Guatemala: Selected Issues and Analytical Notes

Guatemala: Selected Issues and Analytical Notes

Abstract

Guatemala: Selected Issues and Analytical Notes

Macro Financial Linkages: Financial Development and Inclusion

A. Financial Development in Guatemala1

This section examines the current state of financial development in Guatemala, as well as the implications for potential growth and stability from further financial deepening. Guatemala lags its regional peers on financial development, in particular, on markets. However, it overperforms the level of development consistent with its macroeconomic fundamentals. Hence, while at the moment there are no indications of a significant risk build up, further financial development should proceed with caution. In the longer term, as fundamentals continue to evolve, Guatemala could reap further benefits from financial development in terms of growth and stability, provided there is adequate regulatory oversight to prevent excesses. However, care should be taken in not promoting excessive market development when financial institutions are underdeveloped. Laying out strong legal and institutional foundations will be important for supporting healthy financial deepening.

Financial Development: Where Does Guatemala Stand?

1. Guatemala’s financial development was assessed using a comprehensive index. Financial development has proven difficult to measure. Typical proxies in the literature such as the ratio of private credit to GDP and, to a lesser extent, stock market capitalization are too narrow to capture the broad spectrum of financial sector activities. To better capture different facets of financial development, we employ a comprehensive and broad-based index covering 123 countries for the period 1995–2013 (see Appendix 1 and Heng et al (2015) for details). The index contains two major components: financial institutions and financial markets. Each component is broken down into access, depth, and efficiency sub-components. These sub-components, in turn, are constructed based on a number of underlying variables that track development in each area.

2. Guatemala’s financial development has improved only marginally over the past decade, and remains low, compared to the regional peers. The small improvements came from growth in financial institutions, in particular, better institutional access and improved efficiency. In contrast, improvements in market development during the 90s were reversed in the 2000s. Overall, Guatemala continues to lag behind its regional and emerging market peers on many dimensions. In particular, it lags other EM groups on all of the subcomponents of financial market development. It is also behind other EMs on some aspects of institutional development, though performance varies by component. In fact, Guatemala compares favorably on institutional access, outperforming all other EM country groupings. Good access reflects a wide network of ATMs and bank branches per 100,000 adults. However, the distribution of access points is not uniform (see Section B below). On the other hand, the country lags behind other EMs on institutional efficiency, reflecting moderately high interest rate spreads, high overhead costs, and high net interest margins. Finally, Guatemala is behind all other country groupings on institutional depth due to the low level of private sector credit to GDP as well as small mutual fund and insurance industries.

Figure 1.
Figure 1.

Financial Sector Development

Citation: IMF Staff Country Reports 2016, 282; 10.5089/9781475531534.002.A005

3. Nevertheless, Guatemala’s financial development is above the level predicted by country’s fundamentals (Figure 2). A simple cross-country comparison above does not account for differences in the underlying macroeconomic conditions. Financial development gaps—the deviation of the financial development index from a prediction based on economic fundamentals, such as income per capita, government size, and macroeconomic stability—can help identify potential under or overdevelopment of Guatemala, compared to countries with similar fundamentals. These gaps suggest that Guatemala’s financial development is above the level predicted by its macroeconomic fundamentals on all but two subcomponents. The exceptions are one narrow measure of institutional efficiency, namely, 3-bank asset concentration and public debt securities to GDP. While positive gaps could be an indication of inefficiencies and financial stability risks, this analysis is not normative and other measures of financial stability have to be employed to thouroughly analyze the situation. In fact, the analysis of financial stability risks suggests that there are no indications of a significant risk build up at this point (see AN on Macro Financial Linkages: Assessing Financial Risks, Section A) but the supervisor should remain vigilant.

Figure 2.
Figure 2.

Financial Development Gaps

Citation: IMF Staff Country Reports 2016, 282; 10.5089/9781475531534.002.A005

The Potential for Raising Growth and Stability through Further Financial Deepening in Guatemala

4. There is a non-linear relationship between growth and stability on the one hand and financial development on the other hand. Financial development gaps do not address the question of the optimal level of financial development in terms of growth and stability. To explore this question, we examine the relationships between financial development and growth as well as financial development and stability (see Heng et al., 2015). We find that these relationships are nonlinear. In other words, the benefits from financial development are rising at the early stages of development as resources are increasingly channeled into productive uses. However, there is a turning point beyond which the positive growth benefits diminish. Similarly, at the early stages, financial development can help reduce instability, for example, by providing insurance services, but these benefits also start to diminish after a certain point. The turning points likely reflect the fact that large financial systems can eventually divert resources from more productive activities, while excessive borrowing and risk-taking by financial institutions can lead to increased instability and lower long-term growth. Indeed, the inverted U-shaped relationship with growth is driven by the depth of financial institutions, or a measure of size. Access and efficiency, on the other hand, yield unambiguously increasing benefits to growth, although with potential stability costs as reduced bank profitability may encourage risk-taking. Lastly, too much market development at the early stages of institutional development may have negative implications for stability. One reason for this could be increased market volatility, which may more easily set in when financial institutions are not strong enough to help guard against shocks. For similar levels of development, however, institutions and markets are complementary for growth and stability.

