WPS6593 Policy Research Working Paper 6593 Improving Access to Banking Evidence from Kenya Franklin Allen Elena Carletti Robert Cull Jun “QJ� Qian Lemma Senbet Patricio Valenzuela The World Bank Development Research Group Finance and Private Sector Development Team September 2013 Policy Research Working Paper 6593 Abstract Using household surveys and bank penetration data at job, and less education and those who do not own their the district-level in 2006 and 2009, this paper examines own home. The findings are robust to using the district- the impact of Equity Bank—a leading private commercial level proportion of people speaking a minority language bank focusing on microfinance—on access to banking in as an instrument for Equity Bank presence. It appears Kenya. Unlike other commercial banks in Kenya, Equity that Equity Bank’s business model—providing financial Bank pursues distinct branching strategies that target services to population segments typically ignored by underserved areas and less-privileged households. Equity traditional commercial banks and generating sustainable Bank presence has a positive and significant impact profits in the process—can be a solution to the financial on households’ use of bank accounts and bank credit, access problem that has hindered the development of especially for Kenyans with low income, no salaried inclusive financial sectors in many African countries. This paper is a product of the Finance and Private Sector Development Team, Development Research Group. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted at rcull@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Improving Access to Banking: Evidence from Kenya FRANKLIN ALLEN, ELENA CARLETTI, ROBERT CULL, JUN “QJ� QIAN, LEMMA SENBET, AND PATRICIO VALENZUELA ∗ JEL CODES: G2; O1; R2. KEY WORDS: Equity Bank, bank penetration, bank account, minority language, microfinance. ∗ Allen is from the University of Pennsylvania, Carletti is from the European University Institute, Cull is from the World Bank, Qian is from Boston College, Senbet is from University of Maryland, and Valenzuela is from the University of Chile. We thank Financial Sector Deepening Trust Kenya (FSD Kenya), and in particular David Ferrand, for providing the household surveys. We have benefited from helpful comments from Thorsten Beck, Sankar De, Samuel Makome, Randall Morck, Jonathan Robinson, Moses Sichei, John Staley, and seminar participants at Wesleyan University, the 2011 NBER African Successes conference in Zanzibar and the 2012 Summer Research Conference in Finance at the Indian School of Business in Hyderabad. We are grateful to the Gates Foundation, NBER, and the authors’ respective institutions for financial support. We are responsible for all the remaining errors. I. Introduction Despite extensive economic and financial sector reforms over the last few decades, many Sub- Saharan African countries still face a severe financial development gap relative to not only the advanced economies but other peer developing economies (Allen, Carletti, Cull, Qian, Senbet, and Valenzuela, 2012; ACCQSV). A key obstacle to financial development is access of the disadvantaged to finance, which would promote economic growth at the broadest scales. Recent studies show that in Sub-Saharan Africa over 80 percent of the adult population lacked an account in the early 2000s, well above the world average of 50 percent (Chaia et al., 2011; Honohan, 2008). The financial development issues cannot be ignored for the African countries in light of the ample empirical evidence for a positive linkage between financial development and economic development (Burgess and Pande, 2005; Levine, 2005; Rajan and Zingales, 1998) and of the widespread poverty in Africa. The positive relationship between financial development and economic development is in fact suggestive of a positive linkage between finance and poverty alleviation (Beck, Demirguc-Kunt and Levine, 2007; Clarke, Xu and Zhou, 2006). The main question is how to spur access to finance, in particular among the disadvantaged. While the success of microfinance institutions (MFIs), such as Grameen Bank, has captured the attention of many economists and policy makers, some MFIs are now beset by non-repayment problems and high cost of financing. Meanwhile, most established commercial banks view the sectors targeted by MFIs as ‘unbankable.’ In fact, we are only beginning to understand the specific policies and institutions that best promote financial inclusion in environments endowed with asymmetric information, weak institutions and the absence of basic infrastructures necessary for banking. In this paper, we examine bank branch penetration and financial access in Kenya, a country which has made significant strides in financial inclusion and development. The study is important for several reasons. First, Kenya has experienced a significant bank branch expansion in recent years, but it is still characterized by an important latent demand for banking services (Dupas et al., 2012). Although informal savings mechanisms, such as rotating savings and credit groups (so-called ROSCAs) and savings in livestock are still prevalent, even rural poor respondents mention ‘a commercial bank’ as their preferred saving mechanism if they had access to all the alternatives. With robust economic growth it is not surprising to witness a strong bank branch expansion in recent years. 2 Second, Kenya provides an interesting ‘laboratory’ for financial access, given the emergence of Equity Bank, a pioneering institution that devised a banking service strategy targeting low income clients and traditionally under-served territories. Listed on the Nairobi Stock Exchange, it has no government ownership share and has played a key role in integrating microfinance with formal finance while being profitable in the process. Figure 1 shows that Equity Bank has experienced an explosive growth of its assets and a significant expansion of its banking services, becoming the fifth largest bank in Kenya in terms of share of gross assets and deposits of the whole system. The number of deposit and loan accounts of Equity Bank represents around 50% and 30% of the total number of deposit and loan accounts in Kenya, respectively. Its model is also expanding to other African countries, with Equity Bank ending the year 2010 with 5.4 million clients in Kenya, 474,000 in Uganda and 28,000 in South Sudan. Third, we overcome the challenge of obtaining micro-level data, merging a new dataset on bank branch penetration of all the commercial banks across all parts of Kenya with nationally representative household surveys of financial usage conducted by the Financial Sector Deepening Trust of Kenya (FSD Kenya) in 2006 and 2009. This is important given that a major challenge for empirical research on financial access is the paucity of micro-level data, which is useful to address potential endogeneity biases stemming from the nonrandom nature of bank expansion (Burgess and Pande, 2005). The use of micro-level data within a country has additional benefits. Because institutional and macroeconomic context tends to be more homogeneous within a single country, we avoid obstacles confronted in cross-country studies of financial development and inclusion regarding omitted (or poorly measured) variables. In addition, as emphasized by Beck et al. (2007), country-level measures of financial outreach assume a uniform distribution of bank outlets within a country’s various areas and across its population, ignoring that bank branches are concentrated in urban areas and that bank expansions tend to target specific segments of the population. We study two related questions. Do different types of banks pursue different branching strategies? And if so, do these different expansion strategies lead to different effects on households’ access to banking services? In our first main set of results, we find that Equity Bank pursues different branching strategies from other banks. While banks of all ownership types (including Equity Bank) opened a greater number of branches in urban, highly populated and English speaking districts, 3 Equity Bank was more likely to expand to underserved districts than other types of banks between 2006 and 2009. In our second main set of results, we find that the presence of Equity Bank branches has a strong positive impact on the residents’ likelihood of having access to both bank accounts and credit. For example, our regressions imply that the penetration of Equity Bank into a new district increases the probability of a household having a bank account between 4 and 9 percentage points. Since the proportion of households that had a bank account rose from 14 percent in 2006 to 23 percent in 2009, the impact of Equity Bank on financial access appears to be economically meaningful. The positive effect of Equity Bank on financial access is particularly strong for individuals with low income and less education, without their own permanent house, and without a salaried job. These findings support the argument that the business model of Equity Bank, which targets underprivileged segments of the population and bring in sustainable profits, has paid off in terms of greater financial inclusion. We also find that the presence of domestic private banks (including Equity Bank), government banks and foreign banks is positively related to use of bank accounts at the district level. Interestingly, the positive impact of the foreign banks is the smallest among the three groups of banks. A potential explanation for these differences is that foreign banks traditionally have targeted the high-income segment of the population. These strategies are reflected in the differential requirements for opening a bank account. For instance, Equity Bank only requires an ID and a photo (which can be taken in a bank branch) to open a bank account. By contrast, foreign banks, such as Barclays, require a minimum balance of 20,000 Ksh. (about US$ 222), a prohibitive amount for the lowest-income segment of the population. We undertake a number of steps to ensure that our main results are not driven by alternative factors and potential endogeneity biases. We include individual and household characteristics to control for potential changes in (survey) sample composition between 2006 and 2009, and district fixed effects to control for time-invariant district characteristics. We also include the number of bank branches, which can be viewed as a parsimonious control for time-varying district characteristics, such as economic development and bank competition. The most important, potential endogeneity bias stems from the fact that banks’ branch penetration is not a random event. In fact, we show that between 2006 and 2009 Equity Bank shifted its branching policy moving more toward less 4 developed districts as compared to other banks. We exploit this policy shift and use the district-level proportion of people speaking a minority language (not English or Swahili) as an instrument for Equity Bank presence. Importantly, the fraction of people from a district speaking a particular language should not be directly linked to whether a particular individual (who may speak the same or different language) has access to a bank account. Indeed, our tests confirm the validity of this instrument, and our instrumental variables (IV) regressions (Probit and GMM) confirm the positive and significant impact of Equity Bank presence on households’ access to financial services. This paper extends the emerging literature on financial access and inclusion in developing economies as well as contemporaneous studies that focus on Kenya. Several papers have recently attempted to identify policies, services, and innovations that have a causal impact on financial inclusion through experiments based on randomized controlled trials (RCTs) conducted in developing countries. 