| Buy the book! It has so much more .... at Amazon, Game Theory: A Non-Technical Introduction to the Analysis of Strategy at Barnes and Noble -- of course! Game Theory: A Non-Technical Introduction to the Analysis of Strategy |
Bankruptcy is badly understood in modern economics. This is equally true at the most elementary and most advanced levels, but, of course, the sources of confusion are different in these different contexts.
For the elementary student, there is the tendency to confuse bankruptcy, the decision to shut down production, and "going out of business," that is, liquidation. The undergraduate textbook encourages this, since it considers only the shut-down decision, and the timeless model usual in the undergraduate textbook makes the shut-down decision appear to be an irreversible one. The textbook discussion of the shut-down observes that the business will shut down if it cannot cover its variable costs, and this illustrates a point about opportunity costs -- fixed costs are not considered because they are not opportunity costs in the short run. Bankruptcy occurs when the firm cannot, or will not, cover its debt service payments: quite a different thing. Debt service costs are usually thought of as fixed, not variable costs.
In real businesses, of course, bankruptcy, liquidation, and shut-down are three quite different things that may appear in various combinations or entirely separately. A business may be reorganized under bankruptcy and continue doing business with the former creditors as equity owners -- neither shut down nor liquidated. The business that shuts down may not be bankrupt -- it may continue to make debt service payments out of its cash reserves and resume production when conditions merit. And a company may be liquidated, for example at the death of a proprietor, although it is able to cover its variable costs and its debt service payments (although this will only occur when the transaction costs of finding a buyer are so high as to make sale of the business infeasible).
Small wonder, then, that the undergraduate economics student finds the shut-down analysis a little confusing -- it abstracts from almost everything that matters! But more advanced economists will find bankruptcy confusing for another reason. The reason is related to the phrase "the firm cannot, or will not, cover its debt service payments." We may think of a lending agreement as a solution to a cooperative game, that is, a game in which both players commit themselves at the outset to coordinated strategies. The repayment of debt service is the strategy the firm has committed itself do. For the firm to fail to pay its debt service contradicts the supposition that the firm had, at the first instance, committed itself. And the creditors are letting the firm out of its contract, and they are losing by that, and why should they do it? It seems that we must fall back on the first part of the statement: the firm cannot make its debt service payments. Some unavoidable (but not clearly foreseen) circumstance makes it impossible for the debt service to be paid. We then interpret the debt contract as a commitment to pay "if possible," or with some other such weasel-words, and we understand why the creditor capitulates: she or he has no choice.
But how can it be that "the firm cannot" pay its debt service? We need to make our picture a little more detailed.
First, uncertainty clearly plays a part in it. If bankruptcy were certain, there would be no lending. Accordingly, we represent uncertainty in the usual way in modern economics: we suppose that the world may realize one of two or more states. At the outset, the state of the world is not known. After some decisions and commitments are made, the state of the world is revealed, and some of the decisions and commitments made at the first stage must be reconsidered. Bankruptcy is such a reconsideration of commitments made in ignorance of the state of the world: it occurs only in some states of the world, and the payoff to the lender in the other states is good enough to make the deal acceptable as a whole.
Second, we must be a little more careful about just who "the firm" is, since it is a compound player. Let us adopt the John Bates Clark model of the business enterprise, and of the market economy, as a first approximation. In this model there are capitalists (lenders, for our purposes), suppliers of labor services, that is workers, and "the entrepreneur," who owns nothing and whose services are those of coordination between the other two groups.
With these specifics in mind, let us return to the shut-down decision as it is portrayed in the intermediate microeconomics text. What leads "the firm" to shut down? What happens is that the state of the world realized is a relatively bad one. That is, the conditions for production and/or demand are poor, so that the enterprise is unable to "cover its variable costs." In other words, it is unable to pay the workers enough to keep them in the enterprise. The key point here is that the workers have alternatives. The revenue of the enterprise is so little that, even if the workers get it all, they do not make as much as they would in their best alternatives. Saying "the firm cannot cover its variable costs" is a coded way of saying "the firm cannot recruit labor with its available revenues." In such a case, there is clearly no alternative to shutting down.
But, as we have observed, a firm may go bankrupt but not shut down, instead continuing to produce under reorganized ownership. How would this occur? The state of the world is not quite as bad: the enterprise can earn enough revenue to pay its workers their best alternative wages, but having done that, there is not enough left to pay the debt service. The entrepreneur has only two choices: to cut the wages below the workers' best alternative pay, lose them all, produce nothing, and default on all of the debt service; or to pay the workers at their best alternative, produce something, and pay something toward the debt service. Clearly, the latter is in the interest of the lenders, so they renegotiate the note.1
In all of this, "the entrepreneur" has played a passive role. John Bates Clark's "entrepreneur" is not much of a player, from the point of view of game theory, anyway. His role is to combine capital and labor in such a way as to maximize profits. In effect, he is an automaton whose programmed decisions define the rules of a game between the workers and the bankers. At the point of bankruptcy, his role is even less active. The choices and commitments are made by the substantive players: capitalists and workers. The essence of bankruptcy is a game played between a lender and a group of workers. We may as well eliminate the entrepreneur entirely, and think of the firm as a worker-cooperative.2 From here on, we shall follow that strategy.
