 1977. Judging from the increases in averagereal deposit and lending rates in the 1977-82 period over the 1970-76period, which are displayed in Table 3, the interest rate performanceimproved on average between the two periods. This improvement occurredthrough the combined effects of the steady rise in nomirnal rates and, mostimportantly, lower average inflation.

3.12 The improvements in interest rate performance were strongest inUruguay, due to the sweeping reforms of 1974-79, and in Bangladesh andNigeria, due to marked declines in inflation. However, Nigerian real ratesremained very low despite this improvement. Table 3 also shows that thetwo other low real rate countries--Peru and Turkey--suffered sharp declinesin the average levels of real interest rates for the second half of the1970-82 period. However, these figures mask two very different patterns inthe development of real interest rates. In Turkey, the lower average islargely due to the extraordinary negative levels reached in 1977-79, beforethe 1980 reforms. In Peru, the decline in the average reflects the failureof the interest rate reforms to keep pace with rising irLflation. This isclearly evident in Table 4, which displays the average rates for the1970-80 and 1981-82 periods. The real rates show an shaLrp increase inTurkey and a slight decline in Peru.

- 16 -

Table 3

Average Real Interest Ratesin 1970-76 and 1977-82 aT

(percent per annum)

1970-76 1977-82Term Gen. Pref. Term Gen. Pref.

Country Deposit Credit Credit Deposit Credit Credit

Pakistan -2.3 0.3 -0.7 3.0 4.4 4.1Morocco b/ -3.2 1.7 -2.2 -0.9 3.2 -1.6Korea -4.4 -0.4 -9.6 -3.8 -1.9 -6.8Thailand -4.6 2.5 -4.1 -3.5 3.3 -5.1Bangladesh c/ -9.9 -5.0 -6.8 -0.8 2.6 0.1Kenya -8.7 - -6.0 -4.5 -1.0 -2.6

Uruguay -10.9 - - 5.3 18.0 -

Turkey d/ -10.9 -8.0 -9.0 -19.5 -17.2 -27.3Nigeria -20.4 -15.4 - -11.7 -7.4 -7.8Peru -11.5 -3.8 -14.1 -26.2 -14.9 -35.7

Sources: Annex 1, Tables 1-3.

a/ See Table 1, Note 1 for definition of average real rate and Note 5 forthe specific categories of preferential credits. A dash indicates lessthan two observations.

b/ All Moroccan series begin in 1974.c/ All Bangladesh series begin in 1971.d/ All Turkish series begin in 1974.

3.13 An examination of Figure 1 reveals another interesting pattern.Each of the ten countries, and the U.S., experienced a noticeable declinein its annual real interest rates around 1973-74 and again around 1979-80,and a noticeable rise around 1981-82. Table 4 highlights the 1981-82upswing by comparing the average real interest rates over the years 1970-80to those of 1981-82 for each of the ten countries and the U.S. As shownthere, real interest rates rose sharply in all the countries except Peru.These contemporaneous swings in the annual real interest rates suggest thatcommon factors were at work in all of the countries, despite theirseemingly different economies and degrees of financial openness. Thefollowing section offers some explanations for these contemporaneous move-ments and provides a discussion of the determination of interest rates on ayear-to-year basis.

- 17 -

Table 4

Average Real Interest Rates,1970-80 and 1981-82 a_

(percent per annum)

1970-80 1980-82Term Gen. Pref. Term Gen. Pref.

Country Deposit Credit Credit Deposit, Credit Credit

Pakistan -0.8 1.4 0.7 5.4 6.2 6.3Morocco b/ -2.9 1.7 -2.7 2.6 6.2 1.5Korea -5.8 -2.6 -10.8 5.8 7.8 6.8Thailand -5.9 1.2 -5.8 6.1 12.8 2.3Bangladesh c/ -7.7 -3.3 5.2 6.8 9.6 5.6Kenya -7.9 -2.4 d/ -5.6 -0.5 2.0 -0.6

Uruguay -15.3 11.2 e/ - 15.6 23.9 -

Turkey f/ -23.0 -22.0 -23.6 17.9 29.6 -6.5Nigeria -17.6 -12.9 -8.4 g/ -10.2 -5.6 -7.0Peru -18.1 -9.4 -21.8 -21.7 -7.2 -39.6

U.S. -1.6 0.8 - 4.6 7.9 -

Sources: Annex 1, Tables 1-3, and the U.S. Federal Reserve Bulletin, forthe U.S. data.

a! See Table 1, Note 1 for definition of average real rate and Note 5 forthe specific categories of preferential credits. Also a dashindicates less than two observations.

b/ All Moroccan series begin in 1974.

c/ All Bangladesh series begin in 1971.

d/ 1977-80; the corresponding rates for term deposits anLd preferentialcredits were -6.4% and -3.4%, respectively.

e/ 1976-80; the corresponding rate for term deposits waim -4.5%.

f/ All Turkish series begin in 1974.

&/ 1978-80; the corresponding rates for term deposits and general creditswere -11.3% and -6.4%, respectively.

- 18 -

IV. THE DETERMINATION OF INTEREST RATES: NOMINAL AND REAL

4.01 As stated in Section 2, the interest rates quoted in this studyare not market rates, with the exception of the post-78 Uruguayan figures.Instead the rates were either set by the government or reflect such statu-tory rate ceilings or floors.12 / Thus, the movement of interest ratesdiscussed in this study is principally the mDvement of government adminis-tered interest rates. An investigation into the determinants of theseinterest rates is therefore really an examination of government behavior insetting interest rates.

4.02 Before proceeding with this investigation, it is necessary topoint out that government intervention in financial markets is pervasive,even in the industrialized countries. First, reserve requirements andtaxes affect the levels of interest rates in all countries. Second,controls are often placed on interest rates. For example, even the U.S.financial system, which is among the most market-oriented, was subject toRegulation Q until 1983. This regulation established ceilings on theinterest rates paid on many important classes of deposits. Since interven-tion exists everywhere, the relevant issue is its impact not its existence.

4.03 A useful basis for investigating the impact of government ratesetting is a comparison between the observed rates and those that mighthave occurred had a free market prevailed. This standard is useful becauseit provides a convenient benchmark for comparison, and because recentfinancial sector work has increasingly called for greater reliance onmarket forces in the determination of interest rates. Thus the questionis: "Did governments set rates so as to simulate market rates."

4.04 To answer this question, it is necessary to have a broad under-standing of what determines market rates. Here there are basically twotheories. The most well known theory, and the one that forms the basis ofthe standard recommendations on real interest rates, argues that competi-tive financial markets would establish nominal interest rates on depositsthat are positive in real terms, because savers must be induced to holdfinancial rather than real assets, and, on average, real assets grow innominal terms at the rate of inflation. Thus, the nominal deposit interestrate must equal the expected inflation rate plus a small underlying realrate. This real rate provides the incentive to hold financial rather thanreal assets. Lending rates, in turn, will also be positive in real terms,since they are based on the cost of deposits--the rate paid todepositors--plus a margin covering the cost of intermediation: reserve

12/ In the case of Pakistan a weighted average of rates subject togovernment controls was reported and in the cases of Turkey and Peruan attempt was made to capture the effect of discounting andcompounding on the ceiling rates.

- 19 -

requirements, taxes, risk, administrative costs, overhead and the return toequity. This theory is know as Fisher's Theorem.13/

4.05 An alternative, more recent theory suggests that the foregoingFisher theorem--Nominal Rate = Real Rate + Expected Inflation--is not theappropriate model for market determined interest rates in an open economy.The more recent theory stresses the potential substitutions between assetsdenominated in domestic currency and assets denominated in foreigncurrency, rather than the substitution between goods and assets that formsthe basis of the Fisher Equation. This theory predicts that the nominalinterest rate on domestic assets will equal the "world" interest rate,adjusted for the expected rate of devaluation (and for risk). According tothe theory, this condition--Uncovered Interest Parity--is assured by fluidinternational capital flows, which rapidly respond to any divergence in thedomestic and world rates. For example, suppose expectations of a largerdevaluation were to develop, because of an expansionary monetary and fiscalpolicy or a decline in export prices. An incipient capital outflow wouldraise local interest rates until the marginal asset holders were justindifferent between foreign and local currency assets, i.e., until theinterest differential or spread between local currency and foreign currencyassets just covered their expectation of devaluation.L 4,

4.06 Notice that the interest parity theory and the Fisher theory mayyield significantly different results if the expected rate of devaluationis not equal to the expected difference between local anid worldinflation.15 / In particular, if the expected rate of devaluation substan-tially exceeds the expected inflation differential, because the currencyhad become obviously overvalued, then local interest rates might reachextremely high real levels. Such high real rates simply would reflect thepublic's increased desire to borrow in local currency and to lend ordeposit in dollars. Thus, the effects of expected devaLuation provides oneexplanation of the abnormally high real rates recently observed in somemiddle income countries.

13/ It is worth noting that (1) Fisher actually did not claim thatpositive real rates would prevail in each and every year, nor even onaverage, see I. Fisher, The Theory of Interest (MacMillian, 1930), and(2) that the effect of inflation would differ, depending on whether itwas foreseen or unforeseen. Fama's work Mp. cit. suggests thatexpected inflation was fully incorporated into U.S. Treasury Billrates. However, recent evidence has raised some questions about thegeneral empirical validity of the theory, see L. Summers, "TheNon-adjustment of Nominal Interest Rates: A Study of the FisherEffect," in J. Tobin Macroeconomics: Prices and Quantities, Brookings,Washington, DC, 1983 and R. Saracoglu op. cit. and works cited there.

14/ A similar argument can be made for lending rates, althoughl that spreadis not investigated in detail here.

15/ The two theories also could yield different results if the riskpremium on one of the currencies is significant.

- 20 -

4.07 In addition to the case where expectations of devaluation exceedthe differential in inflationary expections, an alternative case exists,namely: If the expected rate of devaluation became significantly less thanthe expected differential in inflation rates, then market-clearinginterest rates in local currency rates would become depressed, possiblyeven negative in real terms according to the Fisher equation. This wouldreflect the public's increased desire to borrow in foreign currency andlend or deposit in local currency. The point here is that recommendationsfor market determined interest rates and recommendations for positive realinterest rates can be inconsistent if market interest rates are determinedin accordance with the Uncovered Interest Rate Parity Model.

4.08 "Do governments set interest rates in accordance with either theFisherian Model or the Uncovered Interest Parity Model?" Judging by thedata collected this study, the answer is no. In general, the administeredrates were at best only partially adjusted to maintain real returns andreal costs of credit in the face of actual inflation or devaluation.

4.09 Figure 2 illustrates the relationship between the real rate ofinterest and inflation. If nominal rates had been set according to theFisher Model, then the real rates would have remained stable and would showup in the graphs as horizontal lines or, given errors in expectations,would vary randomly around a constant level. A quick glance at Figure 2reveals that real interest rates do not follow this pattern (the U.S.included). Instead they are basically a mirror image of the inflationrate. This means that governments typically adjusted the nominal interestrate by much less than actual inflation, leaving variations in the realrate to be largely determined by variations in inflation.16/

4.10 The observation that the inflation rate predominantly determinedthe real rates also explains the contemporaneous variation in the annualreal interest rates across countries (again see Figure 2). Specifically,real rates moved together because domestic inflation rates moved together,and because domestic nominal rates were not fully adjusted for futureinflation. In particular, real rates declined sharply around 1973-74 andagain around 1979-80 in both the sample countries and the U.S.,

16/ This pattern also shows up in some regression analyses of nominalinterest rates undertaken as background for this paper. Regressionsof the nominal interest rates on future and lagged inflation yieldedstatistically insignificant coefficients for both variables, ratherthan coefficients summing to one, as would be implied by the FisherianTheory. In other words, the nominal interest rate was fairlyindependent of the inflation rate, whether future or lagged. This, inturn, implies the real rate is largely a function of the inflationrate. These regression results also support the view that it matterslittle whether lagged or future inflation is used as a proxy forexpectations; since nominal rates were fairly constant over theperiod, the difference between using lagged and future inflation ratesonly affects the timing of the variation in real rates, not theaverage real rate. (See also Footnote 3).

- 21 -

due to the oil-price shocks and the rise in worldwide Lnflation.17/ In1981, and in most cases 1982, real rates were sharply higher. This wasprimarily due to a general decline in world inflation, although increasesin local nominal rates also played a role (See Sections 3 and 6).

4.11 Finally, differences in average real interest rates over theperiod largely reflect differences in average inflation rates: Countrieswith higher average real rates tended to be countries with lower averageinflation. Lower rates of inflation permitted these countries to maintainlow nominal interest rates without having real rates which were excessivelynegative. Countries with highly negative real rates on average tended tobe countries with higher average inflation. This indicates that in thesecountries governments were typically unwilling or unable to set nominalinterest rates in line with average inflation. The exception is in Uruguayafter 1978, when rates were set in the free market. Given this historicalpattern of government policies toward nominal interest rates it seems thatmaintaining a low rate of inflation is critical to maintaining a stablereal interest rate policy.

4.12 If anything, the Uncovered Interest Parity Ilodel fares even worsethan the Fisher equation as a theory of government rai:e setting. Figure 3graphs the local nominal deposit rate in each country and the U.S. Treasurybill rate adjusted for the ex-post rate of devaluation. If uncoveredinterest parity had held, then the two lines would have moved roughly inparallel. However, Figure 3 shows that the gap betweean the two ratesfluctuated wildly. This fluctuation indicates governments did not adjustthe deposit rate to reflect future devaluation as would have occurred withmore open capital markets.

4.13 As mentioned above, Uruguay, from 1978 onward, was the only casein the sample where interest rates were allowed to be freely determined inan open capital market. After the reforms, Uruguayans could deposit andborrow locally in either dollars or pesos and there mere no restrictions incapital flows. In addition, reserve requirements were eliminated in 1979and there were no taxes on interest rates. Thus, an examination of Uruguayin this period provides an idea of the behavior of a free market interestrate in a financially open, developing economy. The following discussionsuggests that the uncovered interest parity model works tolerably well,provided it is assumed that expectations of devaluation were fairlyconstant.

17/ It could be argued that governments kept nominaL interest rates low inan effort to manipulate inflationary expectatioas, i.e., not to beseen as admitting that inflation will be higher in the future.Clearly, such a policy is extremely limited in its potentialeffectiveness, since its credibility is rapidly eroded by negative andreal rates, ex-post.

- 22 -

FigEur 2

Ex-post Real Term Deposit Rate versus Inflation

- EX-POST REAL INTEREST RATE ON TERM DEPOSITS <8 MONTH)

.INfLATION RATE CDECEHBER TO JUNE, ANNUALIZED)

130

le0

70

X 40,-^

-20 /

-50 BANGLADESH

-e0

130-

70-

X 48-

-20-

-35 KENYA

-I I I I I Ioo

138-

78-

X o 4 - - .. . .'-------.... -

-20-._.8

-50 KOREA

-as I97 I I I1978 I I 1

1970 1972 1974 1978 1978 190e 1982

- 23 -

Figure 2 (continued)

- EX-POST REAL DiTEREST RATE ON TERM DEPOS1TS CO MONTH)

---- INFLATION RATE CDECEB4ER TO JUNE, ANNUALCZED)

130-

lee-

70-

X 48-

1 ............... ...................... ,,,,... -.... ,,,._

-20-

6-8 MOROCCO

138-

leg-

78-

1800

130-

70-

4I . , . …

-20-

-50 PAKISTAN

-80- I I I I I I I I I .

1970 1972 1974 1976 19711 1980 1982

- 24 -

Figure 2 (continued)

EX-POST REAL INTEREST RATE ON TERH DEPOSITS (a MONTH)

--. INFLATION RATE (DECEMBER TO JIJNE, ANNUALIZED)

160-

130-

188-

708

408

-20

160-

130-

t00-

70-

A 40 -.

-20-

-50- THAILAND

-80- I - X l l l

180 - :

130-

100- : '

70- . ''.

40-

10- .... * *...--- .. ..

-20

-50 TURKEY

-80-

1970 1972 1974 1076 1978 1980 1982

- 25 -

Figure 2 (continuedp

EX-OST REAL INTEREST RATE ON TERIM DEPOSITS CO MONTH)

.INFLATION RATE (DECEtUER TO %ANE, ANNUALIZED)

160-

130-

le- . ---.

709

i80-

10..-20-

-so -

-70- UITED STATES-69a- UNTE TAE

1970 1972 1i974 1970 1978 198n 1982

- 26 -

Figure 3

Nominal Term Deposit Rate vs Devaluation-adjusted US T-Bill Rate

- NOMINAL INTEREST RATES ON TERM DEPOSITS CO MONTH)

.NOMINAL 6 MO. US T-3ILL RATE, ADJUSTED FOR DEVALUATION

120-. .

x .9........39-

O~~~~~~~~~- ------,,---, . ... . .

-39- BANGLADESH

X~ ~ ~~~~~~~~~~---- ---------.

-30- KENYA

-69-

120-

90-

X eo- ' ....

30-

-30 KOREA

1970 1972 1974 1978 1978 1980 1982

- 27 -

Figure 3 (continued)

NOMINAL INTEREST RATES ON TERM DEPOSITS C6 MONTH)

. NOMINAL 6 MO. US T-SILL RATE, ADJUSTED FOR DEVALUATION

160

31.9

O-,9.-. . ... *-*

-So MOROCCO

-68-1 I * -- T

120-

3130- .....

139

-SO- NIGERIA

128 -

X 69 ' ''' *

SO- .. " "O- '. . -~~~~..... .. ........

-358 PAKISTAN

-60- - * I - I I I I I I

1970 1972 1974 1976 1978 19s 196Z

- 28 -

Figure 3 (continued)

- NOMINAL INTEREST RATES ON TERM DEPOSITS CO MONTH)

.-- NOM33rAL 6 MO. US T-FILL RATE, ADJUSTED FOR DEVALUATION

50-

30-

60-

-30 - PERU

10

90

e460-

30

0.

-30 THAILAND

-60-

150- ." "

120-

90- . .".

30-

0-. ,*--

-30- TURKEY

-80- I 97I 149 1 1 9- I 9

1970 1972 1974 1978 1978 Igoe 1 98Z

- 29 -

Figure 3 (continued)

-NOMINAL INTEREST RATES ON TERM DEPOSITS CO MONTH)

..... NOMINAL B MO. US T-WILL RATE, ADJUSTED FOR DEVALUATION

159- :

90-~~ .' '." .

x 0 69 - '-

=i:: r URUGUAY

19710 1972 1974 1976 1976 1969 1962

- 30 -

FigureiNominal Rates on Domestic Currency and US$ Term Deposits

Uruguay: 1978-82

80-

70-

80-

se-

%40

38-

s _

2e0

10 -........... 0 -.- " .

1 2 3 4 1 2 3 4 1 2 3 4 i 2 3 4 t 2 3 4

1978 1 979 1980 1 981 1982

- DOMESTIC CURRENCY RATES

.... DOLLAR RATES

- 31 -

4.14 Figure 4 plots the nominal interest rates on Feso and dollardeposits in Uruguay in each quarter between 1978 and 1982. As shown there,it took a few quarters for the market to adjust to its new freedom, butafterwards the spread between peso and dollar deposit rates remained fairlyconstant from the end of 1978 through the end of 1981. 18/ In particular,the spread fell from around 40%, in the first three quarters of 1978, tothe 30-35% range in the last quarter, where it generally remained untilmid-1982. After mid-1982, the spread shot up, reflecting the public'sgrowing expectations of a maxi-devaluation--one that finally took place inNovember. Three of the observations with spreads outside the 30-35% range(below) occurred in the second and third quarters of 1981 and the first of1982, when the Central Bank and the Banco de la Republica respectivelyoffered inexpensive forward contracts for peso depositors. 19/

4.15 The relatively constant spread between peso and dollar depositrates implies that the real interest rate in local currency rose over mostof this period. This rise was due to (1) the rise in dollar interest ratesworldwide, and (2) the falling rate of inflation in Uruguay. By the end of1981, the real deposit rate in pesos had reached 32.9%, ex-post. Thisexperience illustrates the point that an open capital maLrket may produceexceptionally high positive real interest rates.

4.16 It should also be noted that the relatively constant spreadbetween peso and dollar deposit rates in fact did not reflect the ex-postrate of devaluation. Between 1978 and mid-1982, the UrLguayan governmentslowed the rate of devaluation as an anti-inflationary umeasure. However,the spread did not decline by the same amount, as might have been expectedfrom the theory. Moreover, during much of this period i:he rate of devalua-tion was pre-announced for the following six months. Tlus, the uncoveredinterest rate parity theorem can only be said to hold if it is assumed thatthe public's expectations were constant. This implies t:hat either (1)expectations were incorrect for about two and one-half years; or (2) thepublic's expectations of devaluation reflected not only the pre-announce-ment, but what was perceived as an increasing risk of a maxi-devaluation in

18/ The actual percentage point spreads between the nomninal 6 month pesodeposit and the nominal 6 month dollar deposit, from 1978 Ql to 1980Q4, are as follows: 41.4, 38.8, 42.8, 34.6, 34.0, 30.3, 32.4, 38.7,34.2, 39.7, 36.7, 35.7, 33.5, 27.8, 29.2, 34.3, 28.6, 33.1, 45.7, and56.0.

19/ In J. Hanson and J. Demelo, "External Shocks, Financial Reforms andStabilization Attempts in Uruguay: 1974-1983," WorLd Development, 1985regression analysis is presented which statistically supports thehypotheses of a constant spread between the peso and dollar depositrates in this period, and of the spread-reducing effects of theguarantees.

- 32 -

excess of the schedule, owing to an increasing overvaluation of theUruguayan peso.20/

V. FINANCIAL SECTOR LIBERALIZATION: THE "NEW" VIEW

5.01 A growing consensus has been forming in the development litera-ture regarding appropriate financial sector policies.21 / This "new" vieweschews the "standard" policy prescription of simply maintaining positivereal interest rates. Instead it emphasizes the need to achieve a generalimprovement in the functioning of the financial sector, through greaterreliance on market forces. With respect to interest rate policies, the newview recognizes that in most developing countries interest rates do notfully reflect market forces but are influenced substantially by governmentinterventions. These interventions often prevent interest rates fromeffectively performing the allocative and incentive functions of marketdetermined prices. Instead, non-price rationing of credit at below-marketrates plays an important role in financial markets and a large number oftransactions either occur in fragmented or foreign markets or aresupplanted by inefficient forms of self-finance. A growing recognition ofthe economic costS of the distortions and market failures induced by manyforms of government intervention in financial markets, plus increaseddoubts as to the effectiveness of many such programs, have provided much ofthe impetus behind the wider acceptance of the more market-orientedapproach to financial sector policy.

5.02 The objections raised to extensive and complex regimes of finan-cial sector interventions center on: (1) the rationality of the implicitincentive structure which develops, (2) the real costs associated with the

20/ This overvaluation reflected many factors: (1) the maintenance of arate of devaluation below the rate of inflation; (2) the 1979 terms oftrade shock; (3) the potential collapse of the Argentine exportmarket--a loss which actually occurred after 1981; (4) the dependenceon volatile foreign capital inflows to maintain the schedule ofdevaluations--an inflow which slowed sharply after the Mexicansuspension of foreign debt service in August 1982; and 5) the riskthat the Uruguayan fiscal deficit could explode at any time andrequire inflationary finance--an explosion which actually occurred inlate 1981. See J. Hanson and J. DeMelo, "The Uruguayan Experiencewith Stabilization and Liberalization,1974-1981", Journal ofInteramerican Studies and World Affairs, Nov. 1983 pp. 477-508 for adiscussion of the pre-announcement policy and its relation to theovervaluation.

21/ Two seminal books for this perspective are Ronald I. McKinnon, Moneyand Capital in Economic Development, (Washington, D.C.: BrookingsInstitute, 1973); and Edward Shaw, Financial Deepening in EconomicDevelopment, (New York: Oxford University Press, 1973). Much of this"new" perspective, as it relates to World Bank work, can be found inMillard Long, "Review of Financial Sector Work," INDFD, 1983.

- 33 -

administration of such regimes, and (3) the effectiveness of the regimes inattaining their professed policy goals. Complex financiaL sector interven-tions are likely to generate an irrational incentive structure, sinceadministered credit schemes, and the rationing mechanisms embodied in them,cannot be expected to effectively manage all the information necessary toensure that available financial resources are matched wit!i the sociallymost productive set of real investment opportunities. In particular,directed credit schemes bias the credit allocation process towards per-ceived low-risk investments, i.e., capital intensive projects taken on bylarge firms (often parastatals) operating behind high tariff barriers orwith local monopolies. The same credit rationing limits the amotnt ofresources allocated to more labor intensive investments and to smallerenterprises with less collateral, uses of credit which would be economic-ally more efficient. The maintenance of such distortions can add up tolarge efficiency losses and worsen the distribution of income.

5.03 In addition, the distributional impact of such regimes is ques-tionable. Non-preferential borrowers, tax payers and depositors end upbearing the burden of directed credit at below-market rates. Non-preferen-tial borrowers pay through queueing costs and, when non-preferential ratesare allowed to rise, through higher borrowing rates. Tax payers ultimatelypay any direct sudsidies granted by the treasury. When the subsidies arecovered through inflationary finance, all holders of finaLncial assets bearthe burden of the inflation tax. Moreover, depositors are often forced tosubsidize preferential borrowers through interest rate ceilings ondeposits. Thus, the net distributional impact of directed credit isuncertain. In addition, the lower real returns implied by the ceilings ondeposit rates induce savers to hold wealth in non-financial forms. Thussome of the potential efficiency gains from formal financial intermediationare lost.

5.04 Administrative costs associated with large scaLe intervention arealso important. The sheer paper work associated with extensive regulationsand the application and screening process absorbs a greal: deal of scarceskilled human resources, as do the various methods of evading control, suchas overbranching to compete for deposits in the presence of deposit rateceilings and sending capital abroad illegally. Other economic lossesassociated with intervention include the unintended suppression of equitymarkets, promotion of "groups" of interlocking banking aad industrialconcerns, and promotion of abuses of discretionary power.

5.05 The development of equity markets is suppresse,1 by low lendingrate ceilings, which makes formal credit artificially cheap for those firmsthat can obtain it. As mentioned previously, low lending rate ceilingscause credit to be allocated to the large, established firms with greatercollateral. However, it is precisely these firms which would be bestsuited to raise capital in equity markets, as opposed to the small andmedium enterprises. As a result, large firms refrain from equity financeand the equity markets do not develop. This overdependence on (debt financeis potentially destabilizing--as recent Latin American experience has madevery clear.

- 34 -

5.06 Low ceilings on lending rates and large directed credit programsalso promote the formation of "groups" of banking and industrial concerns.These "groups" form either because the firms buy up the banks to assurethemselves of access to cheap credit, or the banks buy up the firms torecapture the transfers to borrowers that are implicit in low lendingrates. These "groups" are potentially destabilizing, because theystimulate less-than-arms-length lending practices. Moreover, the provisionof low interest credit to the "groups" worsens the income distribution.

5.07 The mere act of "administering" financial transactions grantstremendous discretionary power to individuals and/or official bodies.Often this power provides too great an opportunity for abuses. Theseabuses can have tremendous moral as well as economic costs. While thetotal losses in equity and efficiency due to corruption and other by-products of financial market intervention are of course unfathomable, theydo represent a potentially serious drain on the country's developmenteffort, and thus every attempt must be made to weigh them against theanticipated gains from intervention.

5.08 In addition to the growing recognition of the costs associatedwith the distortions introduced by government intervention, there aregrowing doubts as to the effectiveness of many government programs inattaining such professed policy objectives as: (1) directing resourcesinto a sector or subsector deemed by the authorities to have high socialrates of return, and (2) increasing the flow of income towards identifiablegroups in the population. In particular, the fungibility of financialresources allows substantial "leakages" of directed credit from targettedsectors into other non-targetted ones. Also, the politicized nature of thenon-price rationing mechanism apparently often leads to income enhancementof politically powerful groups, rather than those ostensibly targetted inthe programs. For example, credit which is supposedly directed to smallfarmers for grain production might wind up in the hands of large,politically influential farmers for the purchase of mechanized harvestingequipment. Or the credit provided by the programs might free up resourcesthat otherwise would have been invested in the sector for investment insuch areas as urban luxury housing.

5.09 In recognition of the above concerns, the market-oriented policyperspective suggests a multidimensional approach to financial sectorpolicy. According to this view there are three broad objectives offinancial sector policies: (1) to mobilize adequate amounts of resources,

- 35 -

both domestically and internationally;22 / (2) to allocate credit to itssocially most productive uses; and (3) to provide a stable financialenvironment, which will encourage development.

5.10 In terms of interest rates, these objectives can best be achievedby allowing rates to vary according to the state of the business cycle,pressures on the foreign exchange market and the risk and maturity of theindividual transactions. Accordingly: (1) Interest rates need not bealways positive in real terms, as local and internationaL pressures maycause markets to clear at nominal rates below expected iiiflation.(2) There should be some correspondence between local interest rates andforeign rates adjusted for expected devaluation, lest exchange markets andcapital flows become unstable. (3) Given that there is not one but manyinterest rates, greater reliance should be placed on market forces forinterest rate determination, so as to adequately reflect all the multiplecombinations of maturity, risk and administrative cost.

5.11 The new perspective recommends that subsidies and below-marketrates generally should be avoided, since they encourage an inefficientallocation of resources, discourage resource mobilization, and haveuncertain distributional implications. Correspondingly, complex systems ofdirected credit also should be avoided, since the interetst ratedifferentials which are established are unlikely to adequately reflect therelative productivities of financial resources across the sectors.Moreover, the fragmentation of financial markets necessary to make such adifferentiation effective is neither feasible nor desirable.

5.12 The new perspective also recognizes the need for greaterfinancial market supervision and greater consistency in the mix of fiscal,monetary, exchange, commercial and trade policy. A re-allocation ofskilled human resources into the supervisory arm of the monetary authorityis essential, since fraud and mismanagement are antitheitical to thedevelopment of a well-functioning financial market. Their potentialnegative externalities create a legitimate role for constructiveregulation of the financial market place.

5.13 Lastly, the new perspective recognizes the complex and pervasiveinterrelationships between the real economy and the financial markets.Given these interrelationships it is crucial that the government closely

22/ Preliminary regression estimates of the demand fot financial assetsusing the data gathered for this study gave statistically significantpositive real interest rate elasticities, both in the time seriesequations for each country and in a cross-sectiondL equation utilizingpooled period-average data. Given the important policy implicationsof the numerical values of these elasticities, and the possibility ofimprecision given the small size of the data sample used here, theauthors have decided to undertake a broader study of the demand forfinancial assets, involving more countries over a longer time period.See also A. Lanyi and R. Saracoglu, Interest Rate Policies inDeveloping Countries, IMF Occasional Paper, 1983, for a recentanalysis of the effects of higher real interest rates.

- 36 -

coordinate its monetary, fiscal, exchange rate and trade policies. Eventhe best financial sector policy regime is unlikely to function properlyin the face of contradictory signals coming from the real-side of theeconomy, i.e., where financial and economic rates of return differ widely.

VI. FINANCIAL SECTOR REFORMS: THE RECORD

6.01 This section assesses the financial sector policies ofBangladesh, Kenya, Nigeria, Peru, Thailand, Turkey, and Uruguay, in termsof the growing consensus regarding the importance of a well-functioningfinancial system and the need for greater reliance on market forces.Evaluating the correspondence between this "new" view and the countries'policies is obviously much more difficult than comparing real interestrates. Both the "new" policy prescription and the evaluation of itsimpact require qualitative judgments of many factors, not simply a mechani-cal application of an interest rate formula. Moreover, all countries, evenmarket-oriented, developed countries, intervene in financial markets tosome extent, typically with government-imposed ceilings on interest ratesand with programs to provide credit on preferential terms to sectors suchas agriculture, housing, and small business. Thus, the central task ofthis analysis is to judge the degree to which such common interventionsdistort the mobilization and allocation of financial resources.

6.02 This section draws on the individual country analyses containedin Annexes 2-8, and the references cited there. It takes the followingform: First the most common types of financial market interventions arediscussed. Second, the degree of intervention across countries isassessed by evaluating the impacts of both the size of directed creditprograms and the distortions implied by the structure of interest rates.Third, the progress toward greater reliance on market determination off inancial markets in the seven countries is assessed in terms of increasesin ceiling rates, simplifications and reductions in the scope of directedcredit programs, and other market-oriented reforms.

6.03 Each of the seven countries maintained directed credit programs.Three types of instruments typically were used: (1) regulations on theportfolio composition of intermediaries; e.g., requirements to devote acertain portion of lending to specific activities; (2) Central Bank redis-counting of credits to priority sectors, usually at subsidized rates; and(3) control of financial intermediaries through direct ownership. Thesegovernment controlled financial institutions included: commercial banks,mortgage banks, agriculture or industrial finance companies, export creditagencies, and institutions specializing in parastatal lending.

6.04 Public sector borrowing to finance the fiscal deficit and publicenterprises was also an important factor in developing countries. Regula-tions often existed forcing the system to hold low-interest governmentdebt: either directly, in the intermediaries' portfolios, or indirectly,through Central Bank reserve requirements, which in turn were used to pro-vide low-interest Central Bank credit to the public sector.

- 37 -

6.05 Both the directed credit programs and the pubLic sector borrow-ings tended to reduce the supply of credit available to non-preferredborrowers, i.e., the well known "crowding out" phenomena. Increasedcompetition for the remaining credit drove up interest rates, in some casesproduced very high rates on non-preferential credits. M4oreover, thespreads between unsubsidized credits and deposit rates t:ended to widen,unless the preferential credits were at near-market rates, or the govern-ment compensated the financial sector for its lost revenues on low interestcredits with direct subsidies.

6.06 Each of the seven countries also maintained some form of adminis-tered interest rate regime, with the exception of Uruguay after 1978.While these regimes varied substantially from country to country, theytypically established ceiling deposit rates and some control over otherinterest rates, including those on both preferential and general bor-rowers. Thus the principal differences between the countries were in therelative size of the directed credit programs and in the interest ratedifferentials between the various types of preferential credits, betweenpreferential and general credits, and between preferential credits anddeposits.

6.07 While it is difficult to produce a precise estimate of the sizeof the directed credit programs, a rough idea can be derived from thefollowing observations from Annexes:

In Nigeria, government control of financial, resources was almosttotal, via establishment of detailed credit allocation guidelinesfor each of the sixteen sectors into which the economy wasdivided. In addition the government directly controlled thecountry's large oil revenue, much of which it: channelled into thepriority sectors.23/

In Turkey, a recent World Bank study estimated that seventy fivepercent of total credit was subject to government regulations ofthe type described above.

In Peru, the Banco de la Nacion, a public institution whichfunctions primarily to finance the government and the publicenterprises, accounted for 21% of total credits in June 1982.This large figure was, nonetheless, less than half of the Banco'saverage share of credit during the period 1976-78. In addition,in 1981 the State Banks and COFIDE, the government's apexindustrial development institution, accounted for another 34% oftotal credit. Thus, the Peruvian authorities had substantial

23/ In oil-based economies the government typically plays a dominant rolein credit allocation, not only through control of oil revenues butoften through financial intermediation in foreigrL capital markets.For example, in Ecuador and in Indonesia after 19174, the governmentundertook foreign borrowings which exceeded its actual needs. Theexcess was allowed to accrue as deposits in the central bank while thecentral bank simultaneously expanded its lending.

- 38 -

influence in the allocation of more than half the country'scredit.

Even in Uruguay, despite the far-reaching liberalizationmeasures, the publicly owned Banco de la Republica and BancoHipotecaria accounted for 37% of all credit in 1982. While muchof this credit was at market terms, a large share remained intargetted lending.

Bangladesh allocated as much as 60-70% of all domestic credits tothe public sector.

In Thailand commercial banks were required to lend 15% of theprevious year's deposits to agriculture. The share of directedcredit in total credit was about 33%.

In Kenya the commercial banks were required to commit 17% oftheir deposits to agricultural lending, though estimates are thatthe actual compliance was only about 13%. Moreover, in Kenya thegovernment channelled a substantial amount of credit to theparastatals via the development banks, financing both investmentsand operating losses.

In sum, government intervention in credit allocation was substantial, evenin the countries with low inflation and relatively high real interestrates.

6.08 To get a better picture of the extent to which such directedcredit policies distorted financial markets, it is necessary to analyze thedifferentials between rates on the directed credit and rates on generalcredits and deposit rates. Clearly, the data cannot provide a preciseestimate of these distortions. However, a rough sense can be gathered fromTable 5, which displays a sample of these differentials as of year-end1982. Moreover, it should be noted that interest rate ceilings limiteddeposit rates in all the countries except Uruguay, as discussed inSection 5.

6.09 A word of caution is required before interpreting the figures inTable 4. Since the preferential rates cited for each country are quitedissimilar and the effective rates cannot be accurately computed, thecalculated spreads are not overly precise. Moreover, there are numerouspreferential rates and only one is cited here. Bearing these warnings inmind, it nonetheless seems that the countries fall into three distinctgroups. The first group--Bangladesh, Kenya, Nigeria--appear to havelimited the degree of subsidization, since the differential between thepreferential rate and the rate on general credits was relatively small.However, this is not to say subsidization was absent, since subsidizationmay take the form of quantity rationing, longer grace periods, easiercollateral requirements, or other adjustments to the lending terms. Themiddle country--Thailand--appears to have engaged in some subsidizationthrough interest rates. In the third group--Peru and Turkey--subsidizationwas extreme. The countries also breakdown along the same lines when

- 39 -

Table 5

Nominal Interest Rate DifferentialsYear-end 1982 a/

(percent per annum)

Preferential over Preferential overCountry General Credits Term Deposits

Bangladesh -4.0 1.0Kenya -4.0 1.2Nigeria -1.5 4.0Thailand -11.0 -3.0Peru -69.0 -42.2Turkey -37.0 -32.0

Sources: Annex 1, Tables 4-6.

a/ See Table 1, Note 5 for the specific categories of preferentiallending.

comparing preferential rates to deposit rates. Since the differentials inthe first group were positive--preferential credit cost more than thereturns on term deposits--their preferential rates were less likely todistort investment incentives and encourage diversion of directed creditthan the preferential rates of the second or third groups.

