Ed Lotterman: Many read far too much into GDP numbers

One swallow proverbially does not make a summer, and one quarter’s economic growth data, good or bad, do not make a trend. Keep that in mind when reading the latest national economics news.

The U.S. Commerce Department’s Bureau of Economic Analysis last week released its “advance” estimate of the value of goods and services produced in the U.S. economy in the first three months of 2015. With output, adjusted for inflation, growing at a tabulated annual rate of 0.2 percent, the news was not encouraging, especially because the last quarter of 2014 had shown growth 11 times as high.

Moreover, like most economic indicators, the definitions and context of the information presented needs to be understood. So let’s review some basics.

Gross domestic product, or GDP, is a money measure of the value of goods and services an economy produces. In general, more things to meet people’s needs are better than fewer things, but don’t make the common mistake of reading more into GDP numbers than is there. It is not some metric that magically tells us everything important about how we are doing.

Remember also that the numbers announced are estimates of the actual underlying production activity. It would be prohibitively expensive to tabulate every single good or service produced over three months. That would be like conducting several decennial censuses in a row in just a few weeks. So an estimate of a total is derived from measurement of a subset of that total.

If done consistently and carefully, the resulting estimates are comparable over time and are of great use. But they are not perfect measurements.

There is a further complication in that for any given quarter, numbers are issued four different times. The ones released April 29 are the “advance estimate.” This is a “quick and dirty” tabulation made from production data that is available very quickly from a bevy of sources. Based on long experience, the Commerce Department knows how these correlate with subsequent tabulations made with additional information that takes longer to accumulate. Given the magnitude of the task, overall accuracy isn’t bad. But occasionally there are very large revisions between the “advance,” “second,” “third” and “latest” versions of the data. The second version comes out one month after the advance. The “latest” revisions take place more than a year later.

In the whole process, one must recognize that some numbers are “lumpy.” An aircraft manufacturer like Boeing has workers doing various tasks every day on planes being built. But the finished planes being sold to an Asian airline enter the “exports” rubric of GDP statistics when ownership of a batch actually passes to the purchaser. This can be hundreds of millions of dollars at a crack.

If such a transfer took place Dec. 28, 2014, it was part of fourth-quarter 2014 “output.” If the official date of delivery was a week later, it would be booked as first quarter 2015. Ditto for large wheat or soybean export sales. So some of the announced quarter-to-quarter variation is an artifact of the tabulation process rather than a feature of actual factory operations. Experienced Federal Reserve officials and other policymakers know this and place less weight on short-term bobbles than many in the media.

Then there is the matter of seasonal variation. Some activities, such as domestic tourism and recreation or construction, are greater in summer than in winter.

With no adjustment, tabulated GDP therefore would decline in the fall. But retail sales burgeon in the fourth quarter because of the holidays. To accommodate all of this, as with many other data series, statisticians apply a seasonal adjustment. But the adjustment does not account for specific weather extremes in some given year that may stop winter construction and recreation or speed up spring tourism. Extreme weather in parts of the country was a factor in the last two quarters nationally.

As with any other indicator, output numbers are of greatest use when taken in context of how they stand in a long-term process and in relation to other sources of information.

There are times, such as in the financially chaotic period from September 2008 through April 2009, when fear and desperation force officials, businesses and even some households to hang on the latest items of information. But even in such times, it is dangerous to rely too much on one data item.

The world economy clearly is in a dicey state. The Federal Reserve clearly has to make some difficult decisions. The decline in announced national output for this quarter may, indeed, be the beginning of a sobering downward trend. But it may also be a temporary blip that will disappear as the year progresses. For most of us, there is little to do but wait and see.