Informal financial markets in Botswana: a case study of Gaborone City

Abstract

The present study investigated informal financial markets in Gaborone, Botswana, with specific focus on the terms and conditions for informal credit, its main uses and the target clientele. The study used primary data, and analytical techniques that included descriptive statistics and analysis of variance. The results show that informal lenders give short-term consumption loans consisting of small amounts, charge very high interest rates, use innovative collateral substitutes such as automatic teller machine (ATM) cards plus personal identification numbers (PINs) and valuable household assets as security, and target mainly the non-poor. The paper makes three policy recommendations: an appropriate regulatory framework for the informal financial sector should be developed, interest rates and other charges should be systematically disclosed so as to encourage competition and reduce the high interest rates, and there should be legislation against the use of ATMs and PINs as security.

Keywords:

• Informal financial markets
• Botswana

1. INTRODUCTION

The financial sector in Botswana is comprised of the formal and informal sectors. The formal financial sector includes the Bank of Botswana (BoB) (as the central bank), commercial banks, merchant/investment banks, insurance companies, leasing finance institutions, a development bank, a savings bank, a building society, a development finance company, a stock exchange, stockbrokers, pension funds, asset management companies and collective investment undertakings (BoB, 2001).

Commercial and merchant/investment banks are covered by a single Act (Government of Botswana, 1995); the remainder are established and supervised under different statutes. The commercial banks, Botswana Savings Bank and the collective investment undertakings are, by law, supervised by the BoB, while the Ministry of Finance and Development Planning serves as the regulator of the National Development Bank, the Botswana Building Society, insurance companies and pension funds. The Stock Exchange Committee, under the overall supervisory oversight of the Ministry of Finance and Development Planning, runs the Botswana Stock Exchange (BoB, 2001).

Historically, the formal financial market in Botswana has been dominated by commercial banks, which are the major suppliers of formal credit to households and private businesses. The recently implemented financial-sector reforms in Botswana were geared towards making bank financial services more easily accessible to the public. The financial sector reforms included licensing of additional commercial banks and other financial institutions, review of the Banking Act to widen the definition of banking beyond the commercial banks and leasing companies to include merchant banks and discount houses, removal of restrictive licensing policies, and reducing the role of government in the financial sector. Despite these attempts, Mosene (2002) noted that smaller borrower households with little collateral are rationed out of formal credit markets, leaving informal credit markets as their alternative source of credit.

The informal financial sector encompasses all financial transactions that take place outside the supervision, regulation and control of the central bank. Currently the BoB does not have a department for non-bank financial institutions (NBFIs). It has a supervision department that deals with NBFIs only if they take deposits or are statutory. However, new legislation to establish an NBFI regulator was recently passed by the Botswana Parliament and is in the process of being operationalised. The informal financial institutions in Botswana are mainly micro-lending firms, which are non-deposit-taking institutions engaged in making small loans from their own resources. Other informal financial sector operators include rotating savings and credit associations (ROSCAs), pawnshops, burial societies and friends or relatives (Mosene, 2002).

The informal financial markets provide financial services to economic agents whose access to formal financial markets is constrained (Bouman, 1995; Okurut & Schoombee, 2007). Smaller borrower households' constrained access to formal-sector credit is argued to be due to both institutional and household-level factors (Nwanna, 1995). At the institutional level, the banks incur high information costs to assess the creditworthiness of small borrowers, and receive low returns because of the small loan amounts involved. This motivates the formal lenders to adopt strict collateral requirements as a screening mechanism to minimise the risk of default, hence keeping the small borrowers out of the formal credit market. At the household level, the low levels of income and asset accumulation, widespread poverty and highly skewed income and asset distribution give the small borrower households a high-risk profile that makes them less attractive to the formal lenders.

