http://www.independent.co.uk/news/uk/crime/legal-selection-system-is-corrupt-admit-judges-706668.html 

          

Judges and senior lawyers admit that the system under which they are appointed is riddled with corruption and open to widespread abuse.

Judges and senior lawyers admit that the system under which they are appointed is riddled with corruption and open to widespread abuse.

Shetreet v SHARP and Lovells / Hogan & Hartson NHS and UK Attorney General v Shetreet

http://www.independent.co.uk/news/uk/crime/draconian-judge-says-riot-sentences-were-fair-7855270.html 

Corrupt judges jailed web girl

Judge ... Michael T Conahan

 

STAFF REPORTER

Published: 19 Feb 2009

http://www.thesun.co.uk/sol/homepage/news/2255203/Corrupt-judges-jailed-web-girl.html

A MODEL student was sent to a detention centre by corrupt judges – for building a spoof MySpace page.

Hillary Transue thought she might get a stern lecture when she appeared before a judge in 2007 for building the webpage mocking her headteacher.

Incredibly the judge sentenced her to three months at a juvenile detention centre on a charge of harassment.

But today the judge, Mark A Ciavarella Jr and his colleague, Michael T Conahan, pleaded guilty to taking more than $2.6million in kickbacks to send teenagers to two privately run youth detention centres.

Prosecutors said that Judge Conahan, 56, secured contracts for the two centres to house juvenile offenders.

Then Judge Ciavarella, 58, carried out the sentencing to keep the centres filled.

Hillary, 17, said: “I felt like I had been thrown into some surreal sort of nightmare.

“All I wanted to know was how this could be fair and why the judge would do such a thing.”

 

Corrupt judges, credit rationing and the political economy of bankruptcy laws

 

Bruno Biais, Universit´e de Toulouse and CEPR, and Gilles Recasens, Universit´e de Reims1

September 2002

1Many thanks to Catherine Casamatta, Roman Inderst, Steven Kaplan, Jean Jacques Laffont, Thomas Mariotti, David Martimort,

Marco Pagano, Howard Rosenthal, Klaus Schmidt, Andrei Shleifer, Jean Tirole, and seminar participants at Toulouse

University and the CEPR Conference on The Firm and its Stakeholders, for very insightful discussions and comments.

ABSTRACT

The liquidation of distressed companies entails social costs, which lenders and managers do not fully internalize.

To mitigate this problem, bankruptcy laws in France and the US (in contrast with the UK or Germany)

favor reorganization. The corresponding expropriation risk faced by creditors worsens credit rationing, especially if bankruptcy judges are corrupt. Consequently tough bankruptcy laws, involving strict enforcement of debt contracts, can be socially optimal. Soft laws are likely to emerge, however, when the majority of citizens are so poor they would be credit rationed even under a tough bankruptcy law. In contrast, tough laws can be chosen when pivotal voters are middle class citizens who benefit from enhancing entrepreneurial opportunities.

 

Corrupt judges, credit rationing and the political economy of bankruptcy laws

 

1 Introduction

Why do we need laws? Laws can provide a useful framework for the enforcement of contracts. In presence of externalities, however, laws can also be useful to limit ex–ante the set of feasible contracts, or interfere ex–post

in their application. With benevolent legislators and honest judges such interference enhances social welfare.

In contrast, when there is corruption it can lead to severe distorsions. The present paper analyzes these issues in the context of bankruptcy laws.

Bankruptcy laws vary quite significantly around the world.2 In the UK and Germany, their main objective is nto enforce debt contracts. In contrast, in the US, France, and Russia, bankruptcy courts can violate contractual clauses, and impose firm reorganization and debt write–offs. Franks and Sussman (1999) and Berglof and Rosenthal (2000) offer very interesting analyses of the historical developments that led to the English and US

bankruptcy laws. They note that the latter was greatly influenced by the public reaction to large railroad bankruptcies in the late nineteenth century. At that time it was widely felt that liquidation of the railroads would go against the public interest.

 

While debtor oriented bankruptcy laws can be ex–post socially optimal, they can have adverse effects ex– ante. When anticipating that their rights as creditors risk to be violated, bankers are reluctant to grant loans.

This can result in credit rationing. Indeed La Porta et al (1997, 1998) find that in countries where creditor rights are well protected, such as the UK or Germany, firms rely extensively on debt financing, which is reflected in large debt to GNP ratios, while in countries where creditor rights are weak, debt financing is more limited. In the face of these results, La Porta et al (1997 and 1998) ask why it is that the law can define such weak

2 See for example, Franks, Nyborg and Torous (1994), White (1994) and Atiyas (1995).

1 creditor rights. From a standard public economics perspective one answer could be that optimal bankruptcy laws trade–off credit rationing with social costs of liquidation. If coping with credit rationing is more important for social welfare than limiting the social costs of liquidation, then the optimal bankruptcy law should be tough,

i.e., it should insist on liquidation whenever debt is not serviced. Alternatively, if the social costs of liquidation are relatively large, the law should besoft, i.e., it should allow for reorganization, to keep the firm in business, even if it implies some violations of creditors’ rights. When one looks at the darker side of human nature, however, the picture becomes a little more blurred.

First, bankruptcy judges may not be honest and benevolent social welfare maximizers. Unfortunately there are numerous cases of bankruptcy judges’ corruption. With corrupt judges soft laws generate credit rationing as well as deviations from socially optimal liquidation decisions ! Second, the laws actually voted may deviate bfrom those which are socially optimal, as median or pivotal voters do not internalize the welfare of all citizens.

The goal of this paper is to analyze the political process through which bankruptcy laws are chosen and its consequences for economic efficiency and social welfare.

We build from a simple corporate finance model `a la Holmstrom and Tirole (1997). Entrepreneurs with investment projects need outside financing to fund these. As they must exert costly but unobservable efforts to make the project profitable there is a moral hazard problem. The latter can generate credit rationing for entrepreneurs with initial wealth below a certain threshold.

We extend this model by assuming that liquidation can create social costs that are not internalized by managers and lenders, and by considering two possible bankruptcy laws. Under the tough law, firms are liquidated whenever they cannot service their debt. Under the soft law, judges decide if firms are liquidated when they cannot serve their debt. Some judges are honest and make liquidation decisions to maximize social welfare, other judges are corrupt and bribes will influence their rulings.

We embed this analysis into a slightly more general model where a continuum of citizens have identical investment projects, but different initial wealth.3 Citizens with initial wealth below a certain threshold are

3This is similar to Biais and Casamatta (1999).

2 credit rationed. In an extension of our analysis we enrich this setting by considering a simple general equilibrium model where equilibrium is jointly determined on the labor market and the credit market. Thus we are able to analyze the consequences of bankruptcy laws on investment and on wages, reflecting the increased labor demand generated by the creation of new firms.

