macroeconomic determinants of bank spread in brazil irae revised[1]

8/18/2019 Macroeconomic Determinants of Bank Spread in Brazil IRAE Revised[1]

1/28

Macroeconomic Determinants of Bank Spread in Latin America:

A Recent Analysis with Special Focus on Brazil*

José Luis Oreiro**

Luiz Fernando de Paula***

Astract: Latin America has one of the highest interest margins in the world; furthermore,

credit to private sector and bank spread are negatively correlated. Brazil, in particular, has

one the highest bank spread in the world – it is even so far the highest one among the Latin

American economies. ndeed, despite of the decline in interest rates since mid!"###, bank

spread in Brazil continues e$tremely high in international terms, and in recent years has

stood at around %& percentage points. 'his paper intends to e$plore in the recent literature

on bank spread the discussion about what determines bank spread in Latin America, with

special focus on the Brazilian case, seeking in particular but not e$clusively to analyze the

macroeconomic determinants of bank spread in recent times.

!ey "ords( bank spread, Latin America, Brazilian banking sector

#$L %ode( )%*; )%%; +"

* The authors thank the anonymous referees whose comments improved the paper. All remainingerrors are the author’s responsibility.-- Associate professor of economics at Universidade de Brasilia and C!" researcher. #$mail%

&lcoreiro'terra.com.br .-** Associate professor of economics at Universidade do #stado do (io de )aneiro and C!"researcher. #$mail% luifpaula'terra.com.br .

+

mailto:[email protected]:[email protected]:[email protected]:[email protected]


2/28

&' (ntroduction

verall the scale of bank lending is low relative to economic activity, financial

depth is limited, banking sector is highly concentrated, and intermediation margins are high

in Latin America. ndeed, limited access to bank credit and uncertainty about financial

stability are factors that have contributed to economic volatility in the region /0ingh et al

&&12. Latin American financial systems are largely bank!based, with security markets

mostly small and ili3uid. 4inancial depth is low compared to developed countries and some

groups of developing countries".

Besides the low level of credit, the pattern of credit growth in Latin America

has been marked by boom and bust cycles. 5redit e$panded sharply in the early "##&s, in

part due to the increase of the capital inflows to the region, but collapsed in many cases

after the banking crises in the mid!"##&s and remained subdued for many years. nly

recently, that is after &&%, credit has begun to recover, due to the more strong economic

growth, easier global monetary conditions and progress in bank restructuring. ndeed, in

most Latin American countries, the unstable macroeconomic environment has been a

critical factor holding back financial system development and has generated a high

volatility of credit growth. 4or e$ample, high short!term interest rates used to fight inflation

or defend the e$change rate has added to banks6 funding costs and increased loan!default

rates.

Brazil is an interesting study case, as the economy has at the same time a low

credit!to!+78 and relatively high bank assets!to!+78 ratio. 7uring the high inflation

period /until "##%2, financial depth did not decrease due to the development of a broader

domestically!denominated inde$ed money and also the increasing development of a

modern clearing system in order to support the clients6 demand for clearing the checks.

9ore recently, the large banking portfolios of interest rate! and e$change rate!inde$ed

" ndeed, according to A7B /&&1, 12 the average ratio banking credit!to!+78 and credit plusmarket capitalisation!to!+78 were, respectively, : and %: in &&* in Latin America, muchlower than in other groups of emerging countries, such as )ast Asia and the 8acific /


3/28

government debt insulated banks against the monetary policy tightening and devaluation of

the currency during e$ternal crises.

n Latin America, credit in general is not only scarce but also costly. ndeed, the

region has one of the highest interest margins in the world and as would be e$pected, credit

to private sector and bank spread are negatively correlated. Brazil, in particular, has one of

the highest bank spread in the world – it is even so far the highest among the Latin

American economies.

A number of international studies have highlighted the importance of

macroeconomic factors – including inflation rate, level of interest rates and interest rate

volatility, +78 growth rate, capacity utilization etc. – in determining bank spread.

5onsidering the macroeconomic instability that has characterized the Brazilian economy

during the last 1 years –for e$ample the stop!go movement of the economy and the

e$tremely high short!term interest rates – it is to be e$pected that such factors would be

significant in e$plaining spread in Brazil. 'his issue has gained in importance as, despite a

decline in interest rates since mid!"###, bank spread in Brazil continues e$tremely high in

international terms, and in recent years has stood at around %& percentage points.* ne of

the main factors preventing credit growth in Brazil is the e$tremely high interest rates

levied on loans, which e$plains at least partly the high profitability of the ma?or retail

banks.

