Economics 27 Professor Patricia M. Anderson

Economics 27 Professor Patricia M. Anderson

Practice Problem Set 4

(ANSWERS)

Multiple Choice Questions

1. Answer is C. Why? We would expect it to be less difficult for smaller firms to monitor workers, i.e. that monitoring costs would vary positively with firm size. Thus, both A and B are incorrect. It would be difficult to get workers to agree to a deferred compensation scheme if they did not believe the firm would be around to fulfill the contract. Thus, smaller start-ups are less likely to use such a scheme, making C correct.

2. Answer is C. Why? Being paid on any type of piece-rate basis makes it costly to the worker to shirk – they only get paid for actual performance. Such schemes tend not to promote teamwork, as in A, but rather to promote competition. Turnover may be increased, as in B, if workers learn that they are better off in a salaried job, but the firm would not find that advantageous. Finally, income variability is likely increased, not reduced as in D. In any case, it is the worker who would likely find reduced income variability advantageous.

3. Answer is C. Why? This would imply that women workers were actually more productive than men, but were not being paid relatively more. Thus, we cannot claim that there is no discrimination, as in A. At the same time, there is no sign of this being statistical discrimination, as in D. Finally, B would imply that men are being discriminated against.

4. Answer is C. Why? Firms that have no taste for discrimination can hire the victims of discrimination for a lower price, lowering their costs (the opposite of A), and placing themselves at a competitive advantage (the opposite of D). Note that if blacks are the victim of discrimination, as supply increases, relative wages would need to fall in order to induce more discriminatory firms to hire them (the opposite of B).

5. Answer is C. Why? This is basically the definition of statistical discrimination, while B refers to the taste models. Similarly, A is true of the employer taste model only. Statistical discrimination can be a competitive advantage, providing a low-cost screening mechanism.

Short Answer Questions

1. A deferred compensation scheme might be used as a screening device in this case. Since this is a short time period, you could do something very simple like pay a lower hourly wage, and then provide a bonus at the end of the summer for workers remaining with the firm. Suppose for example that 500 hours is a full summer, so that overall they would make $3000. You could pay $4 an hour and provide a bonus of $1000, plus the interest earned each week from not paying $6.

2. A minimum wage of $6 would make it impossible to use a deferred compensation scheme, since you cannot underpay at any point. If you tried it, you would have to hope for lax enforcement!

3. In this case, the personal interview may be a noisy signal. Assuming that you are human, you may find it easier to evaluate the humans’ intentions about how long they work than the demons’ intentions. Thus, you would prefer to either just hire the humans that you have evaluated as likely to stay the summer, or would only be willing to hire the demons at a lower wage. The lower wage would be compensation for the higher risk that your evaluation is incorrect.

4. In this case, X simply represents observable characteristics of the humans and demons. The lawyer ran 2 regressions – one for humans, one for demons – of wages on these characteristics. This is where the a’s (intercepts) and b’s (slopes) came from.

5. The first bracketed term is the amount of the difference in mean wages that is due to differences in the characteristics of the two groups. This is usually referred to as the explained portion, and this amount is generally considered not to be due to discrimination. It may be an underestimate of the explainable amount if we have omitted important X’s from our regressions, or perhaps have badly measured it for one group. Note that if this is an underestimate, the second term will be an overestimate.

6. The second bracketed term is the amount of the difference in mean wages that is due to differences in the treatment of the two groups. This is usually referred to as the unexplained portion, and this amount is generally considered to be due to discrimination. It may be an underestimate of discrimination if we have included X’s that are themselves the result of discrimination. Again, note that if this is an underestimate, the first term will be an overestimate.