The Effect of Political Instability on Bank Profitability: Evidence from Ethiopia

The study aims to examine the effect of political instability on Bank profitability. To achieve this object, the researcher used the World Bank database and data from annual audit financial statements of 15 selected commercial Banks for the period 2012-2016. The researcher applied the generalized method of moments to analyze the effect of political instability on Bank profitability. Both descriptive analysis and inferential statistics are made in this regard. The result of the generalized method of moments estimation shows that political instability positively affects Bank profitability. Bank specific variables such as credit risk ratio, Non-interest income to total asset ratio, and Cost to income ratio are essential factors for Bank profitability. Growth of gross domestic product and inflation are significant determinants of Bank profitability. Bank size, liquidity and credit risk ratio are insignificant. The study is helpful to higher government officials, policymakers, Banks shareholders, Banks top management, etc. Moreover, the study contributed to the existing literature by linking political instability and Bank profitability. profitability. Few researches have been conducted, to the best of the researcher’s knowledge, on the relationship between political instability and Bank profitability in Ethiopia. The specific objectives of the study are; (1) To determine the relationship between political instability and Bank profitability; (2) To analyze the relationship between Macro-economic variables and Bank profitability; and (3) To investigate the association between Bank- specific variables and Bank profitability. The study scrutinizes empirically the impact of political instability on the profitability of Ethiopian Banks for the period from 2012-2016 employing evidences from Ethiopian commercial Banks and international organizations such as WB. The study will contribute to the existing literature by linking political instability and Bank profitability. It will also help policy makers, political elites and various stakeholders to better understand the relationship between political instability and Bank profitability. The rest of the paper is organized as follows: Section 2 review of literature, section 3 research methodology, section 4 data analysis and results, section 5 conclusions, Limitations and Implications for further research. The study analyzed the effect of political instability on Bank profitability using empirical evidences from Ethiopian commercial Banks and others. The stud has contributed to the existing literature by linking political instability with Bank profitability. The study found political instability influence Bank profitability positively. Actually there result is against the theory and the researcher’s expectation. Bank specific variables such as CAR, NIITAR, and COIR are important factors for Bank profitability. GGDP and INFL are significant determinants of Bank profitability. In this study political instability is measured by political instability index which is the average value of other indexes which may provide unexpected result. The average political instability index is used in the study. Researchers can analyze the effect political instability on Bank’s profitability by factoring individual political instability variables for recent Bank data.