'his paper intends to e$plore in the recent literature on bank spread the

discussion of what determines bank spread in Latin America, with special focus on the

Brazilian case, seeking particularly but not e$clusively to analyze the macroeconomic

determinants of spread in recent times. 'he paper is structured into % sections plus this

introduction. 0ection develops the analytical approach of the determinants of spread

based on the conventional literature, while 0ection * briefly evaluates some case studies,

with special reference to Latin America. 0ection % sets out an analysis of the recent

evolution of bank spread in Latin America and Brazil and also assess the evidences on its

* Bank spread in Brazil is calculated by the 5entral Bank of Brazil, that uses the followingdefinition( @bank spread is defined as the difference between lending and deposit rates for 57Bscertificates of bank deposit. 'he average 57B rate for the set of financial institutions wascalculated from the average of the individual rates weighted by each institution6s net depositsC/Banco 5entral do Brasil, &&, p. 1&2. 4or the different meanings of bank spread see footnote %.

-


4/28

determinants in Brazil based on the empirical studies. 4inally, 0ection D summarizes the

paper6s main conclusions.

)' Determinants of ank spread: an analytical approach ased on the conentional

literature

'he conventional theoretical literature on the determinants of bank spread% has

developed around two ma?or approaches. 'he first /@monopoly modelsC2 grew out of a

seminal study by Elein /"#here D is the volume of deposits @producedC by the bank and L is the volume of loans. 'hetraditional assumption is made that the marginal cost of loans and deposits is positive andincreasing.


5/28

n such a conte$t, let r be the prevailing interest rate on the inter!bank market; r l

the interest rate charged on loans made by the bank; r d the interest rate paid by deposits

with the bank; α the compulsory reserves as a proportion of the bank6s deposits; εL the

interest elasticity of loan demand; ε7 the interest elasticity of deposit supply; 5HL the

marginal cost of loan services; and 5H7 the marginal cost of deposit services. 'hen,

supposing that the bank is risk neutralD and that its behavior is directed to ma$imizing

profits, it can be shown that the optimal interest margin on loans and deposits is given by


6/28

4rom e3uations /*2 and /%2 it can be shown that the bank6s interest margins on

loan operations and deposit taking is a growing function of its market share. 'herefore, any

reduction in the number of banking firms – resulting, for instance from bank mergers and

buyouts – will increase bank concentration and thus interest margins. ne of this model6s

results is thus that bank spread is a growing function of the degree of overall bank sector

concentration.

'he second approach grew out of a seminal study by Go and 0aunders /"#:"2#,

and conceives the bank not as a firm, but simply as an intermediary between the final loan

taker /firms2 and the final lender /households2. Gowever, this intermediation activity is

sub?ect to two types of uncertainty. 4irstly, there is uncertainty due to lack of

synchronization between deposits and loans. 'his lack of synchronization entails an interest

rate risk for the bank. n order to understand why, let us imagine that the bank encounters

une$pectedly high loan demand, e$ceeding the volume of deposits and its free reserves. n

this case, it will be forced to finance the surplus credit demand on the inter!bank market,

thus incurring a refinancing risk in the event the interest rate rises /9audos and +uevara

&&*, %2. n the other hand, if the bank encounters une$pectedly high deposit supply,

e$ceeding the volume of loans granted by the bank in the same period, it will then have to

apply those surplus funds on the inter!bank market. n that way, the bank will be incurring a

reinvestment risk in the event the interest rate falls /9audos and +uevara &&*, %2.

0econdly, the intermediation activity e$poses the bank to uncertainty regarding

the rate of return on loans. 'hat uncertainty results from the fact that a part of its loans will

not be recovered because of non!payment, voluntary or otherwise, by loan takers. 'he

percentage of non!performing loans, however, is not a variable known e$!ante by the bank,

which can only estimate a likelihood of default.

ne feature the Elein and Go and 0aunders approaches have in common is the

assumption that banks have market power, i.e. both approaches assume that banks are free

to set the interest rates charged on credit operations and paid on deposits. Jnlike the Elein

approach, however, Go and 0aunders assume that the bank is a risk!averse agent. n other

# n what follows, we will work with the most recent e$tension of the Go and 0aunders approachdeveloped by 9audos and +uevara /&&*2. 0ee, also, Allen /"#::2, 9c5hane and 0harpe/"#:12,and Angbazo /"##


7/28

words, the bank6s goal is not to ma$imize e$pected profit, but rather to ma$imize the

e$pected utility of profit. n that conte$t, they show that optimum spread /s-2 is given by

/9audos and +uevara, &&*, D2(

( ) ( ) ( )[ ] 21/2,,2H/2KK/

%

"2/2/

,

"

