Microsoft Word EcRefChina Oct06. doc

efficiency
of the growth path. There are several reasons to expect that 
the growth path has not been very efficient. One reason is the previously mentioned 
malfunctioning of factor markets, in particular credit and capital markets. While the 
financial markets in China have succeeded in intermediating the large volume of 
household saving to firms, the allocative efficiency of the process seems to have been 
poor. Obvious indicators are that the geographical mobility of capital and credit is low, 
that firms plow back profits rather than return them to the capital market, and that public-
sector firms are favored as compared to private firms.
There are also indications that the 
government tends to reallocate funds from regions with high returns to regions with low 
returns.
41
39
An alternative interpretation of the gradualism would be that it was an unintended result of the 
political process. The experimental nature of the process suggests, however, that gradualism was quite 
intended. An often quoted slogan among the political leaders was that “the river should be crossed with 
the feet solidly touching the stones”. 
40
For instance, Maddison (1998 and homepage); Young (2003); and Garnaut and Song (1999) estimate that 
GDP growth during the reform period was about 7.5 rather than 9.5 percent per year. By contrast, official 
statistics on GDP growth during the last decades are, in fact, expected to be revised upwards somewhat in 2006, 
with a yearly growth rate since the early 1990s of slightly above 10 percent per year. 
41
For various indications of inefficient allocation of capital in China, see, for instance, Levine (1997); 
Kahn and Senhadji (2000); Wen and Zhan (2001); and Boyreau-Debray and Wei (2005). 

---

28
As a result of the priorities given to state firms in financial markets, large capital-intensive 
firms have been favored relative to small, medium-sized and labor-intensive firms.
42
Indeed, according to estimates by the People’s Bank of China (2004), small and medium-
sized firms received less than ten percent of the bank loans in the early 21
st
century, 
although they produced more than half of GDP.
43
There is a multitude of explanations of 
why the state banks favor large firms, such as lower perceived risk (because of implicit 
government guarantees on the loans to large firms) and political pressure from local 
authorities that fear severe employment problems if large firms run into financial 
difficulties. The alleged discrimination of SMEs in financial markets has, of course, also 
considerably disfavored private firms, and hence indirectly the service sector, both of 
which are generally regarded as disfavored also in other ways, for example by various 
regulations, including the allocation of land-lease contracts, building permits, etc. (World 
Bank, 2003b). The emergence and expansion of informal financial markets have, of 
course, mitigated the distortions of the allocation of resources brought about by 
deficiencies in formal financial markets. However, the development of informal financial 
markets cannot possibly have eliminated the distortions, since the duality of financial 
markets in itself is a distortion, reflected in different levels of interest rates.
The large resource costs of the growth path in China are also reflected in the high 
investment ratio, about 43 percent of GDP in recent years (“fixed capital formation”). 
This reflects not only the exceptionally large industrial sector (about 46 percent of 
GDP) but also the high capital-output ratio 
within
that sector.
44
By comparison, during 
similar development phases, Hong Kong, Taiwan/China, and South Korea grew nearly 
as fast, with smaller investment ratios (about 35 percent of GDP). The high capital 
intensity in China is also reflected in the relatively high marginal capital/output ratio – 
in the interval of 4-5, while more “normal” ratios are 2.5 to 3.5.
45
The marginal capital/ 
output ratio has also increased during the last decade in connection with a rise in the 
investment share from 35 percent of GDP in the mid-1990s. It is tempting to interpret 
42
SMEs are defined in China as enterprises with between 8 and 2.000 employees, less than US $ 50 
million assets, and less than US $ 37 million sales. 80 percent of these firms are estimated to have been 
privately owned in 2001 according to Citybank. 
43
For similar calculations, see McKinsey (2006, Exhibit 3.5).
44
As in many other developing countries, this figure may be biased upward because statistics are likely 
to be more complete on investment spending than on GDP. Presumably, the revision of the national 
accounts reported in December 2005 has taken care of this problem, at least to some extent.
45
With an investment ratio of 0.45, the marginal capital/output ratio is 4.7 when GDP growth is 9.5 
percent.

---

29
this as a fall in the macroeconomic return on real investment
46
-- a fall to be expected 
when the capital stock rises extremely rapidly.
The large aggregate investment in real capital assets may be compared with the 
modest spending on education, which is often reported to be 4-5 percent of GDP 
(4.3 percent according to the OECD, 2006), of which 2.7 percentage points are 
financed by the public sector.
47
It is unlikely that these proportions can be 
rationalized with reference to the relative returns on these two types of 
investments
48
.
Hence, there is probably a general efficiency argument for 
reallocating spending from (aggregate) investment in physical assets to (aggregate) 
education in China, a point made, for instance, by James Heckman (2005). In 
principle, resources could then be freed for the consumption of both ordinary 
consumer goods and human services, such as health care, without jeopardizing fast 
GDP growth. (Today, private consumption is not more than 40 percent of GDP.) 
49
The modest size of the service sector is often regarded as another indicator of 
inefficiencies in the allocation of resources across production sectors in China. This 
point has recently been weakened, although not eliminated, by revisions of the 
national accounts (in December 2005), whereby the reported GDP share of services 
was raised from 32 percent to 41 percent – still, however, a fairly modest figure for 
a country at China’s current stage of development. 
Naturally, China’s capital-intensive growth strategy has also constrained the ability 
of non-agricultural sectors to absorb the surplus labor in agriculture. This helps 
46
McKinsey Global Institute (2006, Exhibit 3.23) calculates that the marginal capital/output ratio has 
increased from 3.30 in the first half of the 1990s to 4.9 after 2001. 
47
Revised national accounts 2005, and UNDP (2005 Figure 3.9). 
48
Studies of the return on secondary and tertiary education in China in recent years, based on wage 
differentials, usually give returns in the interval 7-8.5 percent. A recent study by Zhang et al. (2005) 
concludes that the return on education has increased from 4.0 percent per year of schooling in 1988 to 
10.2 percent in 2001. By contrast, a twin-study (to avoid selection bias due to innate ability) gives lower 
figures (Li et al., 2005), basically because of an asserted zero return on high school education, which the 
author asserts to function just as a device for screening ability. By contrast, one more year of