The case for stronger first-quarter growth

The Bureau of Economic analysis recently reported that GDP grew at only 0.2 percent in the first quarter of this year. However, the San Francisco Fed believes the procedure used to adjust GDP for seasonal effects has flaws that resulted in an estimate that's too low and that the actual number is around 1.8 percent.

That's still not a booming economy, but it's a lot better than the official estimate.

The seasonal variation in GDP is relatively extreme. For example, GDP drops by approximately 10 percent in the first quarter of each year and then rises by around 20 percent in the second quarter. When these fluctuations are left in the data, detecting changes in the underlying rate of growth becomes difficult.

For example, the following graph from the SF Fed report shows GDP from 2000 through 2006 unadjusted for seasonal variation along with the seasonally adjusted measure. The report notes that it would have been very difficult to see the 2001 recession (the shaded area in the figure) using the unadjusted figures.


Seasonally adjusted data gives a much clearer picture of underlying trends, but a problem apparently exists with the adjustment that causes first-quarter estimated growth rates to be too low. The SF Fed finds that, in the 1990s, real GDP growth was on average 1 percentage point lower in the first quarter as compared to the subsequent three quarters of each year. From 2000 to 2014, this difference increased so that first-quarter estimates are now around 2.3 percentage points lower than in subsequent quarters.

To correct for this "residual seasonality," the SF Fed put the GDP figures through a second round of seasonal adjustment and obtained the following results:


This is an important issue for the Fed as it tries to determine when to begin raising interest rates. If first-quarter GDP grew at 1.8 percent instead of 0.2 percent, the Fed is more likely to raise interest rates sooner rather than later.

However, it should be noted that not everyone at the Fed agrees with the SF Fed analysis. A research note from economists at the Federal Reserve Board of Governors in Washington, D.C., said "our analysis here does not find convincing evidence of material residual seasonality in GDP in recent years." They suggest that other factors, such as unusually harsh weather, may be responsible for the low rate of economic growth at the beginning of this year.

Other measures of economic activity such as changes in employment have been fairly solid in recent months, and that points to the SF Fed's estimate as being more accurate, but the uncertainty can't be fully resolved. In the end, the Fed will have to determine the course of monetary policy based on imperfect knowledge about the true state of the economy.