Table 5 Market discipline in the Latin American deposit market

Martinez-Peria and Schmukler (2001)

The first version was published in 1998 as a working paper

Maximum of 155 Argentinean banks, period of 1993–1997

Maximum of 37 Chilean banks, period of 1981–1986 (monthly), and 1991–1996 (quarterly)

Maximum of 34 Mexican banks, period of 1991–1996 (quarterly)

Regressions with panel data, LS with fixed effects, and pooled estimates

1) Implicit interest rate: calculated by dividing the total interest rate expenses by the total interest-bearing deposits

2) Growth of time deposits

1) CAMEL methodology: capital adequacy, asset quality, management, earnings, and liquidity. Using an index

2) Insured and uninsured deposits, and periods with and without deposit insurance, and before, during, and after episodes of stress (crisis)

Depositors in Argentina, Chile, and Mexico punish banks for risky behavior, both by withdrawing their deposits and by requiring higher interest rates. Both large and small depositors discipline banks. Deposit insurance does not necessarily decrease market discipline, because of lack of credibility. The relative importance of market discipline rises after banking crises for all types of deposits

Barajas and Steiner (2000)

25–33 Colombian commercial banks

Panel data from 1985 to 1999 (half-yearly)

Regressions with least squares (LS) with fixed and random effects. And instrumental variables (2SLS)

1) Growth rate of deposits

2) Interest rate on deposits

3) Reaction of banks: bank fundamentals as dependent variables, explained by growth rate of deposits

1) Bank fundamentals: non-performing loans to total loans, non-performing loans to total assets, loans loss provisions to total assets, capital asset ratio, return on equity and total reserves to assets

2) Macro control variables: GDP growth and interest rate on government securities

3) Return variable and bank specific variables: interest paid on deposits and number of branch offices. Dummy variables for state and foreign ownership. And bank’s size

Positive evidence of market discipline, despite the deposit insurance. Depositors prefer banks showing strong bank fundamentals and banks tend to improve bank fundamentals after being punished by depositors (a test of market influence)

Calomiris and Powell (2000)

Panel data for 119–205 Argentinean institutions

Period of 1993–1999 (quarterly)

LS with fixed firm and time effects, and random effects

1) Interest rate on deposits, converted into dollar equivalent yields

2) Deposit growth

1) Measures of asset risk (loans/other assets, non-performing loans/loans, and the loan interest rate), a measure of the liquidity of non-loan assets (cash/government bonds), and the (book) capital ratio

Positive evidence of market discipline by price and quantity. Market perceptions of default risk are mean reverting, indicating that market discipline encourages banks to respond to increases in default risk by limiting asset risk or lowering leverage

Baquero-Latorre (2000)

Ecuadorian private banking sector, during 1995–1999

LS for panel data with fixed effects, and simultaneous equations

1) Deposits, by maturity and currency

1) Bank fundamentals: capital ratio, non-performing loans to total loans, management, earnings, and liquidity

2) Interest rate on deposits and bank’s size

3) Macro control variables: international reserves to M2, inflation, and devaluation

4) Systemic variables: total deposits in the banking system and preference for liquidity

Lack of evidence of market discipline. Microeconomic variables that were used in this research are not the relevant indicators that allow depositors to impose control actions over the banks. Macroeconomic factors showed higher explanatory relevance

Mayorga-Martínez and Muñoz-Salas (2002)

21 Costa Rican banks

Panel data, period of 1995–2001 (monthly)

Pooled least squares with fixed effects

1) Amount of deposits in real terms by maturity and currency

1) CAMEL methodology and bank’s size

2) Macro control variables

3) Systemic risk: money issue to total deposits in the banking system

Weak (or incipient) evidence of market discipline. Most of the bank fundamentals do not seem to have influence on the behavior of the deposits

Bundevich and Franken (2003)

Maximum of 24 Chilean banks

Panel data from 1989 to 1997 (quarterly) and 1998 to 2001 (monthly)

Dynamic panel data model with fixed effects, a estimation with maximum likelihood

1) Deposits by size, maturity and institutional sector. As deviations of the total bank system

2) Interest rate on deposits by maturity. As deviations of the total bank system

1) CAMEL methodology

2) Rating of agency (Fitch Chile)

3) Bank’ size and bank concentration (Herfindahl index)

Weak evidence in favor of market discipline hypothesis by quantity mechanism, and stronger evidence in favor of the hypothesis by price mechanism. The rating agency does not contribute to discipline

Levy-Yeyati, Martinez-Peria and Schmukler (2004)

Argentina and Uruguay. Period of 2000–2002 (monthly and daily information)

LS with fixed effects and robust standard errors. Vector autoregressive regression (VAR) model. And DIF GMM estimator of Arellano and Bond (1991) with heteroskedasticity robust standard errors

