In lieu of an abstract, here is a brief excerpt of the content:

  • The Return to Capital in China
  • Chong-En Bai, Chang-Tai Hsieh, and Yingyi Qian

China has one of the highest investment rates in the world, over 40 percent of its GDP in recent years. A natural question to ask is: Does China invest too much? On the one hand, China is still a low-income economy, with a capital-labor ratio that is low compared with those of advanced economies, and thus the potential returns to investment could be high. On the other hand, as Robert Lucas pointed out,1 other constraints, such as low levels of human capital, backward technology, and low quality of institutions, may limit the realization of these potential high returns in China as in other developing countries. The fact that capital often flows from poor to rich countries reminds us that the return to capital is not always higher in poor countries.

What does it mean to say that China invests too much? A natural metric to use in answering this question is the return to capital. For example, China's economic growth rate might have been so high that the return to capital has fallen little, if at all, despite high investment rates. Put differently, the investment rate in China might be high precisely because the return to capital in China is high. The questions to be asked, then, are: Has the return to capital in China fallen significantly over time? Is it now low relative to returns in other countries?

Another issue concerns the allocation of investment within China—whether China has invested too much in certain sectors or certain regions and too little in other sectors and regions. Does the return to capital differ [End Page 61] significantly across sectors and provinces in China? Has this dispersion of returns changed over time?

This paper measures the return to capital in China, calculated using data on the share of capital in total income, the capital-output ratio (where both capital and output are measured at market prices), the depreciation rate, and the growth rate of output prices relative to capital prices. Although the approach is conceptually straightforward, the major challenge is the data, which we discuss below before presenting our estimates.

Although we are not aware of any other papers that estimate the aggregate return to capital in China, many papers have reported estimates of the capital stock in the course of estimating productivity growth in China.2 Our estimates of the capital stock in China differ from these earlier estimates in two principal ways. First, we make use of the updated data reported by China's National Bureau of Statistics (NBS) after the 2004 census. Second, we calculate the capital stock in market prices rather than in constant prices. We do this because our goal is to calculate the return to capital, which is a function of the capital-output ratio measured at market prices.

We begin by discussing the methodology we use to estimate the return to capital. We next discuss the data and address several potential measurement problems. We then present our estimates of the aggregate return to capital in China, first for a base case using simple aggregate measures and then for a number of alternatives. These include alternative sectoral concepts that remove residential housing, agriculture, and mining, and alternative capital concepts that include inventories and consider various depreciation rates for fixed capital. We also measure after-tax returns for our base case, and we compare our base case estimate of the return to capital for China with estimates for other economies. Finally, we consider the efficiency of capital allocation in China by measuring the dispersion of the return to capital across sectors and regions and how it has changed over time.

Our base case estimate shows that the aggregate rate of return to capital in China fell from roughly 25 percent between 1979 and 1992 to about 20 percent between 1993 and 1998 and has remained in the vicinity of 20 percent since 1998. These rates of return are above those for most advanced economies calculated on a similar basis. They are also high relative to a large sample of economies at all stages of development. Estimates...

pdf

Share