The GDP numbers for the second quarter of 2020 came out today, and they are ugly. During the months of April, May, and June, the US economy contracted at an annualized rate of about 33%. The first graph below shows real GDP, which measures production by controlling for changes in the price of the produced goods and services. The drop during the second quarter is unprecedented and makes the Great Recession look like a small road bump. The second graph, which shows the percentage quarterly change in real GDP on an annualized basis, is even more shocking.
This terrible outcome is not completely surprising. In my introductory remarks during a recent panel discussion, I had mentioned that in the second quarter the decline in GDP would be much bigger that that in the first quarter, and far greater than what we saw during the Great Recession. I was correct not because of some supernatural fortune-telling ability, but because of Okun's Law. The law takes its name from Yale professor of economics Arthur Okun, who in the early 1960s estimated a relationship in the data between changes in unemployment and the deviation of output from its potential. Okun's original estimate suggested that every 1 percentage point increase in unemployment above its natural rate should be associated with a 3.33% drop in GDP below its potential.
Since Okun's pioneer work, several researchers have tried to estimate the relationship between changes in unemployment and the growth in GDP using different techniques and more recent data. Much of the work finds that the estimate is unstable over time and that the coefficient has fallen to around 2% since the 1980s, meaning that a 1 percentage point increase in unemployment should be associated with only a 2% drop in GDP. This time, however, Okun got closer. As the graph below shows, the rate of unemployment rose by 9.2 percentage points, from 3.8% in quarter I to 13% in quarter II. By Okun's original estimate, if each percentage point increase in unemployment is associated with a 3.33% drop in GDP, then an increase in unemployment by 9.2 percentage points should be associated with a drop in GDP by about 29% after we take into account that GDP should have increase by about 2% (9.2 times 3.33 minus 2%). The drop was actually greater than what Okun's estimate suggests, not lower as the more recent estimates would claim. In that respect, the news are worse than what one would have expected. Nevertheless, using the change in unemployment during April and May, one could have predicted in June, a month before the GDP numbers were released, that the drop in the second quarter would be enormous. So why don't you try some predictions yourselves from now on and see how well you do? It could be fun!
This terrible outcome is not completely surprising. In my introductory remarks during a recent panel discussion, I had mentioned that in the second quarter the decline in GDP would be much bigger that that in the first quarter, and far greater than what we saw during the Great Recession. I was correct not because of some supernatural fortune-telling ability, but because of Okun's Law. The law takes its name from Yale professor of economics Arthur Okun, who in the early 1960s estimated a relationship in the data between changes in unemployment and the deviation of output from its potential. Okun's original estimate suggested that every 1 percentage point increase in unemployment above its natural rate should be associated with a 3.33% drop in GDP below its potential.
Since Okun's pioneer work, several researchers have tried to estimate the relationship between changes in unemployment and the growth in GDP using different techniques and more recent data. Much of the work finds that the estimate is unstable over time and that the coefficient has fallen to around 2% since the 1980s, meaning that a 1 percentage point increase in unemployment should be associated with only a 2% drop in GDP. This time, however, Okun got closer. As the graph below shows, the rate of unemployment rose by 9.2 percentage points, from 3.8% in quarter I to 13% in quarter II. By Okun's original estimate, if each percentage point increase in unemployment is associated with a 3.33% drop in GDP, then an increase in unemployment by 9.2 percentage points should be associated with a drop in GDP by about 29% after we take into account that GDP should have increase by about 2% (9.2 times 3.33 minus 2%). The drop was actually greater than what Okun's estimate suggests, not lower as the more recent estimates would claim. In that respect, the news are worse than what one would have expected. Nevertheless, using the change in unemployment during April and May, one could have predicted in June, a month before the GDP numbers were released, that the drop in the second quarter would be enormous. So why don't you try some predictions yourselves from now on and see how well you do? It could be fun!
No comments:
Post a Comment