6.10 In addition to the differentials between preferential lendingrates, non-preferential rates and deposit rates, the differentials betweenvarious categories of preferential rates establish important incentiveswhich may, or may not, be consistent with relative productivities ofinvestments or overall policy objectives. There was considerable diversityamong the countries in the complexity of preferential lending rates and inthe degree of detailed specification of the targetted subsectors. Thisdetail is difficult to summarize concisely, but some idea can be obtainedfrom the tables in the Annexes (see Table 4, "The Pattern of Lending Rates"in each of the respective Country Annexes, 2-8). As shown there, thedegree of specificity and the interest differentials vary widely. Ingeneral terms, the countries of the low-inflation group maintainedrelatively small differentials between the different sectors, although insome cases they did apply detailed breakdowns of subsidized credits.

6.11 The small differentials in the low inflation-high real ratecountries probably reflect the fact that interest rates are low in thesecountries and thus the differentials simply could not be very great. Incontrast, nominal rates in the high inflation countries tended to besomewhat higher. These higher nominal rates allowed mnore room for interest

- 40 -

differentials particularly in Turkey and Peru.24/ In Uruguay after theliberalization, however, the existence of market determined deposit rateslimited the extent to which the state-owned financial institutions couldsubsidize credit, despite the important role of these institutions in themarket. 25/

6.12 Most of the seven countries in the sample initiated financialsector reforms in recent years, reducing the complexity of the preferentialrate systems, raising nominal rates to accommodate market pressures andeven releasing control over selected credits and deposits. On a country bycountry basis, starting with the three countries in the low inflation-highreal rate group--Bangladesh, Kenya, Thailand--followed by the four higherinflation countries--Nigeria, Turkey, Uruguay and Peru--the reforms aresummarized below:

6.13 Bangladesh markedly raised most ceilings on interest rates inlate 1980. Recently the government also moved to ease the rigidities ofits institution-specific ceiling on credit expansion. This was done bymodifying some of the restrictions which previously prevented the trans-ferral of banks' resources to banks constrained by their credit meilingsbut with a surplus of investment opportunities, from banks below theircredit ceilings but lacking good investment opportunities.

6.14 In Kenya, the structure of interest rates was substantiallyraised over the 1980-82 period. Beyond these readjustments, Kenya did notundertake any sweeping reforms. In large part, this was attributable tothe country's history of relative price stability and relatively market-oriented economic policies, which kept pressures from building to a levelthat would force major reforms. However, the consequences of nearly adecade of fiscal imbalance and an abrupt constriction in the availabilityof foreign credits have resulted in a serious economic crisis, out of whichsome financial sector and macroeconomic policy reforms can be expected inthe future. In particular, the authorities must find a way to finance

24/ The larger differential possible during inflationary periods alsoincrease the "demand" for such subsidies. Given the politicizednature of the credit allocation process in most developing countries,this dynamic should be recognized as an important social cost ofinflation.

25/ The Banco de la Republica has been accused of price leadership oninterest rates--setting rates which provided a profitable cushion forthe other banks. However, the evidence suggests that if thisoligopolistic behavior did exist, then it was largely eliminated bythe end of restraints on the entry of new financial intermediaries.See P. Spiller and E. Favaro, "The Effect of Entry Regulation onOligopolistic Interaction: The Uruguayan Banking Sector", BellJournal of Economics, 1984.

- 41 -

scaled-down public sector deficits domestically without stimulating infla-tion or burdening the relatively well-functioning financial system withlarger quantities of forced lending.

6.15 Due to the country's long history as an open and relatively pricestable economy, the policy makers in Thailand were nol: forced to undertakemassive reforms. However, the government's overriding policy commitment toa freely convertible currency did require the authorilties to make somepolicy changes to manage developments which threatened convertibility. Inparticular, after roughly five years of increasing protectionism andsharply rising world interest rates, the authorities instituted a set ofreforms, in 1980, which included an upward adjustment in the interest ratestructure.

6.16 Since 1978, Nigeria has been raising its interest rate structureslowly, but rates have remained far below historical inflation. Up until1981, when the countries oil revenues collapsed, the Nigerian authoritiesapparently felt little need to promote the domestic financial sector, asmost of the country's investments were undertaken directly by the govern-ment with the oil earnings. However, since the long term outlook hassoften for the oil sector, the need to stimulate domestic resource mobili-zation, and financial intermediation in general, probably will provide theimpetus for further financial sector reforms in the nLear future.

6.17 In Turkey the financial sector was highly controlled by theauthorities. Among the important government policy instruments were largedirected credit programs and a highly complex set of interest rate con-trols. (A description of this regime is provided in Annex 7, Table 4.) In1980 some these interest rates were liberalized, including term depositrates and the ceiling rates on general credits. In addition, the share oftotal credit committed to the directed credit prograrns was scaled backsomewhat. Moreover, the authorities have announced a policy of continuingto reduce the scope and complexity of the preferential credit schemes.

6.18 Peru initiated a shift in interest rate po:Licy in 1978, whichcontinued through 1982. The shift included higher ceiling rates andallowed compounding on deposits and discounting on loans. These adjust-ments were part of a set of major reforms expressly aimed at giving marketforces a larger role in the economy, which also included an opening of boththe goods and capital markets and the legalization of dollar transactionssuch as deposits and loans. As part of these reforms, the Peruvianauthorities reduced the vast number of controlled interest rates by simpli-fying the preferential lending categories (see Annex 5, Table 4 for adescription of this scheme as of end of year 1982). They also reduced thereserve requirements and began paying interest on them.

6.19 Despite Peru's massive upscaling of its interest rate structure,the adjustments in the ceiling rates lagged behind increases in inflation,so real interest rates generally remained quite negative, with the excep-tion of 1981. While much of the increase in inflation was attributable tothe government's use of the inflation tax to finance its deficit, this taxwas simultaneously being rendered less effective by the decline in reserverequirements. The lower reserve requirements reduced the base of the tax

- 42 -

and thereby increased the inflation resulting from a given increase incentral bank credit.26 / Given this potentially explosive situation andthe lack of progress towards higher real rates, the Peruvian reforms wereof rather limited success.

6.20 Uruguay, like Peru and Turkey, had a history of extremely highinflation, though its experience began much earlier. The combination of astagnating economy, balance of payments problems and high inflationprompted the authorities to begin a liberalization process in 1974 that isalmost without precedent in its dimension. By 1979 virtually all interestrate controls were dismantled, the fiscal deficit was closed, sectoralcredit guidelines and reserve requirements on domestic currency depositswere eliminated, foreign currency transactions were legalized, includingbank deposits and dollar loans. With the important exceptions of theoperations of the publicly-owned Banco de la Republica and the BancoHipotecario del Uruguay and the authorities' control over the exchangerate, control over the financial sector resided primarily in the hands ofthe private sector.

6.21 In sum, the above evidence indicates that in each of the sevensample countries, with the possible exception of Nigeria, there weredefinite moves toward greater market-oriented financial sector policies.These ranged from major reform movements, such as in Uruguay, to morelimited realignments in the structure of nominal interest rates anddirected credit programs. This could well be interpreted as a growingacceptance of the value of financial sector liberalization, in accordancewith the "new" market-oriented perspective. However, it must be emphasizedthe influence of market forces still remains quite constrained by manyfactors, including a wide array of government regulations.

6.22 A final point which the authors wish to strongly emphasize isthat, based on the evidence presented above, financial sector liberaliza-tion is neither a simple matter, nor a painless one. Specifically, thesuccess of the financial sector liberalization process is highly dependenton the mix of domestic fiscal, monetary, exchange, commercial and tradepolicies, particularly when the domestic financial market is open tointernational capital flows.

6.23 On this point, the record shows that both Peru and Uruguay mademajor changes in their financial sector policies in the latter part of theseventies which, among things, opened the financial sector to international

26/ See D. Mathieson and R. McKinnon, "Foreign Exchange and FinancialPolicies for Repressed and Liberalizing Economies," IMF Staff Papers,1983 for a discussion of the difficulties created by cutting reserverequirements when the fiscal deficit is not under control. Thedollarization of the Peruvian economy also reduced the base of theinflation tax.

- 43 -

markets.27 / However, when the exchange rate policy became inconsistentwith domestic monetary, fiscal, and wage policies, a common pattern emergedof either high real interest rates in the local currencies, and/or"dollarization" and capital flight. This experience can be contrasted tothat of Thailand, which also began to experience large capital outflows atthe end of the seventies through its very open financial system. However,Thailand was able to escape the full impact of the crisis which struckPeru, Uruguay and much of the rest of Latin America, because its initialmacro-policy framework was much more consistent, and because it adjustedits monetary, fiscal and trade policies more rapidly to the changing condi-tions. While a full exploration of these two experiences lies beyond thescope of this paper, there is general agreement that one of the cruciallessons of the liberalization process in Latin American was the centralityof a consistent macroeconomic policy package when the capital account is"open." Policy-makers contemplating financial sector liberal:Lzations andproponents of such reforms should take note of this lesson.

27/ A similar opening-up occurred in many of the other major LatinAmerican countries at roughly the same time. While a full discussionof these experiences and of the pros and cons of financial opening-upis beyond the scope of this paper, it is worth noting that theseopening-ups were motivated not only by the poteni:ial benefits of tyingthe local capital market to the international market, but also by thepotential anti-inflationary aspects of the linkagJe. See N. ArditoBarletta, M. Blejer, and L. Landau, Economic Liberalization andStabilization Policies in Argentina, Chile, and lJruguay...", WorldBank, and J. Hanson, and J. DeMelo (1983,1984) for more extensivediscussions of these experiences.

- 45 -

ANNEX 1

SUMMARY TABLES AND COUNTRY TABLES

Summary Tables:

Table 1. Ex-Post Real Return on Term DepositsTable 2. Ex-Post Real Interest Rates on General CreditsTable 3. Ex-Post Real Interest Rates on Preferential CreditsTable 4. Nominal Term Deposit RatesTable 5. Nominal Interest Rates on General CreditsTable 6. Nominal Preferential Lending RatesTable 7. Devaluation Adjusted Yields on U.S. T-BillsTable 8. Devaluation Six Months ForwardTable 9. Inflation Six Months ForwardTable 10. Inflation over Previous Twelve MonthsTable 11. Weighted Average of Real Return on Financial AssetsTable 12. Ratio of Financial Assets to GNP (GDP)

Figure IA Nominal Term Deposit Rates versus Inflation over thePrevious Twelve Months

Country Tables:

Bangladesh Table 1. Selected Interest RatesTable 2. Levels of Selected Financial Assets

Kenya Table 1. Selected Interested RatesTable 2. Levels of Selected Financial Assets

Korea Table 1. Selected Interest RateTable 2. Levels of Selected Financial Assets

Morocco Table 1. Selected Interest RatesTable 2. Levels of Selected Financial Assets

Nigeria Table 1. Selected Interest RatesTable 2. Levels of Selected Financial Assets

Pakistan Table 1. Selected Interest RatesTable 2. Levels of Selected Financial Assets

Peru Table 1. Selected Interest RatesTable 2. Levels of Selected Financial Assets

Thailand Table 1. Selected Interest RatesTable 2. Levels of Selected Financial Assets

Turkey Table 1. Selected Interest RatesTable 2. Levels of Selected Financial Assets

Uruguay Table 1. Selected Interest RatesTable 2. Levels of Selected Financial Assets

SUMMARY: Table 1

Ex-Post Real Interest Rates on Term Deposits (6 month)

End of Year, in percent per annum

Country 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Bangladesh #N/A -18.2% -20.0% -32.2% 17.5% 12.4% -8.5% -0.4% -8.5% -5.8% -2.7% 13.6% 0.3%

Kenya 0.3% 0.0% -5.9% -12.5% -18.0% -6.2% -16.9% -11.1% -0.6% -8.5% -5.2% -3.5% 2.5%

Korea 1.8% -2.0% 5.0% -19.1% -15.8% 0.6% 1.5% -6.4% -8.8% -16.3% -0.4% 7.0% 4.0%

Morocco #N/A #N/A #N/A fN/A -0.1% -4.5% -5.0% -3.5% 1.4% 0.0% -8.1% -0.1% 5.5%

Nigeria -24.8% -3.9% -18.0% -10.1% -40.6% -12.2% -27.2% -15.5% -9.0% -3.5% -20.9% -2.4% -17.3%

Pakistan -0.8% 1.6% -12.7% -5.9% -7.6% 5.4% 5.4% 5.9% 3.0% -3.4% 1.9% 8.4% 2.5%

Peru 1.2% 0.0% -9.0% -15.3% -17.1% -16.1% -21.7% -42.1% -22.3% -13.7% -32.3% -3.4% -36.6%

Thailand 5.3% -1.6% -12.6% -21.6% 3.2% 4.2% -6.1% -4.7% -3.2% -16.7% -6.8% 7.4% 4.9%

Turkey #N/A IN/A IN/A -9.3% -14.3% -6.5% -13.4% -21.2% -38.2% -59.5% -0.4% 13.8% 22.1%

Uruguay -4.5% -41.6% -38.1% -34.9% -17.8% 9.9% -22.1% 7.6% -18.9% 2.2% 14.6% 32.9% 0.5%

Sources: Country Tables in Annex 1.

Notes: Real rates were calculated ex-post, i.e., they reflect the inflation rate 6 months forward from December.

SUMMARY: Table 2

Ex-Post Real Interest Rates on General Credits (short term)

End of Year, in percent per annum

Country 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Bangladesh #N/A -14.1% -15.9% -28.8% 24.7% 19.3% -3.8% 3.8% -4.7% -1.9% -0.2% 16.6% 3.0%

Kenya - - - - - - - -7.2% 3.8% -4.5% -1.4% -0.9% 5.0%

Korea 8.0% 4.5% 11.9% -13.8% -15.4% 2.1% 3.6% -4.6% -7.3% -15.0% 2.2% 9.2% 6.3%

Morocco #N/A #N/A #N/A #N/A 3.7% 1.0% 0.4% 2.0% 5.7% 4.2% -4.7% 3.6% 8.9%

Nigeria -21.1% 2.7% -12.9% -4.4% -36.8% -7.0% -21.8% -13.0Z -3.7% 1.5% -16.8% 2.7% -13.2%

Pakistan 1.9% 4.7% -10.1% -3.4% -5.0% 7.6% 7.4% 8.3% 4.5% -2.1% 3.5% 9.0% 3.5%

Peru 10.0% 8.7% -1.1% -7.9% -9.9% -8.9% -14.5% -36.9% -11.1% 0.2% -21.4% 17.3% -26.6%

Thailand 13.2% 5.9% -6.1% -15.7% 10.9% 12.0% 1.0% 2.4% 4.1% -10.5% 0.0% 14.1% 11.6%

Turkey #N/A #N/A #N/A -5.2% -11.9% -3.8% -10.9% -19.0% -37.7% -60.0% -5.0% 25.1% 34.3%

Uruguay #N/A #N/A #N/A #N/A #N/A #N/A -3.1% 25.5% -2.6% 14.1% 25.9% 44.1% 6.6%

Sources: Country Tables in Annex 1.

Notes: Real rates were calculated ex-post, i.e., they reflect the inflation rate 6 months forward from December.'-' indicates not defined.

SUMMARY: Table 3

Ex-Post Real Interest Rates on Preferential Credits

(Various categories--see Notes), End of Year, in percent per annu

Country 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Bangladesh #N/A -15.3% -17.1% -30.4% 21.9% 16.7% -6.0% 2.4% -5.9% -3.2% -3.6% 12.6% -0.6%

Kenya 4.0% 3.6% -2.5% -10.8% -16.0% -3.9% -14.1% -8.1% 2.8% -5.4% -3.2% -2.6% 1.4%

Korea -7.6% -9.2% 2.7% -20.1Z -20.2% -5.4% -5.2% -11.1% -15.1% -22.1% -2.0% 7.4% 6.3%

Morocco #N/A #N/A #N/A #N/A 1.1% -3.6% -4.1% -2.6% 0.9% -0.5% -9.8% -1.1% 4.1%

Nigeria - - - - - - - - -5.5% -0.3% -18.3% 0.9% -14.3%

Pakistan 0.9% 3.5% -11.5% -3.8% -6.0% 6.5% 7.3% 8.3% 3.9% -2.6% 2.7% 9.2% 3.6%

Peru -1.6% -2.8% -11.6% -17.7% -19.4% -18.5% -24.5% -44.7% -29.1% -20.6% -37.7% -23.6% -52.2%

Thailand 6.3% -0.6% -11.8% -20.9% 3.2% 4.2% -6.1% -4.7% -3.2% -16.7% -9.3% 3.5% 1.2%

Turkey #N/A #N/A #N/A -6.7% -12.7% -4.7% -11.8% -19.7% -38.8% -59.7% -14.4% -9.0% -4.0%

Uruguay #N/A #N/A #N/A #N/A #N/A #N/A #N/A IN/A IN/A IN/A IN/A #N/A #N/A

Sources: Country Tables in Annex 1.

Notes: Real rates were calculated ex-post, i.e., they reflect the inflation rate 6 months forward from December.

a/ Bangladesh: loans against jute, jute goods and tea; Kenya: loans from Ag. Finance Corp.; Korea: exportloans; Morocco: cotton warrants; Nigeria: first class/prefered sector max.; Pakistan: credits fromthe Ag. Devel. Bank; Peru: rediscounts to the Banco Agrario; Thailand: export credit; Turkey: ag. credits;Uruguay: not available.'-' indicates not defined.

SUMTARY: Table 4

Nominal Term Deposit Rates (6 month)

End of Year, in percent per annum

Country 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Bangladesh #N/A 4.8% 4.8% 4.8% 6.5% 6.5X 7.5% 7.5% 7.5% 7.5% 13.0% 13.0% 13.0%

Kenya 3.8% 3.8% 3.8% 5.4% 5.4% 5.4% 5.'4% 5.4% 5.4% 5.4% 6.8% 11.0% 13.2%

Korea 16.8% 14.4% 8.4% 8.4% 15.0% 13.8% 15.6% 13.8% 17.1% 17.1% 16.9% 14.6% 7.6%

Morocco #N/A #N/A IN/A #N/A 4.0% 4.5% 4.5% 4.5% 6.0% 6.0% 7.5% 7.5% 8.5%

Nigeria 3.0% 3.0% 3.5% 3.5% 3.5% 3.0% 2.5% 3.0% 5.0% 5.5% 6.0% 6.0% 8.5%

Pakistan 5.2% 5.1% 5.6% 6.7% 8.2% 8.9% 9.1% 9.6% 10.0% 10.1% 10.2% 10.2% 9.9%

Peru 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 11.0% 14.0% 31.5% 31.5% 31.5% 63.0% 71.2%

Thailand 6.0% 6.0% 6.0% 6.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 10.0% 11.0% 11.0%

Turkey #N/A #N/A IN/A 4.0% 6.0% 6.0% 6.0% 6.0% 9.0% 12.0% 32.0% 50.0% 50.0%

Uruguay 15.0% 15.0% 15.0% 18.0% 30.0% 30.0% 30.2% 51.4% 42.6% 50.6% 50.3% 47.4% 66.2%

Sources: Country Tables in Annex 1.

SUMMARY: Table 5

Nominal tnterest Rates on General Credits (short term)

End of Year, in percent per annum

Country 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Bangladesh #N/A 10.0% 10.0% 10.0% 13.0% 13.0% 13.0% 12.0% 12.0% 12.0% 16.0% 16.0% 16.0%

Kenya - - - - - - - 10.0% 10.0% 10.0% 11.0% 14.0% 16.0%

Korea 24.0% 22.0% 15.5% 15.5% 15.5% 15.5% 18.0% 16.0% 19.0% 19.0% 20.0% 17.0% 10.0%

Morocco #N/A #N/A #N/A #N/A 8.0% 10.5% 10.5% 10.5% 10.5% 10.5% 11.5% 11.5% 12.0%

Nigeria 8.0% 10.0% 10.0% 10.0% 10.0% 9.0% 10.0% 6.0% 11.0% 11.0% 11.5% 11.5% 14.0%

Pakistan 8.1% 8.3% 8.7% 9.4% 11.2% 11.1% 11.2% 12.1% 11.7% 11.6% 11.9% 10.9% 11.0%

Peru 16.3% 16.3% 16.3% 16.3% 16.3% 16.3% 21.2% 24.2% 50.4% 52.7% 52.7% 98.0% 98.0%

Thailand 14.0% 14.0% 14.0% 14.0% 15.0% 15.0% 15.0% 15.0% 15.0% 15.0% 18.0% 18.0% 18.0%

Turkey #N/A #N/A #N/A 8.8% 9.0% 9.07. 9.0% 9.0% 10.0% 10.8% 26.0% 65.0% 65.0%

Uruguay #N/A #N/A #N/A #N/A #N/A #N/A 62.0% 76.6% 71.2% 68.1% 65.1% 59.8% 76.3%

Sources: Country Tables in Annex 1.

Notes: '-' indicates not defined.

SUMARY: Table 6

Nominal Preferential Lending Rates

(Various categories-see Notes), End of Year, in percent per annun

Country 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Bangladesh IN/A 8.5% 8.5% 7.5% 10.5% 10.5% 10.5% 10.5% 10.5% 10.5% 12.0% 12.0% 12.0%

Kenya 7.5% 7.5% 7.5% 7.5% 8.0% 8.0% 9.0% 9.0% 9.0% 9.0% 9.0% 12.0% 12.0%

Korea 6.0% 6.0% 6.0% 7.0% 9.0% 7.0% 8.0% 8.0% 9.0% 9.0% 15.0% 15.0% 10.0%

Morocco IN/A IN/A IN/A IN/A 5.3% 5.5% 5.5% 5.5% 5.5% 5.5% 5.5% 6.5% 7.0%

Nigeria - - - - - - - - 9.0% 9.0% 9.5% 9.5% 12.5%un

Pakistan 7.0% 7.0% 7.0% 9.0% 10.0% 10.0% 11.0% 12.0% 11.0% 11.0% 11.0% 11.0% 11.0%

Peru 4.0% 4.0% 4.0% 4.0% 4.0% 4.0% 7.0% 9.0% 20.0% 21.0% 21.0% 29.0% 29.0%

Thailand 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0%

Turkey IN/A fN/A #N/A 7.0% 8.0% 8.0% 8.0% 8.0% 8.0% 11.5% 13.5% 20.0% 18.0%

Uruguay IN/A IN/A #N/A IN/A IN/A #N/A IN/A #N/A tN/A IN/A IN/A IN/A #N/A

Sources: Cbuntry Tables in Annex 1.

Notes: Bangladesh: loans against jute, jute goods and tea; Kenya: loans from Ag. Finance Corp.; Korea: export

loans; Norocco: cotton warrants; Nigeria: first class/prefered sector max.; Pakistan: credits from

the Ag. Devel. Bank; Peru: rediscounts to the Banco Agrario; Thailand: export credit; Turkey: ag. credits;

Uruguay: not available.'-' indicates not defined.

SUMMARY: Table 7

Yields on Six Mbnth U.S. Treasury Bills, Adjusted for Ex-Post Devaluation

End of Year, in percent per annum

Oountry 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

U.S. 6 Mo. T-bill 4.9% 4.0% 5.3% 7.5% 7.1% 5.5% 4.5% 6.5% 9.5% 11.9% 13.7% 12.3% 8.1%

Bangladesh #N/A 13.5% -12.9% 1.5% 205.9% 7.3% 12.6% 16.6% 12.3% -0.9% 41.1% 39.4% 11.9%

Kenya 4.9% 4.0% -1.7% 15.2% 7.1% 9.9% 4.3% 2.8%. 11.6% 11.3% 53.8% 27.8% 17.1%

Korea 43.8% 19.3% 5.3% 8.3% 7.1% 5.5% 4.5% 6.5% 9.5% 73.6% 22.6% 25.5% 16.3%

Morocco 4.8% -4.6% -26.8% 10.7% -7.3% 18.6% 4.4% 0.8% 10.5% 15.5% 79.9% 50.5% 29.4%

Nigeria 4.9% 4.0% 5.3% -5.4% 3.3% 5.5% 10.7% 3.8% -6.9% 4.7% 60.3% 26.3% 31.6% u.

Pakistan 2.6% 451.0% -15.2% 7.5% 7.1% 5.5% 4.5% 6.5% 9.5% 11.9% 13.7% 70.5% 13.7% 1

Peru 4.9% 4.0% 5.3% 7.5% 7.1% 120.1% 38.6% 49.7% 44.2% 45.2% 71.5% 100.2% 177.5%

Thailand 4.9% 4.0% 5.3% 7.5% 7.1% 5.5% 4.5% 6.5% 9.9% 11.6% 17.8% 12.3% 8.1%

Turkey 185.9% -6.5% 5.3% -0.2% 9.4% 20.0% 17.6% 79.6% 114.5% 455.5% 71.5% 72.5% 52.1%

Uruguay 4.9% 148.6% 51.2% 56.3% 112.0% 54.1% 41.2% 27.9% 37.9% 26.5% 30.3% 30.5% -1.0%

Sources: U.S. 6 month T-bill rate data from the U.S. Federal Reserve Bank, Bulletin.Exchange rate data from the TMF, IFS, data tape.

Notes: The country rates are the U.S. T-bill rates adjusted for the country's ex-post rate of devaluation from

December to the following June.

SUMMARY: Table 8

Devaluation Six Months Forward

End of Year, in percent per annum

Country 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Bangladesh #N/A 9.1% -17.3% -5.6% 185.6% 1.7% 7.8% 9.5% 2.6% -11.4% 24.0% 24.1% 3.6%

Kenya 0.0% 0.0% -6.7% 7.2% 0.0% 4.2% -0.2% -3.5% 2.0% -0.5% 35.2% 13.9% 8.3%

Korea 37.1% 14.7% 0.0% 0.8% 0.0% 0.0% 0.0% 0.0% 0.0% 55.2% 7.8% 11.8% 7.6%

Morocco -0.1% -8.3% -30.6% 3.0% -13.5% 12.4% -0.1% -5.4% 0.9% 3.2% 58.1% 34.1% 19.8%

Nigeria 0.0% 0.0% 0.0% -12.0% -3.6% 0.0% 5.9% -2.5% -15.0% -6.4% 40.9% 12.5% 21.8%

Pakistan -2.2% 429.7% -19.5% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 51.9% 5.2%

Peru 0.0% 0.0% 0.0% 0.0% 0.0% 108.6% 32.6% 40.6% 31.7% 29.8% 50.8% 78.3% 156.8%

Thailand 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.4% -0.2% 3.6% 0.0% 0.0%

Turkey 172.6% -10.1% 0.0% -7.1% 2.2% 13.8% 12.5% 68.7% 96.0% 396.7% 50.8% 53.7% 40.7%

Uruguay 0.0% 139.0% 43.5% 45.4% 98.0% 46.1% 35.1% 20.1% 26.0% 13.1% 14.6% 16.2% -8.4%

Source: The IMF, IFS, data tape.

SUMMARY: Table 9

Inflation Six Months Forward

End of Year, in percent per anmun

Country 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Bangladesh -4.1% 28.0% 30.9% 54.6% -9.4% -5.3% 17.5% 7.9% 17.5% 14.2% 16.2% -0.6% 12.6%

Kenya 3.4% 3.8% 10.3% 20.5% 28.5% 12.4% 26.8% 18.6% 6.0% 15.2% 12.6% 15.0% 10.5%

Korea 14.8% 16.7% 3.3% 33.9% 36.5% 13.1% 13.9% 21.6% 28.4% 39.9% 17.4% 7.1% 3.5%

Morocco 3.6% 0.0% -2.9% 14.9% 4.1% 9.4% 10.1% 8.3% 4.5% 6.0% 17.0% 7.7% 2.8%

Nigeria 37.0% 7.1% 26.3% 15.1% 74.2% 17.2% 40.7% 21.8% 15.3% 9.3% 34.1% 8.6% 31.3%

Pakistan 6.0% 3.4% 20.9% 13.3% 17.1% 3.2% 3.5% 3.4% 6.8% 13.9% 8.1% 1.7% 7.2%

Peru 5.7% 7.0% 17.6% 26.3% 29.1% 27.6% 41.8% 97.0% 69.1% 52.3% 94.3% 68.8% 169.9%

Thailand 0.7% 7.7% 21.3% 35.2% 3.7% 2.6% 13.9% 12.3% 10.5% 28.4% 18.0% 3.4% 5.8%

Turkey 20.3% 6.4% 22.6% 14.7% 23.7% 13.3% 22.4% 34.5% 76.5% 176.8% 32.6% 31.9% 22.9%

Uruguay 20.4% 97.0% 85.8% 81.2% 58.2% 18.3% 67.2% 40.7% 75.8% 47.4% 31.2% 10.9% 65.3%

Source: CPI data from the IMF, IFS, data tape.

SUMKARY: Table 10

-Inflation over Previous Twelve Months

End of Year, in percent per annum

Country 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Bangladesh #N/A 15.3% 46.1% 34.2% 76.1% -12.6% -0.1% 17.1% 9.6% 14.0% 13.2% 14.3% 4.9%

Kenya 1.5% 7.2% 3.3% 15.2% 16.0% 20.3% 7.-6% 21.0% 13.7% 9.1% 13.1% 19.3% 13.3%

Korea 10.5% 12.3% 7.8% 8.5% 26.5% 25.4% 10.5% 11.0% 16.4% 21.2% 34.6% 11.7% 4.8%

Morocco 2.6% 4.7% 2.6% 8.7% 14.4% 6.1% 13.4% 9.0% 9.7% 9.0% 9.7% 13.2% 6.7%

Nigeria 13.0% 15.0% -3.5% 17.8% 9.9% 43.1% 12.4% 31.3% 10.3% 11.5% 13.7% 17.4% 6.7%

Pakistan 5.2% 5.8% 7.8% 37.8% 20.6% 12.7% 10.3% 7.3% 5.5% 9.0% 15.1% 10.4% 3.8%

Peru 5.8% 7.6% 4.4% 13.8% 19.1% 23.9% 44.7% 32.5% 73.7% 65.8% 60.8% 72.7% 65.6%

Thailand -1.3% 1.3% 8.8% 20.2% 17.9% 4.4% 3.4% 8.9% 7.8% 15.0% 16.4% 12.3% 2.6%

Turkey 12.3% 17.6% 9.2% 16.8% 16.7% 19.6% 17.0% 44.6z 36.6% 81.1% 86.2% 3u.3a 32.8%

Uruguay 20.1% 35.2% 95.6% 77.6% 106.8% 67.3% 39.4% 57.6% 45.8% 83.1% 42.9% 29.3% 20.5%

Source: CPI data from the IMF, IFS, data tape.

SUMMARY: Table 11

Weighted Average of Real Return on Financial Assets

End of Year, in percent per annum

Country 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Bangladesh #N/A -20.8% -22.6% -34.4% 12.6% 7.9% -12.4% -4.4% -12.3% -9.5% -8.5% 7.2% -5.0%

Kenya -1.4% -2.7% -6.6% -14.6% -18.7% -9.7% -16 5% -13.8% -4.0% -10.0% -7.6% -8.2% -2.8%

Korea -2.1% -4.7% 2.6% -20.8% -20.6% -4.3% -4.2% -10.8% -14.8% -20.7% -5.8% 2.2% 1.7%

Morocco #N/A #N/A #N/A IN/A -3.6% -8.2% -8.7% -7.2% -3.6% -4.8% -13.47. -5.9% -1.2%

Nigeria #N/A -5.6% -19.8% -11.9% -41.7% -13.5% -28.1% -17.0% -11.8% -6.7% -23.8% -5.8% -21.2%

Pakistan -3.8% -1.3% -15.2% -9.2% -12.0% 0.4% 0.4% 0.8% -2.2% -8.5% -3.6% 2.7% -2.4%

Peru -3.1% -4.0% -12.7% -18.7% -20.6% -19.1% -26.4% -45.8% -31.4% -21.7% -32.6% -12.0% -30.0%

Thailand 2.4% -4.1% -14.8% -23.5% 0.5% 1.8% -8.2% -6.8% -5.3% -18.5% -9.47% 4.5% 2.4%

Turkey IN/A #N/A #N/A -11.2% -16.9% -9.4% -16.2% -23.9% -42.1% -62.9% -18.8% -8.2% 1.2%

Uruguay IN/A #N/A -43.7% -40.5% -21.2% 0.5% -27.4% -13.6% -28.9% -17.0% -4.2% 13.3% -33.4%

Sources: COuntry Tables in Annex 1.

Notes: The average return was calculated by weighting the ex-post real return on each instrument by its share oftotal financial assets.

SLUMARY: Table 12

Ratio of Total Financial Assets to GNP (GDP)

End of Year

Country 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Bangladesh #N/A #N/A 18.9% 15.0% 9.9% 12.4% 15.8% 15.7% 16.0% 16.9% 17.0% 17.7% 19.5%

Kenya 28.9% 28.5% 29.8% 31.4% 28.6% 28.5% 27.3% 33.4% 34.7% 34.8% 33.4% 33.2% 32.7%

Korea 30.3% 30.4% 32.4% 32.0% 28.5% 26.9% 26.6% 29.2% 32.5% 33.1% 34.5% 38.9% 43.1%

Morocco 29.0% 30.1% 32.6% 30.9% 32.7% 35.5% 36.1% 37.3% 40.6% 41.2% 41.5% 42.8% 42.6ZU,

Nigeria a/ #N/A 13.7% 13.9% 11.4% 11.6% 15.9% 19.8% 24.9% 21.8% 23.7Z 33.3% 35.2% 36.7%

Pakistan 41.3% 40.4% 37.6% 36.4% 30.2% 29.2% 31.8% 31.8% 33.9% 34.6% 33.9% 32.7% 34.4%

Peru 21.0% 21.6% 22.1% 21.5% 21.3% 19.3% 16.5% 14.7% 14.3% 16.6% 19.4% 20.5% 20.2%

Thailand 29.4% 31.3% 31.7% 30.2% 31.3% 33.0% 34.3% 35.1% 35.5% 33.7% 34.9% 36.7% 41.9%

Turkey 26.1% 26.2% 25.9% 24.8% 23.6% 24.3% 23.6% 22.9% 19.4% 16.8% 16.3% 21.6% 25.5%

Uruguay a #N/A #N/A 20.6% 14.9% 15.1% 16.0% 20.8% 22.7% 25.4% 27.6% 31.7% 40.7% 59.8%

Sources: Country Tables in Annex 1.

a/ For Nigeria and Uruguay the ratio is with repect to GDP, in all other cases it is with repect to GNP.The GNP and GDP figures have been logrithmically interpolated between the current year and one year forwardso as to be expressed in December prices.

- 58 -

Fiaure 1A

Nominal Term Deposit Rate versus Inflation

- NOMINAL INTEREST RATES ON TERM DEPOSITS (6 MONTH)

..... INFLATION RATE CPAST DEC. TO DEC. OF THE CURRENT YEAR)

46-_

x~~~~~~~~

2S

48-

30:

Oli ......

K~~K-Y

40.

38-

X

25- KOR.'

-10i I6- I I T

1-o-72 17 97 96ise16

- 59 -

Figure IA (continued)

NOMINAL INTEREST RATES ON TERM DEPOSITS C0 MONTH)

.-- INFLATION RATE CPAST DEC. TO DEC. oF THE CJRRENT YEAR)

3.9-

25-230,

- ...... ... .... *" ,.

-5- MOROCCO

-Ie-

3- U. . * _ S

3 S

Is U

NIGERIk

40-

30 - .' "

20-

_197 17 1 I I I 1 I 98 1

1970 1972 1974 1978 1978 19801 1982

- 60 -

Figure 1A (continued)

NOI.NAL INTEREST RATES ON Tr DEPOSITS Co MONTH)

.. Z -- FLATION RATE CPAST DEC. TO DEC. Of THE CURRENT YEAR)

10-

88-

80-

PERU

20

.4380

800

240- .' ' .

20- , -'

8 TURKEY

-0-2

1970 1972 1974 19715 1978 1980 1982

- 61 -

Figure 1A (continued)

NOMINAL INTEREST RATES ON TERM DEPOSIT'; CO MONTH)

.l- INFLATION RATE CPAST DEC. TO DEC. OF THE CURRENT YEAR)

100- ****'"t*. s*, ,, u""

80-

0- URUGUAY

-20-

1970 1972 1974 1978 1978 1988 1982

- 62 -

BA1UALES~: Table I

Selected Nminal and E-post Real Interest Rates In Sdveduled Bdks

Erd of Year (amalized)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Deposits, darnstic denamn.

DBMdNozLnal N/A O.04 o.04 0.0% 0 .1 0. .a O.% 041% O.CZ% o.0% 0.0% 0.0% 0.0%Ex-past real #NA -0.2 -0.2 -0.4 0.1 0.1 -0.1 -0.1 -0.1 -0.1 -0.1 0.0 -0.1

Savings (n rn-clidg)Naomnal IN/A 4.5% 4.5% 4.5X 6.0% 6.4 7.C% 7.C% 7.C1 7.0% 1OX1 10.41% 10.0OEx-post real IN/A -18.4% -20.1% -32.4% 16.9% 11.9% -8.9% -0.8% -8.9% -6.3% -5.3% 10.6% -2.3%

Temr, 6 mnthNominal #N/A 4.&C 4.8% 4.8t 6.5% 6.5% 7.5% 7.5% 7.5Z 7.5t 134C 13.0% 13.0%Ex-post real #WA -18.2% -20.0% -32.2% 17.5% 12.4% -8.5% -0.4% -8.5% -51X% -2.7% 13.6% 0.3%

Weighted average rate ofreturn an above assetsplus aurry

Ex-post real #WA -20.8% -22.6% -34.4% 12.6% 7.9% -12.4% -4.4% -12.3% -9.5% -8.5% 7.2% -5.0%

lndiax, damestic denam.

