Brief Thoughts on BIG PAP and Employment Levels

In light of our recent discussion of BIG PAP, I wonder: when we see restraint in direct public sector hiring, do we see similar restraint in the expansion of employment in public administrative proxies? Health sector employment, for example, has continued to expand robustly in recent years, and of course the health sector is a key element of BIG PAP. I’m not sure of the answer here; there are forces pushing in both directions.

A number of observers have (rightly) observed that public sector employment has been shrinking as a share of total employment in the post-recession period, and this has contributed to weak job growth. Calculated Risk observed the following back in March:

A big difference between Mr. Bush’s first term and Mr. Obama’s presidency has been public sector employment. The public sector grew during Mr. Bush’s term (up 900,000 jobs), but the public sector has declined since Obama took office (down 590,000 jobs). These job losses are at the state and local level, but they are still a significant drag on overall employment.

My crude sense is that part of the story during the Bush years was that the polarization of the private sector labor market. While private sector job growth was relatively weak during this period, particularly for less-skilled and mid-skilled workers, impressive productivity gains in knowledge-intensive services, which were tied to investments in organizational capital, led to higher wages in those sectors. This may have contributed to revenue increases that were then translated into public sector hiring in politically popular domains like K-12. At the same time, a large number of less-skilled workers exited the workforce during this period, partly because these workers weren’t really eligible for public sector work and their private sector options weren’t substantially more attractive than going on disability or off-the-books.

Cutting public sector jobs in a weak economic climate is indeed pro-cyclical. The difficulty, however, is that it is very difficult to cut jobs in a healthier economic climate, as the pressure for spending discipline in the public sector weakens when there isn’t a cash crunch. (Indeed, it weakens in the private sector as well.) If we accept that state and local governments (a) have more employees than they need or (b) have the wrong mix of employees or (c) are severely constrained in their ability to redeploy employees across job functions to deliver public services more effectively or some combination of all of these, it is easy to see why some degree of churn can be useful. But again, there are structural forces that make churning of this kind more difficult in the public sector than in the private sector.

This doesn’t change the fact that we usually do see an increase in public sector employment in the aftermath of recessions, so the period we’re in really does represent an exception (rooted, perhaps, in the notion that state and local workforces have grown unsustainably expensive).

Back in 2007, Vincenzo Quadrini and Antonella Trigari released a working paper on public employment and the business cycle.

Starting from the calibrated model we ask how the presence of the public sector affects the business cycle properties of the economy. The main finding shows that the presence of the public sector increases the volatility of total employment by a factor of two or four, depending on the sample period: before or after 1970.

The key feature of the government policy that generates this result is the low cyclicality of public wages and employment, coupled with a premium paid on average to public workers. After a negative productivity shock, public jobs become more attractive for two reasons: lower creation of jobs in the private sector and higher public wage premium (due to the fall in private wages). Therefore, during a recession, public jobs become more attractive and more workers will search in the public sector. This reduces the probability of filling a vacancy in the private sector, which further decreases the creation of new jobs. In this way the presence of the public sector amplifies the macroeconomic impact of aggregate shocks.

A further question we address is whether the volatility of employment (and output) is reduced when the government actively pursues a pro-cyclical employment policy, that is, it hires more workers during expansions and less workers during recessions. Because of the large crowding out effect of public employment on private employment, pro-cyclical employment policies reduce the business cycle volatility. The cyclical properties of public wages are also central to the business cycle properties of the economy. In particular, the macroeconomic volatility is greatly reduced when public wages follow private wages. By adjusting the compensation of public workers to the compensation received in the private sector, fewer workers switch searching from one sector to the other after a negative productivity shock. [Emphasis added]

I think it’s safe to say that not everyone will embrace the model advanced by Quadrini and Trigari. One of its potential implications, however, is that it might be useful for managers in the public sector to have the freedom to adjust compensation levels in response to a revenue shortfall rather than turn immediately to layoffs. Collective bargaining presumably makes this somewhat more difficult.

This is all very provisional, and these are all subjects to which we’ll return.