Give Thanks for those (Soon to Disappear?) Benefits

A striking thing about the jobs most commonly chosen by newly minted graduates of elite law schools, is that they come with valuable fringe benefits. Base salaries are widely imitated and therefore fairly uniform, but firms compete a bit more when it comes to benefits. Health plans, inclusion of same-sex partners, fitness centers, part-time work options, and other things are on the minds of employers and employees, but one particularly expensive and nonuniform benefit is parental leave. At Thanksgiving dinner this year I sat next to a new associate at a Boston firm who seemed thrilled that her firm offered maternity and paternity leave, for biological or adopted children, of three paid (plus benefits) months. The law requires only some leave without pay, so we can think of this as a benefit worth more than $50,000 per child. I think my friend's reaction was partly social/political, as she prefers a world in which parenting is encouraged, or at least not penalized, in this manner, and I think it is also economically rational. She does not think of this costly benefit as coming out of her paycheck, although I think it unlikely that she cashes in on this benefit with this employer, because she correctly sees her own compensation as moving in step with other firms' base salaries. I think her reaction would be the same even if the associate in the office next door had five children in the next five years, and even if that associate's partner also received paid leave from his or her employer. Of course, the birthrate in this group is likely to be low. The benefits come out of the partners' returns, though they may of course encourage new associates to join this firm.

Meanwhile, other guests at the meal represented the other trend in the new generation of professionals: freelancers with no benefits at all. We now see a large group of affluent citizens who buy their own health care, decide their own level of vacationing, and presumably bear the full costs of parenting decisions, unless they are more likely to be attached to a partner who works for one of the high-benefit employers. This bifurcation of high-end employment, if it is that, can be traced to tax law in part, but also to a sort of signalling theory. Firms signal and compete with the benefits they offer, even as some of these benefits are tax-favored. The non-favored benefits need not be offered to all employees, and they are generally not so offered. The mailroom staff is unlikely to enjoy much in the way of paid leave. There is, in short, some bifurcation within firms as well as across industries and job classifications.

In an earlier post (and eventually in conventionally published work) I note that political change is less likely where important interest groups already enjoy a benefit; there would be more of a political move for subsidized, universal parental leave if fewer privileged citizens already enjoyed such leave, just as there would be more of a cry for health care "reform" and perhaps better public education if there were not lavish private alternatives. I do not mean to say that uniformity would necessarily be good, just that the interest group politics are important to bear in mind. But after thinking about that parental leave and doing some back-of-the-envelope calculations, I will hazard the guess that the current multiple equilibria are unstable. My guess is that employers will compete for a while with these benefits, but then cut back on them substantially in favor of some sort of cafeteria plan or a dramatic salary increase, accompanying a cutback in benefits. Yes, I know, we can think of paid parental leave as a sort of insurance policy, particularly for new professionals at law firms and investment banks who are unlikely to know ex ante about the role children will play in their lives over the next five years. But there is so much turnover in this group that some adverse selection is likely. Second, the levels of compensation are such that insurance hardly seems necessary. Finally, the contrast with lower-paid employees in the same firm, not to mention with professionals to whom work can be outsourced, will begin to seem politically incorrect or simply too expensive.

Comments

[Cory Hojka]

One thing I'd question when comparing multiple benefits packages is should it be based on what is offered or utilized? For example, I'm under the impression that law firms can offer all sorts of benefits with the full knowledge that many will be quite underutilized. The reason is that the fear of underperforming relative to one's co-workers will drive people to avoid using them. Thus, a firm or some similar industry can likely offer the moon without much fear that such benefits will actually be used.

If we separate benefits into those that are utilized versus those that are not, I wouldn't be surprised that with the later that we would still see considerable diversity. Real benefits could likely conform to an industry standard while the disinclination of workers to utilize other benefits might allow employers to offer all sorts of wacky packages that risk little chance of actually being used.

[Frank]

Only for the sake of nitpicking:
You might reconsider citing 'signaling theory', which involves asymmetric information. I imagine the contracts are quite transparent. What do these workers want to know that is 'signaled' by these benefits, but not indicated explicitly? Corporate culture? Future utilization of the benefits is not a candidate, of course.

Also, isn't the dependence of "political change" on the support of "important interest groups" a truism? As far as the three closing points go, I wonder if
(i) you've confused adverse selection with attrition; churn does not suggest a bad haul unless you're taking into account relationship-specific investments; the latter cannot be too large a factor if the fields you're interested in can be outsourced to freelancers
(ii) the necessity of those benefits is obvious from the demand for them; and
(iii) the lo-hi contrast has always been there, and if there has been any political backlash involving the use or abuse of freelancers at all.

[saul levmore]

Thank you for the comments. The signaling I have in mind is that the firm with high parental benefits "signals" that it is and will be an employer friendly to parents. It is more likely than most to accommodate with short term leaves, or to be flexible when a parent needs to run out of the office when there is an emergency. The signal is credible because the expensive parental leave policy shows a willingness to invest in such employees and shows that it is likely that over time more parents will be at the firm, or perhaps more non-parents who are sympathetic to large expenditures on other parents, and these members seem much more likely to be supportive of co-workers with child-related emergencies. If we insist on seeing an asymmetry, it is the knowledge of how those already at the firm are likely to react to future workplace issues.

As for the high-low benefits equilibrium and the question of whether it is sustainable, I did not mean a national political backlash, but rather a market reaction. If there is adverse selection (people likely to have many children go to the firm with high benefits, and they guess correctly) or attrition (people who see they are unlikely to have children defect because they do not wish to "pay" for the benefit enjoyed by co-workers), eventually the benefit will be very expensive on a per-worker basis, or will simply fail to provide insurance because it will be priced at its delivery cost.