Global Market Shares

Yesterday, I linked to an interesting post by Michael Pettis, discussing the rise and fall of Japan in global share of GDP, and suggesting that perhaps the same thing would happen to China at some point going forward.

This morning, I made some graphs from IMF data, looking at this in more detail.  First off, here is the data looked at in the way Michael did it with market exchange rates.  In addition to the US, China, and Japan, I added the European Union and India for additional context.

You can see the story Michael is telling - Japan grew it's share massively in the 1980s and early 1990s, and then lost share, to where China is now crossing Japan.  In the meantime the US and Europe have had broadly constant shares at around 25-30% of total world product.

However, this data is computed with market exchange rates.  Roughly speaking, what that means is that we compute the share of Japan, say, relative to the US, by taking the Yen-Dollar exchange rate at any given time, and using that to translate Japanese official GDP figures in Yen into dollars.  That seems reasonable, but there is one issue with it: exchange rates can fluctuate for a lot of reasons that have nothing to do with the current size of national economic production, but rather reflect the attractiveness of investing in those currencies/nations.  Thus factors like asset price trends and interest rates can cause exchange rate movements that affect the graph above.

For that reason, economists long since figured out "Purchasing Power Parity" comparison methods which are based on figuring out what exchange rate would be required to make an equivalent basket of goods and services in each country cost the same.  When we are trying to compare national outputs, that's generally a better way of performing the comparison (if we want to know who makes more stuff, we'd ideally like to do that comparison in a way in which we count the stuff the same way in both countries).

So the IMF also publishes PPP GDP figures, and if we do the comparison that way, we get a rather different story:

Now Japanese share of global output was roughly flat through the early 1990s, and then started declining.  However, we see that Europe's share of output has been declining for decades, and the US share has been declining since the late 1990s.  Meanwhile, China's share (and India's, and no doubt a bunch of other Asian countries) has been steadily increasing.

So this is more of a Alan Greenpan Age-of-Turbulence type story about the more rapid growth in the ex-socialist countries that have figured out how to do modern industrial capitalism and are growing rapidly as they apply those methods to their countries.  And Japan becomes just another developed country whose market share is shrinking.

That story probably has a ways to run yet.

In particular, I think it's important to remember the issue of labor costs:

Japanese labor costs are within the ballpark of US and European costs, and once that became so, Japanese heroic growth came to a stop, as Japan essentially became subject to the same growth constraints as the rest of the developed world (the difficulty of innovating new techniques) in addition to the after-effects of its bubble/overshoot.  As usual, bubbles get going based on some real factor in the fundamentals (the rapid growth of Japan as it adopted industrial production methods), and then lead to an overshoot after the fundamental trend hits its limits (Japan pretty much reached equilibrium with the rest of the developed world in its labor costs and level of labor productivity/technological sophistication).

Chinese labor costs are still a few cents on the dollar, so massive Chinese growth probably is not about to hit that particular wall for a while - China still has a long way to grow just based on adopting well-known and established techniques to its very low cost labor force, rather than needing to innovate entirely new ones.  So while there may well be hiccups along the way, it's not clear how relevant the Japanese example is just yet.