Macroeconomists who criticize micro-data beware, government macro statistics also rely on surveys:
If the first-quarter G.D.P. report is wrong, can we expect it to be changed? The B.E.A. will release annual revisions at the end of this month, but there is not likely to be a lot of new information by then. As a result, any change might not be reported until 2018, the next scheduled comprehensive revision. By then, figures for this year’s first quarter are likely to be of little interest.
Whether or not the estimates are changed soon, the wildly fluctuating reports for the first quarter remind us that “there are literally hundreds of preliminary assumptions that go into estimates of economic barometers,” said Robert J. Barbera, a longtime Wall Street economist who is now the director of the Center for Financial Economics at Johns Hopkins University. “Much of the art of economic forecasting is knowing when to say, ‘Nah.’ ”
This is, he said, one of those times. “The first-quarter real G.D.P. number simply makes no sense.”