Oops, I Underestimated China’s GDP

February 26, 2012 10:00 AM

More than a year ago, I argued that China’s GDP in purchasing power parity (PPP) dollars had overtaken that of the United States in 2010. This calculation is used in my book, Eclipse: Living in the Shadow of China’s Economic Dominance and plays a role in my conclusion that China has overtaken or is close to overtaking the United States in terms of a broader index of economic dominance. (The broader index combines measures of a country’s GDP measured both at PPP and market dollars, its trade, and its net international creditor/debtor position.) 

My argument was twofold. Based on the work of Angus Deaton and Alan Heston, I argued that the International Monetary Fund’s GDP estimate for China for 2005 was understated by 27 percent. In fact, I used for 2005 the number in the Penn World Table (series China, version 2 in PWT 7; available online since June 2011).

My second argument was that between 2005 and 2010, the International Monetary Fund (IMF) had overstated the increase in the relevant PPP prices in China, and hence understated the increase in GDP between these dates by 20 percent. Conceptually, the “mistake” that the IMF made (and continues to make) is to project these PPP prices based on the evolution of the macroeconomic real exchange rate (changes in a country’s nominal exchange rate vis-à-vis the dollar deflated by changes in aggregate prices between that country and the US dollar). But the computation of the relevant price index for the PPP calculations requires the evolution in the internal real exchange rate, measured as the change in domestic prices of tradable goods to non-tradable goods. In the absence of actual data on these prices, one way to estimate them relies on the Balassa-Samuelson relationship, which links the relevant PPP price to changes in the growth rate of GDP per capita in one country relative to the United States. I used plausible magnitudes for this link to increase China's GDP by about 20 percent relative to what the IMF had done.

Combining these two factors, my GDP (in PPP dollars) estimate for 2010 for China was greater than that of the IMF by 47 percent.

It turns out that I have understated China's GDP, possibly substantially. In a new and important paper, Robert Feenstra and three co-authors argue that the IMF’s estimate of China’s GDP for 2005 (which is basically the World Bank’s estimate for that year) may have been underestimated by as much as 50 percent. Their arguments are partly technical and partly relate to those identified by Deaton and Heston, especially the fact that there was an urban bias in the prices that went into the 2005 estimate for China.  But their analysis is important, not just because it is carefully done, but also because Professor Feenstra will be leading the effort to produce the next generation of the Penn World Table GDP (PPP-based) estimates.

If they are right, and if I therefore might have underestimated China’s 2010 GDP (PPP dollars) by about 20 to 23 percent, three conclusions follow.

First, the IMF’s projection that China will overtake the United States later this decade, which a number of commentators routinely invoke, is incorrect. Either that happened in 2010, as I argued, or it happened even before, according to Professor Feenstra and coauthors.

Second, a key message in Eclipse: Living in the Shadow of China’s Economic Dominance: Namely, the imminence of China’s economic dominance is reinforced because of my possible underestimation of China’s GDP. 

Third, if one believes in the process of convergence or catch-up, which implies that countries slow down as they get richer, future growth projections for China might have to be revised downwards because current standards of living for the average Chinese person are much greater than previously supposed.