زیرساخت، امور مالی دولت جایگزین و رشد تصادفی درونزا

This paper constructs a stochastic version of an endogenously growing economy with a public good that raises the productivity of private capital. We explore how growth and welfare are influenced by changes in the mean and variance of productive public spending under two alternative financing methods, mixed money–bond financing and wealth-tax financing. In addition, to evaluate the differences between money financing and bond financing, we consider mixed money–bond financing, under which a larger ratio of bonds to money is utilized to finance a given increase in public spending.

Government expenditure financing is regarded as an important factor explaining variations in economic performance, and many economists have extensively examined the theoretical interaction between aggregate activity and the way that public spending is financed. For example, based on the framework of a nonmonetary Ramsey growth model, Turnovsky (1992) examines the steady-state effects of changes in public spending under different financing methods. Using the monetary growth model of Sidrauski (1967), Gokan (2003) investigates how alternative government financing affects the speed at which capital stock converges to a steady state. Ploeg and Alogoskoufis (1994) develop Sidrauski's model with noninterconnected overlapping generations to compare the growth effects of public spending financed by creating money, issuing bonds and imposing lump-sum taxes. Palivos and Yip (1994) utilize cash-in-advance model of Stockman (1981) to assess the relative merits of tax financing and money financing. However, the present paper differs from the existing literature in two respects. First, the above economists assume that public spending has no direct impact on the behavior of the private sector. It can be interpreted as being either a real drain on the economy or alternatively as some good that does not influence the productivity of private capital. In the study of public spending, it is convenient to distinguish between government consumption expenditure and government infrastructure expenditure. The economists above restrict their attention to government consumption expenditure, and the concept of government infrastructure expenditure is excluded from their analyses. Clearly, an alternative discussion of the role of government spending is required within this topic. To capture the role of infrastructure capital, this paper incorporates public capital into the production function as an input. However, owing to the analytical complexities, we are restricted to considering a steady-state equilibrium and examining the steady-state effects of alternative government financing. Second, the stochastic factors are completely ignored in the existing alternative government financing literature.1 Unforeseen policy shocks occur over time, and the economic activities are inherently subject to the unforeseen risk. Thus, the importance of the interaction between the stochastic factors in the public sector and the real economy should be emphasized in this field. We extend a general equilibrium model to consider a richer, more realistic picture of government sectors. This paper allows us to clarify whether the mean and variance of productive public spending could mitigate the size of business cycle fluctuations in the steady state. In the present paper, productive public capital is introduced in a stochastic AK model, as studied in Turnovsky (1993) and Grinols and Turnovsky (1993). It should be noted that public spending corresponds to a flow of public capital. There are two different financing methods to raise a given increase in public spending: mixed money–bond financing and financing through taxes on wealth. Hence, we can compare how the level of welfare and the mean and variance of real growth are affected by increasing the first and second moments of productive public spending under the two different financing schemes. In addition, to explore the qualitative differences between money financing and bond financing, this paper considers a mixed money–bond financing under which government relies on a larger ratio of bonds to money to raise the necessary revenue. As for tax and monetary policies, we examine the growth and welfare effects, maintaining fixed levels of the first and second moments of public spending. To put it briefly, the purpose of this paper is to clarify how government should behave to raise a given increase in public deficits or to obtain a given level of public deficits from growth and welfare perspectives. To express the productive services derived from publicly provided infrastructure, there exist a number of papers analyzing dynamic general equilibrium models where public spending or a public good is introduced in a production function. See Barro (1990), Futagami et al. (1993), Turnovsky and Fisher (1998), Cazzavillan (1996), Glomm and Ravikumbar (1997), Zhang (2000) and Raurich-Puigdevall (2000). In the above literatures, however, the stochastic disturbances are completely ignored and the role of alternative government financing is not emphasized. On the other hands, Turnovsky (1999) considers the economic uncertainty but not the different sources of government revenue. Fisher and Turnovsky, 1998 and Turnovsky, 2004 explore the alternative modes of government financing, but ignore the stochastic factors in the economy.

In this paper, taxes on wealth or both money and bonds are used to finance a given increase in productive public spending. Let us summarize the merits and demerits of mixed money–bond financing relative to wealth-tax financing in a stochastically growing economy. (i) As the mean rate of public spending is increased, the expected rate of real growth becomes larger, whereas the variance of real growth becomes smaller. On the one hand, the increase in the expected rate of growth is larger under mixed money–bond financing, whereas on the other hand, the decrease in the variance is larger under wealth-tax financing. It is ambiguous whether the level of welfare improves or deteriorates as a result of increasing the mean rate of public spending. However, the positive (negative) effect is smaller (larger) under mixed money–bond financing than under wealth-tax financing. Therefore, from a growth (welfare) perspective, mixed money–bond financing (wealth-tax financing) is preferred. (ii) As uncertainty about government spending increases, the expected rate of real growth becomes smaller, whereas the effect on the variance of real growth is indeterminate. Mixed money–bond financing leads to a larger fall in the expected rate of growth, whereas wealth-tax financing, on the other hand, leads to a larger increase (a smaller fall) in the variance. In addition, how the level of welfare is affected by a greater uncertainty regarding government spending is ambiguous, but mixed money–bond financing results in a smaller fall (larger increase) in the level of welfare. Therefore, from a growth (welfare) perspective, instability of government spending under mixed money–bond financing is more (less) serious than under wealth-tax financing. To explore the qualitative difference between money financing and bond financing, we consider mixed money–bond financing when a larger ratio of bonds to money is utilized to finance a given increase in public spending. We can summarize how such a decision affects the consequences of mixed money–bond financing as follows: (iii) As the ratio of bonds to money increases, the positive effect of an increase in the mean rate of public spending on the rate of nominal interest becomes larger. However, the growth and welfare effects are unaffected by government's decision. (iv) As the ratio of bonds to money increases, the negative effect of an increase in the uncertainty about public spending on the rate of nominal interest becomes smaller. However, even so, the growth and welfare effects are not influenced. From growth and welfare perspectives, it is not significant whether the government prints money or issues bonds to raise the necessary revenue in this stochastic AK model. Finally, let us state the growth and welfare effects of changes in policy shocks, such as the mean and variance of the rate of monetary growth and the rate of taxes on wealth. (v) As the mean (variance) of the money growth rate increases, the expected rate of real growth becomes larger (smaller), as does the variance of real growth, whereas the welfare level is smaller (larger). (vi) As the mean rate of taxes on wealth increases, the expected rate of real growth becomes smaller, as does the variance of real growth, whereas the welfare level increases. However, the variance of the rate of taxes on wealth has no impact upon the first and second moments of real growth and the level of welfare. Suppose that the government deficits in this period are given, i.e., g is fixed in period t. Then, from a growth (welfare) perspective, it is more desirable that government relies on the rate of monetary growth (the rate of taxes on wealth) to obtain the given revenue.