Believe those who are seeking the truth. Doubt those who find it. Andre Gide

Monday, February 2, 2009

A Stimulating Lesson from Japan?

Here we go again. Economic growth is slowing. Stock markets have plunged. Banks are failing. Perennial doomsayers are basking in a glow of perverted pleasure. A plethora of pundits are earnestly explaining the dire need for "stimulative" government spending measures to reverse the course of what will otherwise be a prolonged depression. In short, par for the course.

Well, perhaps not quite par. This time around, many governments appear to be taking seriously the notion that a massive government "electric shock therapy" is needed. Things certainly do seem gloomy out there. Is there any merit in the view that a massive government fiscal action can rescue the day? Apparently, there must be. Why would all these learned people be advocating a policy prescription that is not solidly backed by economic theory and the historical evidence?

To learn more about this view and what underlies it, I decided to visit the arch-liberal Professor Krugman. He begins a recent article by stating that

Everyone’s looking back to the 1930s for policy guidance—and that’s a good thing. But we don’t have to go back that far to see how fiscal policy works in a liquidity trap; Japan was there only a little while ago.
Japan? Well, this sounds promising. What does he have in mind exactly? Evidently, it is the evidence provided in Adam Posen's new book, Chapter 2, entitled Fiscal Policy Works When it is Tried.

Posen argues that, contrary to common perception, Japanese fiscal policy during the 1990s was not really expansionary. On pg. 49 he highlights an exception to this case; a brief period in September 1995. He interprets the corresponding increase in GDP in late 1995 and early 1996 as being attributable to the September stimulus package. This is his evidence that fiscal policy works when it is tried.

Well, maybe he's right. Or maybe not. Quite frankly, I doubt that anyone really knows. In any case, I thought it would be interesting to see how this hypothesis might square up with another look at the data. For this purpose, I chose to consult the Penn World Tables; an international data set frequently used to make international comparisons.

In what follows, I compare Japan to the US. Annual data is available for the time period 1950-2004. I take a look first at levels of real per capita income (GDP) across these two countries. The data look a lot like what one would expect. Relatively stable growth in the U.S., interrupted by the occasional recession. The post-war Japanese growth "miracle," interupted briefly in the 1970s, the late 1980s boom followed by the "lost decade" of economic stagnation.

Next, let's take a look at the available measure of government spending (government purchases; which excludes that sizable chunk of government spending in the form of entitlements). The data reveal a general upward trend, as one would expect in growing economies. Significant bumps in US government spending occur during the Korean war, the peak of the Vietnam war, and in the late 1980s. The growth in government spending in Japan appears more stable and rapid. Note that the 1995 stimulus bill (and subsequent contraction) highlighted by Posen shows up as a relatively minor blip on this chart. Moreover, the pattern appears to have been sharply reversed in the late 1990s and into the 2000s.

A better measure of the "size" of government is to compare it to the "size" of an economy; say, as measured by its GDP. The next figure plots the ratio of government purchases to GDP. By this measure, the relative size of the U.S. government has been in secular decline since the Korean war. There is an even sharper secular decline in the relative size of the Japanese government in the early part of the sample. This latter trend appears to have ended by 1970; and has shown signs of accelerating upward since 1990. On the basis of this data, I think that one might be forgiven for adopting the "mistaken" impression that Japanese fiscal policy has been largely expansionary since 1990.

The last figure I present considers the relative magnitudes of GDP across Japan and the U.S.; as well as the relative magnitudes of the government spending (purchases) share (GPS) across these two economies.
The blue line plots relative income (real per capita GDP). That is, in 1950, real per capita income in Japan was roughly 20% of that of the U.S. By 1970, it was roughly 70%; and by 1991, it was roughly 85%. By any measure, this constitutes a remarkable economic achievement. And this must especially be the case when one considers the devastated state of the Japanese economy at the end of WW2.

The question I want to ask here relates to the behavior of the relative size of government spending over this episode; this is plotted by the red line in the figure. The downward plunge in the early part of the sample reveals that Japanese government spending (as a fraction of GDP) declined rapidly relative to its U.S. counterpart. The upward rise during the latter part of the sample reveals that Japanese government spending (as a fraction of GDP) increased rapidly relative to its U.S. counterpart. The most interesting thing to note is that the blue and red lines are virtual mirror images of each other.

Of course, there are several ways in which the patterns displayed in this data might be explained or interpreted. I am especially eager to learn how this evidence might be construed as supporting the notion that fiscal policy "works" and how this lesson from Japan might be fruitfully employed to cure the current economic crisis.

No comments:

Post a Comment