The small borrowers with constrained access to formal-sector credit are obliged to have recourse to the informal financial sector to meet their credit demand. This constitutes the demand side of the informal credit markets (Montiel et al., 1993). Informal credit may be demanded for productive investment (agricultural production or business) or consumption smoothing. Production credit facilitates the purchase of needed inputs for the production process, thereby increasing the household's productivity. The increased output can increase household income, other factors being constant. Business credit enhances the establishment or expansion of existing businesses (such as the marketing of agricultural products and general trade in goods and services), thereby increasing the household's income and generating employment opportunities. Credit therefore enables poor people who are not risk-averse to overcome their liquidity constraints, undertake some investments, increase productivity to boost income levels and create employment opportunities. In the process, informal credit may help alleviate poverty among low-income households (Binswanger & Khandker, 1995).

Consumption credit enables economic agents to smooth out consumption patterns over time (Heidhues, 1995; Mohieldin & Wright, 2000). This is particularly important for households whose main source of livelihood is the income from the agricultural sector, which is characterised by high variability owing to production and price shocks. Consumption credit therefore helps poor rural households to maintain their productive capacity and is thus argued to be productive in the long run (World Bank, 1989).

On the supply side of informal credit markets, the major players are the relatives, friends and moneylenders, and informal institutions such as cooperative savings and credit societies or unions, rotating savings and credit associations, and non-governmental organisations (Yadav et al., 1992; Aryeetey & Nissanke, 1998:61–77). Informal lenders avoid the risk of default, which reduces the profitability of lending operations, by carefully screening their potential borrowers and collecting as much information as they can about them to assess their creditworthiness. Besides this, they have developed collateral substitutes, such as joint liability contracts and tied credit contracts, to hedge against the risk of default (Okurut et al., 2005).

Mosene (2002) observes that informal financial activities continue to grow in Botswana, with most operations taking place in the urban and semi-urban areas and major villages. The present study therefore examined the terms and conditions of informal credit, the target clients and the main use of informal credit in Gaborone City.

2. THE MODEL

The study used descriptive statistics (percentage distributions) and the analysis of variance (ANOVA) model to analyse the key variables of interest across the various informal financial institutions. The main informal financial institutions the study identified were moneylenders, pawnshops, ROSCAs, savings and credit cooperatives, and community-based organisations. The ANOVA model was used to test the hypotheses that the means of variables of interest were the same across the various informal financial institutions (Billingsley et al., 1986):

where H ₀ is the null hypothesis, H ₁ is the alternative hypothesis, and μ _ik is the mean of the ith variable (loan amount, interest rate, etc.) for informal institution in the kth category (where k=5).

The ANOVA F-statistic for each variable was computed as follows:

where σ is the significance level, v1 is the numerator degrees of freedom (equal to k−1, where k is the number of categories), v2 is the denominator degrees of freedom (equal to n _T−k, where n _T is the total sample size), S ² _B is the between-group variance and S ² _W is the within-group variance.

In terms of the interpretation of the ANOVA results, if the computed F-statistic is greater than the critical F-statistic (at least at the 10 per cent significance level), then the null hypothesis will be rejected – otherwise accepted.

3. DATA SOURCES

The study used primary data collected from informal financial markets in Gaborone City in the months of October and November 2006 using a structured questionnaire. The total sample size was 111, which was generated using the snowball sampling technique. The rationale for using this technique was that there was no comprehensive list of informal financial institutions that could be used as the sampling frame.

It should also be noted that most of the moneylenders and pawnshops had a network of branches in Gaborone City. Since the motivation of the study was to examine their institutional characteristics (features of the financial products they offered, their charges and their target clientele), when the moneylender had more than one branch only one was selected for interview in each case. The rationale was that all the branches of a given moneylender or pawnshop have the same institutional characteristics.

4. EMPIRICAL RESULTS

This study investigated the landscape of informal financial markets in Gaborone, Botswana. Its specific focus was to identify the various supply-side players in informal financial markets, the range of financial products offered, the terms and conditions of informal credit, and the main clients targeted by informal financial institutions.

