Regulating collateral, bankruptcy and the case of Brazil

Regulating collateral, bankruptcy and the case of
Brazil
Aloisio Araujo
Rafael V. X. Ferreira
Bruno Funchal
March 20, 2013
1
Introduction
For a long time now, economists recognize the importance of studying real life
phenomena, such as default, bankruptcy and crisis. Even before the recent bank
and debt crisis, economists understood that even if default could be avoided it is
not Pareto optimal to do so in the presence of incomplete markets.
More importantly, lack of commitment is an important characteristic of the real
world. So, economists start seriously building models with lack of commitment to
try to understand what regulation could be devised to ameliorate this situation.
As expected, this intellectual activity became more intense after the crisis.
This literature is immense and have been approached from several angles. In
this paper, we focus on the models we particularly have worked with and that we
find more relevant.
The main model presented in the paper is a two period general equilibrium
model, with total lack of commitment and where loans must be backed by collateral. We can call this type of model simple mechanism design model. We also
study bankruptcy models and its optimal design.
More precisely, our main focus is on how collateral regulation and bankruptcy
law design promote financial development, economic efficiency and stability in
periods of crisis.
In section 2, we discuss regulation of collateral requirements – a much debated
topic in the aftermath of the 2008 financial crisis – by revisiting Araujo et al.
(2012b). One of their results is that it is never optimal to regulate the market for
subprime loans. The authors also show that regulation of collateral requirements
never lead to a Pareto-improvement for all agents.1
In sections 3 and 4 we review some of the main reasons why having good
bankruptcy laws in place is important. Section 3 addresses the topic of corporate
1
These results are valid under certain conditions.
1
bankruptcy laws and discusses their importance for promoting economic recovery
following periods of crisis. A good bankruptcy law can work as an intrument to
accelerate economic recovery in periods post-crisis, as stated by Miller and Stiglitz
(2010) and Bergoeing et al. (2002). From the efficiency standpoint, in the ex-ante
sense there is a lot of theoretical and empirical work showing that creditor-friendly
bankruptcy laws are important to provide incentives; from the ex-post sense, when
firms are economically viable, it may be optimal to give managers and shareholders
a second chance.
Section 4 discusses personal bankruptcy laws. The 2005 Personal Bankruptcy
Law Reform is often cited as cause of the increase in subprime mortgage foreclosures. For instance, Morgan et al. (2009) argue that the 2005 Bankruptcy Reform
Act contributed to the surge of subprime foreclosures by shifting risk from unsecured credit lenders to mortgage lenders. The design of a bankruptcy procedure
for individuals should balance the incentive to repay debts – thus reducing moral
hazard – and the need to protect borrowers from a bad state of nature, allowing
them a fresh-start. Araujo and Funchal (2005b) addresses this issue empirically
and shows that the relationship between debtors’ punishment – to hinder moral
hazard – and the size of the credit market is not always increasing. It follows that
an intermediate level of punishment is optimal for credit market development.
Finally, section 5 presents a sucessful case of corporate bankruptcy law reform.
Araujo et al. (2012a) study the effect of the Brazilian bankruptcy law reform on
firm debt contract terms. Looking at firm accounting data, the authors found a
reduction in the cost of debt, an increase in total debt, and an even larger increase
in long-term debt, following the legal reform in Brazil. Section 6 concludes.
2
Collateral
When borrowers with limited commitment to repay their debts can default on
their promises, it is usually optimal for them to do so. Therefore, the possibility
of default creates a need for “incentives” (enforcement mechanisms) for debtors to
keep their promises.
Nonetheless, it is not possible in practice to devise a mechanism capable of
ensuring that promises are kept in all circumstances. In addition, such a mechanism could even be undesirable, either for its capacity of preventing efficient risk
sharing or for its cost.
There are several enforcement mechanisms commonly used in the economy,
but debt collateral is one of the most common.2 A great portion of debt contracts
grants lenders the ability to seize tangible assets belonging to the debtor, in case
2
See Araujo et al. (2002).
2
of default. Home mortgages are one example of collateralized debt, and so are
many corporate bonds, that are backed by equipment and plants.
The most recent financial crisis highlighted the consequences of widespread default when promises greatly exceed physical endownments. It also strengthened
the argument for more regulation of the borrower-lender relationship, with regulators demanding for collateral requirements to be increased. Also in the aftermath
of the crisis, investors started asking for safer assets, mostly by pressing for more
collateral to be given in order to secure their funds. All of this translates into a
greater demand for high-quality collateral, drawing attention to scenarios in which
the collateralizable good is scarce.
Araujo et al. (2012b) consider a general equilibrium model to study welfare
under different types of collateral regulation, with scarce collateral. In their setup,
individuals have to put up durable goods as collateral when they want to take short
positions in financial markets.3 Agents are allowed to default on their promises
without any punishment, other than the seizing of the collateral, which is distributed among creditors. Default will occur whenever the market value of the
collateral is lower than the face value of the associated promise.4
More precisely, they consider a pure exchange economy over two periods, t ∈
{0, 1}. In period 1 there is uncertainty regarding the realization of the state of
nature s ∈ {1, . . . , S}. The economy is populated by H agents, and there are
L = 2 goods. Good 1 is perishable and good 2 is durable, without depreciation.
S∗L
Initial endownments are denoted by eh ∈ R+
, where S ∗ = S + 1, and for all
goods there are complete spot markets, with spot prices across states denoted by
∗
S∗L
−→ R+ representing individual
p ∈ RS L . There are utility functions uh : R+
∗
h
S L
preferences over consumption x ∈ R+ , and J real assets, with prices denoted by
q ∈ RJ+ . Assets are indexed by j ∈ J = {1, . . . , J} and their promises are made
in units of the perishable good. These promises are denoted by Aj ∈ RSL .
Also, for each asset there is an associated collateral requirement Cj ≥ 0, denoted in units of the durable good and always held by the borrower. A key aspect
of the model in Araujo et al. (2012b) is that it assumes all assets promise a safe
payoff. Thus, assets differentiate themselves only with regard to collateral requirements.