autocratic in practice, which violets democratic and basic human rights. EPRDF is criticized for creating regional inequalities, group discriminations, economic inequalities, lack of good governance, injustice corruption, etc. Opposing the undemocratic practices of the EPRDF public protest began in 2016 in a more organized manner and continues till now. The public protest began in Gonder (historical city) and expanded to almost all part of the country. It causes overall political, societal and economic unrest such as mass displacement, mass killings, cross-group discriminations, cabinet changes, violence, lockdown of federal roads, lockdown of many private and public enterprises such insurance companies, transportation, Banks, etc. Given these facts the effect of that political instability was not studied. Hence, the researcher is interested to scrutinize its effect more specifically on the profitability of the Banking sector.
The world governance index survey of World Bank (2016) shows profitability of Ethiopian Banks continuously declined for the period from 2012-2016. According to the report return on asset (ROA) was 5.39% and declined to 3.75% in 2016. Similarly return on equity (ROE) was 75.26% in 2012 and decreased to 54.08% in 2016. In the same period political stability index was negative that shows Ethiopia's week political stability. So, the basic question is does political instability reduce Bank profitability?
The key feature of political instability is uncertainty of environment where investment take place that reduces returns and incentives to attract saving and investment (Gyimah-Brempong & Traynor, 1999). From the investment point of view investors require high interest rate to compensate their risky and unsafe investment in uncertain environment. This interest rate reduces interest margin and Bank profitability. Otherwise business people and investors prefer to keep their money at home or at a more preferable place, which may cause Banks illiquid. Political instability creates frustration among customers. In such scenarios, Bank customers deposit less and withdraw more, which reduces Bank's interest margin.
Tremendous researches have been conducted on the determinants of profitability of Banks (Dietrich & Wanzenried, 2011;Obamuyi, 2013;Petria, Capraru & Ihnatov, 2013). Dietrich and Wanzenried (2011) studied determinants of profitability of commercial Banks in Switzerland that include 372 Banks over the period from 1999 to 2009. In this study authors consider only the macro-economic variables such as GDP and Inflation and Bank specific variables like size, ownership structure, and age. Petria et al., (2013) analysed Bank profitability and its determinants using evidences from the EU Banking system for the period from 2004-2011. The study split Bank profitability determinants in to two parts as internal factors and external factors. Internal factors refer to Bank specific factors that may include Credit risk, liquidity risk, management efficiency, the diversification of business. External factors represent macro-economic variables such as market concentration/competition and the economic growth. Obamuyi (2013) investigated determinants of Bank profitability in developing economies using evidences from 20 Nigerian Banks for the period of time from 2006 to 2012. Authors took in to account GDP growth rate and interest rate as macro-economic variables and capital, Bank size, and expense management as Bank specific variables.
Despite many efforts are exerted in studying Bank profitability and its determinants, researchers gave less attention to examine the impact of political instability on Banks profitability. Few researches have been conducted, to the best of the researcher's knowledge, on the relationship between political instability and Bank profitability in Ethiopia. The specific objectives of the study are; (1) To determine the relationship between political instability and Bank profitability; (2) To analyze the relationship between Macro-economic variables and Bank profitability; and (3) To investigate the association between Bank-specific variables and Bank profitability. The study scrutinizes empirically the impact of political instability on the profitability of Ethiopian Banks for the period from 2012-2016 employing evidences from Ethiopian commercial Banks and international organizations such as WB. The study will contribute to the existing literature by linking political instability and Bank profitability. It will also help policy makers, political elites and various stakeholders to better understand the relationship between political instability and Bank profitability. The rest of the paper is organized as follows: Section 2 review of literature, section 3 research methodology, section 4 data analysis and results, section 5 conclusions, Limitations and Implications for further research.

LITERATURE REVIEW
In Ethiopia, the Banking industry has around 100 year's old history. According to Fasil and Merhatbeb (2012) and Maury (2014) (Fasil & Merhatbeb, 2012). The intention was to isolate the central Banking function from Commercial Banking function.
Then after a number private commercial Banks joined the banking industry in different times. Currently the number of commercial Banks including private Banks reached 17 and at least four commercial Banks are under establishment including Amhara Bank S.C.
The government of Ethiopia made many changes since the resignation of the former Prime Minister Haylemariam Desalegn. Following his resignation a new Prime Minister came to power and has been making significant reformations. Such as the change in cabinets, the replacement of regional state presidents, the reshuffling of state owned Banks' top management, and so on. In this time mass displacement, mass killings, killings of higher government officials, kidnapping of university students, public protests, and state-federal government disagreements became common and made Ethiopia socially, economically and politically instable. This political instability disordered the lives of thousands of individuals, women, children, etc. It left many individuals homeless and jobless; many also lost their life. Although this political unrest affects almost every aspect of the individuals and companies, I would like to study its effect on the Banking sector. The researcher limits his study only to commercial Banks (1 state owned and 14 private) because Banks have more organized data than other sectors. Political instability is defined by various individuals. Fosu (1992) defined political instability as: instability of government's, regimes, and communities within a nation is likely to result from forcefully overthrowing or subjecting existing authorities to a relatively high probability of involuntary removal. Political instability is a condition in political systems in which the institutionalized patterns of authority break down, and the expected compliance to political