5. Guatemala has not yet reached the levels of institutional and market development that yield maximum benefits to growth and stability. In Latin America and the Caribbean, Brazil and Chile are closest to reaping the maximum benefits (Figure 3). Guatemala, in contrast, is still far away from reaping the maximum benefits to growth and stability, in particular, in terms of financial market development. Note that these estimates stem from a partial analysis that assumes that all other growth determinants (such as income level, inflation, government size etc.) are held constant while financial development is consistent with the level of macroeconomic fundamentals. Thus, in the longer term, reaping maximum benefits from financial development for growth and stability would also require adjusting Guatemala’s macroeconomic fundamentals, which in turn would support further development of the financial systems. This is an interactive process whereby financial systems are shaped by fundamentals, and fundamentals evolve partly as a function of more developed financial systems. Estimates should, however, be interpreted with caution since it is difficult to disentangle causality in econometric terms, even though instrumental variables were used to address potential endogeneity issues.2

Figure 3.
Figure 3.

Financial Institutions and Market Development, and Economic Growth

Citation: IMF Staff Country Reports 2016, 282; 10.5089/9781475531534.002.A005

Source: IMF staff calculations.Note: surface shows the predicted effect in growth for each level of the indices, holding fixed other sets of controls.

Policy Recommendations

Short-Term:

  • Remain vigilant about financial excesses since the level of development is above that predicted by the macroeconomic fundamentals though there are no indications of a significant risk buildup at the moment

  • To stimulate the development of the secondary bonds market it will be important to dematerialize government securities market and standardize government securities. Adopting Securities Market Law and Public Debt Law are also key priorities.

Long-Term:

  • Continue developing regulation and supervision that are consistent with the existing level of financial development and embed enough flexibility to address future challenges of financial deepening.

  • Given that sequencing of reforms could be important, care should be taken in not promoting excessive market development when financial institutions are underdeveloped, since this could jeopardize macroeconomic and financial stability.

  • More generally, preparing adequate legal and institutional ground for a well-functioning financial system will be key. This includes developing strong property and ownership rights, including strengthening protection of minority shareholders, improving efficiency of the legal system, reducing corruption and improving corporate governance, as well as continuing to develop provision of adequate financial information.

B. Financial Inclusion in Guatemala3

This section examines the current status of financial inclusion in Guatemala, identifies the remaining gaps, and analyzes the impact of removing impediments to financial inclusion on growth and inequality. Guatemala has excelled in providing physical access to financial infrastructure but lags behind peers on creating an enabling regulatory environment for financial inclusion and on the use of financial services by households and firms. While low use to a large extent reflects Guatemala’s current state of economic development, strengthening regulatory environment and reducing entry costs—with the latter helping stimulate growth and reduce inequality—are matters of high priority. In the longer run, creating better conditions for income growth, improving education, particularly, of women, reducing the size of the informal economy and strengthening the rule of law will help raise financial inclusion further.

Introduction

6. Guatemala can benefit from further financial inclusion. Financial inclusion can help boost economic growth and reduce poverty and inequality by mobilizing savings and providing households and firms with greater access to resources needed to finance consumption and investment and to insure against shocks. Financial inclusion can also foster formalization of the economy, helping, in turn, to boost government revenues and strengthen social safety nets. Given a relatively low level of income per capita, high inequality, low savings and investment as well as high labor informality, the benefits from further financial inclusion can be pronounced in Guatemala.

7. The note takes three separate approaches for examining different faucets of financial inclusion and its impediments in Guatemala.4 First, an empirical approach focuses on measuring financial inclusion, identifying financial inclusion gaps, and their underlying drivers. It is based on composite measures of household and firm financial inclusion as well as a measure physical access to financial infrastructure using recently updated FINDEX dataset (World Bank), Enterprise Survey (World Bank), and Financial Access Survey (IMF). These measures help place Guatemala in a temporal and cross-country perspective. Second, a regression analysis is employed to identify characteristics of the individuals who use financial services (i.e. those financially-included). Third, a novel theoretical framework is employed to identify the most binding financial sector frictions that impede financial inclusion in Guatemala. This framework allows examining the implications of alleviating financial frictions on inequality and growth.