,

"&

,,

&

-

LM M L

L

L

D

D

L M D L L LW U

W U

D

DC

L

LC

s σ σ σ β

α

β

α

−++++−

++

+=

! where α7 is the linear intercept of the probability function of a deposit being

made at the bank, β7 is the sensitivity of the probability of a deposit being made at the bank

to variations in the deposit interest rate, αL is the linear intercept of the probability function

of a loan application to the bank, βL is loan application sensitivity to variations in the credit

operation interest rate; 5/L2L is the average cost of credit operations; 5/727 is the mean

cost of deposit!taking operations; W is the bank6s final stock of wealth;

−

2H/

2HH/

W U

W U is

the bank6s absolute degree of risk aversion"&; σL is the standard deviation of the yield on

loans /a measure of the bank6s credit risk2; σ9 is the standard deviation of the yield on

applicationsloans on the inter!bank market /a measure of the bank6s interest rate risk2; σL9

is the co!variance between credit risk and interest rate risk; L& M is the bank6s starting stock

of loans; and 9& is the bank6s initial net position on the inter!bank market.

4rom e3uation /12, it can be concluded that the determinants of bank spread are(

• 'he market structure and the level of competition in the banking sector( the greater

the interest elasticity of loan demand and deposit supply /i.e. the lower the values of βL e

β72, the smaller will be the optimum spread.

• 'he bank6s average operating cost(

+

D

DC

L

LC 2/2/

• 'he bank6s degree of risk aversion(

−2H/

2HH/

W U

W U

• 'he volatility of market loan interest rates( σ9

• 'he credit risk( σL

"& =ote that, as a result of the risk aversion hypothesis, JH/.2 N & and JHH/.2 O &.

1


8/28

• 'he co!variance between loan risk and interest rate risk( σL9

• 'he average size of the credit and deposit operations undertaken by the bank(

/LP72.

ne important aspect of the Go and 0aunders approach is that it leaves room

for the influence of macroeconomic variables in determining bank spread /0aunders and

0chumacher &&&, :"12. 'he volatility of interest rates levied on loans on the inter!bank

market is a direct reflection of the country6s macroeconomic stability. 'he less stable a

country6s economy – e.g. the greater the variation in the inflation rate and e$change rate –

the greater will be the resulting volatility of the basic interest rate "" and, conse3uently, the

greater the bank spread. n such a conte$t, spread can be reduced by macroeconomic

policies to reduce interest rate volatility.

9acroeconomic instability can affect bank spread through two other channels.

'he first is related to the degree of banks6 risk, that must to some e$tent reflect the

instability of the market environment where they operate. 'he less stable the environment,

the greater banks6 risk must be – as can be the case of interest rate risk. " 'hus, in a country

with a history of ma?or macroeconomic instability /high inflation. for instance2 banks

should face greater risks in their intermediation activity.. 'he second channel is the

covariance between interest rate risk and credit risk. A highly volatile basic interest rate will

be e$pressed to some e$tent in a highly variable level of real output. n such a conte$t,

firms6 profits will also be highly variable, increasing the likelihood of default at times when

profits fall below e$pected values. 'hus, macroeconomic instability is reflected not ?ust in a

highly volatile interest rate, but also in high credit risk, i.e. such instability generates high

co!variance between yield on loans and yield on inter!bank market applications. 4rom /12,

it can be seen that the greater such co!variance, the greater will be bank spread.

ne final remark on e3uation /12( the spread given by this e$pression should beunderstood as @pureC bank spread /9audos and +uevara &&*,


9/28

other variables that e$plain banks6 net interest margin, but which are difficult, if not

impossible, to incorporate into the theoretical model. 'hese variables reflect institutional

and regulatory aspects of banking activities. As a result, actual net interest margin

comprises two elements( @pureC bank spread /s-2 and the net interest margin /f2 e$plained

by institutional and regulatory factors.

+' Some (nternational %ase Studies with Special Reference to Latin America

A vast empirical literature on the determinants of bank spread has developed in

recent years. ne ma?or component of the literature has been concerned with testing

empirically the theoretical model of bank spread developed by Go and 0aunders /"#:"2.

Among the most important studies taking this approach are 0aunders and 0chumacher

/&&&2 and 9audos and +uevara /&&*2, and some of these studies will be described below.

9ost of this work uses the @pure spreadC estimation methodology pioneered by

Go and 0aunders. 'he methodology assumes that actual spread comprises @pureC spread

ad?usted upwards or downwards by implicit interest e$pense /e$emption from bank charges

for certain classes of customer2, by the opportunity cost of holding reserves and by capital

re3uirements resulting from regulatory standards and bank supervision. +iven that conte$t,

@pureC spread is estimated in a two!step process. 'he first step involves running a cross!

section regression for each bank6s net interest margin in the chosen country in a given year