1) Log of time deposits, and the change in deposits, by currency

2) The interest rate on deposits

1) Bank fundamentals: Non-performing loans to total loans, the ratio of equity to capital, return on assets, bank’s size (log of assets)

2) Systemic risk: measures of country and exchange rate risks

Systemic risk (driven by macroeconomic factors) exerts a significant impact on the behavior of depositors. For Argentina and Uruguay market discipline is indeed quite robust once systemic risk is factored in

Galindo et al. (2005)

A maximum of 375 banks for 13 Latin American countries

LS for panel data with fixed effects, and panel vector auto-regression

1) Interest paid to depositors

2) The growth rate in deposits

3) Capital ratio: to test the bank reaction to depositors’ actions

1) Bank fundamentals: non-performing loans as a percentage of total assets, capital divided by total assets, cash reserves as a percentage of total assets, and percentage return on assets

2) Dummy variables for country and deposit insurance schemes. And a measure of the degree of compliance with the basel core principals

Market discipline is only strong in countries with higher basel core principles of banking supervision (BCP) compliance. Private banks do respond to market discipline but the result is driven by high BCP compliant

countries (a test of market influence)

Goday et al. (2005)

Sample of a maximum of 29 Uruguayan banks

From January 2000 to December 2004

Regression analysis: DIF GMM estimator of Arellano and Bond (1991) for dynamic panel data

1) The logarithm of the stock of total dollar denominated deposits

2) The logarithm of the stock of dollar denominated time deposits

3) The spread between the bank’s marginal-weighted-average interest rate paid on time deposits and the system’s marginal-weighted-average interest rate paid on deposits

4) The logarithm of the marginal-weighted-average maturity of dollar denominated time deposits

1) Bank fundamentals: equity capital as a percentage of total assets, liquid assets as a percentage of liabilities, non-performing loans as a percentage of total loans, return on assets expressed as a percentage, non-financial losses as a percentage of total assets, the exposure to the government as the percentage of loans to the public sector plus holdings of government bonds on total assets, the total business with non-residents as the percentage of loans plus deposits of non-residents on the total loans and deposits

They found strong evidence that supports the hypothesis that depositors discipline riskier banks by withdrawing their deposits and weaker evidence on the hypothesis that depositors require higher interest rates and reduce the maturity of their deposits as disciplining actions. Additionally, we find that banks react to depositors’ actions-depositors’ discipline is effective specifically in the post-crisis period

Ioannidou and de Dreu (2006)

A maximum of 16 Bolivian commercial banks

Unbalanced panel data covering the period of 1998-2003 (monthly)

Least squares (LS), pooled estimates, and fixed effects

1) Implicit interest rate on savings deposits

2) Growth rate of deposits, using savings deposits denominated in US dollars

1) Bank risk: equity to total assets, non-performing loans to total assets, loan loss reserves to total assets, overhead expenses to total assets, returns to total assets, liquid assets to total assets, and size (the log of total assets)

2) Dummy variables for foreign bank, deposit insurance and coverage

3) Control variables: GDP in Bolivia and the US inflation, and dummy variables for political instability

Market discipline is at work; an increase in bank risk leads to higher interest rates on deposits and lower deposit volumes. Most market discipline comes from large depositors. Deposit insurance causes a significant reduction in market discipline, depending on the coverage rate

Romera and Tabak (2010)

Sample of 146 Brazilian banks

Panel data by semester, period of 1994–2004

LS with random and fixed effects, and White correction for heteroskedasticity

1) Price: interest rates charged on deposits. The implicit interest rate represents the volume of interest rate expenses paid by banks in all their deposits, divided by the total number of deposits

2) Quantity: First, logarithmic difference (growth rates) of deposits. Second, the logarithm of the time deposits balance

1) CAMEL methodology

2) Individual bank characteristics (bank’s size, state bank), macroeconomic and systemic conditions

Evidence of market discipline is found, both via quantity (withdrawal of deposits) and prices (increases in interest rates). Quantity mechanism is more powerful during the crisis period, and the price mechanism during the period of stability. The result agreed with the “too big too fail” hypothesis, meaning that Brazilian depositors believe that large banks are safe, or that the government wouldn’t allow their failure

Márquez (2011)

Sample of 13 Colombian banks

Panel data from 2001 to 2008 (quarterly)

Least squares (LS) with fixed effects

Deposits by size, there are six categories, but the key comparison is between small and large deposits. And kind of account

1) Bank fundamentals: Non-performing loans divided by total loans, provisions for loan losses divided by total loans, expected default frequency of Moody’s, Asset quality, return on earnings, and capital ratios

2) Control variables: GDP, merger, bank’s size and funding rate of the bank

Larger depositors react to economic fundamentals such as the bank’s loan quality and solvency, but the smaller depositors do not exert discipline