Microsoft Word EcRefChina Oct06. doc

firms
has been smaller (and slower) in industry 
and services than in agriculture, the shift to private ownership of 
assets
has been larger. In 
particular, while the dominant physical asset in agriculture – land – is publicly owned, the 
physical assets used by private firms outside agriculture are often privately owned, 
although rental arrangements exist here as well. (A reason why the shift is larger in the 
ownership of firms than of assets in the non-agriculture sector is just that some real assets 
are rented.) Private accumulation of assets has taken place not only via high private 
saving but also as a result of “asset stripping”, when public wealth has been turned over to 
7
A cross-regional regression study by Guo and Yao (2005) finds that not only large overstaffing but also 
huge debt to banks, in fact, have reduced the probability that a firm will be privatized. 
8
By privately controlled firms I mean all production units in China except wholly or majority-owned SEOs and 
those “collective firms” that are still owned mainly by local governments. OECD (2005a, Table 2.1) estimates 
that about 60 percent of aggregate production in China in 2004 was accounted for by the private sector; Tseng 
and Zebregs (2002) have presented similar figures. We would expect that the recent revision of the national 
accounts (presented in late 2005) has increased the statistically recorded private share, since the revision has 
raised the recorded size of the tertiary sector by about 9 percent of GDP. Indeed, in a speech
to the yearly 
congress of Chinese Economists Association on July 3, 2006, the Deputy Finance Minister of China, Jiwei Lou, 
stated that private firms today account for 70 percent of aggregate output.
Since about half the population still works in agriculture, the private share of the Chinese economy is even 
larger in terms of employment.

---

12
private individuals, for instance, in connection with management buy-outs. It is likely that 
the private share of (physical and financial) assets will continue to rise further in the 
future, due mainly to the large volume of plowed-back profits in private firms and the 
high household saving rate (20-25 percent of households’ disposable income). Of course, 
this prediction is based on the assumption that neither capital formation in state firms nor 
government holdings of financial assets will increase even more dramatically. 
The delay of the privatization of firms and assets has not prevented fairly speedy reforms 
in other dimensions of the economic system. In particular, economic decision-making has 
largely been decentralized to households (in the case of consumption) and to firms (in the 
case of production) – schematically depicted in dimension 3 in Figure 1. For the time 
being, however, it is difficult to say whether the degree of decentralized decision-making 
is greater in agriculture or in industry. While individual household farms today have 
basically full authority to make production decisions themselves, government authorities 
continue to intervene frequently in individual SOEs, in particular in large ones, as well as 
in remaining truly collective firms (basically TVEs still owned by towns and villages); 
see, for instance, Chow (1997, 2002). The autonomy of publicly owned firms is also 
(indirectly) constrained by the fact that political authorities often appoint the managers of 
such firms, in particular of large firms. From these specific points of view, 
decentralization of decision-making has not progressed as far in industry as in agriculture. 
On the other hand, the absence of private ownership of land is a more severe constraint on 
the autonomy of investment decisions in farming than in other sectors. In Figure 1, I have 
therefore simply assumed that it is impossible to say in which sector the decentralization 
of decision-making has advanced the most. 
It should be observed that the characterization of the economic system in China today as 
rather decentralized refers to the relation between government authorities, on one hand, 
and firms and households, on the other hand – regardless of whether the relevant 
government agencies represent the national government, provinces or local authorities. By 
contrast, the 
government administration
itself is highly (regionally) decentralized, in spite 
of the centralization of broad political powers to Beijing. Indeed, such (regional) 
administrative
 
decentralization within the public sector existed already before the 
initiation of the economic reforms; see, for instance, Wong (1985). In this sense, the 

---

13
administrative decentralization may be regarded as an inheritance from the pre-reform 
period.
Concretely the decentralization of public-sector administration in China shows up, for 
instance, in great provincial and local government powers in the fields of infrastructure 
investment, government regulations of trade, and various social arrangements (social 
insurance as well as human services). The regional decentralization of public-sector 
administration is therefore an important factor behind the economic and social 
heterogeneity in China. It is also reflected in the small share of public-sector employees 
that work for the central government.
9
An obvious advantage of the administrative 
decentralization is that local initiatives, and local preferences, can play an important part. 
It is, however, often difficult to clarify who is responsible for what in cases when central 
political ambitions are not followed up at local levels. 
An important prerequisite for the shift to decentralized economic decision-making to 
households and firms in China is, of course, that markets and hence price formation have 
largely replaced administrative processes as the basic mechanism for allocating resources 
and coordinating decentralized decision-making (dimension 4 in Figure 1). Indeed, 
markets are the only realistic method for coordinating decentralized decision-making and 
hence exploit decentralized and fragmented knowledge in society (a point emphasized, in