A Look at Recessions Part 1: By the President’s Political Party
I am going to have a few posts on recessions over the next few weeks. This one is short – I’m trying to get another batch of copy edits out to my publisher. In this post, I’m focusing on the period from 1929 on – that’s the period for which real GDP per capita is available. Because I’m swamped, I’m going to put up some graphs and let them do their own talking.
The graph below shows the length of each recession in our time period, in months. It also assigns a recession to the party of the President who happened to be in the Oval Office at the time the recession started.
Now, perhaps a recession that starts shortly after a president takes office is not entirely that President’s fault. Nor can you unambiguously blame someone who has already left office, as the former President might well have taken steps that might have avoided the recession had he not been in office. So the next graph only assumes a recession to a party if that party was in office for three or more years by the start of the recession. Three years seems a bit much for me, but I don’t have the time to argue right now, so I’m picking a big enough figure that nobody should complain. (Gray bars mean a recession falls into the three year no man’s land.)
What about the severity of the recession? Well, before I put up those graphs, a couple comments…
1. the BEA has quarterly data from 1947 on, and annual prior to that
2. I didn’t do any smoothing – I don’t want anyone complaining about what method I chose to do it. Thus, I simply assign each month the closest reported real GDP per capita. Thus, for instance, I assign Jan, Feb, and March of 1981 the real GDP per capita of 1981 Q1. Its not perfect, but it is what is and I don’t have to defend when I start downturns and how I’m going to smooth things when the recessions don’t start or end at the start or end of a quarter or year.
So here’s the annualized loss by recession, also by party. Note that a very short recession could have a bigger annualized loss than a recession which lasted longer but had a bigger overall drop in real GDP per capita. Thus, there may be a trade-off between recessions length and monthly loss. (A topic for another week.)
And here it is with the 3 year deadzone pulled out:
Have at it.