Journal Issue
Share
Article

West African Economic and Monetary Union: Selected Issues

Author(s):
International Monetary Fund. African Dept.
Published Date:
April 2015
Share
  • ShareShare
Show Summary Details

Financial Inclusion in the Waemu1

WAEMU countries lag behind benchmark countries in several dimensions of financial inclusion: Access to finance is low, especially for the most vulnerable parts of the population, and the financial sector appears to only modestly contribute to the population’s ability to deal with shocks as well as firms’ investment programs. Private sector credit-to GDP ratios, however, appear broadly in line with WAEMU countries fundamentals, and this note points to policies, such as investments in infrastructure and the social sectors, which could help closing these gaps. From the firms’ perspective, policies to reduce participation costs in the financial sector and to lower collateral requirements could increase firms’ access to financing, and thus significantly boost GDP.

A. Benchmarking Financial Access

1. Financial access in the WAEMU remains comparatively low. Figure 1 compares different indicators of financial access in the WAEMU against a group of fast growing regional and Asian benchmark countries.2 It shows that:

  • WAEMU countries on average lag behind benchmark groups in the provision of basic financial infrastructure, such as the density of ATMs and the number of bank branches.
  • The relative amount of deposit and loans at commercial banks is broadly in line with African benchmark groups, but significantly lower than those in Asian benchmark countries; the number of people with deposits at commercial banks is relatively low.
  • The short-comings in financial access are also revealed by enterprise surveys in each WAEMU country, with more than half of respondents identifying access to finance as a major constraint for their businesses.

Figure 1.WAEMU: Financial Access

2. While modest in general, financial access appears to be lowest for the most vulnerable parts of the population (Figure 2). Young adults and the population at the bottom of the income distribution (bottom 40 percent) are the groups with the lowest relative number of bank accounts (less than 5 percent of the respective part of the population), but the population living in rural areas, with less education and women are also less often in the possession of a financial account than the average WAEMU inhabitant. In general, accounts are most often used for business purposes or to receive payments such as wages or remittances.

Figure 2.WAEMU: Demographical Characteristics of Financial Access

3. The main modes to access finance and make deposits are similar to those in benchmark countries, but several payment methods are less pronounced in the WAEMU (Figure 3). As in benchmark countries, the use of a bank teller is the main way to make a deposit. Checks and electronic payments, however, are much less developed modes of payments than in the comparator groups, and a much smaller share of the WAEMU’s population is in the possession of a credit or debit card. The following note provides a separate benchmarking exercise for the use of mobile payments.

Figure 3.WAEMU: Deposit and Payment Modes

4. The use of loans and purposes of saving points to relatively weak social protection and only a modest contribution of the financial sector in shock mitigation (Figure 4). While the share of the population with outstanding loans for educational fees is comparable to benchmark countries, the share indebted people due to health issues or other emergencies is relatively high in the WAEMU. The population appears relatively less covered by health insurance and, with the exception of Mali, by agricultural insurance. Less people (are able to) save for emergencies in the future. While pointing to absolute and relative weaknesses in social protection, these indicators also suggest that the financial sector does only provide insufficient help to the population to insure against or deal with shocks.

Figure 4.WAEMU: Use of Loans

5. The banking sector’s contribution to firms’ investment programs also appears limited (Figure 5). Enterprise surveys indicate that, while most firms possess a bank account, less than 30 percent of firms access a loan or a line of credit in most WAEMU countries. The majority of loans require collateral. The value of such collateral on average exceeds the value of the loan, indicating problems with the liquidation of the collateral. Loans from banks constitute only a small fraction of firms’ investment financing, while internal funds appear to be the main source of financing investments.

Figure 5.WAEMU: Firms

B. Explaining Private Sector Credit Gaps

6. Private sector credit to GDP ratios are broadly in line with the benchmark for the WAEMU on average, but there are variations across countries (Figure 7). Following the methodology in Al Hussainy (2011) and Barajas et al. (2013), this note estimates a benchmark ratio of private sector credit to GDP based on a number of structural factors in a panel of over 120 emerging and developing countries for the period from 1986 to 2013.3 The fitted values from these regressions serve as the private sector-to-GDP benchmark. While generally following the dynamics of the benchmarks well, actual credit-to-GDP has been lower than the benchmark in 2013 in four countries (Benin, Burkina Faso, Côte d’Ivoire, and Guinea-Bissau), higher in three (Mali, Niger, Togo), and broadly consistent with the benchmark in Senegal.

