« Comment on Tenure-BECKER | Main | Tenure--Posner's Reply to Comments »



Feed You can follow this conversation by subscribing to the comment feed for this post.


Posner writes:

"The steep decline of unionization in the private sector is a convincing "Darwinian" refutation of the argument one used to hear that unions actually promote efficiency."

How tidy. This account, however, ignores, for instance, the role that union leadership has had in organizing and improving the productivity of various industries. Consider the impact that John L. Lewis and the United Mine Workers had on their industry.

I would argue that it is just as fallacious to believe in the inherent efficiency of an ìat-will workforceî over union workforce as it is to believe the opposite. Too many other factors, beyond union membership, determine whether an enterprise is efficient.

Now, returning to Posnerís larger point, that there is little compelling justification for tenure at just about any level of the workforce. This is persuasive enough when talking about the professoriateóa workforce with highly specialized skills and (presumably) the ability to demand automatically a living wageóbut what about, say, line workers at a factory that assembles vacuums? Mightnít we have some interest (out of a value of fairness, perhaps) to assure that these folks can band together to eke out a slightly better grade of pay, and therefore might these workers benefit from the protection of ìtenureî conferred on them via union membership. Without ìtenure,î of course, organizing efforts are liable to be stumped.

Chetly Zarko

Reducing tenure in K-12 I think is particularly important, as is some kind of pay-scale gradation amongst teachers who perform measurably better. This is not to say that there shouldn't be some kind of tenure for teachers that a hit a certain plateau (arbitrarily, let's say, 10 solid years of measured performance), itself as an incentive.

There is no question in my mind that college tenure systems are over-done as well, but the damage done there is far more minimal since there is actually (some, not enough) competition among many colleges and some valve for the failures of tenure. With American public schools in K-12 though, if you're born into certain social classes and geography, you're pretty much locked into whatever system you're closest too. Since eliminating public education is a political impossibility (NEA, etc.), and probably not desirable anyway since the transition costs would be so massive and unpredictable, the key is in introducing elements of competition into the public system. That takes creativity from policy makers, and courage from politicians, of which there is unsurprisingly little of on either front.


I am surprised to see no treatment of the argument that tenure encourages individuals to invest in specialized skills. In what is labelled the "varieties of capitalism" literature, this is often presented as a trade-off: high levels of job protection encourage greater investment in skills and hence greater productivity especially in technical professions, yet they undermine the flexibility of the workforce


I fail to see why tenure is required in order to encourage a greater investment in skills. It seems that there would be a demand for greater productivity and compensation for greater investment would be reflected in that demand.

And, ceteris paribus, at-will employment is demonstrably more efficient than a unionized workforce.

Another cost of tenure in academics could be the decline in teaching quality. I know that this is not true for many professors, but it is obvious that there is no incentive in place for some.


"Until morale and productivity returns, the floggings will continue." Or in other words; Higher, faster, further, more, more, more with less, less, less. And Oh! by the way, don't expect any increased compensation for the efforts. Tenure, hasn't that always been one of the perks and plums of an Academic career? If used properly it can function as a device for controlling salary increases.


In some fields, such as mathematics, there is generally a significant falling off in academic output at a young age,...Not having seen the studies, it is impossible to comment on this directly.Suppose, however, that universities had a policy of hiring people who had recently been struck by lightening and a study was done to compare the rate of lightning strikes on university faculty before they were hired and after they were hired. One would find a dramatically lower rate of lightning strikes on the faculty after they were hired relative to before they were hired but this would not indicate that being hired by a university protects against lightening strikes.Essentially, people who make major mathematical discoveries at a young age are likely to pursue careers in mathematics but people who don't often go into other careers and give up on mathematics so they don't make the discoveries later in life that they would have made had they been hired as mathematicians.Only in a world where universities hired people to be mathematicians without any regard for their output early in their careers would it be possible to do a direct statistical analysis of productivity with age.


