First GDP, now productivity: Another data flaw?
Critical government reports on the nation's productivity growth appear to have the same seasonal problems that CNBC uncovered in the gross domestic product data.
Looking over a 30-year period, CNBC found first-quarter productivity runs nearly a full percentage point below the average for the economy. Productivity is a measure of output per hour, or efficiency, and is a critical data series looked at by policymakers to judge economic health.
"The most important factor determining living standards is productivity growth," Fed Chair Janet Yellen said in a speech last week. "Here the recent data have been disappointing."
But when it comes to the recent 1.9 percent decline reported in the first quarter of 2015, and the massive near-5 percent plunge a year ago, there could be at least some statistical noise at work. Focusing on nonfarm productivity, the gauge most often followed by economists, CNBC found that it has averaged 1.1 percent in the first quarter over the past 30 years, compared with overall productivity growth of 2 percent. Over the past six years, the efficiency record of the country in the first quarter has been nothing short of abysmal, shrinking 1.2 percent on average since 2010, while all the other quarters show some growth. In four of the past six years, first quarter productivity has been flat or negative.
Read MoreWhy is the economy always so weak in the first quarter? Nobody really knows
Of the 20 biggest quarterly declines in productivity over the past three decades, 11 of them have been in the first quarter, including the top three most severe. The single worst decline was the 4.7 percent productivity plunge in the first quarter of 2014.
In April, CNBC revealed that the gross domestic product data, the broadest and most-followed measure of economic growth for the country, was potentially flawed because of unexplained and persistent weakness in the first quarter. First-quarter GDP data averaged 1.9 percent while the economy overall grew 2.7 percent. Economists told CNBC that such a finding shouldn't exist if data were properly adjusted for seasonal patterns.
A string of recently weak first quarters have confounded Fed officials and investors by raising ongoing concerns about the strength of the economic recovery.
The Bureau of Economic Analysis, which compiles GDP, has since acknowledged problems in the data and announced "a multi-pronged action plan to improve its estimates" including reviewing existing seasonal adjustments. Data for the past three years is expected to include new estimates with the July 30 release of second-quarter GDP.
The research by CNBC does not suggest overall growth should be higher, though that is a possibility. Some but not all of the first-quarter GDP weakness is picked up in somewhat stronger second-quarter growth, according to the CNBC findings.
Given the prior research, CNBC's finding on productivity should not be surprising. Productivity relies on GDP data, which provides the "output" part of the output-per-hour calculation. Problematic GDP data should lead to troubles in the productivity data. CNBC's findings show they do.
Read More1Q growth could be stronger than reported: SF Fed
"Anything that's in the BEA data will be in the productivity data," said John Glaser, a supervisor in the productivity department at BLS. Glaser said BLS will be forced to revise its productivity data after BEA adjusts GDP. CNBC found some first-quarter weakness in the hours-worked part of the equation, but most of the problem appeared to come from the output data.
It's unclear how much the weak first-quarter productivity is responsible for the general and worrisome recent trend of crashing productivity in the economy. Since 2010, productivity has averaged just 0.7 percent compared with 2.25 percent over the previous 25 years.