Figure 6.WAEMU: Drivers of the Financial Gap

(One WAEMU Standard Deviation Increase, in Percent of GDP)

Figure 7.WAEMU: Credit to the Private Sector

(As a Share of GDP)

7. A number of policies could help countries to increase private sector credit relative to the benchmark (Figure 6, Table 1). In the next step, a regression of the financial gap (actual private sector credit-to-GDP minus its benchmark) on macroeconomic, institutional and policy variables helps identifying the drivers of the deviations from the benchmark for 2004-2013. Table 1 highlights the factors which help increasing private sector credit relative to the benchmark, while Figure 3 depicts the change in the private sector credit-to-GDP relative to the benchmark if these underlying factors are changed by one standard deviation.4 Factors which relate positively to private sector credit-to-GDP include trade openness and FDI inflows on the external side; lower inflation and higher social and educational spending on the macroeconomic (policy) side, as well as better infrastructure and institutions (here ICRG index).5

Table 1.WAEMU: Determinants of Financial Inclusiveness Gaps, 2004-2013
(1)(2)(3)(4)(5)
Economic Environment
Growth−0.004 ***−0.004 ***
(−3.08)(−3.04)
US Federal Funds Rate10.0030.008 **
(0.69)(2.17)
External Stance
FDI/GDP0.002 ***0.002 ***
(2.92)(3.17)
Trade Openess0.209 ***0.217 ***
(3.49)(3.40)
Capital Controls0.076 ***0.115 ***
(3.96)(5.73)
Policies
Fiscal Balance (cycl. adjusted)/GDP−0.185 **−0.247 **
(−2.16)(−2.52)
Inflation−0.004 ***−0.003 ***
(−5.50)(−4.48)
FX Regime0.007
(1.15)
Health Spending/GDP1.575 ***1.202 ***
(3.61)(2.57)
Institutions and Infrastructure
Institutions (ICRG)0.295 ***0.234 ***
(4.38)(3.14)
Telephone Lines0.000 ***0.000 ***
(11.44)−11.74
Internet Use0.001 **0.001 *
(2.01)(1.75)
Credit Information Depth−0.002
(−0.69)
Constant0.019 *−0.119 ***−0.033−0.183 ***−0.308 ***
(1.76)(−5.03)(−1.17)(−5.74)(−7.38)
Number of observations10551055105510551055
R-squared0.010.040.040.090.18

Proxy for external environment.

Robust t-statistics in parentheses; significance levels at 10 percent (*), 5 percent (**), and 1 percent (***) levels, respectively.

Proxy for external environment.

Robust t-statistics in parentheses; significance levels at 10 percent (*), 5 percent (**), and 1 percent (***) levels, respectively.

C. Identifying the Most Binding Constraints to Firms’ Financial Inclusion6

8. A micro-founded general equilibrium model helps identifying the most binding constraints to financial inclusion from firms’ perspective. In this section, the micro-founded general equilibrium model by Dabla-Norris et al. (2015) is calibrated to quantify the most binding constraints to financial inclusion and, as a consequence, growth, productivity and a more equal income distribution. Agents in the model differ from each other in wealth and talent and can choose to become entrepreneurs or supply labor for wages. They face three financial frictions:

  • Participation costs ψ which limit access to credit, in particular for smaller and poorer entrepreneurs
  • Intermediation costs χ due to asymmetric information between banks and borrowers which result in deposit-lending spreads
  • Imperfect enforceability of contracts which results in collateral requirements and thus smaller collateral leverage ratios λ.

To determine the values of the parameters ψ, χ and λ, as well as other parameters for the calibration, a range of macroeconomic and financial indicators are fed into the model (Table 2).

