Who provides the capital for Chinese growth: the public or the private sector?

ABSTRACT

We focus on the role of the government in the provision of investment in China, through the medium of a Dynamic Stochastic General Equilibrium model of the economy in which the form of the production function reflects this governmental role. Using indirect inference, we estimate and test for the elasticity of substitution between government and nongovernment capital in both Constant Elasticity of Substitution (CES) and Cobb–Douglas technologies. The results underscore the strong substitution relationship between government and nongovernment capital from 1949, supporting CES rather than the Cobb–Douglas technology. They also show that the orientation of public investment changed after the start of the ‘Socialist Market Economy’ in 1992: government capital became more complementary to nongovernment capital as it focused more on infrastructure and withdrew from industrial production, intervening only in times of crisis, for stabilization purposes, indirectly via the state banks.

KEYWORDS:

• China
• government investment
• indirect inference
• economic growth

JEL CLASSIFICATION:

• E22
• E62
• O47

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ Singh (2012) suggests the long-run crowding-in effects of public capital on private capital in India.

² Throughout this article, ‘China’ refers to the mainland China, excluding Taiwan, Hong Kong and Macau.

³ In this article, private investment is a part of nongovernment investment since nongovernment sector also includes state owned enterprises and township collective enterprises.

⁴ The private sector has occupied a larger and larger share in the nongovernment capital since the 1978 reform.

⁵ Since 2011, extra-budgetary revenue and expenditure were included in intra-budgetary revenue and expenditure.

⁶ The base year for calculating the GDP deflator is 2000. GDP deflator from 1960 to 2012 is from World Development Indicators (WDI) & Global Development Finance (GDF), The World Bank (issued in July, 2013); GDP deflator from 1952 to 1959 is calculated from the nominal GDP and indices of GDP (1952–1960). There are no price indices for government investment and nongovernment investment, therefore we use GDP deflator to calculate their real values.

⁷ PIM: The net capital stock at period t, $\mathrm{C}\mathrm{a}\mathrm{p}\mathrm{i}\mathrm{t}\mathrm{a}{\mathrm{l}}_{\mathrm{t}}$, can be written as a function of its previous value, $\mathrm{C}\mathrm{a}\mathrm{p}\mathrm{i}\mathrm{t}\mathrm{a}{\mathrm{l}}_{\mathrm{t}-1}$, gross investment in period t, $\mathrm{I}\mathrm{n}\mathrm{v}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{m}\mathrm{e}\mathrm{n}{\mathrm{t}}_{\mathrm{t}}$, and the depreciation of capital in previous period, $\mathrm{D}\mathrm{e}\mathrm{p}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{i}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}{\mathrm{n}}_{\mathrm{t}-1}$: $\mathrm{C}\mathrm{a}\mathrm{p}\mathrm{i}\mathrm{t}\mathrm{a}{\mathrm{l}}_{\mathrm{t}}=\mathrm{C}\mathrm{a}\mathrm{p}\mathrm{i}\mathrm{t}\mathrm{a}{\mathrm{l}}_{\mathrm{t}-1}+\mathrm{I}\mathrm{n}\mathrm{v}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{m}\mathrm{e}\mathrm{n}{\mathrm{t}}_{\mathrm{t}}-\mathrm{D}\mathrm{e}\mathrm{p}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{i}\mathrm{a}\mathrm{t}\mathrm{i}\mathrm{o}{\mathrm{n}}_{\mathrm{t}-1}$. Assuming depreciation at a constant rate δ, then we can rewrite the capital stock as: $\mathrm{C}\mathrm{a}\mathrm{p}\mathrm{i}\mathrm{t}\mathrm{a}{\mathrm{l}}_{\mathrm{t}}=\left(1-\delta \right)\mathrm{C}\mathrm{a}\mathrm{p}\mathrm{i}\mathrm{t}\mathrm{a}{\mathrm{l}}_{\mathrm{t}-1}+\mathrm{I}\mathrm{n}\mathrm{v}\mathrm{e}\mathrm{s}\mathrm{t}\mathrm{m}\mathrm{e}\mathrm{n}{\mathrm{t}}_{\mathrm{t}}$. The method is called ‘perpetual’ because all assets are forever part of the inventory of capital stocks. The quantity of services provided by an asset declines as it ages, but it never reaches zero. PIM is a popular method of estimating capital stock, for example, Chow and Li (2002) and Zhang and Zhang (2003) used this method in calculating China’s capital stock, and Kamps (2006) adopts it to provide internationally comparable capital stock estimates for 22 OECD countries. The estimation results in Sections VII and VIII also show the depreciation rates in different periods, which is another contribution to the empirical literature on China.

⁸ The production can be write as ${\mathrm{y}}_{\mathrm{t}}={\mathrm{A}}_{\mathrm{T}\mathrm{F}\mathrm{P}}{\left({\mathrm{l}}_{\mathrm{t}}{\mathrm{h}}_{\mathrm{t}}\right)}^{1-\theta }{\left[{\alpha \mathrm{k}}_{\mathrm{t}}^{\mathrm{r}}+\left(1-\alpha \right){\mathrm{k}}_{\mathrm{G}\mathrm{t}}^{\mathrm{r}}\right]}^{\frac{\theta }{\mathrm{r}}}\mathrm{e}\mathrm{x}\mathrm{p}\left({\nu }_{\mathrm{t}}^{\mathrm{y}}\right)$, here $\hspace{0.17em}{\mathrm{A}}_{\mathrm{T}\mathrm{F}\mathrm{P}}$ total factor productivity, $\hspace{0.17em}{\mathrm{l}}_{\mathrm{t}}$ and $\hspace{0.17em}{\mathrm{h}}_{\mathrm{t}}$ are the quantity and quality of the labour force, hence, ${\mathrm{A}}_{\mathrm{t}}={\mathrm{A}}_{\mathrm{T}\mathrm{F}\mathrm{P}}^{1/\left(1-\theta \right)}{\mathrm{l}}_{\mathrm{t}}{\mathrm{h}}_{\mathrm{t}}$. Until now, there has been almost no comprehensive method that directly measures the quality of the labour force in China. Moreover, there is no official data on the total factor productivity. Nevertheless, we can evaluate $\hspace{0.17em}{\mathrm{A}}_{\mathrm{t}}$ from Equation (5) since we already have the data of output and the two different physical capital stocks.

⁹ The initial value of capital does not have a large effect on the future capital after several years due to a large depreciation rate. Our experiments show that different sets of initial values of nongovernment and government capital only lead to rather small gaps between their corresponding empirical results, especially in the period from 1993 to 2012.

¹⁰ IN means ‘inside the 95% confidence interval’; OUT means ‘outside the 95% confidence interval’.

Additional information

Funding

The authors thank the financial support from China National Social Science Fund (Grant No. 11CJL017), and the Social Science Funding of Hunan Province (Grant No. 2010YBA215)