Empirical Approach I: Cross-Country Data

A. Where Does Guatemala Stand on Financial Inclusion Compared to Peers?

8. Guatemala has excelled on physical access to financial infrastructure though access points are concentrated around Guatemala City (Figure 4). We measure physical access by the number of commercial bank branches and ATMs per 100,000 adults and per 1,000 square kilometers (see the Figure 4 and Appendix 2 for details). Guatemala has made substantial progress on improving physical access to financial infrastructure in the past decade with the number of commercial bank branches per 100,000 adults rising from 18.8 in 2004 to 37 in 2014. In fact, Guatemala now stands as one of the champions on physical access to financial infrastructure in Latin America and the Caribbean (LAC) with 37 branches and 36 ATMs per 100,000 adults, outpacing LAC averages of 24 for branches and 25 for ATMs. In addition, Guatemala has a network of bank correspondents, which are used more widely than elsewhere in LAC.5 The distribution of access points, however, is not uniform. Commercial bank branches are concentrated in the area surrounding Guatemala City while banking correspondents are located in remote areas.

9. Guatemala, however, lags behind peers on creating an enabling regulatory environment for financial inclusion. It scores below average of other emerging markets on the Global Microscope index, which assesses the regulatory environment for financial inclusion across 12 indicators and 55 countries. This is in contrast to many LAC countries, which score well on this index with Peru being the world champion. In 2015 Guatemala ranked relatively well on two Microscope indicators: (i) regulation and supervision of branches and agents and (ii) requirements for non-regulated lenders. It was on par with other LAC countries on average (though generally below emerging Asia) on (i) regulation and supervision of deposit-taking activities, (ii) regulation and supervision of credit portfolios, (ii) regulatory and supervisory capacity for financial inclusion, and (iii) regulation of insurance for low income population. However, Guatemala underperformed on other sub-components of the Global Microscope Index, including, (i) government support for financial inclusion, largely due to the absence of a formal comprehensive financial inclusion strategy, (ii) regulation of electronic payments, (iii) market conduct rules, (iv) grievance redress and operation of dispute resolution, (v) credit reporting systems, and (vi) prudential regulation. The authorities, however, are already working on addressing some of the weaknesses, including the recent adoption of the regulation of mobile financial services and the microfinance law as well as the preparation of the national financial inclusion strategy.

Figure 4.
Figure 4.

Physical Access to Financial Infrastructure and Enabling Environment

Citation: IMF Staff Country Reports 2016, 282; 10.5089/9781475531534.002.A005

10. The use of financial services by households and firms is low, compared to other emerging markets. We employ multi-dimensional indices to capture different facets of the use of financial services by households and SMEs. The diagram in Appendix II illustrates indicators included in each of the indices. While the use of financial services by households has improved in the past 4 years (Figure 5), Guatemala lags behind other emerging and LAC countries on this dimension. 6 The low score reflects low share of the population having an account at a formal financial institution, and using ATMs, debit and credit cards. Indeed, while the share of adult population having an account at a formal financial institution increased dramatically over the past few years – from 22 percent in 2011 to 41 percent in 2014 – it remains below that of other LAC countries on average (46.5 percent) and even more so of emerging Asia (60.35 percent). Guatemala, however, is at par with the regional peers on savings and borrowing from a formal financial institution. Nonetheless, informal finance remains important with the share of population using savings clubs in 2014 as large as that saving at a formal financial institution (12 percent versus 8 percent in LAC on average) and 20 percent of the population reporting borrowing from family and friends, compared to only 14 percent in LAC. Guatemala also does not compare favorably to its regional and emerging market peers on the use of financial services by firms. This reflects low account ownership by SMEs with only 60 percent of firms reporting having a checking or savings account, compared to a LAC average of 92 percent. Low share of firms using banks to finance investments and working capital and somewhat high collateral value, compared to emerging markets outside of LAC, also contribute to a lower firm score.

Figure 5.
Figure 5.
Figure 5.

Use of Financial Services by Households and Firms

Citation: IMF Staff Country Reports 2016, 282; 10.5089/9781475531534.002.A005

1/ Residuals from the regression among household financial inclusion index on non-interest income, bank net interet margin, 3 bank asset concentration, overhead costs to total asset, microscope and distance to default.2/ Residuals from the regression among firm usage of financial services index on non-interest income, bank net interest margin, 3 bank asset concentration, overhead costs to total asset, microscope and distance to default.

B. Where Does Guatemala Stand on Financial Inclusion Compared to Macroeconomic Fundamentals?

11. Financial inclusion gaps help account for the differences in the underlying macroeconomic conditions. We compute financial inclusion gaps with respect to own fundamentals as deviations of financial inclusion indices from the values predicted by the exogenous domestic factors such as income per capita, education, size of the shadow economy, the rule of law, the share of foreign-owned firms, and the importance of fuel exports. The calculated negative gaps could capture possible distortions or market frictions while positive gaps may reflect financial excesses or inefficiencies.7 We find that financial inclusion is higher in countries with the following characteristics (Table 1 and Dabla –Norris et al, 2015): higher income per capita (for households and firms), higher education (for households), stronger rule of law (for households), lower degree of informality (for households), lower prevalence of foreign-owned firms (for firms and access to financial institutions), lower fuel exports (for firms and access to financial institutions). In the longer run, as domestic fundamentals continue to evolve, there may be scope for further gains in financial inclusion. To identify such possibilities, we have constructed gaps with respect to an Asian benchmark—a recognized success story on financial inclusion in countries with relatively strong fundamentals.