Dependent Variables
Bank profitability (ROA) -The study aims to investigate the effect of political instability on Bank profitability by using evidences from panel of Ethiopian Commercial Banks and some other international organizations. In this study Bank profitability is the dependent variable. There are a number of alternative measurements for Bank profit but return on asset is (ROA) preferred for this study due to its wider application in measuring profitability and financial performance.
authorities is replaced by political violence (Morrison & Stevenson, 1971). According to Arriola (2009), political instability takes different forms such as, communal violence, rural insurgency, urban riots, coups d'état, and civil wars. Morrison and Stevenson (1971) defined three conceptual dimensions of political instability: elite instability, involving conflicting elites in successful or attempted removals of the incumbent political authorities by violence or the threat of violence; communal instability, involving communal groups in conflict with political authorities or their supporters in violent attempts to restructure the political status of the communal groups; and mass instability, involving mass movements in conflict with the national regime in violent attempts to reorganize the distribution of power and the goals of policy within the nation. TI anti-corruption help desk report (2013) revealed that political instability is driven by political and economic factors (i.e., power and resource). Elite disagreements are consequential for instability in Africa because they can spiral into other mass forms of violence. Similar to other companies political instability also affects the profitability of the Banking sector. First political instability causes political risk and then the political risk affects both microeconomic and macroeconomic aspect of the country (Yalcinkaya, Celik & Huseyni, 2017). According to the authors, political instability influences the finance sector in general and the and more specifically the Banking sector. In relation to Banks, political instability has dual effect: on the first hand, it causes political risk and increase interest rate and on the other hand it increases the Cost of credit and reduces the demand for credit.
Empirical evidences that documented political instability-Bank profitability are limited. However, there is a general consensus that political instability results political risk and uncertainty which diminishes saving, investment, performance, economic development and growth (Chawdhury, 2015;Gyimah-Brempong & Traynor, 1999;Mbaku, 2016 (2000). They divided political instability as informal and formal (political motivated). Their finding shows that informal political instability has direct and significant effect of financial development. However, formal instability has not effect on financial development.
Determinants of Bank Profitability-Factors that determine Bank profitability can be broadly classified as Bank specific variables and macro-economic variables. In this section the dependent variable (Bank profitability), macro-economic variables, and Bank specific variables are presented, respectively.
Literatures in the area of Finance evidenced, that there are numerous factors that affect the profitability of Banks. The study considers, however, the following factors due to data limitations. Political instability (PI) is one of macro-economic variables that determine Bank profitability (Gyimah-Brempong & Traynor, 1999;Yahya et al., 2017). According to the World Bank report ( GDP growth rate (GDP GR) -affects Bank profitability. GDP growth rate refers to the annual growth rate and the data is accessed from World Bank data base. The result of the studies of (Gul, Irshad & Zaman, 2011;Kosmidou, Tanna & Pasiouras, 2008;Moges, 2017;Rahman, Hamid & Khan, 2015;Sanlisoy et al., 2017;Yahya et al., 2017) show that GDP growth rate positively and significantly affect Bank profitability. Alper and Anbar (2011) found no statistically significant relationship between GDP growth rate and Bank profitability.
Inflation rate (IR) influences Bank profitability. Research findings about the effect of inflation rate on Bank profitability are inconclusive. Davydenko (2010); Kosmidou et al. (2008); Vong and Chan (2006) and found that inflation rate has positive and significant effect on Bank profitability. However,  argued that inflation rate negatively and significantly affects Bank profitability. Other studies (Alper & Anbar, 2011;Moges, 2017) found no significant correlation with Bank profitability.  Kosmidou et al. (2008) concluded that there negative and significant correlation between Bank size and Bank profitability. Capital adequacy ratio (CAR) -shows the Bank's strength is measured by the ratio of total capital to total asset. Previous studies (Hassan & Bashir, 2003;Kosmidou et al., 2008;Sanlisoy et al., 2017) show positive and significant correlation among capital adequacy ratio and Bank profitability.
Credit risk ratio (CRR) -is one of the Bank specific determinants of Bank profitability. Credit risk ratio is measured by the ratio of total non-performing loan to total asset. According to  credit risk ratio has positive and significant effect on Bank profitability. The influence of credit risk ratio on Bank profitability is statistically significant and negative (Imad & Thair, 2011;Osuagwu, 2014).