1 While RCTs represent an important breakthrough in the study of financial inclusion (and other topics in development economics), they may carry some limitations: sample sizes are small, contexts differ, and small variations in experimental design can affect results (Ravallion, 2009). Thus, assessing the external validity of results generated in a specific experimental setting is a thorny issue in developing policy advice. Moreover, important developments in financial inclusion, such as the one studied in this paper, are taking place without the aid of experiments and are not amenable to study via randomization. We, therefore, view our nationally-representative evidence on branch penetration and financial access as complementary to the experimental evidence from RCTs, which usually focuses on a small set of communities. 2 Our study is closest to Burgess and Pande (2005) and Bruhn and Love (2009). Using state-level data and exploiting a shift in the branch licensing policy in India, Burgess and Pande (2005) investigate the impact of access to banking services. They find that bank branch expansion lowered rural poverty, and increased non-agricultural output. Bruhn and Love (2009) show that the entry of Mexico’s Banco Azteca, which opened branches in the stores of Grupo Elecktra, a large retailer of consumer goods, in 2002, was associated with an increase in the start-up of small businesses and a 7 1 Some recent papers using RCTs to study the impact of financial inclusion in developing economies are Banerjee et al. (2010), Brune et al. (2011), Dupas and Robinson (2009), Dupas et al. (2012), and Karlan and Zimmerman (2010, 2011). 2 Our paper also expands international evidence on the determinants of access to financial services in less developed economies. Using survey data for 29,000 households from 29 transition economies, Beck and Brown (2011) investigate how access to financial services is related to household characteristics, bank ownership structure and the development of the financial infrastructure. Beck et al. (2010) and Beck et al. (2011) also use the same surveys that we do to describe aspects of financial inclusion in Kenya. 5 percent increase in income levels. The Kenyan episode that we investigate lacks the quasi- experimental characteristics of the Azteca study, because the increase in bank branch penetration was not random. However, the sheer magnitude of the increase in financial sector participation makes this an episode worthy of study. Moreover, similar in spirit to Burgess and Pande (2005), we exploit a shift in a bank branching strategy and use an IV approach to mitigate the potential endogeneity problems emerging from nonrandom branch expansions. In addition to the emergence of Equity Bank, the Kenyan financial landscape has witnessed another innovation over the last decade that has gained global attention: the introduction of M-PESA, which operates through mobile telephones and greatly facilitates money transfer and remittance by the poor. With the same household surveys that we use, Mbiti and Weil (2011) show that M-PESA is primarily used for transferring money from individual to individual rather than as a vehicle for saving, and that its use increases the probability of being banked. M-PESA can, therefore, be seen as being complementary to the fuller set of financial services offered by large financial institutions, especially Equity Bank. The remainder of the paper is organized as follows. Sections II and III describe the data and the methodologies, respectively. Section IV presents the empirical results stemming from baseline specifications and Instrumental Variables (IV) analyses, and a set of robustness checks and a discussion of those results. Finally, Section V concludes. II. Data Our dataset on bank presence is based on branch-level information that we aggregate to construct a district-level panel on the number of branches by bank. This data was constructed from multiple sources, including phone calls, official websites, banks’ annual reports and government publications. The panel data set covers 45 commercial banks that operated in 65 Kenyan districts in the period between 2006 and 2009. This new dataset allows us to explore the relationship between bank expansion and access to banking services over time and to exploit within-district variation in bank presence. 3 In addition, this dataset set allows us to characterize the expansion of the whole banking 3 Although Kenya is currently divided into 46 districts, we take advantage of a more disaggregated country division as of the 1999 census. In that census, Kenya comprised eight provinces (Central Coast, Eastern, Nairobi, North Eastern, Nyanza, Rift Valley, and Western) that were subdivided into 69 districts; the survey data cover 65 of those districts. 6 system as well as the expansion of specific banks (e.g., Equity Bank) and ownership characteristics of banks (e.g., private domestic banks, foreign banks and government banks). As shown in Table 1, the Kenyan banking system expanded greatly in terms of the number of branches in recent years. Between 2006 and 2009, the total number of bank branches in Kenya increased from almost 600 to almost 1,000 (a 68 percent increase). The expansion involved all ownership categories of banks, especially domestic private banks, among which Equity Bank played an important role. The number of Equity Bank’s branches increased from 44 to over 110 in the period 2006-2009, representing an expansion of 155 percent. No other bank experienced such expansion, in terms of the number of branches, in the same period. The number of bank branches of Barclays and Kenya Commercial Bank (i.e., the banks with the largest branch networks together with Equity Bank) grew only 92 percent and 44 percent, respectively. As Table 2 shows, the strong branch expansion in Kenya occurred not only in urban districts but also in rural, arid and semi-arid districts. Our data on bank branches across districts are augmented by the FinAccess’ households surveys. The 2006 surveys consist of 4,420 completed interviews, while the 2009 surveys consist of 6,598 completed interviews. The sample was constructed using a stratified three-stage design. In the first stage, the number of households to be allocated to each district was assigned. In the second stage, clusters were randomly selected from the district with each cluster comprising an equal sample of 10 households. Finally, in the third stage, individuals within the household were randomly selected based on a listing of all household members aged 16 years and above, using the Kish Grid-diagram. 4 We use the FinAccess surveys in our analysis for a number of reasons. First, it captures important dimensions of financial inclusion as it was specifically designed to measure access to financial services in Kenya. Second, it also provides information on individual and household characteristics, such as education level, gender, asset ownership, income source, age group, language, location and household size. Third, it is a nationally representative household survey that covers the universe of districts in Kenya, alleviating potential sample selection bias problems. Fourth, its implementation over different years allows us to explore the within-district time variation in financial inclusion. 4The Kish Grid-diagram is a method that uses a pre-assigned table of random numbers for selecting individuals within a household to be interviewed. 7 Based on the surveys, we construct two measures of access to banking services. The first measure is a dummy variable indicating whether an individual has a bank account. The second is another dummy indicating whether an individual has a loan from a bank. Table 3 shows that the proportion of individuals having a bank account in Kenya increased considerably in the sample period, going from 14 percent in 2006 to almost 23 percent in 2009. The proportion of individuals having a loan from a bank showed a more modest increase from 3 percent to 4 percent. While 95 percent of the individuals that had a loan from a bank also had a bank account, the remaining 5 percent did not. Table 3 also reports descriptive statistics for the other individual and household-level variables that we employ from the FinAccess survey, and these will be used as controls in our tests below. III. Methodologies With data on bank branching and household surveys, we examine two related questions. Does Equity Bank pursue different branching strategies from other commercial banks in Kenya? And if so, do these different expansion strategies lead to different outcomes on households’ access to banking services? For the first question, we relate the number of branches by districts to a number of district-level characteristics. Our baseline specification takes the form: Branchesdt = α + µ t + β X d + ε dt , (1) where Branches is the number of bank branches in district d at time t for Equity Bank; for other banks it is the number of branches over the total number of banks in each group (based on ownership types) in district d at time t, to put them on a comparable basis with Equity Bank branches at the district level. µt is a dummy variable for 2009, and Xd is a set of district level control variables for 2006; these variables indicate whether the district is a rural or arid and semi-arid district, the population density of the district and the proportion of the population speaking a minority language or Swahili. We estimate this model by ordinary least square (OLS), and cluster standard errors by district for four types of banks: foreign (private) banks, government-owned and government-influenced banks, domestic private banks (excluding Equity Bank) and Equity Bank. Furthermore, to explore if the branch expansion of Equity Bank between 2006 and 2009 targeted more under-served districts, we augment our baseline specification with an interaction term between the district level proportion of people speaking a minority language and the 2009 indicator. 