The Federal Reserve Bank of San Francisco said today a statistical quirk is probably partly responsible for persistently weak economic growth during the first three months of every year.
In the latest volley of a debate embroiling economists from Wall Street to Washington, researchers at the central bank’s San Francisco office said the economy in the first quarter was “substantially stronger” than the government reported last month. After adjusting the data, they found gross domestic product actually expanded at a 1.8 percent annualized rate versus the 0.2 percent gain initially estimated by the government last month.
That meshes with a separate analysis by Barclays Plc Chief U.S. Economist Michael Gapen and by researchers at the central bank’s Philadelphia branch, and puts them at odds with their colleagues at the board in Washington. The disagreement makes it that much harder for policy makers to decipher whether the world’s largest economy is strong enough to warrant an increase in interest rates.
Although the Bureau of Economic Analysis, which produces the GDP data, adjusts its figures for seasonal variations, growth in any given first quarter still tends to run lower than in the remaining three, researchers Glenn Rudebusch, Daniel Wilson and Tim Mahedy wrote in a paper. It’s a phenomenon economists call “residual seasonality.”
The San Francisco Fed researchers applied an additional layer of seasonal adjustments to come up with the stronger reading. From 2000 through last year, first-quarter GDP has averaged 2.3 percentage points lower than the remaining quarters, they found.
“Taking this correction at face value, real GDP growth in the first quarter was stronger and much closer to the economy’s sustainable rate of trend growth,” the economists wrote.
Barclays’s Gapen said that means policy makers and market participants should put more weight on other measures of economic strength.
“The GDP data are less informative right now,” he said. “If you want a contemporaneous view on momentum in the U.S. economy, look to things like payroll growth, the unemployment rate and the ISM,” said Gapen, referring to the Institute for Supply Management’s manufacturing and services surveys. “Those are more likely to be a more accurate picture of where the economy is in any point in time.”
Data released since the government issued its first GDP report last month indicates the economy performed even worse that initially estimated.
Economists’ forecasts compiled by Bloomberg ahead of the revised figures due May 29, unanimously project GDP shrank in the first quarter. Nariman Behravesh, the second-best forecaster of GDP over the past two years according data compiled by Bloomberg, is projecting a 1 percent decline.
The BEA has acknowledged its methodology may produce some distortions. For example, turning monthly data that already been seasonally adjusted into quarterly number could introduce some problems. Also, aggregating different types of line items into bigger categories can create leftover seasonal bias.
“We are continually looking for ways to minimize this phenomenon,” Nicole Mayerhauser, chief of BEA’s National Income and Wealth Division and one of the top economists overseeing the production of the GDP data, said in an e-mail. “One of the areas we’re currently reviewing is possible residual seasonality in measures of federal government defense services spending.”
The BEA is still conducting research as to why this issue shows up in first quarters versus other times of the year, agency spokeswoman Jeannine Aversa said.
The debate surrounding weak first-quarter growth has recently escalated. Federal Reserve economists in Washington published research on May 14 that found “no firm evidence” of leftover bias and that “the first-quarter weakness appears to be driven by a couple of outlier years and by soft readings in a varying subset of underlying components.”
That same day, the Federal Reserve Bank of Philadelphia released a paper arguing the “BEA’s seasonal adjustment procedures are not filtering out all the intra-year movements in the data,” resulting in lower first-quarter estimates. The trend makes economists’ job harder because they “must decide whether an observed slowdown in first-quarter growth represents residual seasonality or the start of a cyclical slowdown.”
Jesse Edgerton, an economist at JPMorgan Chase & Co. in New York, has also joined the fray, saying the trend seems to be more coincidence than a systematic issue with the BEA’s methodology. Conversely, Peter D’Antonio, an economist at Citigroup Global Markets Inc., found that the leftover seasonal bias shaves 1.5 percentage points from growth in the first quarter.
“The message from this is we want to be cautious about reading too much into any weakness that you get in the first quarter, and likewise, into the strength of the rebounds we get from this, because they’ve been overstating the underlying trend,” D’Antonio, who is based in New York, said in an interview. “People kept getting swept away in the lower numbers, ‘Wow things are really bad,’ and then, ‘Wow they’re really good,’ but in fact they haven’t really changed much.”