Finance and Rural Market. By : Netti Tinaprilla

Transcription

1 Finance and Rural Market By : Netti Tinaprilla 1

2 INTRODUCTION The bulk of the world poor is in rural areas Rural finance remains a major policy concern In the 1960s, there were agricultural banks, development banks and rural cooperatives to serve rural needs. Large nonperforming loans experience of such institutions led the World Bank and others to rethink rural finance strategy. In the 1970s and early 1980s, micro finance institutions led by Grameen Bank in Bangladesh started a new trend. 2

3 WHAT IS RURAL FINANCE? Rural finance is defined as financial services offered and used in rural areas by people of all income levels. Rural finance covers all the savings, lending, financing and risk minimizing opportunities (formal/informal) and institutions in rural areas. Rural financial markets are part of the domestic financial system and are therefore affected by government and central bank policies. Rural financial markets tend to be fragmented and consist of formal, semi formal and informal financial intermediaries. 3

4 COMPLETE RURAL FINANCE Important to remember that rural finance must cover the following : Payment services, including remittances Seasonal credit for farming, consumption and investment General insurance and cover against uncertainties 4

5 RECENT LESSONS Formal financial institutions, including development banks, do not always meet the needs of the rural poor, because intermediation costs are too formal and high. They do not understand rural needs. Rural credit is complex because it involves different types of crops, with specialist skills for each crop. It was always the ready access to credit that mattered, rather than the cost of credit. Formal rural financial institutions often prefer to lend to large borrowers, rather than small borrowers. However, the non-performing loan losses were often higher than those for small borrowers. 5

6 MICRO-FINANCE In recent years, efforts to expand access to credit to the poor have focused largely on funnelling (penyaluran) credit to the poor through micro-finance institution (MFIs). A clear strength of such programs has been their ability to reach the poor and to do so with astonishingly (mengherankan) small default rates. However, as these programs have proliferated (berkemb) a number of concerns have arisen about the capacity of MFIs to adequately serve the financial needs of poor households on a sustainable basis. Evaluation of micro-credit programs to date provide little evidence that they actually enable borrowers to accumulate productive assets or move out of poverty. 6

7 HISTORY OF RURAL CREDIT The growing monetization of the Malay peasant economy in the 19 th century, increased the role of credit. The need for production credit as well as consumption credit was most acutely felt by those involved in seasonal agricultural and fishing, livelihoods subject to the monsoons. Credit in past was provided by traditional sources, including rural money-lenders 7

8 AGRICULTURE AND DEBT ACCUMULATION Rural credit varied with the different requirements of those involved in rubber cultivation, rice farming, fishing, vegetable gardening or mixed farming. Although most seasonal loans were usually settled at harvest time, bad harvests and natural disasters cause debt carried over. Over time, this could cumulate and result in chronic debt. 8

9 RURAL INFORMAL CREDIT MARKETS Two broad types: 1. Traditional sector based kinship, community and religion, which generally stress helpfulness and generosity. 2. Economic and monetary to facilitate consumption, production and trade, based on family friends, shopkeepers, traders, landlords, farm labourers and moneylenders. 9

10 MORAL ASPECTS OF DEBT Receiver becomes indebted in two ways: 1. first, a debt to the economic value of the actual goods or services received and 2. second: a moral debt of gratitude. Hutang emas boleh dibayar Hutang budi dibawa mati. Formal systems have impersonalized debt relationships 10

11 RURAL FINANCE CAN IMPROVE HOUSEHOLD FOOD SECURITY Three ways to resolve transitory and chronic food insecurity: 1. Capital for financing inputs, labour and equipment for income generation. 2. Households can adopt more effective precautionary savings strategy. Insurance services can reduce the cost of bearing risks. 3. Financial services could stabilize consumption of food and other essential goods during lean times. 11

12 Demand for credit Pentingnya pasar kredit : 1. Kebutuhan modal saat awal berdiri atau expansi (pasar kredit utk investasi) 2. Untuk pembiayaan aktivitas produksi (pasar kredit utk working capital) karena lag antara kebutuhan input dan penjualan 3. Demand kredit untuk konsumsi

