It's a commonplace observation that the better the mathematician, the worse he or she generally is at balancing a checkbook. It's not a paradox: simple arithmetic is just too boring to hold a bright, abstract thinker's attention. So perhaps this can account for what happened in Chapter 10, "How Math Can Fix the Economy." In the table titled "The S&P Tulip Index, 1975-2007," the author looks at the ratio over a period of years of the year-end closing price of the S&P500 to average household income. Unfortunately, he used REAL (inflation adjusted) household income instead of CURRENT household income. The S&P500 closing prices, of course, are in CURRENT dollars, so the ratio, to make sense, has to be CURRENT dollars to CURRENT dollars.

I immediately noticed something was wrong when I saw in the table that average household income supposedly rose only 40 percent from 1975 to 2007. In fact average household income rose 391 percent in CURRENT dollars! (I used the same Census Bureau spreadsheet as the author, but the correct column — "Current dollars.") With this correction in place, the "S&P Tulip Index" isn't nearly as exuberant at the 1999 stock market peak: a ratio of 4.10 instead of the author's ratio of 16.29.

There is also a conceptual problem with the author's "S&P Tulip Index." It doesn't take into account population growth. If the number of households doubled from 1975 to 2007, you might reasonably expect the S&P500 to double as well, even if average household income stayed constant.

Sadly, the same mistake of REAL versus CURRENT dollars afflicts the next table too, "The Average Home Price Tulip Index, 1975-2007."