NDnnalN&minal IN/A 10.0% 1O.OX 10.0% 13.0% 13. 13.C% 12.0% 12.0% 12.0% 16.0% 16.0% 16.0%

E--post real IA -14.1% -15.9% -28.8% 24.7% 19.3% -3.8% 3.8% -4.7% -1.9% -0.2% 16.6% 3.0%

Against jute, jute gDodsard tea

Noinal IN/A 8.5% 8.5% 7.5% 10.5% 10.5% 10.5% 10.5% 10.5% 10.5% 12.0% 12.0% 12.0%Ex-post real #N/A -15.3X -17.1% -30.64 21.9% 16.7% -6.0% 2.4% -5.9% -3.2% -3.6% 12.6% -4.6%

Inflatimn (CPI) 6 nmthsForward frao December -4.1% 28.C% 30.9% 54.6% -9.4% -5.3% 17.5% 7.92 17.5% 14.2% 16.2% -0.61 12.6%

Devaluatimx 6 mnthsForward fran Declrber #N/A 9.1% -17.3% -5.6% 185.6% 1.7% 7.8% 9.5% 2.6% -11.4% 24.0% 24.1% 3.6%

Inflatian (CPI) overPrior December #N/A 15.3% 46.1% 34.2% 76.1% -12.6% -0.1% 17.1% 9.6% 14.0% 13.2% 14.3% 4.9%

Scurces: Interest rates are fran the Berladesh Bark, Ecnomic Trends. CPI data are fran the Iff, IFS, data tape.Weights for wighted average are fran BargLadesh Table 2. (See note regardiig disposition of savings.)

Notes: Real rates wre calailated ex-post, i.e., they reflect the inflation rate six nonths forvard fram Deceier. In 1977, 1978anl 1979, premiums were paid to ural depositors of .75 & 1.5 percentage points an savirgs and tenn deposits, respectively.

- 63 -

P&GLALESH: Table 2

Levels of Selected Financial Assets

End of Year(in biili.ic. of Taka)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

xerrcy, dmesticNordnal IN/A 2.1 2.9 3.2 4.0 3.7 3.9 5.0 6.6 7.2 8.7 9.7 9.9In 1980 prices #N/A 10.3 9.9 8.2 5.8 6.2 6.4 7.1 8.5 8.2 8.7 8.5 8.3% of GNP #tN/A [EN/A 5.9% 4.2X 2.9% 3.3X 3.5Z 3.7% 4.0% 3.9% 3.9% 3.8% 3.6%% of total assets #N/A 37.8% 31.0Z 27.8% 29.4% 26.6% 22.0% 23.3Z 25.2% 22.9% 23.1% 21.7% 18.7%

Deposits, doxmestic currency

DedNotdnal IN/A 1.8 3.9 4.9 5.4 5.5 6.8 7.5 8.8 10.2 10.6 12.3 14.8In 1980 prices N/A 9.0 13.3 12.4 7.7 9.2 11.3 10.6 11.3 11.6 10.6 10.8 12.4% of GNP #N/A #N/A 7.9% 6.3X 3.9% 4.97 6.1X 5.4% 5.4% 5.4% 4.8% 4.9% 5.4%% of total assets #N/A 33.17. 41.7% 42.37. 39.1% 39.5X 38.87 34.77. 33.5% 32.3% 28.4% 27.5% 27.9%

TermNoidnal #N/A 1.6 2.5 3.4 4.3 4.8 6.9 9.1 10.8 14.2 18.2 22.8 28.5In 1980 prices IN/A 7.9 8.7 8.8 6.2 7.9 11.4 12.9 13.9 16.0 18.2 19.9 23.7% of 1P AN/A. #N/A 5.2X 4.57 3.1Z 4.27. 6.2% 6.6% 6.6% 7.6% 8.2Z 9.0% 10.4%% of total assets 1tN/A 29.0% 27.3% 29.9% 31.5% 33.9% 39.2% 42.1% 41.2% 44.8% 48.5% 50.8% 53.5%

TotalNinal #N/A 5.5 9.3 11.5 13.7 14.0 17.6 21.6 26.2 31.7 37.5 44.9 53.2

In 1980 prices WN/A 27.3 31.9 29.3 19.8 23.2 29.1 30.6 33.8 33.8 37.5 39.3 44.4% of GMP #N/A #N/A 18.9% 15.0% 9.9% 12.4% 15.87 15.7% 16.0% 16.9% 17.07 17.7% 19.57

CPI, December (Dec.80-100) 17.4 20.0 29.3 39.3 69.2 60.5 60.4 70.8 77.5 88.4 100.0 114.3 119.9

GNP, Dec., n.rnal ItN/A #N/A 49.3 76.8 137.9 112.8 111.3 138.0 .63.9 187.8 221.0 253.9 273.3

Sources: Asset data are from the Bangladesh Bank, Bangladesh Barn Bulletin, and Econcioac Trends. GNP data are frcnBangladesh Bureau of Statistics, Fconaiic Irdicators of Bargladesh, and from the Bangladesh Bark, Econrsinc Trends.CPI data are fran the INF, IFS, data tape.

Notes: GNP data are for the fiscal year begirnnig in the current year, therefore incone is expressed in Defcanber prices.

- 64 -

KENL: Tdhle I

Salected -a-1 ai bc-post 1 al sntearat Rite

End of Yer (annalized)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Deaprits. uinin- rate

N UK.OX O.X 0.0% 0.OX O.OX 0.0% O.OX 0.0% O.OX O.X% 0.0% .OXZ 0.0%

fl-pwt ial -3.31 -3.72 -9.12 -17.0% -22.n -11.0 -21.Z -15.7% -5.7% -13.22 -11.22 -13.0% -9.52

N.0x3.0x3.0% 3.0% . M.0 5.0 5.0% 5.0 5.0% 5.0% 5.0% 6.x% 10.0% 12.52ftP eal -0.47 -0.8x -6.6x -14.5x -18.3x -6.6x -17.Z1-11.5x -0Xx.9x -s.6.X -58 4.3t 1.9x

Trem, 6 unthNoodnal 3.82 3.R2 3.8x 5.4% 5.42 5.4% 5.42 5.42 5.42 5.42 6.82 11.0% 13.2Xfl-Vost real 0.3X 0.0% -5.92 -12.52 -18.0% -6.2 -16.92 -11.11 -0.62 -. 52 -5.22 -3.52 2.52

Privae Flxiial Institutions

Nmal O.O% O.X% 0.0% O.OX 0.0% 0.0% 0.0% 0.0% 0.0% 0.0 % M0.0M 0.0%nx-poat real -3.31 -3.72 -9.3 -17.0% -22. Z -11.0% -21.2x -15.7% -5.7% -13.2n -11.2x -13.0% -9.5s

SavinipNoadml 3.0% 3.0% 3.0 t 3.0% 5.0% 5.0% 5.0t 5.0% 5.0t 5.0% 8.0% 10.0% 12.52ft-post real -0.4% -0.8 -6.61 -14.57 -18.3 -6.61 -17.1 -11.52 -0.92 -8.92 -4.0% -4.3 1.9%

Term, l_Wat cateqoryz-j)at real 2.52 2.1 L -3.9% -10.82 -16.4 -4.42 -14.9% -8.3 2.61 -5.6Z -1.4% -0.4% 5.1

Post offc SffIo8 anNomiral 3.0% 3.0% 3.0% 3.0% 3.0% 5.0% 5.0% 5.0% 5.0% 5.o 6.0% 10.0% 10.02&C-Post real -0.4 -0.82 -6.61 -14.5 -19.9% -6.61 -17.2 -11.52 -0.9% -8.9% -5.8s -4.3 -0.4t

W'gbed ! rte of

retur=an abw s amts

pl curec

Ei-pat ral -1.4Z -2.7% -6.61 -14.61 -18.7% -9.7% -16.2 -13.8 -4.0% -10.0% -7.6% s.2n -2.82

L, drg, de n.

Mmdmm (less than 3 yr.)N l - - - - - - - 10.02 10.02 10.0% 11.0% 14.0216.0%

- et - - 7.2A 3.8% --4.5% -1.4% -0.9% 5.0o

tkadrul 7.0% 7.0% 7.0% 7.0% 8.0 8.0% - - - -s

lt-post real 3.5% 3.U.2 -3.0% -11.22 -16.0% -3.9% - - - -

Nintal-E- Jlnece Corp 7.5% 7.5Z 7.52 7.5% 8.t1 8.0% 9.02 9.' 9.C7 9.C% 9.0% 12.0% 12.0%E&c-t real 4.0% 3.6% -2.5% -10.8% -16.0% -3.9% -14.U% -8.U% 2.8% -5.4% -3.22 -2.6% 1.4X

Inflation (CPI) 6 mthsForard frcm Decmbr 3.4% 3.8% 10.31 20.5% 28.5X 12.4% 26.8% 18.6% 6.0% 15.1 12.6% 15.0% 10.5%

rvalu3atimn 6 nmihatForward frm a 0.0% O.Z -6.7% 7.2Z 0.0% 4.22 -0.22 -3.5% 2.0% -0.5% 35.22 13.92 8.3Z

Inflatijc (CPI) overPrior lcmr 1.5% 7.2% 3.12 15.22 16.'J2 20.3X 7.6Z 21.0% 13.7% 9.t2 13.1% 19.3% 13.3%

SWces: Interest rates are froa T1e Central Bark at Kenya, Foc and 781nan1 Review, k aal Report, mid &real ofSatistics FiLrxce ard Plannirg, Enmic &irvey. CPI datE are frao the IMF, IFS, data tape. Weights for wg$iteiaverage are ftrm Kenya Table 2.

NDotes: Real rates wre calculated ex-post, i.e., they reflect &te inflatimn rate six smths forward frar Decemr. Ihe averagerate of tetur s calaolated asumrg each asset yielded the ainimn rate and that aLl term depcsitS in the PrivateFinadcal instiutlcxu were In the laget tens cat ory. '-' indicates not defined.

- 65 -

l2YA: Table 2

Levels of Selected Financial Assets

Fzd of Yrear(in biLiUal of 9illings)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Currency, dampsticNcoinaLe 0.7 0.7 0.9 1.0 1.1 1.2 1.6 2.2 2.3 2.7 3.0 3.6 3.7In 1980 prices 2.3 2.2 2.6 2.5 2.4 2.3 2.8 3.1 2.8 3.0 3.0 3.0 2.8X of (NP 5.8X 5.6% 6.0% 5.6% 5.2x. 4.9% 5.2% 5.8% 5.6% 5.7X 5.5% 5.72 5.2%% of total assets 2).7% 19.5% 20.1% 17.8% 18.1% 17.3% 18.9% 17.5% 16.0% 16.3X 16.6% 17.3% 15.8%

Deposits, dczestic currency

Camercial Bank

DeaMnNouinal 1.2 1.3 1.5 2.0 2.1 2.4 3.0 4.3 4.7 5.6 5.2 5.6 6.0In 1980 prices 3.9 3.9 4.5 5.0 4.7 4.4 5.1 6.1 5.8 6.3 5.2 4.7 4.4% of af 9.9% 9.7% 10.2% 11.Z% 10.L% 9.7% 9.5% 11.6% 11.4% 11.8% 9.6X 9.0% 8.32% of total assets 34.4% 34.0% 34.1% 35.5% 35.4% 34.0% 34.7% 34.8% 32.8% 33.8% 28.6% 27.1% 25.5%

SavinNmLnal 0.7 0.8 0.9 1.0 1.2 1.3 1.5 2.1 2.4 2.7 3.0 3.5 3.9In 1980 pries 2.2 2.3 2.5 2.6 2.6 2.4 2.6 2.9 3.0 3.0 3.0 2.9 2.9% of GP 5.7% 5.7% 5.8% 5.77 5.6% 5.7Z 4.9% 5.6% 5.8% 5.6% 5.4% 5.6% 5.5%% of total assets 19.9% 19.9% 19.4% 18.7% 19.6% 18.37 17.8Z 16.8% 16.6% 16.7% 16.Z% 16.9% 16.7%

TermNokrinal 0.5 0.6 0.6 0.9 0.8 1.1 1.4 2.3 3.0 3.1 3.7 4.1 5.6In 1980 prices 1.7 1.8 1.8 2.2 1.8 2.0 2.3 3.3 3.6 3.5 3.7 3.5 4.2% of (N' 4.5% 4.4% 4.2% 4.9% 3.9% 4.5% 4.4% 6.2X 7.1% 6.5% 6.8% 6.6% 7.8%% of total assets 15.4% 15.5% 14.0% 15.7% 13.7% 15.7% 16.0X 18.6% 20.5% 18.6% 20.4% 20.0% 24.0%

Private Finarnial Inst.

DadNkinal 0.0 0.0 0.0 0.0 0.0 0.1 0.0 01.3 0.2 0.3 0.1 0.2 0.2In 1980 prices 0.1 0.1 0.0 0.0 0.0 0.2 0.0 0.5 0.2 0.3 0.1 0.2 0.2% of GNP 0.3X 0.7% 0.1% 0.1% 0.0% 0.4% 0.LX :x.9% 0.4% 0.5% 0.2% 0.4X 0.3%% of total assets 1.L% 0.6% 0.4% 0.3% 0.1% 1.3% 0.3% 2.8% 1.3% 1.6% 0.5% 1.1% 1.0%

Savings a/NomLnaT 0.1 0.1 0.1 0.2 0.2 0.1 0.3 0.2 0.5 0.5 0.6 0.6 0.6% of GNP 0.6% 0.8% 0.9% 1.1% 1.0% 0.6% 0.9% 0.6% 1.1% 1.0% 1.1% 0.9% 0.9%% of total asses 2.0% 2.6% 3.1% 3.5% 3.5% 2.1% 3.2% 1.9% 3.3% 2.7% 3.2% 2.8% 2.7%

Term a/kio*inal 0.1 0.2 0.3 0.4 0.4 0.7 0.6 0.8 1.2 1.5 2.3 2.6 2.9In 1980 prices 0.4 0.6 0.8 1.0 1.0 1.2 1.0 1.1 1.4 1.7 2.3 2.2 2.1% of GN4P 1.1% 1.4% 1.9% 2.1% 2.1% 2.7% 2.0% 2.1% 2.8% 3.2% 4.2% 4.2% 4.0%% of total assets 3.8% 5.0% 6.2X 6.8% 7._7 9.3% 7.7% 6.37 8.0% 9.3% 12.7% 12.7% 12.7%

Post (fice Savins BankNordt,al 0.1 0.1 0.1 0.1 0.1 0.1 0.2 0.2 0.2 0.3 0.3 0.4 0.5In 1980 prices 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0-3 0.4 0.4% of IN' 0.9% 0.8% 0.8% 0.7% 0.7% 0.6% 0.5% 0.5% 0.5% 0.5% 0.6% 0.7% 0.7%% of total assets 3.L% 2.9% 2.6% 2.3% 2.X% 2.0X 1.9% 1.5% 1.5% 1.5% 1.9% 2.t% 2.L%

TotalNMcinal 3.4 3.8 4.4 5.5 6.0 7.1 8.6 12.5 14.4 16.4 18.3 20.6 23.6En 1980 prices 11.2 11.5 13.0 14.1 13.2 13.0 14.6 17.5 17.8 18.6 18.3 17.3 17.4% of IN' 28.9% 28.5% 29.8% 31.4% 28.6% 28.5% 27.3% 33.4% 34.7% 34.8% 33.4% 33.D. 32.7%

CPI, Dti17ber (Dec.80-100) 30.8 31.0 34.1 39.3 45.5 54.8 58.9 71.3 81.0 88.4 100.0 119.3 135.2

GNP, Dec. adjusced, nirnal 11.9 13.3 14.9 17.6 21.0 25.0 31.4 37.4 41.5 47.2 54.8 62.2 72.1

Sorces: Asset data are fr-c Thne Central Bank of Kenya, Ectnomic ad Firnancial Review.GNP and CPI data are tram tte IIF, LFS, data tape.

Notes: The GNP data have been logrittincally interpolated betuee the current year and one year fo ward so as to he expressed inDeeasber prices.

a/ Savirn and term deposits in the Private Financial Instiutions were assured to be distribu:ed 1:2 for tths years 1970-72,and in aLl other years as reported.

-66- KOA__On _

Selected N rl ar l-post i lIwtert bkt

B31 Of Year (animlzed)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 198a 1981 1932

Deposits, dmestic derx.

DodChddng

Nkulnal 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.O%Ex-pet real -12.9% -14.3t -3.2% -25.13 -26.7% -11.6% -12.2% -17.7% -22.1% -28.5% -14.8% -6.6% -3.4%

PassbookNominal 1.8% 1.8% 1.8% 1.8% 1.8% 1.8% 1.8% 1.8% 1.8% 1.B% 1.8% 1.8% 1.8%E3-l"t reel -11.32 -12.8% -1.4% -24.0S -25.42 -10.0% -10.6% -16.2% -20.7% -27.32 -13.3 -4.9% -1.6X

TecpotaryNlrndni 1.0% 1.0% 1.02 1.0% 1.0% 1.0% 1.0% 1.02 1.0% 1.0% 1.0% 1.8% 1.8%Fz-ixst real -12.0% -13.5% -2.2% -24.6% -26.(X -10.7% -11.4% -l6.9% -21.3l -27.8% -14.0% -4.92 -1.6%

SIVpand tem

Noinal 9.6% 8.7% 4.8% 4.8% 4.8% 0.0% 0.0% 0.0% 0.0% - - - -

Ec-pOst reAl -4.5% -6.9% 1.5% -21.7% -23.2X -11.6% -12.2% -17.7% -22.1% - - - -

NoticeNominal 5.07 5.0% 3.7% 3.7% 3.7% 6.0% 6.0% 10.0% 10.02 10.0% 10.5% 12.0 -

ia-pet real -8.57 -10.1% 0.4% -22.6% -24.1% -6.3% -7.0% -9.5% -14.13 -21.4% -5.9% 4.6% -

Installment savins1al 23.0% 23.0% 12.0% 12.0% 13.2% 13.2% 14.0% 14.2% 15.Z% 18.2% 19.5% 16.2% 8.0%

Err-pt real 7.2% 5.4% 8.5% -16.4% -17.1% 0.1% 0.2% -6.0% -10.3% -15.5% 1.8% 8.5% 4.32

Worma's proert a,Nuinal - - - - - - 19.2% 19.2% 23.0% 23.2% 24.5% 21.22 21.0%

E-"t real - - - - - - 4.6Z -1.9X -6.42 -12.0% 6.0% 13.2X 16.9%

Nasinel - - - - - - - 13.2% 12.6% 12.6% 12.3X 14.4% 8.0%t- t real - - - -6.9% -12.32 -19.5% -4.3S 6.8% 4.31

Term, 6 wnnthNIbinal 16.8% 14.4% 8.4% 8.4% 15.0% 13.8% 15.6% 13.8% 17.12 17.1% 16.9% 14.6% 7.6%t-Veot real 1.8% -2.0% 5.0% -19.1% -15.8% 0.6% 1.5% -6.47. -8.8% -16.3 -0.4% 7.0% 4.0%

Di eits, foreign denm.

esidet eporters

Ntd.nal(in U.S.$) 11.0% 11.0% 8.0% 9.07 15.0% 15.0% 15.0% 15.0% 15.0% 15.0% 15.0% 17.62 14.5%EK-pl t reel (in wm) 32.6% 9.1% 4.6% -18.0% -15.8% 1.7% 0.9% -5.42 -10.4% 27.5% 5.6% 22.8% 19.0%

d avere rate ofrtn on aboe assM

FE-post ree -2.1U -4.7% 2.6% -20.8% -2D.6% -4.3 -4.22 -10.8% -14.8% -20.7% -5.8% 2.20 1.7%

Lersing. daetic dmu.

Genera6nblrre 24.0% 22.0X 15.5% 15.5% 15.5% 15.5% 18.0% 16.0% 19.0% 19.0% 20.0% 17.0% 10.0%Ex-poet real 8.0% 4.5% 11.9% -13.8% -15.4% 2.17 3.6% -4.6% -7.3X -15.02 2.22 9.2% 6.3

For exportsNconLnat 6.0% 6.0% 6.0% 7.0X 9.0% 7.02 8.0% 8.0% 9.0O 9.02 15.0% 15.0% 10.0%Ex-poet real -7.t% -9.2X 2.7% -20.1% -20.2% -5.4. -5.22 -11.17 -15.1% -22.1% -2.(% 7.4% 6.3X

Inflation (CPI) 6 mxthsForward froi Deceer 14.9% 1.77 3.37. 33.9% 36.5% 13.1% 13.9% 21. f- 28.4% 39.9% 17.42 7.17 3.5X

Devaluation 6 rmothsForward fr.e fW1cenber 37.1U 14.7% 0.0 0.8% 0.0% 0.0% 0.0% 0.0% 0.0% 55.2% 7.8% 11.8% 7.6%

Inflatires (CPI) overPrior f eC or 10.52 12.3 .8% 8.5% 26.5% 25.4. 10.5% 11.O% 16.4% 21.2% 34.6% 11.7% 4.8%

Serces: Interest rate data are Eree The sanik of Knreea. ?ntshy Ecoemic Statistics. (YI, Echage Rate and LIkBR data are fromthe fIT, IFS, data tape.

Noxes: Real rates were caculaced ex-posc, i.e., they reflect rhe inflation rate six wontbs forward Eras Dtosereer.

a/ For 1976-81, the interest rate paid on l-rkmsn's ?roperty Forartion accoarts wea assurd to le the rate paid onInstallment Savirp acas,ts plos 5 percentge points. In 1982, the rate is the actual nrdnss.

6/ Er weighted average ~as calculated assurting all Istailnent acnmnts were paid the rate orn Installjrent Savings accountlTe weights were derived frm IKorea Table 2.

-b7I

lAvels of Selscted financial %sset$ in DOEptei Nos aka

tod at Year

1970 1971 1972 1973 1974 13975 1924 1977 1978 1979 1980 1981 1982

.0 ,11t ~~~~133.7 162.1 217.7 111.4 411.5 507.2 676,9 953.4 1364,4 1604.0 18 56.- 2025.0 2573.7

In 1980 prices 647.7 699.4 8'1.8 11-.96 1198.4 1180.4 1425.4 1809. 8 2225.0 2158,2 18 56.,4 1812.7 2197.62 of GNP 4.40 4.32 4.60 4,90 4..70 4.30 4.3Z 4.60 5.30T 4.70 4.5Z 4.22 4.72I of total assets 14.50 14.3% 14.20 1 5 .3% 1s.0 .92 16.02 [5.60 15.30 14,22 13.31 10.71 10.90

Deposits, domestic currency

Demand

ChockingNominal 27.0 29.8 45.9 74 4 i134.4 12'S8 I4 1, 239.9 3 53. 9 07 .9 541.41 710.6 772.0in 1980 prices 130.6 12 4 .3 .83.8 274.6 324.4 192,9 ~02,4 453.4 577.0 414.3 541.4 636.L 659.2I of GNP? 0.92 0.80 1.02 i.22 1.20 1.2% 0.92 2.12 1.30 0.92 1.3% 1.51 1.42I of total aSsets 2.92 2.52 3.00 3.70 4.17 "I"%1 3,40 3,91 4.02 2.72 3.82 3.82 3.31

PassbookNominal 103.9 117.4 165.5 246.7 293.4 3.9.', '.24. 623.4 940. 4 900'.4 979.8 1105.6 1905.8to 1990 prices 503.2 506.6 664.6 9iG.8 951.7 89., 40-3.5 1183.3 1370.5 1211.5 979.9 989.5 1627.32 of GNP 3.40 3.12 3.50 1.90 3.42 0,17 2.7% 7.02 3.12 2.60 2.41 2.32 3.522 Of total assets 11.32 10.32 10.80 12.12 11~.70; 11.00 1). 20 10.2Z 9.40 8.00 6.92 5.82 8.12

TemporaryNomLIna 73.0 110.2 101.4 173.3 701I.?- 2Y7.C, 395.4 739.9 1323.5 1869.0 2145.3 3403.9 4303.6In 1980 prices 3533.5 475.7 '24.2 639.6 3899. 644.S 802.7 1404.3 2158.3 2514.9 2145.3 3047.0 3676.70 of ON? 2,4Z 2.90 3.8z 2.70 2.321% 2.32 2.5% 3.52 4.82 5.52 5,21 7.02 7.82I of total assets 7.90 9.7% 11.8% 9.50 9. 11 9970 94Z4 12.12 14.8% 16.60 33.12 18.02 19.22

Savings and Tar,

Now HoousehoidNominal 30.8 37.4 59.6 66.9 43.2 0.0 , .1I 0.1 - - - - -in 1980 prices 149.0 161.3 234.5 25' .. 124" 1 15 C.3 0.2 - - - -2 of ON? 1.02 i.00 1.22 1.1% 0,30% 4.0 '2 0.02 - - - -2 Of total assets 3.30 3.3% 3,41 3..2 3.77 11,10 2,0 .02 - - -

NoticeNomina.l 17.3 11., 23.1 39.1' 39.~ i8. 133.3 247.9 c19.5 592.3 803 2 815.7 -to 1980 prices 84.9 47.9 80.7 142.7 11C4- le3.4 293.3 470.4 682.5 797.0 803.2 730.2 -2 of ONP 0.60 13. 30 .40 C. 62 0.02 ,7t 0. 9% 1.22 1.32 1.70 9.92 1.72 -2 of total "esst. 1.90 1.00 1.30 1.90 1.4 2.0 .3.3 4.02 ,. 71 5.32 5.62 4.32 -

instalmentsNomin.l. 203.4 2 5 5.7 335.4 49S.0 649.3, 809.2 ij21- 1386.5 1902. 3 2536.3 2722.6 3173.6 3046.2In 1910 prices 985.5 1103.5 1342.9 1827.4 1906,2t 20149.4 236-1. 2631.9 3LO57. 3 3372.3 2722.6 2840.8 2599.3% of ON? 6.72 6.62 .1 7.: % 7. 30 1 '105 '. 6.62 4,92 7.42 6.62 6.52 9.522 of total "Ssets 22.12 22.52 21.9%234.32 24,10 27.90 26.60 22.nt 21..32 22.22 19.12 16.82 12.92

Workman's PropertyNominal - - - 2'.8 104,2 2C5.1 283.6 406,4 626.0 1083.6to 1980 prices - 5 4.6 1.96.0 334.5 3S1. 4286.4 560.4 925.22 of ON? - - -12% 0.52 0. 72. 0.92 1.00 1.32 2.02I of total, assets - .. - -- lia 1.72 7.311, 2.5% 3.02 3.32 6.62

savingsso,s± n.l - - - .- 142.3 339.3 545. 1 694.4 1593.7 3280.0Ont 1980 price - -. - 194.2 553.3 479.6 694.4 1426).6 2783.62 of ON?P - - .50 .2:: 1.3T 3.70 3.32 5.92O of total. "sets - - .1.7% 3.911 4.30 4.90 9.40 13.82

TermNominal. 318.0 395.8 485.4 593,.5 '04.' ' 4 .4247. 1615.3 210.4 2631.4 3923.2 5285.7 6269.0In 1980 prices [340.5 1708,0 1947.3 2399.3 203S6.' 2.5. 2 . 096.3 3419.1 35 40.9 3923.2 '.731.4 5352.92 of ON? [0.52 10.62 17.32 4, ~4% B.z .0 9 1. 815 1.7% 7 . 71 7.1% 9.32 10 .90 11.402 of total aSsets 36.52 34.92 31.70 29,3% 29 .0 0.2 29. 4% 26.42 2 3.61: 20.3% 27.5% 27.90 26.52

Depos.its foreian core"ocy) omn l. Tin olars) 0.0 0.0 O.1 0. 31 . 0.2 31.2 0.1 0.2 0.3 0.6

NomInal (in von) ~~~~13.9 1. 249 2. 3.3 24.4 .7.3 113.6 92.9 71.9 150).5 199.0 445.2In 1980 prices b7.2 79.4 80.5 ., 17 .3.11.4 571, 99.6 215.6 1035.2 Onb.' 150).5 169.2 380.12 of GON? 0.50 0.3% 1,40% ).7.3 1.. .2 1).72 0,.30 7, 3 0. 0.42 0.42 0.82I of total assets 1.52 1 .1n -10 1.4% .i '.92 .12 1.92 r .91 16 1.12 1.02 1.92

4021.0 :136.7 1731.9 2012.9 948-. 53.18- 41.14 4123,2 89 336. 11271.9 142431.2 18929.6 2 36 57.1IIn 1980 prices 4447.1 480,15 41303.3 733)4,9 723'1.2 '0.39. 98,4. 11621.. 1.57'.C 734~6.' 1424 1.3 15943.7 20199.92 of ONP 30. 30 3C.4Z 17.,% 02.' 28.3 726. 154 , -2 29.22 30,.72 33.12 3cL. 50 38.90 43.11

CPI. December (Dec.80-100) 20.6 23,2 .0 27.1 34.3 '.. i 52.7 bi., 7c3 '.0 31.7 117.1

Ezcbaaog rate, Dme..ebr 316.7 371:2 178,3 3747, ,44 . 8-r.~6 . 494.,. 34.2 441 6c'3.9 700.3 748.8

CH?. Dec. Wd3oted, nomelnl 3038.7 3744.1 4724.9 4352.6 97.0:A 1 1915.r, 07. 211449.8 25133.' 3409',1 4i1258.2 49688.2 36937.3

Sources: Asset date are tram Th. SBank of Xcres, 9,3>_rnr ccsGNP sond CPI dateaore troo the 1".9 19, otoLoe

Notes: The GNP? data hae" been 1o.gr4tr,alc.118 lrtro.e ,'nc .crc,y r-r n a.trod srO. be .opr .... d 1aDecember prices. i-te1ens lnc'de 1--1oIo-n .Ž..,,8... 4.lanrad 1-1 lrg 0~.1e,o

- Indicates not defi-ed.

- 68 -

MOROXXCO: Table I

Selected Naninal ard FK-post Real Interest Rates

End of Year (arnnalized)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Deposits, damestic denom.

DemandNinal 0.07 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.07. O.O/ O.C. 0.0%

Ex-post real -3.5% 0.0% 3.0% -13.0% -4.0% -8.6% -9.1% -7.7% -4.3% -5.7X -14.5% -7.1X -2.7%

Term, 6 wmnthsEx-post real IJN/A ltN/A #N/A JN/A -0.1% -4.5% -5.0% -3.5% 1.4% 0.0% -8.1% -0.1% 5.5%

ImportationNaomLnal ltN/A lN/A #lN/A JtN/A 2.0% 2.3% 2.3% 2.3% 3.0% 3.0% 3.87 3.8% 4.3%Ex-post real I/N/A #N/A ltN/A ltN/A -2. L% -6.6% -7.1% -5.6% -1.5% -2.8% -11.T3% -3.6% 1.4%

Weighted average rate ofreturn on above assets

plus currency.

Ex-poet real IrN/A MIN/A JtN/A #N/A -3.6% -8.2X -8.7X -7.2% -3.67 -4.8% -13.4% -5.9% -1.27

Lading, dcmestic denon.

General short term (max)Nominal #N/A #N/A #iN/A ltN/A 8.0% 10.5% 10.5% 10.5% 10.5% 10.5% 11.5% 11.5% 12.0%Ex-post real ltN/A JIN/A (TN/A IlN/A 3.7% 1.0% 0.4% 2.0X 5.7% 4.2% -4.7% 3.6% 8.9%

Cotton wzrantsNominal ltN/A MIN/A ltN/A JtN/A 5.3X 5.5% 5.5X 5.5% 5.5% 5.5% 5.5% 6.5% 7.0XEXc-po"t real #N/A ltN/A JkN/A OVN/A 1.1% -3.6% -4.1% -2.6% 0.9% -0.5Z -9.8% -1.1% 4.1%

Inflation (CPI) 6 monthsForward from Decembr 3.6% 0.0% -2.9% 14.9% 4.1% 9.4% 10.17 8.3% 4.5% 6.0% 17.0% 7.7% 2.8%

1ivaluation 6 monthsForward from December -0.1% -8.37. -30.6% 3.0% -13.5% 12.4% -0.1% -5.4X 0.9% 3.2Z 58.1% 34.1% 19.8%

Inflation (CPI) overPrior December 2.67 4.7r 2.6% 8.7% 14.4X 6.1% 13.4% 9.0% 9.7% 9.0% 9.7% 13.2% 6.7%

Sources: Interest rate data are fran le Banque du Maroc, Rapport.CPI and Exchange Rate data are froan the EMF, IFS, data tape.Weights for weighted average are fran Morocco Table 2.

Notes: Real rates were calculated ex-post, i.e., they reflect the inflation rate six months forward from December.I[he interest rate paid on Thportation Deposits were assumed to be half the rate paid on 6 mDnth term deposits.

- 69 -

TCO: Table 2

levels of Selected Firanil Assets

End of Year(in billicar of Dirhans)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Currency, daosticNoinal 2.3 2.5 2.9 3.4 4.1 4.7 5.7 6.7 7.7 9.0 9.8 11.1 12.0in 1980 prices 5.2 5.4 6.3 6.7 7.0 7.5 8.2 8.7 9.2 9.9 9.8 9.8 10.0% of GNP 11.o% 11.1% 12.2% 11.5% 11.2Y 11.6% 12.3% 12.4% 12.8% 13.4% 13.2% 13.4% 13.0%% of total assets 38.0% 36.87 37.3 37.1% 34.3X 32.6% 33.9% 33.2f 31.6% 32.5% 31.8% 31.3 30.4%

Deposits, daestic cirrency

NoLlnal 3.3 3.7 4.4 5.2 6.8 8.2 9.4 11.2 13.0 14.3 15.5 17.9 19.9In 1980 prices 7.6 8.2 9.4 10.2 11.,6 13.3 13.5 14.7 15.5 15.7 15.5 15.8 16.5% of GNP 16.07. 16.9% 18.2% 17.37 18.7% 20.3 20.1% 20.9% 21.6% 2L.3% 20.9% 21.5% 21.5%X of total assets 55.1% 56.0% 55.7% 56.1% 57.17 57.47 55.7% 55.9% 53.3Z 51.6% 50.3 50.2Z 50.4%

Temn'md.nal 0.4 0.5 0.5 0.6 1.0 1.4 1.8 2.2 3.0 :3.7 4.6 5.9 7.3In 1980 prices 0.9 1.0 1.2 1.2 1.7 2.3 2.5 2.9 3.5 4.0 4.6 5.2 6.1% fc GP 2.0% 2.1% 2.3Z 2.1% 2.8. 3.6% 3.7% 4.1% 4.9% 5.5% 6.3% 7.1% 7.9%% of total assets 6.9% 7.17 6.9% 6.7% 8.5% 10.1% 10.4% 10.9% 12.2% 13.2% 15.1% 16.5% 18.5%

ImportationNirnal 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.7 0.8 0.9 0.7 0.3In 1980 prices 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.9 0.8 0.9 0.6 0.2% of GNP 0.0% 0.0% 0.0% 0.0% C1.0 0.0% 0.0O 0.0% 1.2% 1.1% 1.2% 0.8% 0.3X% of total assets 0.0% 0.0% 0.0% 0.0a 0.0. 0.0( 0.0% 0.0% 2.9% 2.7% 2.8% 2.0% 0.77.

TotalNbilnal 6.0 6.7 7.9 9.2 11.8 14.3 16.9 20.0 24.3 27.8 30.8 35.6 39.6In 1980 prioes 13.7 14.7 16.9 18.1 X).4 23.2 24.2 26.3 29.1 30.5 30.8 31.4 32.7% of GNP 29.0% 30.1% 32.6% 30.9% 32.7% 35.5% 36.1% 37.3 40.6% 241.2% 41.5% 42.8% 42.6%

CPI, Deember (Dec.8D'100) 43.4 45.5 46.7 50.8 58.1 61.6 69.9 76.2 83.7 91.2 100.0 113.2 120.8

GNP, Dec. adjusted, noaLnal 20.5 22.2 24.2 29.8 36.2 40.2 46.8 53.8 59.9 i57.4 74.3 83.2 92.9

Saurces: Asset data are fram le Banque du Mroc, Rapport. L ard CPI data are fran the 2F, llS, data tape.

Notes: The GNP data have been logrithmically interpolated between the carrent year and 4xv year forward so as to be expressed inDecember prices.

- 70 -

N(ERLA: Table 1

Selected Ntoinal and EK-post Real Interest Rates

Eid of Year (annalized)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Deposits, domestic denon.

DadNomdnal 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%Ex-poet real -27.0% -6.7% -20.8% -13.1% -422.6% -14.7% -28.9% -17.9% -13.3% -88.5% -25.4% -7.9% -23.8%

SaviTWNominal 3.0% 3.0% 3.0% 3.0% 3.0% 4.0% 4.0% 4.0% 5.0% 5.0% 6.0% 6.0% 8.5%Ex-post real -24.8% -3.9% -18.4% -10.5% -40.9% -11.3% -26.1% -14.6% -9.0% -4.0% -20.9% -2.4% -17.3%

Term, 6 mnthNominal 3.0% 3.0% 3.5% 3.5% 3.5% 3.0% 2.5% 3.0% 5.0% 5.5% 6.0% 6.0% 8.5%Er-post real -24.8% -3.9% -18.0% -10.1% -40.6% -12.2% -27.2% -15.5% -9.0% -3.5% -20.9% -2.4% -17.3%

Weighted average rate ofreturn on abowe assetsplus awrerwry

Er-post real #N/A -5.6% -19.8% -11.9% -41.7% -13.5% -28.1% -17.0% -11.8% -6.7% -23.8% -5.8% -2 1.2%

Lenirg.dcaestic dern.