4.1 Informal financial services

The informal financial institutions in Gaborone provide a range of financial services that include credit and savings (see Table 1). The general interpretation of the statistics reported in all of the tables in this paper is as follows. The row percentage, in bold and italics at the bottom of the table, is the sub-total in each category expressed as a percentage of the total respondents who answered the particular question. In Table 1, for example, of the 111 informal financial institutions, 49 per cent provided only credit, 20 per cent provided only savings and 32 per cent provided both credit and savings. The other percentages reported in the table refer to the distribution within each category. For example, of the 54 institutions that said they provided only credit, moneylenders accounted for 54 per cent. The rest of the tables should be interpreted in a similar way.

Table 1:  

Type of informal institution	Financial services
	Credit only	%	Savings only	%	Both credit and savings	%	Total	%
Moneylenders	29	53.7	0	0.0	0	0.0	29	26.1
Pawnshops	9	16.7	0	0.0	0	0.0	9	8.1
ROSCAS	16	29.6	5	22.7	23	65.7	44	39.6
Savings and credit cooperatives	0	0.0	1	4.5	9	25.7	10	9.0
Community-based organisations	0	0.0	16	72.7	3	8.6	19	17.1
Total	54	100.0	22	100.0	35	100.0	111	100.0
Percentage	48.6		19.8		31.5		100.0
Note: Pearson χ^2(8) = 115.74. Prob = 0.000.

The informal financial institutions were independently distributed across the categories of financial services provided (χ² = 121.2 and Prob = 0.000).

The moneylenders and pawnshops provided strictly credit only. An interesting observation is that some of the ROSCAs said they provided credit services only, which is a slightly different model from the traditional ROSCAs as described in the literature. According to Bagachwa (1995), ROSCAs undertake financial intermediation through a clearly defined mechanism for mobilising resources and allocating credit. Resources are mobilised by regular pooling of members' savings, and credit is allocated mainly by lottery, where credit is allocated to the members on a rotational basis until each person has had a turn. The advantage of a ROSCA is that it give members access to a lump sum of money at one go, which they do not have when saving individually. The credit contract may be either interest free or charged market interest rates, depending on the rules set by the group. Even in the case of interest-free contracts, an implicit interest rate is paid by the members in the form of interest foregone on pooled savings deposits over time as one waits for one's turn (Bouman, 1995).

However, most of the ROSCAs in Botswana have the characteristics of microlenders, where they pool the resources to create a revolving fund for lending to both members and non-members. Because all the pooled money is lent out, the ROSCAs do not see themselves as being involved in savings. The interest from the lending operations is then shared among the members periodically.

The community-based organisations in Botswana are mainly burial societies that contribute money to help the bereaved family with funeral expenses. The kinds of contributions vary: in some of these groups, the members make a small regular contribution (usually monthly) that accumulates over time, in others they pay periodically (e.g. contributing P50 each time a death occurs).

The centrality of credit operations among informal institutions is underscored in their response to the question as to how the mobilised savings are mainly used (see Table 2). In the category of informal financial institutions that used savings as a revolving loan fund, the majority were ROSCAs (91 per cent).

Table 2:  

Type of informal institution	Revolving loan fund	%	Social events	%	Invest group projects	%	Total
ROSCAs	20	90.9	2	9.1	3	42.9	25
Savings and credit cooperatives	2	9.1	1	4.5	4	57.1	7
Community-based organisations	0	0.0	19	86.4	0	0.0	19
Total	22	100.0	22	100.0	7	100.0	51
Percentage	43.1		43.1		13.7		100.0

4.2 Duration of informal financial institutions

The data collected in this study suggest that most of the informal financial institutions have been in existence for more than 10 years. However, they had grown significantly during the previous 3 years (see Table 3). Their emergence may be linked to developments in the formal financial sector in Botswana that have obliged economic agents whose access to formal financial services is constrained to have recourse to the informal financial sector.

