The sequels to the East Asia crisis of 1997 have demonstrated, once again, how strongly financial variables affect the business cycle. In spite of the recent advances in the area (see literature review below), there still exists a wide gap between the complex facts and abstract theory. This paper aims to narrow this gap by exploring the business cycle implications of default, liquidations, and bankruptcy law in a dynamic general equilibrium setting. Our analysis builds on elements taken from the corporate finance literature. We find that in the presence of financial imperfections the effect of liquidations on the price of capital goods first highlighted by Shleifer and Vishny (1992) and recently supported by some empirical studies can generate endogenous fluctuations in output, investment, and indicators of financial distress that reproduce some important aspects of actual business cycles.
Our model has overlapping cohorts of entrepreneurs. Each entrepreneur has a project that is active across a start0up period and a production period. All projects yield income with the same present value, but some may suffer a shortage of cash as revenue is delayed until the end of the production period. Yet, should contracts be complete, no project would ever be liquidated and the economy would converge to a stationary equilibrium right away. This result changes dramatically in the presence of capital markets imperfections: both liquidity and the market price of the capital goods (henceforth, machines", for brevity) become relevant to the relationship between entrepreneurs and financiers. In this respect, we incorporate the insights of Hart and Moore (1998) whereby, if income flows are observable but not verifiable, payments to financiers are enforced by providing them with the right to foreclose a project in default. In particular, we follow Bolton and Scharfstein (1996) to model the link between external financing and the extent to which liquidation threats are necessary for enforcement: the higher the need for external financing, the higher the probability of liquidation conditional upon default. Hence, some of the projects which suffer a liquidity shortage will be liquidated, which will entail a significant dead weight loss at project level and a negative impact on the price of machines at market level.
We show that equilibrium cycles (with a periodicity of two) can emerge in this setup. The economy will feature an alternating sequence of bust and boom periods. A bust period is characterized by a large number of liquidations, which depresses the price of machines. GDP is low due to deadweight losses and the low production of new machines induced by the abundance of older ones in the second hand market. Conversely, a boom period is characterized by few liquidations, high GDP, and high machine prices. To see how the alternation of bust and boom periods follows, consider an entrepreneur who starts up in the boom. Clearly he needs a lot of external finance since he is buying the machine at a high (boom) price and may be forced to liquidate it at a lower (bust) price. The high financing needs will push the boom start up into a financial contract characterized by a high probability of liquidation in case of default. Since all start0ups in the boom cohort are at the same position, liquidity shortages will produce a relatively high aggregate number of liquidations in the next period, moving the economy from a boom to a bust. Once in the bust, a symmetric argument explains why the bust will be followed by a boom, and so on.
Such an economy has a stationary equilibrium as well, but it is not stable. Hence, if the initial conditions are just slightly away from the steady state, the economy will develop oscillations that will converge, eventually, to the limit cycle described above. Along the process, there is a positive relationship between liquidations and the capital losses due to the fluctuations in the price of machines. Through these prices, liquidations become increasingly correlated across firms, in the sense that the probability of liquidating an individual project grows with the number of projects that are expected to be liquidated at the same time. Crucially, agents are pushed towards such correlated liquidations by market forces. This result points out at the dramatic effect imperfect capital markets may have upon a competitive equilibrium. While standard complete markets economies feature a tendency to smooth" economic activity over the cycle, the very opposite may happen when capital markets are imperfect. To confirm the practical relevance of these results, we check through simulations whether an equilibrium cycle can be supported by parameters of a plausible order of magnitude. Our results show that this is indeed the case.
Two additional results complete the list of findings of the paper. The first is about the timing of credit rationing. It is shown that in case some cohorts are credit0rationed along the cycle, these are boom start ups. The reason is that boom start0ups buy machines at high prices (and expect lower capital gains on them) so external financing poses a larger and more costly burden on them. This somewhat surprising result finds some support in recent empirical evidence. The second finding refers to the impact of bankruptcy law on business cycles. We show that if the law shifts bargaining power from financiers to corporate borrowers (like in a debtor oriented bankruptcy code such as Chapter 11) the economy becomes more cyclical. Hence, though smoothing the business cycle might have been one of the intentions of Chapter 11, it happens to actually operate in the opposite direction. The financiers foresee the greater difficulties of enforcing repayments on entrepreneurs, and demand more liquidation rights. As these rights are exercised on liquidity short projects, they increase business fluctuations.
Download
A Stylized Model of Financially Driven Business Cycles