Cerral advancesNkdrnal 8.0% 10.0% 10.0% 10.0% 10.0% 9.0% 10.0% 6.0% 11.0% 11.0% - - -Er-post real -21.1% 2.7% -12.9Z -4.4% -36.8% -7.0% -21.8% -13.0% -3.7% 1.5% - - -

First class avancesNominal 7.0% 7.07 7.0% 7.0% 7.0% 6.0% 6.0% 6.0% 7.0% 7.5% - - -Ex-post real -21.97 -0.1% -15.3% -7.0% -38.6% -9.6% -24.7% -13.07. -7.2% -1.7% - - -

General nwdmnNardnal - - - - - - - - 11.0% 11.0% 11.5% 11.5% 14.07Ex-post real - - - - - - - - -3.7% 1.5% -16.8% 2.7X -13.2Z

Perfered sector marinumNosdnal - - - - - - - 9.0% 9.0% 9.5% 9.5% 12.5%Ex-post real - - - - - - - - -5.5% -0.3% -18.3% 0.9Z -14.3%

Inflation (CPI) 6 moznthsForward from December 37.0% 7.1% 26.3% 15.1% 74.2% 17.2% 40.7% 21.8% 15.3% 9.3% 34.1% 8.6% 31.3%

Devaluation 6 monthsForward from Deceuber 0.0% 0.0% 0.0% -12.0% -3.6% 0.0% 5.9X -2.5% -15.0% -6.4% 40.97 12.5% 21.8%

Inflation (CPI) overPrior December 13.0% 15.0% -3.5% 17.8% 9.9% 43.1% 12.4% 31.3% 10.3% 11.5% 13.7% 17.4% 6.7%

Sources: All pre-19 82 interest rates are from the Central Bank of Nigeria, Economic and Financial Review or Amial Report.except the gereral mrdnun and prefered sector maxini lerndirg rates and all 1982 figures, which are from WorldBank data. CPI data are from the IMF, , data tape.

Notes: Real rates were calculated ex-post, i.e., they reflect the inflation rate six nonths forward from December.The weighted average uses weights from Nigeria Table 2 and assumes all tinm deposits vield at the 6 onth rate.

-' indicates not defined.

- 71 -

NIGR: A. le 2

levels of Setetad ?inmidal Aosets

Ehd of Year(in bd-llions of Naira)

1970 1971 1972 1973 1974 1975 1976 1977 1978 197Sc 1980 1981 1982

0, datic1oinl #N/A 0.4 0.4 0.4 0.6 1.0 1.4 1.9 2.2 2.4 3.2 3.9 4.2In 1980 price IN/A 1.3 1.5 1.4 1.7 2.1 2.5 2.7 2.7 2.7 3.2 3.3 3.4X of gp IN/A 4.8% 4.6% 3.5% 3.1% 4.5% 5.1% 6.8% 6.4% 5.i% 7.3 8.7% 9.2%% of total aswts eN/A 34.9% 33.2% 30.8% 26.4% 28.5% 25.6% 27.5% 29.Z% 23.1% 22.0? 24.8% 25.0%

Dqlits, d-tiC currency

DTdNomirl EN/A 0.3 0.3 0.4 0.6 1.0 1.9 2.9 2.6 3.8 6.0 5.9 5.8In 1980 pric #N/A 1.1 1.2 1.3 1.8 2.1 3.6 4.0 3.3 4.3 6.0 5.0 4.72 Of d IN/A 3.9% 3.8% 3.2% 3.3X 4.4% 7.3 iO.1% 7.8% 9.U% 13.9% 13.3% 12.7%2 of total assets #N/A 28.1% 27.1% 27.7% 28.2% 28.0% 36.8% 40.4% 35.5% 38.5% 41.8% 37.8% 34.5%

S-qNbslml #N/A 0.2 0.2 0.2 0.3 0.5 0.7 0.9 1.1 1.3 1.7 2.0 2.3In 1980 price N/A 0.6 0.8 0.7 0.9 1.1 1.3 1.3 1.4 1.15 1.7 1.7 1.92 df (1W ENA 2.2% 2.4% 1.9% 1.6% 2.3% 2.7% 3.3Z 3.22 3.1% 3.8% 4.5% 5.0%2 of toa assets EN/A 16.7% 17.3 16.2 13.5 14.6Z 13.6% 13.3 14.72 13.1% 11.5% 12.8% 13.7%

Iml IN/A 0.2 0.3 0.4 0.7 '.1 1.3 1.3 1.5 2.4 3.6 3.8 4.5In 1983 prii Es/A 0.8 1.0 1.2 2.0 2.2 2.3 1.9 1.9 2.7 3.6 3.3 3.62 oi CUP IN/A 2.92 3.1% 2.9% 3.7% 4.6% 4.8% 4.7% 4.5% 5.8% 8.2% 8.6% 9.8%

Nnal EN/A 1.0 1.2 1.4 2.2 3.6 5.3 7.1 7.4 9.9 14.5 15.5 16.9In 198D prio EN/A 3.8 4.4 4.6 6.4 7.5 9.7 9.9 9.4 11.2 14.5 13.2 13.5% of G1W EN/A 13.7% 13.9% 11.4% 11.6% 15.9% 19.8% 24.9% 21.8% 23.7% 33.3% 35.Z% 36.7%

CPI, leber (Dec.80-100) 23.6 27.1 26.1 30.8 33.9 48.4 54.4 71.5 78.8 87.9 100.0 117.4 125.2

GDP. Dec. adijusted, tnir 6.3 7.4 8.3 12.4 18.6 22.8 26.7 28.4 33.9 41.6 43.4 44.2 46.0

Sourcs: Asset data are frum the Central Beric of Nigerls, satLX x. CPI ard GDP datL are froa the DIF, IFS, data tape.

Notes: The CP data ham been logrithdica.ly interpolated beoieen the current year and oe year forwarrd so as to be expressed irDecwber prics.

-72 -

PAKISTAN: Table 1

Selected Nominal and Ex-post Real Interest Rates

End of Year (annualized)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Deposits, domestic denom.

DemandNominal 0.0% 0.0% 0.0% 0.1% 0.1% 0.1% 0.1% 0.1% 0.3% 0.1% 0.1% 0.1% 0.0%Ex-post real -5.7% -3.3% -17.3% -11.7% -14.5% -3.0% -3.3% -3.2% -6.1% -12.1% -7.4% -1.6% -6.7%

SavingsNominal 4.2% 4.1% 4.9% 5.8% 6.1% 6.6% 6.7% 7.6% 7.6% 7.6% 7.6% 7.6% 7.6%Ex-post real -1.7% 0.6% -13.2% -6.7% -9.3% 3.3% 3.1% 4.0% 0.7% -5.6% -0.5% 5.8% 0.4%

Term, 6 monthsNominal 5.2% 5.1% 5.6% 6.7% 8.2% 8.9% 9.1% 9.6% 10.0% 10.1% 10.2% 10.2% 9.9%Ex-post real -0.8% 1.6% -12.7% -5.9% -7.6% 5.4% 5.4% 5.9% 3.0% -3.4% 1.9% 8.4% 2.5%

Weighted average rate ofreturn on above assetsplus currency

tx-post reail -3.8% -1.3% -15.2% -9.2% -12.0% 0.4% 0.4% 0.8% -2.2% -8.5% -3.6% 2.7% -2.4%

Lending, domestic denom.

Weighted average ofall advances from theScheduled Banks

Nominal 8.1% 8.3% 8.7% 9.4% 11.2% 11.1 11.2% 12.1% 11.7% 11.6% 11.9% 10.9% 11.0%tx-post real 1.9% 4.7% -10.1% -3.4% -5.0% 7.6% 7.4% 8.3% 4.5% -2.1% 3.5% 9.0% 3.5%

Short term credits fromthe Ag. Deve]Lopment Bank

Nominal 7.0% 7.0% 7.0% 9.0% 10.0% 10.0% 11.0% 12.0% 11.0% 11.0% 11.0% 11.0% 11.0%Ex-post real 0.9% 3.5% -11.5% -3.8% -6.0% 6.5% 7.3% 8.3% 3.9% -2.6% 2.7% 9.2% 3.6%

Inflation (CPI) 6 monthsForward from December 6.0% 3.4% 20.9% 13.3% 17.1% 3.2% 3.5X 3.4% 6.8% 13.9% 8.1% 1.7% 7.2%

Devaluation 6 mionthsForward from December -2.2% 429.7% -19.5% 0.0% 0.07 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 51.9% 5.2%

Inflation (CPI) overPrior December 5.2% 5.8% 7.8% 37.8% 20.6% 12.7% 10.3% 7.3% 5.5% 9.0% 15.1% 10.4% 3.8%

Sources: Interest rates are from The State Bank of Pakistan, Bulletin.CPI and exchange rate data from the IMF, IFS, data tape.Weights for the weighted average of returis are from Pakistan Table 2.

Notes: Real rates were calculated ex-post, i.e., they reflect the inflation rate six months forward from December.Deposit rates are weighted averages of interest actually paid of all deposits.

- 73 -

PAKISIAN: Table 2

Levels of Selected Financial Assets

Enid of Year(in billions of Rupees)

1970 1971 1972 1973 1974 1975 1976 1977 1978 L979 1980 1981 1982

Currency, dmesticNmdnal 8.1 7.3 6.5 8.6 10.6 11.7 13.8 17.3 21.0 26.4 32.5 34.5 41.1In 1980 prices 27.0 23.3 19.0 18.4 18.7 18.4 19.6 22.9 26.3 30.4 32.5 31.2 35.9% of CNP 16.4% 14.1% 10.7% 11.2% 10.6% 9.57 9.5% 10.0% 10.4% 11.3% 11.6% 10.6% 11.0%% of total assets 39.8% 34.8% 28.5% 30.9% 35.1% 32.4% 29.8% 31.4%; 30.8% 32.8% 34.3Z 32.3X 31.9%

Deposits, dui-stic currency

DemandNomxnal 3.6 4.0 5.2 6.0 6.4 7.0 9.1 9.5 11.7 14.1 16.4 17.1 18.8In 1980 prices 12.0 12.8 15.2 12.7 11.3 11.0 12.9 12.5 14.7 16.2 16.4 15.5 16.4% of CNP 7.32 7.7% 8.6% 7.8% 6.4% 5.7X 6.2% 5.5S 5.8% 6.1% 5.9% 5.2% 5.0%% of total assets 17.6% 19.1% 22.9% 21.4% 21.3% 19.4% 19.6% 17.Z: 17.2% 17.5% 17.3% 16.0% 14.6%

SavinsNomLnal 4.3 5.5 6.5 7.9 8.3 10.5 14.0 17.2 22.0 25.3 28.6 34.2 41.8In 1980 prices 14.4 17.4 19.0 16.9 14.6 16.5 19.9 22.7 27.6 29.1 28.6 31.0 36.4% of GNP 8.7% 10.5% 10.7% 10.3% 8.3% 8.5% 9.6% 9.9%,9 11.0% 10.9% 10.2% 10.5% 11.2.% of total assets 21.1% 26.IZ 28.5% 28.4Z 27.5% 29.0% 30.2% 31.3! 32.3% 31.5% 30.Z% 32.1% 32.4%

TermNm.nal 4.4 4.2 4.6 5.4 4.8 6.9 9.4 11.0 13.4 14.6 17.3 20.9 27.1In 1980 prices 14.6 13.3 13.4 11.5 8.5 10.9 13.4 14.6 16.8 16.9 17.3 19.0 23.6% of GNP 8.9% 8.1% 7.6% 7.0% 4.8% 5.6% 6.5% 6.4t 6.7% 6.3% 6.2% 6.4% 7.2%% of total asets 21.5% 19.9% 20.2% 19.3% 1.6.1% 19.2 20.4% 20.0Y 19.7% 18.2% 18.2% 19.6% 21.0%

TotalInal 20.3 21.1 22.7 27.9 '30.1 36.2 46.3 54.9 68.1 80.4 94.8 106.7 128.8In 1980 prie 68.0 66.8 66.7 59.5 53.2 56.7 65.8 72.7 85.5 92.6 94.8 96.6 112.4% of GM 41.3% 40.4% 37.6% 36.4% :0.2% 29.2% 31.8% 31.8% 33.9% 34.6% 33.9% 32.7% 34.4%

CPI, Dlcw er (Dec.8DI100) 29.8 31.6 34.0 46.9 56.6 63.8 70.3 75.5 79.7 86.9 100.0 110.4 114.6,

QGP, Dec. adjusted, nowinsl 49.1 52.2 60.4 76.7 99.6 123.8 145.6 172.9 200.7 232.6 279.6 326.1 374.4

Sources: Asset data are from Ihe State Bark of Paldstan, Billetin. GNP and CPI data are frao the IDF, IFS, data tape.

Notes: The GNP data have been logrithuically interpolated betwei the current year and one year forard so as to be expressed inDecanber prices.

- 74 -

PERU: Table I

Selected Effective and Ex-Post Real Interest Rates a/

End of Year (annualized)

1970 1971 1972 1973 1974 1975 L976 1977 1978 1979 1980 1981 1982

Deposits, domestic denom.

DemandNominal 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 2.0% 2.0% 2.0% 2.0% 2.0% 2.0% 10.0%Ex-post real -5.4% -6.5% -15.0% -20.8% -22.5% -21.6% -28.0% -48.2% -39.7% -33.0% -47.5% -39.6% -59.2%

SavingsNominal 5.0% 5.0% 5.0% 5.0% 5.0% 5.0% 9.0% 11.5% 29.0% 30.5% 30.5% 50.5% 55.0%Ex-post real -0.7% -1.9% -10.7% -16.9% -18.7% -17.7% -23.1% -43.4% -23.7% -14.3% -32.8% -10.9% -42.6%

Term, 6 monthNominal 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 11.0% 14.0% 31.5% 31.5% 31.5% 63.0% 71.2%Ex-post real 1.2% 0.0% -9.0% -15.3% -17.1% -16.1% -21.7% -42.1% -22.3% -13.7% -32.3% -3.4% -36.6%

Morgage bank certificatesNominal 9.0% 9.0% 9.0% 9.0. 9.0% 9.0% 11.0% 14.0% 31.5% 33.0% 33.0% 61.6% 71.2%Ex-post real 3.1% 1.9% -7.3% -13.7% -15.6% -14.6% -21.7% -42.1% -22.3% -12.7% -31.5% -4.3% -36.6%

Deposits, foreign denom.

Nominal (in U.S.$) 5.0% 5.0% 5.0% 5.0% 5.0% 5.0% 5.0% 4.6% 9.1% 11.4% 14.9% 11.0% 6.7%Ex-post real (in Soles) -0.7% -1.9% -10.7% -16.9% -18.7% 71.7% -1.8% -25.4% -15.0% -5.1% -10.8% 17.2Z 1.6%

Weighted average rate ofreturn on above assetsplus currency

Ex-post real -3.1% -4.0% -12.7% -18.7% -20.6% -19.1% -26.4% -45.8% -31.4% -21.7% -32.6% -12.0% -30.0%

Lending, domestic denom.

GeneralNominal 16.3X 16.3% 16.3% 16.3% 16.3% 16.3% 21.2% 24.2% 50.4% 52.7% 52.7% 98.0% 98.0%Ex-post real 10.0% 8.7% -1.1% -7.9% -9.9% -8.9% -14.5% -36.9% -11.1% 0.2% -21.4% 17.3% -26.6%

Rediscount rates to theBanco Agrario

Nominal 4.0% 4.0% 4.0% 4.0% 4.0% 4.0% 7.0% 9.0% 20.0% 21.0% 21.0% 29.0% 29.0%Ex-post real -1.6% -2.8% -11.6% -17.7% -19.4% -18.5% -24.5% -44.7% -29.1% -20.6% -37.7% -23.6% -52.2%

Inflation (CPI) 6 monthsForward from December 5.7% 7.0% 17.6% 26.3% 29.1% 27.6% 41.8% 97.0% 69.1% 52.3% 94.3% 68.8% 169.9%

Devaluation 6 monthsForward from December 0.0% 0.0% 0.0% 0.0% 0.0% L08.6% 32.6% 40.6% 31.7% 29.8% 50.8% 78.3% 156.8%

Inflation (CPI) overPrior December 5.8% 7.6% 4.4% 13.8% 19.1% 23.9% 44.7% 32.5% 73.7% 65.3% 60.8% 72.7% 65.6%

Sources: 1970-77 interest rates are from La Superiritendencia de Banca y Seguros, Boletin Estadistico. 1978-82 interest rates arefrom El Banco Central de Reserva del Peru, Boletin, CPT, exchange rate and LIBOR rate data are from the IMF, IFS,data tape. Weights for the weighted average are frum Peru Table 2.

Notes: Real rates were calculated ex-post, i.e., they reflect the inflation rate six months forward from December.

a/ The 1982 ceiling rate on demand deposits was 55%, actual rates paid were typically 10%, with 20% paid on large deposits.For 1981, term deposits and morgage hank certificates were compounided quarterly. For 1982, compounding was monthly.For 1970-76, dollar deposits were assumed to yield 5x. For 1977-82, the yield was assumed to be LIBOR minus threepercentage poiits.The effective, general lending rates were the ceiling rates plus 2 percentage points commission adjusted for 100%prepayment of interest and commisslons.

- 75 -

PERU: Table 2

Levels of Selected Finracial Assets in tte Banking System

End of Year(in bilUions of Soles)

1970 1971 1972 1973 L974 1975 1976 1977 1978 1979 1980 1981 1982

Currenxy, damesticlNoinal 16.3 18.9 21.9 27.2 33.5 42.6 49.6 60.8 91.0 162.0 273.4 436.2 627.9In 1980 prices 272.7 294.0 326.2 356.1 368.2 377.9 304.0 281.4 242.5 260.4 273.4 252.6 219.6Z of GP 6.5% 6.8% 6.8. 6.8% 6.8%. 6.6% 5.6% 4.7% 4.2% 4.3% 4.3% 4.1% 3.3%% of total assets 31.2% 31.6% 30.7% 31.8% 32.0% 34.5% 34.1% 32.0% 29.1% 25.8% 22.3% 2D.0% 16.6%.

Deposits, daaestic qwrency

Demwdkdna1 16.7 17.4 22.0 25.9 33.9 37.2 47.5 60.3 86.0 154.3 268.5 367.8 461.0in 1980 prices 279.4 270.6 327.7 339.1 372.6 330.0 291.1 279.0 229.2 248.1 268.5 213.0 161.2% of UP 6.77. 6.3% 6.8% 6.5% 6.9% 5.8Z 5.47 4.7% 3.9% 4.17. 4.3Z 3.5% 2.5%% of total assets 31.9% 29.1% 30.9% 30.3X 32.4% 30.1% 32.7% 31.7% 27.5% 24.6% 21.9% 16.9% 12.2%

SavNomlnal 8.0 8.9 10.0 11.3 12.8 15.0 17.1 21.6 30.2 67.0 137.8 355.7 680.2In 1980 prices 133.8 138.4 149.0 148.0 140.7 133.1 104.8 100.0 80.5 107.7 137.8 206.0 237.9% of (NP 3.2Z 3.2% 3.1% 2.8% 2.6% 2.3% 1.97 1.7% 1.4% 1.8% 2.27 3.3% 3.6%X of total assets 15.37 14.9X 14.0% 13.27 12.27 1Z.2% 11.8% 11.4% 9.6Z 10.7% 11.2% 16.3% 18.0%

TermNomtnal 5.8 7.3 7.6 8.6 9.1 9.7 9.9 16.4 24.4 39.5 60.8 114.5 222.1

in 1980 prices 97.0 113.5 113.2 112.6 100.0 86.1 60.7 75.9 65.0 63.5 60.8 66.3 77.7% of QP 2.37 2.6% 2.4% 2.2% 1.8% 1.5Z 1.1% 1.3% 1.1% 1.0% 1.0% 1.1% 1.2%% of total assets 11.1% 12.2% 10.7% 10.1% 8.7% 7.9% 6.8% 8.6% 7.8% 6.3% 5.0% 5.3% 5.9%

MbrSW bsrk certificatesNominal 5.1 7.1 9.5 12.1 14.6 18.2 19.8 23.8 30.8 59.1 107.1 242.4 322.3In 1980 prices 85.3 110.4 141.5 158.4 160.5 161.5 121.4 110.1 82.1 95.0 107.1 140.4 112.7% of GP 2.0% 2.6% 2.97 3.0% 3.0% 2.8% 2.D. 1.8% 1.4% 1.6% 1.7% 2.3% 1.7%% of total assets 9.8% 11.9% 13.37. 14.2% 13.9% 14.7% 13.6% 12.5% 9.8% 9.4% 8.7% 11.1% 8.5%

teposits, foreign wrrencyN1binal in dollars 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.4 0.7 1.5 1.9 2.9Nominal in sole 0.4 0.2 0.3 0.4 0.8 0.7 1.5 7.2 50.6 146.6 378.2 662.1 1470.4in 1980 prices 6.7 3.1 4.5 5.2 8.8 6.2 9.2 33.3 134.8 235.7 378.2 383.4 514.2% of total assets 0.8% 0.3% 0.4% 0.5% 0.8% 0.6% 1.0% 3.8% 16.2% 23.3% 30.9% 30.4% 38.9%

TotalNmdral 52.3 59.8 71.3 85.5 104.7 123.4 145.4 190.1 313.0 628.5 1225.8 2178.7 3783.9In 1980 prios 874.9 930.1 1062.1 1119.5 1150.8 1094.7 891.2 879.7 834.1 1010.4 1225.8 1261.6 1323.3% of QP 21.0% 21.6% 22.17 21.5% 21.3% 19.3% 16.5% 14.7% 14.3% 16.6% 19.4% 20.5% 20.2%

CPI December (Dec.80-100) 6.0 6.4 6.7 7.6 9.1 11.3 16.3 21.6 37.5 62.2 100.0 172.7 285.9

Ewcclwe rate, Daember 38.7 38.7 38.7 38.7 38.7 38.7 45.0 69.4 130.4 196.2 250.1 341.2 506.2

GNP, Dec. adjusted, ixinal 249.6 277.0 322.2 397.2 492.1 640.6 882.7 1290.4 2188.6 3790.7 6308.5 10639.2 18760.0

Sources: Asset data are fromu the Banco Central de Reserva del Perm, mria. GNW and CPI data are from the IMF, MFS, data tape.

Notes: The GNP data have been logrtt±mically interpolated betwen the ctrrent year and am year forward so as to be expressed inDecermr prices. The figures exclude the non-baxk financial sector, except for deposits in the banking system.

- 76-

THAILAND: Table I

Selected Nominal and Ex-Post Real Interest Rate Ceilings

Erd of Year (annualized)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Deposits, domestic denom.

DermandNominal 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%Ex-post re!al -0.6% -7.1% -17.6% -26.0% -3.6% -2.6% -12.2% -11.0% -9.5% -22.1% -15.3% -3.3% -5.5%

SavingsNominal 3.5% 3.5% 3.5% 3.5% 4.5% 4.5% 4.5% 4.5% 4.5% 5.5% 8.0% 9.0% 9.0%Ex-post real 2.8% -3.9% -14.7% -23.5% 0.8% 1.8% -8.3% -7.0% -5.4% -17.9% -8.5% 5.4% 3.1%

Term, 6 months.Nominal 6.0% 6.0% 6.0% 6.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 10.0% 11.0% 11.0%Ex-post real 5.3% -1.6% -12.6% -21.6% 3.2X 4.2% -6.1% -4.7% -3.27 -16.7% -6.8% 7.4% 4.9%

Weighted average rate ofreturn on above assetsplus currency

Ex-post real 2.4% -4.1% -14.8% -23.5% 0.5% 1.8% -8.2% -6.8% -5.3Z -18.5% -9.4% 4.5% 2.4%

Lending, domestic denom.,

Ceneral loarLs & overdraftsNominal 14.0% 14.0% 14.0% 14.0% 15.0% 15.0% 15.0% 15.0% 15.0% 15.0% 18.0% 18.0% 18.0%Ex-post real 13.2% 5.9% -6.1% -15.7% 10.9% 12.0% 1.0% 2.4% 4.1% -10.5% 0.0% 14.1% 11.6%

Export (discount rate)Nominal 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0%Ex-post real 6.3% -0.6% -11.8% -20.9% 3.2% 4.2% -6.1% -4.7% -3.2% -16.7% -9.3% 3.5% 1.2%

Inflation (CP.) 6 monthsForward from Decenber 0.7% 7.7% 21.3% 35.2% 3.7% 2.6% 13.9% 12.3% 10.5% 28.4% 18.0% 3.4% 5.8%

Devaluation 6 monthsForward fromn December 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.4% -0.2% 3.6% 0.0% 0.0%

Inflation (CPI) overPrior December -1.3% 1.3% 8.8% 20.2% 17.9% 4.4% 3.4% 8.9% 7.8% 15.0% 16.4% 12.3% 2.6%

Sources: Interest rates are from the Bank of Thailand, Monthly (Quarterly) Bulletin.CPI data are from the IMF, IFS, data tape.Weights for weighted average are from Thailand Table 2.

Notes: Real rates were calculated ex-post, i.e., they reflect the inflation rate six months forward from December.

- 77 -

THAILWAD: Table 2

Levels of Selected Financial Assets

End of Year(in billions of Baht)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Curreny, d>TPsticNoxinal 11.9 13.1 15.3 18.6 20.4 22.3 25.7 28.5 33.0 40.6 45.7 47.5 53.7In 1980 prices 31.4 34.2 36.7 37.3 34.7 36.2 40.3 41.2 44.2 47.3 45.7 42.3 46.6% of GNP 8.4% 8.5% 8.1% 7.7% 7.2% 7.0% 7.1% 6.7% 6.5% 6.7% 6.4% 6.0% 6.2%% of total assets 28.7% 27.1% 25.5% 25.5% 22.9X 21.3% 20.6% 19.1% 18.4% 19.9Z 18.3% 16.4% 14.9%

Deposits, domestic uirrcy

Demntbidnal 6.9 7.7 8.8 10.6 12.5 13.0 15.1 16.4 21.1 22.2 25.1 25.0 23.5In 1980 prices 18.4 20.2 21.2 21.1 21.1 21.2 23.8 23.7 28.2 25.8 25.1 22.3 20.4% of GNP 4.9% 5.0% 4.7% 4.4% 4.4% 4.1% 4.2Z 3.8% 4.2% 3.7% 3.5% 3.2% 2.7%% of total assets 16.8% 16.0% 14.8% 14.4% 13.9% 12.5% 12.1% 11.0% 11.8% 10.9% 10.0% 8.6% 6.5%

SavinpNmuinal 2.7 3.0 3.9 4.9 6.3 7.2 8.9 10.7 14.2 1'.2 27.1 36.8 60.0In 1980 prioes 7.2 7.7 9.3 9.8 10.7 11.8 14.0 15.4 19.1 20.0 27.1 32.7 52.1: of CQP 1.9% 1.9% 2.0% 2.C% 2.27 2.3% 2.4% 2.5% 2.8% 2.8% 3.8% 4.6% 7.0%% of total assets 6.6% 6.1% 6.4% 6.7% 7.1% 6.9% 7.1% 7.1% 8.0% 8.4% 10.8% 12.7% 16.6%

TennNoautal 19.8 24.5 31.9 39.1 50.1 62.1 74.9 94.1 110.7 124.0 152.3 181.1 223.3In 1980 prics 52.4 64.2 76.6 78.2 85.1 100.9 117.7 135.8 148.3 144.4 152.3 161.3 193.9% of total assets 47.9% 50.9% 53.3% 53.4% 56.1 59.3X 60.1X 62.9% 61.9% 60.8% 60.9% 62.4% 61.9%

TotalNomial 41.3 48.3 59.8 73.2 89.3 104.6 124.6 149.7 179.1 204.0 250.3 290.4 360.5In 1980 prices 109.5 126.3 143.8 146.4 151.7 170.1 195.8 216.0 239.7 237.5 250.3 258.6 313.0% of GW 29.4% 31.3% 31.7% 30.2Z% 31.3X 33.0% 34.3% 35.17 35.5% 33.7% 34.9% 36.7% 41.9%

CPI, Deer (Dec.80-100) 37.7 38.2 41.6 50.0 58.9 61.5 63.6 69.3 74.7 85.9 100.0 112.3 115.2

GNP, Dec. adjusted, rmdnal 140.5 154.2 188.4 242.5 285.1 316.9 362.7 426.2 503.8 606.2 716.9 791.6 860.1

Soarces: Asset data are fran the Bark of Thailand, Quarterly (mthly) Bulletin. GNP and CPPI data are fron the IMF, IFS, datatape.

Notes: The GNP data have been logrithmicaUy interpolated between the current year and oa- year forward so as to be expressed inDeaenber prices. Deposits are nCn-goveLrmvt deposits in the ccnmrcial banks, Wlich are 80-90% of all private deposits.

- 78 -

TURKEY: Table I

Selected Nominal and Ex-post Real Interest Rates

End of Year (annualized)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Depositss Ioustic denom.

DemandNominal #IN/A #N/A ltNiJA 0.0% 2.0% 2.0% 2.0% 2.0% 0.0% 0.0% 0.0% 0.0% 0.0%Ex-post real #N/A #N/A ItN/A -12.8% -17.5% -10.0% -16.7% -24.2% -43.3% -63.9% -24.6% -24.2% -18.6%

SavingsNominal #N/A 9iN/A ON/A 2.5Z 3.0)% 3,0% 3.0% 3.0% 3.0% 3.0% 5.0% 5.0% 5 0%Ex-post real i#N/A #N/A #tN/A -10.7% -16.7% -9.1% -15.8% -23.4% -41.6% -62.8% -20.8% -20.4% -14.5%

Term, 6 monthNominal #N!A #N/A #N/A 4.0% 6.0% 6.0% 6.0% 6.0% 9.0% 12.0% 32.0% 50.0% 50.0%Ex-post real #hN/A dN/A #N/A -9.3% -14.3% -6.5% -13.4% -21.2% -38.2% -59.5% -40.4% 13.8% 22.1%

Weighted average rate ofreturn on above assetsplus currency

Ex--p06t real #N/A #NR/A #N!A -11.2% -16.9% -9.4% -16.2% -23.9% 42.1% -62.9% -18.8% -8.2% 1.2%

Lendi.g. domestic denom.

General, saort termNominal #NN/A 'A N/A #N/A 8.8% 9.0% 9.09% 9.0% 9.0% 10.0% 10.87 26.0% 65.0%a/ 65.0% a/Expost real #N/A #1N/A #1N/A -5.2% -11.97 -3.8% -10.9% -19.0% -37.7% -60.0% -5.0% 25.1%. 34.3%

Ag. credits, short termNominal #N/A i#N/A (#N/A 7.0% 8.0Z 8.(% 8.0% 8.0% 8.0% il.5% 13.5% 20.0% 18.0%Ex-post real #N/A #/N/A ('N/A -6.7% -12.7% -4.7% -11.8% -19.7% -38.8% -59.7% -14.4% -9.0% -4.0%

inflation (CPI) 6 monthsForward from December 20.3% 6.4% 22.6% 14.7% 23.7% 13.3% 22.4% 34.5% 76.5% 176.8% 32.6% 31.9% 22.9%

Devaluation 5 monthsForward from December 172.6% -10.1Z 0.0% -7.1% 2.22 3. 8% 12.5% 68.7% 9h.0% 396.7% 50.8% 53.7% 40.7%

Inflation (CPT) overPrior Deceamber 12.3% 17.6% 9.2% 16.8% 16.7% 19.6% 17.0% 44.6% 36.67 81.17 86.2% 30.3% 32.8%

Sources: Interest rates are from the Central Bank Quarterly Bulletin, and World Bank data. CPI data are from the ŽF, IFS,data tape. Weights for weighted average are from Turkey Table 2.

NJotes: Real rates were calculated ex-post, i.e., they reflect the inflation rate six months forward from December.

a; Wcrld Bank staff estimate.

- 79 -

TURKEY: Table 2

Levels of Selected Financial Assets

End of Year(In billions of Turkish Lira, TI.)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Currency, domestic

Nominal 11.9 13.9 16.0 20.7 26.2 32.9 42.5 63.0 93.8 143.7 217.5 280.6 411.9In 1980 prices 194.1 192.7 203.2 225.0 244.1 256.3 283.0 290.1 316.3 267.5 217.5 215.4 238.1X of GNP 7.1% 6.5% 5.9% 5.7% 5.5% 5.5% 5.5% 5.9% 5.6% 4.6% 4.0% 3.7% 4.1%X of total assets 27.0% 24.6% 22.6% 23.0% 23.2% 22.5% 23.5% 25.9% 28.6% 27.4% 24.7% 17.2% 16.1%

Deposits, domestic currency

DemandNominal 6.6 8.7 11.8 16.0 22.6 32.1 44.9 63.0 86.0 154.5 286.0 458.5 651.3In 1980 prices 107.7 120.6 149.8 173.9 210.5 250.1 299.0 290.1 i90.0 287.7 286.0 352.0 376.5% of GNP 3.9% 4.0% 4.3% 4.4% 4.7% 5.3% 5.8% 5.9% 5.1% 4.9% 5.3% 6.1% 6.5%Z of total assets 15.0% 15.4% 16.7% 17.8% 20.0% 22.0% 24.8% 25.9Z 26.3% 29.4% 32.5% 28.1% 25.5%

SavingsNominal 16.8 20.9 24.9 32.9 39.7 52.2 62.7 82.4 1.03.3 143.7 197.4 228.5 275.5In 1980 prices 274.0 289.8 316.2 357.6 369.8 406.6 417.5 379.5 :148.3 267.5 197.4 175.4 159.32 of GNP 10.0% 9.7% 9.1% 9.0% 8.3% 8.7% 8.2% 7.8% 6.1% 4.6% 3.7% 3.0% 2.8%% of total assets 38.1X 37.0% 35.2% 36.5% 35.1% 35.7% 34.7% 33.9% 31.5% 27.4% 22.5% 14.0% 10.8%

TermNominal 8.8 13.0 18.0 20.5 24.6 29.0 30.8 34.4 44.4 83.3 177.9 665.1 1212.2% of GNP 5.2% 6.0% 6.6% 5.6% 5.1% 4.8% 4.0% 3.2% 2.6% 2.7% 3.3% 8.8% 12.1%% of total assets 20.0% 23.0% 25.5% 22.8% 21.8% 19.8% 17.0% 14.2% 13.6% 15.9% 20.2% 40.7% 47.5%

TotalNominal 44.1 56.5 70.7 90.1 113.1 146.2 180.9 242.8 327.5 525.2 878.8 1632.7 2550.9In 1980 prices 719.3 783.4 897.8 979.3 1053.5 1138.9 1204.6 1118.1 1104.2 977.8 878.8 1253.3 1474.72 of GNP 26.1% 26.2% 25.9% 24.8% 23.6% 24.3% 23.6% 22.9% 19.4% 16.8% 16.3% 21.6% 25.5%

CPI, December (Dec.80-100) 6.1 7.2 7.9 9.2 10.7 12.8 15.0 21.7 29.7 53.7 100.0 130.3 173.0

GNP, Dec. adjusted, nominal 168.7 215.4 273.1 363.8 478.3 601.3 767.6 1061.4 1684.9 3123.3 5391.5 7560.5 10017.9

Sources: Asset data are from the Central Bank of the Republic of Turkey, Quarterly Bulletin.CPI and GNP data are from the MF, IFS, data tape.

Notes: The GNP data have been logrithmically interpolated between the current year and one year forward so as to be expressed inDecember prices.

- 80 -URUWAY: Table I

Selected Nasinal and Ex-post Real Interest Rates

End of Year (annualized)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 a/ 1982

Deposits, domestic deno..

DemandNominal 0.0% 0.0% 0.0% 0.0% 10.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%Ex-post real -16.9% -49.2% -46.2% -44.8% -30.5% -15.5% -40.2% -28.9% -43.1% -32.1% -23.8% -9.8% -39.5%

SavirgsNominal 6.0% 6.0% 6.0% 8.0% 18.0% 18.0% 21.6% 25.5% 22.8% 24.0% 25.2% 24.0% 24.2%Ex-post real -12.0% -46.2% -43.0% -40.4% -25.4% -00.2% -27.3% -10.8% -30.1% -15.9% -4.6% 11.8% -24.9%

Term, 6 monthNominal 15.0% 15.0% 15.0% 18.0% 30.0% 30.0% 30.2% 51.4% 42.6% 50.6% 50.3% 47.4% 66.2%Ex-post real -4.5% -41.6% -38.1% -34.9% -17.8% 9.9% -22.1% 7.6% -18.9% 2.2% 14.6% 32.9% 0.5%

Deposits, foreign denom.

DemandNominal (in U.S.S) tN/A JN/A 5.5% 5.5% 5.5% 5.5% 5.6% 5.4% 5.5% 5.5% 5.7% 5.5% 5.8%Ex-post real (in NS) tN/A JN/A -18.5% -15.3% 32.0% 30.3% -14.7% -10.0% -24.4% -19.0% -7.7% 10.5% -41.4%

Term, 6 monthNominal (in U.S.$) tN/A JN/A 8.0% 8.0% 8.0% 8.0% 7.4% 7.5% 8.0% 11.9% 14.6% 13.1% 10.2%Ex-post real (in N$) tN/A #N/A -16.6% -13.3% 35.2% 33.4% -13.2% -8.2% -22.6% -14.1% 0.1% 18.5% -39.0%

Weighted average rate ofreturn on above assetsplus currency

Ex-post real (1) tN/A #N/A -43.7% -40.5% -21.2% 0.5% -27.4% -13.6% -28.9% -17.0% -4.2% 13.3% -33.4%Ri-post real (2) IN/A #N/A -43.7% -4Y.4% -20.8% 1.1% -26.9% -12.8% -28.2% -15.1% -1.5% 16.9% -31.7%

iing, domestic denom.-xed term up to 6 mo.)

AverageNominal IN/A IN/A #N/A IN/A IN/A IN/A 62.0% 76.6% 71.2% 68.1% 65.1% 59.8% 76.3%Ex-post real IN/A #N/A #N/A tN/A tN/A tN/A -3.1% 25.5% -2.6% 14.1% 25.9% 44.1% 6.6%

PrimeNominal #N/A #N/A *N/A #N/A IN/A IN/A 47.6% 65.8% 59.7% 49.9% 49.8% 46.5% 56.7%Ex-post real IN/A IN/A IN/A tN/A IN/A IN/A -11.7% 17.9% -9.1% 1.7% 14.2% 32.1% -5.2%

PreferentialNomLinal IN/A IN/A #N/A #N/A #N/A IN/A #N/A #N/A #N/A IN/A IN/A #N/A #N/AEx-post real IN/A IN/A #N/A #N/A IN/A IN/A #N/A IN/A #N/A IN/A #N/A #N/A IN/A

Lening, foreign deno .