Three thoughts are below.

1. Availability bias: Sure, employees are employed at-will. However, how an organization treats an employee during his\her final days at an organization, even a true "loser" employee, will have a disproportionately large influence on how the (former) employee remembers the organization and his experience. Treating "losers" rudely at the end of the contest happens all the time in the workplace. Companies go under because of it. It is somewhat like poor sportsmanship and ungracious winners on tv.



"additional motivation may be provided by a tournament-style promotion system."

Real wages in IL may be down because the tournament is broken in chicago and IL - not sure. People with money, the right friends, the right age, and the common lifestyle (married, kids) get jobs and unconventional or different people do not. Given this realization that the game is rigged, or even the realization that the probability of game-rigging has increased: people will not bother to enter the tournament. To do so would be stupid and a waste of time. People will leave. You will have an adverse selection problem: "winners" will be dolts.


"The greater cost of tenure is simply in forcing retention of inferior employees."

Yeah, and at some point, is this so bad in all instances? If you force out some 40-year-old inferior person, where does he go? Does he live off social services provided by the superior person's wages that are donated to a church or parish? At some point, there may be an idea of "insurance" behind tenure and protecting weaker employees. As long as this insurance is not excessive, I do not see how it is so bad.

Much more dangerous is the dolt who wins a rigged or incomplete tournament.


One more: It may be hard to take an absolute position on tenure at all points in time. It seems that trends toward more tenure may be smart during certain time periods, and less smart during other periods. The answer may be "it depends".

One might try to argue that a minimalist tenure system, vs. no tenure system, can lead to more stability. Given stability, people and society may be more likely to invest relatively larger amounts.

The variety within the U.S. by state is interesting and valuable. Some states are more unionized (protect jobs - somewhat like tenure?). Other states are more right-to-work (no unions). Would we want to make it all one way or the other? Probably not. Some people thrive in one type of state, and others thrive in another. This may change during different parts of a person's life. Different types of states may provide diversification of outcomes and lower the riskiness of a society. We may get better risk-adjusted returns given a little bit of everything in the U.S.



Einstein didn't work at Princeton University. He worked at the Institute for Advanced Study, located in Princeton, NJ.


Einstein didn't work at Princeton University.Oops. I was wondering whether I should fact check that before I used it as an example.According to the Seeley G. Mudd Library at Princetonhttp://www.princeton.edu/mudd/news/faq/topics/einstein.shtml
While an important member of the larger intellectual community of Princeton, Einstein was not a member of the Princeton University faculty although he did have an office on campus. Then again, according to:http://nobelprize.org/physics/laureates/1921/einstein-bio.html He became a German citizen in 1914 and remained in Berlin until 1933 when he renounced his citizenship for political reasons and emigrated to America to take the position of Professor of Theoretical Physics at Princeton* With the foot note that *Albert Einstein was formally associated with the Institute for Advanced Study located in Princeton, New Jersey. Well, at least I didn't use Linus Pauling and Cal. Tech. as an example. According to:
http://lpi.oregonstate.edu/lpbio/lpbio2.htmlIn 1964 Linus Pauling left his tenured professorship at Caltech because of pressure from administrators and conservative trustees who disapproved of his prominent, persistent antinuclear and international peace-promoting activities. Pauling had been at the Institute for 42 years - first as a graduate student, then as a faculty member. (In 1937 he was appointed Chairman of its Division of Chemistry and Chemical Engineering and Director of the Gates and Crellin Laboratories' positions that he had abdicated in 1958 under administrative pressure.)

