Ok folks, this one is a little nerdy, but there are some important issues here. In the last year or so the Bureau of Economic Analysis' measure of national income growth has exceeded its measure of GDP growth. This has led some analysts to jump on the income numbers and say that the economy is actually growing somewhat more rapidly than the GDP measures that were routinely follow, which are measured on the output side.

Just to backtrack for a moment, in principle we should be able to measure GDP on either the income side or the output side and come up with the same number. The income side means that we add up all the incomes generated in production: wages, interest, profits, rents, etc. On the output side we would be measuring the value of all the goods and services produced over the course of a quarter or year. If we had perfect measurements of the economy then the two should end up being the same.

As a practical matter they never are the same. Most often the output side is somewhat higher (@ 0.5 percent) than the income side. The conventional wisdom is that income is more poorly measured because people have a reason to lie about their income (i.e. taxes). So economists have traditionally looked at the output side as the better measure of growth.

While the output side has historically shown a somewhat higher measure of GDP, there have been some exceptions to this, most notably in the late 1990s and the peak of the last business cycle. These were both periods in which there were large amounts of capital gains generated by bubbles in the stock market in the 1990s and housing market in the last decade.

While capital gains are not supposed to count as income for GDP purposes, it is almost certain that some do. (Short-term capital gains are taxed the same way as normal income, so there is no reason for people to distinguish between capital gains and normal income on tax forms.) Anyhow, if we hypothesize that some amount of capital gains end up being counted as normal income then any time we have more capital gains in the economy we would expect that the income side measure of GDP would rise relative to the output side measure.

David Rosnick and I tested this view a couple years back, measuring the extent to which the statistical discrepancy (the gap between the output measurement and income measurement of GDP) fluctuated in response to recent trends in capital gains. And, we got a very solid fit. So we were not surprised when the recent run-up in the stock market, coupled with a recovery in the housing market, caused income growth to again exceed output growth. But hey, there can be lots of good jobs struggling over this one.

Anyhow, this matters both for the answer to the question as to how fast the economy is growing, but also how we view savings. Savings are defined as the difference between income and consumption. If capital gains are showing up as income, then we are overstating income and therefore overstating savings. We think this was very likely the case at the peaks of both the stock and housing bubbles, which means that the consumption driven by these bubbles was even larger than the conventional data show.


Note: Typos corrected -- thanks Joe and Medgeek.