AverageNominal (in U.S.S) IN/A IN/A IN/A #N/A IN/A #N1/A 12.0% 13.8% 14.2% 16.3% 18.5% 18.4% 18.2%Ex-post real (in NS) IN/A #N/A IN/A IN/A #N/A #N/A -9.5% -2.8% -18.1% -10.3% 3.5% 24.1% -34.5%

PrimeNominal (in U.S.S) IN/A #N/A #N/A #N/A IN/A #N/A lN/A IN/A #N/A 15.8% 17.4% 16.8% 17.2%Ex-post real (in NS) #N/A #N/A IN/A #N/A IN/A #N/A IN/A #N/A #N/A -11.1% 2.5% 22.4% -35.1%

Inflation (CPI) 6 monthsForward from December 20.4% 97.0% 85.8% 81.2% 58.2% 18.3% 67.2% 40.7% 75.8% 47.4% 31.2% 10.9% 65.3%

Devaluation 6 monthsForward from December 0.0% 139.0% 43.5% 45.4% 98.0% 46.1% 35.1% 20.1% 26.0% 13.1% 14.6% 16.2% -8.4%

Inflation (CPI) overPrior December 20.1% 35.2% 95.6% 77.6% 106.8% 67.3% 39.4% 57.6% 45.8% 83.1% 42.9% 29.3% 20.5X

Sources: Interest rate data are from Banco Central del Uruguay, boletin Estadistico.CPI and exchange rate data are fro the IMF, IFS, data tape.Weights for the weighted average are from Uruguay Table 2.

Notes: Real rates were calculated ex-post, t.e., they refLect the tiflation (and devaluation) rate six months furward fromDecember. Weighted average (1) assumes that all foreign assets are savings deposits; (2) assumes all time deposits.

a/ Since the real rates look forward six months from December, the 1981 figures do not reflect the November 1982 devaluation.

- 81-

UN.: Table 2

teves of Selected Fiandal Asets

zId Of Year(leyels in billias of Nw Peae or U.S.S)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 a/

Cwurrcy, dsticNakntnl 0.1 0.1 0.2 0.2 0.3 0.5 0.8 1.1 1.8 3.2 5.1 6.1 7.8In 1980 prices 7.8 8.5 10.4 5.8 4.4 3.9 4.7 4.2 4.7 4.5 5.1 4.7 5.0X of (DP #N/A #N/A 11.3X 5.9% 5.2X 4.6% 4.9% 4.5X 4.3t 4.3t 4.6% 4.8% 5.1%% of total assets #N/A #N/A 54.8% 39.6% 34.2% 28.8% 23.8% 19.6% 16.8% i5.6% 14.7% 11.7% 8.5%

Deposits, dcmstic orrency

DemarkNdmlrnd #N/A #N/A 0.1 0.1 0.2 0.4 0.5 0.8 1.3 2.6 3.6 3.6 3.6In 1980 pric FN/A FN/A 3.6 3.8 3.2 3.0 3.3 2.9 3.3 3.7 3.6 2.8 2.3X of GDP #N/A FN/A 3.9% 3.8% 3.7% 3.5% 3.4% 3.05 3.0% 3.6% 3.3t 2.8% 2.4%% of total assets FM/A #N/A 18.8% 25.6% 24.6% 21.7% 16.6% 13.31 11.8% 12.9% 10.5% 6.8% 3.9%

SavingN?kenal #N/A #N/A 0.0 0.0 0.1 0.1 0.3 0.4 0.8 1.5 2.7 3.9 4.3In 1980 prices FN/A FN/A 1.6 1.4 1.0 1.2 1.7 1.5 2.0 2.2 2.7 3.0 2.8X of C( FN/A FN/A 1.8% 1.4% 1.2% 1.4% 1.8% 1.6% 1.8X 2.1% 2.5% 3.0S 2.8%% of total assets #N/A #N/A 8.7% 9.6% 8.0% 8.7% 8.7% 7.2% 7.1% 7.6% 7.9% 7.4% 4.7%

TermNcoinal #N/A PN/A 0.1 0.1 0.2 0.3 0.6 0.8 2.4 5.3 11.4 15.1 15.5In 1980 prices FN/A FN/A 2.6 2.7 2.5 2.8 3.4 3.2 6.2 7.6 11.4 11.7 9.9X Of GOP FN/A #!/A 2.9% 2.7% 2.9% 3.2t 3.6% 3.4% 5.6% 7.2% 10.4% 11.8% 10.0%% of total assets FN/A FN/A 13.9% 18.31 19.5% 20.1% 17.1% 15.0% 22.2% 26.1% 32.9% 28.9% 16.8%

Subtotal dastic assets',ufasl /#NA FN/A 0.4 0.5 0.8 1.3 2.2 3.1 6.2 12.6 22.8 28.7 31.3In 1980 pri±es FN/A FN/A 18.2 13.7 11.1 10.8 13.1 11.9 16.2 18.1 22.8 22.2 20.1X of GP N/A FM/A 19.8% 13.9% 13.0% 12.7% 13.7% 12.5% 14.7% 17.2% 20.9% 22.3% 20.3t% of total assets #N/A FN/A 96.2% 93.1% 86.1X 79.2% 66.2% 55.1% 57.9% 62.2% 65.92% 54.8% 33.9%

Deposits. foreign currency

d and termNoidnal U.S.S F/A FMA 0.0 0.0 0.1 0.1 0.3 0.5 0.6 0.9 1.2 2.0 1.8Naninal, damtic #N/A FM/A 0.0 0.0 0.1 0.3 1.1 2.5 4.5 7.7 11.8 23.7 61.1In 1980 prices #NMA FM/A 0.7 1.0 1.8 2.8 6.7 9.7 11.8 11.0 11.8 18.3 39.2X of (IF FN/A FN/A 0.8% 1.0% 2.1% 3.3% 7.0% 10.2% 10.7% 10.4% 10.8: 18.4% 39.5%

Tbtal domestic ard foreignNrinal #N/A #N/A 0.4 0.5 0.9 1.6 3.3 5.6 10.7 20.3 34.6 52.4 92.4In 1980 prices FN/A FN/A 18.9 14.7 12.9 13.7 19.8 21.5 28.1 29.0 34.6 40.5 59.3% of DP #N/A #N/A 20.6% 14.9% 15.1% 16.0% 20.8% 22.7% 25.4X 27.6% 31.7% 40.7% 59.8%

CPI, December (Dec.80-100) 0.7 1.0 1.9 3.4 7.1 11.9 16.6 26.2 38.2 70.0 10D.0 129.3 155.9

Fzdunw rate, Demr 0.3 0.4 0.7 0.9 1.7 2.7 4.0 5.4 7.1 8.5 10.0 11.6 33.8

GIF, Dec. adjusted, ntmnal #N/A #N/A 1.8 3.4 6.1 10.2 15.9 24.8 42.2 73.6 109.1 128.7 154.5

Sources: Asset andl CUP data are fraa the Banao Central del Ungusy, Boletin Estadistico. Excharig rate ard CPI data are fran thehF, In, data tape.

Notes: The GW data have been logrithoically interpolated between the xirrent year and one year fcivard so as to be expressed inDecember prices.

a/ The draiutic rise in assets for 1982 is a result of a rtusive devaluation. This shld not. be interpreted as financialdeepenirg, since foreign lUabilities have also risen.

- 83 -

ANNEX 2

BANGLADESH

Government Intervention in Financial Markets

1. Since Independence, the authorities in Bangladesh have! assumed alarge and influential role in virtually every aspect of the country'sfinancial system. Shortly after December 1971, when Bangladesh wasformally severed from West Pakistan, many of the major enterprises werenationalized. Through the government's jurisdiction over the nationalizedcommercial banks and the central bank, the authorities controlled thecreation and allocation of nearly all formal credit. The demand forcredits was also largely under the control of the authorities, through thejurisdiction over the public enterprises. This situaticn has changed onlyslightly in recent years. Despite the growth in private demand for credit,claims on the public sector still represent 60-70% of domestic credit, downfrom 70-80% in the first half of the 1970's.

2. The monetary authorities at the central bank---The BangladeshBank--managed the quantity and allocation of credit with the standardpolicy instruments found in most developing countries. Among them werereserve requirements on the commercial banks' total deposits (5%) andliquidity requirements on demand and time deposits (25%), in the form ofcash, deposits with the Bangladesh Bank, and "unencumbered approvedsecurities,' i.e., government securities and debentures of either thepublic enterprises or financial institutions. Other poLicy instrumentsincluded an administered regime of interest rates and a comprehensive setof credit expansion ceilings for each financial institution. Theseceilings were issued on a quarterly basis for the commercial banks andannually for the Bangladesh Shilpa Bank, an industrial development financeinstitution, and for the Bangladesh Krishi Bank, an agriculturaldevelopment bank.

3. In addition to the above instruments, the Bangladesh Bank washeavily involved in the allocation of credit by way of its directed creditprograms. These programs designated priority sectors, imposed compulsorylending targets and established refinancing facilities. Among the sectorsreceiving preferential treatment in some form, were agriculture, smallindustry, exports, residential housing, transportation, public enterprisesand the private jute trade.

4. Although the regime of interest rate and credit ceilings wasby-and-large quite rigid, there was some limited moves toward financialsector liberalization recently. Since the institution-by-institutioncredit ceilings created rigidities which discouraged inter-bank transfers,the authorities recently introduced three categories of bank credits,making the credit ceilings more flexible and freeing-up this importantchannel for transferring resources among intermediaries. The threecategories of bank credits were: (1) advances not sub'Ject to theceilings--primarily government securities and public sector debentures;(2) advances that can be made in excess of the ceiling, if the bank canmeet certain liquidity requirements; and (3) advances l:hat are strictly

- 84 -

subject to the ceilings. The effect of establishing the second categorywas to allow banks with surplus funds to lend to banks with a surplus ofinvestment options, even though the recipient bank had extended its maximumallowance of credit. In light of the economy's heavy reliance on bankintermediation for savings mobilization this was a constructive change.

5. A large upward shift in the level of the interest rate ceilingswhich occurred in October of 1980 can also be viewed as an importantfinancial sector reform, one that was in response to market pressures andin the direction of market-determined interest rates. However, the factthat some lending rates were reestablished at levels below those paid ontime deposits is an indication of how far the system remains from marketdeterminated rates (see Table 1, also Tables 3 and 4).

The Development of Interest Rates

6. Despite the upward ratcheting of the nominal levels in theinterest rate regime, which is evident in Table 1, rates were fairly rigidover time. The nominally fixed rates, together with large changes in therate of inflation, produced wide swings in the real rates ex-post. From1971 up until the 1975 stabilization program, Bangladesh experienced highand rising inflation rates. The resultant highly negative real depositsand lending rates can be seen in Table 1, extending from the end of yearrates for 1971 through 1973. The 1975 devaluation of the Taka andreduction of the government deficits produced a deflationary period from1975 through 1976. As a result, the ex-post real rates swung highlypositive in December 1974 and December 1975. From 1976 to 1980 the realrates remained a few points negative, as inflation fluctuated between 8%and 18%. A sharp drop in inflation during 1982, produced relatively highpositive real rates at the end of 1981 (13.6% and 16.6%, for term depositsand normal credits, respectively), which declined as inflation acceleratedduring the first half of 1983.

7. Despite the predominance of negative real deposit rates,beginning in 1975 financial assets grew steadily, doubling in real termsfrom 1974 to 1982, and growing from 9.9% of GNP to 19.5% (see Table 2)There seems to be some evidence that remittances from Bangladesh nationalsworking in the oil producing regions and elsewhere played an important partto the growth in financial assets. Between 1974 and 1982 remittancestotaled about 30 billion Taka, versus an increase in financial assets ofabout 40 billion. However, the yearly increments in financial assets donot correlate well with the yearly remittances, suggesting that the linkageis not strong. In particular, an examination of the growth in financialassets reveals a pattern of steady but moderately accelerating growth.Remittances, on the other hand, had a very steep growth path. Thus, in theearly part of the period, annual remittances were much less than theincrement to financial assets, while, in the latter part, remittancessubstantially exceeded asset growth. In sum, the causal link betweenremittances and growth of financial assets is unclear.

- 85 -

BAUES: Table 1

Selected Nmur1nal and Ex-xst Real Interest Rates in Sche&aed BaTis

Erd of Year (arualized)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Deposits, domatic denu.

DwmedNcudral. #N/A O.% 0.1% O.% O.CZ% o.a O..% 0.0% O.C0 O.X% 0.% o..:1 0.0%EC-post real #N/A -0.2 -0.2 -0.4 0.1 0.1 -0.1 -0.1 -0.1 -0.1 -0.1 0.0 -0.1

Sirgs ( c )Nominal #N/A 4.5% 4.5% 4.5% 6.C% 6.a% 7.0: 7.0% 7.C% 7.0% 1O.CX 10.% 10.0%Er-post real N/A -18.4% -20.1% -32.4% 16.9% 11.9% -8.9% -0.8% -8.9% -6.3% -5.3% 10.6% -2.3%

Tenn, 6 munthNao.nal #N/A 4.8% 4.8% 4.8t 6.5% 6.5% 7.5% 7.5% 7.5% 7.5Z 13.0% 13.0 13.0%E -post real #N/A -18.2% -20.0% -32.2% 17.5% 12.4% -B.5% -0.4% -6.5% -5.8% -2.7% 13.6% 0.3%

Weighted averagpe rate ofreturn an above asstsplus currency

Ei-post real #NA -20.8% -22.6% -34.4% 12.6% 7.9% -12.4% -4.4% -12.3% -9.5% -8.5% 7.2% -5.0%

Iendirg, domstic denm.

NoD22Nominal #N/A 10.0% 104.0 10.0% 13.0% 13.0% 13.C0 12.aP 12.0% 12.0% 16.C% 16.0% 16.0%

E -post real #WA -14.1% -15.9% -28.8% 24.7% 19.3% -3.8% 3.0 -4.72 -1.9% -0.2% 16.6Z 3.

Agairst jute, jute gods

nd tea

Nominal #N/A 8.5% 8.5% 7.5% 10.5% 10.5% 10.5% 10.5; 10.5% 10.5% 12.0% 12.0% 12.0%

Ex-post real #N/A -15.3Z -17.1% -30.4% 21.9% 16.7% -6.0% 2.4: -5.9% -3.2% -3.6% 12.6% -0.6%

Inflatim (CPI) 6 mnxths

Fonard fran Decmer -4.1% 28.0% 30.9% 54.6Z -9.4% -5.3% 17.5% 7.9,2 17.5% 14.2Z 16.2% -0.6% 12.6%

Devaluatian 6 muths

Foraird frm December #/A 9.1% -17.3% -5.6% 185.6% 1.7% 7.8% 9.51 2.6% -11.4% 24.0% 24.1% 3.6%

Inflation (CPI) over

Prior Decsber #N/A 15.3% 46.1X 34.2% 76.1% -12.6% -0.1% 17.1X 9.6% 14.0% 13.2% 14.3% 4.9%

Souroes: Interest rates are fran the Bawgladesh Bak, Ecrmic Trends. CPI data are fran the IMF, IFS, data tape.

Weights for weighted average are fran Bargiadesh Table 2. (See note regardirg dispositian of savirgs.)

Notes: Real rates were calculated se-post, i.e., they reflect the inflation rate six ianths fDroard fran December. In 1977, 1978

and 1979, premiums were psid to rural depositors of .75 & 1.5 percentage points on savirgs and tenm deposits, respectively.

- 86 -

WaANd: Tale 2

Levels of Seected FinncIl masts

End of Year(in billia of Taka)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Qzecy, d sticNmdnal IN/A 2.1 2.9 3.2 4.0 3.7 3.9 5.0 6.6 7.2 8.7 9.7 9.9In 1980 prices YN/A 10.3 9.9 8.2 5.8 6.2 6.4 7.1 8.5 8.2 8.7 8.5 8.3% of GNP YN/A #N/A 5.9% 4.2Z% 2.9% 3.3t 3.5% 3.7% 4.0% 3.9% 3.9% 3.8% 3.6%% of total tI/A 37.8% 31.0Z 27.8% 29.4% 26.6X 22.0% 23.3 25.2Z 22.9% 23.1% 21.7% 18.7%

eposits, dotic currmy

DemiNominal #WA 1.8 3.9 4.9 5.4 5.5 6.8 7.5 8.8 10.2 10.6 12.3 14.8In 1980 prices #N/A 9.0 13.3 12.4 7.7 9.2 11.3 10.6 11.3 11.6 10.6 10.8 12.4% of G MN/A IN/A 7.9% 6.3Z 3.9% 4.9% 6.1% 5.4% 5.4% 5.4% 4.8% 4.9% 5.4%% of total sN/A 33.1% 41.7% 42.3% 39.1% 39.5% 38.8% 34.7% 33.5% 32.3 28.4% 27.5% 27.9%

TermN's1zl #N/A 1.6 2.5 3.4 4.3 4.8 6.9 9.1 10.8 14.2 18.2 22.8 28.5In 1980 prices MIA 7.9 8.7 8.8 6.2 7.9 11.4 12.9 13.9 16.0 18.2 19.9 23.7% of GNP #WA #I/A 5.2% 4.5% 3.12 4.2% 6.2% 6.6% 6.6% 7.6% 8.2% 9.0% 10.4%X of total asses IN/A 29.% 27.3% 29.9% 31.5% 33.9% 39.2% 42.1% 41.2% 44.8% 48.5% 50.8% 53.5%

Totalruiilral IN/A 5.5 9.3 11.5 13.7 14.0 17.6 21.6 26.2 31.7 37.5 44.9 53.2In 1980 prices IN/A 27.3 31.9 29.3 19.8 23.2 29.1 30.6 33.8 35.8 37.5 39.3 44.4% of GNP IWA IN/A 18.9% 15.C% 9.9% 12.4% 15.8% 15.7% 16.0Z 16.9% 17.0% 17.7% 19.5%

CPI, Deember (Dec.80-100) 17.4 20.0 29.3 39.3 69.2 60.5 60.4 70.8 77.5 88.4 10X.0 114.3 119.9

GNP, Dec., nauinal IN/A #N/A 49.3 76.8 137.9 112.8 111.3 138.0 163.9 187.8 221.0 253.9 273.3

Souroes: Asset data are fron the Bargladesh Bfnk, Banglad.h Bak lletin, and Econocic Trends. GNP data are fronBAngladesh Bureau of Statistics, Econumic Indicators of Bengladesh, and frmn the Bangladesh Bank, Economic Trends.CPI data are fran the IF, IFS, data tape.

Notes: GNP data are for the fiscal year beginning in the airrent year, therefore inccse is expressed in Decet er prices.

- 87 -

8. Overall, the interest rate regime was not suEficiently flexibleto dampen the movement of the real rates by adjusting the nominal rates toaccommodate swings in inflation. Although, some positive real ratesoccurred, these were only in deflationary periods.

The Pattern of Deposit Rates, End of Year 1982

9. Table 3 displays the nominal interest rates, as of December 1982,on savings and term deposits of various maturities. At first glance, itseems that the two percentage point spread between six month and three yeardeposits was inadequate to compensate asset holders for the additional riskof committing funds long term, particularly in light of the country'srecent experience with inflation rates in the teens. However, long termdeposits remained a very large share of all term deposits. In fact, termdeposits with maturities in excess of three years were roughly 60% of allterm deposits throughout the 1978-82 period, rising from 34% in 1975.Furthermore, the figures show an eleven-fold increase in deposits of 3+year maturities over deposits with 2-3 year maturities, where there is onlya 0.5 percentage point spread. This suggests that the explanation for thepopularity of the 3+ year term deposits involves issues beyond the interestrates. Precisely what these issues are and what part, if any, remittancesplay remains an open question.

Table 3

The Pattern of Deposit Rates, End of Year 1982

Savings Deposits Nomina].

with checking 8.5%without checking 10.0%

Time Deposits

3-6 months 12.0%6-12 months 13.0%1-2 years 14.0%2-3 years 14.5%3+ years 15.0%

Source: The Bangladesh Bank, Economic Trends.

The Pattern of Lending, End of Year 1982

10. The first observation that can be made regarding the lendingrates displayed in Table 4 is that many were at or below rates the paidon time deposits. Thus, even without accounting for reserve andliquidity requirements or the cost of intermediation, it appears thatthe lending rates were well below the marginal cost oE the funds and asubstantial amount of subsidization of borrowers was taking pilace.

- 88 -

11. As far as the structure of the lending rates is concerned, themost concessional rate charged, the rate on non-traditional export credits,was approximately two thirds of the rate charged for general lending. Sucha disparity may be questioned on efficiency grounds, particularly if asubstantial portion of the concessionary credits were merely diverted intodeposits paying rates higher than the borrowing rate. However, anassessment the size of the distortions introduced by the credit programsrequires a close examination of the returns on the actual uses of the fundsand those foregone by the diversion of the funds and the cost ofadministering the programs, including the subsidies.

Table 4

The Pattern of Lending Rates, End of Year 1982

nominal

Agricultural production 12.0%

Industry 14.5%

Specific industries in lessdeveloped areas 13.0%

Export credittraditional items 12.0%non-traditional items 11.5%

Loans for Socio-economicobjectives 13.0%

Loans given in the ChittagongHill Tracts 13.0%

General loans 16.0%

Source: The Bangladesh Bank, Economic Trends.

- 89 -

ANNEX 3

KENYA

Government Intervention in Financial Markets

1. The financial sector in Kenya is quite well developed incomparison to most countries in its income range ($341) U.S. pier capita in1982). It is vigorous and sophisticated and, as measured by the ratio offinancial assets to GNP, relatively deep (see Table 2). Institutionally,the financial sector contains a number of financial institutions of varioustypes, with some of the more advanced forms, such as insurance companies,having an importance not usually associated with countries at Kenya'sincome level. In contrast with many developing countries, thisinstitutional diversity is not associated with badly fragmented financialmarkets, but rather there appears to be a fair amount of competition acrossinstitutional types and between institutions in each category.

2. While the authorities have allowed market forces to play arelatively influential role in the financial system, the governmentmaintained a formidable presence in the financial market place. The mostimportant facets of the government's intervention in the financial sectorwere: (1) ownership of commercial banks, finance coumpanies, the largestpension fund, and an insurance company, which provided the government withextensive direct control over credit allocation; (2) a regime of minimuminterest rates on deposits and maximum lending rates; and (3) extensiveborrowing by the government to finance its deficit arLd to relend to theparastatal enterprises. Other important policy instruments used by theauthorities included such traditional measures as: reserve requirements,liquidity requirements and regulations governing the financialinstitutions' portfolio composition, including the share of depositscommitted to agricultural lending and the share held as low yieldinggovernment securities. The government also guaranteed many of the creditsextended to the parastatals, and placed some restrictions on foreign ownedenterprises' access to domestic credit.

3. By imposing ceilings on the lending rates the authorities reducedthe amount of credit that the financial intermediaries could profitablyextend by: (1) preventing intermediaries from charg:Lng a premium to coverthe additional costs and risks of term finance and oil lending to smallerborrowers and those with little collateral, thereby reducing such lending,and by (2) limiting the interest rates payable on deposits, therebysuppressing the mobilization of financial savings and thus restricting thepool of loanable funds. As a direct result, the commnercial banksmaintained loan portfolios composed primarily of short term credits to themajor private firms and the parastatals. While the finance companiesenjoyed somewhat looser regulation and were able to extend a larger shareof their loans to the smaller firms, the interest rate ceilings weregenerally binding on such lending and formal credits to these enterprisesremained far less than demanded.

- 90 -

4. The bias towards short term lending to the large firms and theparastatals has serious distributional and efficiency implications.Investments undertaken by the major firms tended to produce relativelyfewer low-skilled employment opportunities than investments by smallerconcerns, and tended to enhance the earnings of the groups which owned thefirms. In addition to the questionable distributional impact ofchannelling an exceptionally large share of the country's financialresources to the parastatals, these enterprises were also notoriouslyinefficient. Economic efficiency was further diminished by the scarcity oflong term credits. Despite the regularity with which short termobligations were rolled-over, the inability of firms to lock-in the termsof their borrowings greatly indreased the risk of undertaking projects withlong gestation or payback periods.

5. The volume of borrowing by the government also threatened to holdback Kenya's growth potential by placing a severe strain on the financialsystem and shifting the allocation of the country's resources excessivelytowards the public sector. Between 1976 and 1982, government expendituresrose dramatically, from approximately one quarter to one third of GNP. Alarge fraction of this growth came through the expansion of theparastatals, financed by government intermediation between the publicenterprises and the financial system and foreign capital markets--oftenwith negative spreads. By engaging in intermediation on such a largescale, the authorities, in effect, established a system of directed creditcomparable to that which many developing countries implement throughregulation of their banking systems. This led to substantial crowding outof private investment, through both the reduction in available credit andthe preemption of private investments opportunities by the parastatals.

6. The massive public sector financing requirements are also aserious impediment to interest rate reforms. While the recent increase ininterest rates raised the allowable returns on private deposits and loans,it also placed upward pressure on the yields the government had to pay onits securities, raising government debt service and increasing pressures onthe fiscal deficit. This presents the authorities with a serious dilemma,since the abrupt decline in the country's access to foreign credits (due toKenya's large foreign debt of $2.4 billion in 1982, and the generalconstriction of credits to the developing countries), has placed a premiumon mobilizing domestic resources and allocating them more efficiently,suggesting the need for further interest rate liberalization.

7. As noted, such reforms would however increase the public sectordeficit, forcing the authorities to resort to either: (1) increasedinflationary finance, or (2) increased forced holding of governmentsecurities, or (3) a reduction in public expenditures. Each of thesepolicy options has its costs. Increased inflation would jeopardize thecountry's tradition of relative price stability, upon which the successfuldevelopment of the financial sector has depended. An increase in forcedholding of government securities at below market rates would increaseintermediation costs, possibly resulting in high real lending rates and/orlower real deposit rates. This would tend to depress the economy furtherand partially defeat the purpose of the initial reforms--increased domestic

- 91 -

resource mobilization. Finally reductions in public expenditures wouldresult in lower short term employment levels, if cuts were made in currentexpenditures, or lower long term growth, if the cuts rere out ofdevelopment expenditures. Apparently, the authorities have chosen to limitthe increases in the level of interest rates, lower ea;penditures primarilythrough lower development expenditures, and resort to some inflationaryfinance. Clearly, this solution is going to be costly in the long term ifit is maintained, and the government is faced with a great challenge to itspolitical and management skills, a challenge with very serious consequencesfor the country's current and future welfare.

The Development of Interest Rates

8. Since 1977, the low ceilings on lending rates provided littleincentive to raise deposit rates much above the minimuLm set by thegovernment. These minimums, which are displayed in Table 1, werepredominantly negative in real terms ex-post. Only recently did thispattern change, due to the combined effect of declining inflation andhigher minimum deposit rates. Table 1 shows that at l:he end of 1982 termdeposits and savings accounts in the commercial banks and private financecompanies yielded positive returns ex-post. In addition to the upwardadjustment in the deposit rate ceilings, the maximum nominal lending rateswere increased over the period, reaching 16.0% in 1982.

9. The ceilings on lending rates implied slighl:ly negative realrates on average over the period. The evidence also suggests that theeffective rates were commonly increased a couple of percentage points abovethe ceiling rates by compensating balances and various other adjustments tothe repayment schedules. This was true particularly on loans from the lesswell regulated Finance Companies, which accounted for a growing share ofthe market. However, these effective rates were not sufficient toencourage substantial lending at term or to non-blue chip entities.

10. A comparison between the rates on loans for land purchases fromthe Agricultural Finance Corporations and the ceiling rates on generalcredits from the commercial banks reveals that the land purchase loans wereonly a couple of percentage points lower than the general category. Thusthese rates were also only slightly negative in real terms on average, andwere not exceptionally concessional. However, the terms of these loans mayhave been considerably easier than those on general credits, making thespread in effective rates somewhat wider and the land purchase lendingcorrespondingly more concessionary.

11. Although the ceilings on lending rates were increased, theminimum rates for deposits were increased more. Thus the spread betweenterm deposits and the ceilings on general credits decLined from 4.6% in1979 to 2.8% by 1982. This decline suggests that the profitability ofintermediation was reduced, and financial institutions increasingly had toresort to devices such as fees and compensating balances to maintainprofits. This squeezing of margins tends to restrict formal finance andencourage expansion of less efficient forms of informal or self-finance.

- 92 -

5el1-t1 -1m1 -d at-Pm 1al Iztamt hzt

VId of Yer (amnzlizmd)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1961 1982

Iqpaits, uinim rat

camdral BwsDean

Nomd .Oa O.O O.OX O.Ot 0.0% O.2 O.O O.0 O.OX O.X O.X O.X 0.02Er-pot ral -3.3X -3.72 -9.13 -17.0X -22.2X -11.OX -21.2X -15.72 -5.72 -13.2X -11.2X -13.0X -9.5X

1 3.0X 3.0 0 3.02 3.0 X 3.02 5.02 5.0 X 5.0 X 5.02 5.02 5.02 6.02 1O.02 12.5XbEpt real -0.42 -0.8X -6.62 -14.52 -18.3X -6.6X -17.2X -11.52 -0.92 -6.92 -5.8X -4.31 1.9X

Term, 6 wnth?N1M1 3.8X 3.8X 3.62 5.4 5.4X- 5.4X 5.4X 5.4X 5.4X 5.4X 6.8X 11.02 13.2%Ex-ixt real 0.31 O.OX -5.92 -12.5X -18.02 -6.2X -16.92 -11.12 -0.62 -8.52 -5.2X -3.5X 2.52

Private F1znal Irstitutia

Nmil OaO OaO O.O OaO OaO OaO O.OX OaO una oao oao oao o.oxrEx.t rea -3.13 -3.7X -9.3 -17.0 -22.2X -11.0 -21.2 -15.7X -5.7X -13.2% -11.2X -13.02 -9.5S

:m l 3.02 3.02 3.02 3.02 5.0 X 50 5.0X 5.02 5.02 5.02 8.02 10.02 12.52b-9It ral -0.4X -0.62 -6.6X -14.5X -18.3X -6.62 -17.2% -11.52 -0.9X -8.92 -4.02 -4.3S 1.92

Term, lonpt tegryb-pot real 2.52 2.12 -3.92 -10.86 -16.4X -4.42 -14.92 -8.31 2.62 -5.6X -1.42 -0.4S 5.2%

Pot Office &vLqp B -k

N 1 3.0kI 3.0 3.0 3.02 3.02 3.02 5.02 5.02 5.02 5.0X 5.02 6.02 10.02 10.0%bE-pot real -0.42 -0.6 -6.6X -14.52 -19.92 -6.62 -17.2% -11.52 -0.92 -8.9X -5.8% -4.3X -0.4X

Agtedt erate of

..- turn an aw aset

i-put ral -1.4X -2.72 -6.62 -14.62 -18.72 -9.72 -16.5X -13.8S -4.02 -10.0% -7.6S -8.2% -2.8S

!-Ai ,detic dunce.

comwial ak~

md.un (less than 3 yr.)Nra - - - - - - - 10.02 10.0 10.02 11.0 14.02 16.0Xb-pot real - - - - - - - -7.2S 3.8X -4.5Y -1.4% -0.9S 5.02

Nkualm 7.02 7.02 7.02 7.02 8.02 8.02 - - - - - - -b-put real 3.5X 3.1U -3.02 -11.2X -16.02 -3.9X - - - - - - -

11 ~ ~ Finalce GDrp 7.5Z 7.52 7.52 7.52 8.02 8.0Z 9.0Z 9.0Z 9.0Z 9.0a 9.0X 12.02 12.02EM-Wet real 4.02 3.62 -2.5Z -10.82 -16.02 -3.9% -14.12 -8.12 2.82 -5.4Z -3.22 -2.6% 1.42

Inflati (CPI) 6 nthsForward frtm Dber 3.42 3.82 10.31 20.5X 28.5Z 12.42 26.8Z 18.6Z 6.0Z 15.2% 12.6Z 15.02 10.5%

1evaluatime 6 onthsForward from Deilber 0.0 0.02 -6.72 7.2% 0.02 4.2Z -0.2% -3.5Z 2.0% -0.5% 35.Za 13.9% 8.3t

Infladtin (CPI) overPrior De-i- 1.52 7.2X 3.31 15.2% 16.0Z 20.3. 7.6% 21.02 13.7Z 9. U 13.1U 19.3X 13.3%

9swces: Interest rates are frcm The Central B*k af Kenya, E mic and Ftn Review, rual Report, and reaj ofStatistics Finrwe ad Plaug, Sey. CPI data are irr the IMF, IFS. data tape. ltigts for wigitedawras are frmm Kenya Table 2.

Notes: Raal rates wm cal-atd ex-Wet, i.e.,* they reflect the inflation rae six mnths forw.rd frum December. lhe aweraerate of retuzr wo calojlaced asmadTi asch set yielded the ldnImn rate and that all term deposits in the PrivateFinal istituticn were In the lmngat term catagory. '-' indicates rot defined.

- 93 -

In fact, the formal financial sector appears to have contracted slightly(See Table 2). The ratio of total financial assets to GNP declined fromapproximately 34.8%, in 1979, to 32.7% by the end of 1982.

12. Alternative explanations for the decline in the ratio offinancial assets to GNP between 1979 and 1982 include: (1) portfolioadjustments the post-coffee boom years; (2) capital flight associated withthe growing economic crisis; and (3) a drawing-down of parastatal depositswhich are not netted out of the asset data. In contrasi: to that decline,Table 2 reveals a sharp upward shift in the ratio of tot:al financial assetsto GNP from 27.3% in 1976 to 33.4%. in 1977, and 34.7% in 1978. This shiftcorresponds to large foreign exchange inflows associated with the coffeeboom in these years and probably reflects a substantialLy more optimisticeconomic outlook. The softening of the economic picture in the post-boomyears could then have accounted for the subsequent declLne in the ratioover 1979 to 1982. The growing economic crisis and the associated internaland external imbalances could also have been responsible for the decline inthe ratio, as agents with access to foreign asset markets--legal orotherwise--shifted capital offshore, thereby reducing the relative amountof domestically mobilized resources. Finally, the paraatatals, under thepressure of rising losses, drew down accounts built-up during earlieryears. Thus, a variety of factors contributed to the reduction in thedepth of the formal financial system.

The Pattern of Deposit Rates, End of Year 1982

13. At the end of 1982, the official minimum deposit rates displayedin Table 3 indicated a rather small spread for maturity--only onepercentage point between deposits of one month and deposits up to twentyfour months. As discussed in paragraph 8, the ceiling on lending rates andthe cost of intermediation combined to limit the rates financialintermediaries paid on short term deposits to roughly the minimums. Givenuncertainties over future interest rates, banks also had little incentiveto offer rates above the minimums on longer term deposits. In combinationwith a liquidity squeeze, this caused a switch into maturities of almostexclusively less than three months.

14. A couple of difficulties arise from a predominantly short termdeposit base. First, it discourages term financing, since banks must keepa shorter loan portfolio to match the greater volatility of theirliabilities (Of course, firms may not desire term finance if there is areasonable chance that interest rates will decline). Second, the bankssuffer greater exposure to potentially destabilizing out-flows of funds.This can have serious repercussions for the entire financial system if thesystem does not have adequate institutional resources to cope with a majorbanking crisis. Since the short maturity of the deposit base can berelated to the lack of an adequate premium for long term deposits, which inturn can be traced partly to the lending ceilings, these dangers add weightto the argument for allowing interest rates to move towardmarket-determined levels.

- 94 -

KgJYA: Table 2

Levels of Selected Financial Assets

Fnd of Year(in biLLions of %ditlins)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 198.) 1981 1982

Currency, drmsticC dmnal 0.7 0.7 0.9 1.0 1.1 1.2 1.6 2.2 2.3 2.7 3.0 3.6 3.7In 1980 pric 2.3 2.2 2.6 2.5 2.4 2.3 2.8 3.1 2.8 3.0 3.0 3.0 2.8X of GNP 5.82 5.6% 6.0% 5.62 5.2% 4.9% 5.2% 5.82 5.62 5.7% 5.5X 5.7X 5.2%% of total assets 30.2% 19.5% 20.1X 17.8% 18.1% 17.3% 18.9% 17.5% 16.0% 16.3% 16.62 17.3% 15.8%

Deposits, danstic currency

Cmrcial Bark

DemnsdNomdzul 1.2 1.3 1.5 2.0 2.1 2.4 3.0 4.3 4.7 5.6 5.2 5.6 6.0In 1980 prics 3.9 3.9 4.5 5.0 4.7 4.4 5.1 6.1 5.8 6.3 5.2 4.7 4.4X of 1W 9.9% 9.7Z 10.2% 11.2% 10.1% 9.72 9.5% 11.62 11.42 11.8% 9.62 9.0% 8.3XZ of total assets 34.4% 34.0% 34.12 35.5% 35.4% 34.0% 34.7X 34.8% 32.8% 33.8% 28.6X 27.1% 25.5%

SavinpNoinal 0.7 0.8 0.9 1.0 1.2 1.3 1.5 2.1 2.4 2.7 3.0 3.5 3.9in 1980 prices 2.2 2.3 2.5 2.6 2.6 2.4 2.6 2.9 3.0 3.0 3.0 2.9 2.9X of 1W 5.77 5.72 5.8% 5.77 5.62 5.2X 4.9% 5.62 5.8% 5.6% 5.4% 5.6X 5.5%% of total assets 19.9% 19.9% 19.4% 18.2% 19.6% 18.3% 17.8% 16.8% 16.62 16.2% 16.2% 16.9% 16.72

TermNminal 0.5 0.6 0.6 0.9 0.8 1.1 1.4 2.3 3.0 3.1 3.7 4.1 5.6In 1983 prices 1.7 1.8 1.8 2.2 1.8 2.0 2.3 3.3 3.6 3.5 3.7 3.5 4.2X of GMW 4.52 4.4% 4.2% 4.9% 3.9% 4.5% 4.4% 6.2Z 7.1% 6.5% 6.8% 6.62 7.8%% of total assets 15.4% 15.5% 14.0% 15.7% 13.72 15.72 16.0% 18.62 20.5% 18.62 20.4% 20.0% 24.0%

Private Financial Inst.