13 Who provides credit? 1. Institutional/ formal lenders (government banks, commercial banks, pegadaian, dll). Main problem: Lack of personal knowledge about characteristics and activities of the borrowers, leads to a problem in monitoring. To reduce the risk of default, banks ask for a collateral before advancing a loan. This means that formal credit becomes often an unfeasible option for poor borrowers

14 2. Informal lenders : e.g. landlord (tuan tanah), shopkeeper (kios), trader (tengkulak), but also cooperatives, credit unions (bank keliling), self-help organizations. They have better information regarding characteristics and activities of borrowers (closer to borrowers and interacting with them); Bersedia menerima jaminan yang tdk diterima lembaga formal.

15 Characteristics of rural credit markets 1. Informational constraints : Use : production or consumption Repayment decision (monthly or quarterly) depending on characteristics of borrowers; 2. Segmentation : perlu waktu utk hubungan personal, shg outsiders have more problems getting funds; 3. Interlinkages between credit transactions and other activities of the moneylender and of the borrower; lender is supplier and borrower is farmer

16 4. Interest rate variation : tgt geografi, sumber dana, karakteristik peminjam. 5. Rationing : existence of upper limits on how much a borrower receives from a lender at the going rate of interest; 6. Exclusivity : moneylenders typically dislike situations in which their borrowers are borrowing also from other sources.

17 Theories of informal credit markets Lender s monopoly: the very high interest rates that are sometimes observed may reflect the fact that the lender has exclusive monopoly power over his clients and can therefore charge a much higher price for loans than his opportunity cost. Problems: Empirically we don t have evidence of the existence of complete monopoly (at best local monopoly ); Theoretically, a high explicit interest rate is not always the best way to maximize profits (especially when inter linkages are present). Tingginya i profit belum tentu tinggi

18 Lender s risk hypothesis: lenders may be earning only enough to cover their opportunity cost. The higher interest rates may then be just the consequence of the substantial risk of default characterizing the rural credit markets. Types of default: borrower view s Involuntary; no choice Strategic : credit for consumption

19 In this model the relation between i (interest rate charged from moneylender) and p (probability of receiving capital and interests back) is the following ( r represents the opportunity cost): i=[(1+r)/p]-1 The informal interest rates will become higher the higher the probability of default (lower p). This takes us to study how do lenders manipulate and lower the default risk (potential default may be important, but actual default rates appear to be low).

20 The lender s risk hypothesis however does not take into account the fact that the probability of default is most likely dependent from the amount to be repaid. Larger amounts to be repaid may lead to a greater risk of default. large loans will have lower probability to be made; in some cases the loan will not be given irrespective (tdk berturut-turut) of the interest rate premium, as the premium itself may affect the chances of repayment

21 We can extend this line of reasoning also to the kind of use to which the loan will be put (e.g. loan to migrate to the city or to invest and reduce future need for credit). In presence of strategic default, the most likely use of loans will be for working capital or consumption purposes (less for fixed investment reducing future needs).

22 Default and collateral: the fear of default creates the tendency to ask for collateral (land, use rights to output, labor, etc.), whenever this is possible. The collateral is mainly of three types: 1. Valued highly from both the lender and the borrower; 2. Valued highly by the borrower but not from the lender. 3. Valued highly by the lender but not from the borrower.

23 Ad.1.Collateral valued highly by both parties has the additional advantage of covering the lender against involuntary default as well. However, if lender values the collateral MUCH MORE than the borrower, this may lead to excessively (sering) high rates of default (as the lender may design the contract in such a way that it would make default the best option for the borrower). Let s see this using a very simple model.

24 Let s consider the case of a small landowner borrowing from a big landowner and giving as collateral a plot of land adjacent (bersebelahan) to the land of the big landowner. The small landowner s evaluation of the value of the plot of land is V S while the value of the plot of land for the big landowner is V B. The borrower will prefer to return the loan if L(1+i)<V S +F The lender will prefer to have his money back if L(1+i)> V B

25 Loan repayment is in the interest of both parties only if V B < V s +F If the opposite (V B > V s +F) is true the lender would prefer the borrower to default. A possible mean of achieving this goal is to drive up the interest rate so that the borrower has an incentive to default.

