Comparing Presidents, Real GDP per capita…All the Lags

by cactus

Comparing Presidents, Real GDP per capita, All the Lags You Can Eat Three Ways From Thursday

I didn’t really wanna write this post, but I keep getting told that when you look at how Preznits did on the economy, lags, lags, and more lags are what matters. I’d like to move on to more useful stuff (i.e., back to how taxes, oil prices, etc. affect growth) so I’m going to bite the bullet and try to cover everything lag related here and now.

Now, before I start, every single bit of information in this post comes from the BEA’s NIPA Table 7.1, line 10. The data and the analysis are also shown here.

To recap the last few episodes, data goes back to 1929, and we looked at the growth rate from the full year before a President took office to his last full year in office… that is to say, the baseline from which he was measured was the period before he took office. For Presidents who left office early due to death or excessive Nixonizing… if they lasted into the second half of the year, I considered that a “full year.” That leaves this as the recurring, somewhat inconvenient character with halitosis and a big ol head of lettuce in his teeth:

Graph 1: Annualized Growth in Real GDP per capita

Perhaps the worst offense in that graph is the FDR performance, and over the past few weeks I tried a number of things to tone that sucker down… but even if you stick to the period up to the ’38 recession he still outshines everyone. So I’m going to pick a period where FDR is safely out of the picture – I’m going to leave out his term, as well as that of his illustrious predecessor and his successor. (I should note – reader Cantab has noted that if you make FDR responsible for only the recessions that occurred during Ike’s administration, FDR no longer looks so good. (Thanks for the tip Cantab!) That also has the advantage of focusing on a period of relative stability with no direct superpower v. superpower global wars or Great Depressions.

Now, to lags… lags are one way to assume that a President’s policies don’t have an immediate effect. But lagging by a year, say, assigns to one President some of the growth which depends on his successor’s behavior. For example, if you lag Carter’s term by one year, you get Reagan’s first year, a year in which Reagan’s long pre-advertised tax cuts came along. Ditto Clinton and GW. Which would mean that you essentially mean that Reagan’s first year in office owes more to the policies of a man that was no longer in office than to the policies that everyone knew Reagan was going to put in place at least since the previous November. For that reason, I prefer to simply leave out the first year of a President’s term to putting in lags, but no matter…. below are graphs showing the real GDP per capita by President, in one case leaving out each one’s first year in office, in the other lagging by one year:

I gotta say, either way, Ford is looking like he’s hoppin’ but the Dems still outperform the Reps, by either measure.

Now, say you want to assume that it takes a full 2 years before a President’s policies have any effect. Thus, if one fine November some slick dude like Clinton tells folks their tax rates are going to be a-rising in the following year, not only will that have no discernible effect when he raises taxes, it won’t have any effect two years later. Put another way… you’re assuming the American public is incredibly dense, because people catch on awfully slow. Now, if that’s your assumption, here’s what it looks like:

Ford no longer looks all that hot in the top graph… because he’s no longer there – the man only served two years, after all, so leaving out his first two years drops him out completely. And Dems still look better than Reps. But in the bottom graph, where we’re talking true lags, Reps marginally outperform Dems and Ford is the best performing of all the Republican Presidents.

Now let’s get more extreme… say we’re going three years before a President’s policies have an effect. Below are graphs for annualized growth in real GDP per capita when the first years of the President’s term are left off, and for annualized growth in real GDP per capita with three year lags:

Well, this kinda gives us mixed results. We’re finally showing what everyone knows is true, namely that Reagan is the bestest Preznit evah in the top half, but it shows Dems outperforming Reps. The second graph – the lag version – no longer has Reps outperforming Dems, but its close. Ford remains the best performing Republican President.

Let’s go with four years:

I’m not going to comment much about this, except to note that the top graph obviously shows the Presidents who served at least one full term plus one year. Also, there’s no point in dragging this charade forward any more since its obvious where things are going.

So let me close this off with a comment… anyone who wants to argue that the data (cherry-picked to leave out FDR) shows that Republican Presidents follow better economic policies than their Democratic counterparts had better understand that there’s only graph in this post that allows you to tell a story that is even almost-kinda-sorta compatible with that point of view. Which means that you had better be familiar with the implications of that graph. That is to say, if you are going to insist that Republican Presidents are more growth-oriented than Democrats, be prepared to:

1. explain why no President has an effect on the economy for his first two years in office
2. say it loud and proud: Gerald Ford is the guy all Republicans should seek to emulate

A reminder… the data comes from the BEA’s NIPA Table 7.1, line 10. The data and the analysis are also shown here.

Now if we can put this lags thing to bed, maybe we can go back to trying to figuring out the why of all of this. Underperforming someone as hapless as the typical Democrat occupying the Oval Office should not, frankly, be difficult, and the failure to do so speaks of some seriously deluded policy and implementation.

by cactus