Dem3WNominal 0.0 0.0 0.0 n.0 0.0 0.1 0.0 0.3 0.2 0.3 0.1 0.2 0.2In 1980 prices 0.1 0.1 0.0 0.0 0.0 0.2 0.0 0.5 0.2 0.3 0.1 0.2 0.22 of GNP .3X% 0.2% 0.1% 0.1% 0.0% 0.4% 0.1% 0.9% 0.4% 0.5% 0.2% 0.4% 0.3%7 of total assets 1.1% 0.62 0.4% 0.3% 0.1% 1.3% 0.3X 2.8% 1.3% 1.6% 0.5% l.U% 1.0%

Savings a/snir 0.1 0.1 0.1 '3.2 0.2 1.1 0.3 0.2 0.5 0.5 0.6 0.6 0.6

2 of (W 0.62 0.8% 0.9t 1.l1 1.0% 0.67 0.9% 0.62 1.1% 1.0% 1.12 0.9% 0.9%2 of total assets 2.0% 2.6X 3.1% 3.5% 3.5% 2.1Z 3.M 1.9% 3.3X 2.72 3.2X 2.8% 2.7%

Tetm a/Sni~nal 0.1 0.2 0.3 0.4 0.4 0.7 0.6 0.8 1.2 1.5 2.3 2.6 2.9In 1980 prices 0.4 0.6 0.8 1.0 1.0 1.2 1.0 1.1 1.4 1.7 2.3 2.2 2.1% of 1W 1.1% 1.4% 1.9% 2.12% 2.1% 2.72 2.0% 2.1U 2.8% 3.Z% 4.2% 4.2% 4.0%Z of total assets 3.8% 5.0% 6.2g 6.8% 7.37, 9.3% 7.2X 6.37 8.0% 9.3% 12.72 12.7% 12.2X

Post Office Swins BankNoinal 0.1 0.1 0.1 0.1 0.L 0.1 0.2 0.2 0.2 0.3 0.3 0.4 0.5In 1980 prices 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.3 0.4 0.4Z of GNP 0.9% 0.8% 0.8% 0.7% 0.7% 0.6% 0.5% 0.5% 0.5% 0.5% 0.62 0.7% 0.722 of total assets 3.U% 2.9% 2.6% 2.32 2.3% 2.07 1.9% 1.5% 1.5% 1.5% 1.9% 2.UL 2.L%

TotalNridnal 3.4 3.8 4.4 5.5 6.0 7.1 8.6 12.5 14.4 16.4 18.3 20.6 23.6In 1980 prices 11.2 11.5 13.0 14.1 13.2 13.0 14.6 17.5 17.8 18.6 18.3 17.3 17.4X of 11W 28.9% 28.5X 29.8% 31.4% 28.6% 28.5% 27.3% 33.4% 34.7% 34.8% 33.4% 33.2% 32.7%

CpI, D tember (Oec.80-100) 30.8 33.0 34.1 39.3 45.5 54.8 58.9 71.3 81.0 88.4 [D)0.O 119.3 135.2

GNF. Dec. adjusted, noinal 11.9 13.1 14.9 17.6 21.0 25.0 31.4 37.4 41.5 47.2 54.8 62.2 72.1

Sources: Asset data are frxau tle Cencrai Bark oft Keya, Economic arv3 Finrmcial Review.GNP acd CPL data are frmn the LIF, LFS, data tape.

Notes: I-re GE data have been I ntr icAWly interpoxlated betven the Lurrenc year ivJ ale year forward so as to be expressed inDecmber prices.

a/ Savinp and term deposits in the Private Financial Institutions were asmsed to be distributed 1:2 for the years 1970-72,and in all other years as reporced.

- 95 -

Table 3

Commercial Bank Minimum Deposit RatesL

On Account Less than Shs 1 million, End of Year 1982

Nominal

Savings deposits 12.5%

Term deposits

7 day call freeone month 12.5%1-3 months 12.5%3-6 months 12.7%6-9 months 13.2%

9-12 months 13.4%12-24 months 13.5%

Source: Central Bank of Kenya, Economicand Financial Review.

The Pattern of Lending Rates, End of Year 1982

15. The structure of lending rates in Kenya at the end of 1982 doesnot appear to have been particularly complex nor highly distortionary, asevidenced by the array of lending rates displayed in Table 4. Outside ofthe rates established for direct credits from the Central Bank, the numberof specified lending categories and rates were few and the spreads rela-tively small. Thus, while other "non-price" rationing mechanisms may haveintroduced substantial distortions, the margins within the regime ofinterest rates do not generate the massive subsidies typically found instatutory interest rate regimes. This is not to say that subsidization wasabsent, but merely that interest rate subsidies and the distortions theyproduce were low in comparison to most developing countries.

16. The agricultural sector benefited most from the preferentiallending rates, i.e., the rates on rediscounts and advances from the CentralBank and loans from the Agricultural Finance Corporation. In addition, theKenya authorities required the commercial banks to commit 17% of their netdeposit liabilities to agricultural lending; the corresponding figure forthe non-bank financial institutions was 10%. However, the evidencesuggests that enforcement of these requirements was loose and thatcompliance was less than full. Evidence also indicates that much of the"agricultural" lending was, in fact, only marginally related to agricul-ture, and that there was substantial diversion of agricultural credits fromthe agricultural sector. This suggest that targetted lending in reasonably

- 96 -

well integrated capital market is likely to be limited in effectiveness orto require large expenditures to monitor. Moreover, even monitoring couldnot prevent potential investors in targetted sectors from borrowing subsi-dized funds to replace their own and subsequently investing their fundselsewhere.

Table 4

The Pattern of Lending Rates, End of Year 1982

Central Bank of Kenya

Rediscount Rate for T-Bills 13.48%Advances against T-Bills 14.50%Bills and Notes Under Crop Finance Scheme

Rediscounts and AdvancesCSFS & AFC 13.75%Finance of Export Bills 14.00%

Other Bills and NotesRediscounts 14.50%Advances 15.00%

Advances against Government Securities 14.50%

Kenya Commercial Banks

Loans and Advances (minimum) free(maximum) 16.00%

Other Financial Institutions

Agricultural Finance CorporationLand Purchase Loan 12.00%Seasonal Crop Loan 14.00%All other loans 13.00%

Hire Purchase Companies and Merchant BanksLoans 16.00%

Building SocietiesLoans 16.00%

Source: Central Bank of Kenya, Economic and Financial Review.

17. In sum, the financial system in Kenya is one of the most marketoriented in the developing countries and among the freest from interestrate induced distortions. However, the size and growth of public sectorborrowing has placed a large strain on the system, one that threatens tolimit its development.

- 97 -

ANNEX 4

NIGERIA

Government Intervention in Financial Markets

1. The Nigerian authorities' intervention in the financial marketswas among the most extensive of all the market-oriented developingcountries, during the 1970-82 period. In addition to extensive use of suchcommonly encountered policy instruments, as administered interest rates,directed credit, minimum reserve and liquidity ratios, &Lnd mandatory bankholdings of government securities, the Nigerian government exercised directand indirect control over a large portion of the country's financialresources and financial transactions. This was primarily due to thegovernment's control over the country's oil revenues and. its ownershipposition in the commercial banks.

2. The public sector used the country's oil revenues to undertake amajor fraction of total investment, by-passing the financial sector. Inaddition, the government used its privileged borrowing position to engagein extensive intermediation between sources of foreign and domestic capitaland the Nigerian public and private sectors. Finally, the authoritiesmaintained extensive indirect influence over the financial resources notunder their direct control by virtue of the government's ownership of mostof the large commercial banks. In short, the authorities exertedsubstantial control over virtually all of the financial savings availableto the economy.

3. The government also intervened in the financial markets through awide ranging and complex system of directives and guideLines. The monetaryauthority annually established a regime of interest rates, which coveredall deposit and lending rates in the formal financial market. This regimeincluded a maximum and minimum lending rate and a set oi differential ratesinvolving a number of preferential categories.

4. Policy goals relating to the financial sector typically werepursued through credit allocation guidelines or specific lendingdirectives. The Central Bank of Nigeria issued guidelines to thecommercial banks specifying, among things, the aggregate credit expansion,the share of credit allocated to each of 16 sectors into which the economywas divided, the share of credit allocated to indigenous borrowers (whichis further subdivided by size of enterprise and other criteria), and theshare of credits by term. So complex and internally inconsistent was thesystem that it defied compliance and violations occurre<l regularly.

The Development of Interest Rates

5. Ceilings on nominal interest rates in Nigeria were set belowinflation rates, which averaged about 15% during the 1970 to 1982 period.In fact, none of the rates paid on deposits yielded posLtive returnsex-post during the entire period, judging from Table 1, which displays aselection of deposit and lending rates in Nigeria from :L970 to 1982. This

- 98 -

NICZL: Tal 1

Selected NIrAl ad E-ost PRal Inteest Pates

Ehd of Year (ar.aliz)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Deposits, daimtic darm.

DodNoinal 0.0% 0.0% O.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.OS 0.0%EK-pat real -27.0% -6.7% -20,8% -13.1% 42.6% -14.7% -28.9% -17.9% -13.3% -8.5% -25.4% -7.9% -23.8%

SaViSNominal 3.0% 3.0% 3.0% 3.0% 3.0% 4.0% 4.0% 4.0% 5.0% 5.0% 6.0% 6.0% 8.5%

E-post real -24.8% -3.9% -18.4% -10.5% -40.9% -11.3% -26.1% -14.6% -9.0% -4.0% -20.9% -2.4% -17.3%

Term, 6 lmathNodinal 3.0% 3.0S 3.5% 3.5% 3.5S 3.0% 2.5% 3.0% 5.0% 5.5% 6.0% 6.0% 8.5%Exrpost real -24.8% -3.9% -18.0% -10.1% -40.6% -12.2% -27.2% -15.5% -9.0% -3.5% -2D.9% -2.4% -17.3%

Weited average rate ofreturn on abow aoete

E-ost real iS/A -5.6% -19.8% -11.9% -41.7% -13.5% -28.1% -17.0% -11.8% -6.7% -23.8% -5.8% -21.2X

Lendirg. donvutic dam.

Gxera adverocaNominal 8.0% 10.0% 10.0% 10.0% 10.0% 9.0% 10.0% 6.0% 11.0% 11.0% - - -Erpost real -21.1% 2.7% -12.9% 4.4% -36.8% -7.0% -21.8% -13.0% -3.7% 1.5% - - -

First class advasN 1dxul 7.0% 7.0% 7.0% 7.0% 7.0% 6.0% 6.0% 6.0% 7.0% 7.5% - - -Er-pot real -21.9% -0. 1% -15.3% -7.0% -38.6% -9.6% -24.7% -13.0% -7.2% -1.7% - - -

Geerd sNonal - - - - - - - - 11.0% 11.0% 11.5% 11.5% 14.0%

Mrpost real - - - - - - - - -3.7% 1.5% -16.8% 2.7% -13.2%

Perfered sector danN;ial - - - - - - - - 9.0% 9.0% 9.5% 9.5% 12.5%Er-ot real - - - - - - - - -5.5% -0.3% -18.3% 0.9% -14.3%

Iflation (CPI) 6 ntmsForward from Deceuber 37.0% 7.1% 26.3% 15.1% 74.2% 17.2% 40.7% 21.8% 15.3% 9.3% 34.1% 8.6% 31.3%

Devlduatim 6 wtbwForward from Deoer 0.0% 0.0% 0.0% -12.0% -3.6% 0.0% 5.9% -2.5% -15.0% -6.4% 40.9% 12.5% 21.8%

Inflation (CPI) overPrior Decmber 13.0% 15.0% -3.5% 17.8% 9.9% 43.1% 12.4% 31.3% 10.3% 11.5% 13.7% 17.4% 6.7%

Souree: All pre-19 8 2 interest rates are frm the Central Bak of Nigeria, oc ad FA al Paviw or A worterept the general da, and prefered sector _aue lening rates ad all 1982 figes, bidch are from WorldBan data. CPI dta are from tte IM, PdS t tse.

Notes: Real rates were olodated ox-post, i.e., they reflect the inflation rate six nxtlu forward from Decmr.The weigIted averap ures weigts frm NiWeria Table 2 ad asas all tine deposits yield at the 6 axth rate.'-' iLicates not defined.

- 99 -

interest rate policy reflected a view held in the government that with itsoil revenues the country had little need to mobilize domestic savings. Onthe lending side, the rates charged for general credits were positive inreal terms only in three of thirteen years, in 1971, 1979 and 1981. Onlyonce, in 1981, did the real rate charged on preferential credits becomepositive ex-post.

6. Despite the repressed level of deposit rates, the holdings offinancial assets grew rapidly between 1971 and 1982. The total ofcurrency, demand, savings and time deposits rose from 13.7% of GDP in 1971to 36.7% of GDP in 1982, an average annual growth rate of over 1]2% (seeTable 2). This growth can be attributed to steady increases in themonetization of the economy, as well as unsterilized capital inflows(1974-76, 1979-80) and monetization of the government deficits (1977-78,1981-82).

The Pattern of Deposit Rates, End of Year 1982

7. At the end of 1982 the interest rate spreads between deposits ofshort and longer maturities were very small, as displayed in Table 3. As aresult, the term structure of bank liabilities was quite short, with about80% of total deposits having maturities of less than one year in 1981.In concert with the low ceiling rates on lending, the shortness of bankliabilities discouraged long term finance. This could become a matter ofconcern, if the decline in oil revenues is sustained and. the private sectoris then expected to increase its total investment.

- 100 -

NIaIA :Tble 2

levels of Selected FTnnal Amets

W of Year(in bficm Of d Kira)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Oiry do-iNowiml If/A 0.4 0.4 0.4 0.6 1.0 1.4 1.9 2.2 2.4 3.2 3.9 4.2In 1980 prim AN/A 1.3 1.5 1.4 1.7 2.1 2.5 2.7 2.7 2.7 3.2 3.3 3.42 of arP #WA 4.8% 4.6% 3.52 3.1U 4.5% 5.U1 6.8% 6.4% 5.7% 7.32 8.72 9.21X of total assets ft/A 34.92 33.11 30.8% 26.4% 28.52 25.62 27.52 29.21 23.92 22.02 24.8% 25.tM

Dposits. dmsic awrowy

DesdNm1 fK/A 0.3 0.3 0.4 0.6 1.0 1.9 2.9 2.6 3.8 6.0 5.9 5.8In 1980 prime *N/A 1.1 1.2 1.3 1.8 2.1 3.6 4.0 3.3 4.3 6.0 5.0 4.72 of UP EN/A 3.9% 3.8% 3.21 3.32 4.4% 7.3% 10.1U 7.8% 9.1U 13.9 13.3% 12.7%2 of total ate tN/A 28.11 27.11 27.7% 28.21 28.02 36.8% 40.42 35.5X 38.52 41.8% 37.8% 34.5%

S-INkbai fN/A 0.2 0.2 0.2 0.3 0.5 0.7 0.9 1.1 1.3 1.7 2.0 2.3In 1980 prim ft/A 0.6 0.8 0.7 0.9 1.1 1.3 1.3 1.4 1.5 1.7 1.7 1.9

of GDP #/A 2.21 2.42 1.92 1.6% 2.3X 2.72 3.3X 3.21 3.11 3.8% 4.52 5.0X2 of total Jeet N/A 16.2 17.3 16.2% 13.5% 14.62 13.62 13.3 14.72 13.1% 11.5% 12.8% 13.7%

N,uLml ft/A 0.2 0.3 0.4 0.7 1.1 1.3 1.3 1.5 2.4 3.6 3.8 4.5In 1980 prim fN/A 0.8 1.0 1.2 2.0 2.2 2.3 1.9 1.9 2.7 3.6 3.3 3.62 of GP #/A 2.92 3.11 2.9% 3.7X 4.6% 4.8% 4.7% 4.5% 5.8% 8.11 8.62 9.8%

TotalNidml AN/A 1.0 1.2 1.4 2.2 3.6 5.3 7.1 7.4 9.9 14.5 15.5 16.9In 1980 price f/A 3.8 4.4 4.6 6.4 7.5 9.7 9.9 9.4 11.2 14.5 13.2 13.52 of GP ft/A 13.7% 13.9% 11.4% 11.6% 15.9% 19.8% 24.92 21.8% 23.7% 33.3 35.2% 36.72

CPI, Dmber (ec.D80-1ID) 23.6 27.1 26.1 30.8 33.9 48.4 54.4 71.5 78.8 87.9 100.0 117.4 125.2

CDP, De.adjusted, ,l 6.3 7.4 8.3 12.4 18.6 22.8 26.7 28.4 33.9 41.6 43.4 44.2 46.0

Soures: Asset data are fromn the (ntl Bark of NIgeria, Mtmnd-y Report. CI and GDP data are from the NF, IS, data tape.

Notes: The GNP data hrae bow logritthicaily interpolatei benie the cuirrt year and cn year forward so as to be expresed inDewber primas.

- 101 -

Table 3

Deposit Rates Effective End of Year 19B2

Nominal

Savings deposits 8.5%

Time deposits7 day notice 7.5%one month 8.0%1-3 months 8.25%3-6 months 8.5%6-12 months 8.75%over 12 months 9.0%

Source: World Bank, Financial Intermediation in Nigeria.

8. Capital flight is another problem which arises out of the overalllow level of the deposit rates, vis-a-vis inflation and vis-a-vis ratespaid in the international markets. Although the authorities administereda system of exchange controls, the large size of the ext:ernal sectorpresented plentiful opportunities for capital flight, such as over-invoicing of imports.

The Pattern of Lending Rates, End of Year 1982

9. All of the lending rates at the end of 1982 were highly negativeex-post, given a December 1982 to June 1983 inflation rate of 31.3%. Thus,despite the upward adjustment of the nominal rates since of 1981, reallending rates continued their historically negative reaL levels.

10. The negative real rates were, however, only a small part of amuch larger problem relating to the efficiency of the whiole financialsystem. The heavy reliance on a wide ranging regime of quantity controlsintroduced a series of distortions into credit allocation. Moreover, thelevel and structure of lending rates, in comparison to deposit rates,retarded the growth of financial intermediation by limiting the profitabi-lity of intermediation. In addition, with a six month term deposit rate of8.5% and concessional rates of 7.0% to 8.0%, there was a distinct possibi-lity that funds targetted for investment in the priority sectors were beingdiverted into deposits or other uses. The likelihood oE such leakages wasparticularly high, given effective interest rates in informal creditmarkets which ran as high as 100%.

11. Such high informal rates also indicate a demand for credits byborrowers who were excluded from the formal markets by its ratJioningmechanisms. Their willingness to pay such high rates also indicates theexistence of investment opportunities with rates of return well above thelending rates in the formal sector. These rates cast doubt on the

- 102 -

efficiency of the formal credit allocation scheme. Furthermore, theefficacy of the credit allocation scheme in achieving social or economicobjectives can also be questioned, given evidence that a large portion ofdirected credit wound up in the hands of either the large enterprises, orlarge farmers, who had access to credit through ordinary channels in theformal credit markets and tended to invest in capital intensive productionmethods.

Table 4

Lending Rates Effective End of Year 1982

Nominal

General minimum 10.5%

General maximum 14.0%

Preferred sector maximum 12.5%

Less preferred sector maximum 14.0%

Agricultural credit guaranteescheme 7.0-8.0%

Residential housing costingless than 100,000 Niara 8.0%

Agricultural production 8.0%

Source: World Bank, Financial Intermediation in Nigeria.

- 103 -

ANNEX 5

PERU

Government Intervention in Financial Markets

1. The Peruvian government played a dominant role in the country'sfinancial system throughout the 1970 to 1982 period. In the early andmid-70s, the government relied heavily on a system of directed credit toachieve a wide range of policy objectives, controlling both the allocationand cost of credit. Two important components of this system were a regimeof detailed interest rate ceilings and an array of publicly owned,specialized financial institutions. A large percentage of the country'scredit was channelled through these institutions by the central bank,through extensive rediscounting of loans made to target borrowers, such asagriculture, housing, public enterprises, etc. Other, though lesser,components of the system included preferential rediscounting of commercialbank lending to targetted sectors and restrictions on the composition ofthe bank portfolios.

2. A growing economic crisis in the mid-70s led to a shLft ineconomic policy towards greater emphasis on market-oriented allocation. In1978 a number of reforms were introduced, aimed at controlling inflationand ending economic stagnation by giving market forces a larger role in theeconomy. These reforms included the liberalization of interest ratepolicies, the easing of reserve requirements, the opening of the capitalmarkets, the legalization of dollar denominated financial transactions,plus a reduction in tariffs and trade restrictions. Beginning in late1978, a series of large upward shifts were made in the nominal interestrate ceilings. The permissible compounding periods were also progressivelyshortened starting in 1981.

3. The escalation of domestic inflation in 1983, to levels exceeding100%, offset most of these reforms producing negative real deposit ratesonce again. The rise in inflation reflected both the effects of adverseweather conditions on output and the necessity of increased inflationaryfinancing to cover the rising government deficit. In addition, the rate ofinflation was increased by the capital market reforms themselves. Thesereforms reduced the base of the inflation tax, and thus increased theinflation which would result from a given amount of inflationary finance.This reduction occurred in two ways: (1) by cutting required reserves andpaying interest on them and (2) by allowing the "dollarization" of theeconomy, which reduced the demand for sol denominated EinanciaLl assets andthus the demand for base money (See para. 6). This interacticin betweenrising inflation and capital market reforms is a good example of thedifficulties that can arise when financial sector liberalization occursbefore the fiscal deficit is under control. (See Mathieson and McKinnonfor a discussion.)

4. In a country with a fair degree of capital market openness, suchas Peru, interest rate policy must be judged in terms of expecteddevaluation as well as expected inflation. Serious complications can

- 104 -

arise in the financial sector from misalignment between domestic interestrates and foreign rates adjusted for expected devaluation. In particular,if the domestic currency becomes overvalued, then expectations ofdevaluation will raise the expected returns, in local currency, on foreigncurrency assets. This will produce a shift away from domestic currencyassets towards foreign currency assets, unless the interest rates ondomestic currency assets are permitted to rise to a competitive level.This situation has occurred to some degree in Peru. Unless expectations ofdevaluation decline, the government is confronted with two ratherunpleasant options. Either it can allow sol deposit rates to rise in linewith the expected return on dollar denominated assets and thereby stem theflow out of soles into dollars. Or it can hold down the sol deposit ratesand let "dollarization" or capital flight continue.

5. Either choice is likely to exert negative pressure on domesticoutpu1t. If, on the one hand, interest rates on sol assets were to rise,then banks' costs of sol deposits would also rise, placing upward pressureon sol denominated lending rates in real terms. This would produce acontractionary impact on the already depressed economy. In addition, ifrates were allowed to rise, then the government would face higher borrowingcosts to service its local debts. On the other hand, if the interestrates on sol assets were not allowed to rise, then funds would move intodollar assets, shrinking the supply of sol credit. This too would have acontractionary impact, since a reduction in the supply of sol credit wouldforce many firms to either take-on expensive dollar liabilities, or borrowin the expensive informal credit market, or restrict their operations.

6. Besides reducing available domestic credit, the shift into dollarassets shrinks the size of the domestic monetary base as mentioned above.This reduces the quantity of assets that the government can tax throughseigniorage. This loss can produce a potentially explosive increase ininflation as follows: A decrease in the domestic monetary base reduces thegovernment's yield from seigniorage, increasing the need for financing fora given deficit. A larger need for financing implies more inflationaryfinanIce and more rapid inflation. Faster inflation, relative to foreigninflation, raises the rate of devaluation required to maintain the realexchange rate. More rapid devaluation, or an increasingly misalignedcurrency, raises the expected sol return on dollar assets. Higher expectedreturns on dollar assets promote capital flight and "dollarization," which,coming full circle, shrinks the domestic monetary base, reinitializing theentire process. Because the process feeds back on itself, increasing therequired inflationary finance with each cycle, the government must be verycautious about pursuing financial reform before its deficit is undercontrol.

7. As of late 1982, the government appeared to be following a pathsomewhere between the two extreme options. Effective interest rateceilings were set quite high relative to past inflation rates. However,ex-post, the yields on sol denominated assets remained well below the theex-post yields on dollar assets assets and "dollarization" steadilyincreased. By the end of 1982, dollar holdings amounted to nearly 40% oftotal assets. In an attempt to capture some of these dollar resources, the

- 105 -

government imposed a 90-100% reserve requirement on the banks' dollardeposits. The central bank invested these dollars overseas at a yieldequal to LIBOR, paying the banks approximately LIBOR minus 2 piercentagepoints, which turned around and payed depositors LIBOR minus 3.

8. By legalizing dollar transactions and maintaining thes highreserve requirement, the government established some control over fundsthat would have leaked out in illegal capital flight to Panama or Miami.However, since the funds were invested overseas, and, thus, not availabledomestically, the process was essentially capital flight through thecentral bank, although it did help the country's measured reserveposition. As previously noted, such a policy does not improve the domesticcredit situation and can be highly destabilizing. With surging inflationand lagging devaluation "dollarization" and its hazards are bound toescalate, underscoring the need (1) to allow interest rates on soldenominated assets to move to levels more competitive vTith the expectedreturn on dollar assets, (2) to reduce expectations of devaluation, and(3) to "de-dollarize."

9. In sum, responding to mounting inflationary and other economicpressures, the Peruvian government recently moved away from a highlycontrolled system of directed credit, toward a more market-oriented policyin the financial sector. It did this by substantially raising theeffective interest rate ceilings, reducing the importance of the directedcredit schemes and initiating discussions on banking law reforms. However,while some financial market liberalization did take place, governmentintervention remained quite extensive: Interest rate ceilings remainedbinding on many financial transactions; the authorities retained controlover many of the country's financial institutions; and the directed creditprograms remained large. In addition, the system cont-Lnued to be subjectedto the strain of high inflation. In particular, high inflation rates leadto unwarranted appreciation of the currency, which, in turn, raisedexpectations of devaluation and therefore the exchange risk of holding soldenominated assets and dollar liabilities.

The Development of Interest Rates

10. As can be seen from the nominal interest rate ceilings displayedin Table 1A and the corresponding effective ceiling rates displayed inTable 1, there was a major upward shift in the structure' of sol denominatedinterest rates in the 1978-82 period, as compared to the 1970-77 period.Despite the upward shift in the effective rates for so:L deposits, whichreflected the higher nominal ceilings and more frequent allowablecompounding, the ex-post real rates for sol deposits r,smained predominantlynegative during 1978-82. In fact, the year-end figures displayed in Table1 were negative throughout 1972-82. However, the adjustments did permitreal sol deposit rates to reach positive levels at timess during the early1981 to mid-1982 period. In particular, a six month term deposit made inJanuary 1982 (after the adoption of a 55% uniform deposit rate!) yielded aneffective rate of 71% and an ex-post return of 2.0%, if compoundedmonthly. After mid-1982 real deposit rates moved highly negative onceagain with the resurgence of inflation. Thus, over the 1970-82 period

- 106 -

PERU: Table IA

Selected Nominal Interest Rate Ceilings

End of Year (annualized)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Deposits, domestic denom.

DemandNominal 0.0% 0.0% 0.0% 0.02 0.0% 0.0% 2.0% 2.0% 2.0% 2.0% 2.0% 2.0% 55.0%

SavirgsNominal 5.0% 5.0% 5.0% 5.0% 5.0% 5.0% 9.0% 11.5% 29.0% 30.5% 30.5% 50.5% 55.0%

Term, 6 monthNominal 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 11.0% 14.0% 31.5% 31.5% 31.5% 52.0% 55.0%

Morgage bank certificatesNominal 9.0% 9.0% 9.0% 9.0% 9.0% 9.0% 11.0% 14.0% 31.5% 33.0% 33.0% 51.0% 55.0%

Lending, domestic denom.

GeneralNominal 12.0% 12.0% 12.0% 12.0% 12.0% 12.0% 15.5% 17.5% 31.5% 32.5% 32.5% 47.5% 47.5%

Rediscount rates to theBanco Agrario

Nominal 4.0% 4.0% 4.0% 4.0% 4.0% 4.0% 7.0% 9.0% 20.0% 21.0% 21.0% 29.0% 29.0%

Sources: 1970 to 1977 figures are from La Superintendencia de Banca y Seguros, Boletin Estadistico.1978 to 1982 figures are from El Banco Central de Reserva del Peru, Boletin.

- 107 -

PERU: Table 1

Selected Effective and Etx-Post Real Interest Rates a/

End of Year (annualized)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Deposits, domestic denom.

DemandNominal 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 2.0% 2.C% 2.0% 2.0% 2.0% 2.0% 10.0%Ex-post real -5.4% -6.5% -15.0% -20.8% -22.5% -21.6% -28.0% -48.2% -39.7% -33.0% -47.5% -39.6% -59.2%

SavingsNominal 5.0% 5.0% 5.0% 5.0% 5.0% 5.0% 9.0% 11.5% 29.0% 30.5% 30.5% 50.5% 55.0%Ex-post real -0.7% -1.9% -10.7% -16.9% -18.7% -17.7% -23.1% -43.4% -23.7% -14.3% -32.8% -10.9% -42.6%

Term, 6 monthNominal 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 11.0% 14.0% 31.5% 31.5% 31.5% 63.0% 71.2%Ex-post real 1.2% 0.0% -9.0% -15.3% -17.1% -16.1% -21.7% -42.1% -22.3% -13.7% -32.3% -3.4% -36.6%

Morgage bank certificatesNominal 9.0% 9.0% 9.0% 9.0% 9.0% 9.0% 11.0% 14.C% 31.5% 33.0% 33.0% 61.6% 71.2%Ex-post real 3.1% 1.9% -7.3% -13.7% -15.6% -14.6% -21.7% -42.1% -22.3% --12.7% -31.5% -4.3% -36.6%

Deposits, foreign denam.

Nominal (in U.S.$) 5.0% 5.0% 5.0% 5.0% 5.0% 5.0% 5.0% 4.6% 9.1% 11.4% 14.9% 11.0% 6.7%tx-post real (in Soles) -0.7% -1.9% -10.7% -16.9% -18.7% 71.7% -1.8% -25.4% -15.0% -5.1% -10.8% 17.2% 1.6%

Weighted average rate ofreturn on above assetsplus currency

Ex-post real -3.1% -4.0% -12.7% -18.7% -20.6% -19.1% -26.4% -45.8% -31.4% -21.7% -32.6% -12.0% -30.0%

Landing, domestic denom.

GeneralNominal 16.3% 16.3% 16.3% 16.3% 16.3% 16.3% 21.2% 24.2% 50.4% 52.7% 52.7% 98.0% 98.0%Ex-post real 10.0% 8.7% -1.1% -7.9% -9.9% -8.9% -14.5% -36.9% -11.1% 0.2% -21.4% 17.3% -26.6%

Rediscount rates to theBanco AgrarioNominal 4.0% 4.0% 4.0% 4.0% 4.0% 4.0% 7.0% 9.0% 20.0% 21.0% 21.0% 29.0% 29.0%Ex-post real -1.6% -2.8% -11.6% -17.7% -19.4% -18.5% -24.5% -44.7% -29.1% --20.6% -37.7% -23.6% -52.2%

Inflation (CPI) 6 monthsForward from December 5.7% 7.0% 17.6% 26.3% 29.1% 27.6% 41.8% 97.0% 69.1% 52.3% 94.3% 68.8% 169.9%

Devaluation 6 monthsForward from December 0.0% 0.0% 0.0% 0.0% 0.0% 108.6% 32.6% 40.6% 31.7% 29.8% 50.8% 78.3% 156.8%

Inflation (CPI) overPrior December 5.8% 7.6% 4.4% 13.8% 19.1% 23.9% 44.7% 32.5% 73.7% 65.8% 60.8% 72.7% 65.6%

Sources: 1970-77 interest rates are from La Superintendencia de Banca y Seguros, Boletin Estadistico.. 1978-82 inrerest rates arefrom Cl Banco Central de Reserva del Peru, Boletin. CPI, exchange rate and LIBOR rate data are from the IF, IFS,data tape. Weights for the weighted average are from Peru Table 2.

Notes: Real rates were calculated ex-post, i.e., they reflect the inflation rate six months forward from December.

a/ The 1982 ceiling rate on demand deposits was 55%, actual rates paid were typically 10%, with 20% paid on large deposits.For 1981, term deposits and mOrgage bank certificates were compounded quarterly. For 1982, compounding was monthly.For 1970-76, dollar deposits were assumed to yield 5%. For 1977-82, the yield was assumed to be LIBOR minus threepercentage points.The effective, general lending rates were the ceiling rates plus 2 percentage potnts commission adjusted for 100%prepayment of interest and commissions.

- 108 -

savers utilizing sol denominated instruments were greatly punished byinflation; correspondingly these assets shrank as a percentage of GNP (seeTable 2).

11. In comparison, the ex-post rate of return, in soles, of dollardenominated assets remained much higher than the return on sol denominatedassets. The dollar denominated assets even turned positive in 1981 and1982. As can be seen in Table 2, this produced a massive shift intodollars, which climbed from 16.2% of total assets in 1978 to 38.9% in1982. The growth in dollar assets was primarily responsible for the risein the ratio of financial assets to GNP, rising from 14.3% in 1978 to 20.2%in 1982, a level which was much closer to the share of GNP that prevailedduring the low inflation period, 1970-1975.

12. The ceiling rates for general sol credits also underwent anupward revision over the 1978-82 period. As a result, the effective realrate on sol credits became positive at the end of 1979 and 1981 (but notyear-end 1980, as the December 1980 to June 1981 inflation rate surged to94%). In early 1981, the nominal lending rate ceiling was raised to47.5%. This rate, plus a two percentage point commission, together withthe common practice of requiring borrowers to prepay interest andcommissions, made possible an effective lending rate of 98.0%. For January1981, this meant the real cost of borrowing for six months could have beenas high as 18.2% (17.9% in January 1982).

13. In contrast to the general lending rates, the rate at which thecentral bank rediscounted loans through the Banco Agrario was highlyconcessionary during the entire period, remaining highly negative in realterms ex-post throughout. Moreover, despite the Central Bank's efforts toreduce its rediscounting activity and move towards more market-orientedrates, a wide variety of special rediscounting categories were maintained.Each category had its own particular interest rate ceiling. These ratesranged between the rate to the Banco Agrario up to the rate on general solcredits. However, the bulk of these credits were extended at highlynegative real rates ex-post.

The Pattern of Deposit Rates, End of Year 1982

14. In early 1982, the authorities adopted a uniform 55% nominalinterest rate ceiling on all sol denominated deposits. This ceiling alsoincluded demand deposits, although the rate actually paid on these accountswas typically 10%, and ranged up to 20% on large accounts. By varying thelength of the compounding period, the banks were able to pay effectiverates higher than the nominal ceilings and provide some differentialbetween types of accounts. Typically, term deposits and mortgagecertificates were compounded quarterly and savings deposits were notcompounded. Dollar deposits were not subject to a ceiling, but were paidroughly LIBOR minus three percentage points, in dollars. Thus the expectedyield on dollar assets, in sol terms, depended on the expected rate ondevaluation. These factors combined to generate the following structure ofdeposit rates:

- 109 -

PEMU: Table 2

Levels of Selected Financial Assets in the Banking System

Ehd of Year(in billions of Soles)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Curreny, dometicNoLnal 16.3 18.9 21.9 27.2 33.5 42.6 49.6 60.8 91.0 162.0 273.4 436.2 627.9In 1980 prices 272.7 294.0 326.2 356.1 368.2 377.9 304.0 281.4 242.5 260.4 273.4 252.6 219.6% of GP 6.5% 6.8% 6.8% 6.8% 6.8% 6.6% 5.6% 4.7% 4.M 4.3A 4.3% 4.17 3.3%% of total assets 31.2% 31.6% 30.7% 31.8% 32.0% 34.5% 34.1% 32.0% 29.1% 25.8% 22.3% 20.0% 16.6%

Deposits, dcmestic currency

Dema-drbulnal 16.7 17.4 22.0 25.9 33.9 37.2 47.5 60.3 86.0 154.3 268.5 367.8 461.0In 1980 prices 279.4 270.6 327.7 339.1 372.6 330.0 291.1 279.0 229.2 248.1 268.5 213.0 161.2% of QGN 6.77. 6.3X 6.8% 6.5% 6.9% 5.8% 5.4% 4.7% 3.9% 4.1% 4.3Z 3.5% 2.5%X of total assets 31.9% 29.1% 30.9% 30.3% 32.4% 30.1% 32.7% 31.7% 27.5% 24.6% 21.9% 16.9% 12.2%

S.ArSLNomnal 8.0 8.9 10.0 11.3 12.8 15.0 17.1 21.6 30.2 67.0 137.8 355.7 680.2In 1980 prices 133.8 138.4 149.0 148.0 140.7 133.1 104.8 100.0 80.5 107.7 137.8 206.0 237.9z of QGP 3.2X 3.2% 3.1% 2.8% 2.6% 2.3% 1.9% 1.7% 1.4% [.8% 2.2X 3.3% 3.6%X of total assets 15.3% 14.9% 14.0% 13.2X 12.2% 12.2% 11.8% 11.4% 9.6% 10.7% 11.2% 16.3% 18.07

Tern;I*nial 5.8 7.3 7.6 8.6 9.1 9.7 9.9 16.4 24.4 39.5 60.8 114.5 222.1In 1980 prices 97.0 113.5 113.2 112.6 100.0 86.1 60.7 75.9 65.0 63.5 60.8 66.3 77.7Z of QiP 2.3% 2.6% 2.4% 2.2% 1.8% 1.5% 1.1% 1.3Z 1.1% 1.0% 1.0% 1.1% 1.2%X of total assets 11.1% 12.1Z 10.7% 10.1% 8.7% 7.9% 6.8% 8.6% 7.8% 6.3Z 5.0% 5.3% 5.9%

MDrge bark certificatesNomnarl 5.1 7.1 9.5 12.1 14.6 18.2 19.8 23.8 30.8 59.1 107.1 242.4 322.3In 1980 prices 85.3 110.4 141.5 158.4 160.5 161.5 121.4 110.1 82.1 95.0 107.1 140.4 112.7X of GNP 2.0% 2.6% 2.9% 3.0% 3.0% 2.8% 2.2% 1.8% 1.4% 1.6% 1.7% 2.3% 1.7%Z of total assets 9.8% 11.9% 13.3% 14.Z% 13.9% 14.7% 13.6% 12.5% 9.8% 9.4% 8.7% 11.1% 8.5%

Deposits, forelip curnencyNImnal in dolars 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.1 0.4 0.7 1.5 1.9 2.9Nouinal in soles 0.4 0.2 0.3 0.4 0.8 0.7 1.5 7.2 50.6 146.6 378.2 662.1 1470.4in 1980 prices 6.7 3.1 4.5 5.2 8.8 6.2 9.2 33.3 134.8 235.7 378.2 383.4 514.2% of total assets 0.8% 0.3% 0.4% 0.5% 0.8% 0.6% 1.0% 3.82 16.1% 2B.3% 30.9% 30.4% 38.9%

TotalNasiiinl 52.3 59.8 71.3 85.5 104.7 123.4 145.4 190.1 313.0 628.5 1225.8 2178.7 3783.9In 1980 pric 874.9 930.1 1062.1 1119.5 1150.8 1094.7 891.2 879.7 834.1 1010.4 1225.8 1261.6 1323.3X of GM 21.0% 21.6% 22.1% 21.5% 21.3% 19.3% 16.5% 14.72, 14.3% 16.6% 19.4% 20.5% 20.2%

CPI, December (Dec.80-100) 6.0 6.4 6.7 7.6 9.1 11.3 16.3 21.6 37.5 62.2 100.0 172.7 285.9

Exclarge rate, Decenber 38.7 38.7 38.7 38.7 38.7 38.7 45.0 69.4 130.4 196.2 250.1 341.2 506.2

GNP, Dec. adjusted, rcminal 249.6 277.0 322.2 397.2 492.1 640.6 882.7 1290.4 2188.6 3790.7 6308.5 10639.2 18760.0

Sources: Asset data are from the Bano Central de Reserva del Peru, Miwria. GNP ad CP1 data are frcm the 11F, IFS, data tape.

