A more accurate measure of economic output

How well is the U.S. economy doing, and where might it be heading in the future?

To answer these questions, we need a way to assess the total amount of goods and services the economy is producing in a given time period. One measure of this quantity, gross domestic product (GDP) is well known. It estimates the total value of new goods and services produced in the U.S. over a given period, usually a quarter or a year. (Also, the goods must pass through organized markets, so black market activity and goods produced in homes aren't counted.)

However, there's another way to arrive at this estimate of total economic activity: gross domestic income (GDI).

GDI is identical to GDP in theory. Money used to purchase goods and services becomes someone's income in one way or another. But GDI differs in practice due to the way the national income and product accounts are constructed.

Thus, because neither measure is perfect on its own, and the errors in GDP and GDI are largely independent, it should be possible to combine the two measures to improve our estimate of how well the economy is performing in a given time period. That's what two recent strands of research are attempting to do.

The first new measure, called gross domestic output (GDO), is now published by the Bureau of Economic Analysis. It's simply the average of GDP and GDI. Adjusted for inflation, here is how it compares to GDP:

GDO

The graph is from a recent Issue Brief published by the Council of Economic Advisors. It discusses this new measure and notes that "the simple average -- what we have called GDO -- of the initial estimates historically has been a better gauge of the latest and presumably most accurate estimates of GDP growth than either GDP or GDI individually as well as a more stable predictor of future economic growth. Moreover, using GDO helps at least partially to resolve some recent economic anomalies. As a result, GDO offers a valuable new source of information for households, businesses, researchers, and policymakers seeking to understand economic issues in real time."

The second new measure, called GDPplus, is an optimally weighted combination of GDP and GDI, with weights that are allowed to evolve over time.

This measure, which is available from the Philadelphia Fed, has some technical advantages over the simple average discussed above. However, while both GDO and GDPplus improve on using GDP or GDI alone, neither alternative overcomes all the problems with GDP and GDI, particularly the lag of several months before data on GDP and GDI first become available. In addition, it can be as long as several years before all the important data revisions are completed, and forecasts beyond a quarter or two ahead are unreliable. However, GDPplus, unlike GDO, can be calculated even if only one of GDP or GDI is available.

Thus, the best approach to characterizing how well the economy is performing at a moment in time, and how well it's likely to do in the future, is to use a measure such as GDPplus in combination with other windows into the state of the economy such as the unemployment rate, industrial production, consumption, investment and so on.