Greg Mankiw gets in touch with his Old Keynesian Roots

He decides to consider current income the only determinant of savings.

He writes of Sonia Sottomayer “My grandmother would have been shocked and appalled to see someone who makes so much save so little.”

True and one can understand his grandmother’s total lack of appreciation for the work of Milton Friedman. Professor Mankiw, however, knows better, as is explained in the blog post with the best title ever and by Brad deLong.

Mankiw mentions his theory and I discuss after the jump.

Prof Mankiw

I once wrote a short paper called The Savers-Spenders Theory of Fiscal Policy based on the premise that there are two types of people: Some save and intertemporally optimize their consumption plans, while others live paycheck to paycheck, spending their entire income as soon as it’s received.[…]

Apparently, the new Supreme Court nominee Sonia Sotomayor is an example of the latter. The Washington Post reports that the 54-year-old Sotomayer has a $179,500 yearly salary but

On her financial disclosure report for 2007, she said her only financial holdings were a Citibank checking and savings account, worth $50,000 to $115,000 combined. During the previous four years, the money in the accounts at some points was listed as low as $30,000.

So evidently 50,000= 0. I mean most people not as expert on economics as prof. Mankiw would not consider someone with 50,000 in the bank to beliving from paycheck to paycheck. The fact that her wealth was much lower at some point in the last 4 years means that her consumption does not track her income. The evidence for significan consumption by “spenders” is mostly that income and consumption are more highly correlated than they would be if all people were intertemporal maximizers without budget constraints (excess sensitivity). This is observed in aggregate data in spite of the case of judge Sottomayer whose wealth is not only positive (wouldn’t show up) but variable (uh oh for the we’re all spenders hypothesis of Prof Mankiw’s colleague prof strawman).

In contrast, as explained by Brad DeLong and Nate Silver in the linked posts, it is relatively easy to reconcile the Sottomayer data with the intertemporal utility maximizing hypothesis.

The idea that one can expect a simple relationship between financial wealth and income would not be shockingly inconsistent with modern economic theory if stated by Prof Mankiw’s grandmother, but it is authentically shocking coming from prof. Mankiw.

Now I actually think quite highly of Prof. Mankiw’s work on consumption. I can remember the day I first saw it presented — October 19, 1987. OK click the link and see that I can look up the date not because the brilliance of the talk seered the date into my memory but because the stock market was crashing.

The first I heard of the crash was that Mankiw was describing instruments used to forecast future aggregate income and mentioned the advantage that stock market indices are available every day (then in an aside sometimes every minute) but that they are not good predictors of future income (then in an aside “I guess we should be glad about that today”). Only when the seminar was over did I find out what he was talking about, so I was reassured in advance that the expected income decline conditional on the crash was, according to him, modest. And so it turned out to be, supporting Mankiw’s work on excess volatility of stock markets (both papers with co-author John Campbell).

The Sottomayer episode shows something almost serious I think. Economists don’t take economic theory seriously. We switch from optimizing models to our grandmother’s model to get a blog post out. We certainly switch back and forth from the model in which everyone is rational, markets are complete and competition is perfect depending on the arugment we want to make. Mankiw takes economics very seriously (compared to say Waldmann) yet even he is willing to embrace pre Friedman thinking at the slightest provocation.