26 This might help explaining the existence of land inequalities in poor societies. This analysis works much better for consumption loans than production loans, as consumption loans are usually characterized by a lower elasticity with respect to the interest rate (e.g. illness in the family). This is one possible explanation of why interest rates may be high for some types of credit transactions but not for others.

27 Interest rate i 2 i 1 Consumption credit demand L S L D Loans

28 Interest rate Default and credit rationing Credit rationing: situation in which at the going rate of interest in the credit transaction the borrower would like to borrow more, but is not permitted to by the lender. L S L D Loans

29 The possibility of default is intimately tied to the existence of credit rationing. This can be shown using a simple model. We assume that a moneylender wishes to maximize the rate of return on his funds. We also assume that, if the interest rate becomes too high [such that the expected farmer s surplus will go below a given value A alternative source of working capital] the farmer will borrow from another source. This can be written as: f(l)-l(1+i) A Participation constraint

30 Output, Costs, Profits Production Function f(l) A L(1+i*) L* Loan size Higher i*, higher L(1+i*), then lower A

31 When we introduce the potential default, another constraint has to be taken into account. If the farmer defaults, he will not be able to borrow from the lender anymore and will be forced to borrow from alternative sources, gaining A from that moment onwards. Assuming his time horizon is of N periods, his choice will be between earning N[f(L)-L(1+i)] without default or f(l)+(n-1)a for default NOT to occur we must have: N[f(L)-L(1+i)] f(l)+(n-1)a which becomes f(l)-(n/n-1)l(1+i) A Not default constraint

32 The no-default constraint is tighter than the participation constraint; The difference between the two cases will be larger (smaller) when N becomes smaller (larger). It is evident that if N becomes 1, no loans will be returned, so no loan will be advanced, while we go back to the previous case with N very large; We can see graphically how introducing this new constraint affects the equilibrium.

33 Output, Costs, Profits Modified cost line Production Function f(l) cost line A A [N/(N-1)]L(1+i**) L(1+i**) N/(N-1)(1+i**) is the marginal product of the loan for the moneylender 1+i** is the marginal cost of the loan faced by the borrower L** L^ If L or i were larger, then the no-default constraint would fail L** optimal size of the loan for the moneylender Loan size

34 At the going interest i**, the borrower would like to borrow more. However, the moneylender cannot increase the interest rate nor expand the loan (at the going interest rate) without increasing the likelihood of a default. This is because a higher loan or higher interest rates increase the returns associated to a default.

35 Informational asymmetries and credit rationing Lending risk may vary significantly from borrower to borrower (individual characteristics, landholdings, access to irrigation, type of crop, etc.) When the factors affecting risk are observable, the lender can select his clients and charge higher rates for higher-risk clients. However, when the lender is not able to distinguish between high and low-risk clients, the interest rate affects the mix of clients that are attracted and, consequently, the probability of default. In this case we might have a situation in which, at the going interest rate, there is an excess demand for loans that would make it possible for lenders to raise interest rates. However, lenders would not raise interest rates for fear of attracting too many high-risk customers.

36 Informal asymmetries and credit rationing: an example Two types of potential customers: safe type and risky type; Both need a loan of the same size, L; The safe type can always obtain a return of R>L; The risky type instead can obtain R >R>L but only with probability p. With probability 1-p he gets 0. If the lender can finance only one project, should he increase the interest rate in order to maximize his expected profits? Not necessarily. In fact, the safe type customer would be willing to take the loan only if i s =R/(L-1), while the risky type would be willing to pay i r =R /L-1 i r > i s

37 If the lender applies i s or below, both types of customers will apply, while if he applies a rate larger than i s, only the risky type will apply (knowing this, once moved away from i s, the lender should then move from i s to i r ) The expected of the lender in the two cases would be: r =p(1+i r )L-L s =1/2 i s L+1/2[p(1+i s )L-L] The lender will be reluctant to charge the higher interest rate when s > r.