8 In order to study the second question, we estimate a Probit model of access to banking services. The dependent variable, Bankidt, is a dummy indicating whether household i has a bank account in district d at the time t. Our main independent variables are, Equitydt, a dummy variable indicating whether Equity Bank operates in district d at time t, and, Branchesdt, the total number of bank branches of other banks operating in district d at time t. The total number of bank branches can be viewed as a parsimonious control for time-varying district characteristics, such as economic development and bank competition. 5 The inclusion of this variable is fundamental to our identification strategy, given the nonrandom nature of bank expansion in which banks favored opening branches in more developed districts. All specifications include district fixed effects to control for time-invariant district heterogeneity. Thus, our baseline specification takes the form: Pr( Bankidt = 1) = F (α + µ d + β Equity dt + γ Branchesdt + φ zidt ) , (2) where Pr denotes probability, F is the standard normal cumulative distribution function, µd is a set of district dummies, and zidt is a vector of control variables at the individual and household level. They indicate household size, education, gender, wealth (measured by an asset score), house ownership, type of job, age and language. 6 The parameters α, β , γ and φ are estimated by maximum likelihood, with error terms clustered at the district-year level to account for possible correlations in errors across households within districts for each specific year. We also augment our baseline regression with a time (2009) dummy and its interaction with Equitydt to test whether the impact of Equity Bank is part of a general trend or not. However, Eq. (2) may still produce biased estimates of the impact of bank penetration on financial access due to potential endogeneity concerns. The source of endogeneity is that bank branch expansion is not a random event. More specifically, in the absence of a policy shift in Equity Bank’s branch penetration, we would expect relatively greater branch penetration in more developed districts or in districts with more growth opportunities, as more growth and growth opportunities generate more demand for financial services. There can be greater demand for financial services even without organic growth in a region: for example, households in the region would like to use bank branches to receive remittances from their relatives working outside Kenya. To summarize, if 5 Note that our results are very similar whether or not we include Equity Bank’s branches in the calculation of the total number of branches. 6 The asset score is higher if the individual owns a greater number of physical assets such as a mobile phone, television, or microwave. There are 20 asset categories and the ownership of each of them increases the asset score by one. 9 districts with more demand (or future demand) for financial services are more likely to provide access to finance, then β would be an overestimate of the true impact of Equity Bank branch expansion. That is, Equity Bank’s presence could be more a feature of a relatively developed banking district than a cause of greater uptake of accounts. On the other hand, if Equity Bank succeeded in its policy of opening more branches in less developed districts or districts with less (perceived) growth opportunities than other banks, then the same logic above suggests that β would underestimate the true impact of Equity Bank branch expansion. In order to alleviate the above endogeneity problems, we rerun our baseline regressions on the probability of having a bank account using an Instrumental Variables (IV) approach. Our goal is to identify the increase in bank accounts that occurred because of Equity Bank’s shift in branching strategy. Specifically, we use the district-level proportion of people speaking a minority language (not English or Swahili) and its interaction with a 2009 dummy variable as instruments for Equity Bank branch expansion. These instruments are based on the fact that Equity Bank implemented a strategy of serving more vulnerable sectors of the population and underdeveloped areas from 2006 to 2009, and of speaking to clients in their native tongue. District-level proportions of people speaking a minority language constitute valid instruments for Equity Bank branch expansion if these variables only affect the probability of having a bank account through bank expansion. It is reasonable to argue that the fraction of people from a district speaking a particular language should not be directly linked to whether a particular individual or household has access to a bank account, which is our outcome variable. However, it is important to note that this assumption does not imply that the language that an individual speaks is not an important determinant of having a bank account. In fact, we control in all our specifications for individual languages. With the instruments, we then estimate Eq. (2) with both an IV (linear) GMM model and an IV Probit model. With this empirical strategy, we try to identify the increase in account uptake that is attributable to Equity Bank’s unique approach to attracting clients. Next, we augment our baseline specification with interaction terms between Equitydt and a set of characteristics at the individual level (i.e., wealth, house ownership, education and type of job) to examine potential heterogeneities in the impact of Equity Bank presence on access to financial services. These heterogeneities allow us to assess whether Equity Bank has effectively increased 10 access to financial services by the poor and low income people and hence fostered financial inclusion of the disadvantaged. We also examine the impact of Equity Bank presence on financial inclusion by estimating an Ordered Probit model of the usage of banking services. In this model, our dependent variable takes the value “0� when the individual has neither a bank account nor a loan from a bank, the value “1� when the individual has a bank account, and the value “2� when the individual has a loan from a bank. 7 Similar to our Probit model in Eq. (2), our main explanatory variables are Equitydt and Branchesdt. Finally, as an additional robustness check, we apply a Difference-in-Differences (DD) estimation strategy to recover the treatment effect stemming from the possible shift of the Equity Bank branching policy that produced a sharp change in its penetration (see Table 1). Our exercise compares the difference in the probability of having a bank account before and after the entry of Equity Bank for districts affected by this branch expansion with the same difference for unaffected districts. In order to conduct our DD estimations, we restrict our sample to districts without any Equity Bank branch in 2006 and include time-fixed effects. Districts without any Equity Bank branch in 2006 and 2009 represent the control group and districts without an Equity Bank branch in 2006 but with at least one branch of Equity Bank in 2009 represent the treatment group. Using this subsample of districts, our specification takes the following form: Pr( Bankidt = 1) = F (α + µ d + µ t + β Equity dt + γ Branchesdt + φ zidt ) , (3) where µt is a time dummy variable. This specification is the most basic DD setup (with two periods and two groups), which is valid under the assumption that changes in the probability of having a bank account over time would have been the same in individuals in both treatment and control districts in the absence of Equity Bank penetration. IV. Empirical Results The primary goal is to establish evidence that the supply-side factors as represented by the expansion of bank branches, and in particular of Equity Bank, has increased access to financial 7 Note that 95.2 percent of the individuals who have a loan from a bank also have a bank account. 11 services. To do this, we first examine the branching strategies followed by Equity Bank and other banks in Kenya, and the results are reported in Table 4. Then, we evaluate the impact of Equity Bank presence and bank branch expansion on the use of bank accounts, and report the results in Tables 5 to 7. Tables 8 to 11 report additional results and robustness checks. IV.1 Banks’ Expansion Strategies Different types of banks are likely to follow different expansion strategies. While foreign banks may cherry-pick a set of elite customers and prefer urban and developed areas, domestic banks may exploit their superior knowledge of culture, social norms, and local communities and have a comparative advantage in rural and underdeveloped areas. For example, as part of its expansion strategy, Equity Bank emphasized that local languages be spoken in its branches, which is important considering that 30-40 percent of the people in central Kenya cannot speak Swahili and only speak a minority language. In addition, government-owned banks may pursue non-profit goals in their branching and expansions. As a first exploration of this issue, Figure 2 presents the number of bank branches in 2006 and 2009 for three bank groups: foreign (private) banks, government-owned and government-influenced banks, and domestic private banks. 8 The figure also splits the districts along different dimensions in terms of urbanization and demographics. The figure illustrates three stylized facts. First, between 2006 and 2009 all bank groups undertook substantial branch expansions. Second, all bank groups showed a higher penetration in urban, English speaking, and highly populated and educated districts. Third, domestic private banks had greater presence, in terms of branch numbers, in underdeveloped districts, while foreign banks had the least presence in those districts. These patterns suggest that domestic private banks, whose expansion has been led by Equity Bank (see Table 1), played a vital role in fostering banking services in underdeveloped districts. The foregoing stylized facts motivate our formal estimation and analysis of the determinants of bank penetration in Kenya. Table 4 reports OLS estimates on the determinants of bank penetration by district (Eq. 1), comparing Equity Bank with all three groups of other banks (by their ownership structure) in the banking system. The dependent variable for regressions on Equity Bank is the 8 Government-owned banks are those in which the government owns a majority of shares. In the ( two) government- influenced banks the government maintains a large, but non-controlling ownership share. Throughout most of the paper, we refer to both groups collectively as “government banks.