Default will happen whenever the face value of the promises made by the agents
is higher than the market value of the durable good held by them. So, the payoff
of asset j in state s will be given by min{1, ps (s)Cj }.
Agent h’s portfolio is given by (θh , ϕh ), where θh = (θ1h , . . . , θJh ) ∈ RJ+ gives the
number of units of each asset bought by the agent, and ϕh = (ϕh1 , . . . , ϕhJ ) ∈ RJ+
denotes the short positions in the assets.
3
4
This model was introduced by Geanakoplos and Zame (2007).
See also Geanakoplos (1996).
3
Agents in this economy choose consumption and portfolios (xh , θh , ϕh ), by solving:
max
uh (xh )
s.t.
0
p(0) · (xh (0) − eh (0)) + q · (θ − ϕ) ≤ (θjh − ϕhj ) min{1, ps (s)Cj } ≤ 0
p(s) · (xh (s) − eh (s)) − p2 (s)xh2 (0) −
j∈J
xh2 (0) −
ϕhj Cj ≥ 0
x≥0,ϕ≥0,θ≥0
j∈J
As usual, a competitive equilibrium is defined by market clearing and agents’
optimality conditions:
h
Definition 1. A GEIC 5 equilibrium for this economy is a vector [(xh , θ , ϕh )h∈H ; p, q]
such that:
h
i) (xh , θ , ϕh )h∈H solves agents’ problem.
h
h
ii)
h∈H (x (0) − e (0)) = 0
h
h
h
iii)
h∈H (x (s) − e (s) − x2 (0)) = 0
iv)
h∈H (θ
h
− ϕh ) = 0
When the set of assets J in the economy is very large, collinearity will render
most assets redundant. In this case, there will be a set J CC , containing only
S assets, that can replace J as the set of actively traded assets without loss of
generality. In this setting, a GEICC equilibrium will be a GEIC equilibrium in
which the set of tradable assets contain J CC ; and a GEIRC equilibrium will be a
GEIC equilibrium with exogenously fixed set of collateral requirements.
In fact, Araujo et al. (2012b) show that, given a GEICC equilibrium, there will
be no other set of assets such that in the resulting GEIRC equilibrium all agents
are better off.6
Araujo et al. (2012b) also study how scarcity and an unequal distribution of
collateralizable durable goods affect risk-sharing and welfare. If the durable good is
plentiful and satisfy the collateral constraints of the agents, the model is equivalent
to a standard Arrow-Debreu model and competitive equilibrium allocations are
Pareto-optimal.
In the presence of scarcity, however, most assets are not traded in equilibrium
and markets are incomplete. Araujo et al. (2012b) quantitatively address the
5
6
General equilibrium with incomplete markets and collateral.
Under the assumption that all agents have identical homothetic utility.
4
question of whether government regulation can lead to welfare improvements if
the collateralizable good is scarce. They provide a series of examples – some of
them illustrative; some realistically calibrated – to lay out their point.
In one particular numerical exercise, calibrated to match real world data,
Araujo et al. (2012b) investigate the role of subprime loans for risk-sharing, as well
as who in the economy gains and who loses through regulation. They assume there
are four states of nature and four types of agents with heterogeneous preferences,
whose endownments are calibrated to match the income and wealth distribution
in the US: agent 1 is considered to be rich; agents 2 and 3 are middle-class individuals; and agent 4 is poor. Endownments for non-durable are interpreted as
income; endownments for the durable are interpreted as wealth.
Then, different equilibria are computed, under different types of collateral regulation. Four GEIRC equilibria are considered:
1. GEICC: a specific GEIRC equilibrium for which adding assets (differing
only by the collateral requirement) does not change the equilibrium allocation;
2. GEIRC 1: regulation that only allows trades in the asset that defaults in
all states;
3. GEIRC 2: regulation that forbids subprime loans;
4. GEIRC 3: regulation that forbids default.
Figure 1 illustrate the welfare implications for each equilibrium.
In the GEICC equilibrium two assets are traded. The rich agent 1 lends in
the sub-prime asset and borrows in the safe asset. Agents 2-4 borrow exclusively
sub-prime, while agents 2 and 3 (the middle-class) actually saves some money in
the safe mortgage.
GEIRC1 corresponds to the equilibrium where there is full default for all traded
assets. The rich agent 1 benefits from lending more units in the subprime asset, due
to higher interest rate, in relation to GEICC equilibrium. Agents 2 and 3 cannot
be made better off through any regulation. Agents 1 and 4 gain simultaneously if
only subprime borrowing is allowed.
GEIRC2 corresponds to the equilibrium in which subprime loans are not available. In this case, agents 1 and 4 lose, since they can’t lend and borrow, respectively, in the subprime asset. Agents 2 and 3 are better off in GEIRC2, through
lending in the safe asset.
Finally, forbiding default – i.e., the GEIRC3 equilibrium – makes all agents
worse off.
5
Figure 1: Calibrated examples from Araujo et al. (2012b)
6
In the examples considered by Araujo et al. (2012b), regulation of collateral
requirements never leads to a Pareto-improvement for all agents, and equilibria
corresponding to different regulated collateral requirements are often not Paretoranked. But it is possible to benefit some agents by regulation.
In addition they find that it is never optimal to regulate the market for subprime
loans, since middle-class individuals lose from such a regulation.
3
Optimal Corporate Bankruptcy Law
As stated in the previous section, collateral is an important mechanism to foment
credit transactions between borrowers and lenders. However, as the number of
lenders increases, it becomes increasingly difficult to write debt contrats – with or
without collateral – since debtors may acquire new creditors and assets as time
passes. It may be very hard to specify how the division process should change as
function of such adjustments. Bankruptcy laws help provide a default option for
this problem of contract incompleteness.
Thus, the ability of economic agentes to incur debt depends not only on their
capability to fulfill their payment promisses, but also on the legal environment
in which both borrowers and lenders operate. An important part of the legal
environment is the legislation governing bankruptcies, as well as the quality of
judicial enforcement, both in terms of efficiency and speed.