Table 3:  

Type of informal institution	Age of business
	<1 year	%	1–3 years	%	3–5 years	%	5–10 years	%	> 10 years	%	Total
Moneylenders	10	35.7	7	23.3	4	19.0	5	35.7	3	16.7	29
Pawnshops	2	7.1	4	13.3	1	4.8	2	14.3	0	0.0	9
ROSCAs	14	50.0	17	56.7	10	47.6	2	14.3	1	5.6	44
Savings and credit cooperatives	0	0.0	1	3.3	2	9.5	1	7.1	6	33.3	10
Community-based organisations	2	7.1	1	3.3	4	19.0	4	28.6	8	44.4	17
Total	28	100.0	30	100.0	21	100.0	14	100.0	18	100.0	111
Percentage	25.2		27.0		18.9		12.6		16.2		100.0

4.3 Target clients

The target clientele of the informal financial institutions in Gaborone (see Table 4) is mainly female (about 60 per cent). This may be because women lack collateral to offer as security for bank loans since many of them do not own property (Okurut et al., 2005). The empirical literature suggests that the high participation of women in informal financial institutions may also be explained by some key features of these informal markets, such as their accessibility, the simple procedures, and the flexibility of credit and its adaptability to many purposes (Hoff & Stiglitz, 1990; Aryeetey, 1994).

Table 4:  

Type of informal institution	Gender of clients
	Males	%	Females	%	Equal representation	%	Total
Moneylenders	5	35.7	11	16.4	13	43.3	29
Pawnshops	3	21.4	1	1.5	5	16.7	9
ROSCAs	2	14.3	35	52.2	7	23.3	44
Savings and credit cooperatives	2	14.3	4	6.0	4	13.3	10
Community-based organisations	2	14.3	16	23.9	1	3.3	19
Total	14	100.0	67	100.0	30	100.0	111
Percentage	12.6		60.4		27.0		100.0

The informal lenders were also asked to indicate the income status of their clients. The responses to the question were ranked as high, medium and low. It should be noted that there were no financial figures attached to any of the categories because to give a precise answer the lenders would have had to supply their clients' detailed personal and household income information of the clients, and this information was therefore not available. Such information could only be obtained from the clients themselves, a task that was outside the scope of the study. The results should thus be interpreted as an index of the lenders' perceptions of the income status of their clients. The results suggest that informal financial institutions in Botswana target mainly clients from the medium-income category (57 per cent), while only 41 per cent are from the low-income category (see Table 5). This finding provides an interesting contrast with the empirical literature that suggests the informal financial institutions target mainly the poor (Hoff & Stiglitz, 1990; Hulme, 2000).

Table 5:  

Type of informal institution	Income bracket
	High	%	Medium	%	Low	%	Total
Moneylenders	0	0.0	21	33.3	8	17.8	29
Pawnshops	1	33.3	4	6.3	4	8.9	9
ROSCAs	2	66.7	27	42.9	15	33.3	44
Savings and credit cooperatives	0	0.0	6	9.5	4	8.9	10
Community-based organisations	0	0.0	5	7.9	14	31.1	19
Total	3	100.0	63	100.0	45	100.0	111
Percentage	2.7		56.8		40.5		100.0

An analysis of the informal financial institutions' clients' ages suggests that most of them are between 30 and 40 years old (60 per cent), followed by those aged 40–50 years (25 per cent), which indicates that most of them are economically active (see Table 6).