12. Guatemala is broadly in line with its own fundamentals on financial inclusion of households and firms though it falls behind the world “frontier” Asian economies. Financial inclusion gaps in relation to domestic fundamentals are virtually zero for both households and firms though there are large negative gaps, compared to Asian emerging markets. This suggests that Guatemala’s relatively low level of financial inclusion, compared to peers can be explained by the constraints imposed by the current level of development, including its current relatively low income level, low education level, including financial education, the large size of the shadow economy, and the relatively weak rule of law. Looking at the gaps on the individual subcomponents, most of the household inclusion gaps are small positive with the exception of the gap on ATM use—an area where some improvement is possible even in the short run. On the firm side, the negative gaps on account holdings and the usage of banks to finance investment and working capital suggest additional areas for improvement. The large negative gaps with respect to Asian benchmarks indicate that in the longer run, as domestic fundamentals continue evolving, there will be scope for further financial inclusion gains.

13. An econometric examination of the factors behind financial inclusion gaps reveals the importance of strengthening regulatory environment in Guatemala. The results of a simple regression analysis (Table 2 and Dabla-Norris et al) suggest that higher (more positive/less negative) financial inclusion gaps with respect to domestic fundamentals are associated with lower non-interest income (for household and firms), lower bank safety buffers (for households), lower bank efficiency, as measured by the overhead costs (for firms) and stronger regulatory environment, as measured by the Global Microscope score (for firms). However, the direction of causality is not clear, in particular, in the case of bank safety buffers and efficiency, which could reflect consequences of inclusion instead of the underlying drivers. In the case of Guatemala, however, one unambiguous conclusion that can be drawn from this analysis is that strengthening regulatory environment for financial inclusion could help improve the inclusion of firms.

Empirical Approach II: Microdata

14. Econometric investigation using micro data confirms the importance of income and education levels, in particular, for women, for being financially included. We estimate a set of probit models that link various measures of financial inclusion with individual characteristics such as education, income level, age, and gender. The dependent variables are dummy variables that are equal to one if (1) an individual has an account in a financial institution; (2) an individual has a debit card; (3) an individual has a savings account; and (4) an individual has a credit card. We find that education and income level have strong positive relationship with all the dependent variables, meaning that more educated and higher income individuals are more likely to have a bank account, a debit or a credit card, or savings holding other variables constant. We also find that all the dependent variables except savings are positively associated with age suggesting that older people are more likely to have a bank account, debit or credit card but are less likely to save. Finally, the results indicate that women are generally less likely to use financial services though there is a mitigating impact of better education—the coefficient on an interaction term between a dummy variable equal to one for women and that equal to one for those with secondary education. Hence, while women generally are less financially included those who have better education have better access to finance.

Theoretical Approach

15. We apply a micro-founded structural model to shed light on the implications of relaxation of various constraints to financial inclusion for fostering growth and reducing inequality. Appendix 2 and Dabla-Norris et al. (2015) provide details of the model description. We group financial constraints into three broad categories:

  • Participation (entry) costs. These typically reflect high documentation requirements by banks for opening, maintaining, and closing accounts, and for loan applications that impede access to finance. These can also reflect various forms of barriers, including red tape and the need for informal guarantors as connections to access finance.

  • Borrowing constraints. The amount firms can borrow (the depth of credit) once they have access to banking systems is generally determined by collateral requirements, which depend on the state of creditors’ rights, information disclosure requirements, and contract enforcement procedures, among others.

  • Intermediation costs. High intermediation costs resulting from information asymmetries between banks and borrowers and limited competition in the banking system can lead to smaller and less capitalized borrowers being charged higher interest rates and fees.

The model’s key parameters are calibrated to match the moments of firm distribution from the Enterprise Survey Data, such as the percent of firms with credit and the firm employment distribution, as well as the economy-wide nonperforming loan ratio, and interest rate spread (Figure 6). We conduct policy experiments to identify the most binding constraints to financial inclusion and examine the macroeconomic effects of removing these constraints. Three illustrative simulation scenarios includes: (i) reducing the financial participation cost to 0, (ii) relaxing borrowing constraints in the form of collateral requirements to the world minimum (iii) increasing intermediation efficiency (i.e., reducing monitoring costs to 0 by equalizing spreads to the proportion of non-performing loans).8

Figure 6.
Figure 6.