/0aunders and 0chumacher, &&&, p.:"#2. 'hat e3uation is given by(

∑ ++=i

i jic jcic u !"M δ γ /D2

! where( ic !"M is the bank6s net interest margin i in country c in the period t;

jic is a vector of control variables /implicit interest e$pense, opportunity cost of re3uired

reserves and capital re3uirements for credit risk e$posure2 for each bank i in country c in

some period t; cγ is the regression constant, which is an estimate of @pure spreadC for all i

banks in country c at any time t, and ui is the residual.

n this first step, e3uation /D2 is processed for each country in the sample over

the study period. 'he @pure spreadC estimates obtained in the first step vary over time and

3


10/28

among countries."* Accordingly, in the second step, a regression is run with panel data from

the @pureC spread estimates obtained in the first step against a series of variables that reflect

the market structure and intermediation risks. 'he e3uation to be estimated is given by(

∑− ++= ""&

ccctc σ θ η θ γ /


11/28

to include operating costs and a direct measure of the degree of competition /Lerner inde$2

as e$planatory variables.

9audos and +uevara used a one!stage panel data regression in order to estimate

the theoretical model they developed of the determinants of spread, measured by net

interest margin, and considering as e$planatory variables a number of bank and country

characteristics for each period. 'he e$planatory variables of the theoretical model, all

e$pected to relate positively with spread are( competitive structure /measured by the Lerner

inde$2, operating costs /in relation to total assets2, degree of risk aversion /ratio of net

worth to total assets2, interest risk, credit risk, interaction between credit risk and interest

risk /measured by multiplying the two variables2 and average size of operations /log of the

volume of loans2.

n addition to the variables of the theoretical model, they also consider, as

e$planatory variables, implicit interest payments /measured by net operating e$penditure of

non!interest revenues as a percentage of total assets2, the opportunity cost of bank reserves

/ratio of li3uid reserves to total assets2 – both e$pected to relate positively to spread – and

3uality of management – e$pected to relate negatively to interest margin. Gowever, as a

pro$y for 3uality of management, they use the ratio operating costsrevenues, an increase in

which lowers 3uality of management, resulting in a smaller interest margin; thence the

negative sign between the ratio and net interest margin is to be e$pected. 'he results of that

study show that most of the variables posited by the theoretical model are statistically

significant and have the e$pected sign, i.e. interest margin relates positively with the Lerner

inde$, operating costs, bank risk aversion, credit risk and interest risk. 0ignificant, positive

coefficients were also yielded by implicit interest payments and opportunity cost of bank

reserves, and significant, negative coefficients by the operating costsrevenues ratio, as

e$pected by the authors.

Brock and Io?as!0uSrez /&&&2 conducted an empirical analysis using panel

data on determinants of bank spread in Latin American countries. Jsing a sample of banks

in si$ Latin American countries /Argentina, Bolivia, 5olombia, 5hile, 9e$ico and 8eru2

over the period "##" to "##D, they investigated why bank spread had not diminished in

these countries in a period of financial liberalization resulting from reforms to the banking

sector, marked particularly by reductions in reserve re3uirements and in direct restrictions

,+


12/28

on credit and interest rates. 4or that purpose, they analyzed the evolution of si$ measures of

e$!post spread /net interest margin2, finding significant differences among these

measurements in all the countries. n addition, they use the model of Go and 0aunders

/"#:"2 with a two!step panel regression using bank!specific variables, in order to estimate

the determinants of spread for each of the countries individually, e$cept 9e$ico. n the first

step, which derived @pure spreadC, Brock and Io?as!0uSrez controlled the microeconomic

factors"1 and, in the second step, they ran a regression of the @pure spreadC for each country

e$plained by the following variables( interest rate volatility, inflation rate and +78 growth

rate.

'he first step results shown that some of the variables relate positively and

significantly in some of the countries( capital!asset ratio /Bolivia and 5olombia2, cost ratio

/Argentina and Bolivia2 and li3uidity ratio /Bolivia, 5olombia and 8eru2. n the other

hand, contrary to e$pectations, non!performing loans ratio did not relate positively with

bank spread in any of the countries, while in two countries /Argentina and 8eru2 the

correlation was negative and significant. 'he authors suggest that this result may be

associated with inade3uate loan loss provisioning( higher non!performing loans would

reduce banks6 income. n the second stage regression, using macroeconomic variables, the

best results were given by interest rate volatility, inflation rate and +78 growth rate. 'hus

macroeconomic uncertainty, represented by interest rate volatility /Bolivia and 5hile2 and

inflation /Bolivia, 5olombia, 5hile and 8eru2, related positively with spread, corroborating

the results from developed countries. 4inally, economic growth rate yielded non!significant

coefficients /of varying sign2 in all the countries. 'he authors conclude, overall, that spread

in Bolivia is e$plained by microeconomic factors; in 5hile and 5olombia, by both macro

and microeconomics factors; while spreads in Argentina and 8eru are not really e$plained

by either macro or micro variables.

ne recent study /+elos &&D2 analyzed the evolution of e$!ante spread and e$!

post spread in Latin America and the determinants of e$!post spread in emerging countries,

considering bank!specific data in the period "### to && for :1 developing countries,

among them "% Latin American countries. 4rom the descriptive evidence, +elos observes

"1 'he variables considered are non!performing loan ratio /non!performing loanstotal assets2,capital ratio /e3uitytotal assets2, cost ratio /overhead and other operating costsperforming loans2and li3uidity ratio /short!term assetstotal deposits2.