To make things more explicit still, let us consider a numerical example. The players are, as before, a banker and a group of workers. If the banker lends and the workers work, the enterprise can produce a revenue that depends on the state of the world. there are three states. The best state is the "normal" one, so we assign it a probability of 0.9. The other two states are bad and worse -- a bankruptcy state and a shut-down state -- with probabilities of 0.05 each. Thus production possibilities are as shown in Table 18-1.
| state | revenue | probability |
| 1 | 3000 | 0.9 |
| 2 | 2000 | 0.05 |
| 3 | 1000 | 0.05 |
We suppose that the safe rate of return (opportunity cost of capital) is .01, and that the lender, being profit oriented, offers a loan of 1000 to enable production to take place. The contract rate of interest is 10%; i.e. 1100 has to be paid back at the end of the period. We suppose, also, that the workers can get an alternative pay amounting to 1500.
If the loan is made, the state of the world is revealed, and then the participants reconsider their strategy choices in the light of the new information. Should the bank make the loan? Should the workers' cooperative accept it? We shall have to consider the various outcomes and then apply "backward induction" to get the answer.
What then happens in state 3? The answer is that in state 3, the members of the cooperative all resign in order to take their best alternative opportunities, at 1500>1000, so that the cooperative spontaneously ceases to exist, and the lender gets nothing.
What about state 1? The enterprise revenue is enough to pay the 1100 in debt service, and the workers' income, 1900, is more than their best alternative, so they do stay and produce, and both the bank and the workers' cooperative are better off.
We now turn to the pivotal state 2. Here, there is enough revenue to pay the debt service, but if it is paid, the workers get only 900<1500. In such a case, again, the worker-members of the cooperative will resign, and the cooperative dissolve for lack of members, and the bank will get nothing. On the other hand, if the bank renegotiates for partial repayment of 500 or less, then the workers get 1500 and the cooperative continues. Thus, in this state, the bank renegotiates and earns 500.
The bank's expected repayment thus is
.9(1100) + .05(500) + .05(0) = 1015 > 1010
Thus the bank makes more than its best alternative and will accept the contract. As for the workers in the cooperative, they make a mathematical expectation of
.9(1900) + .05(1500) + .05(1500) = 1860 > 1500
And so they, too, accept the contract. Thus the loan is made, despite a .05 probability of bankruptcy and a .05 probability of outright default.
In many games of this kind one or another player can obtain a better result if he can commit himself credibly at the outset to a strategy which may seem less advantageous, once the state of the world is known and others have made their decisions. Would the bank be better off it if could commit itself not to renegotiate? The answer is that it would not. Its payoffs would be
.9(1100) + .05(0) + .05(0) = 990 < 1010
The lenders would be worse off and, if (for example) statute law forbade them from renegotiating, they would refuse to make the loan!
But what about the workers? It is their desertion that leads the enterprise to be abandoned if the debt service is paid in state 2. What if they could be somehow bound to the firm? Slavery offers one possibility. In a system that permits slavery, "the entrepreneur" might buy slaves instead of hiring free workers. In state of nature 3, "the entrepreneur" would rent out the slave work force for 1500, pay the 1100 debt service, and pocket the profits (assuming the cost of food necessary to keep the slaves productive is less than 400). In state 2, "the entrepreneur" would require the slaves to work in the firm, produce 2000, pay the debt service, and pocket 900 less the cost of their food. The bank would get its debt service in every state (barring slave starvation) and might well prefer to lend to slavemasters rather than worker cooperatives or John Bates Clark style firms.
In the context of the John Bates Clark firm, the desertion of the workers in states 2 and 3 comes as no surprise to us -- the workers are hired by "the entrepreneur" at mutual convenience and are expected to leave whenever it benefits them to do so. In this example, however, the loan is made to a cooperative association of the workers, their own association. If it were made to them individually, they would be no less responsible for it after they had moved on to their other, better-paying jobs. But the obligation to pay the loan has been assumed by a group of workers, as a group, and the group can continue to exist only so long as it is in the interest of the workers as individuals for it to do so. And this does not reflect the constitution of the firm, but the liberal constitution of society, that holds that no agency, even one constructed by the workers, may require a person to work without offering a payment sufficient to get the worker's assent.
And this is the essence of the case for the proprietary or corporate enterprise as well. The proprietor or investor-owned corporation is no more than a middleman between a group of workers and a bank, so far as bankruptcy is concerned. The essence of bankruptcy is a renegotiation of the loan contract between a lender and a group of workers, and laws exempting the creditor from the full amount of the debt, in appropriate circumstances, are laws for the protection of the creditors, not of the debtors.