� 12 number of branches by district. For other groups of banks, we divided the number of branches in a district by the number of banks in the group, to put them on a comparable basis with Equity Bank branches at the district level. Our independent variables indicate the degree of urbanization, the population density of the district and the proportion of non-English speaking Kenyans. In addition, we include a dummy variable for 2009 in order to explore the extent of branch expansion during the period 2006-2009. The results suggest some similarities, as well as some disparities, between Equity Bank and the others in terms of their branching strategies. On the one hand, all types of banks including Equity Bank have greater branch penetration in urban and English speaking districts (Models 1 to 4). On the other hand, Equity Bank differs on several dimensions. First, the coefficients associated with rural and arid and semi-arid districts are either statistically insignificant or marginally significant at standard confidence levels for Equity Bank, while they are negative and highly significant for the other banks. Second, the expansion of Equity Bank, as indicated by the positive coefficient on the 2009 dummy variable, is highly significant. That coefficient is also significant for the rest of the banking system, but its size is only one-tenth to one-fifth that of the coefficient for Equity Bank. Third, the sensitivity of branch penetration to population density is lower for Equity Bank than for others. In fact, the coefficient for population density is negative and highly significant for Equity Bank in Model 4, indicating a preference for less densely populated districts (other factors held constant). This coefficient is also negative for other banks, but is not significant or only marginally significant. Fourth, the intercept/constant of the regression models remains much larger for Equity Bank’s branching than for other bank types. This shows Equity Bank’s greater general tendency to open branches regardless of district characteristics. Overall, the results in Models 1 to 4 suggest that Equity Bank has experienced a significant branch expansion relative to other banks between 2006 and 2009, particularly in underdeveloped non-urban districts. The sensitivity of branch penetration to population density is particularly interesting in view of the African financial development gap observed at a macro level (ACCQSV, 2012). One of the major findings is that population density appears to be more important for banking sector development as measured by liquid liabilities and private credit (relative to GDP) in Africa than in other developing countries. More specifically, while population density affects financial development in all developing countries, the relationship is much steeper for Africa. It is challenging to resolve the African 13 financial development gap, but the evidence on Equity Bank is encouraging, since it points to the potential for financial institutions with novel strategies to foster financial inclusion and help promote convergence of the African financial development model with the rest of the developing world. One component of Equity Bank’s expansion strategy has been to target the segment of the population speaking local languages. There are two ways of doing this. The first one is penetrating districts characterized by a high proportion of the people speaking minority languages. The second way is to use minority language in the branches. We start by exploring whether Equity Bank has moved more than other banks to districts in which English or Swahili is not the predominant language. By interacting the district-level proportion of people speaking Swahili or minority languages with the 2009 dummy variable, we get a better sense of the types of districts that Equity Bank was entering between 2006 and 2009. The interactions between the 2009 dummy variable and the language variables are positive and significant (Model 8). While the negative coefficients for minority dialects and Swahili speakers are larger (in absolute value) than the coefficients on the interactions, the positive coefficients on the interactions do indicate that Equity Bank moved toward serving non- English speakers from 2006 to 2009. A similar pattern holds for non-Equity Bank branches (Models 5 to 7), but: (a) the coefficients on the interactions between minority language and the 2009 dummy are not significant, (b) the coefficients on the interactions between Swahili and the 2009 dummy are much smaller than for Equity Bank, and (c) the ratio of the (absolute value of the) coefficient for the language dummies to the coefficient for the interactions is much larger for the non-Equity Bank sample. Therefore, it appears that Equity Bank moved more toward minority-language speakers than other banks between 2006 and 2009. Thus, the overall evidence from Table 4 suggests that Equity Bank’s branching strategies are distinct from those of other banks; such branching patterns suggest that Equity Bank was promoting financial inclusion during the period of 2006-2009. IV.2 Bank Penetration and Financial Access: A Closer Look at Equity Bank Our previous results suggest that Equity Bank has pursued different branching strategies from other commercial banks. Our next step is to explore whether its expansion strategy has led to different 14 effects on households’ access to banking services. Table 5 reports the marginal effects when estimating our baseline Probit model of households’ access to banking services (Eq. 2) by maximum likelihood with standard errors clustered by district and year (see Petersen, 2009). We report first a regression with individual/household characteristics, district level bank branch expansion by other banks and district fixed effects (Model 1). Next, we differentiate banks by their ownership structure (i.e. foreign banks, government banks and domestic private banks). We run these regressions with and without the dummy variable associated with Equity Bank presence (Models 2 and 3); note that the domestic banks category includes Equity Bank in those models. Finally, instead of using an indicator denoting the presence of a certain type of bank, we use the number of branches for each bank ownership type. Again, we run these regressions with and without the dummy variable associated with Equity Bank presence (Models 4 and 5), and again, the domestic banks’ branches include those of Equity Bank. Table 5 shows, first, that the presence of Equity Bank in a specific district is strongly positively related to the residents’ probability of having a bank account, which goes beyond the effect of bank expansion and presence of other commercial banks (Models 1, 3 and 5). This effect is highly statistically significant regardless of the model specification. The coefficient in Model 1 suggests that the penetration of Equity Bank in a new district increases the probability of having a bank account by roughly 4 percentage points after controlling for the total number of other banks’ branches. Model 3 also suggests that the impact of the presence of Equity Bank in a specific district goes beyond the impact of the presence of a standard domestic bank by around 2 percentage points. As noted, according to the FinAccess survey, the proportion of people having a bank account in Kenya was 14 percent in 2006 and 23 percent in 2009. Therefore, the magnitude associated with Equity Bank presence on financial access is also economically meaningful. This finding is consistent with the notion that the business model of Equity Bank, which targeted middle- and low-income segments of the population using such strategies as the local language requirement in its branches, has paid off in terms of greater financial inclusion. Second, we find that the presence of domestic private banks (including Equity Bank), government banks or foreign banks all have a positive impact on access to bank accounts at the district level (Model 2). However, we find a hierarchy in the effects of different types of banks: the effect of foreign banks is 2.5 percent (and statistically insignificant), the effect of government banks is 3 15 percent, and the effect of domestic private banks is 4 percent. A potential explanation for these differences is that while domestic private banks have followed a strategy of targeting all segments of the population, foreign banks traditionally have targeted the high-income segment of the population. This is consistent with the finding in Beck and Brown (2011) that foreign banks may cherry-pick their clients among households in transition economies. These strategies are also reflected in the differential requirements for opening a bank account. Third, the number of bank branches by district, a measure of banking sector outreach and competition, appears to be an important determinant of usage of banking services. The coefficient on the total number of bank branches is positive and highly statistically significant (Model 1). Perhaps surprisingly, the results suggest that the positive impact of a new branch is marginally stronger for foreign banks (Models 4 and 5). Most of the control variables come in significantly in the expected directions. For example, individuals with large families and with no knowledge of English are less likely to have a bank account, while education, wealth (measured by an asset score), house ownership, age and income stability (i.e. salaried workers) are positively related to having a bank account. The overall fit of our model is good, with the Pseudo R2 around 0.33 in all specifications. 9 In Table 6, we test whether Equity Bank had a unique impact on the probability of having a bank account or whether the Equity Bank presence dummy variable is just picking up the effect of a larger (country-level) trend on bank expansion. Thus, Table 6 augments our main baseline regression (in Table 5) with a 2009 dummy and with its interaction with the bank penetration variables. Model 1 shows a systemic component increasing (of 6.4 percent), on average, a household´s probability of having a bank account in Kenya between 2006 and 2009. However, as shown in Models 2 to 4, the positive and highly significant coefficients on the interaction between Equity Bank presence and the 2009 dummy, and the insignificant coefficients on the other interaction terms as well as the 2009 dummy, suggest that the increased probability of having a bank account is mainly driven by the expansion of Equity Bank between 2006 and 2009. 