Greg Hammond

N.E. Hatfield writes:
If used properly it [tenure] can function as a device for controlling salary increases.
Mmm, perhaps, but I am dubious, because I believe such an exchange leads to lower productivity and hence greater costs to the organization. Having watched the university employment game from an early age, I am convinced that tenure is/was first and foremost a security device for protection against "worry," chiefly the worry from arbitrary discharge for any reason, not merely political, but generally not economic. [Tenure did afford some measure of protection against purely economic layoffs, but only if there were untenured folks behind you.] Otherwise, it was truly a "resting" point. That is not to say that many tenured professors do not work hard after achieving tenure. Many do, but they continue to provide value and receive it as well for their labors, either in increased recognition, continued promotions, sabbaticals, paid speaking, etc. Those for whom tenure is a substitue for salary increases are likely to repay that "ceiling" with reduced productivity. I think we see that all around, notwithstanding the occasional--perhaps apocryphal in the case of the UMW and John L. Lewis--exception. Just my $.02.


Could Josh or anyone else here please cite and comment on the research that proves "ceteris paribus, at-will employment is demonstrably more efficient than a unionized workforce?"

I'm curious, since if this were well established by a number of studies, why would Posner have to cite the decline of unions in recent years as "a 'Darwinian' refutation of the argument that labor improves efficiency?" I mean, couldn't Posner have just cited the research itself?

Thanks in advance for helping out this non-economist.


From a theoretical perspective, unions raise wages above market clearing prices which increases the unemployment rate - there is an obvious dead weight loss. Think of it in terms of opportunity cost: if an elevator mechanic can add 1 dollar of value per hour but a union contract requires him to be paid 2 dollars an hour, he will not be hired and will be forced to accept a job where his value added is presumably less. In short, unions allocate resources inefficently.


In my view, academic tenure remains a necessary safeguard against suppressing and punishing unpopular ideas. I'm not convinced that Judge Posner's proposal for replacing tenure with contractual protections - i.e., make politics an impermissible ground for termination - would work in practice.

Assuming that contractual provisions could adequately cover all of the independent thought protected by tenure, there remains the problem of pretextual discharges and demotions. I suspect that numerous firings in academia, as in other employment contexts, are cloak-and-dagger affairs. Moreover, there is a heightened risk of politically-motivated pretextual retaliation in fields where performance measures are more susceptible to subjective factors - e.g., fields in which journals accept and reject papers based not only on analytical quality but also on viewpoint (expressed in the paper, or otherwise known to be held by the author). Since many of the most subjective academic fields are also among the most politically charged - sociology, law, political science - politically-motivated pretextual retaliation would be an even bigger problem than it already is (e.g., to non-tenured faculty) if the tenure system were abolished.

Contractual employment rights are only valuable to the extent that they can be enforced ? including against violations based on pretext. Many employment rights, for example against race-based animus, are valuable and enforceable in large part because they enjoy a highly supportive legal environment (extensive statutory protection, heightened standards of judicial scrutiny, sympathetic courts). Would professors who are unpopular for challenging politically correct taboos receive a similar level of legal protection?

As shown by the recent excoriation of Harvard president Larry Summers over his comments on women in science, these are very real concerns. If unpopularity can bring a Harvard president to his knees, imagine what it could do to less powerful professors.


Thanks, Josh. I don't mean to loose sight of the rest of the conversation by pressing what must seem to most people here to be a settled point; nonetheless, I still canít help but wonder if there are any studies that demonstrate that union workforces are inherently less efficient than non-union ones.

Certainly we have evidence of an anecdotal sort (e.g. unions must be less efficient because the market has all but done away them, with unions representing a mere 8% of the workforce); certainly we have theoretical explanations, too. Maybe together these are compelling and do the job. But I canít help but think not. Thereís simply too much inefficiency in every workforce, no matter what kind of productivity gains they say are driving the economy (or it least it seems to be so, to me, from my little cubicle).

So how about it? Can you cite so research for me to look into on this?


There is a lot to say in response, but I will limit myself to two comments.

First, I find it distressing that Judge Posner does not realize the destructive potential of politicizing the entire civil service. Certain elected officials, who favor friendship and partisanship over competence, will fill the payroll with hacks and not only hurt productivity but create cronyism and corruption. Perhaps the civil service has been "clean" for too long that Posner has forgotten what it was like before. And, there is no reason why one's political affiliation should matter for non-supervisory civil service jobs (unless one favors cronyism or corruption). Given the scandals in today's headlines, clean government should be a command, not a suggestion.

