The Monkey Cage

I immediately see three alternative explanations - perhaps at the low- to mid- level positions, political appointees replace poorly performing bureaucrats in addition to those with ideologies that clash with that of the administration, thus quality bureaucrats are overrepresented. The performance of bureaucrats would, on average, worsen if fewer political appointees were appointed. The effect could also be confounded a second way: perhaps political appointees have a shorter shelf life and need to spend time adjusting (because they’re recycled at least every four years,) and careerists have been in office longer and have already adjusted. The second effect still supports Lewis’s conclusions, however. The third alternative explanation is that political appointees are placed in positions where, for whatever reason, it is tougher to get results.

Or, political appointees’ average performance is brought down by a few high profile failures, like that of Michael “Brownie” Brown at FEMA.

I’d be curious to see a similar analysis of the performance of “rotator”-type bureaucrats, of the sort common at NIH, NSF, NIJ, and elsewhere. I wonder which type they more closely resemble…

What’s a rotator bureaucrat?

Is the story here one of differences in incentives or differences in skills? The quote above seems to suggest the latter, but this is not so obvious to me.

One might imagine that administrations hold political appointees to different standards than career bureaucrats. After all, if appointments are rewards for past loyalty and service, it seems logical that political appointees are allowed greater leeway to use their office for their own ends than career officials. Career civil servants, on the other hand, seem likely to be held to stricter account for administrative performance.

If this is so, the conclusions are likely to be the same as suggested by Lewis. But the mechanisms seem worth exploring.

It would be interesting to try to generalize an experience v. background effects to other jobs - teaching, business management, etc. I know that’s not what this study is concerned with, but I still think it would be interesting.

I imagine that Chris means those people, like NSF types, that come from outside the federal bureaucracy, do something bureaucratic for a few years, and then go back where they came from.

A few clarifying points might help here. First, there are some cases undoubtedly where appointees improve performance. It can be the case that a very competent appointee can come in and improve performance as long as he or she serves in that position. The deleterious consequences of politicization on performance may not show up until later, when a second or third appointee has assumed office and the ripple effects of the politicization have played themselves out. While some programs are fortunate enough to be administered by very competent appointees, it is much less common that programs are administered by a string of effective appointees.

Second, there are two types of harms from appointee management, one related to the differences in backgrounds of appointed managers vs. careerist managers and one that is harder to observe but arguably more pernicious. I argue that maintaining a high number of appointees can hurt agency performance even if all of the appointees selected are of consistently high quality. The large number of appointee managers creates predictable problems for organizational management that go deeper than the persons filling management positions at any given time. It makes it difficult for agencies to recruit and retain high-quality civil servants; it reduces incentives for careerists to develop expertise; and it leads to increased management turnover—three factors that can hurt performance even under the best of conditions.

Third, while it is possible that all difficult-to-manage programs are run by appointees and all easy-to-manage programs are run by careerists, I am dubious that this explains the results. If this is the case, however, I recognize that any relationship between appointment status (appointee or careerist) and performance could be spurious because the ultimate cause of the low performance would be the inherent difficulty of the program itself, not whether the manager is an appointee or a careerist. In the paper the models attempt to control for the management environment, including the characteristics that would make a program or bureau hard or easy to manage. I also estimated a set of instrumental variables regressions where the appointment authority of the bureau chief is estimated along with the effect of appointment authority on performance. I won’t bore you with the details about the instruments I selected for the models.

When the results are pushed further to see what differences between appointees and careerists explains the performance gap, it is two factors—previous experience working in the agency before you become its chief and length of tenure (less turnover). This suggests that appointees with previous agency experience and appointees committed to long service should perform comparably to careerists.

On the rotator question, my guess would be that programs run by rotators would be managed slightly less well than other programs. This does not necessarily mean that they would not know a good project to give money to when they saw one. It would mean they’d be less likely to have long term goals, strategic plans, performance measures, evaluate performance against these goals, and measures, etc.

Thanks, David, for addressing these issues.

On the rotator question, my own observations, dating back to when I was a rotator at NSF, would be as follows:

(1) A rotator-permanent employee dummy variable would leave a huge proportion of the variance in effectiveness unexplained. Many of the permanent people were competent paper pushers, and a lot of the job involved pushing paper. So on an “efficiency” dimension they would score high. At the same time, many (most?) of them were not, shall we say, keenly attuned to developments in their field. They functioned primarily, at decision time, more as secretaries and numbers counters rather than well informed decision makers or even motivated to be. Moreover, on the motivation dimension they seemed to be an uncommon number of burn-outs, as well as a few who simply seemed incompetent.

(2) Among the rotators, some were helpless with respect to the routine aspects of the job, and would have to score very low on an efficiency dimension. At the same time, they knew much more than the permanent people about their discipline or field.

(3) From (1) and (2), I’d say in general that there was no sharp distinction between the two, given the variance within groups, and also that which group was doing a better job would have to depend on what you mean by “doing a better job.” For informed decision making, activism, and energy, it would be the rotators hands-down. For normal bureaucratic activity, it would be the permanent people.

These, of course, are cartoons, and they’re cartoons composed by a rotator to boot, and they’re based on the observations by one person for two or three years. So take ‘em for what they’re worth.

Lee makes some interesting points. I wonder, for the sake of argument, how good the career professionals would be if they knew that they had a shot at the top positions where they themselves would have a higher salary and influence over where the money went. If the rotators get to make all the decisions that does not create incentives for good people to come or stay or invest in expertise in the field. If, however, career professionals had to prove competence to get into the top jobs with higher salaries and more influence over outputs we might get better careerists.

In general, though, this distinction between area expertise and making the trains run on time is an interesting one. The performance measures I evaluated in the research at issue here focuses more on making the trains run on time.

In other research I have looked at whether agency employees agree with statements like “my agency’s senior leaders generate high levels of motivation and commitment in the workforce” and “I have a high level of respect for my agency’s senior leaders”. The patterns are the same here with regard to appointees and careerists. I don’t know but wonder how the rotators would do on these dimensions. Worth a look.