Table 6:  

Type of informal institution	Age of clients/members
	20–30 years	%	30–40 years	%	40–50 years	%	>50 years	%	Total
Moneylenders	1	8.3	24	35.8	4	14.3	0	0.0	29
Pawnshops	2	16.7	7	10.4	0	0.0	0	0.0	9
ROSCAS	8	66.7	25	37.3	10	35.7	1	25.0	44
Savings and credit cooperatives	1	8.3	3	4.5	6	21.4	0	0.0	10
Community-based organisations	0	0.0	8	11.9	8	28.6	3	75.0	19
Total	12	100.0	67	100.0	28	100.0	4	100.0	111
Percentage	10.8		60.4		25.2		3.6		100.0

4.4 Mechanisms for hedging against loan default risk

The information asymmetry that prevails in credit markets gives rise to default risk (Van Tassel, 1999; Ghatak, 1999). Informal financial institutions in Botswana have therefore developed innovative mechanisms to hedge against this risk, in the form of prerequisites for potential clients and security requirements.

4.4.1 Prerequisites for clients

The key prerequisites for the clients are that they must have a monthly income (59 per cent) or be employed (19 per cent). These are used as the basis for assessing the potential borrowers' repayment capacity, especially by the moneylenders and ROSCAs (see Table 7). A discussion with lenders revealed that loans granted to clients in paid employment are linked to their net pay (on average 50 per cent of net pay) to ensure that both principal and interest will be fully recovered from the monthly earnings. The borrowers have to provide a pay slip and their national identity card (omang) or passport as proof of eligibility for the loan.

Table 7:  

Type of informal institution	Prerequisites for clients
	Employed	%	Monthly income	%	Member of community	%	Have asset	%	Relationship with a member	%	Total
Moneylenders	10	58.8	19	35.8	0	0.0	0	0.0	0	0.0	29
Pawnshops	0	0.0	3	5.7	0	0.0	6	100.0	0	0.0	9
ROSCAs	7	41.2	21	39.6	4	80.0	0	0.0	8	88.9	40
Savings and credit cooperatives	0	0.0	8	15.1	0	0.0	0	0.0	1	11.1	9
Community-based organisations	0	0.0	2	3.8	1	20.0	0	0.0	0	0.0	3
Total	17	100.0	53	100.0	5	100.0	6	100.0	9	100.0	90
Percentage	18.9		58.9		5.6		6.7		10.0		100.0

For pawnshops, the critical prerequisite is that the potential client must have valuable assets to surrender to the pawnbroker as security. The pawnbroker links the loan amount granted to the client to a fixed proportion of the estimated market value of the asset (on average about 25 per cent). This provides an incentive to the client to repay the loan because the opportunity cost of losing the pawned asset is higher than the loan, hence minimising loan default risk. Among the ROSCAs, the additional prerequisite for the potential clients is that they must be members of the community or be introduced by an existing member.

It can be concluded that informal financial institutions in Botswana use dynamic selection criteria to identify potential clients (in the form of prerequisites). They have cost-effective mechanisms for collecting information about potential clients. This is consistent with the empirical literature. These criteria enable them to overcome the information asymmetry constraints prevalent in credit markets and assess the creditworthiness and strategic behaviour of potential borrowers effectively, thus minimising the default risk and improve the rate of loan repayment (Stiglitz, 1990; Besley & Coate, 1995; Conning, 1999; Conlin, 1999; Coleman, 1999; Varian, 1990).

4.4.2 Loan security requirements

Informal lenders require security for their loans. There were two main types. About 20 per cent, mostly the pawnshops, required household assets as security. However, just over 70 per cent of the respondents said they ask for automatic teller machine (ATM) cards (see Table 8). This use of ATM cards as security is a new and surprising feature of the informal financial sector. Moneylenders have taken advantage of the development of this new product in the formal financial sector to hedge against default risk. A discussion with the moneylenders revealed that the modus operandi is that clients must surrender their ATM cards plus personal identification numbers (PINs). The moneylenders withdraw loan repayments directly from banks at the month end, after which the ATM cards are returned to the borrowers. The perfect predictability of paydays facilitates the process: most institutions in Botswana publish the expected paydays for all the months of the year at the beginning of each calendar year. This enables the moneylenders to gauge accurately when salary deposits will be made and thus forestall withdrawals borrowers might make using cheques or vouchers over the bank counters.