38 Once the probability of repayment, p goes below a certain threshold, the lender will not raise his interest rate to i r (that would attract only the risky type) and will prefer instead to charge i s and take the chance of getting a safe customer. Of course, in this case one of the two customers will not receive the loan even though he would be willing to pay the going interest rate we have rationing. This time the cause of rationing is the fact that the higher interest rate would drive away the safe borrower, and the higher possible return would not be enough to compensate for the higher probability of default.

39 Default and enforcement The borrower s dealings with the moneylender must yield him a greater profit than he could get elsewhere by defaulting in order for him not to default. Algebraically: f(l**)-(1+i**)l**>f(l**)-(n/n-1)(1+i**)l**=a Where f(l**)-(1+i**)l** represent the borrower s profits incase of no-default and A represents the profits associated with the alternative source of financing. The probability of default tends to increase as the borrower gives less weight to the potential future consequences of a default (has lower N time horizon) or as the profits associated with the alternative source of financing (A) increase the lender will be able to charge higher rates without risk of default only if N is large enough or if A is small (getting the loan from alternative sources is costly).

40 How can lenders reduce the probability of default? By making it more costly for the borrower to get a new loan once he has defaulted, for example by building a system of reputations. This requires, however, the rapid spread of default information, which is not always a reasonable postulate. As a matter of fact the way information is distributed evolves with the evolution of a society. And we might think at this as following a U pattern: In traditional village societies with limited mobility, community networks are very strong and information circulates rapidly. At the other extreme, industrialized countries, credit histories are tracked on computer networks. Between these two extremes, we have societies in which mobility is increasing and traditional ties fall apart - together with informal information networks while formal networks have not been fully in place.

41 In situation where information about the previous behaviour of borrowers are not easily available, lenders have two sorts of reactions: Approaching new potential borrowers very carefully or not at all (Ray, box page 559); Devoting a lot of effort and money checking the new borrower s credentials (past history) as this information might help establishing whether the borrower is an intrinsically bad prospect. Collecting information about past history, however is worth only if we have two types of potential borrowers: cheaters and opportunists. If defaults come only from poorly specified contracts and not from the presence of cheaters, than there is no useful information to be found in the borrower s past history. On the other side, if we have no cheaters and there is no screening, the whole system will fall apart and no loans will be granted.

42 So screening efforts by a lender have large positive externalities on other lenders as they increase the cost of default for their borrowers. Unfortunately, as it happens in presence of positive externalities, the lenders do not take into account the existence of the externality and devote time and resources to screening only if they think this benefits them (as in the case when there are indeed some borrowers who are intrinsically bad risks). What is interesting is that, in this case, the credit market can work exactly because a market failure (lack of information about intrinsic types) helps solving another kind of information failure (lack of information on past defaults) by giving lenders an incentive to screen borrowers (collecting information or providing small test loans at the beginning of the relationship).

43 Interlinked transactions In developing countries it is common to find loan transactions linked with dealings in some other market, such as the market for labor, land, or crop output. To the extent that the lender can directly benefit from the assets owned by the borrower, this makes credit transactions easier to enforce. It does appear that in the event of coincident occupations, the interlinking moneylender has an edge over other moneylenders in credit dealings. In fact, in many parts of developing world the pure moneylender figure is disappearing.

44 There are several reasons why interlinkage is an observed mode of credit transactions: Hidden interest: in some societies the explicit charging of interest is forbidden or shunned e.g. in Islamic societies under the Shaariat law. In these cases the loan results as interest-free and the interest is paid in secondary form; Interlinkages and information: thanks to an interlinked bargain, the lender can dispense with some of the costs of keeping track of the activities of the borrower (e.g. trader with crops and landlord with labor); Interlinkages and enforcement: used sometimes to prevent strategic default ( one carrot used as two sticks e.g. landlord and tenant/borrower. Surplus to ensure effort now also to prevent default).

