Scott Walker's War on Tenure: An Institutional Economics Analysis

Scott Walker has declared war on tenure, and it has been reported on nationally (in the NYTimes or Inside HigherED and the Washington Times) and blogged about by some economists (Tyler Cowen and Menzie Chinn) But Scott Walker only half-understands tenure, and the half he understands leads him to miss the point of it.
Granting a faculty member indefinite tenure is a mutually-beneficial contracting mechanism that benefits both faculty and colleges. To understand this, we have to understand the different aspects of a faculty member's productivity:
1. Teaching. At most colleges, teaching is the most important service a faculty member provides for her institution. This involves teaching a subject area well, and defending academic standards.
2. Service to the Institution. This might include curriculum/course development, advising, committee work, or generally helping out with administration of academic programs.
3. Research. This mostly includes publications.
4. Service to the Profession. This might include serving as a referee for journals, membership and service in professional organizations, etc.
Aspects (1) and (2) involve "firm-specific human capital investments" and mostly benefit the institution, whereas (3) and (4) involve "general human capital investments," and mostly benefit the individual faculty member (and her ability to find a job elsewhere!). More generally speaking, there are a lot of high-skill jobs that involve similar types of tradeoffs in the allocation of effort for investing in human capital, but this is especially pronounced, I think, in Academia.
This presents an interesting contracting challenge. Institutions would like high levels of effort in all areas, but would especially like high effort in firm-specific areas (teaching and administration). Tenure partially helps achieve this, and this is where the half of tenure Walker doesn't understand comes into play. On the one hand, Walker is correct that tenure has a general tendency to reduce the overall incentives facing a faculty member and thus may reduce total effort (although even this is not fully clear since tenure reduces aggregate uncertainty, and uncertainty can sometimes be a disincentive to work). On the other hand, Walker fails to recognize that tenure also increases the relative incentive for making "firm-specific human capital investments" - investments in teaching and curriculum development. (Side note: tenure is also creates similar substitutions within the allocation of research time away from small "marginal discoveries" and towards bigger projects that may attract acclaim for the institution.)
Hence, university presidents should be wary of abolishing tenure, and not only for altruistic reasons. By increasing uncertainty about continued employment, educational quality suffers. Regular faculty members will invest more time in research (who knows, I may need a job somewhere else?) and retreating on standards (students better like my teaching - and populate my classes - or I won't get to stay).