Table 8:  

Type of informal institution	Security type
	Household assets^a	%	Cars	%	Houses	%	ATM cards	%	Guarantors	%	Total
Moneylenders	1	11.1	0	0.0	0	0.0	25	78.1	0	0.0	26
Pawnshops	7	77.8	2	100.0	0	0.0	0	0.0	0	0.0	9
ROSCAs	1	11.1	0	0.0	0	0.0	3	9.4	0	0.0	4
Savings and credit cooperatives	0	0.0	0	0.0	1	100.0	4	12.5	1	100.0	6
Total	9	100.0	2	100.0	1	100.0	32	100.0	1	100.0	45
Percentage	20.0		4.4		2.2		71.1		2.2		100.0
^aThe household durable assets include television sets, music systems and computers.

It can be concluded that informal financial institutions in Botswana use effective collateral substitutes. This is consistent with the empirical literature. However, the practice of asking for ATM cards and PIN numbers as security may be argued to be highly undesirable. First, it violates the rules of the banks that issue these cards. Secondly, such clients are vulnerable to loss and fraud and will get no recompense from the banks if there is fraudulent use of the cards and PINs. It is recommended that the forthcoming regulatory framework for NBFIs should protect the public from such risky practices.

In addition, this analysis of securities demanded by informal financial institutions in Botswana lends further credence to the argument that they serve the non-poor. The rationale is that the poor (e.g. those who are either unemployed or occupied in the informal sector or in traditional subsistence agriculture) may not have ATM cards and valuable property to offer as security.

The Gaborone informal financial institutions have high loan repayment rates (of 80–100 per cent), as reported by just over 70 per cent of the respondents (see Table 9).1 These high repayment rates may be explained by the mechanisms the lenders have adopted, in the form of security requirements, for hedging against the risk of default. This is consistent with empirical literature that suggests informal financial institutions are able to extend credit and recover their loans even from those clients the formal banks consider a bad risk (Rhyne & Otero, 1992; Morduch, 1999).

Table 9:  

Type of informal institution	Repayment rate
	<20%	%	21–40%	%	41–60%	%	61–80%	%	81–100%	%	Total
Moneylenders	3	60.0	2	66.7	3	30.0	1	16.7	20	30.3	29
Pawnshops	0	0.0	0	0.0	4	40.0	0	0.0	5	7.6	9
ROSCAs	0	0.0	1	33.3	3	30.0	4	66.7	32	48.5	40
Savings and credit cooperatives	2	40.0	0	0.0	0	0.0	1	16.7	6	9.1	9
Community-based organisations	0	0.0	0	0.0	0	0.0	0	0.0	3	4.5	3
Total	5	100.0	3	100.0	10	100.0	6	100.0	66	100.0	90
Percentage	5.6		3.3		11.1		6.7		73.3		100.0

4.5 Key features of informal financial products in Botswana

The discussion in this section focuses only on credit, because the savings mobilisation by informal financial institutions is largely geared to capitalising the loan fund. The key credit features are loan amounts, loan periods, interest rates charged, repayment frequencies and how the clients use the credit. The ANOVA model was used to test for differences of means of product features across the different categories of informal financial institutions (see Table 10).

Table 10:  

Type of informal institution	Mean loan amount (P)	Mean loan period (months)	Mean interest rate per month (%)
Moneylenders	1375.9	1.2	23.4
Pawnshops	900.0	1.1	25.6
ROSCAS	1362.5	1.2	17.5
Savings and credit cooperatives	1580	1.3	18.3
Community-based organisations	1078.9	1.1	16.7
Overall mean	1299.5	1.2	20.4
F-statistic	1.17	0.57	6.91
Prob>F	0.3272	0.6871	0.0001

4.5.1 Loan amounts

The mean loan amounts granted to clients by informal financial institutions are generally small (approximately P1299.5).2 The savings and credit cooperatives granted the highest mean loan amount of P1580, while the pawnshops granted the lowest at P900. However, on the basis of the ANOVA test, the mean loan amounts were not statistically different across the various informal financial institutions (at least at the 10 per cent significance level, as indicated by the F-statistic of 1.17 and Prob > F = 0.3272).