Second, Posner evaluates the "costs" of tenure purely from the standpoint of management. What about thinking of it from the standpoint of the worker, given that most Americans are workers and only a few are managers? We should ask whether the average worker would be willing to trade the "cost" of some form of tenure (that is, slightly higher prices) for the job security it provides. I would bet that most Americans would gladly take the trade-off, if they had the option.

My suggestion would be to make it easier to fire workers for cause (e.g., incompetence, negligence, etc.) but more difficult to fire them for reasons unrelated to job performance. Workers with demonstrated competence issues should not be able to file wrongful discharge suits. But on the other hand, no good, established worker should be fired to make room for the boss's nephew.


Related observation: hasn't tenure, in whole or in part, driven up the cost of a higher education (to a degree far greater than the rate of inflation, or any other comparable indicator)?


> But the situation is no different in "knowledge" firms such as software

Computer scientists in corporate research labs once had de facto tenure. They could take big research risks (Bell Labs, Xerox PARC). Today, most corporate research is very short-sighted, incremental work because researchers have a bit less job security. If you measure a scientist's performance every year, then he has to do short-term research to show new results.

Marc Shepherd

Some of Judge Posner's proposed productivity measures can have invidious outcomes. Take the proposal for measuring FBI agents (number of arrests leading to convictions, weighed by the severity of the sentence and/or property recovered).

This proposal would be a serious disincentive to agents who investigate crimes that are serious, but complex. Agents might gravitate towards a high volume of easy-to-prove convictions.

Beyond that, isn't the first duty of the FBI to ensure that justice is done? Paying agents by the length of the sentences imposed would encourage the government to always argue for the longest conceivable sentence, regardless of whether it is just.


"With their productivity continuously measurable, there is no need for job protection."

Its almost as if people haven't read history... Foremen standing behind workers and docking their pay if they talked to each other... People getting fired for no reason other than to instill fear in other workers... Crushing drudgery and constant suspicious monitoring... the late 1800s weren't pleasant, massive numbers of people found it to be horrible working under constant performance monitoring.

And now we get one of the most privileged people in the whole workforce endorsing neo-Taylorism.

I don't mean to be disrespectful but... Why should we care what a federal judge with one of the world's cushiest retirement plans and a lifetime of tenured positions says about job security. Has Posner EVER been laid off? Is he renouncing his tenure or his retirement plan? (He could, job security has helped him to get rich)

"The 80-year-old mathematician may be working hard, but he may be incapable of achieving the output of the 25-year-old mathematician who would take his place were it not for tenure."

That is exactly why people bargain, (collectively before the Right made union a bad word) for job security, to stop employers who paid them less than their marginal output in their youth from firing them in thanks.

To quote Death of a Salesman (thanks to a reminder from my employment law professor): you can't eat the fruit and throw away the peel

When people get fired for no reason, especially when they are old or sick or when childcare demands reduce their output, they don't just die. Rather they start to draw unemployment, welfare, medicaid. Given the overhead of these programs, it may cost as much or more to support people off the payroll as on. The difference? In one case it is paid out of profits, the other, out of taxes.

So in reality, drop tenure and you raise profits, but you also increase the burden on taxes. Who pays taxes these days? Corporations pay less and less, the middle class pays more. Who gets profits? Is it really so hard to see the motivation from the result? Or else you stick it into deficit spending, which is just a way of postponing payment of boomer retirement onto my generation.

Incidentally, the same effect occurs when Walmart pays people below a maintenance wage and gives no health care. People still get sick, medicaid and public emergency room costs go up. They work two jobs but still have to take welfare to pay for day care.