9These results are also consistent with international evidence on the determinants of the use of bank services. See, for example, Beck and Brown (2011). 16 IV.3 Addressing Potential Endogeneity Problems As discussed earlier, there are potential endogeneity problems arising from non-random bank branch expansion. We use an IV approach to address the endogeneity issue. Table 7 reports the results from our two-stage IV procedures; the instruments are the proportion of people speaking a minority language (not English or Swahili) in a given district and its interaction with the 2009 dummy. The appropriateness of our instruments is based on our previous results showing that the proportion of people speaking a minority language is directly associated with the penetration strategy of Equity Bank (Table 4), and the reasonable assumption that this variable does not directly affect the individuals´ likelihood of having a bank account. It is important to note that this assumption does not contradict that the language spoken for a specific individual can be an important determinant of her/his own probability of having access to banking services. In fact, all our regressions control for languages at the individual level. We estimate IV GMM and IV Probit models for our baseline regression reported in Model 1 of Table 5. 10 The results from the second-stage (Models 1, 3 and 4 of Table 7) indicate that the impact of Equity Bank presence on the probability of having a bank account remains positive and highly statistically significant. The main difference with our baseline Probit estimator is that the magnitude of the coefficient increases from 0.04 (Table 5, Model 1) to 0.09 (Table 7, Models 1 and 3), which suggests an attenuation bias in our baseline Probit estimates. This is consistent with the premise that, if more underdeveloped districts are less effective at providing access to finance, the baseline Probit estimates would underestimate the true impact of Equity Bank branch expansion. The results from the first-stage (Columns 2 and 5) suggest that Equity Bank altered its branching policy in the period 2006-2009. The negative coefficient on the district-level proportion of people speaking minority languages and the positive coefficient on the interaction between people speaking minority languages and a 2009 dummy variable suggest that Equity Bank moved toward serving more clients who speak a minority language. These results are similar to the ones obtained from Table 4 that indicates that Equity Bank moved toward serving minority language speaking districts. Table 7 also presents the p-values for the Partial R-squared of excluded instruments and Hansen’s J test of over-identifying restrictions (Baum, Schaffer and Stillman, 2003). The R-squared of the 10In unreported regressions, we also conduct our IV approach for all of our specifications. The main results in this paper remain qualitatively unchanged regardless of the specification that we choose. 17 excluded instruments indicates that the instruments and endogenous variables are correlated, even after netting out the effects of all other exogenous variables. In addition, the J test cannot reject the null hypothesis that all the instruments are valid. These results are consistent with the validity of the instruments for Equity Bank’s branching strategies. Overall, the results in Table 7 suggest that the positive impact of Equity Bank on access to banking services is robust to controlling for potential endogeneity bias. IV.4 Some Extensions a. Who Benefits from Banking Services?: The Case of Equity Bank Until now we have presented evidence consistent with the propositions that Equity Bank’s branch strategy has targeted under-served geographic areas and had a substantial impact on access to banking services in Kenya. Specifically, the results in our previous sections suggest that Kenyans residing in districts in which Equity Bank started operations have a higher probability of having a bank account. Since Equity Bank’s business model is based, at least in part, on providing financial services to the rural poor, we would expect the impact of its presence to be stronger for the underprivileged segments of the population (in terms of physical and human capital and income stability). With this idea in mind, Table 8 explores potential heterogeneities in the impact of Equity Bank presence on the probability of having a bank account. In particular, Models 1 to 4 augment our baseline regression in Table 5, Model 1 with four interaction terms respectively: the interaction of the Equity Bank presence dummy with a score for the ownership of a set of physical assets, a dummy indicating if the individual is a homeowner, a dummy indicating whether the individual has a secondary and/or tertiary education level and a dummy indicating whether the individual has a salaried job. All the coefficients associated with the interactions terms have a negative sign and are highly significant, suggesting that the impact of Equity Bank on access to banking services is, indeed, heterogeneous across individuals with different characteristics. It is stronger for Kenyans with less wealth, without own permanent house, without secondary and/or tertiary education and without a salaried job. Therefore, these results are consistent with Equity Bank’s mission of extending financial services to the population segments generally ignored by traditional commercial banks. Finally, it is worth noting that all the main results and the signs of the coefficients associated with the new 18 interaction terms remain unchanged when all the interactions terms are included at the same time (Model 5). In order to ensure that the interactions terms are not picking up the effects of a general bank expansion, we augment the specifications reported in Table 8 with new terms corresponding to the interactions between dummy variables indicating the presence of foreign banks, government banks and domestic banks (including Equity Bank) and the individual characteristics used in the interactions terms in Table 8. The results from unreported regressions indicate that our previous results are robust to the inclusion of the new interaction terms. On the one hand, the coefficients associated to the interactions between Equity Bank presence and individual/household characteristics remain qualitatively unchanged in eleven of our twelve new specifications. On the other hand, all coefficients associated with the new interaction terms (with the exception of one coefficient) are not statistically significant suggesting that the heterogeneous impact of Equity Bank is not necessarily part of a general trend in the banking system in Kenya. b. Access to Credit We expand our baseline results to consider the impact of Equity Bank presence and bank penetration on the probability of having a loan from a bank. Table 9 reports the coefficients and marginal effects from an Ordered Probit model of usage of banking services. Our dependent variable takes the value "0" if the individual has neither a bank account nor a loan from a bank, the value "1" if the individual has a bank account, and the value "2" if the individual has a loan from a bank. The results indicate that Equity Bank presence increases the probability of both having a bank account (column 3) and a bank loan (column 4). In line with our baseline results in Table 5, Equity Bank presence increases by almost 4 percentage points the probability of having a bank account, and by 1 percentage point the probability of having a bank loan. IV.5 Robustness Checks a. Rural and Arid and Semi-Arid Districts Since bank expansions are likely to be nonrandom with banks favoring opening branches in more developed urban areas, we replicate our main results excluding urban districts (i.e., Nairobi and Mombasa) from the sample to ensure that the higher degree of development does not bias the 19 results. These regressions are reported in Table 10. The results remain qualitatively unchanged, with the impact of Equity Bank’s presence on the probability of having a bank account still positive and highly statistically significant, though somewhat smaller than in our baseline results. However, given that the share of the population with bank accounts tends to be much lower outside Nairobi and Mombasa, the 2-3 percentage point increase in bank account usage in rural and arid and semi-arid districts still indicates a substantial impact of Equity Bank branching. b. Difference-in-Differences Estimator Table 11 presents the results from our difference-in-differences estimator for the probability of having a bank account. This table replicates the specifications in Models 1, 3 and 5 in Table 5, using only a subsample of districts without the presence of Equity Bank in 2006 and including time fixed effects in our baseline specification. Therefore, the coefficients reported in the table indicate the change in the probability of having a bank account in districts where Equity Bank started operations between 2006 and 2009 relative to districts without Equity Bank presence during the same period. The results suggest that the impact of the penetration of Equity Bank in a new district on the probability of having a bank account remains positive and highly statistically significant. The main difference with our baseline results is that the magnitude of the impact decreases slightly from 4 percentage points (Model 1 in Table 5) to 3 (Model 1 in Table 11) percentage points. 11 IV.6 Interpretations and Limitations A potential concern regarding the interpretation of our results is that Equity Bank might not have actively pursued financial inclusion by entering underdeveloped areas. Rather, our main result could be accounted for by either demand side factors (e.g., more developed areas have more demand for financial services) or by Equity Bank’s anticipation of growth opportunities in certain areas that they hope to capture by opening new branches ahead of other banks. We note that even if such alternative interpretations hold, Equity Bank seems to have seized upon these opportunities more quickly than other banks and, regardless of Equity Bank’s motivation, the end result has been greater financial inclusion. 11Again, smaller impact outside urban areas is partly driven by the low base usage rates in the districts included in the difference-in-differences analysis. 