Bank profitability: Insights from the rural banking industry in Ghana

Figures & data

Table 1. Tiers of Ghana’s financial system

Tier	Definition	Institutions	Services	Clients	Outreach
Formal	Licensed by central bank	Commercial banks, development banks	Deposits, loans, foreign exchange, cash transfer, insurance	Large businesses, government	Urban
		Rural and community banks	Deposits, loans, money transfer, payments, social investments	SMEs, large enterprises	Rural
Semi-formal	Provisionally licensed as of January 2013	Credit unions	Deposits, loans for members only	Low income, self-employed	Rural
		Savings and loans companies (microfinance institutions), financial NGOs	Deposits, loans,	Microenterprises, entrepreneurial poor
Informal	Not legally registered at national level	Susu institutions, informal moneylenders	Deposits, loans	Self-employed, poor	Rural

Source: Mann et al. (2010), Steel (2006) and Bank of Ghana (2013).

Table 2. Variables, definitions, notations and expected signs

Variable	Definition	Notation	Expected sign
Dependent variable
Return on assets	Profit before interest and tax divided by total assets (%)	ROA
Return on equity	Profit before interest and tax divided by total assets (%)	ROE
Independent variables
Bank size	Natural logarithm of total assets	BSIZE	?
Liquidity risk	Cash and due from balances held at other depository institutions to total assets	LRISK	–
Credit risk	Total loans divided by total assets	CRISK	+
Funding risk	Z-score = Total deposits to asset ratio + Equity to assets ratio divided by the standard deviation of total deposits to assets ratio	FUNDRISK	+
Bank stability	Profit before interest and tax to assets ratio plus equity to assets ratio divided by the standard deviation of profit before interest and tax to assets ratio	BSTAB	+
Diversification	Total investment in financial securities (short and long-term) divided by total assets	DIV	+
Capitalization	Equity capital divided by total assets	CAP	+

Source: Author’s construction (2015).

Table 3. Descriptive statistics

	ROA (%)	ROE (%)	FUNDRISK	BSIZE (Ghana cedis)	LRISK	CRISK	DIV	BSTAB	CAP
Mean	2.26	85.43	1.64	9,131,214.	16.38	45.12	0.29	2.29	0.04
Median	2.06	62.42	1.47	6,179,371.	14.23	41.14	0.28	2.09	0.03
Maximum	69.41	4,721.77	6.44	1.54E+08	2,664.49	4,973.99	0.86	69.41	0.56
Minimum	−26.97	−663.84	0.09	285,670.6	−1.66	0.000	0.00	−26.97	0.00
Std. Dev.	2.85	141.66	0.73	93,32450.	56.87	142.96	0.14	2.84	0.04
Observations	2,200	2,200	2,200	2,200	2,200	2,200	2,200	2,200	2,200

Notes: ROA is the return on assets, ROE is the return on equity, FUNDRISK is the funding risk, BSIZE is the bank size, LRISK is the liquidity risk, CRISK is the credit risk, DIV is the diversification in the business model, BSTAB is the bank stability and CAP is the bank capital.

Source: Author’s construction (2015).

Table 4. Pearson correlation matrix

	BSIZE	LRISK	CRISK	FUNDRISK	BSTAB	DIV	CAP
BSIZE	1
LRISK	−0.2025	1
CRISK	0.1535	−0.1499	1
FUNDRISK	−0.1279	0.1542	0.0922	1
BSTAB	0.1712	−0.0500	0.0793	0.1127	1
DIV	0.0244	0.0657	−0.4271	0.1805	0.1212	1
CAP	−0.5306	0.0840	−0.0115	0.6035	0.0442	0.0311	1

Notes: BSIZE is the bank size, LRISK is the liquidity risk, CRISK is the credit risk, FUNDRISK is the funding risk, DIV is the diversification in the business model, BSTAB is the bank stability and CAP is the bank capital.

Source: Author’s construction (2015).

Table 5. Results of panel least squares regression. Dependent variable: ROA

Variable	Full model (2009Q1–2013Q4)			2009Q1–2011Q2			2011Q3–2013Q4
	Coefficient	t-Value	p-Value	Coefficient	t-Value	p-Value	Coefficient	t-Value	p-Value
BSIZE (−1)	0.3284	4.4448	0.000^***	0.5404	3.698411	0.0002^***	0.2760	2.2955	0.0219^**
LRISK (−1)	0.0485	1.8032	0.0715	0.0056	0.1202	0.9043	0.0328	0.8958	0.3706
CRISK (−1)	0.0345	0.4322	0.6657	−0.0894	−1.1117	0.2666	0.0462	0.5996	0.5489
FUNDRISK(−1)	−0.1514	−0.9714	0.3315	−0.1624	−0.5446	0.5862	−0.0846	−0.3368	0.7363
BSTAB (−1)	0.4368	22.298	0.0000^***	0.2482	7.5514	0.5862	0.3881	14.168	0.0000^***
DIV (−1)	0.0785	2.3438	0.0192^***	−0.0141	−0.1813	0.8562	0.0626	1.4522	0.1468
CAP (−1)	0.0345	0.4322	0.6657	−0.0627	−0.4082	0.6832	0.0451	0.3535	0.7238
CONSTANT	−4.4238	−3.5783	0.0004^***	−7.6239	−3.2010	0.0004^***	−3.8492	−1.8766	0.0609^*
	R^2 = 0.71; N=1,901			R^2 = 0.73; N = 872			R^2 = 0.76; N = 1,029
	Adj.R^2 = 0.68			Adj.R^2 = 0.69			Adj.R^2 = 0.72
	Durbin–Watson stat = 2			Durbin–Watson stat = 2			Durbin–Watson stat = 2
	F-statistic = 31.78^***			F-statistic = 16.74^***			F-statistic = 22.29^***
	Wald test: χ^2(8) = 5,909.78^***			Walt test: χ^2(7) = 980.72^***			Wald test: χ^2(8) = 4,652.74^***
	Hausman test: χ^2 = 315.57^***
	Likelihood ratio (χ^2) = 1,899.16^***