Within this framework, we analyze the emergence of bankruptcy laws resulting from the votes of this population of agents.

Our analysis generates the following insights:

² The adverse effect of soft laws on access to financing is worsened by judges’ corruption. Tough laws do not grant judges any discretion regarding liquidations. Hence there is no scope for bribery. With a soft law, in contrast, the judge has discretion over the liquidation decision. In this context corrupt judges extract bribes. These play the role of a tax on creditors, reducing the return they can expect from their loans, and thus their willingness to fund projects. When the judicial system is corrupt, it can be preferable to

opt for tough laws, even if it means inefficient liquidations.

² Yet, soft laws are likely to emerge in democracies when the majority of the citizens are so poor that they are credit constrained, whatever the bankruptcy code. While poor citizens suffer from the social costs of liquidation generated by tough laws, the extent to which they benefit from enhanced financing opportunities is limited. Correspondingly they are not very sensitive to the adverse effect of soft laws or judges’ corruption on credit rationing.

² In contrast, tough laws are more likely to emerge in democracies where the pivotal voters are middle class citizens benefitting from enhanced entrepreneurial opportunities. In this case, access to financing is facilitated, which spurs investment and growth.

² Furthermore the determinants of the softness of bankruptcy laws are correlated with the business cycles.

In slumps, the social costs of bankruptcy are likely to be particularly large and soft laws are more likely to be passed than in booms.

3 Our paper builds on the substantial literature analyzing the design of bankruptcy laws (see e.g. Harris and Raviv, 1993, White, 1989, Bebchuck, 1988, Aghion, Hart and Moore, 1992, Berkovitch, Isarel and Zender, 1997, and Berkovitch and Israel, 1999.) Relative to this literature, our contributions are i) to analyze the differences between (bankruptcy) laws and (financial) contracts arising in presence of externalities, ii) to endogenize the bankruptcy law as resulting from an electoral process, iii) to study the consequences of bankruptcy judges’

corruption, and iv) to delineate the impact of the law on social welfare (in particular through credit rationing and social costs) and on financing choices.

Our political economy approach is in the line of the insightful paper by Bolton and Rosenthal (1999).

Some of the major differences between our paper and theirs include the following: In their analysis voting on moratoria occurs ex–post. In ours citizens vote for the bankruptcy law ex–ante, and then financial contracts are written and economic decisions taken, reflecting the legal context. Also, we emphasize the role of (possibly corrupt) judges, while Bolton and Rosenthal (1999) emphasize more the legislative process. Finally, their focus on how bankruptcy laws complete contracts by making their application contingent on macro–shocks, differs from our focus on how laws take into account externalities imposed on third parties by the parties of financial contracts.

Our analysis of the consequences of bankruptcy laws on access to financing is also in the line of the law and finance body of research, initiated by La Porta et al (1997, 1998). Our paper proposes to push this research nagenda one step further, by studying the political economy of the emergence of legal systems, and offering a rationale for their imperfections.

The next section presents institutional features of bankruptcy procedures which motivate our analysis.

Section 3 analyzes corporate financing choices with a tough law. Section 4 analyzes the case of a soft law.

Section 5 analyzes the socially optimal law and studies which law results from voting. Section 6 presents

extensions of our basic model. Section 7 offers a brief conclusion. Proofs not given in the text are in the

Appendix.

4

2 Institutional background on bankruptcy laws in different countries

 

Bankruptcy laws vary considerably across countries.4

As shown by the very interesting historical analysis of Franks and Sussman (1999), “the English procedure was developed by lenders and borrowers, exercising their right to contract freely.” It was left to the parties of debt contracts to determine their mutual obligations. “The role of the state in this process was relatively limited, largely confined to enforcing the contract”. In line with this historical evolution, the current UK bankruptcy code emphasizes the protection of creditor’s rights.5

Similarly to the UK law, the German law emphasizes the protection of creditors’ rights (see Kiefer, 2000).

In most cases, when companies default on their debt repayment obligations, they end up liquidated, and the proceeds are distributed to the debtholders.

In contrast, as explained by Franks and Sussman (1999), the US constitution gave Congress large powers to create bankruptcy laws resulting in interferences with the application of contracts. The US law took a decisive turn towards the end of the nineteenth century, when very important railroads companies failed. As explained by Franks and Sussman (1999): “It was largely felt that the lenders’ liquidation rights stood in conflict with the public interest”. Similarly, Berglof and Rosenthal (2000) note that: “It was argued that the liquidation of the

railroads would lead to significant costs for the US economy, e.g., cutting off the West from important supply lines.” In this context, as noted by Franks and Sussman (1999), “the Federal Courts innovated new procedures to preserve the railroads sometimes in blunt violation of pre–contracted agreements.” The current US law, in particular the Chapter 11 procedure, can be used to maintain firms in operations, even when creditors do not agree. For example, in the US, if creditors disagree with the reorganization plan, the judge can decide to use the “cram down” procedure to implement the plan in spite of their opposition.6

4 See for example, Franks, Nyborg and Torous (1994) for a comparison of the US, UK and German insolvency codes. White

(1994) and Atiyas (1995) also offer interesting international comparisons.

5Franks and Sussman (2000) offer an empirical analysis of the workings of the bankruptcy process in the UK.

6Franks and Torous (1989 and 1994) offer an empirical analysis of the workings of the bankruptcy process in the US, and Fisher

and Martel (1995, 1999, 2000) compare it to its Canadian counterpart.

5

The French bankruptcy law goes even further than the US law as regards the violations of creditors’ rights (see Biais and Mal´ecot, 1996). Its first stated objective is to save failing firms and avoid laying off workers. To reach this goal, judges enjoy large discretionary powers. If, based on their analysis of the firm and its social context, they feel that keeping the firm in operation is essential, they can unilaterally write–off the creditors’ rights. The French law, which was voted in 1985 by the socialist majority in parliament, reflects the popular feeling that other stakeholders than the creditors are concerned by bankruptcy procedures, and that judges should aggregate the different preferences of the different parties, to implement socially optimal decisions.

When that law was voted, very severe industrial restructurings were taking place in France (for example in the steel and coal industries in the North and the East of the country) and the social costs of liquidation were quite visible to French citizens.

This very brief comparison of bankruptcy procedures in four major industrial economies suggests that two different philosophies can underly these laws. On the one hand, bankruptcy laws can be designed to enforce the contract between two parties: the creditors and the debtors ; this is the approach taken in Britain and Germany. We refer to such laws as tough. On the other hand, bankruptcy laws can be designed with a view at taking into account the welfare of other parties, on which the application of the debt contract could have

external effects. In this second approach, taken by the US and France, the judge is given discretionary powers, and is allowed by the law to violate the contractual rights of the creditors. We refer to such laws as soft.