Independent variables
The aim of this study is to examine the effect of political instability on Bank profitability using empirical evidence from Ethiopian Commercial Banks. To achieve this objective audited financial data is examined for each Banks to measure Bank specific factors. Data related to macro-economic variables is obtained from World Bank data base. The study included fifteen Ethiopian Commercial Banks (1 state owned and 14 private). The study covers the time period of five years from 2012-2016 and only fifteen Banks based on the availability of complete data. The researcher prefer this period for two reasons: one is complete data is available only for this time series and second World Bank data base (2012)(2013)(2014)(2015)(2016) shows ROA and ROE of Commercial banks are almost steadily declined. Full data for the variables considered are available only for this period and selected Banks. Quantitative research method is employed based on the nature of variables. The researcher employed panel data that is a cross-section of comprises fifteen Commercial Banks and a time series of five years. Use of panel data has various advantages over using only cross sectional data or time series data alone. Panel data increase degrees of freedom, reduce multicolinearity, omitted variables biases, and increase efficiency of econometric model (Hsiao, 2003).

Model specification
Liquidity ratio (LQR) -computed as the ratio of total loan to total asset, represents Bank's available liquid assets that can be used to settle its debt. It also indicates the risk of not having enough cash reserves to meet the demands of withdrawals from depositors . Kosmidou et al. (2008) found negative and significant correlation between Cost to income and Bank profitability.
Non-interest income to total asset ratio (NIITAR) is computed non-interest income divided by Bank's total asset. A research result by  indicates positive and significant relationship between NIITAR and Bank profitability. The specific econometric model is presented as follows: The econometric model that consists of the dependent variable Bank Profitability (ROA) and specific independent Bank profitability determinants is presented as follows: Where; CAR is the ratio of total capital to total asset, CRR is ratio of total non-performing loan to total asset; LQR is the ratio of total loan to total asset; NIITAR is non-interest income divided by Bank's total asset; IR is the annual inflation rate from world Bank data base; GDP GR-the annual growth rate from World Bank data base; and Political Instability (PI) -World Governance Index survey from World Bank data base. β0 is constant; ε-is stochastic error term; i-represents specific Banks, and t-represents the time.

Descriptive Analysis
From table 1 above, one can see that the maximum value of political instability index is -1.34 and the minimum value is -1.62, which results a mean political instability index value of -1.486. This index is based on world governance index (WGI). According to WGI political instability index ranges from -2.5 to +2.5. As it can be seen from table 1 that the maximum, minimum and average Bank profitability are 0.067, 0.000, and 0.28 respectively, which shows that Banks earn on average ETB 0.282 of profit from each asset owned. Table 2 represents the correlation matrix, which shows the correlation among each variable.
The correlation matrix has a value that range between +1 (perfect positive correlation) and -1 The    According to Roodman (2006) the Sargan /Hansen test for joint validity of the instruments is standard after GMM estimation. These tests tell us there is no endogenity problem in the model. Similarly in the GMM estimation result one can see that the p-value for first and second order autocorrelation is not significant which shows absence of serial correlation.
As it can be observed from table 3, political instability and Bank profitability are positively and significantly correlate each other. The result is similar with previous findings such as Campos et al., (2012); Yahya et al. (2017). The answer for the question "Does political instability reduce Bank profitability?" is No. Rather, it has positive contribution for Bank profitability, according to the finding of this study under question. CAR and NITTAR have positive and significant (5%) relationship with Bank profitability. Banks can increase their profitability by increasing their capital adequacy ratio (CAR). (Sanlisoy et al., 2017;Kosmidou et al., 2008;Bashir, 2003) found the same. CAR shows the capital strength of Banks. Therefore, the more Banks get strong the more they earn profit. Table 3 above revealed the increase in Non-interest income to total asset ratio (NIITAR) results in an increase in Bank profitability.  found similar relationship between NIITAR and Bank profitability. According to the result Banks can increase profitability if they The study analyzed the effect of political instability on Bank profitability using empirical evidences from Ethiopian commercial Banks and others. The stud has contributed to the existing literature by linking political instability with Bank profitability. The study found political instability influence Bank profitability positively. Actually there result is against the theory and the researcher's expectation. Bank specific variables such as CAR, NIITAR, and COIR are important factors for Bank profitability. GGDP and INFL are significant determinants of Bank profitability. In this study political instability is measured by political instability index which is the average value of other indexes which may provide unexpected result. The average political instability index is used in the study. Researchers can analyze the effect political instability on Bank's profitability by factoring individual political instability variables for recent Bank data.