Journal Issue

Liberia: Staff Report for the 2012 Article IV Consultation and Request for Three-Year Arrangement Under the Extended Credit Facility–Background Notes

International Monetary Fund. African Dept.
Published Date:
December 2012
  • ShareShare
Show Summary Details

Drivers of Net Interest Margins and Bank Profitability in Liberia1

This note provides an analysis of regional drivers of net interest margins (NIMs)—the difference between interest income acquired by banks and interest costs paid on interest bearing liabilities such as deposits over total interest bearing assets. The analysis suggests that NIMs in Liberia are low when considering the high bank operation costs relative to peer countries. This may explain why indicators of profitability for Liberian banks are significantly lower than in the region.2 This also suggests that the Central Bank of Liberia (CBL) should be mindful of the low bank returns when using moral suasion to lower bank lending rates as this may weaken the financial system while pressuring smaller banks out of the credit market. The note presents background on the Liberian banking system followed by an analysis of the determinants of net interest margins in the region and implications for Liberia.


1. Bank profitability has been weak in Liberia with returns on equity near zero or negative during 2003–11 due to:

  • A weak credit environment with limited supporting infrastructure (the lack of a credit reference system or credit bureau, no collateral registry, weak property rights, and limited legal enforcement on debt repayment);

  • Capacity constraints at banks including poor internal controls and risk management;

  • High costs of operation;

  • Absence of a domestic fixed-income securities market to provide a secure investment instrument;

  • Low returns on investment abroad or in foreign assets due to loose monetary policy in advanced countries (Banks are only allowed to invest in foreign assets with very short maturity and issued by top rated commercial banks and national authorities).

2. Despite the anemic bank profitability bank lending and deposit rates have been relatively rigid. While the increase in the number of banks may have played a role in containing lending rates through competition, Liberia’s banking system is dominated by one bank holding about 50 percent of commercial bank demand deposits and lends around 40 percent of private sector credit.

Lending and Deposit Rates in Liberia

Source: Central Bank of Liberia

3. Despite a rapid increase in bank lending, interest income continues to play only a secondary role in bank revenue. Non-interest income as share of revenue has remained above 50 percent declining by less than 10 percent over 2008–2011, while private credit growth averaged a brisk 34 percent per year over the same period.

Bank Income Source and Private Credit Growth

Source: Central Bank of Liberia and IMF staff estimates

Regional Drivers of NIMS

4. The literature highlights high operational costs relative to assets and weak property rights as two dominant factors behind relatively high lending margins in sub-Saharan Africa. We examine the drivers of NIMs in a dynamic panel of 86 banks across six ECOWAS countries.3 Property rights are generally and similarly weak within the region according to the World Bank CPIA ratings, which allows us to focus on the impact of operational costs on NIMs within the sample. We estimate the following two models using Generalized Method of Moments (GMM) and Ordinary Least Squares (OLS).

We estimate equation (1) using system GMM–Blundell and Bond (1998) in Table (1) and equation (2) using system GMM–Arellano and Bond (1991) in Table (2), where NIEPA is non-interest expense as percent of average interest bearing assets and X and M are matrices of control variables. We also include OLS estimates for each model for robustness.

Table 1:Estimation of Determinants of Net Interest Margins in ECOWAS Banks
Independent Variable: NIMGMM(Blundell/Bond 98) CoefficientsOLS-Coefficients
L. Return on Equity(ROE)−.007**.001
L. Return on Assets (ROA).076*.045
Country Fixed EffectsYesYes
Year Fixed EffectsYesYes
Number of Banks8686
Number of Observations403403
Where ***, **, and * indicate significance at 99, 95 and 90 percent confidence respectively. D indicates time difference and LD indicates lagged time difference. Standard error in parenthesis.
Where ***, **, and * indicate significance at 99, 95 and 90 percent confidence respectively. D indicates time difference and LD indicates lagged time difference. Standard error in parenthesis.
Table 2Estimation of Determinants of Net Interest Margins in ECOWAS Banks Using Log Differences
Independent Variable: D.NIMGMM(Arellano/Bond 91) CoefficientsOLS-Coefficients
LD. Return on Equity(ROE).010*.004
LD. Return on Assets (ROA)−.278***−.170***
Country Fixed EffectsYesYes
Year Fixed EffectsYesYes
Number of Banks5976
Number of Observations224311
Where ***, **, and * indicate significance at 99, 95 and 90 percent confidence respectively. D indicates time difference and LD indicates lagged time difference. Standard error in parenthesis.
Where ***, **, and * indicate significance at 99, 95 and 90 percent confidence respectively. D indicates time difference and LD indicates lagged time difference. Standard error in parenthesis.