Note however that such discretionary powers can enhance social welfare only if judges are benevolent.

Unfortunately, there has been ample recent evidence in France that some judges use their powers in order to obtain bribes, rather than to maximize social welfare. A recent investigation, undertaken by the French Parliament, uncovered major dysfunctionings in bankruptcy courts.

Lambert–Mogilianksy, Sonin and Zhuravskaya (2000) offer an interesting analysis of the Russian bankruptcy law. Similarly to the American Chapter 11 procedure and the French bankruptcy law, Russian courts have significant discretion in bankruptcy procedures. As noted by Lambert–Mogiliansky, Sonin and Zhuravskaya (2000): “The judge does not need to follow the creditor’s request. This clause in the law was motivated by the fact that creditors may opt for inefficient liquidation.” The analysis of Lambert–Mogiliansky, Sonin and

6

Zhuravskaya (2000) suggests that corruption of Russian bankruptcy courts is rather frequent.

 

3 Corporate financing with a tough bankruptcy law

3.1 Model

Consider a continuum of risk-neutral entrepreneurs (also referred to hereafter as “managers”). Each entrepreneuri has access to an investment project, requiring initial investment I. While all the investment projects are identical, the entrepreneurs differ in terms of their initial wealth, Ai. The total mass of the population of entrepreneurs is normalized to one. For simplicity, assume there are only three types of entrepreneurs:

the rich (with initial wealth: Ar), the middle class (with initial wealth: Am), and the poor (with initial wealth:

Ap), where Ar > Am > Ap. The masses of the three categories are ¹r; ¹m and ¹p respectively. The average initial wealth of the population is: E(A) = ¹rAr+ ¹mAm + ¹pAp. To undertake the investment project, the manager with initial wealth Ai needs to raise outside funds: I ¡Ai. Competitive risk neutral outside financiers are willing to lend as long as they break even in expectation. For simplicity their required rate of return is normalized to 0, and, as they are competitive, their participation constraint is saturated. If the investment is undertaken, the project can yield payoff R or 0. If the manager exerts effort, and incurs disutility e, then the

probability that the payoff is R is ph; while if she does not exert effort, the probability of success is lowered to pl. We assume effort is unobservable, which raises a moral hazard problem between the entrepreneur and the outside financier. We also assume there is limited liability. So far, our framework is directly inspired by Holmstrom and Tirole (1997). In the remainder of the paper we build on this basis, adding ingredients to model bankruptcy laws and their social and political environment.

After the cash flow (R or 0) is obtained, the firm can continue to operate or be liquidated. In the latter case, liquidation proceeds L are obtained. For simplicity we do not model explicitly the case where the firm is maintained in activity. We simply assume that in this case, the manager obtains non–transferable private benefits

B. These can be thought of as reflecting the psychological satisfaction of the entrepreneur. Alternatively, one can think of B as the non–pledgeable rents the manager could earn from continued operations. We assume

7

that B > L, to focus on the case where liquidation is inefficient. Indeed, soft bankruptcy laws, which we want to analyze, are often justified as a way to avoid inefficient liquidation.

We also assume that the project has positive net present value if and only if i) the manager exerts effort, and ii) the firm is not liquidated except possibly in the bad state:

phR + B ¡ e ¡ I > ph(R + B) + (1 ¡ ph)L ¡ e ¡ I > 0

> ph(R + L) + (1 ¡ ph)B ¡ e ¡ I > phR + L ¡ e ¡ I:

3.2 The first best

In this context, the first best is to undertake the project, and never liquidate it. Denote:

S1 = phR + B ¡ I ¡ e;

the surplus created by the project in that case.

Is that outcome incentive compatible? The corresponding contract would involve a monetary transfer T to

the outside financier in the good state and nothing in the bad state. In this case the incentive compatibility

condition of the entrepreneur is:

ph(R + B ¡ T) + (1 ¡ ph)B ¡ e > pl(R + B ¡ T) + (1 ¡ pl)B:

The participation constraint of the outside financier is:

I ¡ A

ph

· T:

Combining the two conditions, the first best can be implemented if and only if:

Ai ¸ I ¡ phR + ph

e

ph ¡ pl

:

Denote AE the right hand side of this inequality. It states the standard result that, with moral hazard and limited liability, agents with wealth below a threshold are credit rationed. Note that, the minimum initial wealth constraint is tightened as the cost of effort (e) increases.

8

3.3 The second best

If Ai < AE the first best cannot be implemented. Is it possible to reach the second best outcome, whereby the firm is liquidated in the bad state? Denote:

S2 = phR + phB + (1 ¡ ph)L ¡ I ¡ e;

the surplus created by the project in the second best.

In that case the incentive compatibility condition of the entrepreneur is:

ph(R + B ¡ T) ¡ e > pl(R + B ¡ T):

The participation constraint of the outside financier is:

I ¡ Ai ¡ (1 ¡ ph)L

ph

· T:

Combining the two conditions, the second best can be implemented if:

Ai ¸ I ¡ phR + ph

e

ph ¡ pl

¡ (phB + (1 ¡ ph)L):

Denote AT the right hand side of this inequality.

Note that AT = AE ¡(phB+(1¡ph)L). Liquidation in the bad state relaxes the condition under which the project can be financed, because it relaxes the participation constraint (by promising liquidation proceeds to the financier) as well as the incentive constraint (by threatening the manager to deprive him from his private when cash flows are low).7 Hence the minimum level of wealth needed to obtain outside financing is lower

when the firm is liquidated in the bad state than when it is never liquidated.

Although liquidation is ex–post inefficient, this contract cannot be renegotiated. The manager has no cash in the bad state, and thus cannot pay the financier to avoid liquidation.

 

7This is not unlike in Hart and Moore (1994, 1998), Bolton and Scharfstein (1990), and Bolton and Rosenthal (1999).

9

3.4 Debt, equity and credit–rationing

The two contracts, implementing the first and second best outcomes respectively, can be interpreted in terms of financial contracts widely observed in practice:

² In the first best contract, which is feasible when Ai ¸ AE, the firm is never liquidated, and cash flows are split between the owner manager and the outside financiers. This corresponds to the case where the firm obtains outside financing by issuing outside equity (without relinquishing control, either because the entrepreneur keeps the majority of the shares, or because of dual class shares).

² In the second best contract, which is feasible when Ai ¸ AT , the outside financiers receives a monetary transfer when cash flows are high, and liquidation proceeds when cash flows are low. This corresponds to the case outside financing is obtained by issuing risky debt, secured by the firm’s assets.8

As the outside financiers are competitive, the entrepreneur captures the entire net value of the project. If his initial wealth is greater than AE he opts for the contract implementing the first best, and thus issues equity.