Country-Specific Financial Constraints1

Citation: IMF Staff Country Reports 2016, 282; 10.5089/9781475531534.002.A005

Source: World Bank Enterprise Surveys.1/ Firms with access to credit represents the percent of firms with a bank loan/line of credit

16. In Guatemala high entry costs appear to be the most binding constraint. Compared to other LAC countries, Guatemala has moderate collateral levels and intermediation costs. Median collateral is 117 percent of the loan value versus 125 percent in LAC while lending-deposit spread of about 8 percentage points is close to the LAC average of 7 percentage points. Moderate levels of collateral and spreads reflect high concentration of credit among a small number of large clients who are well-known to the bank. However, Guatemala’s firms have low level of access to credit, which offsets the positive effects created by moderate borrowing and intermediation costs. Among firms, 49 percent have lines of credit with the bank compared to the LAC average of 55 percent while among the SMEs only 60 percent of firms have a bank account, compared to a LAC average of 92 percent.

17. Given the high level of inequality in Guatemala, reducing entry costs should be a matter of first priority. According to the model results (Figure 7), the loosening of any of the three constraints to financial inclusion will generate an improvement in growth but lowering the spreads and the collateral level would also worsen inequality. Intuitively, this is because, due to their already moderate levels, the loosening of these two constraints generates much larger marginal benefits for those at the top of the talent and wealth distribution. In this situation, very talented or very wealthy entrepreneurs can significantly increase their leverage and their production. In contrast, the loosening of the participation constraint (reduction in entry costs) will both raise growth and reduce inequality. Given the high level of income inequality in Guatemala (Gini coefficient of 52), policies should focus on loosening participation constraints in the first instance.

A05ufig2

Poverty and Inequality

(Percent of population below poverty line of $4 a day; Gini)

Citation: IMF Staff Country Reports 2016, 282; 10.5089/9781475531534.002.A005

Figure 7.
Figure 7.

Impact of Reducing Financial Constraints on GDP and Inequality

Citation: IMF Staff Country Reports 2016, 282; 10.5089/9781475531534.002.A005

Policy Recommendations

Short-term

  • Improve regulatory environment for financial inclusion, in particular, finalize and adopt the national financial inclusion strategy, improve regulation of electronic payments, market conduct rules, and credit reporting systems. The recently adopted mobile money regulation was an important step and continuing developing e-money will facilitate access in remote areas.

  • Reduce entry/participation costs such as documentation requirements/guarantees while safeguarding for financial stability. The introduction of the simplified accounts currently under consideration would be an important milestone in this regard.

Long-term

  • Create better conditions for income growth, improve education, particularly for women, and pay special attention to financial education, reduce the size of the informal economy. This will likely require raising fiscal spending, in particular, on education.

  • As demand-side barriers are addressed, regulators should examine financial institutions’ lending practices and credit concentration limits.

  • As entry/participation barriers are relaxed and previously unbanked businesses enter the financial system, credit bureau implementation should be improved in order to lower information costs and collateral requirements, especially for new clients. At the same time, increased competition in the banking sector should be promoted to improve efficiency and maintain low spreads.

Table 1.

Financial Inclusion and Fundamentals

article image
Robust standard errors in parentheses*** p<0.01, ** p<0.05, * p<0.1
Table 2.

Determinants of the Financial Inclusion Gaps

article image
Robust standard errors in parentheses*** p<0.01, ** p<0.05, * p<0.1
Table 3.

Characteristics of Financially-Included Population

article image
Standard errors in parentheses*** p<0.01, ** p<0.05, * p<0.1

References

  • Amidžić, G., A. Massara, and A. Mialou. 2014. “Assessing Countries’ Financial Inclusion Standing—A New Composite Index,IMF Working Paper WP/14/36.

    • Search Google Scholar
    • Export Citation
  • Arellano, M. and O. Bover. 1995. “Another Look at the Instrumental Variable Estimation of Error-components Models.Journal of Econometrics, Vol. 68, pp. 2951.

    • Search Google Scholar
    • Export Citation
  • Beck, T., A. Demirguc-Kunt, and R. Levine, 2007, “Finance, Inequality and the Poor,Journal of Economic Growth, No. 12 (1), pp. 2749.

    • Search Google Scholar
    • Export Citation
  • Blundell, R. and S. Bond. 1998. “Initial Conditions and Moment Restrictions in Dynamic Panel Data Models.Journal of Econometrics, Vol. 87, pp. 115143.

    • Search Google Scholar
    • Export Citation
  • Camara, N., and D. Tuesta, 2014, “Measuring Financial Inclusion: A Multidimensional Index,BBVA Working Paper (14/26), BBVA Research.

    • Search Google Scholar
    • Export Citation
  • Camara, D. Tuesta, and Urbiola (2015), “Extending access to the formal financial system: the banking correspondent business model,BBVA Working Paper (15/10), BBVA Research.