5. Both GMM and OLS estimates show that high NIEPA is associated with higher NIMs across banks while lagged increases in NIEPA are associated with increasing NIMs within banks (Tables 1 and 2). Moreover, decreasing lagged returns on assets (ROA) are strongly associated with increasing NIMs (Table 2), which is likely due to banks raising lending rates and cutting deposit rates when returns weaken.

6. These results are difficult to reconcile with developments in net interest margins in Liberia. Operational costs are high in Liberia with an average NIEPA for Liberian banks in the sample at 21 percent, which is more than one standard deviation above the sample mean of 11 percent. Costs are high in Liberia due to poor infrastructure—with expensive electricity and limited roads—and lack of human capital. Yet, the cumulative NIM for Liberian banks at 7 percent in 2011 is lower than the 8 percent average NIM for all banks in the sample. Also, returns on assets and equity have been persistently very low in Liberia relative to the region. In 2011, despite a sharp increase in nonperforming loans, lending and deposit rates have remained stable while net interest margins declined.

Net Interest Margins and Overhead costs (2008-2010)


Sources: Bankscope and Central Bank of Liberia.

Implications for Liberia

7. The analysis shows that overhead costs relative to assets drive net interest margins across banks in the region, which implies that Liberian banks should be expected to earn relatively high net interest margins. The Liberian financial system has expanded rapidly since 2006 with banks nearly doubling in number, but most banks remain very small with high overhead cost and relatively few assets. Moreover, lending is highly concentrated, with two banks out of a total of nine accounting for nearly 75 percent of demand deposits and private sector credit, leaving a very small market for most banks. Additionally, poor lending controls and risk management are raising bank operational risk. While some of these factors are mitigated by most of the banks being subsidiaries of larger foreign banks that may be able to backstop liquidity and capital needs if necessary, low profits make the banking environment less attractive to potential entrants and hampers financial development.

8. Consequently the analysis raises concerns on the repercussions of the CBL’s policy to use moral suasion to lower commercial bank lending rates. Small Liberian banks with higher operational costs relative to size are particularly vulnerable to this policy. Several of the smaller banks are operating in niche markets such as agriculture and microfinance where credit risks and operational costs are higher. The exercise of moral suasion on the small banks may risk pressuring them out of the credit market and discourage them from expanding operations, which would hamper competition. Additionally, this policy may exacerbate banking sector fragilities stemming from the anemic returns.


    ArellanoM. and S.Bond. (April1991). Some tests of specification for panel data: Monte Carlo evidence and an application to employment equations. The Review of Economic Studies 58. pp. 277297.

    BlundellR. and S.Bond. (1998). “Initial conditions and moment restrictions in dynamic panel data models.Journal of Econometrics87(1) 115143.

    Demirgüç-KuntAsliLucLaeven and RossLevine.2004. “Regulations, Market Structure, Institutions, and the Cost of Financial Intermediation.Journal of Money Credit and Banking36 (3): 593622.

    HonohanPatrick and ThorstenBeck.2007. “Making Finance Work for AfricaThe World Bank.

Prepared by Kareem Ismail (AFR).

EBS/11/164 – Liberia- Staff Report for the Seventh Review under the Extended Credit Facility.

Bankscope data from Côte d’Ivoire, Guinea Bissau, Liberia, Nigeria, Sierra Leone and Togo.

Other Resources Citing This Publication