If his initial wealth is lower than AE but greater than AT , he opts for the contract implementing the second best and issues debt. Finally, if the initial wealth of the would–be entrepreneur is below AT , then he faces credit–rationing, and the project cannot be undertaken, although it has positive net social value.

 

4 Soft bankruptcy laws, social costs and judges

4.1 Social costs and bankruptcy judges

So far, we have focused on outside investors and managers. It is plausible, however, that firm liquidations can have external effects on third parties. In particular, they can entail social costs corresponding to the destruction of firm specific capital acquired by the employees of the firm, or firm specific investments made by suppliers or ncustomers. Additional costs are borne by citizens, whose everyday life was linked to the existence of the firm.

 

8Our simple model of debt contracts to discipline managers is in the line Hart and Moore (1994, 1998) and Bolton and Scharfstein

(1990), and consistent with empirical analyses such as Kaplan (1989).

10

For example, in France, in the eighties and nineties, there was a severe crisis in the textile industry. Many small businesses, located in small towns or villages were liquidated. This generated large social costs, as the whole life of these small towns was disrupted: as workers were laid off and had to move, and the population of these small towns decreased and became poorer, local shops and schools had to be closed, and valuable social

networks were distroyed. Tirole (2001,, page 3) points at the importance of such costs:

“Managerial decisions ... exert externalities on a number of “natural stakeholders” who have an innate relation with the firm... There is no denying that such externalities may be substantial; for example, the closure of a plant by a major employer in a depressed area has dramatic consequences for its workers and the local economy.”

In the present paper we do not explicitly model all these costs and externalities. Rather, we take a short–cut and assume that these costs (denoted c) are diffuse and borne by all the citizens.

As shown in the previous section, entrepreneurs with wealth above AE can obtain outside equity financing and undertake the investment project, without creating any liquidation risk. With a tough bankruptcy code, somewhat less wealthy entrepreneurs, with Ai 2 [AT ;AE], can issue risky debt. But if these firms default, this creates social costs. The purpose of soft bankruptcy laws is to interfere with the application of debt contracts to reduce these social costs of bankruptcy. We hereafter study the consequences of such laws on welfare, taking

into account credit–rationing as well as social costs.

We assume social costs are not known exactly ex–ante. We model these costs as a random variable, with expectation E(c), and, for positive real number k, we denote G(k) the probability that social costs are greater than k. Furthermore, we assume that social costs can be observed, ex–post, by the judge when the liquidation decision has to be made. We model the soft law as follows: The bankruptcy law defines a threshold value of the cost,c¤, and states that defaulting firms should be liquidated if the cost is lower than this threshold,

and reorganized otherwise. When the firm is reorganized, the creditors cannot receive liquidation proceeds, while the manager still enjoys the non–transferable private benefit from continuation, and the social costs of liquidation are not incurred. If the firm is liquidated, creditors receive L, the manager receives nothing, and social costs are incurred.

Judges are not always honest, and do not always act as stated by the law. We assume that with probability ¹ the judge is incorruptible, while with probability 1 ¡ ¹ she is corrupt. Tough laws do not grant judges any discretion regarding liquidations. Hence there is no scope for bribery. With a soft law, in contrast, the judge has discretion over the liquidation decision. Consequently, interested parties might attempt to bribe him to influence the court’s ruling. Managers can’t bribe judges when the return of the firm is 0, since they have no cash to do so. Furthermore, as in Bennedsen (2000) or Shleifer and Vishny (1994), we assume that the public is disorganized, so the entire population of citizens cannot get together to convince or bribe the judge to be efficient. The bank, in contrast, can use its financial resources to bribe the corrupt judge, so that he rules in favor of liquidation. For simplicity, we do not explicitly model the corruption game ; we just assume the

amount of bribe is equal to a fraction of the liquidation proceeds: ±; which reflects the bargaining power of the bankruptcy judge. Thus, the corrupt judge obtains: ±L while the bank obtains (1 ¡ ±)L. As analyzed below, the bribe left to the judge reduces the cash flow which can be pledged to the creditors, and consequently the bank’s willingness to grant a loan.9

Denote ¼ the ex–ante probability of reorganization in case of default when the judge is honest. It is given by:

Pr(c > c¤) = G(c) = ¼:

 

4.2 Equilibrium under the soft law

Under the soft law the incentive compatibility condition of the manager is the following:

ph(R ¡ T + B) + (1 ¡ ph)¹¼B ¡ e

> pl(R ¡ T + B) + (1 ¡ pl)¹¼B:

This is equivalent to:

9In a previous version of this paper, we also studied the case where the managers could bribe the judge. The main qualitative results were unaffected. Indeed, irrespective of the identity of the party who can collude with the judge, bribes reduce pledgeable income and thus enhance credit rationing problems.

12

R + B ¡

e

ph ¡ pl

¡ ¹¼B > T:

Comparing this incentive compatibility condition to its counterpart obtained for the tough law, we see that there is an additional term on the right–hand–side: ¹¼B, which reflects the adverse effect of the softness of the bankruptcy law on the incentives of the manager : when they can hope for reorganization, managers are less incentivized to exert effort than when they are threatened by systematic liquidation. Correspondingly, we denote:

AEI = ¹¼B;

where AEI stands for Adverse Effect on Incentives. Note that AEI is equal to the additional expected private benefit obtained by the manager in the bad state, relative to what he would have obtained under the tough law.

Under the soft law the participation constraint of the bank is:

T ¸

I ¡ A ¡ (1 ¡ ph)[¹(1 ¡ ¼) + (1 ¡ ¹)(1 ¡ ±)]L

ph

;

or:

T ¸

I ¡ A ¡ (1 ¡ ph)L + (1 ¡ ph)[1 ¡ ¹(1 ¡ ¼) ¡ (1 ¡ ¹)(1 ¡ ±)]L

ph

:

Comparing this condition to its tough law counterpart, we see that there is an additional term on the right–

hand–side: (1 ¡ ph)[1 ¡ ¹(1 ¡ ¼) ¡ (1 ¡ ¹)(1 ¡ ±)]L, which reflects the adverse effect of the softness of the

bankruptcy law on the participation constraint of the bank. Correspondingly, we denote:

AEP = [1 ¡ ¹(1 ¡ ¼) ¡ (1 ¡ ¹)(1 ¡ ±)]L;

where AEP stands for Adverse Effects on Participation. It is equal to the loss in expected liquidation proceeds obtained by the bank in the bad state, relative to what would be obtained under the tough law.