In this section, we review some of the main reasons why having a good corporate
bankrupcy law in place is important.
Why a good corporate bankruptcy law for crisis
Bankruptcy laws have the ability to promote financial development and induce
economic growth. In times of economic crisis, the number of bankruptcies usually
increase, so bankruptcy laws can play an important role in aliviating crisis and
accelerating turnarounds. Figures 2 and 3 show the number of bankruptcies in
selected countries during the 2008 financial crisis, compared to the same number
a few years before.7
It is often the case that during crisis, some banks find themselves holding claims
of bankrupt debtors. If the bankruptcy law in place produces low recovery rates
and slow insolvency resolution, defaults from borrowers can lead to defaults of
lenders. The 2008 financial crisis produced numerous examples of creditors facing
liquidity problems – and ultimately having to file for bankruptcy themselves – for
not being able to recover much of the debt owed to them by defaulting individuals
and firms. Johnson et al. (2000) report that many bank collapses following the 1998
7
Source: OECD (2012), Entrepreneurship at a Glance 2012, OECD Publishing.
7
Figure 2: Number of bankruptcies in selected countries
Italy
Japan
Netherland
Norway
Spain
United Kingdom
United States
400
350
300
250
200
150
100
50
0
Russian devaluation were associated with creditors recovering very few of what was
owed to them by defaulting debtors. Similar cases were reported throughout Asia
during the late 1990s crisis. Even lenders who are not forced into bankruptcy
might find themselves with liquidity problems, and forced to reduce or deny credit
to borrowers in sound financial situation.
Yet, if increased recovery rates during crisis comes at the expense of a higher
number of liquidations, it might lead to fire sales. If liquidating firms have their
assets sold in times when many potencial buyers are facing difficulties themselves,
the selling price tends to be negatively affected.
Miller and Stiglitz (2010) bring attention to the externalities that arise when
bankruptcy laws designed mostly to deal with issues in the firm-level are widely
used in times of crisis. Following the realization of aggregate shocks with the
potential to produce widespread defaults, bankruptcy courts concerned only with
idiosyncratic events might fail to consider the potential of large-scale asset disposal
to produce fire-sales. Miller and Stiglitz (2010) make the case for a “Super Chapter
11” that internalizes these externalities.
8
Figure 3: Number of bankruptcies in selected countries
Australia
Canada
Denmark
Finland
France
Germany
Iceland
400
350
300
250
200
150
100
50
0
Also, avoiding liquidation of viable firms is important not only for efficiency
reasons, but also to accelerate turn arounds, since with many firms exiting the
economy, unemployment levels rise.
Still on crises, Bergoeing et al. (2002) compare the recoveries of the Mexican
and Chilean economic crises in the early 80’s. Chile reformed its bankruptcy
code in 1982, clearly defining the rights of each creditor and improving efficiency
of bankruptcy procedures. In contrast, Mexico had an obsolete and unwieldy
bankruptcy law, that dated back to 1943 and stayed in place until 2000. The
authors concluded that, despite many similarities in initial conditions, the reform
of bankruptcy procedures in Chile affected capital accumulation and accelerated
the recovery of the Chilean economy.
both on the incentives to accumulate capital and on the efficiency with which
that capital was accumulated, which explain the faster recovery of Chilean economy.
Efficient corporate bankruptcy law
An important issue in the economic literature on bankrupcy law concerns its optimal design. There is a known trade-off regarding ex-ante and ex-post efficiency
of pro-creditor and pro-debtor legal environments.
From the ex-ante sense, there is a lot of theoretical and empirical work show9
ing that creditor-friendly bankruptcy laws are important to provide incentives.
The main argument is that if bankruptcy is considered sufficiently threatening,
managers are less inclined to take excessive risk, expropriate cash flow from the
firm, or falsely report its true returns, for instance. Also, since pro-creditor laws
usually increase the recovery rate of credit claims in case of default, creditors can
lend at lower rates, increasing the attractiveness of safer projects, and limiting
risk-taking.8
From the ex-post sense, there are different aspects to consider. When default
has already happened, a good bankruptcy law should maximize and preserve firm
value, and also lead to welfare-increasing asset realocation.
An important goal of bankruptcy is to drive inefficient firms out of the market.
But when firms are economically viable, it may be optimal to give managers and
shareholders a second chance. This is the case of firms whose managers have specific knowledge about the business and are the only ones capable of restructuring
the firm. It may also the case of firms that are going through a temporary financial distress and are having difficulties in obtaining financing, due to some sort of
credit market friction.
Thus, a good bankruptcy law must strike a balance, combining pro-creditor and
pro-debtor features. It must also provide the right incentives for both creditors
and debtors, while avoiding inefficient liquidations.
Other important aspects to observe regarding the adoption of Chapter 11 like
legislation are the costs of liquidation and reorganization procedures and the relative share of physical capital in the economy. Araujo and Funchal (2013) solve and
simulate a general equilibrium model with incomplete markets and bankruptcy, in
which the optimal bankruptcy law for the economy will depend not only on the
costs of each procedure, but also on the structure of the productive sector – i.e.,
on how important physical capital is for the economy, relative to variable inputs.
Their model has three types of agents: managers; secured creditors, responsible
for financing the fixed inputs; and unsecured creditors, who sell the variable input.
They compare two types of bankruptcy laws: a pro-reorganization law, in which
managers have the right to choose between a reorganization and a liquidation procedure; and a pro-liquidation law, in which firms in financial distress are promptly
liquidated.
Figure 4 displays the optimal bankruptcy laws, as prescribed by the simulation
8
La Porta et al. (1997, 1998) bring evidences of the key role of creditor protection and debt enforcement in supporting credit markets. They addressed the issue of actually measuring creditor
protection, and used the created measure to analyse forty-nine different countries with respect
to the design of their commercial laws. Since then, a growing number of empirical work have
focused on the importance of certain aspects of the legal environment to promote financial development and growth, with a vast empirical literature bringing evidence of a positive correlation
between creditor protection and credit market development.