Notes: BSIZE (−1) is the lagged bank size, LRISK is the lagged liquidity risk, CRISK (−1) is the lagged credit risk, FUNDRISK (−1) is the lagged funding risk, BSTAB (−1) is the lagged bank stability, DIV (−1) is the lagged diversification and CAP (−1) is the lagged capital.

^** Significance level at 5%.

^*** Significance level at 1%.

Source: Author’s construction (2015).

Table 6. Results of panel least squares regression. Dependent variable: ROE

Variable	Full model (2009Q1–2013Q4)			2009Q1–2011Q2			2011Q3–2013Q4
	Coefficient	t-Value	p-Value	Coefficient	t-Value	p-Value	Coefficient	t-Value	p-Value
BSIZE (−1)	0.0284	0.3420	0.7324	0.3636	2.2265	0.0263^**	−0.1801	−1.4383	0.1507
LRISK (−1)	0.0616	2.0638	0.0392^**	0.0642	1.2227	0.2218	−0.0003	−0.0076	0.9939
CRISK (−1)	0.0153	0.3125	0.7547	−0.1310	−1.5789	0.1148	0.01257	0.1746	0.8614
FUNDRISK (−1)	−1.4166	−18.265	0.0000^***	−1.1162	−7.5159	0.0000^***	−1.0299	−7.1791	0.0000^***
BSTAB (−1)	0.4326	19.675	0.0000^***	0.2540	6.7344	0.0000^***	0.4006	13.792	0.0000^***
DIV (−1)	0.1097	3.0248	0.0025^***	0.0527	0.6212	0.5347	0.0613	1.4584	0.1451
CONSTANT	3.9873	2.9368	0.0034^***	−0.8123	−0.3221	0.7475	7.3652	3.5458	0.0004^***
	R^2 = 0.75; N = 1,901			R^2 = 0.76; N = 872			R^2 = 0.81; N = 1,029
	Adj.R^2 = 0.73			Adj.R^2 = 0.72			Adj.R^2 = 0.78
	Durbin–Watson stat = 1.9			Durbin–Watson stat = 2			Durbin–Watson stat = 2
	F-statistic = 40.22^***			F-statistic = 19.56^***			F-statistic = 31.12^***
	Wald test: χ^2(7) = 136,330^***			Wald test: χ^2 (7) = 51,311.83^***			Wald test: χ^2(7) = 108,615.3
	Hausman test: χ^2 = 144.2^***
	Likelihood ratio: χ^2 = 1,913^***

Notes: BSIZE (−1) is the lagged bank size, LRISK is the lagged liquidity risk, CRISK (−1) lagged credit risk, FUNDRISK (−1) is the lagged funding risk, BSTAB (−1) is the lagged bank stability, and DIV (−1) is the lagged diversification. Capitalization has been dropped from the above model in line with the position of the literature that capitalization should be dropped from the model when ROE is used as proxy for profitability (Athanasoglou et al., 2008).

^** Significance level at 5%.

^*** Significance level at 1%.

Source: Author’s construction (2015).

Mann, C., Tinsey, J., Tedjo, G., & Nwadei, T. (2010). Ghana’s rural finance system and climate regime. World Wildlife Fund. Retrieved from http://www.bu.edu/gdp/files/2011/01/Capstone-2010-A-Ghanas-rural-finance-system-and-climate-regime.pdf

 Google Scholar

Steel, W. (2006). Extending financial systems to the poor: What strategies for Ghana? Paper presented at the 7th ISSER-Merchant Bank Annual Economic Lectures, University of Ghana, Legon.

 Google Scholar

Bank of Ghana. (2013). Provisional microfinance licensees. Retrieved from http://www.bog.gov.gh/index.php?option=com_content&view=article&id=854&Itemid=123

 Google Scholar

Athanasoglou, P., Brissimis, S., & Delis, M. (2008). Bank-specific, industry-specific and macroeconomic determinants of bank profitability. Journal of International Financial Markets, Institutions and Money, 18, 121–136.https://doi.org/10.1016/j.intfin.2006.07.001

 Google Scholar