Credit Market Problems in Developing Countries

Transcription

1 Credit Market Problems in Developing Countries November 2007 () Credit Market Problems November / 25

2 Basic Problems (circa 1950): Low quantity of domestic savings major constraint on investment, especially in manufacturing Dualistic credit market,! formal sector commercial and government run banks serving urban sector and large landowners,! informal sector small scale lenders and agricultural cooperatives serving small scale rural borrowers and the informal urban sector Skewed distribution of credit access a source of persistent and widening wealth disparity. () Credit Market Problems November / 25

3 The Development Planning Approach Dominant paradigm (1960s / 70s) Monopolistic moneylenders, esp. in rural areas Lack of nancial depth,! low savings due to uncertainty regarding potential bank failure and ine ective legal systems,! lack of alternatives nancial assets catering to varying needs of potential savers/lenders. () Credit Market Problems November / 25

4 Policy Strategy: put the moneylender in his place,! development banks to lend at low rates to rural borrowers initially ll savings gap with tax revenue and foreign loans/aid Eventually banks would become self nancing BUT,! failed to drive out informal moneylenders,! development banks did not become nancially viable,! access to formal credit remained skewed towards large landowners () Credit Market Problems November / 25

5 The Chicago School Approach Informal sector is competitive, but risky and costly due to,! seasonal activity,! systemic risk,! geographically dispersed customers,! absence of collateral land is often untitled,! di culty of enforcing repayment ) lack of information re ects acquisition costs ) high interest rates re ect risk not monopoly power () Credit Market Problems November / 25

6 Lender s risk hypothesis: Pro t per $ loaned = p(1 + r) (1 + i) where r = lending rate i = marginal cost of funds p = probability of repayment,! Competition ) zero economic pro ts,! risk adjusted interest rate r = 1 + i p 1 () Credit Market Problems November / 25

7 Interest Rate r d S(r) r * r I(r) S(r) I* S(r d ) Savings, Investment Figure: Implications of Interest Rate Ceiling () Credit Market Problems November / 25

8 Policy implication: government intervention is bound to fail,! removal of interest rate ceilings BUT,! information is a public good social bene ts exceed private bene ts,! localized information limits competition,! interest rates often negatively related to risk,! no explanation for other institutional characteristics () Credit Market Problems November / 25

9 Example: Credit Markets in Chambar, Pakistan Irfan Aleem (1990) Informal sector = 75% of rural lending Mean Interest Rates: Formal = 12%, Informal = 79% High degree of duality Non specialization by lenders Interlinking of loan and commodity contracts Informal sector: no collateral, but low default rates Formal sector: collateral but high default rates. () Credit Market Problems November / 25

10 New Institutional View Credit market transactions are typically incomplete institutions: private sector response to market failure signi cant entry, BUT,! informational constraints,! market segmentation,! local market power ) monopolistic competition role for government, but must recognize informational disadvantages () Credit Market Problems November / 25

11 Agency Problems Reasons for absence of formal credit in village economies A result of limited liability (lack of collateral) and asymmetric information Even when titled land is available, formal banks may not accept it as collateral Two main rationales for intervention,! E ciency: are productive investments not being undertaken?,! Distribution: is access to credit equitable?,! need not be a trade-o () Credit Market Problems November / 25

12 Adverse Selection Example Investment requires $1, but borrowers have no wealth A fraction q of borrowers are safe : earn certain output y A fraction 1 q of borrowers are risky : ȳ with probability p Output = 0 with probability 1 p Bank cannot distinguish borrower types Equal expected return: pȳ =y. Gross cost to bank per $1 lent = k, where y > k Bank charges gross (interest plus principle) lending rate = R () Credit Market Problems November / 25

13 How does the bank s expected pro t vary with R? Given R, the bank s expected return per dollar lent is [q + (1 q)p] R De ne the break-even value of R as R b [q + (1 q)p] R b = k R b = k q + (1 q)p (1 q)(1 p)k R b = k + q + (1 q)p R b = k + A Bank s expected pro t: [q + (1 q)p] R k if R < y π = pr k if R > y () Credit Market Problems November / 25

14 π 0 k k+a y k/p y/p R () Credit Market Problems November / 25

15 π 0 k y k+a k/p y/p R () Credit Market Problems November / 25

16 Implications Raising interest rates need not increase pro ts linearly If p falls, the bank may not be able to break even at a rate low enough for safe borrowers ) Banks will only serve risky borrowers,! this is ine cient (since y > k) and inequitable,! credit rationing () Credit Market Problems November / 25