,,


13/28

that in the Latin American countries the credit+78 ratio is low, while e$!ante and e$!post

spread levels are high by international standards. n his econometric estimations, the

e$planatory variables he uses for interest margin are bank!level characteristics /measured

by bank size, bank e3uity, overheads costs and a dummy for foreign ownership2, several

country!level characteristics /competition, reserve re3uirements, deposit rates, indirect

ta$es, legal protection and availability of information about potential borrowers2 and

macroeconomic characteristics /+78 growth, inflation, volatility of inflation and country

risk ratings2.

+elos /&&D2 estimated @cross!countryC regressions for && and the results

suggest that, of the bank!level characteristics, only bank size and overhead costs are

significant /and relate positively2. f the country!level and macroeconomic features,

deposit rate and reserve re3uirements are associated positively with bank spread, while

+78 growth displays a significant negative correlation, a result associated with banks6

e$ercising their market power. Gowever, concentration does not correlate significantly with

spread, which the author associates with the significant relationship between concentration

and overhead costs. Ge also estimated panel regressions with data for "### and &&,

confirming the relationships of the significant variables in the previous regression, although

reserve re3uirements showed reduced significance because the related data do not vary over

time. 'he estimation also confirms the significance of the positive coefficients for legal

structure and ta$es and the negative coefficient for foreign ownership.

0umming up, empirical works show evidences that high banking spread in Latin

America is determined for both microeconomic /high operating costs, poor loan 3uality,

high capitalization, and reserve re3uirements2 and macroeconomic factors /interest rate

volatility, +78 growth and inflation2. )mpirical evidence suggests that microeconomic

factors have been the main determinant of spreads in Bolivia; micro and macroeconomic

factors impacted on spreads in 5hile and 5olombia /Brock and Io?as 0uarez, &&&; Bara?as

et al, "##:, &&&2; macroeconomic factors were more important in determining Brazilian

spreads that were particularly high at large banks due to market power /Afanasieff, Lhacer

and =akane &&2"D; neither micro or macroeconomic factors ade3uately e$plained the

evolution of spreads in Argentina and 8eru /Brock and Io?as 0uarez, &&&2. A comparison

"D 0ee more on the determinants of bank spread in Brazil in the ne$t section.

,-


14/28

of spreads in Latin America with the industrialized countries found that the main difference

is the effect that non!performing loans have had on spreads, with deteriorating loan 3uality

associated with narrower spreads in Latin America. t is suggested that this feature was

indicative of inade3uate loan loss provisioning, or that banks with large amounts of bad

loans lowered spreads in an attempt to grow out of difficulties /Brock and Io?as 0uarez

&&&2. 4inally, +elos6 findings suggest that in Latin American countries, interest rates are

higher, banks less efficient and reserve re3uirements greater than in other emerging

countries, and that these factors have significant impact on spread.

,' -eriew of ank spread in Brazil

%.". )volution of bank spread in recent times( some empirical evidences

Loan interest rates charged in Brazil figure among the highest in the world,

according to 94 figures. 4igure " shows that, in "##%, the average spread for both

corporate and the personal sectors was around "& in the Brazilian banking system(

appro$imately eight times higher than the second!highest rate charged in any country in the

sample."


15/28

(egion of margins 7percentage private sector

countries 7percentage8 of assets8 79 of :;!8

;eveloped countries + -illiams /&.

,/


16/28


17/28

wary of the risk of contagion by the 9e$ican crisis, until reaching a plateau of

appro$imately %& at the start of &&&. n particular the change in the e$change rate

regime in "### – from a semi!fi$ed e$change rate to a floating e$change rate – was a

structural factor that contributed for the reduction of bank spread, as during the Ieal 8lan

/"##%!"###2, short!term interest rate was used to face speculative attacks on domestic

currency in order to preserve a stable e$change rate. Gowever, spread has continued at

those – still e$tremely high – levels ever since. 0pread has continued at those – still

e$tremely high – levels ever since.

=ote( Average bank spread related to operations with preset interest rate.

