Retirement is now a public good. Accordingly, the market will not co-ordinate to produce enough of it. Therefore, taxation is justified to bring about an optimal amount of retirement.

Or so the story goes. According to a study by University of Michigan professor of business economics, Joel Slemrod,

“[W]orkaholics misjudge the benefits and costs of working more hours in their careers and working beyond their intended retirement date.”

Mr. Slemrod proposes that the government tax people, in order that they work less. How much less, exactly? Well, the study apparently finds that really smart people, who’ve gone to college might be working on average a year or so past the date they had planned on retiring. Specifically, it says:

“In fact, each additional year of schooling reduces the likelihood of being fully or partly retired by about as much as one-third of a year’s increase in the planned age of retirement.”

So, if you went to college for 6 years and picked up you Master’s Degree along the way, you might on average remain employed 2 years past your expected retirement date.

In order to proceed with this discussion, it’s necessary to accept the tenet that raising income taxes does in fact discourage productivity. I bring this up only because leftists of all shapes and sizes often dispute the actual effects of progressive income taxes. It’s imperative to accept this, because if it were not the case, a tax policy of this nature would be entirely impotent, it would never be proposed, and I wouldn’t take the time to refute it. That being said, I seriously question the desirability of any tax policy whose stated aim is to take a person’s property in order to discourage his future production of property.

Slemrod’s study seems to take the position that retirement early is better than retirement later, and that any amount of work beyond one’s expected retirement is detrimental to society and must be prevented. The press release does not give the relevant information: How is “expected retirement date” determined — is it a midlife decision, a decision made when one is very young, or one that is made largely by people who are very near to the age at which they are first permitted to retire with benefits, or do different people make this assessment at different points in their lives? But even assuming uniformity, the result is that we’re talking about several months to maybe 2 years or so over the course of a lifetime. People err in judgement, and people’s tastes change over time.

But it doesn’t stop there. If we are to believe Slemrod, we must accept that intelligent, hard-working individuals routinely and systematically fail to make accurate subjective decisions about their own livelihood. And we also must assume that government tax policy is markedly better at determining individual utilities, across the board. This of course, is absolute rubbish. People make retirment plans because they believe they will no longer want to work – but sometimes we discover that working really isn’t so bad after all. I submit that people who further their education may have a propensity to enjoy their jobs more than those who don’t, for example:

I wouldn’t be devoting my free-time as a 25 year old to the study of Carl Menger and Eugen Boehm-Bawerk if I didn’t enjoy Economics to begin with. It follows that I’ll likely enjoy a career within the field of my specialization, and I will probably stick around longer in such employ. In economic terms, my opportunity cost of retiring from a job I enjoy is greater than that of retiring from a job I don’t particularly care for.

But you’ve also got to question the practicality of such measures. Do we tax based on degree earned? Based on years of post-secondary education? Based on net income? Based on average weekly hours worked? And I would suggest that we’ve also got to consider what really is the optimal retirement date for a person? The idea that that we can tax people into optimality, that we can tax them into retiring 4 months earlier, with pinpoint, individual accuracy is nothing short of nonsense. And the belief that we can develop a feasible tax policy that will target the workaholics like a smart-bomb, without collateral damage in the form of high-wage earners, people who have a legitimate need for working overtime, etc., is likewise nonsense. Taxing people’s overtime pay may actually be counterproductive to Slemrod’s goal, encouraging them to work fewer hours per week, but more hours per lifetime – making them work even further into their golden years.

h/t to Libertarian Jason who tuned me into this.