4.5.2 Loan periods

The informal financial institutions granted only short loan periods, on average 1.2 months. The highest mean loan period (1.3 months) was extended by savings and credit cooperatives, while the lowest was by pawnshops and community-based organisations (1.1 months). The ANOVA test suggested that the mean loan periods were not statistically different across the informal financial institutions (see Table 10). These short loan periods offered by informal financial institutions in Gaborone is consistent with evidence in the empirical literature on other developing countries.

4.5.3 Interest rate on loans

The overall mean interest rate on informal loans is 20.4 per cent per month (see Table 10). The highest mean rate was charged by the pawnshops (25.6 per cent per month), followed by moneylenders (23.4 per cent per month). The mean rates were significantly different across the various informal financial institutions (at the 1 per cent significance level). The informal lenders apply compounded monthly interest rates. In addition, the study results suggest that informal-sector interest rates in Botswana are significantly higher than formal bank rates. For example, for a loan of P2000 at an interest rate of 20 per cent per month, informal lenders charge interest of P400 for one month. For a similar loan of P2000 at an interest rate of 20 per cent per annum, formal banks charge P33.3 for one month. The commercial bank prime lending rates at the time of the survey were 16.50 per cent per annum, while NBFIs' short-term loan interest rates were 19.50 per cent per annum. The BoB rate (i.e. the rate at which the Central Bank lends to commercial banks) was 15.0 per cent per annum (BoB, 2006).

The high informal-financial-sector interest rates in Botswana may be explained by monopolistic competition in these markets (Basu, 1983). Given evidence from the empirical literature that informal financial institutions have lower administrative costs (Timberg & Aiyar, 1984) and lower transaction costs (Aryeetey, 1994) than the formal banks, the mean interest rate of 20.4 per cent per month may be argued to be too high.

The practice of informal financial institutions in Gaborone charging much higher interest rates than formal banks is consistent with the empirical literature. Timberg and Aiyar (1984), in a study of urban moneylenders in India, found that they charged higher interest rates (18–24 per cent per annum) than the formal banks (13–16 per cent per annum). Further evidence by Bagachwa (1995) also showed that moneylenders' interest rates in Tanzania of 6–9 per cent per month were significantly higher than those of the formal sector of 21–31 per cent per annum.

From a policy perspective, it is recommended that the new NBFI regulatory framework should enhance competition in the informal financial sector through a policy of full disclosure of interest rates and other charges so as to enable clients to shop around and compare rates. This could drive informal interest rates down to levels where informal lenders would earn normal risk-adjusted profits, and the lower rates would of course have welfare benefits for the clients.

4.5.4 Loan repayment frequency

The repayment frequency for informal loans is monthly as reported by almost all of the respondents (see Table 11), which is consistent with the short-term lending practices of most informal financial institutions as reported in the empirical literature.

Table 11:  

Type of informal institution	Repayment frequency
	Weekly	%	Monthly	%	Total
Moneylenders	0	0.0	29	33.7	29
Pawnshops	0	0.0	9	10.5	9
ROSCAs	0	0.0	37	43.0	37
Savings and credit cooperatives	1	100.0	8	9.3	9
Community-based organisations	0	0.0	3	3.5	3
Total	1	100.0	86	100.0	87
Percentage	1.1		98.9		100.0

4.5.5 Uses of informal credit

The informal lenders were also asked about the main use to which their clients put the loan funds. Their answers were based on the reasons usually given by clients at the time of asking for loans. The well-known problem of fungibility of credit (where the actual use of the credit may differ from the reason given at the time of application) was assumed to be minimal, since informal lenders are not very rigid about how the funds should be used. The results in Table 12 suggest that approximately 85 per cent of credit from informal financial institutions is for consumption purposes (i.e. household consumption about 45 per cent and family emergencies just under 40 per cent). Interestingly, some of the informal loans are used to repay other debts, which in the long run may plunge these borrowers into a vicious cycle of debt and increase their vulnerability to poverty. This is consistent with observations by Ardington et al. (2003) based on their study in South Africa.