    • Search Google Scholar
    • Export Citation
  • Dabla-Norris, E., Y. Deng, A. Ivanova, I. Karpowicz, F. Unsal, E. VanLeemput, and J. Wong. 2015. “Financial Inclusion: Zooming in on Latin America,IMF Working Paper No. WP/15/206.

    • Search Google Scholar
    • Export Citation
  • Dyna Heng, Anna Ivanova, Rodrigo Mariscal, Uma Ramakrishnan, and Joyce Cheng Wong. 2015. “Advancing Financial Development in Latin America and the Caribbean,IMF Working Paper WP/16/81.

    • Search Google Scholar
    • Export Citation
  • Kaufmann, D., A. Kraay and M. Mastruzzi. 2010. “The Worldwide Governance Indicators: A Summary of Methodology, Data and Analytical Issues.World Bank Policy Research Working Paper, No. 5430.

    • Search Google Scholar
    • Export Citation
  • La Porta, R., F. Lopez-de-Silanes and A. Shleifer. 2008. “The Economic Consequences of Legal Origins.Journal of Economic Literature, Vol. 46, No. 2, pp. 258332.

    • Search Google Scholar
    • Export Citation
  • Sahay, R., M. Cihak, P. N’Diaye, A. Barajas, D. Ayala Pena, R. Bi; Y. Gao, A. Kyobe, L. Nguyen, C. Saborowski, K. Svirydzenka, and R. Yousefi. 2015a. “Rethinking Financial Deepening: Stability and Growth in Emerging Markets.IMF Staff Discussion Notes, No. 15/8.

    • Search Google Scholar
    • Export Citation
  • Townsend (1979) Townsend, R., 1979, “Optimal Contracts and Competitive Markets with Costly State Verification.Journal of Economic Theory 21, no. 2 (1979): 26593.

    • Search Google Scholar
    • Export Citation

Appendix 1. Measuring Financial Development

To measure financial development we employ the same framework as in Heng et al., 2015. 1

A05ufig3
Source: IMF staff calculations.1 Stock of debt by local firms is based on residency concept.

Sources and Data Processing

  • The data generally cover the period 1995 to 2013 with gaps, in particular, for countries in the Middle East, Sub-Sahara Africa and Latin America. For some variables, e.g., ATMs per thousands of adults, the data were only available starting in 2004. Our data came from numerous sources: World Bank’s World Development Indicators (WDI), FinStats, Non-Bank Financial Institutions database (NBFI), Global Financial Development database (GFD); International Monetary Fund’s International Financial Statistics (IFS); Bureau van Dijk, Bankscope; Dealogic’s debt capital markets statistics; World Federation of Exchanges (WFE); and Bank for International Settlements’ debt securities statistics.

  • After a gap filling process to generate a balanced panel, all variables were normalized using the following formula:
    Ix,it=xitmin(xit)max(xit)min(xit),

where Ix,it is the normalized variable x of country i on year t, min(xit) is the lowest value of variable xit over all i-t; and max(xit) is the highest value of xit. For variables capturing lack of financial development, such as interest rate spread, bank asset concentration, overhead costs, net interest margin, and non-interest income, one minus the formula above was used.

The weights were estimated with principal component analysis in levels and differences, factor analysis in levels and differences, as well as equal weights within a subcomponent of the index. For most of the methods the weights were not very different from equal weights and econometric results were robust to the method of aggregation. For simplicity, we use an index with equal weights.

Regression Frameworks

  • Regressions use 5-year averages in order to abstract from cyclical fluctuations, and estimated using dynamic panel techniques common in the growth literature.

Financial Development Gaps

  • The benchmarking regressions link financial development (FD), institutions (FI) and markets (FM) development indices to fundamentals. Following the literature on benchmarking financial development (Beck and others 2008) fundamentals (XitFI) included initial income per capita, government consumption to GDP, inflation, trade openness, educational attainment proxied by the average number of years of secondary schooling for people 25+, population growth, capital account openness, the size of the shadow economy (given its importance for the LAC region) and the rule of law. Instruments (Zit) for financial development such as the rule of law and legal origin dummies were also used. Predicted norms were computed using the following equation:
    FIit=δ1XitFI+δ2Zit+htFI+eitFI,

    where FIit stands for one of the financial indices (FD, FI or FM). Gaps shown are the difference between the actual values of the index and the calculated norms.

Financial Development, Growth, and Stability

  • The link between financial development, growth and stability was examined using a dynamic panel regression framework. Real GDP growth (DYit) is linked to financial development allowing for a potential non-linearity by adding a square of financial development while controlling for other factors that are likely to affect growth (below). In the case of individual sub-components of FI and FM, the interaction term between these two indices is included. The controls for the growth regression XitY were the same as in the benchmarking regression (XitFI) with two additional variables: ratio of FDI to GDP and capital account openness.