Table 2.WAEMU: Target Moments
Savings (in Percent of GDP)14.5
Collateral (in Percent of Loan Value)170
Firms with Credit (in Percent of Firms)20
Non-Performing Loans (in Percent of Loans)17
Interest Rate spread7.4

9. Preliminary results point to participation costs and high collateral requirements as the main borrowing constraints on average in the WAEMU. Based on calibration, Figures 7-9 depict the effects of relaxing individually each of the three financial constraints on the number of firms accessing credit, GDP, productivity, income inequality, interest rate spreads and the NPL ratio. They suggest that, while both lower participation costs and lower collateral requirements could yield significant GDP gains, they have differentiated effects on other variables, in particular:

  • Increasing financial access (Figure 8): Lowering participation costs, such as transaction costs, institutional impediments, and bureaucratic hurdles, could increase the fraction of firms with credit substantially. With more access to credit which leads to higher investments GDP increases significantly. From the WAEMU’s current position, lower participation costs could also decrease income inequality as measured by the Gini coefficient, because previously constrained (less wealthy) entrepreneurs over-proportionately benefit from the change when they enter the market. Overall productivity may decline for the same reason.
  • Lowering collateral constraints (Figure 10). Policies which could help decrease collateral requirements, such as the introduction of collateral registries, could also yield large GDP gains and increase productivity through gains in efficiency. The latter effect differs from the impact of policies which increase financial access described above as it over-proportionately benefits more talented entrepreneurs. While relaxing the collateral constraints allows all firms to borrow more, less talented businesses do not scale up their businesses by the same magnitudes as their maximum business scale is sooner achieved. As a consequence, the policy may lead to an increase in income inequality.

Figure 8.WAEMU: Lowering Participation Costs

(from left to right, dot indicates initial position)

Figure 9.WAEMU: Lowering the Cost of Intermediation

(from left to right, dot indicates initial position)

Figure 10.WAEMU: Lowering Collateral Constraints

(from left to right, dot indicates initial position)
References

    Ed Al HussainyAndrea CoppolaErikFeyenAlainIzeKatieKibbuka and HaocongRen (2011): A Ready-to-Use Tool to Benchmark Financial Sectors Across Countries and Over Time,FinStats2011 (Washington: World Bank).

    • Search Google Scholar
    • Export Citation

    Dabla-NorrisEraYanJiRobertTownsend and FilizUnsal (2015) “Identifying Constraints to Financial Inclusion and Their Impact on GDP and Inequality,NBER Working Paper No. 20821

    • Search Google Scholar
    • Export Citation

    BarajasAdolfoThorstenBeckEraDabla-Norris and SeyedReza Yousefi (2013) “Too Cold, Too Hot, or Just Right? Assessing Financial Sector Development Across the Globe,IMF Working Paper WP/13/81

    • Search Google Scholar
    • Export Citation
1Prepared by Monique Newiak (AFR) and Rachid Awad (MCM), with valuable contributions from Filiz Unsal, Era Dabla-Norris (both SPR) and Eva Van Leemput (University of Notre Dame). The findings should not be reported as representing the views of the IMF. Section C “Identifying the Most Binding Constraints to Firms’ Financial Inclusion” is a part of a research project on macroeconomic policy in LICs supported by UK’s Department of International Development (DFID). The findings should not be reported as representing the views of DFID.
2African benchmark countries include: Ghana, Kenya, Lesotho, Rwanda, Tanzania, Uganda, and Zambia. Asian benchmark countries include: Bangladesh, Cambodia, India, Laos, Nepal, and Vietnam. Sub-Saharan Africa is provided as a comparator in many cases as well.
3It regresses the ratio of private sector credit-to-GDP on: (i) the log of GDP per capita and its square, (ii) the log of the population to proxy for market size, (iii) the log of population density to proxy for the ease of service provision, (iv) the log of the age dependency ratio to account for demographic trends and the related savings behavior, (v) an oil exporters dummy, and time dummies to control for globale factors.
4Standard deviation calculated over WAEMU country time series from 2004 to 2013.
5ICRG: International country risk guide.
6We thank Eva Van Leemput for providing the calibrations.

Other Resources Citing This Publication