Dropping tenure is just an externalization of costs. Instead of the university paying for the old age of its former stars, they are fired and you and I foot the bill. Who cares you say, so long as it is efficient? Well, it is also significantly less human. Read Death of a Salesman again. Profit is not the only function of employment.

p.s. It is shifty to spend a lifetime joined in the project of disparaging unions in the public sphere and then cite the decline of unions as "darwinian evidence" that you were right. All that proves is that people like Posner and Reagan can influence public opinion.


Posner said:
"Quality of research is readily measurable by grants, prizes, and above all by citations to the professor's scholarly publications, weighted by the quality of the journal in which the citations appear."

So is the fact that I cited you in my law review note going to increase your pay or decrease it? :) Does it matter that I was disagreeing with you? I don't know if I'll get published or how my journal rates on the scale though. If you do get a bonus, I hope it is enough to buy a beer or a glass of wine on me.


Commentor said:
"Think of it in terms of opportunity cost: if an elevator mechanic can add 1 dollar of value per hour but a union contract requires him to be paid 2 dollars an hour, he will not be hired and will be forced to accept a job where his value added is presumably less. In short, unions allocate resources inefficently."

I don't want to think of it in terms of opportunity costs. If all the elevator mechanics are unionized, then the value of the elevator repair is set at the value of the union contract. The elevator in Macy's has to work! They have to pay $2. So the elevator mechanic gets an extra dollar an hour, then maybe he saves up so he can go to Macy's and buy a shirt for his daughter.

Profits at Macy's go either to management and shareholders, or to employees. The argument is about what share goes where. If you define "value" to be the minimum Macy's can get away with in the absense of collective bargaining, then you are begging the question.
If efficiency equals maximum profit in the absense of unions then you've rigged the game.

The proper economic question is, is there a deficiency in the bargaining process that makes it necessary for employees to join together in order to be on an equal footing with Macy's. Can Macy's fire people when they ask for a cost of living increase and scare everyone so much that they eventually drive wages down to $5.15 an hour? Yes. Can they do that when there is a union? Probably not. Should everyone but the overlords get paid $5.15 an hour? You decide.

You didn't tell us where you came up with $1 for the value. Was the process that set that wage fair? Were both parties represented by counsel or just one. Macy's can hire or fire one person and barely feel it, but the mechanic is desperate to feed her kids. Is that equal footing?


"Related observation: hasn't tenure, in whole or in part, driven up the cost of a higher education (to a degree far greater than the rate of inflation, or any other comparable indicator)?"

Nope. That's due to dramatic increase in salaries that university administrators' are being paid.


I am an employee of the Federal government. My salary is substantially below market rate for my profession, but I have tenure which confers job security. I have decided to trade the higher salary for security. (Individuals with my skill set receive far greater salaries than I do in the private sector, but I have seen them lose their jobs when the company no longer needs their services.) If I didn't have the security, I'd move to the private sector immediately. So, doesn't the taxpayer gain?


Michael above is correct. Every government employee I know of makes less than their private sector counterparts, but in turn enjoy greater security, a less stressful workplace, and (often) a sense that they contribute to the greater good rather than to stockbrokers and clients.

To me the great danger is a growing Wal-mart-ization/cash-dollar-value of the academy. We support tenure because often what is the most valuable in the long-term (economically and otherwise) often has no cash value in the short-term. Alan Turing developed the theoretical basis for the computer in the 30s - but it tooks decades before it could even translate into hardware (everyone at the time thought it a poor investment, as it would be too slow for calculation).

Should he have worked on it? Turing was a smart man, and no doubt could have used his math as an economics professor (as posner suggests) to do work that would have been very useful to then-current-day corporations. But we may be decades behind in computers right now if he hadn't. And if he wasn't in part protected from market forces, he may have has no incentive to not.

Demand that doesn't pay out, say for abstract computer models or literary theory, has a tough time generating supply without a push. Why not give it a push?

The comments to this entry are closed.

Become a Fan

May 2014

Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31