10
results from Araujo and Funchal (2013). For sectors intensive in physical capital,
the best law is pro-liquidation, since it permits secured creditors to recover their
claims immediately, making the cost of capital lower. For sectors intensive in
variable input the best law is pro-reorganization, since it gives trade creditors
another chance to recover their credit, making the cost lower. For extremely
high levels of liquidation costs the best law is pro-reorganization, regardless of the
physical capital proportion.
Figure 4: Optimal corporate bankruptcy laws.
See Araujo and Funchal (2013)
Share of
physicalcapital
3/4
C hapter 7
1/2
C hapter11
1/4
1/4
1/2
3/4
(Relative)D epreciation
in liquidation
.
Araujo and Funchal (2013) also conduct an empirical exercise, by using the
estimated value of the bankruptcy cost for the United States9 to assess reorganization and liquidation costs, and the U.S. sectorial spent share of materials and
physical capital10 to calibrate the proportion of physical capital and variable input.
They use data from U.S. industry sector (considered to be industry representative),
and hope to identify the technical component – common to the industry in every
country – of industry physical capital intensity.
The authors use this data to analyze the optimal bankruptcy law for each
country, first by calculating the value added share of each industry sector for
each country (to infer the size of each sector); then they sum the share of each
sector that should have a pro-liquidation (or pro-reorganization) procedure; and,
finally, if the share of the pro-liquidation sectors is bigger than 50%, they consider
9
10
See Bris et al. (2006).
source: NBER-CES Manufacturing Industry Database
11
a pro-liquidation bankruptcy law to be the best for the country. Otherwise, a
pro-reorganization bankruptcy law is considered to be the best for the country.
They find that 27 (out of a sample of 44) countries apply procedures aligned
with their suggestions. Approximately 80% of the countries (35 out of 44) should
apply a pro-reorganization bankruptcy law. They also find that managers keep a
higher proportion of their capital in place when reorganization is available.
These results assume that all countries have bankruptcy costs equal to those in
the United States. If bankruptcy costs are allowed to vary from country to country,
Araujo and Funchal (2013) find that countries with higher costs of liquidation
(such as Brazil, for instance) should move torward a pro-reorganization procedure,
while countries with low liquidation costs (like Sweden) should have a more proliquidation bankruptcy law.
4
Efficient Personal Bankruptcy Law
Each time there is a financial crisis, the debate on debt repayment following bad
states of nature grows. Recent discussion in the media11 emphasizes the consequences of bankruptcy due to job losses and the difficulty for consumers to erase
their debts and get a fresh-start, mainly after the 2005 Bankruptcy Reform Act.
However, in the recent past, the debate centered on the importance of inducing
debt repayment in order to reduce moral hazard. An example is the discussion
on the personal bankruptcy reform, known as Bankruptcy Abuse Prevention and
Consumer Act of 2005 and whose main goal was to stop consumers from using
bankruptcy to walk away from debt they could afford to repay.
The personal bankruptcy law reform is often cited as having contributed to
the increase in subprime mortgage foreclosures. Morgan et al. (2009) argue that
the 2005 Bankruptcy Reform Act contributed to the surge of subprime foreclosures by shifting risk from unsecured credit lenders to mortgage lenders. Before
the bankruptcy reform, any household could file for Chapter 7 bankruptcy and
have credit cards and other unsecured credit discharged. The bankruptcy reform
blocked this possibility by adding a means test to force those households in better
conditions to file for bankruptcy using Chapter 13, where they must continue to
pay unsecured lenders using their future income. Depending on the means test,
cash constrained householders who might have saved their home by filing for Chapter 7 are more likely to face foreclosure. As a result, Morgan et al. (2009) found
a substantial impact of the bankruptcy reform on subprime foreclosure, where the
foreclosure rate during the seven quarters after the reform was 12.6% higher.
11
See for example “Downturn Pushes More Toward Bankruptcy” (The New York Times,
04/03/2009)
12
Several works that formalize theories on private credit show that when lenders
can more easily force repayment they are more willing to extend credit (e.g.
Townsend (1979), Aghion and Bolton (1992),Hart and Moore (1994, 1998)). Dubey
et al. (2005) and Dubey and Shubik (1979) approach the problem through the
debtors’ side, arguing that the degree of punishment of debtors in case of bankruptcy
influences the amount of credit in the market. Dubey and Shubik (1979) show that
when markets are complete, the optimal level of penalty is extremely high. Dubey
et al. (2005) show that in presence of incomplete markets,12 assuming that certain
contingencies cannot be written into contracts, the intermediate level of penalty
that encourages some amount of bankruptcy provides a higher level of individual
credit in the economy.13 Santos and Scheinkman (2001) develop a model of financial intermediation and show that when society punishes default and intermediaries
can impose collateral requirements to limit the size of debt positions.
A simple model Consider a consumer who lives for two periods and maximizes
utility over her consumption c. The consumer is born with some amount of durable
goods of value D (like a house, a car, etc) that she consumes in both periods. The
durable good depreciates at rate δ. Period 1 income w1 is observed but the second
period income is uncertain, varying according to the realization of the state of
nature
w2s ∈ [w21 , ..., w2S ]. Each state occurs with probability ps , where ps > 0 ∀s
ps = 1. The wage is observed by the borrower, but the lender can only
and
s
verify its value at a monitoring cost, denoted by γB. 14
There is a large number of agents divided in two different groups: borrowers
and lenders. Borrowers can be thought of as consumers; the financial institutions
that offer standard debt contracts are the lenders.15 Each lender is endowed with
enough money to supply credit to consumers. These lenders’ endowments may
be used either to lend to a borrower with rate r, or to purchase a risk-free asset
paying an exogenously given rate of return rf .
If the borrowers report bankruptcy, part of the debt will be discharged, and
12
The standard debt contract (non-contingent repayment of principal plus interest) that is
usually offered to individuals and small businesses makes the market incomplete, since there is
no contract that is offered contingent on the successful states of nature.
13
Araujo et al. (1998) generalize the result proving the equilibirum existence with a continuum
of states and without imposing any assumptions on ex-post endowments.