Combining the participation constraint of the bank and the incentive compatibility condition of the manager, the project can be financed by debt with a soft law if and only if:

A > AT + ph¹¼B + (1 ¡ ph)[1 ¡ ¹(1 ¡ ¼) ¡ (1 ¡ ¹)(1 ¡ ±)]L

= AT + phAEI + (1 ¡ ph)AEP:

Denote AS the right hand side of this inequality. The minimum level of initial wealth needed to avoid credit rationing in the case of debt financing with a soft law (AS) is greater than its counterpart in the case of debt financing with a tough bankruptcy law (AT ). Since the flexibility of the bankruptcy law weakens the incentives of the manager to exert effort, and tightens the participation constraint of the bank, it worsens the credit rationing problem.

The minimum level of initial wealth under which there is credit rationing under the soft law (AS) is increasing in the probability of continuation ¼. The greater this probability, the more difficult it is to induce the manager to exert effort and the outside financiers to provide funding. The soft bankruptcy law generating the smallest possible amount of credit rationing corresponds to the case where ¼ is set to 0 and AS is equal to:

AS(¼ = 0) = AT + (1 ¡ ph)(1 ¡ ¹)±L:

Building on the analysis above the following proposition obtains:

Proposition 1

If Ai < AT , firm i faces credit rationing, irrespective of the bankruptcy law. If AT · Ai < AS(¼ = 0), then entrepreneur i can finance the project with debt under the tough law but faces credit rationing under the soft law. If AS(¼ = 0) · Ai < AE, then entrepreneur i can finance the project under the tough law, and the soft law can be designed such that the project can be financed. Finally, if Ai ¸ AE, then the project can be financed by outside equity, irrespective of the bankruptcy law.

4.3 Properties of the equilibrium

First note that since AS(¼ = 0) > AT ; the minimum level of cash necessary to obtain financing under the soft law is always greater than its tough law counterpart. The wedge between AS(¼ = 0) and AT reflects the presence of corrupt judges, who are bribed into not liquidating the firm, irrespective of the relative levels of c and c¤. Hence, we can state the following proposition.

Proposition 2 Whatever the threshold level of social costs (c¤) stated in the soft law, there is more credit rationing with that law than with the tough law.

Note also that the smallest possible amount of initial cash below which there is rationing under the soft law (AS(¼ = 0)) is increasing in 1 ¡ ¹, the proportion of corrupt judges. Hence we can state the following proposition.

Proposition 3 The greater the proportion of corrupt judges the more credit rationing is generated by the soft law.

To shed further light on the difference between the soft law and the tough law, consider the role of L under the two regimes. Credit rationing problems are mitigated by the firm’s ability to pledge the liquidation value of its assets (L) as collateral. Note however that the derivative of AT with respect to L is more negative than the derivative of AS with respect to L. Hence, we can state the following proposition:

Proposition 4 The effectiveness of collateral to reduce credit rationing is stronger with a tough law than with a soft law.

The proposition arises because under the tough law collateral can be credibly pledged, while under the soft law creditors face the risk to be expropriated from their claims on the liquidation value of the firm.

5 The optimal law and the actual law

Hereafter in the paper we focus on the case where the poor are really poor, and do not have access to financing, while the rich are really rich, and can undertake the project, whatever the bankruptcy law, while the investment opportunities of the middle class citizens can be affected by the law. Thus we assume:

Ap < AT < Am < AE < Ar:

15

5.1 The socially optimal law

Credit rationing problems are more severe with the the soft law than with the tough law. When middle class

citizens are relatively poor, in the sense that: Am < As(¼ = 0), they are credit rationed with the former, while they can be financed with the latter. Hence with the soft law, the utilitarian social welfare is equal to:10

Wsoft = E(A) + ¹rS1;

while with the tough law it is:

Wtough = E(A) + ¹rS1 + ¹m(S2 ¡ (1 ¡ ph)E(c)):

Thus, the tough law is ex–ante socially optimal if financing middle class entrepreneurs is optimal, i.e., if the net present value of the project is greater than the expected social costs of liquidation it generates. This is stated in the next proposition:

Proposition 5 If the middle class is relatively poor, in the sense that: Am < As(¼ = 0), the tough law is socially optimal if and only if the net present value of the investment project when there is liquidation in the low cash flow state, is greater than its expected social costs, that is:

S2 > (1 ¡ ph)E(c):

On the other hand, when the middle class citizens are relatively rich, in the sense that Am > As(¼ = 0),

they have access to financing even with the soft law, provided the threshold level of social costs, c¤, is not too low. Consequently, the soft law is optimal, as it can be designed to optimally reduce social costs without worsening credit rationing. This is stated in the next proposition.

Proposition 6 If the middle class is relatively rich, in the sense that Am > As(¼ = 0), then the soft law is socially optimal.

10Since we focus on the utlitarian social welfare, transfers cancel out. Hence, transfers to corrupt judges have no direct impact of welfare. But they have an indirect effect: they reduce welfare because of their adverse impact on the participation constraint of the bank.

 

5.2 Voting on the bankruptcy law

The investment opportunities of the poor and the rich citizens are unaffected by the bankruptcy law. From their perspective the only difference between the soft law and the tough law is that social costs are greater with the latter. As these costs are diffuse and borne by all citizens, the poor and the rich prefer the soft law rather than the tough law.11 Consequently if the coalition of the rich and the poor includes more than half the

population (i.e., if ¹r + ¹p > 1

2 ), the soft law is chosen by majority voting, irrespective of its social optimality,

and in particular irrespective of its consequences on credit rationing. This raises the possibility of a conflict between the outcome of majority voting and social optimality. This can lead to excessively soft laws, as stated in the following proposition:

Proposition 7 If the coalition of the rich and the poor has the majority (¹r + ¹p > 1

2 ); the middle class is relatively poor (Am < As(¼ = 0)), and social costs are relatively limited (S2 > (1 ¡ ph)E(c)), then majority voting selects the soft law although this is socially suboptimal.

The proposition reflects the fact that the poor and the rich fail to internalize the adverse effect of the soft law on middle class citizens which it deprives from access to credit.

Now consider the individual preferences of the middle class citizens. If they are relatively poor, in the sense that Am < As(¼ = 0), then their expected utility under the soft law is simply: Am, as only the rich are financed and there are no liquidations. On the other hand, under the tough law their expected utility is:

Am + S2 ¡ ¹m(1 ¡ ph)E(c):

Hence they prefer the tough law if and only if:

S2 > ¹m(1 ¡ ph)E(c):

This condition is very similar to the condition under which the tough law is socially optimal (stated in Proposition

5). The only difference is that the right–hand–side of the inequality , corresponding to social costs, is

lower than in Proposition 5. This reflects the fact that individual citizens do not fully internalize the social

11Our analysis of the preferences of different social classes towards economic policies is in line with Biais and Perotti (2002).