  • The impact of financial development on financial and macroeconomic instability used a similar framework. Financial instability (FSit) is measured by the first principal component of the inverse of the distance to distress (z-score),2 real credit growth volatility, and real and nominal interest rate volatility. This combined variable allows capturing different facets of financial instability, thus improving over previous research which typically focused on a single variable. Growth volatility (GVit) is measured by the standard deviation of GDP growth. The controls included initial income per capita, government consumption to GDP, trade openness, changes in terms of trade, growth in per capita income, capital flows to GDP, exchange rate regime, a measure of political stability, and an indicator for whether a country is an offshore financial center.

  • The following three equations were estimated using the Arellano-Bond approach:
    DYit=(a01)ln(Yit1)+bf(FinDevit)+g'XitY+htY+niY+eitYFSit=a0FSit1+bf(FinDevit)+g'XitS+htS+niS+eitSGVit=a0GVit1+bf(FinDevit)+g'XitV+htV+niV+eitV

    Where f(FinDevit) have two forms, one with the aggregated index: f(FDit)=b1FDit+b2FDit2

    and one with the subcomponents:
    f(FIit,FMit)=b1FIit+b2FIit2+b3FMit+b4FMit2+b5FIit×FMit

Table A5.1 shows the results of the estimated equations for growth and instability.

Estimated Equations

article image
Source: IMF staff calculations.Note: Standard errors in parentheses*** p<0.01, ** p<0.05, * p<0.1

Appendix 2. Construction of the Financial Inclusion Index

This Appendix explains the construction of Indices of Financial Inclusion and its components and provides an overview of the data and its processing for the construction of indices.

Measuring Financial Inclusion

Since there is no commonly accepted definition of financial inclusion we used a practical definition for the purpose of this note, namely, access and effective usage of financial services by households and firms. We employ multi-dimensional indices to capture different faucets of financial inclusion. In particular, we construct three multi-dimensional indices capturing different angles of financial inclusion: (i) usage of financial services by households (Findex); (ii) usage of financial services by SMEs (Enterprise Survey); and (iii) access to financial institutions (Financial Access Survey). The diagram below illustrates indicators included in each of the indices. We chose indicators that cover the most important aspects of financial inclusion emphasized in the literature, while taking into account data constraints. For example, the household inclusion index encompasses information on the use of bank accounts, savings, borrowing, and payment methods but omits information on insurance due to data constraints. We also chose not to combine the three indices into a single index, notably because cross-country data coverage across households and firms varies substantially. Instead, we compare Guatemala and other LAC countries to other regions and for households across time,1 separately on each dimension.2

A05ufig4

Financial Inclusion Index

Citation: IMF Staff Country Reports 2016, 282; 10.5089/9781475531534.002.A005

Data Sources and Processing

Table below shows the main data sources. The data from Global Findex covers the period for 2011 and 2014 only. The data point from enterprise survey is the latest observation available.

From the components to the composite index

All variables were normalized using the following formula: Ix,it

Ix,it=xitmin(xit)max(xit)min(xit)

Where Ix,it is the normalized variable x of country i on year t, min(xit) is the lowest value of variable xit over all it; and max(xit) is the highest value of xit. For those variables that capture a lack of financial inclusion, such as Value of collateral needed for a loan and percent of firms identifying access or cost of finance as major constraint, the reverse formula was used:

Ix,it=1xitmin(xit)max(xit)min(xit)

Several methods were used to estimate the weights: principal component analysis with the variables in levels and in differences, factor analysis with the variables in levels and in differences, as well as equal weights within a subcomponent of the index. For most of the methods the weights were not very different from equal weights and econometric results were robust to the method of aggregation. Thus, for simplicity of exposition the paper presents an index with equal weights.

article image

Household Inclusion Index

article image

Household Inclusion Index

article image

Firm Inclusion Index

article image

Access Index

article image

Most recent year available.

Appendix 3. Model Calibration

The model features an economy where agents differ in their talent and wealth. Each person has to decide whether to become a worker (earn wages) or an entrepreneur (earn profits) and whether to pay a fixed participation cost to be able to borrow from the banking system. Entrepreneurs then decide on how much of their wealth to invest in their business, whether and how much to borrow at the going interest rate, and how many workers to employ at the going wage rate. The output from business projects depends on the amount of capital invested, the amount of labor hired, as well as on the entrepreneur’s talent. In the model, the magnitude of the participation cost represents the cost of financial contracting. The higher is this cost, the more agents remain in credit autarky. Moreover, it tends to disproportionately exclude poor but talented individuals as the fixed cost amounts to a larger fraction of their wealth.