14
A similar model applies to small businesses. Suppose that, instead of a two period economy,
there is only one time period in which the small firms’ owners choose an amount B to invest in
their project. The output ws B α is uncertain, since it depends on the realization of the state of
nature ws . In this setup, we reach the same results as those in the consumption model.
15
Townsend (1979) and Williamson (1986, 1987) show that the standard debt contract is the
optimal contract for competitive financial markets. Ying Yan (1996) shows that the standard
debt contract is the optimal debt contract for non-competitive financial markets.
13
some of the individuals’ assets, including personal goods (D) and their current
income will be exempted up to an amount E. The bankruptcy law determines the
level of E exogenously, so we call E the bankruptcy exemption level in this paper.
Debt contracts are subject to this bankruptcy law. Notice that part of borrowers’
goods serves as an informal collateral imposed by the law to unsecured credit.
Definition 1 Strategic bankruptcy16 : It occurs when the borrower has enough
wealth to pay her debts but she chooses not to do it.
Definition 2 Bankruptcy by bad fortune: It occurs when the realization of states
of nature is bad in such way that borrowers are unable to fulfill their repayment promises.
Consumption in the first period defines the level of debt B at the beginning of
period 2:
B = (c1 − D − w1 ),
which means that the agent consumes more than the sum of her wage and durable
goods.
A loan contract between the borrower and the lender consists of a pair (r, B),
where B is the loan size and (1 + r) is the loan rate, subject to the legal imposition
on the exemption level E. It applies to the situation in which the borrower does
not repay the debt (1 + r)B.
If at least some debt will be held, so that B > 0, we can divide the borrowers’
actions in three distinct choices:
C1 does not file for bankruptcy if:
w2s + δD ≥ (1 + r)B
and (1 + r)B ≤ max(w2s + δD − E, 0)
C2 strategic bankruptcy if:
w2s + δD ≥ (1 + r)B
and (1 + r)B > max(w2s + δD − E, 0)
C3 bad fortune bankruptcy if:
w2s + δD < (1 + r)B and theref ore (1 + r)B > max(w2s + δD − E, 0))
16
Moral hazard is relevant because borrowers have a choice not to repay their debts.
14
It is optimal for consumers to file for bankruptcy if and only if their gains in
bankruptcy are bigger than their gains when they choose not to file for bankruptcy,
i.e., if and only if (1 + r)B > max(w2s + δD − E, 0). That is, the consumer will
default whenever her debt in the second period exceeds the level of assets that
can be seized and whenever her debt cannot be fully enforced. So, the consumer
delivers min[(1 + r)B, max(w2s + δD − E, 0)].
We can write the probability of no
ps ιs (1 − ιd ) and the probability of
bankruptcy as (1 − pbankruptcy ) = p(C1) =
s bankruptcy as pbankruptcy = p(C2) + p(C3) =
ps [ιs ιd + (1 − ιs )], where ιs = 1
s
if w2s + δD ≥ (1 + r)B and ιd = 1 if (1 + r)B > max(w2s + δD − E, 0).
The wealth in each situation for the borrowers is given as follows:
w2 + δD − (1 + r)B
if no bankruptcy
W2 =
w2 + δD − max(w2s + δD − E, 0) if bankruptcy
Thus the lender can receive in case of bankruptcy a payment between w2s +
δD (if the bankruptcy exemption is zero) and zero (if the bankruptcy exemption
overcomes the debtors’ wealth in the second period).
The expected return on lending must be no less than the risk-free return.
Therefore, the lender’s participation constraint is given by:
(1 + rf )B ≤
s
+
ps ιs (1 − ιd )(1 + r)B +
ps [ιs ιd + (1 − ιs )] [max(w2s + δD − E, 0) − γB]
(1)
s
The extra interest rate paid, r − rf , is exactly the one needed to offset the loss
incurred by the financial institution when the consumer goes bankrupt. It is the
same as a risk premium.
Given a menu of contracts, the consumer chooses a pair (r, B) that maximizes
her expected utility function:
max u(c1 ) + Eu(c) = u(c1 ) + θ
S
(r,B)
s=1
ps u(c2s )
c1 = w 1 + D + B
c2s = w2s + δD − min[(1 + r)B, max(w2s + δD − E, 0)] ∀s
The constraint (1) is always valid with equality, since a smaller rate of return
r makes the borrower strictly better and still satisties the lender’s participation
15
constraint. Also, since the lender pays the monitoring cost to verify the wage value
(w), the contract specified above is such that in bankruptcy states borrowers have
no incentives to falsely report the state of nature.
Observe that the lenders’ expected return, described by their participation
constraint, determines the supply of credit in the economy. The supply of credit
depends directly on the punishment level imposed by the local legislation.
Proposition 1. Any value of exemptions above the critical value E* makes the
supply of credit to individuals zero.
Proof: See Araujo and Funchal (2005a).
Proposition 2. As the bankruptcy exemption decreases, the interest rate charged
to individuals reduces.
Proof: See Araujo and Funchal (2005a).
Differently from the supply side, if the bankruptcy exemption increases (reducing the debtors’ punishment), the consumer has more incentive to demand credit.
This happens because the cost of building another asset that is more aligned with
debtors’ interests decreases, since they can keep a bigger amount of their personal
goods if bankruptcy occurs. Such asset – that allows debtors to file for bankruptcy
at a cost lower than the bankruptcy exemption – acts as a substitute of the original
debt contract. At the limit, if the exemption is unlimited, the cost of bankruptcy
goes to zero, making the demand for credit even more attractive. On the other
hand, if the bankruptcy exemption goes to zero, individuals can lose everything
they have in case of a bad realization of the sate of nature, inhibiting their demand
for credit.
Proposition 3. As the bankruptcy exemption falls, the individuals’ demand for
credit decreases.
Proof: See Araujo and Funchal (2005a).
Therefore, there are two distinct forces acting in the proposed problem. A
decrease in E expands the supply of credit, thus reducing the interest rate charged
to borrowers, since the chances of creditors being repaid are bigger, and they
receive more in bankruptcy states. On the other hand, the demand is repressed
since the debtors fear the punishment for losing their goods. With an increase in
E, there is an incentive for consumers to demand credit, since they can build assets
aligned with their needs. On the other hand, such level of exemption inhibits the
lenders’ supply of credit, since the chances and the amount of repayment fall.
