The “Things Are Not Always As They Seem” Economy
September 2022
“Things are not always as they seem; the first appearance deceives many.” Okay, so maybe Phaedrus wasn’t ruminating on the nature of economic data, but it fits nonetheless. Even in the most normal of times, the economic data can send often conflicting and confusing signals on the state of the economy. Since the onset of the pandemic, however, the signals being sent by the economic data have been even harder to decipher. The disruptions stemming from the economy coming to a sudden stop and the distortions caused by what in the U.S. was an extraordinarily aggressive policy response have yet to be fully unwound. One implication is that what for decades were normal seasonal patterns in a wide swath of economic activity have been cast into disarray, such that the headline numbers on a given data release are now often at odds with the underlying details. Clearly, the U.S. economy has slowed under the weight of elevated inflation and rising interest rates. What is far less clear, however, is the extent to which the economy has truly slowed and the likelihood that this slowdown will progress into recession.
The second estimate from the Bureau of Economic Analysis (BEA) showed real GDP contracted at an annual rate of 0.6 percent in Q2, compared to the initial estimate of a 0.9 percent contraction. While some have taken the contractions in real GDP in each of the first two quarters of 2022 as “proof” that the economy is in recession, we’ve been on record with our view that the GDP data have said more about the quirks of GDP accounting than the underlying health of the U.S. economy. For instance, because they grew at a slower pace in Q2 than was the case in Q1, business inventories knocked 1.83 percentage points off the quarterly change in real GDP in Q2. Moreover, the readings on real GDP are at odds with the readings on real Gross Domestic Income (GDI). Real GDI expanded at annual rates of 1.8 percent in Q1 and 1.4 percent in Q2, not great by any stretch, but neither is this a sign of a contracting economy.
In theory, GDP and GDI are measuring the same thing, simply from different vantage points; GDP is an expenditures-based measure of all final goods and services produced in a given period, while GDI measures the income generated in the production of those final goods and services. When the two measures have diverged in the initial estimates, revised data have tended to lean toward where the initial estimate of GDI was, and we won’t be surprised if that is the case this time around. Our premise will be put to the test on September 29, when the BEA releases their annual comprehensive revisions to the recent historical GDP data.
Even if the revised data show real GDP tracking real GDI more closely over 1H 2022 than has thus far been reported, that will still leave a decided disconnect between the GDP data and the employment data. Total nonfarm payrolls increased by 2.663 million jobs over the first half of 2022 as real GDP is reported to have been contracting, but the disconnect is even more pronounced than has appeared. The Bureau of Labor Statistics (BLS) released the preliminary results of its annual benchmark revisions to the data from its monthly establishment survey, the basis of the BLS’s estimates of nonfarm employment. Each year, the establishment survey data are benchmarked to comprehensive counts of employment based on payroll tax returns, which all employers must file, as of the month of March. Based on the preliminary 2022 results, the level of nonfarm employment as of March 2022 will be revised up by 462,300 jobs, equivalent to 0.3 percent of total nonfarm employment and much larger than the average revision (0.1 percent) over the past ten years.
The upward revision is not a surprise. Given how the dynamics of the labor market have shifted since the onset of the pandemic, particularly the meaningfully higher degree of turnover, we’ve suspected the errors associated with the initial estimates of job growth have been larger over recent quarters. Still, while the direction of the revision to nonfarm employment isn’t surprising, the magnitude of the revision is, with one implication being an even larger disparity between the GDP data and the employment data. While the pending revisions to the GDP data may narrow that disparity, they won’t eliminate it.
As the GDP data now stand, nonfarm labor productivity is reported to have declined at annual rates of 7.4 percent in Q1 and 4.1 percent in Q2. With the revised employment data yielding a meaningful upward revision to the estimate of aggregate hours worked, the declines in productivity over 1H 2022 figure to be even larger than first reported, which is hard to fathom. Unless of course you believe that firms were taking on workers at a break-neck pace in order to have them sit idle all day every day. Or, it could be that “quiet quitting” is really a thing and the only people who haven’t caught on to it are the bosses who don’t know their subordinates aren’t really working.
Looming revisions to key data have the potential to re-shape perceptions of the economy’s path over 1H 2022. Then again, it may be that the revised data offer no more clarity than did the initial estimates, particularly since the more recent data have been just as muddled. For instance, a “surge” in labor force participation pushed the unemployment rate up to 3.7 percent in August from 3.5 percent in July despite a sizable increase in employment. A substantial portion of August’s increase in the labor force, however, is no more than seasonal adjustment noise, leaving a much smaller actual increase in the labor force. This is no trivial matter given the extent to which depressed labor force participation remains a drag on job growth and an accelerant to wage growth. While the ISM Manufacturing Index held steady at 52.8 percent in August, the survey details show more industry groups reported declines in new orders and production than reported increases, the takeaway being that the expansion in the factory sector is slower and less broadly based than had been the case. Finally, while lower gasoline prices have led to significantly smaller monthly changes in the Consumer Price Index, the over-the-year change remains above 8.0 percent.
As they attempt to assess the economy’s capacity to absorb further increases in interest rates, FOMC members are trying to process incoming economic data that continue to send decidedly mixed messages. What does seem clear is that, while the economy has slowed, the labor market remains notably robust, with the unemployment rate still well below what many FOMC members would associate with full employment, while inflation remains far above the FOMC’s 2.0 percent target rate. All of which leads us to expect another 75-basis point hike at this month’s FOMC meeting.
Sources: Bureau of Economic Analysis; Bureau of Labor Statistics; Institute for Supply Management