45 Interlinkages and creation of efficient surplus: interlinkages may arise also in order to prevent distortions that lower the total surplus available to be divided between lender and borrower. We are going to analyse two cases. Loan repayment in labor; Loan repayment in output;

46 Peak consumption Peak consumption Loan repayment in labor (Example of consumption smoothing) Higher interest (less slack consumption) Lower wage Lower interest Lower wage W W W* Risk-free interest rate B A L(1+i*) A L(1+i) W* L(1+i) B C C L(1+i*) L Slack consumption L Slack consumption Lender s return from a contract (w*, i*)

47 Peak consumption W If the lender can find a different combination of w and i keeping the borrower on the same utility curve while increasing his own utility, we can say that he will have found a contract that dominates the previous one. W* A B A L(1+i) In our case, the contract maximizing lender s utility while keeping constant the borrower s utility is leading to a surplus A C >AC. W C C U The result is optimal as there is no more distortion in the size of the loan (L >L) because i is now = to the true opportunity cost for the lender. L(1+i*) L L Slack consumption

48 Peak consumption W W A B A L(1+i*) L(1+i) In this (second) case, the contract maximizing lender s utility while keeping constant the borrower s utility is leading to a surplus A C >AC. W* C C U The result is optimal as there is no more distortion in the size of the loan (L <L) because i is now = to the true opportunity cost for the lender. L L Slack consumption

49 Value of output Loan repayment in ouput (Financing working capital) Production function If the borrower could borrow directly at the same opportunity cost of the lender/trader, he/she would ask for a loan of size L B S C D L* L Real cost line Opportunity cost line Loans However, as this is not possible, he/she will have to contact a lender. Assuming the lender/trader chooses a pure credit contract, he/she will charge a higher interest rate (between the opportunity cost and the cost that would be paid by the borrower on the market) and pay the market price of rice. In this case, however, the borrower would ask for a loan of size L*<L Profits (BC) will be lower than in the optimal case (S ). The only way to reap the additional surplus for the lender is by charging i and paying p< market price, reducing marginal cost of production and the price in the same proportion so that the ratio remains unchange relative to p/(1+i).

50 Loan repayment in ouput, cont. (Financing working capital) Algebraically, we can write the maximum surplus as: S =pq -(1+i)L If we impose a profits tax of t per dollar on the combined operation and set t such that ts =A (where A is profit the borrower makes financing working capital on the credit market), we obtain: ts =ptq -(1+i)tL We can define then p pt and i such that 1+i =(1+i)t. p <p and i <i. Now, if the borrower tries to maximize profits, he will not change the quantity of L he demands. The borrower will still earn A (outside option) while the lender will earn S -A. In this case the interlinked contract is optimal and involves lower buying price coupled with a lower interest rate (lower even than the true opportunity cost for the lender).

51 Alternative credit policies The needs of rural credit cannot be adequately served with the use of large financial institutions such as commercial banks, mainly because of the micro information needed for these operations. In response to this observation one can adopt two kinds of policies: 1. using the informal lenders to grant and recover loans from small borrowers creating vertical formal-informal links and expanding formal credit to informal lenders; 2. Design credit organizations at the microlevel that will take advantage of local information in innovative ways.

52 Alternative credit policies: vertical formal-informal links Expansion of formal credit to informal lenders (landowners, traders, cooperatives, etc.) instead of attempting to displace one system by the other. Does this lead to increased competition between lenders and better conditions for borrowers? Not necessarily, because of: Costs of monitoring (increased number of outside options lower cost of default for the borrower increased administrative/monitoring costs also the equilibrium interest rate might increase); Collusion (if the relation between competing lenders can be described as a strategically sustained cooperative gain there are several reasons easier to detect deviations if scale of operations increases, larger losses in case of retaliation, potentially diminishing marginal gains to think collusion may actually increase); Differential information (increased siphoning off of best borrowers may drive out of the market the lenders with less information).

53 Alternative credit policies: microfinance Another possibility is that institutional lenders mimic and try to exploit some of the features of informal lending. For example acting both as lender and as miller or moving from individual to group lending. In fact, in cases when it is impossible for financial institution to provide credit to an individual, it may be possible for the same institution to give credit to a group. The most famous example of an institution engaged in group lending is that of the Grameen Bank in Bangladesh.