Notes: The GNP data have been logrithalcaly interpolated between the current year and one year forward so as to be expressed inDecebr prices. The figures exclude the no-bark financial sector, except for deposits in the barkig system.

- 110 -

Table 3

The Pattern of Effective Deposit Rates, End of Year 1982

Deposits in Soles: Nominal Ex-post Real

Demand 20% -59.2%Savings 55% -42.6%Term 71.2% -36.6%Mortgage Bank Certificates 71.2% -36.6%

Deposits in Dollars: 6.7% 1.6%

Source: Table 1.

15. Even with the shortest allowable compounding period, the realeffective interest rate ceiling on sol deposits at the end of 1982 washighly negative ex-post and well below the yield on dollar deposits. Thus,the primary issue for savers was not whether to make long term sol depositsor not, but whether to shift their financial assets into dollars fromsoles. Judging from the ex-post yields, it seems clear that those whoswitched into dollars by the end of 1982, and held them for the six months,did extremely well compared to those that did not.

The Pattern of Lending, End of Year 1982

16. Due to the Peruvian government's traditional reliance on directedcredit, the financial sector contains a large number of specializedfinancial institutions, each established to channel credit to particularareas in accordance with perceived needs. To support this institutionalframework, the authorities maintained a regime of nominal lending rateceilings. A sense of the breadth and complexity of the regime can bederived from Table 4, which displays the regime in effect as of December of1982. Moreover, it should be noted that this regime represented asimplification of the system of directed credit which prevailed before1978.

- 111 -

Table 4

Lending Rate Ceilings Effective End of Year 1982

Loans in Soles:

Credit institutionsMortgage and Housing Banks and

Credit Unions 55.5%Loans for single dwellings

Mortgage, Housing and Savings Banks 48.5%Credit Unions 52.0%

Financial Enterprises 54.0%Insurance Co. loans to policy holders

with policy guarantees 10.0%Inter-financial institution loan and

deposit rate 60.0%Other loans

short term 47.5%long term 54.0%

Readjustment loans in accordance withLaw 23327for single dwellings 17.0%with present rates up to 47.5% 17.0%with present rates over 47.5% 22.5%

Lines of Credit:

Selective Regional Creditto financial intermediaries 41.5%to final users 44.5%

Selective Agrarian Creditto financial intermediaries 43.5%to final users 49.5%

Agricultural Promotion Creditto financial intermediaries 5.0%to final users 10.0%

Promotional Funds:

FONCAP and FIREin soles

to financial intermediaries 49.0%to final users 54.0%

in U.S. $to financial intermediaries 12.0%to final users 15.0%

FONEXin U.S. $

to financial institutionsup to 1 year 6.0%up to 2 years 7.0%up to 3 years 9.0%up to 5 years 10.0%

- 112 -

Loans from the Central Bank to:

Commercial and Savings Banks 44.5%Regional Banks 43.5%Financial Enterprises 52.5%Agrarian Bank 29.0%Bank of the Nation 37.0%Bank of IndustryFENT linein U.S. $ 0.0%in Soles 40.0%

other loans 42.0%Mining Bank

mining acceptances 3.0%other loans 42.0%

COFIDE 42.0%FONAPS 42.0%Loan for subsidized single dwellings 31.0%

Source: Banco Central de Reserva del Peru.

17. Without a better understanding of how the ceiling rates in Table4 translated into actual costs of borrowed funds, it is not possible topass final judgement on the interest rate regime either in terms of equityor efficiency. However, it is likely such a complex system fell far shorton both accounts, either through the incidence of various implicit taxes,or through built-in rigidities.

18. A very serious consequence of the directed credit schemes and theresultant proliferation of financial institutions was the high degree ofmarket fragmentation. The operations of many of the financial institutionswere narrowly restricted, but they also enjoyed special sanctions thatinhibited other institutions from entering their markets. The result was amarked lack of competition in many sub-markets and a limitation on theinter-sectoral mobility of financial resources.

19. These factors, in combination with the lack of scale economiesand the high labor costs, contributed to excessive intermediation costs.Intermediation costs in Peru were high, even by developing countrystandards. Operating costs of Peru's commercial banks recently wereestimated at 13.3% of total assets, which is far in excess of estimates forthe OECD countries and substantially above estimates for Turkey andColombia developing countries which have fairly high-cost financialmarkets. Such inefficiencies place a serious drag on the Peru's ability tomobilize financial resources and thus lower its growth potential.

- 113 -

ANNEX 6

THAILAND

Government Intervention in Financial Markets

1. In comparison to most developing countries, Thailand's financialsector is quite advanced, open and deep. Market forces, in particular,international market forces, govern the financial sector to a much greaterdegree than most countries at Thailand's income level. Correspondingly,the Thai government plays a relatively less important role, although it isfar from a passive participant.

2. The relative importance of the market and intervention swung backand forth during the 1970 to 1982 period. Since 1970, the economy passedthrough episodes of relatively liberal economic policy (70-75) and ofincreasing intervention (75-80). Recently, more market-oriented policieshave again become the rule.

3. Despite a financial system on the free market end of the spectrumof developing countries, government intervention remained an importantfeature of the financial markets. However, the latitude for interventionwas constrained in many areas by the government's overall policy objectiveof a stable and freely convertible currency.

4. Although the authorities imposed interest rate ceilings, theseceilings could not be set so low as to push the domestic deposit rates orthe lending rate to prime borrowers much below the correspondinginternational rates. Due to free convertibility, such a disparity wouldinduce capital flows that would worsen the mobilization of financialresources and threaten the stability of the Baht. In other words, ifdeposit rates were set too low, savers would be induced to place theirdeposits overseas, reducing the pool of domestically available resourcesand running down exchange reserves. Low lending ceilings would eitherpressure banks to lower the deposit rates to maintain their spreads,sending depositors overseas, or squeeze the profitabilit:y of domesticlending operations. Such a profit squeeze would cause the banks toconstrict their domestic lending, and shift towards greater net foreignlending. Moreover, borrowers who obtain credit would fi.nd it profitablesimply to deposit the funds abroad. Thus, low interest rate ceilings wouldadversely affect the mobilization of domestic and foreign resources andwould put pressure on international reserves.

5. Even though the lending ceilings did not bind on short termcredits to prime borrowers, i.e., the large and medium sized enterprises,the ceilings were set too low to adequately compensate Eor riskier and morecostly lending operations, such as, term finance and lending to smallerenterprises. As a result, most term finance was undertaken throughcontinuous refinancing short term debt, and many enterprises were forced togo into the informal credit market or queue up for rationed credit throughthe directed credit programs maintained by the Thai authorities.

- 114 -

6. These directed credit programs were quite sizable. Theyaccounted for as much as a third of total credit extended and involved amajority of the country's financial institutions including the centralbank, the commercial banks, plus a number of specialized lendinginstitutions.- The Bank of Thailand (the central bank) participated inthese schemes in a number of ways and is an important source of funds. Itdid so by providing cheap rediscounting on bills of target borrowers withthe commercial banks and the specialized lending institutions, such as theGovernment Housing Bank and the Bank for Agriculture and AgriculturalCooperatives (BAAC).

7. The authorities also required the commercial banks to lend 13% oftheir previous year's total deposits to agriculture, either by directlending, by purchasing government bonds or by making deposits in BAAC.Since the rates of return to the commercial banks on these assets were lessthan the market rate on bank lending, they also formed a source ofsubsidized funds to the directed credit schemes. However, again the needto maintain domestically mobilized savings in the commercial banking systemlimited the amount of subsidization the authorities could impose.Otherwise, forced investments in low interest credits would force thecommercial banks to lower their deposit rates, driving away somedepositors. At the end of 1981, the average return to the commercial bankon assets connected with the agricultural credit schemes was about threequarters the ceiling rate on general commercial bank lending. Althoughthis represented a healthy subsidy, it was close to the marginal cost ofbank deposits, represented by the rate paid to long term deposits.

8. Other forms of government intervention in the financial marketsincluded compulsory holding of government securities by the banks and arequirement that committed 60% of bank deposits gathered from a ruralregion to lending in that region.

9. In short, while the Thai financial sector could be counted amongthe developing world's most market-oriented, and certainly among the mostopen, the authorities maintained important institutional controls.However, the distortions created by these controls were limited by theoverriding policy objective of maintaining a convertible currency and anopen economy.

The Development of Interest Rates

10. The regime of interest rate ceilings met with somewhatlimited success in adjusting to inflationary developments andmaintaining positive real rates ex-post--judging from Table 1, whichdisplays a selection of nominal interest rate ceilings and thecorresponding ex-post real rates, for the 1970-82 period. During thisperiod, two waves of moderately high inflation swept through the Thaieconomy, in 1972-74, and again in 1977-81. These two surges closelycorresponded to the movements of world inflation, both in timing andmagnitude, with the Thai inflation running a little above world rate inthe first, and approximately the world rate in the second. The impactof these inflationary episodes is readily apparent in the ex-post real

- 115 -

rates, for they moved decidedly negative during both the 1971-73 and1976-80 periods. However, in comparison with most developingcountries, the regime's performance was quite respectable, as nm!asuredby the frequency of positive real rates. Taken as a group, the yearend figures for the ex-post real ceiling rates on savings deposits, 6month term deposits, general loans and concessional export credits,were positive in 24 out of 52 instances, for the years 1970 to 1.982.

11. Of all the interest rate ceilings displayed in Table It the rateon general loans and overdrafts remained positive in real terms mostfrequently. Of the thirteen end year figures, from 1970 to 198:2, the realrate moved significantly negative only three times, in 1972, 19,73 and1979. This could be attributed to the fact that this rate must stay inline with the corresponding international rate, as discussed earlier.

12. In contrast to the rate on general credit, the ceilings on thediscount rate on export bills, the savings and term depcosit rate wasnegative, in real terms, more frequently. They all dropped below the rateof inflation in the same eight years, during the two inflationary surges, acommon pattern worldwide. Moreover, after a couple of years of acceleratedinflation the interest rate ceilings were moved upwards and maintained evenafter inflation fell off somewhat. This upward ratcheting processcontributed to the subsequent appearance of positive reaLl rates. The late1980 reforms were clearly an example of this process. Table 1 shows asubstantial increment of the ceiling rates for deposits and generallending. Savings deposits shifted upward from 5.5% to 8.0%, six month termdeposits from 7.0% to 10.0%, and the general loans and overdrafts rose from15.0% to 18.0%. These increases were sustained, and in some casesaugmented, through 1981 and 1982 while inflation declined.

13. Conspicuously absent in 1980 was an upward adjustment of thediscount rate for export bills. Also, the discount rate charged foragricultural bills, although it is not displayed in Table 1, was unaffectedby the reforms, remaining unchanged at 10.0% since 1972. Despite theauthorities' announced intentions to pursue more market oriented rates,apparently the reform program had its limitations.

14. In Table 2, which displays the levels of financial assets held inthe commercial banks by the private sector, the effects of the 1980 reformsshow up markedly. Total assets grew 32.1%, in real terirns, from the end of1979 to the end of 1982, and, as a share of GNP, rose to 41.9% in 1982,compared to 33.7% in 1979. The increase in the ceiling rates also affectedthe composition of asset holdings, accelerating the ongoing shift intoquasi money from Ml. Thus, the 1980 reforms generated significantfinancial deepening.

The Structure of Deposit Rates, End of Year 1982

15. The ceiling rates for various deposit accounts in the commercialbanks at the end of 1982 are shown in Table 3. Some sp:read exists betweenmaturities, but it is questionable whether that spread was adequate to

- 116 -

THAILAID: Table I

Selected Nominal and Es-Post Real Interest Rate Ceilings

End of Year (annualized)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Deposits, domestic denom.

DenandNominal 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%Ex-post real -0.6% -7.1% -17.6% -26.0% -3.6% -2.6% -12.2% -11.0% -9.5% -22.1% -15.3% -3.3% -5.5%

SavingsNominal 3.5% 3.5% 3.5% 3.5% 4.5% 4.5% 4.5% 4.5% 4.5% 5.5% 8.0% 9.0% 9.0%Ex-post real 2.8% -3.9% -14.7% -23.5% 0.8% 1.8% -8.3% -7.0% -5.4% -17.9% -8.5% 5.4% 3.1%

Term, 6 monthsNominal 6.0% 6.0% 6.0% 6.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 10.0% 11.0% 11.0%Ex-post real 5.3% -1.6% -12.6% -21.6% 3.2% 4.2% -6.1% -4.7% -3.2% -16.7% -6.8% 7.4% 4.9%

Weighted average rate ofreturn on above assetsplus currency

Ex-post real 2.4% -4.1% -14.8% -23.5% 0.5% 1.8% -8.2% -6.8% -5.3% -18.5% -9.4% 4.5% 2.4%

Lending, domestic denom.,

General loans & overdraftsNominal 14.0% 14.0% 14.0% 14.0% 15.0% 15.0% 15.0% 15.0% 15.0% 15.0% 18.0% 18.0% 18.0%Ex-post real 13.2% 5.9% -6.1% -15.7% 10.9% 12.0% 1.0% 2.4% 4.1% -10.5% 0.0% 14.1% 11.6%

Export (discount rate)Nominal 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0% 7.0%Ex-post real 6.3% -0.6% -11.8% -20.9% 3.2% 4.2% -6.1% -4.7% -3.2% -16.7% -9.32 3.5% 1.2%

Inflation (CPI) 6 monthsForward from December 0.7% 7.7% 21.3% 35.2% 3.7% 2.6% 13.9% 12.3% 10.5% 28.4% 18.0% 3.4% 5.8%

Devaluation 6 monthsForward from December 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0% O.0% 0.4% -0.2% 3.6% 0.0% 0.0%

inflation (CpI) overPrior December -1.3% 1.3% 8.8% 20.2% 17.9% 4.4% 3.4% 8.9% 7.8% 15.0% 16.4% 12.3% 2.6%

Sources: Interest rates are from the Bank of Thailand, Monthly (Quarterly) Bulletin.CPI data are froa the IW, IFS, data tape.Weights for weighted average are frce Thailand Table 2.

Notes: Real rates were calculated ex-post, I.e., they reflect the inflation rate six months forward from December.

- 117 -

TIYAND: Table 2

levels of Slected Final Assets

End of Year(in biilions of BEht)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Nominal 11.9 13.1 15.3 18.6 20.4 22.3 25.7 28.5 33.0 40.6 45.7 47.5 53-7In 1980 prices 31.4 34.2 36.7 37.3 34.7 36.2 40.3 41.2 44.2 47.3 45.7 42.3 46.6% of GNP 8.4% 8.5% 8.1% 7.7% 7.2% 7.C0 7.1% 6.7% 6.5% 6.7% 6.4% 6.0% 6.2%% of total ats 28.7% 27.1% 25.5% 25.5% 22.9% 21.3% 20.6% 19.1% 18.4% 19.9% 18.3 16.4% 14.9%

Deposits, dmstic airreny

DemrdN*ml 6.9 7.7 8.8 10.6 12.5 13.0 15.1 16.4 21.1 22.2 25.1 25.0 23.5In 1980 prices 18.4 20.2 21.2 21.1 21.1 21.2 23.8 23.7 28.2 25.8 25.1 22.3 20.4% Of GNP 4.9% 5.0% 4.7% 4.4% - 4.4% 4.1% 4.2X 3.8% 4.2X 3.7% 3.5% 3.2% 2.7%% of total assets 16.8% 16.0X 14.8% 14.4% 13.9% 12.5% 12.1% 11.0% 11.8X 10.9% 10.0C 8.6% 6.5%

SavinpNomiral 2.7 3.0 3.9 4.9 6.3 7.2 8.9 10.7 14.2 17.2 27.1 36.8 60.0In 1980 prims 7.2 7.7 9.3 9.8 10.7 11.8 14.0 15.4 19.1 2X.0 27.1 32.7 52.1% of GIP 1.9% 1.9% 2.0% 2-.% 2.2% 2.3% 2.4% 2.5% 2.8% 2.8% 3.8% 4.6% 7.0%X of tota assets 6.6% 6.1% 6.4% 6.7% 7.1% 6.9% 7.1% 7.1% 8.0% 8.4% 10.8% 12.7% 16.6%

TermN112Unl 19.8 24.5 31.9 39.1 50.1 62.1 74.9 94.1 110.7 124.0 152.3 181.1 223.3In 1980 prioss 52.4 64.2 76.6 78.2 85.1 100.9 117.7 135.8 148.3 144.4 152.3 161.3 193.9

% of total asets 47.9% 50.9% 53.3% 53.4% 56.1% 59.3% 60.1% 62.9% 61.9% 60.8% 60.9% 62.4% 61.9%

TotalNomLil 41.3 48.3 59.8 73.2 89.3 104.6 124.6 149.7 179.1 2D4.0 250.3 290.4 360.5In 1980 prices 109.5 126.3 143.8 146.4 151.7 170.1 195.8 216.0 239.7 237.5 250.3 258.6 313.0% of MP 29.4% 31.3t 31.7% 30.2T 31.3% 33.0% 34.3t 35.1% 35.5% 33.7% 34.9% 36.7% 41.9%

CPI, Dnmsr (Dec.80-100) 37.7 38.2 41.6 50.0 58.9 61.5 63.6 69.3 74.7 85.9 100.0 112.3 115.2

GIP, Dec. adjusted, nml 140.5 154.2 188.4 242.5 285.1 316.9 362.7 426.2 503.8 606.2 716.9 791.6 860.1

SaLrces: Asset data are frm the Bark of Thailand, Qarterly (Mmthly) balletin. Gl d CPI data are frao the DI, IFS, datatape.

Notes: The GP data have been logriticaUlly interpolated betimen the irrent year and ore year foramrd so as to be expressed inDember prices. Deposits are navVovernnt deposits in the commcial baks, *tih are 80-90% of all private deposits.

- 118 -

mobilize longer term deposits. In particular, the 3% spread between sixmonth and two year deposits was probably not sufficient to cover the riskof committing funds for an additional year and a half, especially wheninflation occasionally reached 20%. The banks, however, moved to reducethis risk somewhat by offering lower early- withdrawal penalties.

Table 3

Interest Rate Ceilings on Deposits, End of Year 1982

Nominal

Demand deposits 0.0%

Savings deposits 9.0%

Time deposits3 to 6 months 11.0%6 to 12 months 12.0%1 to 2 years 13.0%over 2 years 14.0%

Source: Bank of Thailand, Quarterly Bulletin.

The Pattern of Lending Rates, End of Year 1982

16. In comparison with the 19% ceiling rate, which binds on mostborrowers, and the 15% to 60% rates paid in the informal credit market forup to half of all credit, the preferential lending rates displayed inTable 4, of 7% to 10%, were difficult to justify on efficiency grounds.Judging from the large volume and widespread credit transactioned at rateswell above the preferential rates many investment opportunities existed atrates much higher than those offered preferentially. This makes it diffi-cult to argue that the net effect of the preferential credit programs wasto mitigate capital market imperfections and direct resources to applica-tions with the highest return.

17. It is also possible to question the credit programs on distribu-tional grounds. By their nature, rationing schemes at below market ratestend to favor the higher income, more creditworthy borrowers in both thetarget and general markets. Furthermore, the suppressed cost of fundspromotes investment in overly capital intensive method of production. Bothof these phenomena.have been observed in Thailand. Clearly, these resultsis at odds with the stated objectives of poverty alleviation.

- 119 -

Table 4

Commercial Bank Lending Rates, End of Year 1982

Nominal

Loans and Overdrafts 19.0%

Discount Rates (rediscountedfrom the Bank of Thailand)

Export bills 7.0%Bills from industrial

undertakings 7.0%Bills from purchases of

agricultural products 7.0%agricultural bills 10.0%

Source: The Bank of Thailand, Quarterly Bu'Lletin.

18. The development of "groups" that hold ownership interests in boththe firms and the banks also reduce the potential improvements in equityand efficiency which originally motivated the programs. Much of theincentive for the formation of such groups can be traced to interventionand the credit rationing process. Firms may try to gain access to underpriced credit by owning the bank, or the banks may attempt to recapture thetransfer from creditor to borrower, implicit in negative real rates, byowning its borrowers.

19. Comparing the 11% rate paid on six month time deposits, a goodbenchmark for the interest cost of commercial banks deposits to the generallending ceiling rate of 19% reveals a healthy gross spread of 8%. Incontrast, the spread for concessional lending was much smaller. Sincethese discount rates were offered on credits, which were, in turn, re-discounted by the Bank of Thailand, the cost of these funds to the commer-cial banks was the rediscount rates of 5%. Thus, on credits discounted at7% the spread was a rather small 2%. Such a small spread was unlikely tofully cover the cost of the lending operation, and some of the cost musthave been borne by the non-preferential borrowers. This transfer, inaddition to the direct cost borne by the government through the centralbank, must be included in an assessment of the economic cost of the fundsused in the preferential credit program.

- 121 -

ANNEX 7

TURKEY

Government Intervention in the Financial Markets

1. Throughout the 1970 to 1983 period, the Turkish governmentintervened extensively in the financial markets, directing much of thecredit allocation process through administered interest rates andsubsidized credit channelled to numerous sectors in the economy--publicenterprises, agriculture, housing, exports, artisans and smal]L traders,etc. A number of additional instruments were used to manipulate the costof credit including: specific quantity directives; specialized financialinstitutions; interest surcharges and rebates; preferential tax treatments;reserve requirements; and the rate of interest paid on reserves. Asrecently as 1983, it was estimated that only about one quarter of totalcredit was free of government controls, and these were generally lent atcomparatively high interest rates.

2. Beginning in the second half of the 1970s, the system ofadministered nominal interest rates was placed under a great deal of stressby sharp increases in inflation rates. During the 1977-80 period,inflation surged from less than 20Z to in excess of 8C%, driving virtuallyall real (ex-post) lending and deposit rates to significantly negativelevels. In June 1980, term deposit rates and some ler.ding rates wereliberalized, granting market forces a freer hand, and allowing some ratesto move upwards sharply.

3. By 1982, the success of the anti-inflationary polic:Les, combinedwith the loosening of interest rate controls and the large amounts ofdirected credit at low interest rates, produced abnormally high realinterest rates in the small part of the credit market with free rates.With the government's strong commitment to anti-inflationary policy, theauthorities responded to the sharp rise in the real ccost of lending byseeking to reduce the cost of deposits, rather than pursuing expansionarymonetary policy. In January 1983, the authorities lowered reserverequirements, raised the ceiling on savings deposit ra.tes, and reimposedceilings on term deposit rates. In some respects, the reimposition of thedeposit ceilings represents a partial retreat from the libera:Lization of1980.

4. At the end of 1982, the Turkish financial market st:ill containedmany elements characteristic of financially repressed economies, such ascredit rationing, compensating balances, and movements into slhort termassets. Thus, in spite of the 1980 liberalization, market forces had onlya limited role in determining interest rates and non-pirice mechanismsremained major outlets for market pressures.

- 122 -

The Development of Interest Rates

5. Table 1 shows the development of various interest rates, nominaland ex-post real. Here the most striking feature is the absence ofpositive real interest rates from 1973 until 1981. At that point the rateon six month term deposits and the rates charged for general credits becameabruptly positive. In 1982 these same rates became highly positive; thereal rate paid for general short term credits reached 34.3% and the sixmonth deposit rate rose to 22.1%. In contrast, the real rate charged forshort term agricultural credit and the rate paid on savings depositsremained negative throughout 1973-82.

6. The 1977-80 inflationary surge broke the 1973 to 1982 period intothree distinct episodes of interest rate behavior: 1973-77, 1977-80, and1980-82. During the 1977-80 period, real rates reached extraordinarilynegative levels, as low as -60%. These highly negative rates greatlystressed the financial system and prompted the reforms of June 1980.However, it is important to note that highly negative real rates were notmerely a result of this surge. They were, in fact, a long standingcondition, since they were also present prior to 1977.

7. The post-1980 development of the nominal rates for generallending and for term deposits reflected the 1980 liberalization, with ratesmoving up sharply. The combination of higher nominal rates and declininginflation produced an abrupt upward shift in the real rates; the termdeposit and general lending rates both moved directly positive. Incontrast, the unliberalized rates for agricultural credits and sightsavings only moved back up to roughly the same negative real levels thatprevailed during the pre-1977 period. Thus, the 1980 liberalization ofterm deposit and general credit rates notwithstanding, the system ofadministered interest rates only partially adjusted to inflationarypressures, and did not provide a wide ranging mechanism for allocatingcredit.

8. The combined effect of the high positive real rate on termdeposits and the massive shift into these assets carried the weightedaverage of returns on financial assets to a positive level in 1982,relieving a long standing pattern of financial repression. As shown inTable 2, the 1980 easing of financial repression sparked a reversal of thegeneral flight from financial assets. The ratio of total assets to GNPrebounded from the extremely depressed level of 16.3% in 1980 to 25.5% in1982. The bulk of this upswing was caused by the massive rise in timedeposits, which increased from 15.9% of total assets in 1979 to 45.7% in1982 in direct response to the high real rates paid on such deposits: 13.8%in 1981 and 22.1% in 1982. Although the liberalization program was highlysuccessful in mobilizing financial savings, the negative side of thisdevelopment was the high average cost of deposits and the resultant highreal lending rates on the portion of credit subject to the liberalizedlending rates.

- 123 -

TURKEY: Table I

Selected Nominal and Ex-post Real Interest Rates

End of Year (annualized)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Deposits, domestic denom.

DemandNominal iN/A #N/A IN/A 0.0% 2.0% 2.0% 2.0% 2.0% 0.0% 0.0% 0.0% 0.0% 0.0%Ex-post real IN/A tN/A tN/A -12.8% -17.5% -10.0% -16.7% -24.21 -43.3% -63.9% -24.6% -24.2% -18.6%

SavitgNominal tN/A tN/A tN/A 2.5% 3.0% 3.0% 3.0% 3.0% 3.0% 3.0% 5.0% 5.0% 5.0%Ex-ost real IN/A tN/A tN/A -10.7% -16.7% -9.1% -15.8% -23.4% *41.6% -62.8% -20.8% -20.4% -14.5%

Term, 6 monthNominal #N/A tN/A ON/A 4.0% 6.0% 6.0% 6.0% 6.0% 9.0% 12.0% 32.0% 50.0% 50.0%Ex-post real ON/A tN/A tN/A -9.3% -14.3% -6.5% -13.4% -21.2% --38.2% -59.5% -0.4% 13.8% 22.1%

Weighted average rate ofreturn on above assetsplus currency

Ex-post real tN/A 0N/A ON/A -11.2% -16.9% -9.4% -16.2% -23.9% -42.1% -62.9% -18.8% -8.2% 1.2%

l,ending. dmtic denom.

General, short termNominal IN/A ON/A ON/A 8.8% 9.0% 9.0% 9.0% 9.0% 10.0% 10.8% 26.0% 65.0%a/ 65.0% a/

x-post real tN/A ON/A ON/A -5.2% -11.9% -3.8% -10.9% -19.0% -37.7% -60.0% -5.0% 25.1X 34.3%

Ag. credits, short termNominal tN/A tN/A hN/A 7.0% 8.0% 8.0% 8.0% 8.0% 8.0% 11.5% 13.5% 20.0% 18.0%Ex-post real ON/A ON/A ON/A -6.7% -12.7% -4.7% -11.8% -19.7% --38.8% -59.7% -14.4% -9.0% -4.0%

Inflation (CPI) 6 monthsForward from December 20.3% 6.4% 22.6% 14.7% 23.7% 13.3% 22.4% 34.5% 76.5% 176.8% 32.6% 31.9% 22.9%

Devaluation 6 monthsForward from December 172.6% -10.1% 0.0% -7.1% 2.2% 13.8% 12.5% 68.7% 96.0% 396.7% 50.8% 53.7% 40.7%

Inflation (CPI) overPrior December 12.3% 17.6% 9.2% 16.8% 16.7% 19.6% 17.0% 44.6% 36.6% 81.1% 86.2% 30.3% 32.8%

Sources: Interest rates are from the Central Bank Q.iarterly Bulletin, and World Bank data. CPI data are from the IMF, IFS,data tape. Weights for weighted average are from Turkey Table 2.

Notes: Real rates were calculated ex-post, i.e., they reflect the inflation rate six months forward f'rom December.

a/ World Bank staff estinate.

- 124 -

TURKEY: Table 2

Levels of Selected Financial Assets

End of Year(In billions of Turkish Lira, TL)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982

Currency, domestic

Nominal 11.9 13.9 16.0 20.7 26.2 32.9 42.5 63.0 93.8 143.7 217.5 280.6 411.9In 1980 prices 194.1 192.7 203.2 225.0 244.1 256.3 283.0 290.1 316.3 267.5 217.5 215.4 238.1% of GNP 7.1% 6.5% 5.9% 5.7% 5.5% 5.5% 5.5% 5.9% 5.6% 4.6% 4.0% 3.7% 4.1%% of total assets 27.0% 24.6% 22.6% 23.0% 23.22 22.5% 23.5% 25.9% 28.6% 27.4% 24.7% 17.2% 16.1%

Deposits, domestic currency

DemandNominal 6.6 8.7 11.8 16.0 22.6 32.1 44.9 63.0 86.0 154.5 286.0 458.5 651.3In 1980 prices 107.7 120.6 149.8 173.9 210.5 250.1 299.0 290.1 290.0 287.7 286.0 352.0 376.5X of GNP 3.9% 4.0% 4.3% 4.4% 4.7% 5.3% 5.8% 5.9% 5.1% 4.9% 5.3% 6.1% 6.5%% of total assets 15.0X 15.4% 16.7% 17.82 20.0% 22.0% 24.82 25.9% 26.3% 29.4% 32.5% 28.1% 25.5%

SavingsNominal 16.8 20.9 24.9 32.9 39.7 52.2 62.7 82.4 103.3 143.7 197.4 228.5 275.5In 1980 prices 274.0 289.8 316.2 357.6 369.8 406.6 417.5 379.5 348.3 267.5 197.4 175.4 159.32 of GNP 10.0% 9.7% 9.1% 9.0% 8.3% 8.7% 8.2% 7.8% 6.1% 4.6% 3.7% 3.0% 2.8%2 of total assets 38.1% 37.0% 35.22 36.5% 35.1% 35.7% 34.7% 33.9% 31.5% 27.4% 22.5% 14.0% 10.8%

TermNominal 8.8 13.0 18.0 20.5 24.6 29.0 30.8 34.4 44.4 83.3 177.9 665.1 1212.22 of GNP 5.2% 6.0% 6.6% 5.6% 5.1% 4.8% 4.0% 3.2% 2.6% 2.7% 3.3% 8.8% 12.1%X of total assets 20.0% 23.0% 25.5% 22.8% 21.8% 19.8% 17.0% 14.2% 13.6% 15.9% 20.2% 40.7% 47.5%

TotalNominal 44.1 56.5 70.7 90.1 113.1 146.2 180.9 242.8 327.5 525.2 878.8 1632.7 2550.9In 1980 prices 719.3 783.4 897.8 979.3 1053.5 1138.9 1204.6 1118.1 1104.2 977.8 878.8 1253.3 1474.7X of GNP 26.1% 26.2% 25.9% 24.8% 23.6% 24.3% 23.6% 22.9% 19.4% 16.8% 16.3% 21.6% 25.5%

CPI, December (Dec.80-100) 6.1 7.2 7.9 9.2 10.7 12.8 15.0 21.7 29.7 53.7 100.0 130.3 173.0

GNP, Dec. adjusted, nominal 168.7 215.4 273.1 363.8 478.3 601.3 767.6 1061.4 1684.9 3123.3 5391.5 7560.5 10017.9

Sources: Asset data are from the Central Bank of the Republic of Turkey, Quarterly Bulletin.CPI and GNP data are from the DLF, IFS, data tape.

Notes: The GNP data have been logrithmically interpolated between the current year and one year forward so as to be expressed inDecember prices.

- 125 -

Deposit Rates, End of Year 1981

9. Judging from the array of deposit rates shown in Table 3, thereal return on deposits was positive only on fixed term deposits of 6months or longer maturity. Although the real rate on 6 month deposits roseto a rather high 13.8X, there was no incentive to commit funds for longerperiods than 6 months, as the interest rate ceilings provided no additionalspread for maturity. The size of the risk premium which savers required tocommit funds long term, given the volatility of the past inflation, and therisk to financial institutions of accepting such long-term liabilities madeit unlikely that longer term deposits would have been important even in afree market.

Table 3

Deposit Rate Ceilings, End of Year 1981

Term Deposits

Nominal Ex-post Real

1 month 25.0 -25.52 months 35.0 -29.83 months 45.0 -18.26 months 50.0 13.81 year 50.0 13.22 years 50.0 #N/A3 years 50.0 #N/A4 years 50.0 #N/Aover 4 years 50.0 #N/A

Source: World Bank data.

Lending Rates, End of Year 1982

10. In the following table, a sample of the wide variety of lendingterms is presented. The nominal rate charged appears iLn the left column,and the effective rate charged, net of the surcharge or subsidy of theIRRF, the transactions tax and bank commission, appears; in the right.However, the effective rate does not reflect the effect:s of compensatingbalance requirements or other taxes.

- 126 -

Table 4

Lending Rate Ceiling, End of Year 1982

(percent per annum)

Short Term Credits: Nominal Effective

Credit from Central Bank to TMOfor agricultural support purchases 10.0 11.0

Agricultural Credits 20.0 20.0Artisan and small traders 20.0 20.0Export Credit

Industrial products 31.5 28.8Other credits 31.5 35.1

General Credits 36.0 47.3Commercial bank credits to SEE's 36.0 42.3

Medium and Long Term Credits:

Investment credits to SEE'sby State Investment Bank 21.5% 21.5%

Agricultural credit from theAgricultural Bank 22.0 22.0

Artisan and small traderscredit from the Halk Bank 22.0 22.0

Housing credits form the housing bank,construction savings system andsocial housing credit 16.0 18.4

Housing credit scheme and cooperativeSociety credit 22.0 25.3Commercial housing 35.5 44.4

Export oriented, inUnderdeveloped regions 29.0 20.8Other regions 29.0 25.2

Other loans in underdeveloped regions 38.0 28.6Rediscountable loans for manufactures

of investment goods 31.5 28.5Rediscountable working capital loans

In underdeveloped regions 38.0 40.3In other regions 38.0 46.0

Other Medium and Long Term Loans Nominal Effective

Investment LoansExport oriented

In underdeveloped regions 45.0 29.3In other regions 45.0 36.0

- 127 -

(Table 4 continued)

Other Medium and Long Term Loans Nominal Effective

OtherIn underdeveloped regions 45.0 31.5In other regions 45.0 45.0

Working capital loansIn underdeveloped regions 45.0 45.0In other regions 45.0 51.8

Source: World Bank data.

11. Despite recent efforts to rationalize the pattern of lendingrates, there obviously remained a wide dispersion of rai-es in 1982, bothnominal and effective. The importance of the subsidies,, taxes andcommissions on the structure is also quite apparent, as they resulted in aneffective rate of 47.3% on general short term credits, compared to thenominal rate of 36.0%, and an effective rate of 29.3% on investment loansfor export oriented enterprises in underdeveloped regions, compared to anominal rate of 45.0%. Additional disparities between nominal andeffective rates were produced by compensating balance requirements andother practices, generally to the detriment of non-preferential borrowers.

12. The most preferential lending rates were offered to theagriculture sector, artisans and small traders, with housing, stateeconomic enterprises and export promotion following. Non-preferentialborrowers were left to compete for the small amount of remaining credit,while paying effective rates of 65% or more when account:ing for the costsof compensating balances. These high real rates threatened many firms withinsolvency.