4igure *. Bank spread in Brazil .&//,0)1123. 0ource( 5entral Bank of Brazil

ne hypothesis to e$plain why spreads are so high in Brazil might be banks6

market power, evidence of which is the increasing concentration of banking in recent times.

ndeed, some recent studies of the Brazilian banking sector – e.g., Belaisch /&&*2 – show

that the market structure prevailing in this sector is essentially non!competitive. n that

conte$t, with few incentives to increase their operating efficiency, banks operate with high

spreads, either as a way of generating revenue sufficient to cover their high costs or as a

result of their ability to price their services at levels substantially above the marginal cost of

producing bank services.

,1


18/28

ne factor supporting the hypothesis that the problem of spread in Brazil results

from banks6 market power is the recent tendency for concentration to increase in the

banking sector. n the period "#::!&&*, the "1 largest banks6 market share in banking

system total assets increased from around # in Tune "#:: to appro$imately %


19/28

rates, i.e. spread. Ge suggests that this phenomenon is caused by the possibility of a re!

allocation among the components of the bank assets, or even incorporating into mark!up the

risk premium involved in credit. n that regard, inflation has a negative effect on level of

activity by inducing an increase in bank loan rates. n the other hand, the statistical tests

suggest that an increase in productive capacity utilization would reduce spread, thus

pointing to a pro!cyclic effect.

Another study by Afanasieff, Lhacer and =akane /&&2 identified two stylized

facts about spread behavior after the Ieal 8lan( /a2 a marked fall in interest rates after

"##1&; and /b2 persistently high dispersion among bank loan rates. 'hese facts provided the

rationale for applying the methodology first used to determine bank spreads by Go and

0aunders /"#:"2. 'he first step involved panel data for "% commercial banks between

4ebruary "##< and =ovember &&&, so as to reflect how spread was influenced by

individual /bank!level2 microeconomic variables", i.e., those relating to bank!specific

characteristics. 4rom that panel, it was possible to obtain an estimate of @pureC spread /see

0ections and * of this paper2. 'he second step involved a structural model to estimate the

long!term influence of macroeconomic variables – market interest rates, a measure of risk

premium /5!bond spread over a J0 'reasury bond of e3uivalent maturity2, inflation rate,

output growth rate, compulsory reserves on sight deposits, and financial ta$ rates – on the

@pureC spread calculated previously.

'he results of the first!step regressions show the following variables to be

statistically significant( non!interest!bearing deposits to total assets, operating costs, service

revenue to total operating revenues – all of which have a positive effect on bank spread –,

as well as a dummy for foreign banks, whose negative sign indicates that such banks charge

smaller average spreads. 'he coefficients estimated in the second step were significant,

suggesting that macroeconomic aspects are prominent as ma?or determinants of spreads in

& A more stable international environment, a fall in the overnight rate and measures adopted by the

5entral Bank of Brazil all contributed to a reduction in spreads /8aula and Alves Tr. &&*, *1:2. 'he5entral Bank measures included particularly a reduction in compulsory reserve re3uirements, from


20/28

Brazil. 'he results of the regression suggest that spread tends to grow with rises in basic

interest rate, risk premium, output growth and ta$es. 5ontrary to e$pectations, the rate of

inflation affects spread negatively, possibly because inflation may be capturing the effect of

banks6 appropriation of seigniorage on spread.

Another important study of determinants of bank spread in Brazil was conducted by

the 5entral Bank of Brazil in connection with the pro?ect @nterest rates and bank spreadC.

8ublished in the form of annual reports starting in "###, this study offers an accounting

breakdown of spread*, in addition to other econometric studies of the determinants of

spread in Brazil. Bank spread in Brazil is broken down on the basis of the margins charged

by a sample of banks – a sample e$tended from &&% onwards, to take in a larger universe

/commercial banks and multi!banks, including state!owned ones2 encompassing all the

banks operating in Brazil for which information /on their fi$ed!rate, freely!allocated credit

operations only2 is available at each base date. 'he following components are considered(

/a2 a residual corresponding, by and large, to bank net margin; /b2 ta$ wedge, including

direct and indirect ta$es /ta$ on financial operations – 4, among others2; /c2 5redit

+uarantee 4und /@ Fundo #arantidor de CréditoC – 4+52%; /d2 overhead costs; /e2

compulsory reserves on banks6 deposits /demand deposits, time deposits and saving

deposits2; and /f2 default /provision e$penses for non!performing loans2.

4igure % shows how each of these components participate in bank spread in

Brazil, from &&& to &&*, now using the methodology revised in &&%1. 7ecomposition of

spread was calculated by 5entral Banks of Brazil using a sample of #* banks in &&&, "&&

banks in &&", D: banks in && and


21/28

commercial banks and universal banks /public and private onesD2. 'he analysis considers

only preset lending interest rates for both individuals and corporate.

4igure %. Accountin4 decomposition of ank spread in Brazil' 0ource( 5entral Bank ofBrazil.

