Microsoft Word EcRefChina Oct06. doc

targeted
interventions to boost productivity growth in low-income regions, 
as well as among low-income groups of citizens regardless of where they live.
104
As compared to many other types of redistributional interventions, policy measures 
specially designed to boost factor income among low-income groups through higher 
productivity have the advantage of enhancing rather than harming economic efficiency.
105
Indeed, empirical studies in China indicate that targeted infrastructure investment in poor 
101
A McKinsey report on China (2006, p. 88) argues that a more efficient allocation of financial funds, in 
the form of a shift of lending from SOEs to private firms, could increase GDP byUS
$
259 billion, or 13 
percent, alternatively keep the same growth rate with US
$
259 billion less investment each year – resources 
that could then be used for other purposes. Clearly, all such calculations are highly uncertain. 
102
See, for instance, empirical studies that compare SOEs and private firms by Anming et al. (2003); Xu et 
al. (2005) and McKinsey (2006).
103
Phillips and Kunrong (2005) find that growth rates are higher in geographical areas with a smaller 
share of SOEs – other factors held constant. Some support for the hypothesis that collectively owned 
TVEs are less productive than privately owned TVEs is found in a study by Chang et al. (2003), 
according to which privately owned TVEs tend to generate a higher return on their assets than 
collectively owned TVEs, even when the latter have incentive-oriented contracts for managers. 
104
An example of such policy measures is the poverty reduction programs coordinated by the State Council’s 
Leading Group for Poverty Reduction (LGPR), established in 1986. 
According to Hussain (2003), the incidence of absolute poverty (less than one dollar per day) varies from 0.68 
to 13.5 across cities.
105
There is, however, a risk that targeted infrastructure investment will result in serious waste, since the 
authorities may be tempted to neglect efficiency concerns when distributional aspects become important for 
investment decisions. Indeed, there are many striking examples of such consequences from other countries. The 
heavy government investment in infrastructure in southern Italy during many decades is a well known example. 

---

52
geographical areas (including investment that increases market access) tend to boost 
productivity considerably among firms, not least agricultural firms, in such areas.
Several 
of these studies also indicate productivity improvements as a result of better nutrition, 
sanitation, basic health services and education in geographical areas with large incidence 
of impoverished individuals.
106 
While those who move to urban areas would be the most obvious winners in terms of 
income, some individuals remaining in the countryside would also gain as a result of 
diminished labor supply in such areas, as well as by remittances from family members 
who move to the cities. Some urban insiders would most likely lose, since the upward 
drift of urban wages would be mitigated;
107
this development would, however, also 
contribute to a more even distribution of factor income in the country as a whole. 
Moreover, the reallocation gains could, in principle, be spread through tax-transfer 
programs favoring low-income groups. 
108
Moreover, in a similar way as the shift from collective farms to family farms in the early 
1980s released potential productivity gains in agriculture, a shift to private ownership of 
agricultural land would be expected to do the same. Not only would farmers be 
encouraged to make long-term investments in the land they cultivate. It would also be 
easier to consolidate fragmented patches of land, with productivity improvements as a 
predicted result – not least through better exploitation of returns to scale, for instance, in 
the case of wheat, vegetables and animal products. The potential efficiency gains of 
consolidating land holdings are vaguely indicated by the fact that there seems to be less 
than half a hectare per farm in China (Zhu, 2006, p. 68).
109
An end to the rationing of 
land-lease contracts would also help remove an important source of corruption.
106
Jalan and Ravallion (2002) find that investment in