20 Throughout the paper we have also provided a series of checks to help rule out competing interpretations. First, we employ the IV approach to address the nonrandom branching decisions of Equity Bank with instruments that capture differences in branching strategies between Equity and other banks; we continue to find similar (in fact, stronger) results. Second, we find that Equity Bank presence not only increased the likelihood of a household having a bank account, but also the chances of obtaining a loan from a bank (though the effect on this latter banking service is weaker, as fractions of households having bank loans remain low in most underdeveloped areas). We have limited ourselves to the question of whether Equity Bank’s branch expansion led to greater financial inclusion. We acknowledge, however, that provision of financial services on the scale of Equity Bank raises a host of additional issues that we do not address. For example, having over half of all bank accounts in a country in the hands of a single institution could raise concerns about the stability of the Kenyan banking system. While the bank accounts are individually small, they are vital to account owners with limited means. Such a concentration of bank accounts is likely to pose challenges for the deposit insurance system. Even if the system has the funds to cover a potential Equity Bank failure, any delay in accessing deposits by account owners as the failure is resolved poses its own set of stability issues. Equity Bank’s expansion into other countries also raises the possibility that poor financial performance in other markets could jeopardize the safety of Kenyans’ deposits. Encouraging competition for this market segment could help alleviate concerns about the concentration of deposits and accounts, but it raises additional issues. For example, it seems competition among micro-lenders in Andhra Pradesh and the over-indebtedness of many poor borrowers has contributed to the recent instability in Indian microfinance. All of these issues are worthy of investigation, but we leave this to future research. This also leads to the question of what is the role of financial inclusion in promoting economic growth. Simply expanding banking services to poor areas and population does not guarantee that the funds will be used properly (e.g., in spawning entrepreneurial activities) and loans repaid. This has been shown by the uneven successes of many microfinance institutions in developing countries. To this point, the gains in financial inclusion spurred by Equity Bank have focused primarily on bank accounts. In the future, however, it seems likely that provision of credit on a wider scale will be more lucrative than expanding provision of savings services. Although Equity Bank has been 21 financially successful thus far (as reflected in their books and stock performance), it remains to be seen whether the bank’s model is sustainable over the long term and whether it can be applied successfully in other African countries. V. Conclusions Most Sub-Saharan African countries face a financial development gap relative to other peer developing economies. A key obstacle to financial development is the access of the unprivileged segments of the population to finance, which would promote economic growth at the broadest scale. In view of the important role of financial inclusion in developing countries, in this paper we study the expansion strategies of Equity Bank and its impact on access to banking services in Kenya. The experience of Equity Bank in Kenya offers an interesting ‘laboratory.’ The banking network in Kenya has experienced rapid growth in recent years, expansion that has been led by Equity Bank, a pioneering institution that devised a banking service strategy targeting low income clients and traditionally under-served districts. We find this branch expansion is strongly associated with greater access to banking services and that Equity Bank has had a significant effect in enhancing households’ access to bank accounts and bank credit, especially for individuals in a vulnerable condition. These results highlight the importance of institutions, such as Equity Bank, with a business model focused on the provision of financial services to the population segments ignored by traditional commercial banks while generating sustainable profits in the process. Such institutions could be an important part of the solution to the financial access problem that has hindered the development of inclusive financial sectors in many developing countries. 22 References Allen, Franklin, Elena Carletti, Robert Cull, Jun Qian, and Patricio Valenzuela, 2012, Resolving the African development gap: Cross-country comparisons and a within-country study of Kenya, forthcoming in S. Edwards, S. Johnson, and D. Weil, eds: NBER Volume on African Economic Successes (University of Chicago Press). Banerjee, Abhijit, Esther Duflo, Rachel Glennerster and Cynthia Kinnan, 2010, The miracle of microfinance? Evidence from a randomized evaluation, Working paper, MIT. Baum, Christopher F., Mark E. Schaffer and Steven Stillman, 2003, Instrumental variables and GMM: Estimation and testing, Stata Journal, StataCorp LP 13, 1-31. Beck, Thorsten, and Martin Brown, 2011, Which households use banks? Evidence from the transition economies, Working paper 1295, European Central Bank. Beck, Thorsten, Robert Cull, Michael Fuchs, Jared Getenga, Peter Gatere, John Randa, and Mircea Trandafir, 2010, Banking Sector Stability, Efficiency, and Outreach in Kenya, in Christopher S. Adam, Paul Collier, and Njuguna S. Ndung’u, eds., Kenya: Policies for Prosperity, Oxford: Oxford University Press. Beck, Thorsten, Asli Demirgüç-Kunt, and Ross Levine. 2007, Finance, inequality, and the poor, Journal of Economic Growth 12, 27-49. Beck, Thorsten, Asli Demirgüç-Kunt, and Maria Soledad Martinez Peria, 2007, Reaching out: Access to and use of banking services across countries, Journal of Financial Economics 85, 234-266. Beck, Thorsten, Samuel Munzele Maimbo, Issa Faye, and Thouraya Triki, 2011, Financing Africa: Through the Crisis and Beyond, Washington, DC: World Bank. Bruhn, Miriam and Inessa Love, 2009, The economic impact of banking the unbanked: Evidence from Mexico, Policy Research Working Paper 4981, World Bank. Brune, Lasse, Xavier Giné, Jessica Goldberg, Dean Yang. 2011, Commitments to save: A field experiment in rural Malawi, Manuscript, University of Michigan. Burgess, Robin, Rohini Pande, 2005, Do rural banks matter? Evidence from the Indian social banking experiment, American Economic Review 95, 780-795. Chaia, Alberto, Aparna Dalal, Tony Goland, Maria Jose Gonzalez, Jonathan Morduch, and Robert Schiff, 2011, Half the world is unbanked, Working paper, New York University. Clarke, George R.G., L. Colin Xu, and Heng-Fu Zhou, 2006, Finance and income inequality: What do the data tell us? Southern Economic Journal 72, 578-596. Dupas, Pascaline, and Jonathan Robinson, 2009, Savings constraints and microenterprise development: Evidence from a field experiment in Kenya, Working Paper 14693, NBER. Dupas, Pascaline, Anthony Keats, Sarah Green, and Jonathan Robinson, 2012, Challenges in banking the rural poor: Evidence from Kenya's western province, Working Paper 17851, NBER. Honohan, Patrick, 2008, Cross-Country variation in household access to financial services, Journal of Banking and Finance 32, 2493-2500. 23 Karlan, Dean and Jonathan Zinman, 2010, Expanding microenterprise credit access: Using randomized supply decisions to estimate the impacts in Manila, Working paper, Yale University. Karlan, Dean and Jonathan Zinman. 2011, Microcredit in theory and practice: Using randomized credit scoring for impact evaluation, Science 332, 1278-1284. Levine, Ross, 2005, Finance and growth: Theory and evidence, in Philippe Aghion and Steven Durlauf, eds.: Handbook of Economic Growth (Elsevier Science, Amsterdam). Mbiti, Isaac and David N. Weil, 2011, Mobile banking: The impact of M-Pesa in Kenya, NBER Working Paper 17129. Petersen, Mitchell, 2009, Estimating standard errors in finance panel data sets: Comparing approaches, Review of Financial Studies 22, 435-480. Rajan, Raghuram and Luigi Zingales, 1998, Financial dependence and growth, American Economic Review 88, 559-586. Ravallion, Martin, 2009, Should the randomistas rule? The Economists’ Voice, 6, 1-5. 24 Table 1 Bank Branches over Time by Bank Name and Ownership Types This table reports the number of Kenyan bank branches by bank name and ownership types for 2006 and 2009, based on various data sources, including phone calls, official websites, banks’ annual reports and government publications. No Domestic private banks 2006 2009 Change % Change 1 African Banking Corporation Ltd. 8 10 2 25 2 CharterHouse Bank Ltd. 10 10 0 0 3 Chase Bank Kenya Ltd. 2 12 10 500 4 City Finance Bank Ltd. 1 1 0 0 5 Commercial Bank of Africa Ltd. 19 20 1 5 6 Co-operative Bank of Kenya Ltd. 52 83 31 60 7 Credit Bank Ltd. 4 5 1 25 8 Diamond Trust Bank of Kenya Ltd. 8 32 24 300 9 Dubai Bank Ltd. 4 4 0 0 10 Equatorial Commercial Bank Ltd. 3 5 2 67 11 Equity Bank 44 112 68 155 12 Fidelity Commercial Bank Ltd. 5 7 2 40 13 FINA Bank Ltd. 6 14 8 133 14 First Community Bank 0 18 18 - 15 Giro Commercial Bank Ltd. 6 8 2 33 16 Guardian Bank Ltd. 6 7 1 17 17 Imperial Bank Ltd. 6 13 7 117 18 Investments and Mortgages 9 17 8 89 19 Middle East Bank Kenya Ltd. 2 3 1 50 20 National Industrial Credit Bank Ltd. 16 16 0 0 21 Oriental Commericial Bank Ltd. 4 8 4 100 22 Paramount Universal Bank Ltd. 4 7 3 75 23 Prime Bank Ltd. 9 14 5 56 24 Southern Credit Banking Corporation Ltd. 9 10 1 11 25 Trans-National Bank Ltd. 9 13 4 44 26 Victoria Commercial Bank Ltd. 1 1 0 0 Sub-Total 247 450 203 82 No Foreign banks 2006 2009 Change % Change 1 Bank of Africa Kenya Ltd. 4 10 6 150 2 Bank of Baroda Kenya Ltd. 7 9 2 29 3 Bank of India Ltd. 4 5 1 25 4 Barclays Bank of Kenya Ltd. 62 119 57 92 5 Citibank N.A. 3 4 1 33 6 ECO BANK 9 19 10 111 7 Gulf African Bank Ltd 0 14 14 - 8 Habib Bank A.G. Zurich 4 5 1 25 9 Habib Bank Ltd. 4 4 0 0 10 K-Rep Bank Ltd. 22 30 8 36 11 Standard Chartered Bank Kenya Ltd. 31 35 4 13 12 UBA 0 4 4 - Sub-Total 150 258 108 72 No Government and government-influenced banks 2006 2009 Change % Change 1 CFC Stanbic Bank Ltd. 16 35 19 119 2 Consolidated Bank of Kenya Ltd. 12 13 1 8 3 Development Bank of Kenya Ltd. 1 2 1 100 4 Kenya Commercial Bank Ltd. 117 169 52 44 5 National Bank of Kenya Ltd. 33 43 10 30 Sub-Total 179 262 83 46 Total 576 970 394 68 25 Table 2 Descriptive Statistics, District Level This table reports descriptive statistics at the district level, based on 2006 and 2009 FinAccess household surveys, 1999 Kenyan census; data sources on bank branches include phone calls, official websites, banks’ annual reports and government publications. District Bank branches Bank branches Swahili (2006) Minority Urban Rural Arid and Area (km2) Population (2006) (2009) Language Semi-Arid (1999) (2006) Baringo 3 4 0.65 0.27 0 0 1 8,646 264,978 Bomet 2 4 0.44 0.50 0 1 0 1,882 382,794 Bondo 0 2 0.10 0.88 0 1 0 987 238,780 Bungoma 8 12 0.64 0.29 0 1 0 2,069 876,491 Buret 1 3 0.52 0.42 0 1 0 955 316,882 Busia 3 6 0.68 0.32 0 1 0 1,124 370,608 Butere Mumias 1 4 0.70 0.27 0 1 0 939 476,928 Embu 6 8 0.34 0.58 0 0 1 729 278,196 