4rom the accounting decomposition of spread, the most important constituent

factors are, respectively, net interest margin /a &&&!&&* average of D.#2 and overhead

/D.&2, followed by ta$ wedge /".D2 and provision e$penses /"#.#2. 5ompulsory

reserve re3uirements, the least important item in the accounting decomposition, came to

represent a relatively more significant effect in && /#." of spread2, as a result of the

imposition of additional compulsory reserve re3uirements that year.

4or econometric tests it is supposed that the following structural e3uation is

valid(

D Jnfortunately 5entral Bank of Brazil does not divulge data on bank spread for individual banks.n any case, one should consider that after the privatization of a lot of state!owned banks during themid!"##&s the remaining public banks – including the two federal giant banks, Banco do Brasil and5)4 – are performing more or less like private banks.

-+


22/28

ln spreadR &β 'rend P "β ln selic P ,β ln adm P *β ln risk P %β ln imp P 1β ln comp

/D2

where( iiβ /iR &,..., 12 are the estimated parameters, 'rend is a deterministic

trend that controls other variables which may affect spread, but are not included in the

e3uation above


23/28

an appro$imation to banks6 gross mark!up, given that time deposits and overnight rates

behave similarly; /ii2 a measure of country risk premium /5!Bond yield over a J0 'reasury

bond yield of e3uivalent maturity2; /iii2 the ratio of overhead to credit volume; and /iv2

indirect ta$es.

'hey test for co!integration among the variables and find the following relative

values for 0eptember &&"( risk component /%12, overhead /&2, indirect ta$es /"#2

and 0elic overnight rate /"D2. n this analysis of bank spreads, risk!related variables

played a greater part than loan!loss costs, as in the study carried out regularly by the

5entral Bank. 'his may be e$plained by the forward!looking nature of the risk!related

variables with regard to future scenarios, while non!performance costs, relating to past

losses, are retrospective. n this way, as &&" was a year of uncertainty in Brazil6s economy,

the influence of the risk component in spread increased, as was to be e$pected. 'he

importance of the 0elic interest rate in determining spread may be understood differently.

Brazilian government debt has an important specificity( a larger share of government bonds

are inde$ed by 0elic interest rate*&. 'hese bonds are called Letras Financeiras do 'esouro

(LF'). 'he inde$ation by 0elic made L4'Hs completely free of the interest rate ris& , so that

they can be considered a perect su+stitute for banking reserves /Barbosa &&D,.*12. 'he

e$istence of such a kind of bonds that are highly li3uid and with a high nominal and real

rate of return imposes a very high opportunity cost for loans to the private sector, increasing

the bank spread /8aula and Alves Tr. &&*, *D"2.

reiro et al /&&D2 analyze the bank spread in Brazil using a multiple regression

analysis with the ob?ective of finding what macroeconomic variables determine, directly or

indirectly, the banking spread in the period "##1!&&*. 4or this purpose they use the

following variables in the model( interest bank spread, 0elic short!term interest rate,

volatility of interest rate /as a pro$y of the bank6s interest rate risk2, industrial output /as a

pro$y of the +782, e$tensive consumer price inde$ /85A2 and reserve re3uirements on

demand deposits /a regulatory variable under 5entral Bank6s control2. 'he results found in

the variance decomposition and impulse!response function showed that the high volatility

of the short!term interest rate /0elic2 and its level, and the industrial output were the main

macroeconomic determinants of the banking spread in Brazil.

*& According to Brazilian 5entral Bank, the share of 0elic inde$ed bonds in internal government netdebt was about *1 in the first 3uarter of &&:. 0ee www.bcb.gov.br

--

http://www.bcb.gov.br/http://www.bcb.gov.br/


24/28

5' %onclusion

'his paper did a survey of the literature about the determinants of bank spread

in Latin America, with both focus on Brazil and the macroeconomic determinants of

spread. According to the literature, macroeconomic instability affect bank spread through

three main channels( first, the volatility of interest rates levied on loans on the inter!bank

market is a direct reflection of the country6s macroeconomic stability, so that the less stable

a country6s economy – e.g. the greater the variation in the inflation rate and e$change rate –

the greater will be the resulting volatility of the basic interest rate and, conse3uently, the

greater the bank spread; second, the degree of banks6 risk, as banks6 risk must to some

e$tent reflect the instability of the macroeconomic environment where they operate( the less

stable the environment, the greater banks6 risk must be; finally, the covariance between

interest rate risk and credit risk, as a highly volatile basic interest rate will be e$pressed to

some e$tent in a highly variable level of real output.

