Larry Summers and Finance

By James Kwak

I think some people didn’t understand the point I was making about the question of whether the government made money on TARP in my earlier post. Summers said, “The government got back substantially more money than it invested.” This is true, at least if you give him some slack on the word “substantially.” The money repaid, including interest on preferred stock and sales of common stock, exceeded the money invested.

My point begins with the observation that, as of late last year, the government had earned an annual return of less than 0.5%. My point itself is that it is silly to evaluate an investment by whether or not it has a positive return in nominal terms. You can only meaningfully evaluate an investment by comparing it to some benchmark. Saying that a nominal return of 0.5% is greater than 0% is meaningless, since the 0% benchmark is meaningless. Most obviously, it doesn’t account for inflation; since inflation has been about 1–2%, the government lost money in real terms.

But even saying whether an investment made or lost money in real terms isn’t very meaningful. You should at least compare it to the risk-free rate of return—the rate you could have gotten by investing in Treasuries. As I said in the post, if the government had instead invested in 5-year Treasury notes, it would have gotten an annualized return of 2.4%.

Actually, you should evaluate your investment against other investments of comparable riskiness. Preferred stock in big banks in October 2008  was pretty risky. If you compare TARP (which was concentrated in one very volatile industry) to investing in the broad stock market, the latter returned more than 18% annually.

And, really, you should evaluate investments ex ante, not ex post, by comparing the amount you paid to the at-that-moment value of what you were buying. And by that measure, the government paid high for preferred stock and sold insurance guarantees low.

It’s fine to argue that TARP was good financial stability policy, and that goodness more than compensates for the fact that it was a bad investment. I may not agree, but that’s plausible. But it’s highly misleading to talk about the government’s positive nominal return as a good thing without evaluating it the way you evaluate any other investment.