The equilibrium level of credit provided by extreme levels of bankruptcy exemption (0 or unlimited) tends to be very low or even zero. An optimal level
16
of bankruptcy exemption E ∗∗ may exist where the the equilibrium of supply and
demand of credit provide a higher level of credit and welfare in the economy.
Empirical Findings Araujo and Funchal (2005a) analyzed empirically the relationship between the degree of debtors punishment and personal credit market
development. To address this topic, the authors used the U.S. personal bankruptcy
exemptions – that vary across U.S. states – to construct a proxy of debtor’s punishment in case of default. By looking at the changes in the degree of debtors’
punishment (across time and states), they show whether and how these changes
are linked to credit market development. Using a two-way fixed effects panel regression to estimate the relationship between debtors’ punishment and a measure
of credit market development, they found that this relationship is not always increasing, and there is actually an intermediate level of punishment that is optimal
for credit market development. Figure 5 displays how the amount of personal
credit in the economy changes in response to the level of debtor punishment.
States with extreme levels of punishment (high or low) tend to have a lower volume of credit relative to states with intermediate levels of protection. Thus, their
main results suggest that punishment applied by bankruptcy legislation should
neither be so harsh that it inhibits credit demand nor so lenient that it reduces
credit supply.
Figure 5: Optimal amount of debtor punishment.
See Araujo and Funchal (2005a)
0,008
Credit Card
Other Personal Loans
0,005
0,003
0,000
-0,003
0,0
0,1
0,2
-0,005
Amount of credit_normalized
-0,008
-0,010
.
Debtors' Punishment
17
0,4
0,5
5
The case of Brazil
Since the 1990s, several Latin American countries have gone through legislative
reforms that changed the legal framework governing corporate bankruptcy. Brazil
was one these countries. In 1993, Brazilian Congress initiated an effort to reform
the Brazilian corporate insolvency legislation. In December 2004, a new law was
finally approved.
The former Brazilian bankruptcy law was part of a very fragmented legal framework governing corporate insolvency, most of which was enacted in the 1940s.
The insolvency process provided both liquidation and reorganization procedures but was very ineffective in preserving asset values, in protecting creditor
rights in liquidation – leading to an increased cost of capital – and in successfully
enabling viable distressed firms to reorganize themselves. Liquidation was characterized by severe inefficiencies; reorganization – called concordata (composition
with creditors) – was extremely rigid and usually unable to provide meaningful
options for the firm to recuperate itself. There was no formal renegotiation between parties, and the procedure also incentivized an informal use of the system
to promote consensual renegotiations, notwithstanding an insufficient legislative
framework capable of fostering workouts.
Figure 6: Average length of insolvency procedures.
See Araujo et al. (2012a).
11
10
9
8
7
6
5
4
3
2
1
0
OECD
ECA
EAP
SSA
LAC
MENA
SAS
Brazil
Resolving insolvencies used to take, on average, ten years to complete, making
Brazil the country with the slowest insolvency procedures in the world.
18
Figure 6 displays the average length of insolvency procedures in Brazil and
in seven groups of countries: the Organization for Economic Cooperation and
Development (OECD), Latin America and the Caribbean (LAC), the Middle East
and North Africa (MENA), Europe and Central Asia (ECA), East Asia and the
Pacific (EAP), South Asia (SAS) and sub-Saharan Africa (SSA).17 Notice that
the average time to close a business in Brazil was more than twice the average
for Latin America. This situation eroded the value of assets and thus lowered the
amount received by creditors.
The process of liquidating the firm’s assets was usually characterized by procedural inefficiency, lack of transparency and the so-called “succession problem”18 ,
whereby tax, labor and other liabilities were transferred to the buyer of the liquidated asset. This liability transfer had the undesirable effect of depressing the
market value of an insolvent company’s assets.
Another important shortcoming was the bankruptcy priority rule, that was
very punitive to creditors. It estipulated that labor and tax claims had priority
over all other types of creditors – including secured creditors.
As a consequence of the weak protection given to creditors of insolvent firms,
financial markets were characterized by a relatively low credit volume and high
interest rates. Distorted incentives and the lack of effective mechanisms to support corporate restructuring resulted in disproportionately high default rates of
potentially viable companies. Exit costs were increased for nonviable companies;
productivity and employment were reduced.
Creditor recovery rates from before the reform illustrate the ultimate effect of
an inefficient procedure with poor creditor protection. Until 2005, the recovery
rate in case of bankruptcy in Brazil was a mere 0.2 cents on the dollar, while the
averages for Latin American and OECD countries were 26 and 72 cents, respectively. The main reason for such low recovery was the beforementioned priority
order, since creditors ranked behind labor and tax claims. After workers and the
Government were paid, the remaining amount was usually insignificant or even
nil.
The ex-ante effect of creditors receiving close to nothing from insolvent firms
was an increase in the interest rate charged to solvent firms. Before the new law,
Brazil had an extremely high interest rate spread (49%), which was more than
four times larger than the average spread for Latin American countries (11%), and
more than twelve times larger than the average for OECD countries (3.87%). All
values refer to the 1997—2002 period. 19
17
The Latin American and Caribbean block is composed of Argentina, Bolivia, Brazil, Chile,
Colombia, Costa Rica, the Dominican Republic, Ecuador, El Salvador, Guatemala, Haiti, Honduras, Jamaica, Mexico, Nicaragua, Panama, Paraguay, Peru, Uruguay, and Venezuela.
18
Problema da sucessão, in Portuguese.
19
See Araujo and Funchal (2005b)
19
On June 2005, six months after being signed into law, the new bankruptcy
legislation (Law 11,101/05) came into effect.20 The new law integrated the insolvency system into the country’s broader legal and commercial systems, which
was an improvement. The new law also provided in- and out-of-court options to
reorganize.
