Here we go again: The latest data show the U.S. economy plodding through another weak first quarter.
But it may not be coincidence. A detailed review by CNBC of 30 years of the government's gross domestic product data, the most followed measure of U.S. growth, suggests a longstanding problem of under-reporting Q1 expansion.
Over some time periods, in fact, first quarter growth is so weak it appears to be measuring a different economy altogether compared to overall growth and the three other quarters. The discrepancy raises the issue of whether investors and policymakers should bet on a second quarter rebound. The government releases its initial estimate for first quarter GDP on April 29.
Over the past 30 years, first quarter growth has been by far the weakest of the four, averaging just 1.87 percent while the economy has grown 2.7 percent, according to the CNBC analysis. Several economists contacted by CNBC said if data were properly adjusted for seasonal variables, such a difference would not show up over a three decade span. They all found the results statistically significant.
"Q1 GDP growth tends to be the lowest of the four quarters," said Tom Stark, manager of the Real-Time Data Research Center at the Federal Reserve Bank of Philadelphia. Stark was one of three economists who duplicated CNBC's results. He said that statistically, "the results seem fairly robust over a number of alternative samples."
Officials at the Bureau of Economic Analysis, which tabulates and compiles the GDP data, initially declined to comment to CNBC. But after the story was published, Brent Moulton, associate director for national economic accounts, said even though the data is adjusted for seasonal patterns, it can still show so-called "residual seasonality."
"BEA is currently examining possible residual seasonality in several series, which may lead to improvements in....the regular annual revision to GDP scheduled for July," he added.
"I find it odd," said UBS economist Drew Matus, whose own work confirmed the CNBC findings. The data, over the three-decade time span, "should have corrected or been corrected." He added, "We have to be cautious in reading too much into data and/or delaying policy because of data that exhibits a bias over long periods of time."
Read More Even after first rate increase, policy will remain easy: Fed's Dudley
Wall Street seems unaware of the phenomenon, or at least does not appear able to anticipate the Q1 effect in its forecasts. An analysis of separate data by CNBC shows that Wall Street has underestimated Q1 GDP 80 percent of the time since 1985, the most for any quarter and with often substantial misses.
The implications for policymakers and investors are potentially significant. CNBC's Rapid Update, a running average of GDP forecasts, shows the first quarter tracking at only 1.2 percent, a full point below the fourth quarter. As the Federal Reserve considers its first interest rate hike in nearly a decade, some Fed officials have said the recent weak economic data give them pause.
The reason for under-performance in the first quarter is ultimately unclear. GDP totals the value of all of the nation's production and is revised multiple times over years. It is seen as a less-than-perfect measure. Data are seasonally adjusted by individual category while the number as a whole is not. It could be that some events that happen fairly regularly in the first quarter—such as really severe weather or declines in defense or business spending—don't happen quite regularly enough to be picked up by seasonal adjustments.
First quarter data since 2010 has been especially depressed, averaging a paltry 0.62 percent while the economy grew 2.3 percent. Those are numbers more like Europe since the recession, not the U.S. The average is heavily affected by two quarters of negative growth, including a 2.1 percent decline last year. That decline was blamed on harsh weather across the nation, but economists still had trouble explaining its recession-like severity, coming after two quarters of strong growth and preceding two quarters of even stronger numbers.
Read More Gasoline, shelter costs lift US consumer prices
And while economists could look out their window and see the lousy weather, they couldn't predict the plunge in GDP in the first quarter of 2014 to minus 2.1 percent: The consensus estimate looked for a positive 1.1 percent, but that was a whopping miss of 3.2 percentage points. The Street was even worse in 2011, looking for 1.9 percent growth in the first quarter compared to minus 1.5 percent.
Average quarterly growth since 1985
But the CNBC analysis shows that the problem is far more persistent than just the post-recession period. CNBC and Stark from the Philadelphia Fed removed the past five years from the analysis and still found the first quarter substantially weaker. Six of the 10 worst quarters for growth since 1985 have been first quarters, including a 5.4 percent plunge in the teeth of the 2009 recession. But even excluding all the negative quarters, Q1 still lags. Whether the economy is in expansion or recession, the first quarter is the weakest of the four.
Working with UBS economist Matus and Stark, CNBC examined whether one or more sectors could be responsible for the first quarter's under-performance. Matus found the effect especially pronounced in commercial structures, business investment and government. Stark looked at 11 separate categories of GDP and found Q1 the worst in eight, especially in exports and federal government spending. Only consumer spending seems immune to the Q1 effect.
Read More
The challenge for investors and policymakers is the extent to which they should look through first quarter weakness. Some of the growth that is not captured in the first quarter appears to end up in the second quarter, where growth runs 60 basis points above normal. That accounts for two-thirds of the under-performance of the first quarter. So it's unclear if overall growth should be higher or if the problem is more an issue of when growth is counted.
What is clear is that the analysis presents a challenge to the government data collectors to investigate whether the process can be improved and help policymakers and investors make better choices, which could ultimately create better economic outcomes.
UPDATE: This story was updated to include a comment from the BEA.