17

costs induced by bankruptcies. This raises the possibility of discrepancies between the optimal law from the point of view of the middle class and the socially optimal law. This is illustrated in the following proposition:

Proposition 8 If the middle class citizens have the majority (¹m > 1

2 ) and are relatively poor (Am <

As(¼ = 0)) then the tough law is chosen by majority voting if: S2 > ¹m(1 ¡ ph)E(c). This is socially suboptimal if: S2 < (1 ¡ ph)E(c).

 

6 Extensions

6.1 Bail out policies versus bankruptcy laws

With soft bankruptcy laws, judges can decide to expropriate creditors to avoid liquidations. Another possibility would be to raise taxes to pay–out the creditors. Potentially this could lead to a better outcome than soft laws as it could avoid the social costs of liquidation, without making banks reluctant to lend. Yet, rich and poor citizens prefer soft laws rather than bail–outs. While both policies mitigate the social costs of bankruptcy, the former does not create a tax burden for rich and poor citizens, while the latter does. To put it differently,

bail–outs amount to subsidies from the rich and the poor to the middle class. The rich and the poor do not favour such policies, given that they do not internalize the effects of bankruptcy laws on credit–rationing for the middle class.

 

6.2 Taking into account the positive consequences of business creations for non–entrepreneurs

So far we did not take into account the job opportunities generated by new businesses for non–entrepreneurs.

In this subsection we offer an extension of our model which takes this aspect into account.12

12This is in the line of the political economy analyses of the interactions between financial and labour markets by Pagano and Volpin (2000) and Bolton and Rosenthal (1999).

18

6.2.1 Modelling jointly the job market and the capital market

Assume that, in addition to I units of capital, each project requires one unit of labor, and that agents who are not entrepreneurs can supply labor. For example suppose that agent i can supply labor li, at a disutility cost of c(li). c(li) is assumed to be increasing and convex. If agent i does not become an entrepreneur, and supplies labor li, she is employed by a firm which generates cash flow with probability pH, and in this case she receives her wage w. Hence her expected utility is:

liphw ¡ c(li):

We assume that the labor market is competitive. Agent i’s first order condition yields:

phw = c0(li);

i.e., the (expected) wage is equal to the marginal disutility of labor. Inverting the marginal cost of labor, we obtain labor supply as a function of expected wages: li = c0¡1(phw):

Consider the case where only the rich have access to financing, while the poor and the middle class are credit rationed. In this case the market clearing condition on the labor market is:

¹r = (¹m + ¹p)c0¡1(phw);

where the left hand side is the labor demand expressed by the business created by the rich, while the right– hand–side is the labor supply offered by the middle class and the poor. In this context the equilibrium wage is:

1

ph

c0( ¹r

¹m+¹p

). Denote it: wpm, where the subscript denotes that it is earned by the middle class and the poor. On the other hand, if both the rich and the middle class have access to credit, the market clearing condition

on the labor market is:

(¹r + ¹m) = ¹pc0¡1(phw):

In this context the equilibrium wage is: 1

ph

c0(¹r+¹m

¹p

). Denote it: wp, where the subscript denotes that it is

earned by the poor. Comparing the two cases, and relying on the convexity of c(:), we obtain the following proposition:

Proposition 9 The equilibrium wage is greater when both the rich and the middle class have access to financing than when the middle class are credit constrained.

When both the rich and the middle class have access to credit, labor demand is greater, and labor supply lower, than when the middle class is credit constrained. Consequently wages are greater. This is particularly pronounced when the marginal cost increases strongly with labor supply, i.e., when c(:) is very convex.

Now turn to the financial market. Everything is as in the previous sections, except that the cash flow available to pay back the financier and incentivize the manager is no longer R, but only R ¡ w. Applying the same logic as in the above sections, entrepreneurs can finance their project with equity if their initial wealth is greater than:

AE = I ¡ ph(R ¡ w) + ph

e

ph ¡ pl

;

while they can raise debt under the tough law if their initial wealth is greater than:

AT = AE ¡ (phB + (1 ¡ ph)L); and under the soft law if it is greater than:

AS = AT + phAEI + (1 ¡ ph)AEP:

These equations are formally similar to those presented in the above sections, except that the wage (w) is deducted from the revenue (R) in AE. Note however that there is another, more subtle, difference. The wage depends on the number of firms which are financed. Hence, to the extent that this number reflects the bankruptcy law, the wage varies with that law.

For brevity, we now focus on what we think is the most interesting case, i.e., the situation where the middle class obtain financing under the tough law, while they are rationed under the soft law (and the rich always obtain equity financing). The corresponding equilibrium conditions for the tough law are:

Ar > I ¡ ph(R ¡ wp) + ph

e

ph ¡ pl

>  

Am > I ¡ ph(R ¡ wp) + ph

e

ph ¡ pl

¡ (phB + (1 ¡ ph)L) > Ap;

20

while for the soft law the equilibrium conditions are:

Ar > I ¡ ph(R ¡ wmp) + ph

e

ph ¡ pl

= Asoft

E ;

and:

Asoft

E ¡ (phB + (1 ¡ ph)L) + phAEI + (1 ¡ ph)AEP > Am:

6.2.2 Social welfare

Assuming that parameters are such that these conditions hold, we now analyze the socially optimal law, and the law chosen by majority voting. Under the tough law, the utilitarian social welfare is:

Wtough ¡ ¹pc(¹m + ¹r

¹p

);

where the first term is the social welfare obtaining under the tough law when labor is not needed (characterized in the previous section) and the second term is equal to the disutility of labour of the poor workers (wages cancel out in the utilitarian function).

Under the soft law, the utilitarian social welfare is:

Wsoft ¡ (¹p + ¹m)c( ¹r¹m + ¹p):

In the previous section we established that, when there was no need for labor supply, the tough law was socially preferable to the soft law if:

Wtough ¡Wsoft = ¹m(S2 ¡ (1 ¡ ph)E(c)) > 0:

To take into account the labor market the condition becomes:

¹m(S2 ¡ (1 ¡ ph)E(c)) > ¹pc(¹m + ¹r¹p’) ¡ (¹p + ¹m)c( ¹r#¹m + ¹p):

In addition to the trade–off between investment and social costs (characterized in the previous section), theabove inequality also emphasizes that, with the tough law, more labor needs to be supplied, and correspondingly the disutility of labor is increased.

 

6.2.3 Political preferences

In this subsection, for simplicity, we assume that c(x) = k 2x2. The greater the parameter k, the greater the convexity of c(:). As shown above, the more convex the cost function, the greater the increase in wages generated by the increase in labor demand due to additional business creations under the tough law.

As in the previous section, since the poor do not internalize the welfare gains of the other classes, their preferences about the law are not necessarily aligned with social optimality. This is stated in the following proposition.

