Yesterday, I gave talks to two classes at the Naval Postgraduate School. Both are classes taught by my good friend Ryan Sullivan. The talks were both titled “How Economists Helped End the Draft.” This has become an annual event in his class and one I look forward to.

As I always do, I drew an upward-sloping curve to show how, when the government institutes a draft and pays less, it causes people who have supply prices that are below the wage that would have been paid in a volunteer military, but above the wage paid in a military with a draft, not to volunteer. (I pointed out that when I looked at every one of the many bills proposed by someone in Congress in 1980 or 1981–I’ve forgotten which–I found that each of them reduced first-term pay, sometimes drastically.) This means that some of these people will be replaced by people with even higher supply prices, people who would not have volunteered even for a volunteer military.

I gave dramatic examples of people who likely had very high supply prices during the draft era, Exhibit A being Elvis Presley. Then I gave less-dramatic examples: someone who knew at age 18 that he wanted to be a doctor, someone who wanted to start a business or get a job, etc.

In Q&A, one student asked if keeping the volunteer military means that there is more income inequality than otherwise. I said no and that the opposite was the truth. I was thinking of all the relatively low-paid people who would still volunteer but would get regular military compensation (which includes room and board) that was 20% to 50% lower than they would have got if they had been in a volunteer military.

Afterwards, the student came up to explain his point. He was thinking about the Elvis Presley’s of the world and of the less dramatic examples of people who had high supply prices because they had high opportunity costs. Their pay would be cut. I was thinking of the 70% of first termers (my guess) who would volunteer at the draft-era wage but would have earned more under an an all-volunteer wage.

I hadn’t thought of his point and readily admitted it: it means, combined with my point, that the effects on income inequality are ambiguous.

But there’s a reason I didn’t think of his point. I keep thinking that many good-willed people who worry about income inequality are really worried about relatively low-income people. This example reminds me that many people who worry about income inequality really are focusing on income inequality and don’t care whether certain steps taken to reduce it hurt both high-income and low-income people. I’m not saying that this student is in that category. He may not have thought of my point.

This is yet another example of the perversity of focusing on reducing income inequality.