سیاست های پولی و اعتباری درچین : یک تحلیل نظری

A three-sector model of the Chinese economy is developed in which the activity of state-owned enterprises (SOEs) is constrained by the state-imposed credit plan for working capital. Our analysis indicates the weaknesses of credit control and nominal interest rate variation as tools for influencing the price level; but the hardening of SOEs' budget constraints is found to be an effective device. The existence of credit and currency controls tends to make devaluation contractionary. Furthermore, because of general equilibrium repercussions, policies that boost industrial exports tend to reduce welfare in the agricultural sector, where poverty is concentrated.

The development of substantial elements of a market system in the real sector in China over the past two decades has been accompanied both by bouts of high inflation and, recently, by deflation. To combat these problems, the authorities have relied primarily on monetary policy. 1 However, commercial banking in China, which is dominated by four large state-owned banks, has remained subject to the type of controls associated with central planning. The main instrument of monetary control is quantitative regulation by the "credit plan," which is set by the People's Bank of China (PBC). 2 In formulating the credit plan the PBC collaborates closely with the State Planning Commission. As a result, the credit plan is used not just to control the money supply; it is also used to pursue more microeeonomic objectives, with detailed priorities set for "policy lending," which, in some cases, is even specified at the level of the individual borrower. Policy lending is employedboth to support chosen growth sectors and to underpin loss-making stateowned enterprises (SOEs), and this leads to a large proportion of bank loans being non-performing) Many loans are effectively subsidies extended under government direction (blurring the distinction between monetary and fiscal policy), while often the banks themselves concede parts of loans to ensure SOEs' survival (World Bank 1996; Zou and Sun 1996). In this paper we develop a theoretical model of monetary policy and credit in China, focusing on the role of working capital. 4 This is a particularly important component of the credit plan, being a priority item (Montes- Negret 1995) and constituting about 60% of planned loans (World Bank 1996). We develop a three-sector model in which agriculture is modeled as a separate sector, but industry is split into two sectors, which produce a nontraded and an export good, respectively. 5 For simplicity, the sector producing the non-traded industrial good is assumed to be composed entirely of SOEs, for whom output is constrained by the availability of credit for working capital. 6 Also, by assuming that a portion of the loans in this sector is not repaid to the banks, we allow for SOEs having soft budget constraints. In contrast, the export sector is treated as facing no limits on the availability of credit fbr working capital. This is intended as a stylized representation of the priority that is given to exports in the credit plan (Montes-Negret 1995). However, during periods of severe credit tightening, the difficulties encountered by SOEs have led the authorities to override other priorities and give the SOEs a larger proportion of the credit available. This happened, for example, in early 1996 (see Sachs and Woo 1997). We therefore also modify the model to allow for credit rationing to the export sector. Throughout, the simplifying assumption is made that agriculture has no working capital needs. Because of the complex mix of plan and market in China and the variety of institutional forms under which production takes place, any stylized model of the Chinese economy should be treated with caution. Nonetheless, we believe our analysis indicates the importance of attempting to model the specific features of the Chinese economy, for we obtain results about the effects of monetary policy that are quite different from thoseobtaining in a free-market economy. We find that the impact of credit controls on the aggregate price level can be of either sign. This is consistent with the belief of some Chinese economists that monetary expansion has not caused inflation (see Qin 1994). Furthermore, in our formulation, an increase in the nominal interest rate leads to a higher aggregate price level, while monetary expansion, in the sense of greater initial money balances for households, has an overall contractionary effect on output. This latter relationship has also been found empirically for the short term (up to seven quarters) by Phylaktis and Girardin (1999). Another of our findings is that devaluation may affect output negatively, which accords with the contradietory empirical evidence on the effect of Chinese devaluations (Brada, Kutan and Zhou 1993; Turay 1995). These "unorthodox" results Of our paper are due to the general equilibrium repercussions of tile credit plan and the control of foreign currency. In Section 2 we set up the basic model. In Section 3 we characterize the macroeconomic equilibrium, while in Section 4 we examine various policy changes. We pay particular attention to variation of the credit plan, and we go on to consider a change in the nominal interest rate and a hardening of the budget constraint for SOEs. We also look at various other policy changes, including devaluation. In Section 5 we consider briefly the effect of the imposition of credit rationing in the export sector. In Section 6 we summarize our main conclusions and discuss their implications.

We have attempted to formulate a macro model of the Chinese economy, taking into account some of the differences between production sectors in China, with particular attention to monetary policy. We find that a tightening of the credit plan for SOEs has an effect on the aggregate price index that is unclear in sign. Of course, this does not indicate that the policy is necessarily ineffective; but it does suggest that in the Chinese institutional framework such a policy should be used with care. And we find that the extension of binding credit controls to the export sector should be avoided because there is then a clear-cut positive effect on the aggregate price index. We also find that another commonly used tool of monetary policy in China, interest rate variation, has the opposite of the desired effect. Given the institutional framework of the credit plan, a higher nominal interest rate is associated with a higher aggregate price index (though if credit controls are extended to the export sector, interest rate variation has no significant effects). Money is not neutral in the model. Consider an increase in the money supply in the form of higher initial money holdings for households or as an increased allocation of working capital to SOEs that is then distributed to workers, rather than used for buying intermediates. The price index therefbre rises, but the effects on output are mixed. The output of SOEs is unaffected, whereas exporters produce less and food production rises. We also examine an increase in financial discipline in the form of SOEs honoring a larger proportion of their debt obligations. The effects are opposite in sign to those of an increase in initial money holdings by households. Since the aggregate price index therefore falls, this indicates that tile hardening of budget constraints is an effective anti-inflationary device. An added bonus, through general equilibrium repercussions, is that exports increase. Devaluation, however, is of rather doubtful benefit in the model. It has an unclear effect on exports and the price index, while causing SOEs to reduce output; and the effect on food production is likely to be negative. Finally, notice that a characteristic feature of the model is that policy changes that lower the price index and raise exports tend to reduce welfare in the agricultural sector. (Agricultural welfare is positively related to the terms of trade Pv/P9 and so also to output AF.) Since poverty in China is largely concentrated in the agricultural sector (World Bank 1997) this indicates a significant conflict facing policy makers.