 

Proposition 10 The poor prefer the soft law, although the tough law is socially optimal, if:

¹mS2 ¡k

2[¹p(¹m + ¹r¹pz)2 ¡ (¹p + ¹m)( ¹r¹m + ¹p)2]> ¹m(1 ¡ ph)E(c) >k

2[(¹m + ¹r=¹p)2 ¡ ( ¹r¹m + ¹p)2]:

The first inequality corresponds to the condition that the tough law be socially optimal. It requires that the welfare gains of the middle class due to the access of credit be large, in particular due to S2 being large.

The second inequality corresponds to the condition that the poor prefer the soft law. It requires that the cost function c(:) be not too convex (and corresponding k be low). This dampens the increase in wages generated by the tough law. Thus the analysis conducted in this subsection shows that, while the increase in wages

stemming from business creations reduces the attractiveness of the soft law for the poor, if that effect is not too pronounced the qualitative effects identified in the previous section are upheld.

Finally note that, in our model, the rich never favour the tough law: Not only does it generate social costs, it also raises wages, and thus reduces their profits.

 

7 Conclusion

This paper proposes a simple model in which tough bankruptcy laws mitigate credit rationing problems while soft laws can reduce the social costs of liquidation. We analyze the political process leading to the adoption of bankruptcy laws and characterize situations whereby the laws emerging from voting are not socially optimal.

22

Our analysis yields several empirical implications regarding the financing of corporations:

² In line with the country level evidence offered by La Porta et al (1997, 1998) and the firm level evidence offered by Giannetti (2000) our model implies that access to debt financing is reduced in countries with soft bankruptcy codes. Thus, soft laws are an obstacle to entrepreneurship.

² Our model also implies that the positive impact of collateral for access to credit should be greater in countries with tough bankruptcy laws. It could be interesting to test this using firm–level data, such as that used by Giannetti (2000).

² A further implication of our model is that the adverse effect of soft bankrutcy laws on credit rationing increases with the proportion of corrupt judges.

Our analysis also yields implications relative to the emergence of different types of laws:

² When the party in power represents the interests of relatively poor citizens, who, whatever the bankruptcy law, cannot raise funds to become entrepreneurs, soft laws should be passed (whether they are socially optimal or not). Indeed, the 1985 French law was chosen by the rather leftist socialist majority which wasin power at the time. In contrast, in countries with a politically influent and potentially entrepreneurial middle class, tough bankruptcy laws are more likely to be passed. It would be interesting to test these

implications in a cross section of countries, including developed and developing or transition economies.

² Our theoretical analysis implies that middle class citizens are in favor of tough laws if the surplus they obtain when setting up business are relatively large relative to the social costs of bankruptcy. Also, relatively poor citizens are less favorable to soft laws when new business creations could result in significantly larger wages for them. Furthermore tough laws are more likely to be passed when the social costs of liquidation are not perceived to be high. These conditions are more likely to hold in upturns than in downturns of the business cycle. Hence, our model implies that soft (resp. tough) laws are more nlikely to be passed in economic downturns (resp. upturns). This is consistent with stylized facts on the history of bankruptcy laws: in the US soft laws tended to be passed after business cycles downturns,

while in the UK a major change in the bankrutcy law was initiated by creditors in 1869, at the height of an expansion. Domowitz and Tamer (1997) offer empirical evidence on the correlation between changes in the bankruptcy laws and the business cycle. It could be interesting to analyze the macroeconomicconsequences of the procyclicality of changes in bankruptcy laws. To pursue this avenue of research it would be interesting to build on the analyses of the politics of macroeconomics offered by Alesina (1987) and Alesina and Tabellini (1990).

Our analysis also yields some policy implications. In countries where credit rationing problems are severe and corruption is a serious issue, our model suggests that very little can be gained, and a lot can be lost, by opting for a soft bankruptcy code. This suggests that, in transition economies and possibly in developing countries, opting for tough bankruptcy codes might well be the best course of action – if the social costs of liquidation are not extremely high. This line of thought could be pursued further by examining the following conjecture: in a dynamic extension of our analysis, soft bankruptcy codes could lead to poverty traps: if  the majority of the citizens cannot invest anyhow, they support soft bankruptcy policies; in turn these policies reduce the ability of relatively poor entrepreneurs to invest, and become richer. This can lead to a vicious circle.

In further research it could be interesting to analyze this issue in a dynamic model, where investment, wealth distribution, political preferences and bankruptcy laws would evolve jointly. This would be in the same spirit as Gali and Zilibotti (1995) who analyze growth dynamics where initially poor economies are stuck in a poverty trap.13 An important difference is that in Gali and Zilibotti (1995) poverty traps arise because of imperfect

competition between firms, while in the approach suggested here it would arise through the interaction between credit market imperfections, wealth distribution and laws.

13It would also be in the line of Aghion and Bolton (1997) who analyze the joint dynamics of credit rationing and wealth  inequality.

 

References

Aghion P., Hart O., Moore J., 1992, “The Economics of Bankruptcy Reform”,

The Journal of Law, Economics and Organization, July, 523-546.

Aghion P., Bolton P., 1997, “A Theory of Trickle Down Growth and Development,” Review of Economic Studies, 151–172.

Alesina, A., 1987, Macroeconomic Policy in a Two–Party System as a Repeated Game, Quarterly Journal of Economics, 651–678.

Alesina, A. and G. Tabellini, 1990, A Positive Theory of Fiscal Deficits and Government Debt, Review of Economic Studies, 403–414.

Atiyas I., 1995, “Bankruptcy Policies: a Comparative Perspective”, working paper, Private Sector Development Department, The World Bank.

Bebchuck L., 1988, “A New Approach to Corporate Reorganization”,

Harvard Law Review, 775–804.

Bennedsen, M., 2000, “Political ownership”, Journal of Public Economics, 559–81.

Berglof, E. and H. Rosenthal, 2000, The political economy of American bankruptcy: The evidence from roll call voting, 1800–1978, Working paper, Princeton University.

Berkovitch E., Israel R., Zender J.F., 1997, “Optimal Bankruptcy Laws and Firm Specific Investments”,

European Economic Review.

Berkovitch E., Israel R., 1999, “Optimal Bankruptcy Laws Across Different Economic Systems”, Review of Financial Studies.

Biais B., Mal´ecot J.F., 1996, “Incentives and Efficiency in the Bankruptcy Process: The Case of France.”,

PSD Occasional Paper n±23, April, Private Sector Development Department, The World Bank. Biais B. and Casamatta C., 1999, “Optimal Leverage and Aggregate Investment”, The Journal of Finance, vol 54, n±4, 1291-1323.