The new reorganization procedure – called “judicial reorganization”21 – was
mostly inspired by Chapter 11 of the U.S. bankruptcy code. The new law’s aim
was to strike a reasonable balance between liquidation and reorganization. It also
significantly improved the flexibility of the insolvency legal system, by permitting
the conversion of reorganization proceedings into liquidation.
20
21
See Araujo and Funchal (2005b) and Araujo et al. (2012a).
Reorganização judicial, in Portuguese.
20
Figure 7: Evolution of bankruptcy related variables before and after the reform.
See Araujo et al. (2012a).
7-A: Liquidation Requests before and after the Reform
2,500
7-B: Reorganization Requests before and after the Reform
120
100
2,000
Pre-Reform Post-Reform
Pre-Reform
Post-Reform
80
1,500
60
1,000
40
500
21
20
jan-03oct-03 aug-04 jun-05 apr-06 feb-07 dec-07 oct-08 aug-09
Month and Year
jan-02
oct-02
7-C: Liquidation and Reorganization Requests after the Reform
1,400
20
jul-03 apr-04 jan-05
Month and Year
oct-05
jul-06
7-D: Creditors Recovery Rate
1,200
15
1,000
Liquidation
Reorganization
800
10
600
400
5
200
jun-05
apr-06
feb-07
dec-07
Month and Year
oct-08
aug-09
2003
2004
2005
2006
2007
2008
2009
Concerning the liquidation and the reorganization procedures, some of the key
changes brought by the new Brazilian bankruptcy law are listed below:
1. Change in the priority order in liquidation. Labor credits are now given
priority up to an amount equal to 150 times the minimum monthly wage to
each worker; residual amounts go to the near-end of the line, together with
unsecured creditors; increased priority to secured creditors, giving secured
claimers priority over tax credits; increased priority to unsecured creditors.
2. Facilitated disposing of the firm’s assets: the distressed firm may be sold
(preferably as a whole) before the creditors’ list is formed.
3. End of the “succession problem”: labor, tax and other liabilities are no longer
transferred to the buyer of an asset sold in liquidation.
4. Restriction to the use of bankruptcy for creditors holding low value claims:
only creditors holding claims higher than forty times the minimum monthly
wages can initiate a bankruptcy procedure.
5. Incentives for lenders of firms under reorganization: any new credit extended
during the reorganization process is given first priority in the event of liquidation.
6. Automatic stay period: firms under reorganization are granted an automatic
injunction protecting them from secured creditors that seek to seize the debt
collateral.
Both the upper bound for labor claims and the increased priority of creditors
had a direct impact on expected recovery rates. The second and third changes,
in turn, have increased the value of the firm in bankruptcy, by speeding up the
firm’s value in the event of bankruptcy and dissociating the firm’s assets from its
liabilities.
Another relevant change concerns the new reorganization procedure. Whereas
under the previous law no renegotiation between the interested parties was allowed
and only a few parties were entitled to claim assets, now management makes
a sweeping proposal for recuperation that must either be accepted by workers,
secured creditors and unsecured creditors (including trade creditors) or the distressed firm will be liquidated. Therefore, creditors were empowered and play a
more significant role in the current procedure than in the previous one, including negotiating and voting for the reorganization plan. These changes motivate
creditors to participate more actively in the bankruptcy process.
As stated by Araujo and Funchal (2009), this new design of Brazilian bankruptcy
procedures brings new incentives to the interested parties. The incentive for
22
debtors to default strategically is reduced, for mainly two reasons: first, the conditions under which debtors can file for bankruptcy are now limited to those
prescribed by law; and second, a reorganization procedure can be converted to
liquidation at creditors’ discretion, a feature that nearly eliminates the use of
reorganization as a bargaining mechanism. 7 Figure 7 shows the number of liquidation requests and the number of reorganization requests, both before and after
the Brazilian bankruptcy reform. Notice that in both cases the number of requests
dropped abruptly after the reform, which is in line with what would be expected
as a consequence of reducing incentives to strategic defaults.
Creditors, on the other hand, now face new incentives to actively participate
in the bankruptcy procedures, due to three key changes: first, creditors now play
a more significant role in the procedures than they previously did, including negotiating the reorganization plan and then voting on its approval; second, they can
file for out-of-court reorganization; and third, their credits’ priority in case of liquidation is higher now than it was under the previous legislation. Figure 7C shows
an increase in the number of reorganizations in periods of crisis – like the mortgage crises period – when the liquidity default problems were more pronounced.
In this case, the number of liquidation requests, which was previously more than
ten times greater than reorganization requests, dropped to practically the same as
the number of reorganization requests. This illustrates the more active behavior
of creditors.
Figure 8: Ratio between corporate private credit and GDP.
See Araujo et al. (2012a).
200
180
private credit-to-gdp ratio (2003 as base 100)
Pre-Reform
Post-Reform
160
140
120
100
80
60
Brazil
Argentina
Chile
Mexico
40
20
jan-01
nov-01
sep-02
jul-03
may-04
mar-05
nov-06
jan-06
Month and Year
23
sep-07
jul-08
may-09
mar-10
Araujo et al. (2012a) study the effect of the Brazilian bankruptcy law reform
by employing some variations of a difference-in-differences setup to analyse the
behavior of a few debt related variables. Looking at firm accounting data on 698
publicly traded firms, ranging from 1999 to 2009, they find a reduction in the
cost of debt, an increase in total debt, and an even larger increase in long-term
debt, following the reform in Brazil. Since secured creditors have benefited more
from the new law than unsecured ones, the effect is more pronounced on longterm debt, which is known to be more correlated with secured debt. They find
no statistically significant effect on short-term debt or loan ownership structure.
Araujo et al. (2012a) also look at aggregate data, contrasting the time series of the
private credit-to-GDP ratio in Brazil with those of Argentina, Chile and Mexico.
Figure 8 brings information on this variable for these four countries, with January
2003 normalized to 100, since this is the period when information is available for
all countries. It shows that around mid-2005 all series are close to each other, and
as time progresses Brazil detaches itself from the other countries, suggesting some
correlation between the reform and the expansion of private credit.