)mpirical works show evidences that high banking spread in Latin America is

determined for both microeconomic /high operating costs, poor loan 3uality, high

capitalization, and reserve re3uirements2 and macroeconomic factors /interest rate

volatility, +78 growth and inflation2. Gowever, the picture is somehow heterogeneous

among the Latin American economies. )mpirical evidence suggests that microeconomic

factors have been the main determinant of spreads in Bolivia; micro and macroeconomic

factors impacted on spreads in 5hile and 5olombia; macroeconomic factors were more

important in determining Brazilian spreads; neither micro or macroeconomic factors

ade3uately e$plained the evolution of spreads in Argentina and 8eru.

n the Brazilian case, macroeconomic aspects are prominent as ma?or

determinants of bank spread, and this can e$plain at least partially why spread is so high

compared to other countries, including Latin American ones. 8articularly noteworthy are

the risk variables /risk premium, interest rate volatility2, output growth and the level of

short!term interest rate. 'hese finding are not surprisingly if one considers that Brazilian

economy has had a Ystop!go6 tendency as general feature in the last 1 years, and more

-


25/28

recently /that is, since the mid!"##&s2 has suffered a lot of speculative attacks on its

domestic currency.

n particular, the importance of the level and volatility of interest rate in Brazil

as macroeconomic determinants of the spread confirms the hypothesis of banks preference

for li3uidity /8aula and Alves Tr &&*2, according to which – in view of the e$istence of an

interest risk!free application combining li3uidity and profitability /inde$ed public bonds,

the L4'6s2 – banks in Brazil came to build a high li3uidity premium into their loan!making

operations. Added to this, 0elic interest rate rises may lead to greater variation in real

output levels and business profitability, thus raising credit risk, which can result in higher

loan rates and increased spreads. Lastly, for the purposes of proposing policies to reduce

bank spread in Brazil, the results of this study seem to indicate that a reduction in the 0elic

interest rate and in its volatility, combined with the substitution of inde$ed bonds by non!

inde$ed ones, are necessary conditions for obtaining any pronounced and lasting reduction

in spread in Brazil.

-/


26/28

References

Afanasieff, '.0., Lhacer, 8.9., =akane, 9.. && 'he determinants of bank interest spread

in Brazil. Mone, airs ZV( ":*!&8#:""&.

Bara?as, A., I. 0teiner and =. 0alazar. &&&. 'he impact of liberalization and foreign

investment in 5olombia6s financial sector. Journal o De/elop%ent Econo%ics D*(

"1

Belaisch, A.. &&* @7o Brazilian banks compete^C "MF Wor&in1 Paper >8&*""*.

Brock, 8.L., 0uarez, L.I. &&". Jnderstanding the behavior of bank spreads in Latin

America. Journal o De/elop%ent Econo%ics D*( ""*!"*%.

-0


27/28

5arvalho, 4.5., 8aula, L.4. and >illiams, T. &. Banking in Latin America3 n '4e Oxord

4and+oo& o +an&in15 ed. 5 A. Berger, 8. 9olyneu$ and T. >ilson. $ford( $ford

Jniversity 8ress /forthcoming2.

5entral Bank of Brazil, http(www.bcb.gov.br . Access in =ovember &&ashington( A7B.

)7. &&%. 0pread no Brasil e no mundo. Carta "ED" n. "&&, /available in

http(www.iedi.org.br , access in ": Tune &&


28/28

9cshane, I.>., 0harpe, .+. "#:1. A time seriescross section analysis of the determinants

of Australian trading bank loandeposit interest margins( "#D!"#:". Journal o

0an&in1 and Finance #( ""1!"*D.

=akane, 9. . &&*. 5oncorr`ncia e spread bancSrio( uma revis[o da evid`ncia para o

Brasil. n( Juros e spread +anc;rio no 0rasil7 a/alia nos do projeto. 5entral

Bank of Brazil, Brasilia.

reiro, T.L., 8aula, L.4., 5osta da 0ilva, +.T., no, 4.G. &&D. 7eterminantes

macroeconmicos do spread bancSrio no Brasil( teoria e evid`ncia recente. 0razilian

Journal o pplied Econo%ics "&( D!D*%.

8aula, L.4., Alves, Tr, A.T. &&*. Banking behaviour and the Brazilian economy after the

Ieal 8lan( a post!Eeynesian approach. 0anca !azionale del La/oro .uaterl, 2e/ie60aunders, A., 0chumacher, L. &&&. 'he determinants of bank interest rate margins( an

international study. Journal o "nternational Mone, and Finance "#( :"*!:*.

0ingh, A., A. Belaisch, 5. 5ollyns, 8. de 9asi, I. Erieger, +. 9eredith and I. Iennhack.

&&1. 0tabilization and reform in Latin America( A macroeconomic perspective on

the e$perience since the early "##&s. "MF Occasional Paper n. *:.

-2

macroeconomic determinants of bank spread in brazil irae revised[1]

Documents