Table 12:  

Type of informal institution	Use of loan funds
	Household consumption^a	%	Working capital	%	Pay debts	%	Family emergency^b	%	Total
Moneylenders	13	31.7	2	28.6	3	42.9	11	32.4	29
Pawnshops	3	7.3	0	0.0	2	28.6	4	11.8	9
ROSCAs	20	48.8	3	42.9	2	28.6	14	41.2	39
Savings and credit cooperatives	5	12.2	2	28.6	0	0.0	2	5.9	9
Community-based organisations	0	0.0	0	0.0	0	0.0	3	8.8	3
Total	41	100.0	7	100.0	7	100.0	34	100.0	89
Percentage	46.1		7.9		7.9		38.2		100.0
^aHousehold consumption includes purchase of food and durable assets. ^bFamily emergencies include funeral expenses, treatment in case of sickness and payment of school fees.

This finding that informal credit is largely for consumption purposes differs from the findings of Morewagae et al. (1995), who in their survey of 1140 micro enterprises in Botswana concluded that informal credit was mainly used for investment (for both start-up and expansion).

In conclusion, it can be argued that informal financial institutions are a double-edged sword for the low-income households (Dallimore & Mgimeti, 2003). In the first instance, the informal financial institutions may be a blessing because they give improved access to credit to low-income households whose access to formal bank credit is constrained. However, the high interest means the sword may cut both ways – in the long run, the low-income households may find themselves deeper entrenched in poverty if they use their loans for debt servicing.

5. SUMMARY AND CONCLUSIONS

The informal financial markets in Gaborone are made up of moneylenders, pawnshops and rotating savings and credit associations. These institutions have features that match those reported in the empirical literature, such as having strict eligibility criteria for selecting potential clients, using innovative collateral as security for loans, giving small short-term consumption loans, and charging high interest rates. However, what makes the Gaborone informal financial institutions unique is that they target not the poor but the non-poor. From the above analysis, three policy recommendations are suggested to improve the performance of the informal financial sector.

5.1 Develop regulations for the informal financial sector

It is critical to develop an appropriate regulatory framework for the informal financial institutions, in line with the newly passed legislation for the NBFIs, to protect both the institutions and the clients. The newly established NBFIs Authority, as the lead agency, will guide the development of this regulatory framework in consultation with the stakeholders. Countries such as Uganda, which recently passed the Microfinance Deposit-taking Institutions Act (2003) to regulate the NBFIs, could provide useful lessons.

5.2 Enhance competition in informal financial institutions

The high interest rates prevail in informal financial markets, and to control interest rates would be inconsistent with the current policy framework of liberalisation. The most logical way of driving informal interest rates downwards would therefore be by encouraging competition in the sector. The policy option is to impose full disclosure of interest rates and other charges in the NBFIs' regulatory framework so as to empower the clients to make informed choices between informal finance providers. The drop in interest rates would translate into positive welfare benefits for low-income households.

5.3 Prevent the use of ATM cards as security

ATMs are financial instruments that must be protected. Adequate provisions should be made in the forthcoming NBFIs regulations to protect clients from the risky practice of informal lenders taking ATMs and PINs as security for loans. A massive fraud involving ATMs could cause a great financial loss to the clients.

The authors acknowledge the financial support for the study from Botswana Institute for Development Policy Analysis.

Notes

¹The loan repayment rate is defined as the amount recovered expressed as a percentage of the amounts due in the period.

²The loan amounts are in Botswana pula (exchange rate US$1 = BWP6.12 at the time of writing).