54 The Grameen style lending 1. These loans are intended for clients too poor to meet the wealth requirements of the formal banking (mostly women). 2. Loans are given to small groups of borrowers; 3. Requires borrowers to keep savings in the bank; 4. Borrowers are jointly liable for the loans granted to each member of their group; 5. Joint liability of the group members is the only guarantee on the loan (no collateral). 6. In the event of default, no group member is allowed to borrow again. [This gives an incentive to the members of the group to monitor (and to put pressure on) each other in a way that would be impossible for the lender.]

55 The Grameen style lending, cont. The literature usually identifies three main advantages of group lending: Reduction of transaction costs (direct costs of lending vary inversely with the size of the loan); Working in groups helps poor people having access to more credit (financial reason) and getting access to the training and organizational inputs they need; Repayment rates are more favorable in group lending. These benefits are due mainly to positive assortative matching (a form of self-selection) and peer pressure. This assumes that borrowers have more information about each other and are better at monitoring each other s activities than the lender could be. If this is true: 1) riskier borrowers could be driven out of the market (safe borrowers would not accept them in their group); 2) there will be pressure for peer monitoring to lower the level of project riskiness (as there is joint liability).

56 The Grameen style lending, cont. Moreover, group lending has been found to offer also nonpecuniary returns (e.g. self-esteem, mutual trust, empowerment, self-discipline, encouraging socialintermediation etc.), especially to women (Berenbach and Guzman, 1993). There are also potential drawbacks associated with group lending: Increases the incentives to generalized default when one member incurs into genuine difficulties. This however does not affect the Grameen style lendings as loans are given sequencially; Peer monitoring may lead to a socially inefficient (too low) level of riskiness (and profitability) of the projects adopted. Group-based schemes may lack flexibility worst-performing member slowing down the group.

57 Evaluation of microcredit programs Several studies have analyzed the effectivenes of microcredit programs, particularly the case of the Grameen Bank. Analysing a microcredit program however is complex, for several reasons: First of all, microcredit programs can be judged along several dimensions (sometimes conflicting with each other).for example: Sustainability vs. outreach (Snow and Buss, 2001) Sustainability: earning sufficient revenues to cover all costs, mobilize local savings and involve its members in governance. This assures the institution is not vulnerable to the withdrawal of donor support. Outreach: getting help to the poor. If doing it as soon as possible is the top priority, rapid outreach desirable. This is only possible if it is available a cash flow sufficient to increase the loan portfolio. As generating the cash needed for expansion from operations alone is difficult in microcredit programs, donor capital is required.

58 Evaluation of microcredit programs, cont. Secondly, even if we focus on the capacity of the programs to improve the condition of the poor (outreach), the analysis is made more complex by the fact that the evaluation requires to compare the final outcome (in time T) NOT with the situation of the beneficiaries ex-ante (in time T-n), BUT with the situation of the same beneficiaries in time T, had they not participated. There are reasons to believe that, failing to compare the final outcome with its counterfactual, but just with the situation of the participants before entering the program may lead to overestimates (selection of good borrowers ) or underestimate (targeting the poor maybe with worse unobservable characteristics) the effectiveness of the program. Moreover, analyzing a microcredit program requires taking into account of both pecuniary and non pecuniary aspects.

59 Lessons: There is at least a study (Pitt and Khandker (1996), for example) who managed to show that the Grameen Bank was effective in raising household income of participants vis-à-vis non participants (even without considering non monetary aspects). Of course there has been a number of cases in which microcredit programs have been implemented (see Hassan, 2002 and Snow and Buss, 2001 for some references), apart from the case of the Grameen case. In some of these cases microcredit programs were successful while in other they were not. This leads us to the conclusion that microcredit programs have to be tailored to the needs of the countries where they are implemented to be effective.

60 Lessons, cont. Risks extend beyond merely wasting resources. Other potential risks involve (Snow and Buss, 2001): microcredit lending being harmful to borrowers (e.g. potential debt trap); new aid dependence, different from the dependence on large project aid countries experienced in the 1960s and 1970s, but still damaging; at least some microcredit programs distort capital markets as other interventions. Microcredit programs may absorb resources needed for education and infrastructure development. To avoid waste of resources or worse, governments and donors, ought to have specific (and explicit) goals for microcredit programs. This will help to measure and evaluate correctly their outcomes and to improve them.

61 Terima Kasih