6
Conclusion
The main goal of this paper is to approach important elements that contribute to
financial development and stability, such as collateral regulation and bankruptcy
laws.
To approach the financial market regulation subject, in times of financial crises,
we presented the Araujo et al. (2012b) discussion on collateral regulation. One of
the most interesting results of the paper is that it is never optimal to regulate the
market for subprime loans.
Moving to the bankruptcy law design, we divided the discussion in three distinct topics: first, the corporate bankruptcy law; second, the personal bankruptcy
law; and finally, the Brazilian case of the bankruptcy law reform.
Looking to the corporate bankruptcy law in the ex-ante sense, there is a lot
of theoretical and empirical work showing that creditor-friendly bankruptcy laws
are important to provide incentives; from the ex-post sense, when firms are economically viable, it may be optimal to give managers and shareholders a second
chance. Similarly, the design of a bankruptcy procedure for individuals should balance the incentive to repay debts to reduce moral hazard and the need to protect
borrowers from the bad state of nature, allowing them a fresh-start. Araujo and
Funchal (2005b) addressed this issue empirically and show that the relationship
between debtors’ punishment to protect creditors against moral hazard and the
size of the credit market is not always increasing. Therefore, the intermediate level
of punishment is optimal for credit market development.
24
Finally, we presented a sucessful case of bankruptcy law reform. Araujo et al.
(2012a) study the effect of the Brazilian bankruptcy law reform on firms debt
contract terms. Looking at firm accounting data the authors found a significant
improvement in the terms of debt, lower price and higher volume of credit, following the reform in Brazil.
25
References
Aghion, P. and Bolton, P. An incomplete contracts approach to financial contracting. The Review of Economic Studies, 59(3):473–494, 1992.
Araujo, A. and Funchal, B. How much debtors punishment. Ensaios Econômicos
da EPGE, 2005a.
Araujo, A. and Funchal, B. Bankruptcy law in latin america: past and future.
Economia, 6(1):149–216, 2005b.
Araujo, A. and Funchal, B. A nova lei de falências brasileira: primeiros impactos.
Revista de Economia Polı́tica, 29(3):191, 2009.
Araujo, A. and Funchal, B. Bankruptcy law and credit market: A generalequilibrium approach. Fucape Working Paper Series, (39), 2013.
Araujo, A.; Monteiro, P., and Páscoa, M. Incomplete markets, continuum of states
and default. Economic Theory, 11(1):205–213, 1998.
Araujo, A.; Ferreira, R.., and Funchal, B. The brazilian bankruptcy law experience.
Journal of Corporate Finance, (0):–, 2012a. doi: 10.1016/j.jcorpfin.2012.03.001.
Araujo, A.; Kubler, F., and Schommer, S. Regulating collateral-requirements when
markets are incomplete. Journal of Economic Theory, 2012b.
Araujo, Aloisio; Páscoa, Mário Rui, and Torres-Martı́nez, Juan Pablo. Collateral
avoids ponzi schemes in incomplete markets. Econometrica, 70(4):1613–1638,
2002.
Bergoeing, R.; Kehoe, P.; Kehoe, T., and Soto, R. A decade lost and found:
Mexico and chile in the 1980s. Review of Economic Dynamics, 5(1):166–205,
2002.
Bris, A.; Welch, I., and Zhu, N. The costs of bankruptcy: Chapter 7 liquidation
versus chapter 11 reorganization. The Journal of Finance, 61(3):1253–1303,
2006.
Djankov, S.; McLiesh, C., and Shleifer, A. Private credit in 129 countries. Journal
of financial Economics, 84(2):299–329, 2007.
Dubey, P. and Shubik, M. Bankruptcy and optimality in a closed trading mass
economy modelled as a non-cooperative game. Journal of Mathematical Economics, 6(2):115–134, 1979.
26
Dubey, P.; Geanakoplos, J., and Shubik, M. Default and punishment in general
equilibrium1. Econometrica, 73(1):1–37, 2005.
Geanakoplos, J. Promises promises. Cowles Foundation Discussion Papers, 1996.
Geanakoplos, J. and Zame, W. Collateralized asset markets. University of California at Los Angeles working paper, 2007.
Hart, O. and Moore, J. A theory of debt based on the inalienability of human
capital. The Quarterly Journal of Economics, 109(4):841–879, 1994.
Hart, O. and Moore, J. Default and renegotiation: A dynamic model of debt. The
Quarterly Journal of Economics, 113(1):1–41, 1998.
Johnson, S.; Boone, P.; Breach, A., and Friedman, E. Corporate governance in
the asian financial crisis. Journal of Financial Economics, 58(1):141–186, 2000.
La Porta, R.; Lopez-de Silanes, F.; Shleifer, A., and Vishny, R. Legal determinants
of external finance. Journal of Finance, 52(3):1131–50, 1997.
La Porta, R.; Lopez-de Silanes, F.; Shleifer, A., and Vishny, R. Law and finance.
The Journal of Political Economy, 106(6):1113–1155, 1998.
Miller, Marcus and Stiglitz, Joseph. Leverage and asset bubbles: Averting armageddon with chapter 11?*. The Economic Journal, 120(544):500–518, 2010.
Morgan, D.; Iverson, B., and Botsch, M. Seismic effects of the bankruptcy reform.
FRB of New York Staff Report, (358), 2009.
Santos, T. and Scheinkman, J. Competition among exchanges. The Quarterly
Journal of Economics, 116(3):1027–1061, 2001.
Townsend, R. Optimal contracts and competitive markets with costly state verification. Journal of Economic Theory, 21(2):265–293, 1979.
Williamson, S. Costly monitoring, financial intermediation, and equilibrium credit
rationing. Journal of Monetary Economics, 18(2):159–179, 1986.
Williamson, S. Costly monitoring, loan contracts, and equilibrium credit rationing.
The Quarterly Journal of Economics, pages 135–145, 1987.
Ying Yan, E. Credit rationing, bankruptcy cost, and optimal debt contract for
small business. Federal Reserve Bank of Cleveland (mimeographed), 1996.
27