17 Moral Hazard Example (Ghosh, Mookherjee and Ray 2000) Indivisible project requires funds L All agents risk neutral Likelihood of high output depend on non-contractible e ort e : Q if good harvest with prob. p(e) Output = 0 if crop failure with prob. 1 p(e) where p 0 (e) > 0, p 00 (e) < 0 () Credit Market Problems November / 25

18 Case 1: Self nanced farmer (benchmark) Farmer chooses e to solve max e p(e).q e L,! FOC p 0 (e ) = 1 Q where e = rst best (e cient) e ort level () Credit Market Problems November / 25

19 Case 2: Debt nanced farmer Total debt: R = (1 + i)l Limited Liability: In case of default can only lose collateral, w < L Farmer chooses e to solve max e,! FOC yields the incentive curve: p(e).(q R) + (1 p(e)).( w) e p 0 (ê) = 1 Q + w,! ê(w, R) is decreasing in R and increasing in w,! if R > L > w, it follows that R (IC) ê < e () Credit Market Problems November / 25

20 Lender s pro t: π = p(e)r + [1 p(e)] w L Competitive lending (Figure 1) ) π = 0 : R = L w p(e) + w = L (1 p(e))w p(e) (ZP),! equilibrium: determined by intersection of (IC) and (ZP) Monopoly lending ) maximize pro ts subject to IC (Figure 2),! if max pro t is π M then isopro t line is R = L πm (1 p(e))w p(e) (MP) () Credit Market Problems November / 25

21 R Isoprofit Curve Incentive Curve R^ E e^ e* e Figure 1: Equilibrium Debt and Effort in the Credit Market.

22 R Isoprofit Curve Incentive Curve E' R^ E e^ e* e Figure 2: Effect of an Increase in the Lender s Profit.

23 An increase in collateral, w (Figure 3),! fall in the equilibrium interest rate and debt, and an increase in the e ort level.,! for a xed π, the borrower s income increases ) interest rate dispersion, even in competitive credit markets ) exaccerbates pre-existing inequality () Credit Market Problems November / 25

24 R E Isoprofit Curve E' Incentive Curve e Figure 3: Effect of Higher Borrower Wealth (Collateral).

25 Enforcement Problems (Ex Post Moral Hazard) Example (Ghosh, Mookherjee and Ray 2000) In nite horizon lending borrowing game with no saving and discount factor δ Borrower s production technology is F (L) where F 0 (L) > 0 and F 00 (L) < 0 r = bank rate of interest (opportunity cost of funds) Self nanced farmer (benchmark): max L F (L) (1 + r)l,! FOC: F 0 (L ) = 1 + r () Credit Market Problems November / 25

26 Partial Equilibrium: Single Lender Assume that a defaulting borrower goes into autarky and receives v (exogenous for now) Borrower s incentive constraint: F (L) + Optimal contract solves δ 1 δ v 1 [F (L) R] (IC) 1 δ max L,R F (L) R subject to where z is lender s minimum pro t R δ [F (L) v] (IC) z = R (1 + r)l (PC) () Credit Market Problems November / 25

27 Two types of solution: (1) if indi erence curves are tangent to isopro t curve between A and B at L, this is the solution and IC is not binding (2) if not, both constraints are binding and solution ˆL(v, z) is at corner B,! in general, the constrained optimal level of lending is L(v, z) = min L, ˆL(v, z) () Credit Market Problems November / 25

28 R R ~ Borrower's indifference curve B Incentive constraint Isoprofit line A ~ L L * L Figure 4: Optimal Solution to the Enforcement Problem.

29 Some Implications Increase in lender s pro t z! PC shifts up (Figure 5),! reduced lending, L,! increase in interest rate, R/L Increase in value of outside option, v! IC shifts down (Figure 6),! reduced lending, L,! increase in interest rate, R/L Can credit rationing arise in equilibrium?,! yes: if z or v become high enough there is no combination of L and R that satis es both (IC) and (PC) () Credit Market Problems November / 25

30 R Incentive constraint Isoprofit line L Figure 5: Effect of an Increase in Lender s Profit.

31 R Isoprofit Line B Incentive Constraint B' L Figure 6: Effect of an Increase in Borrower s Outside Option.