Philadelphia Reflections

BEA Monitors the Economy

Global Interdependence Center

The Global Interdependence Center meets at the Philadelphia Federal Reserve, organizing frequent seminars of outstanding quality about finance. This week, the speaker was Andrew Hodge, head of Profits Research, U.S. Department of Commerce, Bureau of Economic Analysis. Someone there once had the brilliant idea that aggregate national income was almost identical to Gross Domestic Product, so national income could be easily derived from tax information at the I.R.S. Originally probably seen as a way of verifying GDP statistics derived in other ways, aggregated income and profits look in some ways to be superior to the data coming from Wall Street earnings reports. As a leading indicator, it appears to be outstandingly effective in predicting an impending upswing in the business cycle, just about at the time everyone is getting discouraged about downswings. It's not so good at predicting market peaks.

The BEA profits report

Seems to be superior to Wall Street earnings reports in four ways. 1) Wall Street is not particularly useful in distinguishing domestic from foreign activity within multinational firms. 2) Wall Street reports generally attempt to avoid seasonality noise by comparing this month with this-month-last- year. If the market direction has changed during the past year, downswings may cancel upswings and such comparisons can be misleading. 3) at market inflection points, volatility gets exaggerated by firms going out of business at the bottom or businesses formed or expanded at the top. 4) Wall Street is only 40% of the economy. The other 60% has private ownership, particularly in S-corporations.

Out of studying the differences between the two types of statistics about the economy, it emerges that the tax-derived BEA statistics are quite good leading indicators, particularly when the economy is in a trough. They are sort of leading indicators of coming market peaks as well, but they lead by longer intervals. A lead of as long as a year isn't very useful as an indicator.

As the jargon goes, that's the take-home message. BEA data is pretty good at predicting market bottoms. But some interesting sidelights appear, as well.

Our economy is becoming less volatile, with milder cycles and less frequent ones. But national income is just as volatile as ever, particularly in stock prices. This would appear to be due to the steadily increasing proportion which is in the financial sector (or decreasing proportion in the manufacturing sector). The financial sector is characterized, worldwide and for a long time in the past, as having "sticky" wages and costs. With the cost side comparatively inert, profits become much more volatile. In the final analysis, the stock market becomes more volatile than the underlying economy.

A final conclusion is my own. If the best personal investment vehicle is a broad index fund representing the whole economy, then you had better be watching national statistics like the BEA, rather than sector statistics. At the moment, the problem is deciphering what's available on BEA.gov in tabular rather than graphics format.

Originally published: Thursday, November 02, 2006; most-recently modified: Tuesday, May 14, 2019