13. Taken as a system, the rationality of the array of rates issuspect. First, even at the government's optimistic inilation projectionof 20% for 1983, there were a number of effective interest rates that werenegative in real terms. Second, the differences in the rates probably didnot accurately reflect the relative social rates of returns, or the socialcost of the funds. While the policy objective of steerLng income towards aparticular group in the economy may have been served by an interest ratedifferential in excess of 27% between credit to agricull:ure and generalborrowers, it was unlikely this differential was well grounded in relativeproductivities. The same can be said for almost any paLr of rates in thearray. Furthermore, the absence of a pronounced spread in the long termlending rates, necessary to cover additional costs and risks associatedwith term finance, dampened the incentive for financial intermediaries tomake such loans.

14. In conjunction with an estimate for the average cost of funds,the lending rates in Table 4 can also be used to roughlv approximate thevarious intermediation spreads. For the first quarter of 1983, the average

- 128 -

cost of deposits available for commercial bank lending was estimated to be30.4%. This figure included the interest paid on the different types ofdeposits, and the net cost of reserve and liquidity requirements. Whilemore precise calculations would require estimates of all the commissions,compensating balances, tax adjustments, etc., the calculations based onTable 4 indicate that a large number of rates were nominally below averagecosts of deposits. Clearly, a great deal of back door adjustment andgovernment intervention was required to sustain this system, castingsubstantial doubt on the system's effectiveness in achieving its policyobjectives.

15. It has been also estimated that, in 1982, the cost of intermedia-tion added six percentage points to the average lending rate. This wasabout two times that of other OECD countries. High administration costs,associated with large quantities of directed credit and the rapid turnoverof funds due to high inflation contributed to this cost of intermediation,as did "over-branching" by the banks. Such over-branching is a typicalresponse to ceilings on deposits rates. While the lending rate ceilingscan be more easily evaded through compensating balances, commissions, etc.,deposit limits cannot be so easily bent, and banks therefore compete fordeposits by branching. The resulting excessive staff and overhead costscontribute to high intermediation costs.

16. In sum, as of early 1983 continued progress towards rationalizingfinancial sector policies in general, and interest rates policies in parti-cular, was sorely needed. Furthermore, such progress could be expected toyield substantial economic gains in terms of the mobilization of financialsavings and the efficiency of their allocation, by stabilizing the finan-cial system and by reducing market fragmentation, subsidization and thecost of intermediation.

- 129 -

ANNEX 8

URUGUAY

Government Intervention in Financial Markets

1. The period 1970 through 1983 witnessed a thorough going overhaulof the government's role in Uruguayan financial markets. The first fewyears of the period were marked by a regime of administered interest rates,credit rationing and the pervasive use of quantity directives. Then,between 1974 and 1979, government intervention was, systematically reduced,as part of a major shift in economic policy towards libearalization. Thesereforms led to a financial market which is among the moist free and open inthe developing world.

2. The principal aspects of the 1974-79 financia'l sector- liberaliza-tion included: the legalization of foreign currency deniominated contractsand assets, including bank deposits (1974); the successive lifting ofinterest rate ceilings towards levels in line with market conditions--withtheir final abolishment in 1979; the elimination of sec-toral credit alloca-tion guidelines (1975); and the termination of reserve r~equirements oncommercial banks (1979). During the same period, the government's budgetdeficit was closed, eliminating the need for the governmnent to captureresources through seigniorage.

3. In the post-reform period, the market, however imperf'ect, becamethe principal determinant of the interest rates. This is not to say,however, that the government abandoned the field entirely. It retained asubstantial role in credit allocation, through the publicly ownLed Banco dela Republica and the Banco Hipotecario del Uruguay (Mortgage Bank). Inaddition, attempts were made to influence the cost of credit thkrough theforeign exchange markets.

4. The government's participation in the allocation of creditdeclined sharply in the seventies. The share of the Central Bank in totalcredit fell dramatically, from about 70% of total credit in 1974 to only12% in 1980. To a large extent this fall reflected the elimination of thegovernment's budget deficit. The other government financial intermedia-ries--the Banco de la Republica and the Banco Hipotecario--also declined inimportance. For example their combined share of private sector, credit in1982 was 37%, compared to 64% in 1974. Correspondingly, private financialintermediaries extended 63% of private sector credit in 1982, v'ersus 36% in1974. Thus, the government's role, though substantially reducetd, remainedlarge and influential. Moreover, the terms on which credit was extendedfrom the publicly owned financial institutions to the priority sectors--agriculture, exports and housing--steadily approached those prevailing inthe market.

5. With the opening of the financial markets and the elimination ofreserve requirements and interest rate ceilings, the government abandonedmost of its control over money creation and interest rates. However, itdid attempt to reduce the rate of inflation and the interest differentialbetween credits denominated in domestic and foreign currency thiroughexchange market policies.

- 130 -

6. During the 1979-1981 period, the government sought to reduceinflation by preannouncing the rate of devaluation. This was a switch fromthe 1975-79 policy of a more or less passive crawling peg, which moved at arate roughly equal to the difference between local and foreign inflation.The shift to an active exchange policy was to be accompanied by a reductionin the level of protection--a reduction that actually occurred quiteslowly. The government's objective was to limit the future rate of infla-tion in import prices, and to remove the impression from the minds ofdomestic producers that they could count on the exchange rate and thesystem of protection to passively validate price increases. The intendedresult was to reduce domestic inflation to approximately worldwide rates,without suffering any loss of employment.

7. The authorities also tried to lower the difference between dollarand peso interest rates by providing inexpensive foreign exchange guaran-tees during the second and third quarters of 1981 and the first part of1982. The guarantees were intended to increase arbitrage between foreignfunds and domestic credit, thereby reducing the interest rate differentialbetween financial transactions denominated in dollars and those denominatedin pesos.

8. Both policies did meet with some success; the rate of inflationdid decline, though it remained above the devaluation adjusted world rate;and the interest rate differential was reduced somewhat, although the pesointerest rates did not actually decline due to the rise in foreign rates.However, both policies had to be abandoned as unsupportable, victims of theincreasingly overvalued real exchange rate, the sudden changes in worldfinancial markets, the sharp changes in the Argentine economy, and a rapidincrease in the the public sector deficit. In November 1982 the exchangerate was allowed to depreciate more than 100%; since then the exchange rateregime has been a "dirty" float. This massive devaluation, combined withthe large build-up of dollar credits before 1982, produced significantfinancial pressures on borrowers and the financial intermediaries them-selves. The Central Bank has been forced to intervene in various banks andsupport them through purchases of some of their liabilities. Meanwhile,the economy remains in the recession that has gripped Latin America.

9. In sum, during the last decade the Uruguayan Government moveddecisively to relinquish its control of the country's financial markets,leaving them predominantly in the hands of the private sector and open tointernational forces. However, rapidly changing world and local conditionsinteracted with the "dollarization" of the financial system to generateserious financial difficulties.

The Evolution of Interest Rates

10. Table 1 displays an array of Uruguayan interest rates, nominaland ex-post real, for the period 1970 to 1982. Table 2 shows the develop-ment of financial assets during the same period. A few words are warrantedregarding the numbers presented. First, all the numbers were year-endfigures, and second, the real rates were calculated ex-post, i.e., they

- 131 -

URUGUAY: Table I

Selected Nrcinal and EK-post Real Interest Rates

End of Year (annualized)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 a/ 1982

Deposits, domestic denoe.

DemandNominal 0.0% 0.0% 0.0% 0.0% 10.0% 0.0% (.0% 0.0% 0.0% 0.0% 0.0% 0.0% 0.0%Ex-post real -16.9% -49.2% -46.2% -44.8% -30.5% -15.5% -40.2% -28.9% -43.1% -32.12 -23.8% -9.8% -39.5%

SavingsNominal 6.0% 6.0% 6.0% 8.0% 18.0% 18.0% 21.6% 25.5% 22.8% 24.0% 25.2% 24.0% 24.2%Ex-post real -12.0% -46.2% -43.0% -40.4% -25.4% -0.2% -27.3% -10.8% -30.1I -15.9% -4.6% 11.8% -24.9%

Term, 6 monthNominal 15.0% 15.0% 15.0% 18.0% 30.0% 30.0% 30.2% 51.4% 42.6% 50.6% 50.3% 47.4% 66.2%Ex-post real -4.5% -41.6% -38.1% -34.9% -17.8% 9.9% -22.1% 7.6% -18.9% 2.2% 14.6% 32.9% 0.5%

Deposits, foreign denam.

DemandNominal (in U.S.$) YN/A IN/A 5.5% 5.5% 5.5% 5.5% 5.6% 5.4% 5.5S 5.5% 5.7% 5.5% 5.8%Ex-post real (in NS) Y N/A dN/A -18.5% -15.3% 32.0% 30.3% -14.7% -10.0% -24.4% -19.0% -7.7% 10.5% -41.4%

Term, 6 monthNominal (in U.S.$) IN/A tN/A 8.0% 8.0% 8.0% 8.0% 7.4% 7.5% 8.0% 11.9% 14.6% 13.1% 10.2%Ex-post real (in NS) IN/A #N/A -16.6% -13.3% 35.2Z 33.4Z -13.2% -8.2% -22.6% -14.1% 0.1% 18.5% -39.0%

Weighted average rate ofreturn on above assetsplus currency

Fx-post real (1) IN/A tN/A -43.7% -40.5% -21.2% 0.5% -27.4% -13.6% -28.9% -17.0% --4.2% 13.3% -33.4%Ex-post real (2) tN/A IN/A -43.7% -40.4% -20.8% 1.1% -26.9% -12.8% -28.2% -15.1% -1.5% 16.9% -31.7%

iing, domestic denom.-xed term up to 6 mo.)

AverageNominal IN/A IN/A #N/A IN/A IN/A tN/A 62.0% 76.6% 71.2% 68.1% 65.1% 59.8% 76.3%Ex-post real IN/A IN/A IN/A IN/A IN/A IN/A -3.1% 25.5% -2.,% 14.1% 25.9% 44.1% 6.6%

PrimNominal IN/A tN/A #N/A #N/A ?N/A IN/A 47.6% 65.8% 59.% 49.9% 49.82 46.5% 56.7%Ex-post real IN/A IN/A ISN/A IN/A ?N/A IN/A -11.7% 17.9% -9.1% 1.7% 114.2% 32.1% -5.2%

PreferentialNoMinal IN/A IN/A IN/A IN/A #N/A #N/A #N/A IN/A IN/A IN/A dN/A #N/A #N/AEx-post real IN/A IN/A 1N/A IN/A IN/A IN/A IN/A IN/A IN/A lN/A IN/A IN/A IN/A

Lendi foreil denom.

AverageNominal (in U.S.S) IN/A IN/A IN/A IN/A IN/A tN/A 12.0% 13.8% 142.:% 16.3% 18.5% 18.4% 18.2%Ex-post real (in NS) IN/A #N/A #N/A #N/A IN/A #N/A -9.5% -2.8% -18.1% -10.3% 3.5% 24.1% -34.5%

Prim

Nominal (in U.S.S) IN/A #N/A IN/A IN/A IN/A #N/A IN/A IN/A A NA 15.8% 17.4% 16.8% 17.2%Ex-post real (in NS) #N/A #N/A IN/A IN/A INjA N/A IN/A IN/A #N,A -11.1% 2.5% 22.4% -35.1%

Inflation (CPI) 6 monthsForward from December 20.4% 97.0% 85.8% 81.2% 58.2% 18.3% 67.2% 40.7% 75.% 47.4% 31.2% 10.9% 65.3%

Devaluation 6 monthsForward from December 0.0% 139.M 43.5% 45.4% 98.07 46.1% 35.1% 20.1% 26.0% 13.1% 14.6% 16.2% -9.4%

Inflation (CPI) overPrior December 20.1% 35.2% 95.6% 77.6% 106.8% 67.3% 39.4% 57.6% 45.8% 83.1% 42.9% 29.3% 20.5%

Sources: Interest rate data are from 3dnco r(-it rl del Uruguay, Boletin Estadistico.CPI and exchange rate data are fr:r the IMF, [FS, data tape.Weights for the weighted average ire from Uruguay Table 2.

Notes: Real rates were calculated ex-post, i.e., they reflect the inflation (and devaluation) rate six months forward fromDecember. Weighted average (1) asiunes that all foreign assets are savings deposits; (2) assumes all time deposits.

s/ Since the real rates look forward six months from December, the 1981 figlures do not reflect the November 1982 devaluation.

- 132 -

ULIY: Table 2

I a£ of Selected Firancial Asets

End of Year(levels in blliz of New Pesos or U.S.$)

1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 a/

Currwwzy, dacesticN 1bsrul 0.1 0.1 0.2 0.2 0.3 0.5 0.8 1.1 1.8 3.2 5.1 6.1 7.8In 1980 prices 7.8 8.5 10.4 5.8 4.4 3.9 4.7 4.2 4.7 4.5 5.1 4.7 5.0Z Of tDP #N/A #N/A 11.32 5.9% 5.22 4.6% 4.9% 4.5% 4.3X 4.32 4.6% 4.8% 5.1%% of total assets N/A IK/A 54.8% 39.6% 34.2Z% 28.8% 23.8% 19.6% 16.8% 15.6% 14.7% 11.7% 8.5%

Deposits, d stic arricy

DmandtNid.ul #N/A IN/A 0.1 0.1 0.2 0.4 0.5 0.8 1.3 2.6 3.6 3.6 3.6In 1980 prices IN/A #N/A 3.6 3.8 3.2 3.0 3.3 2.9 3.3 3.7 3.6 2.8 2.3% of (P IN/A IN/A 3.9% 3.8X 3.7% 3.5% 3.4X 3.0% 3.02 3.6X 3.3 2.8% 2.4%% of total assets Mi/A IN/A 18.8% 25.6% 24.6% 21.7% 16.6% 13.3X 11.8% 12.9% 10.5% 6.8% 3.9%

SavinpNominal IN/A IA 0.0 0.0 0.1 0.1 0.3 0.4 0.8 1.5 2.7 3.9 4.3In 1980 prices IN/A IN/A 1.6 1.4 1.0 1.2 1.7 1.5 2.0 2.2 2.7 3.0 2.8% of GP tN/A #I/A 1.8% 1.4X 1.2% 1.4% 1.8% 1.6% 1.8% 2.1% 2.5% 3.02 2.8%% of total assets IN/A IN/A 8.7% 9.6X 8.0% 8.7% 8.7% 7.2% 7.1% 7.6% 7.9% 7.4% 4.7%

Termmnal IN/A #N/A 0.1 0.1 0.2 0.3 0.6 0.8 2.4 5.3 11.4 15.1 15.5

In 1980 prices IN/A IN/A 2.6 2.7 2.5 2.8 3.4 3.2 6.2 7.6 11.4 11.7 9.9% OfP N/A IN/A 2.9% 2.7% 2.9% 3.2% 3.6% 3.4% 5.6X 7.22 10.4% 11.8% 10.0%% of total amets IN/A IN/A 13.9% 18. 19.5% 20.1% 17.1% 15.0% 22.2% 26.1% 32.9% 28.9% 16.8%

Subtotal. dcuticulmeteaidlA ItAat #W N/A 0.4 0.5 0.8 1.3 2.2 3.1 6.2 12.6 22.8 28.7 31.3In 1980 prios IN/A IN/A 18.2 13.7 11.1 10.8 13.1 11.9 16.2 18.1 22.8 22.2 20.1% of GP IM/A #N/A 19.8% 13.9% 13.0% 12.7% 13.7% 12.5% 14.7% 17.2%2 2.9% 22.3203.32% Of total sets IA N/A 96.2% 93.1% 86.3 79.2z 66.2% 55.1U 57.9% 62.2% 65.9% 54.8% 33.9%

1bposits, foign aurmncy

D *i arA temsNominal U.S.S N/A #N/A 0.0 0.0 0.1 0.1 0.3 0.5 0.6 0.9 1.2 2.0 1.8Natnal, d tic IN/A IN/A 0.0 0.0 0.1 0.3 1.1 2.5 4.5 7.7 11.8 23.7 61.1In 1980Oprices IN/A #N/A 0.7 1.0 1.8 2.8 6.7 9.7 11.8 11.0 11.8 18.3 39.2% ofGD I/A IN/A 0.8% 1.0% 2.1% 3.32 7.0% 10.22 10.7% 10.4% 10.8% 18.4% 39.5%

Naina tic and fore4 #NWA #N/A 0.4 0.5 0.9 1.6 3.3 5.6 10.7 20.3 34.6 52.4 92.4

In 1980 prices IN/A IN/A 18.9 14.7 12.9 13.7 19.8 21.5 28.1 29.0 34.6 40.5 59.3% of *P VN/A IN/A 20.6% 14.9% 15.1% 16.0% X2.8% 22.7% 25.4% 27.6% 31.7% 40.7% 59.8%

CPI, temr (Dec.80-100) 0.7 1.0 1.9 3.4 7.1 11.9 16.6 26.2 38.2 70.0 100.0 129.3 155.9

Fwige rate, Deber 0.3 0.4 0.7 0.9 1.7 2.7 4.0 5.4 7.1 8.5 10.0 11.6 33.8

<DP, Dec. adjusted, ndnal IN/A #N/A 1.8 3.4 6.1 10.2 15.9 24.8 42.2 73.6 109.1 128.7 154.5

Sozuer: Asset andi GP data are fram the BDano Central del Urugjay, Boletin EAtadistico. DodanF ae ard CPI data are frau the11FF, IF, data tape.

Notes: The GP data lae been logrithmically interpolated betmei the current year and am year forward so as to be e,pressed inDeoemer prices.

a/ Tk drmatic rise in asets for 1982 is a result of a massive deyaluation. This should not be interpreted as finranaldeepanirig. snce foreign lIabilities hae also risen.

- 133 -

reflect the inflation rate (and the rate of devaluation, in the case offoreign denominated accounts) six months forward. As a result, the realinterest rate figures shown for 1981 and 1982 do not capture thLe impact ofthe massive devaluation that occurred at the end of November 1982. Thisdevaluation produced enormous ex-post rates of returns for holders ofdollar assets up to November 1982. In contrast, the ex-post real rates onpeso denominated assets remained only slightly positive during thisperiod.

11. Judging from Table 1, the effect of the liberalization programwas to permit relatively high and flexible nominal interest rates in pesos,from 1975 onward, and dollar rates which approximated rates in worldmarkets. Despite the free movements of the interest rates and the elimina-tion of controls in 1979, real rates were often negative ex-post. Since1975, the ex-post lending and deposit rates in Table 1, taken as a group,are negative somewhat more often than positive. In fact of the nineinterest rates shown in Table 1, negative real rates appeared 39 times outof 65. Generally speaking, the rates were positive only when inflationsubsided in 1976, 1981 and 1982. Apparently, free interest rates in anopen capital market do not guarantee positive real interest rates ex-postat every point in time; other factors, such as expectations and capitalflows, matter. But, in comparison to most other highly inflationary devel-oping countries, where positive real rates were quite rare, positive rateswere common in Uruguay. Moreover, one must recall that real rates werenegative in most developed countries during much of the seventies.

12. As measured by the ex-post rate of return, the most attractivefinancial asset was the six month term deposit. Since 1975 the real rateon six month term deposits was positive six times. No other saving instru-ment, denominated in domestic or foreign currency, was positive in realterms more than three times. As one might expect, in response to favorablereturns domestic currency term deposits have risen dramatically, both inreal terms or as a percent of GDP (see Table 2).

13. Given the predominately negative ex-post real rates on foreigncurrency denominated deposits from 1976 to 1982 and the positive realreturns on peso denominated assets, one might not expect a massive shifttowards dollar assets. Nonetheless, Table 2 shows such a shift. BetweenDecember 1979 and December 1981 the amount of dollar deposits doubled, thistranslated into a tripling, in peso terms, through the change in theexchange rate. While a large inflow of funds from Argentina constituted animportant part of the increase in dollar assets, it also seems thatdomestic asset holders switched into dollars, apparently anticipating thatthe peso would be devalued faster than it was, even during the pre-announc-ement period. In late November, 1982, a precipitous devaluation of over100% did take place, contributing to a 200% devaluation between December1981 and December 1982. As a result, dollar asset holders made extremelylarge profits. Correspondingly, debtors in foreign currency were badlybattered.

14. In the loan market, pessimistic expectations regarding the valueof the peso were reflected in an aversion to assuming the exchange risk offoreign liabilities. Evidence for this can be seen in Table 1, where the

- 134 -

real rates for foreign currency denominated loans, i.e., the rate adjustedfor devaluation and domestic inflation, remain about twenty percentagepoints or more, below the rates on credits in pesos (ten or more points forpreferential credits). Apparently, there were not enough agents willing totake on the foreign currency risk of borrowing in dollars or lending inpesos to bring down this spread, even though both activities were consis-tently profitable through 1982. The banks, in particular, did not engagein dollar to peso arbitrage, instead they maintained roughly balanced netpositions.

15. Like the deposit rates, the real lending rates were often nega-tive during the 1976-82 period. Of the four rates displayed in Table I(domestic and foreign, prime and normal), the rate charged on normalborrowing in domestic currency was positive most often. It was lastnegative 1978, and did not become negative in 1982 despite surging infla-tion, as did the other three lending rates.

16. With respect to credits denominated in foreign currency, duringmost of the period an appreciating real exchange rate applied downwardpressure on the real cost of credit, keeping the real rate negative and lowin comparison to peso credits, up until the devaluation. By the mid-1982,the real exchange rate had appreciated to the point where, given thechanging international conditions, a massive devaluation was required. Asmentioned at the beginning of this section, this devaluation does not showup in the figures in Table 1 because the real rates were calculated lookingsix months forward from December. Thus, the 1981 figures did not catch theshift and the 1982 figures reflect developments occurring after the shift.However, for those dollar liabilities taken on before and due after thedevaluation the real rate was very high ex-post.

17. In sum, although the liberalized Uruguayan credit markets had arespectable record of adjustment to price and exchange rate developments,relative to most developing countries, the volatility of the real ratescreated some major difficulties for the economy. In particular, the highlypositive real lending rates in local currency and the recent devaluationsgenerated large debt service burdens for the Uruguayan firms. Debt-equityratios also increased as a result of refinancing this debt service and themassive increase in the peso equivalent of the firms' dollar liabilities.Moreover, while the corresponding high deposit rates encouraged financialsavings, they also made equity finance unattractive. These financialmarket difficulties combined with a sharp fall in external demand, partic-ularly from Argentina, have left a large portion of the private sector inUruguay in a precarious financial position.

The Pattern of Deposit Rates, End of Year 1982

18. As evidenced in Table 3, there was virtually no incentive fordepositors to commit funds for longer than six months at the end of 1982.Although the spread between saving and six month term deposits was substan-tial, the spread between term deposits of six months and those of over ayear was negligible. Since these rates were market determined, this levelyield curve indicates that neither the banks nor depositors were interestedin long term commitments.

- 135 -

Table 3

Deposit Rates, End of Year 1982

Domestic Denomination:

Savings 24.2%Term 6 mo. or less 66.2%Term over 1 year 67.3%

Source: Banco Central del Uruguay, Boletin Estadistico.

The Pattern of Lending, End of Year 1982

19. At the end of 1982, publicly owned lending institutions providedabout forty percent of all credits extended to the private sector (Banco dela Republica, 6/83). Most of these credits were extended to prioritysectors on near market terms, though some subsidized credits continued tobe made available. In particular, credits for agriculture and for housingappear to have borne below market rates. The remaining credits were lentlargely at short terms by private institutions. The average lending rateswere about twenty percentage points above the prime rates for credits indomestic currency, and about one percentage point above the prime rates forcredits in foreign currency.

20. Among the features that permitted the public institutions to lendat below market rates are: tax exempt status; large non-interest bearingdeposits from the public sector; special fees and commissions relating tocustoms charges; and the backing of the Treasury on somne priority lending.these factors reduced the public intermediaries' cost of funds and requiredspreads. With respect to private lending, a very rough measure of thespreads can be derived from Table 4. Ignoring possible currency arbitrageand using the term deposit rates as proxies for the marginal cost of funds,the intermediation spreads on average loans were as foLlows:

Table 4

Intermediation Spreads, End of Year 1982

NominalDomestic Denominations

average credits 76.3%6 month term dep. 66.2%

"spread" 10.1%

Foreign Denominationsaverage credits 18.2%6 month term dep. 10.2%

"spread" 8.0%

Source: Table 1.

- 136 -

The spread for intermediation in domestic currency exceeded that for inter-mediation in foreign currency by about two percentage points, a differencewhich largely reflects the expectations of devaluation. It is also worthnoting that the spread between the prime lending rates and the depositrates (in pesos) declined sharply after 1978, reflecting the increasedcompetition in banking (see Spiller and Favaro). However, -spreads onaverage loans were high and increased somewhat over the same period.

- 137 -

Sources and References

Ardito Barletta, N., M. Blejer, and L. Landau (1983), EconomicLiberalization and Stabilization Policies in Argentina, Chile,and Uruguay: Applications of the Monetary Approach to the Balance ofPayments, World Bank.

Banco Central de Reserva del Peru (various dates), Boletin, (Lima: BancoCentral de Reserva del Peru, various issues).

Banco Central de Reserva del Peru (various dates), Memnoria, (Lima: BancoCentral de Reserva de Peru, various isses).

Banco Central del Uruguay (various dates), Boletin Est:adistico (Uruguay:Banco Central del Uruguay, various issues).

Bangladesh Bank (various dates), Bangladesh Bank Bulletin (Dhaka:Bangladesh Bank, various issues).

Bangladesh Bank (various dates), Economic Trends (Dhaka: BangladeshBank, various issues).

Bangladesh Bureau of Statistics, Ministry of Finance and Planning(various dates), Monthly Statistical Bulletin of Bangladesh (Dhaka:Government of the People's Republic of Bangladesh, various issues).

Bank of Thailand (various dates), Quarterly Bulletin (Bangkok: Bank ofThailand, various issues).

Central Bank of Kenya (various dates), Economic and Financial Review,(Nairobi: Central Bank of Kenya, various issues,'.

Central Bank of Nigeria (various dates), Annual Report., (Lagos: CentralBank of Nigeria, various issues).

Central Bank of Nigeria (various dates), Economic and Financial Review,(Lagos: Central Bank of Nigeria, various issues)'.

Central Bank of Nigeria (various dates), Monthly Report, (Lagos: CentralBank of Nigeria, various issues).

- 138 -

Central Bank of the Republic of Turkey (various dates), QuarterlyBulletin (Ankara: Central Bank of the Republic of Turkey, variousissues).

Darby, M. (1975), "The Financial and Tax Effects of Monetary Policy onInterest Rates," Economic Enquiry, June 1975, pp. 266-276.

Fama, E. (1975), "Short Term Interest Rate as Predictors of Inflation,"American Economic Review, April 1975, pp. 325-338.

Fisher, I. (1930), The Theory of Interest (MacMillan, 1930).

Hanson, J., and J. DeMelo (1983), "The Uruguayan Experience withStabilization and Liberalization,1974-1981", Journal ofInteramerican Studies and World Affairs, Nov. 1983, pp. 477-508.

Hanson, J., and J. Demelo (1985), "External Shocks, Financial Reforms andStabilization Attempts in Uruguay: 1974-83," World Development,August 1985 (forthcoming).

Lanyi, A., and R. Saracoglu (1983), Interest Rate Policies in DevelopingCountries, IMF Occasional Paper #22, Oct. 1983.

Long, M. (1983), "Review of Financial Sector Work," World Bank, INDFD,1983.

Mathieson, D., and R. McKinnon (1981), "How to Manage a RepressedEconomy," Essays in International Finance, No 145, (Princeton:Princeton University, Dept. of Economics, December 1981).

Mathieson, D., and R. McKinnon (1983), "Foreign Exchange and FinancialPolicies for Repressed and Liberalizing Economies", IMF StaffPapers, 1983.

Ministry of Economic Planning and Community Affairs (1979), DevelopmentPlan 1979 to 1983, (Nairobi: Ministry of Economic Planning andCommunity Affairs, 1979).

Ministry of Economic Planning and Community Affairs (1983), DevelopmentPlan 1983 to 1988, (Nairobi: Ministry of Economic Planning andCommunity Affairs, December 1983).

- 139 -

McKinnon, R. (1973), Money and Capital in Economic Development,(Washington, D.C.: Brookings Institute, 1973).

Saracoglu, R. (1984), "Expectations of Inflation and Interest RateDetermination," IMF Staff Papers, March 1984, pp. 141-178.

Shaw, E. (1973), Financial Deepening in Economic DeveLopment, (New York:Oxford University Press, 1973).

Spiller, P., and E. Favaro (1984), "The Effect of Entry Regulation onOligopolistic Interaction: The Uruguayan Banking Sector", BellJournal of Economics, 1984.

Summers, L. (1983), "The Nonadjustment of Nominal Interest Rates: A Studyof the Fisher Effect," in J. Tobin Macroeconomics: Prices andQuantities, (Brookings, Washington, DC, 1983).

La Superintendencia de Banca y Seguros (various dates),, BoletinEstadistico, (Lima: La Superintendencia de Banca y Seguros,various issues).

Tanzi. V. (1980), "Inflationary Expectations, Economic: Activity, Taxes,and Interest Rates," American Economic Review, Ma,rch 1980,pp.12-21.

The World BankH Publications Order FormSEND TO: YOUR LOCAL DISTRIBUTOR OR TO WORLD BANK PUBLICATIONS

(See the other side of this fom.) P.O. BOX 37525WASHINGTON, EI.C. 20013 U.S.A.

Date

Name Ship to: (Enter if different from purchaser)

Title Name

Firm Title

Address Firm

City State_ Postal Code_ _ Address

Country Teblephone I ) City _ _ St3te_ FPostal Code

Purchaser Reference No. _ Country - 'IIephone

Check your method of payment.Enclosed is my E Check L international Money Order O Unesco Coupons l Inlernational Postal Coupon.Make payable to World Bank Publications for U.S. dollars unless you are orderin, from your local distributor.

Charge my O VISA U MasterCard U American Express U Choice. iCredit cards accepted only for orders addressedto World Bank Publications.)

Credit Card Account Number Expiration Date Signature

U Invoice me and please reference my Purchase Order No.

Please ship me the items listed below.

Customer InbernalStock Number Autbeou Title Routing Code Quantity Unit Price Total Amount S

All prices subject to change. Prices may vary by country. Allow 6-8 weeks for idelivery.

Subtotal Cost $

Ibtal copies - Air mail surcharge if desired ($2.00 each) $____

Postage and handling for more than two complimentary items ($2.00 each) S___

T_tal $

iBR-0o053 Thanrg you for yrour order.

DISTRIBUTORS OF WORLD BANK PUBLICATIONSALGERIA FINLAND KOREA, REPUBLIC OF SPAINOffce des PubbWa Akateeiiin Kiriakauppa Pan Korea Book Corporation Mundi-Prensa Libw, S.A.

UnivrrUr P.O. Box 128 P.O. Box 101, Kwangwhamun Catelo 371, place centrel de UlAknoun SF-00101 Seoul 28001 MadridAlgiers Helsnki 10 KUWArI SRI LANKA AND THE

ARGENTINA FRANCE MEMRB MALDIVESCarlos Hirsch SRL World Bank Pubikations PO. Box 5465 Lake House BookshopGaleria Guemes 66 Avenue d'iena MALYSIA Box 244Florida 165, 4th Floor-Okc. 4531465 75116 PYi UniveAity of MalaSya Cooperative ta°dmer tla hA1333 Buenos Aires GERMANY FEDERAL Bookhop, Limited GnerM watha

AUSTRALIA, PAPUA NEW REPUBUC OF P O. Box 1127, Jalan Pantai Baru 2GUINEA, Fill, SOLOMON UNO-Verlag Kuala Lumpur SWEDENISLANDS, AND VANUATU D-5300 Bonn I For singe titlesInfo-Line Sinrockstrasa 23 1 FTEC ABCE! PzesKuni.

Overeas Document Delivery GREECE San Lorenzo 153-11. Col. del HovbokhandelSydney, NSW 2001 MEMRB Valle Deleg. Beniito Juarez At:M.Ed eebcSydney, NSW 2001 24, Ippodamou Street 03100 Mexico City Regeringsgatain, Box 16356

AUSTRIA Athens-11635 ~~~~~~~~~~~~~~S-103 27 StockholmGeSrolA Aanen-Co. MOROCCO For Subscription orders:A-1011 Wien HONG KONG, MACAU MEMRB Wennergren-Williams ABGa ben 31 Asia 2000 Ltd. 2 Rue Molere Racine Box 30004

Graben 31 6 Fl., 146 Prince Edward Road, W Casablanca S-104 25 Stockholm

BAHRAIN Kowloon NETHERLANDS SWeTVZERLANDPOMRBo 2on1Kon Medica Books Europe, BV (MBE) Librairie payotM0 x 2 10 HUNGARY Noorderwal 38, 6 Rue GrenusManama Town 317 Kultura 7241 BL Lochem Case postal 381

BANGLADESH PO. Box 139 NEW ZEALAND CH 211 Geneva 11Micro Industries Development 1389 Budapest 62 R Hill nd Son Ltd TANZANIA

Assistance SocBety ( AS) INDIA Private Bag Oxford University PressGDPhO Box 80 UBS Publishers' Distributors Ltd. New Market P.O. Box 5299Dhsk.a Post Box 7015 Auckland Dar es Salaam

BELGIUM New Delhi 110002 NIGERIAPublications des Nations Unies 10 First Main Road University Press Limited Central DartmentAv. du Roi Gandhi Nagar Three Crowns Building Jericho 306 Silom Road1060 Brussels Bangalore 560009 Private Mail Bag 5095 BangkokBRAZIL Apeejay Chambers, P.O. Box 736 Ibadan TRINIDAD4cTOBAGOPublicacoes Tecnicas 5 Wallace Street NRASyTeINticD Stde UOB it

Internacionais Ltda. Bombav 400001 NORWAY Systematcs Studies UR tRua Peixoto Gomide, 209 8il-B, Chownnghee Lane PO0 Boxh11775entum CErnpaein Roa01409 Sao Paulo, SP Calcutta 700016 Osl ox 117TrSnidad.Cur WestIdCANADA 7/188, I(CA). Swarup Nagar PAKISTAN TUNISIALe Diffuseur Kampu O PKSATUIALa.8,151Apr Sitfestrivamang Roa Mirza Book Agency Societe Tunisienne de DiffusionBocherv8 150 Amperec Stee uigngamRoadka 65, Shahrah-e-Quaid-e-Azam 5 Avenue de CarthageBoucherville6 Quebec NMagas 6 P.O. Box No. 729 Tunis

J4B 5E6 Madras 600034 ~~~~~~Lahore 3 TURKEYCHILE INDONESIA PAAAHaset Kit vA AS.Editorial Renacimiento Pt. Indira Limited EdcinAMAbei Clua 6, sTaZsMiraflores 354 JI. Sam Ratulangi 37 nes LibrA Cultural 4, [lCaddesiSantiago Jakarta Pusat AnamEna 1A UGANDA

COLOMBIA P.O. Box 181 Avn 7 Esana 16 Uganda BookshopEnlace Ltda. IRELAND PERaU P.O.Ugnd Boxk7h45Apartado Aero 34270 TDC Publishers Editorial DR arrollo SA KampalaBogota DE. 12 North Frederick Street Edito er AmApartado Aereo 4430 Dublin Apartado 384 UNITED ARAB EMIRATESCali, Valle Lima MEMRB

ISRAEL PHIUPPINES P.O. Box 6097COSTA RsCA Jerusem Post National Book Store Sharjah

Callee11-13 Jerusalem Post Building 701 Rizal Avenue UNIDKINGDOMANDCalle 11-13 ~~~P.O. Box 81 Mtr aiaUIE IGO NAv. Femandez Guell Romena Jerusalem 91000 Metro ania NORTHERN IRELANDSan Jose PORTUGAL Microinfo Ltd.

C6TE D'IVOIRE I~~TALY . Livraria Portugal P. 0. Box 3Cente d'Editon et de R iffusion Licosa Conuntisionaria Sansonji Rua D Carino 70-74 Alton, Hampshire GU 34 2PG

Afncaines (CEDA) Via Lamarmora 45 1200 Lisbon England04 B. P 541 Casella Postale 552 SAUDI ARABIA VENEZUELAAbidjan 04 Plateau 50121 Florence Jarir Book Store Librena del EsteCYPRUS JAPAN P 0. Box 3196 Aptdo. 60.337MEMRB Eastern Book ServiceP.O. Box 2098 37-3 Hongo 3-Chome, SINGAPORE, TAIWAN, WESTERN SAMOANicosia Bunkyo-ku 113 BURMA Wesley Bookshop

DENMARK Tokyo Information Publications PO. Box 207SamfundaLitteratur JORDAN Private, Ltd. ApiaRosenoerns Alle 11 MEMRB 02-06 1st Fl., Pei-Fu Industrial ZIMBABWEDK-1970 Copenhagen V RO Box 3143 BLdg., 24 New Industrial Road Textbook Salf Pvt. Ltd.EGYPT ARAB REPUBUC OF Jabal Singapore Box 3799Al Ahram Arnrnan SOUTH AFRICA HarareAl Galaa Street KENYA Oxford University PressCairo Africa Book Service (E.A.) Ltd. pSuhBoern A41ca

P O. Box 45245 Cape Town 8000Nairobi

The World Bank

Headquarters European Office Tokyo Office U1818 H Street, N.W. 66, avenue d'1ena Kokusai BuildingWashington, D.C. 20433, U.S.A. 75116 Paris, France 1-1 Marunouchi 3-chome

Telephone: (202) 477-1234 Telephone: (1) 47.23.54.21 Chiyoda-ku, Tokyo 100, JapanTelex: WUI 64145 WORLDBANK Telex: 842-620628 Telephone: (03) 214-5001

RCA 248423 WORLDBK Telex: 781-26838Cable Address: INTBAFRAD

WASHINGTONDC

ISSN 0256-2235Cover design by Bill Fraser ISBN 0-8213-0782-7