Garissa 1 8 0.02 0.98 0 0 1 44,952 329,939 Gucha 0 0 0.33 0.59 0 1 0 661 460,939 Homa Bay 5 6 0.08 0.90 0 1 0 1,160 288,540 Isiolo 2 5 0.73 0.23 0 0 1 25,698 100,861 Kajiado 6 21 0.83 0.03 0 0 1 21,903 406,054 Kakamega 10 12 0.64 0.30 0 1 0 1,395 603,422 Keiyo 1 0 0.78 0.13 0 0 1 1,439 143,865 Kericho 7 9 0.59 0.39 0 1 0 2,111 468,493 Kiambu 10 19 0.17 0.66 0 1 0 1,324 744,010 Kilifi 5 11 1.00 0.00 0 1 0 4,779 544,303 Kirinyaga 5 9 0.01 0.96 0 1 0 1,478 457,105 Kisii 7 13 0.35 0.58 0 1 0 649 491,786 Kisumu 20 31 0.21 0.66 0 1 0 919 504,359 Kitui 3 7 0.22 0.78 0 0 1 20,402 515,422 Koibatek 0 1 0.40 0.57 0 0 1 2,306 138,163 Kuria 1 1 0.93 0.05 0 1 0 581 151,887 Kwale 4 8 1.00 0.00 0 1 0 8,295 496,133 Laikipia 7 13 0.54 0.26 0 1 0 9,229 322,187 Lamu 2 3 1.00 0.00 0 1 0 6,167 72,686 Lugari 0 0 0.50 0.35 0 1 0 670 215,920 Machakos 9 16 0.35 0.53 0 1 0 6,281 906,644 Makueni 4 7 0.29 0.71 0 0 1 7,966 771,545 Malindi 4 11 1.00 0.00 0 1 0 7,751 281,552 Mandera 0 2 0.00 1.00 0 0 1 26,744 250,372 Maragua 0 0 0.03 0.83 0 1 0 868 387,969 Marakwet 1 1 0.63 0.28 0 0 1 1,588 140,629 Marsabit 1 2 0.85 0.13 0 0 1 61,296 121,478 Mbeere 0 2 0.13 0.85 0 0 1 2,093 170,953 Meru 19 31 0.13 0.83 0 1 0 8,017 1,300,000 Migori 2 5 0.59 0.39 0 1 0 2,005 514,897 Mombasa 60 83 0.90 0.00 1 0 0 230 665,018 Moyale 1 3 0.73 0.27 0 0 1 9,390 53,479 Mt.Elgon 1 0 0.93 0.03 0 1 0 944 135,033 Murang'a 12 16 0.07 0.86 0 1 0 930 348,304 Mwingi 1 3 0.26 0.74 0 0 1 10,030 303,828 Nairobi 230 388 0.58 0.01 1 0 0 696 2,100,000 Nakuru 26 40 0.58 0.23 0 1 0 7,242 1,200,000 Nandi 4 7 0.83 0.16 0 1 0 2,899 578,751 Narok 4 6 0.90 0.04 0 0 1 15,098 365,750 Nyandarua 4 6 0.00 0.93 0 1 0 3,304 479,902 Nyando 0 0 0.10 0.88 0 1 0 1,168 299,930 Nyeri 17 20 0.06 0.89 0 1 0 3,356 661,156 Rachuonyo 0 0 0.06 0.90 0 1 0 945 307,126 Siaya 2 4 0.11 0.77 0 1 0 1,520 480,184 Suba 0 0 0.33 0.65 0 1 0 1,055 155,666 Taita Taveta 3 11 0.95 0.00 0 1 0 17,128 246,671 Tana River 2 1 0.98 0.00 0 0 1 38,446 180,901 Teso 0 1 1.00 0.00 0 1 0 559 181,491 Tharaka 0 0 0.03 0.97 0 0 1 1,570 100,992 Thika 12 22 0.18 0.69 0 1 0 1,960 645,713 Trans-Nzoia 9 15 0.96 0.01 0 1 0 2,487 575,662 Transmara 1 0 0.73 0.25 0 0 1 2,846 170,591 Turkana 2 3 0.94 0.06 0 0 1 68,388 450,860 Uasin Gishu 16 23 0.78 0.12 0 1 0 3,328 622,705 Vihiga 0 4 0.57 0.30 0 1 0 563 498,883 Wajir 2 4 0.02 0.98 0 0 1 55,501 319,261 West Pokot 1 1 0.66 0.34 0 0 1 9,064 308,086 26 Table 3 Descriptive Statistics, Household Level This table reports descriptive statistics at the household level, based on the 2006 and 2009 FinAccess household surveys. 2006 2009 Variable Obs Mean Std. Dev. Min Max Obs Mean Std. Dev. Min Max Bank account 4360 0.1420 0.3491 0 1 6562 0.227 0.419 0 1 Bank Loan 4360 0.0290 0.1680 0 1 6562 0.044 0.204 0 1 Household size 4360 4.4722 2.3382 1 16 6562 4.959 2.600 1 20 Education: Primary 4360 0.3172 0.4654 0 1 6562 0.186 0.389 0 1 Education: Secondary 4360 0.2289 0.4202 0 1 6562 0.124 0.329 0 1 Education: Tertiary 4360 0.0234 0.1512 0 1 6562 0.007 0.082 0 1 Female 4360 0.5585 0.4966 0 1 6562 0.419 0.493 0 1 Asset score 4360 2.2011 2.1053 0 15 6562 2.723 2.761 0 18 Own permanent house 4360 0.1298 0.3361 0 1 6562 0.186 0.389 0 1 Income type: Agriculture 4360 0.4165 0.4930 0 1 6562 0.417 0.493 0 1 Income type: Waged 4360 0.1729 0.3782 0 1 6562 0.152 0.359 0 1 Income type: Business 4360 0.2865 0.4522 0 1 6562 0.286 0.452 0 1 Age group: 18-24 4360 0.2032 0.4024 0 1 6562 0.178 0.383 0 1 Age group: 25-34 4360 0.2837 0.4509 0 1 6562 0.258 0.438 0 1 Age group: 35-44 4360 0.1982 0.3987 0 1 6562 0.208 0.406 0 1 Age group: 45-54 4360 0.1232 0.3287 0 1 6562 0.134 0.340 0 1 Age group: 55+ 4360 0.1459 0.3530 0 1 6562 0.183 0.387 0 1 Language: Swahili 4360 0.4764 0.4995 0 1 6562 0.647 0.478 0 1 Language: Minority language 4360 0.4266 0.4946 0 1 6562 0.231 0.422 0 1 Urban 4360 0.1135 0.3173 0 1 6562 0.158 0.365 0 1 Rural 4360 0.6564 0.4750 0 1 6562 0.649 0.477 0 1 Arid and Semi-Arid 4360 0.2300 0.4209 0 1 6562 0.193 0.395 0 1 Total bank branches 4360 29.3475 67.4684 0 230 6562 61.189 124.190 0 388 Bank branches (excluding Equity Bank) 4360 27.7094 64.6335 0 220 6562 56.249 116.231 0 362 Equity ban branches 4360 1.6381 2.9712 0 10 6562 4.939 8.093 0 26 Equity bank presence 4360 0.477 0.50 0 1 6562 0.873 0.333 0 1 27 Table 4 Comparing Banks’ Expansion Strategies This table reports OLS estimates on the determinants of bank branching by districts. For Equity Bank, the dependent variable is the number of bank branches in a given district; for other bank groups, the dependent variable is the number of branches at district level over the number of banks in each group. Robust standard errors are in parentheses. Standard errors are clustered by district. ***, **, * indicate significance at the 1%, 5% and 10% levels, respectively. (1) (2) (3) (4) (5) (6) (7) (8) Dependent Variable: Number of branches Foreign Govern. Domestic Equity Foreign Govern. Domestic Equity over number of banks Rural -4.032** -4.368*** -2.958** -6.440 -4.032** -4.367*** -2.958** -6.440 (1.677) (1.470) (1.360) (3.898) (1.684) (1.476) (1.365) (3.913) Arid and Semi-Arid -4.146** -4.598*** -3.051** -7.276* -4.146** -4.598*** -3.051** -7.276* (1.681) (1.478) (1.362) (3.929) (1.688) (1.484) (1.368) (3.945) Minority language -5.801** -6.808*** -4.588** -17.880*** -5.819** -6.842*** -4.601** -18.302*** (2.691) (2.474) (2.192) (6.334) (2.692) (2.465) (2.192) (6.311) Swahili -5.902** -6.953*** -4.680** -18.666*** -5.983** -7.102*** -4.751** -19.229*** (2.741) (2.512) (2.232) (6.452) (2.783) (2.574) (2.267) (6.618) log(Density population) -0.080 -0.102* -0.063 -0.311** -0.080 -0.102* -0.063 -0.311** (0.052) (0.057) (0.043) (0.134) (0.052) (0.058) (0.043) (0.135) Dummy(2009) 0.118** 0.213*** 0.102** 1.045*** (0.052) (0.072) (0.045) (0.282) Minority language x Dummy(2009) 0.038 0.071 0.027 0.846*** (0.038) (0.062) (0.031) (0.295) Swahili x Dummy(2009) 0.164* 0.305** 0.144* 1.129** (0.093) (0.137) (0.080) (0.510) Constant 10.054*** 11.720*** 7.711*** 25.869*** 10.112*** 11.824*** 7.761*** 26.390*** (2.858) (2.595) (2.329) (6.585) (2.893) (2.632) (2.358) (6.722) Observations 129 129 129 129 129 129 129 129 Adjusted R-squared 0.750 0.747 0.717 0.524 0.748 0.744 0.714 0.512 Table 5 Access to Banking Services and Bank Presence: Baseline Models This table reports estimates from a Probit model of the probability of a household having a bank account. The data consists of 4,360 individuals interviewed in 2006 and 6,562 individuals in 2009. We include district fixed effects in all the models. Robust standard errors are in parentheses. Standard errors are clustered by district- year groups. ***, **, * indicate significance at the 1%, 5% and 10% levels, respectively. Dep. variable: Bank account (1) (2) (3) (4) (5) Household size -0.0060*** -0.0057*** -0.0058*** -0.0056*** -0.0060*** (0.002) (0.002) (0.002) (0.002) (0.002) Education: Primary 0.0006 -0.0089 -0.0078 -0.0030 0.0017 (0.012) (0.013) (0.014) (0.012) (0.013) Education: Secondary 0.1181*** 0.0964*** 0.0981*** 0.1123*** 0.1193*** (0.022) (0.019) (0.020) (0.021) (0.022) Education: Tertiary 0.2886*** 0.2144*** 0.2168*** 0.2762*** 0.2895*** (0.049) (0.061) (0.062) (0.049) (0.050) Female -0.0572*** -0.0564*** -0.0564*** -0.0576*** -0.0569*** (0.007) (0.007) (0.007) (0.007) (0.007) Asset score 0.0283*** 0.0295*** 0.0294*** 0.0286*** 0.0282*** (0.005) (0.004) (0.004) (0.005) (0.005) Own permanent house 0.0531*** 0.0498*** 0.0499*** 0.0533*** 0.0520*** (0.016) (0.016) (0.016) (0.016) (0.016) Income type: Agriculture 0.0330** 0.0260* 0.0263* 0.0321** 0.0322** (0.014) (0.014) (0.014) (0.014) (0.014) Income type: Waged 0.2152*** 0.2032*** 0.2043*** 0.2124*** 0.2148*** (0.025) (0.027) (0.027) (0.025) (0.025) Income type: Business 0.1512*** 0.1438*** 0.1440*** 0.1508*** 0.1505*** (0.019) (0.017) (0.017) (0.019) (0.019) Age group: 18-24 0.1448** 0.1652** 0.1636** 0.1480** 0.1430** (0.066) (0.064) (0.064) (0.066) (0.065) Age group: 25-34 0.3200*** 0.3344*** 0.3330*** 0.3235*** 0.3190*** (0.072) (0.072) (0.072) (0.072) (0.071) Age group: 35-44 0.4052*** 0.4175*** 0.4161*** 0.4091*** 0.4042*** (0.082) (0.083) (0.082) (0.082) (0.082) Age group: 45-54 0.4227*** 0.4341*** 0.4327*** 0.4259*** 0.4220*** (0.082) (0.082) (0.082) (0.082) (0.082) Age group: 55+ 0.4056*** 0.4135*** 0.4122*** 0.4083*** 0.4047*** (0.083) (0.085) (0.085) (0.083) (0.083) Swahili -0.1085*** -0.1051*** -0.1057*** -0.1080*** -0.1084*** (0.014) (0.015) (0.015) (0.014) (0.014) Minority language -0.1345*** -0.1350*** -0.1340*** -0.1387*** -0.1341*** (0.011) (0.013) (0.013) (0.011) (0.011) Bank branches 0.0010*** (0.000) Equity Bank presence 0.0424*** 0.0222** 0.0555*** (0.011) (0.009) (0.011) Foreign bank presence 0.0255 0.0273 (0.018) (0.018) Gov. and gov.-influenced banks presence 0.0298** 0.0287* (0.015) (0.015) Domestic private banks presence 0.0392** 0.0216 (0.016) (0.017) Foreign bank branches 0.0044 0.0103*** (0.003) (0.002) Gov. and gov.-influenced bank branches 0.0019 -0.0019 (0.003) (0.004) Domestic private bank branches -0.0028 -0.0072*** (0.003) (0.002) Observations 10,922 10,922 10,922 10,922 10,922 Pseudo R-squared 0.332 0.326 0.327 0.331 0.333 District Fixed Effects YES YES YES YES YES 29 Table 6 Access to Banking and Bank Presence: Comparing 2009 and 2006 This table reports estimates from a Probit model of the probability of a household having a bank account. The data consists of 4,360 individuals interviewed in 2006 and 6,562 individuals interviewed in 2009. We include district and year fixed effects in all the models; we also include but do not report coefficients on individual/household controls. Robust standard errors are in parentheses. Standard errors are clustered by district-year groups. ***, **, * indicate significance at the 1%, 5% and 10% levels, respectively. Dep. variable: Bank account (1) (2) (3) (4) Dummy(2009) 0.0642*** 0.0121 0.0169 0.0145 (0.013) (0.013) (0.019) (0.013) Bank branches 0.0067*** (0.002) Equity Bank presence -0.0129 -0.0472** 0.0006 (0.015) (0.021) (0.015) Foreign bank presence 0.0185 (0.020) Gov. and gov.-influenced banks presence 0.0126 (0.018) Domestic private banks presence 0.0175 (0.026) Foreign bank branches 0.0183** (0.008) Gov. and gov.-influenced bank branches 0.0015 (0.005) Domestic private bank branches -0.0074* (0.004) Interaction Effects Bank branches x Dummy(2009) -0.0024*** (0.001) Equity Bank presence x Dummy(2009) 0.0528*** 0.0700** 0.0646*** (0.016) (0.035) (0.017) Foreign bank presence x Dummy(2009) -0.0165 (0.022) Gov. and gov.-influenced banks presence x Dummy(2009) -0.0180 (0.025) Domestic private banks presence x Dummy(2009) 0.0277 (0.041) Foreign bank branches x Dummy(2009) -0.0060 (0.006) Gov. and gov.