Once in the banking system, the amount of credit available is constrained by other financial frictions. If an entrepreneur has paid the participation cost, he or she can borrow from the banking system at the going interest rate. The model assumes that a business can fail for external reasons (“bad luck”), with some probability. Given imperfect enforceability of contracts, entrepreneurs have to post personal wealth as collateral for the loan. Since banks runs the risk that entrepreneurs can defraud them, this constrains the amount that can be borrowed. Therefore, weak contract enforceability leads to lower leverage, imposing borrowing constraints on entrepreneurs. A second friction is modeled as arising from asymmetric information between the bank and the borrower. The underlying intuition is that if the entrepreneur does not pay back the loan, the bank cannot be sure whether the business actually failed. Banks have to pay an audit or monitoring cost to find out. Otherwise, entrepreneurs could benefit from claiming failure and keep the profits. These costs—measure of the degree of intermediation costs in the economy—are recuperated by banks through interest rates and high overhead fees.1

In the baseline, the model is calibrated to data for 12 LAC countries. Firm-level data for 2005 from the World Bank Enterprise Survey are used, in addition to standard macroeconomic and financial variables (savings rate, non-performing loans (NPLs), and interest rate spreads) for 2010 or the latest year available. While lack of financial inclusion is an even more acute problem for firms in the informal sector, the model focuses primarily on formal sector firms. The model’s key parameters are jointly chosen to match the simulated moments, such as the percent of firms with credit and the firm employment distribution, with the actual data for each country (see Dabla-Norris et al., 2015, for details).

1

Prepared by Anna Ivanova and Victoria Valente

2

We use system GMM estimation (Arellano and Bover, 1995; Blundell and Bond, 1998) to address the dynamic dependence of our variables of interest and potential endogeneity of control variables. We also employ additional instrumental variables used in the literature, namely, rule of law (Kaufmann, Kraay and Mastruzzi 2010) and a set of dummies for the country’s legal origin (La Porta, Lopez-de-Silanes and Shleifer 2008).

3

Prepared by Noelia Camara (BBVA), Yixi Deng, Anna Ivanova, and Joyce Wong (all IMF).

4

The first and the third approaches follow closely those proposed in the IMF working paper “Financial Inclusion: Zooming in on Latin America” by Era Dabla-Norris, Yixi Deng, Anna Ivanova, Izabela Karpowicz, Filiz Unsal, Eva VanLeemput, and Joyce Wong.

5

Banking correspondents are non-financial commercial establishments that offer basic financial services under the name of a financial services provider, facilitating access points to the formal financial system, in particular, for low-income customers. The establishments are spread across diverse sectors (grocery shops, gas stations, postal services, pharmacies, etc.) as long as they are brick-and-mortar stores whose core business involves managing cash. In its basic form, banking correspondents carry out only transactional operations (cash in, cash out) and payments but, in some cases, they have evolved as a distribution channel for the banks’ credit, saving and insurance products. See Camara, Tuesta, and Urbiola (2015) for details.

6

Alternative multidimensional indices of financial inclusion developed in Camara and Tuesta (2014) suggest similar results.

7

The regressions explain a large portion of the variation in financial inclusion, with R-squares close to 0.7 in the regressions for households and firms. Nonetheless, the lack of a solid theory on the factors driving financial inclusion implies that the correct model specification is subject to uncertainty. Hence, the gaps should be interpreted with due caution, in particular, with respect to causality. Nevertheless, they could be useful in indicating a possible area where financial inclusion is lacking. The explanatory power for access to financial institutions regression is low and we omit the discussion of the results of this regression for Guatemala.

8

Specifically, we focus on changes in the steady state of the economy when these constraints changes. These examples are illustrative, however, as the targets for the illustrative scenarios are chosen arbitrarily (i.e. there is no reason why participation costs could or should be zero in Guatemala, for example). Moreover, in practice, as many reforms are implemented on various fronts contemporaneously they are likely to affect the frictions in unison with additive effects.

1

The framework in Heng and others, 2015, in turn, follows Sahay and others (2015). For further details, see “Advancing Financial Development in Latin America and the Caribbean,” forthcoming, IMF working paper.

2

Z-score is a measure of financial health. Z-score compares the buffer of a country’s commercial banking system (capitalization and returns) with the volatility of those returns.

1

Findex data is available for two years: 2011 and 2014.

2

We explore different aggregation methods, namely, weights derived from the principle component analysis (Camara, N., and D. Tuesta, 2014), factor analysis (Amidžić et al., 2014) and equal weights. The results are similar when using alternative measures (see Appendix 2). For simplicity of exposition we present the results for indices constructed using equal weights.

1

In the model, the bank’s optimal verification strategy follows Townsend (1979), whereby verification only occurs if the entrepreneur cannot pay the face value of the loan. This happens when the entrepreneur is highly leveraged and also faces a production failure. As a result, banks only monitor if a production failure is reported and the loan contract is highly-leveraged. A low-leveraged loan implies that entrepreneurs are not borrowing much from the bank and therefore the required repayment is small.

Guatemala: Selected Issues and Analytical Notes
Author: International Monetary Fund. Western Hemisphere Dept.