Problem Set 2 -- Solutions

Due April 25, 1997

1.    In Jensen and Meckling's piece, debt financing serves to provide entrepreneur/managers incentives toward non-value-maximizing behvior.

Entrepreneurs using debt financing do not fully bear the downside risks.  They have an incentive to undertake riskier projects than they otherwise would because they gain fully if things go will, but they bear only part of the losses if they do not.
In Jensen's piece on takeovers, he asserts that high debt levels can serve to improve managers incentives. Jensen argues that while managers may always have incentives to engage in non-value-maximizing activities, it is easiest for them to do so when there is "free cash flow."  That is, when there are resources leftover after all profitable investment opportunities for the firm have been exhausted.  Jensen argues that it is difficult for equity holders to induce managers to return "free cash flow" to investors as dividends.  Managers will instead use the cash flow for non-value-maximizing activities.  He argues that it is much easier for debt holders to prevent this because defaulting on debt payments is costlier than breaking an unwritten rule to return "free cash flow" to equity holders. Debt  financing has more than one effect on entrepreneur/managers' behavior.  One thing it does is provide incentives toward risk-taking.  But at the same time entrepreneur/managers are deterred from defaulting when they are actually able to make payments because they bear at least some of its costs. (This was assumed away in Jensen and Meckling, which largely bankruptcy costs which are borne by the individual but not the firm -- such as the loss of his reputation.)  This encourages them to make good on debt payments (return "free cash flow" to the firm's owners)  rather than using firms' revenues in non-value-maximizing ways.  The costs of default "tie managers' hands" in certain situations.

2.    What is the winner's curse?

This refers to the tendency for those with the most optimistic signals about an objects value to win the object when bidding against other buyers, but pay a price which is greater than what the object is truly worth.

Suppose there is a single object up for sale.  Potential buyers do not know the exact true value of the object.  Each of them has some private information ("a signal") concerning the object's value.  On the average, these signals reflect the object's true value.  But across individual buyers, some have information which makes the object look more valuable than it really is, and some have information which makes it look less valuable than it really is.  If they bid according to their signal, the buyer with the most optomistic signal will win the object, but will pay more than its true value.  This phenomenon is known as the winner's curse.

How does this phenomenon affect the market for corporate control?

Bidders, realizing this phenomenon, will discount particularly optimistic information and shade their bid down.  It also provides incentives to acquire better information about target firms.

 Does it tend to make this market work better or worse?

In retrospect I am not sure that this is all that good a question.  Define "working better" as improving the efficiency of who wins the object -- or which management team obtains control.  The winner's curse affects how bidders incorporate private information into their bid.  But because it affects all bidders in the same way, it does not change who will win the bidding -- so it has no efficiency implications.  It only affects how much the winner bid (and pays).

Hard question.  Don't worry too much if your answer is different than this.

3.    You are the owner of a small trucking firm in LA.  You employ several similarly risk averse drivers who value both income and leisure.  The firm's output, and thus profit, is a function of your drivers' effort -- high effort levels mean that your firm is able to serve more customers because your trucks are on the road and moving more of the time.  Effort, however, is costly to drivers.

Assume that individual truck drivers' output is observable (the number of "hauls" he makes, how long they are, etc.) to you.  Output is a function of driver effort and of factors outside of his control, such as traffic.  Assume that you can directly observe only output and not effort.

Your drivers run two types of routes.  One goes back and forth between West LA and Bakersfield.  On the average, there is very little traffic, and not much difference in the traffic from day to day.  The other goes back and forth between West LA and Pomona.  Because it goes through Downtown LA, this route tends to have a lot of traffic, and a lot of variablility in traffic.

You are trying to design optimal compensation contracts for drivers on these different routes.  Compensation consists of a fixed wage plus a per-mile rate.  Assume that the market for truck drivers is perfectly competitive -- truck drivers' utility from working for you is the same as that in their next best opportunity.

The fact that drivers (but not firms) are risk averse and that only one dimension of effort matters to the firm means that the simple model developed in class applies.  Differences in traffic patterns mean that a) there is more variance in factors outside of drivers' control when they are going to and from Pomona than to and from Bakersfield, and b) drivers going to and from Pomona are less able to respond to incentives because they are more prone to be stuck in traffic.  Both of these imply that you should pay the drivers going to and from Bakersfield stronger performance incentives (higher per mile rates) than drivers going to and from Pomona.  In the framework in class, the first refers to differences in V and the second refers to differences in C"(e).  (Don't worry if you did not pick up on both -- one or the other is sufficient.)

Both types of drivers will be no better off than their next best opportunity by assumption.  The Bakersfield-bound drivers are earning higher commissions on the average.  It is therefore likely that they will be paid lower fixed wages.  (This holds as long as drivers are not very risk averse.  One needs to make stronger assumptions about their degree of risk aversion for this to be strictly true from the theory.)  With respect to fixed wages, either "Bakersfield is lower" or "ambiguous" are acceptable answers.

See above.
You hook yourself up to the internet.  On its website, Caltrans has a page which offers up to date traffic reports on all major Southern California highways.  This provides a good, but imperfect, signal of the conditions your drivers face. Using the device means that you can better measure factors outside of your drivers' control.  This lowers V in the framework.  If V is lower, it is optimal to pay stronger performance incentives.  So per mile rates will go up and fixed wages will go down.  The change will probably be greatest for the Pomona drivers -- where there was the most variability to begin with.  (Having the web say "there is no traffic going to Bakersfield" most of the time does not improve your information because it is something you already knew.)

4.. Read the following two Wall Street Journal articles about Ben and Jerry's Homemade, Inc. and answer the following questions.

Company Background. News Story.

a) How are the stated objectives of Ben and Jerry's different than most firms'?

Ben and Jerry's clearly maximizes more than profits or the value of their own firm. Evidence for this is that it values the well-being of Vermont farmers, whether they sell their ice cream to countries (such as France) who have policies to which they object, etc.

b) Considering only the interests of Ben and Jerry's employees and shareholders, are Ben and Jerry's policies likely to have been
value-maximizing before the firm went public in 1984? After the firm went public? Explain.

Ben and Jerry were the only residual claimants. Like the 100% owner in Jensen and Meckling's article, they bore all the costs associated with activites which diverged from profit-maximization. If they took an action (such as using high-price suppliers) which decreased profits, the value of their shares in the company went down. Since B and J fully bore the costs and benefits of these decisions, they may well have been efficient.

c) Discuss Ben and Jerry's management problems since then in light of moral hazard. Be sure to state within your answer: who the
principal(s) and agent(s) are, how their objectives differ, and on which types of decisions conflicts of interests arise.

The new CEO was an agent; the shareholders (including Ben and Jerry) were the principals. CEO maximizes his own utility. Shareholders maximize theirs. Most shareholders probably did not receive utiltiy from anything but the value of their investment. Clearly, the shareholders/founders Ben and Jerry do. The CEO was caught in a bind due to the conflict of interest among the shareholders. He was a agent to more than one principal, and these principals' objectives differed. Conflicts of interest arose when decisions that would increase the price of the firm's shares did not coincide with the founder/shareholders' social agenda -- such as whether to move into sorbets and away from ice cream, or whether to sell products in France.