Biais, B., and E. Perrotti, 2002, “Machiavellian Privatization”, forthcoming American Economic Review, 240–258.

Bolton, P. and D. Scharfstein, 1990, “A theory of predation based on agency problems in financial contracting.”

American Economic Review, 94–106.

Bolton, P., and H. Rosenthal, 1999, “Political intervention in debt contracts: Moratoria and bail–outs,”

Working paper, Princeton University. Bond, P., 2000, Judicial Agency, Contracts and Laws, Working Paper, Northwestern University.

Domowitz, I., and E. Tamer, 1997, “Two hundred years of bankruptcy: A tale of legislation and economic fluctuations.” Working Paper, Northwestern University.

Fisher T.C.G., Martel J., 1995, “The Creditors’ Financial Reorganization Decision: New Evidence from Canadian Data”, Journal of Law, Economics and Organizations, n±11, 112-126.

Fisher T.C.G., Martel J., 1999, “Should We Abolish Chapter 11? Evidence from Canada”, Journal of Legal Studies, n±28, 233-257.

Fisher T.C.G., Martel J., 2000, “The Bankruptcy Decision : Empirical Evidence from Canada”, Working Paper, THEMA, Universit´e de Cergy-Pontoise.

Franks J.R., Torous W.N., 1989, “An Empirical Investigation of US Firms in Reorganization”, The Journalnof Finance, vol 44, n±3, 747-767.

Franks J.R., Torous W.N., 1994, ” A Comparison of Financial Recontracting in Distressed Exchanges and Chapter 11 Reorganizations ”, Journal of Financial Economics, vol 35, n±3, 349-370. Franks J.R., Sussman O., 1999, “Financial Innovations and Corporate Insolvency”, working paper, London Business School.

Franks J.R., Sussman O., 2000, ” An Empirical Study of Financial Distress of Small Bank-Financed UK Companies : A Reassessment of English Insolvency Law ”, working paper, London Business School.

Franks J.R., Nyborg K., Torous W., “A Comparison of US, UK and German Insolvency Codes,” Financial Management, 86–101.

Gali, J. and F. Zilibotti, 1995, “Endogenous growth and poverty traps in a Cournotian model,” Annales d’Economie et de Statistique, 197–210.

Giannetti, M., 2000, “Do better institutions mitigate agency problems? Evidence from corporate financing choices,” Working Paper, Stockholm School of Economics.

Harris M., Raviv A., 1993, ” The Design of Bankruptcy Procedures ”, working paper n±137, University of Chicago.

Hart, O., and J. Moore, 1994, “A theory of debt based on the inalienability of human capital.” Quarterly Journal of Economics, 841–879.

Hart, O., and J. Moore, 1998, “Default and renegociation: A dynamic model of debt.” Quarterly Journal of Economics.

Holmstrom B., Tirole J., 1997, ” Financial Intermediation, Loanable Funds and the Real Sector ”, Quaterly Journal of Economics, n±112, 663-691.

Kaplan, S., 1989, “The effects of management buy–outs on operating performance and value,” Journal ofnFinancial Economics, 217–254.

Kiefer R., 2000, ” Distressed Equity Issues, Underinvestment and the German Financial System ”, working paper, European University Institute.

Lambert–Mogiliansky, A., C. Sonin, and E. Zhuraskaya, 2000, Capture of bankruptcy: Theory and evidence from Russia, Working paper, CEFIR and NES.

La Porta R., Lopez de Silanes F., Shleifer A., Vishny R., 1997, “Legal Determinants of External Finance”, The Journal of Finance, vol 52, n±3;1131–1150.

La Porta R., Lopez de Silanes F., Shleifer A., Vishny R, 1998, “Law and Finance”, Journal of Political Economy, vol n±106; n±6; 1113–1155.

Pagano, M., and P. Volpin, 2000, The political economy of corporate governance, Working paper, CSEF– Salerno University.

Shleifer, A., and R. Vishny, 1994, “Politicians and firms”, Quarterly Journal of Economics, 995–1025.

Tirole, J., 2001, “Corporate governance”, Econometrica, 1–36.

White M.J., 1989, ” The Corporate Bankruptcy Decision ”, Journal of Enonomic Perspectives, vol 3, n±2, 129-162.

White M.J., 1994, ” The Costs of Corporate Bankruptcy : A U.S.-European Comparison ”, in ” Bankruptcy

: Economic and Legal Perspectives ”, J. Bhandari, ed. Cambridge University Press, 1-37.

28

Appendix: Proofs

Proof of Proposition 1:

As ¼ increases from 0 to 1, AS increases from: AS(¼ = 0);to: AS(¼ = 1): Hence, for all intermediary values of the initial wealth: Ai 2 [AS(0);AS(1)];there exists a value of ¼ 2 [0; 1] such the incentive compatibility and participation constraints hold as equalities, i.e., AS(¼) = Ai. Correspondingly, there exists a value of the threshold c¤ such that the incentive and participation constraint hold for that firm. Equivalently, the soft law can be designed so that this firm has access to credit.

QED

Proof of Proposition 6:

If Am 2 [As(¼ = 0);As(¼ = 1)], then the middle class can undertake the investment under the tough law.

The middle class can also undertake the investment if the probability of reorganization is lower than or equal to ¼m defined by: AS(¼m) = Am, or equivalently if the threshold cost c¤ is greater than or equal to cm defined as: cm = G¡1(¼m). If the total surplus generated by the project is large enough relative to the social costs,

i.e., if:

S2 + (1 ¡ ph)(¹¼m + (1 ¡ ¹)) > (1 ¡ ph)¹(1 ¡ ¼m)E(cjc < cm);

then, under the soft law it is optimal to set: c¤ = cm. In that case, social welfare is greater under the optimal soft law than under the tough law. Otherwise, it is optimal to set c¤ above cm and social welfare is the same under the tough law and under the optimal soft law.

QED

 

Proof of Proposition 10:

The expected utility of poor citizens under the tough law is:

Ap + ¹m + ¹r¹pc0(¹m + ¹r¹p) ¡ c(¹m + ¹r¹p) ¡ ¹m(1 ¡ ph)E(c);

which simplifies to:

Ap + k2(¹m + ¹r¹p)2 ¡ ¹m(1 ¡ ph)E(c);

Their expected utility under the soft law simplifies to:

Ap + k2( ¹r¹m + ¹p)2:

Hence, the poor prefer the soft law if the increase in wages brought about by business creations under thetough law is more than compensated by social costs, i.e. if:

¹m(1 ¡ ph)E(c) >k2[(¹m + ¹r¹p)2 ¡ ( ¹r¹m + ¹p)2]:

To obtain the condition in the proposition, combine that condition with the condition under  which the tough law is socially optimal: