On The Optimal Size Of Government R&D Expenditure

Since twenty-first century, especially after financial crisis in 2008, the competition of world economy has becoming more and more fierce. In order to face up to the new challenges from the third industrial revolution, and the falter of the economy resurgence, countries around the world pay more attention to the effect of Science and Technology(S&T) to the economic growth, in which the government has become more and more important. It has never been of booming development in this macro background, however, the economic growth of China is facing the challenges from the strategic transformation of the economic structure, and the the new normal of the economy growth. It's a very important thing, therefore, that how to fully understand the effect of government and to define it's border in the field of research and development(R&D).As an important one of public expenditures, the government R&D expenditure plays a very important role to the economic growth, and this has been paid much more concerned in the previous studies. However, it's not enough yet to study the border of the behaviors of government from economic growth. These studies paid more attentions to the effect of government subsidies to firms, and the direct effect of government R&D expenditure to economic growth, and the policies of government to create a good environment to R&D. It's undeniable that these studies have made lots of successful achievements, but they are only indirect side of government behaviors on R&D. The government will participate the R&D in very direct ways, such as providing funds to universities, R&D institutions, and firm, and could even been in R&D activities by itself. So it's very important meaning in theories and realities, that to define the optimal size of government R&D expenditure.Therefore, it's focused on the optimal size of government R&D expenditure. The main objects are to clarify the relationship between government R&D expenditure and economic growth or accumulation of knowledge, and estimate the government R&D expenditure which is based on the Chinese data. We can make the policy according to the results.In this paper, the optimal size of government R&D expenditure is divided into two aspects, they are, the optimal lump-size of government R&D expenditure, and the optimal structural size of government R&D expenditure. In order to study the problems better, it assumes that the productivity of R&D is set. The advantages of this is that it lets us to focus on the main problem.According the objects, some research methods will be used as follows:First of all, the normative analysis is combined with the positive analysis. In the process of positive analysis, the results of normative analysis are based on the positive analysis to reduce the stochastic error which comes from normative analysis. Secondly, the horizontal comparison is combined with the vertical analysis. It not only compares China with other main countries all over the world, but analyze different levels of data from vertical angle. Thirdly, the qualitative analysis is combined with quantitative analysis. Fourthly, different econometric methods are used. The OLS and IV are used to estimate the optimal lump-size of government R&D expenditure, and the SUR(or SURE) is used to estimate the optimal structural size of government R&D expenditure. And last, macro data and micro data are used. We can grasp the overall situation of government R&D expenditure through macro data, and we can find out the problem according to micro data.The main contents of this paper can be divided into three aspects:Firstly, the optimal lump-size of government R&D expenditure. Based on the previous studies, it establishes a basic endogenous economic growth model containing government R&D expenditure, and this model also has a indirect mechanism to show the effect of government R&D expenditure to economic growth, which offer a theoretical support for optimal structural size of government R&D expenditure.In this basic model, the economy has been divided into four sectors, they are final goods sector, the intermediate goods sector, the research sector, and family. The economic behaviors of these four sectors are defined, and the equilibrium solution is obtained by unbalanced growth path. So the size of government R&D expenditure which maximize the equilibrium economic growth, is the optimal lump-size of government R&D expenditure. However, there is no physical capital in this basic model, so there are no dynamics in this model. Therefore, we can't understand the indirect mechanism of the growth of knowledge and economy to the equilibrium growth path. In order to solve this problem, the government R&D expenditure is added into the basic model of new knowledge production function, and physical capital is added into the basic model. The effect of "learn by-doing" is considered, because it's not only connected with physical capital, but a very important indirect accumulation of knowledge. But it's not considered at the same time of direct and indirect ways of accumulation of knowledge. It's very easy to obtain this by some transformations.Secondly, the optimal structural size of government R&D expenditure. The previously mentioned indirect mechanism itself can't study the optimal structural size of government R&D expenditure. The ways of studying this is different to obtain the equilibrium solution of optimal lump-size of government R&D expenditure. A theoretical demand function is constructed by the theory of Revealed Preference Theory's median voter model. The demand for government R&D expenditure of research sector, is the function of tax, real output, other R&D funds, and the number of R&D personnel. The theoretical demand function supports the empirical study of the optimal structural size of government R&D expenditure.Thirdly, the empirical study of the optimal size of government R&D expenditure. According to model (4.27), a econometric model is established, and different levels of data are used to estimate the optimal lump-size of government R&D expenditure. And then, based on previously mentioned theoretical demand function, Seemingly Unrelated Regression(SUR) are used to estimate the optimal structural size of government R&D expenditure.The main results of this paper are:Firstly, the optimal lump-size of government R&D expenditure. (l)There is a nonlinear relationship between the lump-size of government R&D expenditure. When optimal lump-size is reached, more government expenditure on R&D will be negative to economic growth, and then, reduction on government R&D expenditure will not be helpful to economic growth. (2) The improvement of the productivity of R&D, which under the ceteris-paribus conditions, will promote the economic growth which is on equilibrium path. And if the stock of human capital was raised, which under the assumption of qualified human capital, there will be in favor of the equilibrium economic growth.(3) According to the results of empirical study, the optimal lump-size of Chinese government R&D expenditure is 29.32%, the provincial data is 27.02%, the east, middle, and west data is 30.47%,16.83%,38.24%respectively. The results of single province are provided in table 6-6.Secondly, the optimal structural size of government R&D expenditure. (1) The optimal structural size of government R&D expenditure of nation wide are 20.40%, 63.65%, and 13.26%(table 6-17); the east is 23.93%,62.12%, and 13.91%, the middle 28.65%,51.08%, and 20.27%, the west is 16.25%,70.86%, and 12.89%(table 6-18). The provincial data is 26.83%,55.41%, and 17.76% respectively?Besides that, according to nationwide data, the faster other R&D funds grow, the demand of university and firm for government R&D expenditure is smaller, but demand of R&D institution is more. From the results of east, middle, and west data, there are differences among universities, R&D institutions and firms around different districts. There is a negative effect of tax on demand for government R&D expenditure of east firms. With the growing number of other funds, more demand for government R&D expenditure will be in middle universities. And there are widespread shortage of funds in west. (2)According to the congestion quotiety in table 6-19, the R&D product provided by west universities is more similar to the pure public goods, while the the R&D product provided by firm in this district is is more similar to the pure pirate goods. Other results can be found in table 6-19.(3)According to preference matching index, there are inefficient of government R&D expenditure in universities nationwide. More details can be found in table 6-18.Based on results above, some policy recommendations are given as follows:Government has to give fully respect to innovators, and protect theirs rights appropriately. In order to grasp the core competition force, the fundamental research has to be given much more attention, more government funds has to support this. Continue to deepen the reform of S&T-finance system, and regulate the behavior of the government. Utilize the different financial instruments comprehensively to support R&D. Guide the firms to develop the technologies which are favor of the development of society and economy, and use these technologies to leverage the productivity of whole industry. Government should encourage the cooperation among university, R&D institution and firm, and make full use of advantages of each R&D sector, to improve the structural effects of government R&D expenditure.