-influenced bank branches x Dummy(2009) -0.0059 (0.004) Domestic private bank branches x Dummy(2009) 0.0029 (0.003) Observations 10,922 10,922 10,922 10,922 Pseudo R-squared 0.332 0.335 0.334 0.336 Individual / Household Controls YES YES YES YES District Fixed Effects YES YES YES YES Time Fixed Effects YES YES YES YES 30 Table 7 Access to Banking and Bank Presence: Instrumental Variables This table reports estimates from IV GMM and IV Probit models of the probability of a household having a bank account. The data consists of 4,360 individuals interviewed in 2006 and 6,562 individuals interviewed in 2009. Equity Bank presence is instrumented with the district–level proportion of people speaking a minority language and its interaction with a 2009 dummy. Robust standard errors are in parentheses. Standard errors are clustered by district-year groups. ***, **, * indicate significance at the 1%, 5% and 10% levels, respectively. IV GMM IV Probit Marginal effects Second stage First stage Second stage Second stage First stage Dep. variable: Bank account (1) (2) (3) (4) (5) Household size -0.0081*** 0.0089*** -0.0068*** -0.0359*** 0.0089*** (0.0014) (0.0009) (0.0015) (0.0078) (0.0009) Education: Primary -0.0239*** -0.0112* 0.0068 0.0352 -0.0112* (0.0080) (0.0061) (0.0100) (0.0512) (0.0061) Education: Secondary 0.1019*** -0.0247*** 0.1284*** 0.5465*** -0.0247*** (0.0120) (0.0069) (0.0139) (0.0502) (0.0069) Education: Tertiary 0.2298*** -0.0254* 0.3051*** 1.0115*** -0.0253* (0.0320) (0.0152) (0.0622) (0.1601) (0.0151) Female -0.0473*** -0.0158*** -0.0558*** -0.2938*** -0.0158*** (0.0065) (0.0048) (0.0070) (0.0380) (0.0048) Asset score 0.0386*** -0.0013* 0.0281*** 0.1469*** -0.0013* (0.0021) (0.0008) (0.0019) (0.0095) (0.0008) Own permanent house 0.0573*** -0.0181*** 0.0534*** 0.2527*** -0.0181*** (0.0106) (0.0059) (0.0106) (0.0458) (0.0059) Income type: Agriculture 0.0314*** -0.0107 0.0333** 0.1712** -0.0107 (0.0091) (0.0076) (0.0144) (0.0729) (0.0075) Income type: Waged 0.1847*** -0.0218*** 0.2200*** 0.8487*** -0.0218*** (0.0132) (0.0082) (0.0234) (0.0736) (0.0082) Income type: Business 0.1268*** -0.0048 0.1523*** 0.6734*** -0.0047 (0.0104) (0.0075) (0.0180) (0.0697) (0.0074) Age group: 18-24 -0.0038 0.0339*** 0.1389** 0.5900*** 0.0339*** (0.0120) (0.0119) (0.0541) (0.1963) (0.0119) Age group: 25-34 0.1022*** 0.0335*** 0.3134*** 1.2218*** 0.0335*** (0.0123) (0.0120) (0.0591) (0.1955) (0.0119) Age group: 35-44 0.1349*** 0.0266** 0.3986*** 1.4130*** 0.0266** (0.0129) (0.0122) (0.0657) (0.1965) (0.0122) Age group: 45-54 0.1273*** 0.0271** 0.4167*** 1.3884*** 0.0271** (0.0142) (0.0129) (0.0714) (0.1987) (0.0129) Age group: 55+ 0.1125*** 0.0254** 0.3990*** 1.3768*** 0.0254** (0.0135) (0.0129) (0.0693) (0.1998) (0.0129) Swahili -0.1737*** -0.0058 -0.1095*** -0.5412*** -0.0058 (0.0159) (0.0067) (0.0119) (0.0550) (0.0067) Minority language -0.2130*** -0.0102 -0.1280*** -0.7956*** -0.0103 (0.0170) (0.0085) (0.0094) (0.0676) (0.0085) Bank branches 0.0012*** -0.0000*** 0.0010*** 0.0052*** -0.0000*** (0.0002) (0.0000) (0.0001) (0.0007) (0.0000) Equity Bank presence 0.0939*** 0.0984*** 0.5989*** (0.0197) (0.0187) (0.1307) Minority language district -0.9086*** -0.9069*** (0.0224) (0.0223) Minority language district x Dummy(2009) 0.2313*** 0.2333*** (0.0187) (0.0186) Observations 10922 10923 10922 10922 10922 Centered R-squared 0.7585 Uncentered R-squared 0.9311 Probability of positive outcome (predict, p) 0.1128 Partial R-squared of excluded instruments: 0.2822 p-value of test of excluded instruments 0.0000 p-value of Hansen J statistic 0.1924 District Fixed Effects YES YES YES YES YES 31 Table 8 Access to Banking and Bank Presence: More Interaction Effects This table reports estimates from a Probit model of the probability of a household having a bank account. The data consists of 4,360 individuals interviewed in 2006 and 6,562 individuals interviewed in 2009. We include district fixed effects in all the models; we also include but do not report coefficients on individual/household controls. Robust standard errors are in parentheses. Standard errors are clustered by district-year groups. ***, **, * indicate significance at the 1%, 5% and 10% levels, respectively. Dep. variable: Bank account (1) (2) (3) (4) (5) Equity Bank presence 0.0708*** 0.0528*** 0.0578*** 0.0521*** 0.0911*** (0.0157) (0.0115) (0.0114) (0.0115) (0.0167) Asset score x Equity Bank presence -0.0153** -0.0119* (0.0066) (0.0067) Permanent own house x Equity Bank presence -0.0506*** -0.0431** (0.0191) (0.0193) Secondary-Tertiary Education x Equity Bank presence -0.0515*** -0.0414** (0.0156) (0.0163) Salaried job x Equity Bank presence -0.0416** -0.0372** (0.0168) (0.0175) Observations 10922 10922 10922 10922 10922 Pseudo R-squared 0.333 0.333 0.333 0.333 0.335 Individual / Household Controls YES YES YES YES YES District Fixed Effects YES YES YES YES YES 32 Table 9 Access to Credit and Bank Expansion This table reports estimates from an Ordered Probit model of the probability of a household’s access to bank services. Bank services take the value 0 if the individual has neither a bank account nor a bank loan; it takes the value 1 if the individual has a bank account; and it takes the value 2 if the individual has a bank loan. The data consists of 4,360 individuals interviewed in 2006 and 6,562 individuals interviewed in 2009. We include district fixed effects in all the models; we also include but do not report coefficients on individual/household controls. Robust standard errors are in parentheses. ***, **, * indicate significance at the 1%, 5% and 10% levels, respectively. (1) (2) (3) (4) Ordered Probit Average Marginal Effects Financial Financial Financial Access=0 Access=1 Access=2 Bank branches 0.0039*** -0.0007*** 0.0005*** 0.0002*** (0.0006) (0.0001) (0.0000) (0.0000) Equity Bank presence 0.2637*** -0.0496*** 0.0360*** 0.0135*** (0.0683) (0.0128) (0.0093) (0.0035) Observations 10922 Pseudo R-squared 0.2918 Individual / Household Controls YES District Fixed Effects YES 33 Table 10 Access to Banking and Bank Presence: Rural and Arid and Semi-Arid Districts This table reports estimates from a Probit model of the probability of a household having a bank account after dropping urban districts. We include district and year fixed effects in all the models; we also include but do not report individual/household controls. Robust standard errors are in parentheses. Standard errors are clustered by district-year groups. ***, **, * indicate significance at the 1%, 5% and 10% levels, respectively. Dep. variable: Bank account (1) (2) (3) (4) (5) Bank branches 0.0042*** (0.0014) Equity Bank presence 0.0247*** 0.0207*** 0.0277*** (0.0091) (0.0072) (0.0102) Foreign bank presence 0.0200 0.0184 (0.0130) (0.0131) Gov. and gov.-influenced bank presence 0.0289*** 0.0221** (0.0096) (0.0103) Domestic private bank presence 0.0369** 0.0278* (0.0144) (0.0144) Foreign bank branches 0.0132*** 0.0132*** (0.0028) (0.0026) Gov. and gov.-influenced bank branches 0.0025 -0.0004 (0.0024) (0.0029) Domestic private bank branches -0.0046 -0.0043 (0.0039) (0.0041) Observations 9388 9388 9388 9388 9388 Pseudo R-squared 0.316 0.315 0.316 0.317 0.317 Individual / Household Controls YES YES YES YES YES District Fixed Effects YES YES YES YES YES 34 Table 11 Access to Banking and Bank Presence: Difference-in-Differences Estimator This table reports estimates from a Probit model of the probability of a household having a bank account; only the districts without Equity Bank presence in 2006 (and with or without Equity Bank presence in 2009) are included in the regressions. We include district and year fixed effects in all the models; we also include but do not report individual/household controls. Robust standard errors are in parentheses. Standard errors are clustered by district-year groups. ***, **, * indicate significance at the 1%, 5% and 10% levels, respectively. Dep. variable: Bank account (1) (2) (3) Bank branches -0.0021 (0.0019) Equity Bank presence 0.0293** 0.0198** 0.0280** (0.0118) (0.0092) (0.0119) Foreign bank presence 0.0038 (0.0109) Gov. and gov.-influenced banks presence 0.0181** (0.0079) Domestic Private banks presence 0.0095 (0.0146) Foreign bank branches -0.0026 (0.0050) Gov. and gov.-influenced bank branches -0.0007 (0.0028) Domestic private bank branches -0.0034 (0.0067) Observations 5314 5314 5314 Pseudo R-squared 0.348 0.348 0.348 Individual / Household Controls YES YES YES District Fixed Effects YES YES YES Year Fixed Effects YES YES YES 35 Number of Customer Customer Deposits (In Millions) (In KShs Millions) 120000 6 100000 5 80000 4 3 60000 2 40000 1 20000 0 0 2006 2007 2008 2009 2010 2006 2007 2008 2009 2010 Gross Loan Portfolio Total Assets (In KShs Millions) (In KShs Millions) 150000 80000 120000 60000 90000 40000 60000 20000 30000 0 0 2006 2007 2008 2009 2010 2006 2007 2008 2009 2010 Profits Before Taxes Shareholders' Funds (In KShs Millions) (In KShs Millions) 10000 30000 8000 25000 20000 6000 15000 4000 10000 2000 5000 0 0 2006 2007 2008 2009 2010 2006 2007 2008 2009 2010 Figure 1. Equity Bank’s performance indicators. This figure shows the evolution of the number of customers, deposits, gross loan portfolio, total assets, profits before taxes and shareholders’ funds over the period 2006-2010. The source is Allen et al. (2012) based on data from the Equity Bank’s Annual Reports and Financial Statements 2010. 36 Urban Districts Rural, Arid and Semi-Arid Districts 200 200 2006 2009 2006 2009 150 150 100 100 50 50 0 0 Foreign banks Gov.-influenced banks Local private banks Foreign banks Gov.-influenced banks Local private banks Low Minority Language Districts High Minority Language Districts 300 300 2006 2009 2006 2009 250 250 200 200 150 150 100 100 50 50 0 0 Foreign banks Gov.-influenced banks Local private banks Foreign banks Gov.-influenced banks Local private banks Highly Populated Districts Low Populated Districts 300 300 2006 2009 2006 2009 250 250 200 200 150 150 100 100 50 50 0 0 Foreign banks Gov.-influenced banks Local private banks Foreign banks Gov.-influenced banks Local private banks Highly Educated Districts Low Educated Districts 350 350 2006 2009 2006 2009 300 300 250 250 200 200 150 150 100 100 50 50 0 0 Foreign banks Gov.-influenced banks Local private banks Foreign banks Gov.-influenced banks Local private banks Figure 2. Bank Penetration in Kenya. This Figure shows the number of bank branches according to different ownership structures: Foreign banks, Government banks, and Domestic private banks. Urban districts include Nairobi and Mombasa districts. Low minority language districts are the ones where the share of the population speaking a minority language is smaller than the median. Highly populated districts are the ones where the density population is higher than the median. Highly educated districts are the ones where the share of the population with secondary or tertiary education is higher than the median. 37