Debt centralization and soft budget constraint.

Author:An, Zhiyong
Position::Report
 
FREE EXCERPT
  1. INTRODUCTION

    China's new Budget Law (yusuanfa) taken effect in 1995 requires local governments to maintain balanced budgets and forbids them from borrowing in the capital market. However, the potential benefits of local government borrowing are substantial. For example, local government borrowing permits the smoothing of taxes over time, as emphasized by Barro (1979). In order to enjoy the substantial benefits of local government borrowing while still complying with the Budget Law, debt centralization has been adopted by the central government of China in practice. By debt centralization, we mean that the central government of China bundles together the borrowing requirements of local governments into a single bond product, and then disburses the proceeds of issuing the bond product to the latter. The local governments, not the central government, are jointly responsible for the repayment obligation. In other words, the local governments simply "delegate" the central government to issue the bond product for them.

    In this study, we show that debt centralization could result in soft budget constraint (SBC) of local governments by building and analyzing a stylized model that might be new to the literature. Our model is adapted from Tirole (2006) and Banal-Estanol et al. (2011). Their models are put forward in the context of project finance. Basically, they assume in their models that there is an entrepreneur who has two independent and identical projects to finance. The entrepreneur has two options: either (1) finance the two projects through two separate nonrecourse loans or, equivalently, through two different limited liability corporations (i.e., separate financing); or (2) finance the two projects through a single loan or, equivalently, within the same corporation (i.e., joint financing). The basic question facing the entrepreneur is whether he (or she) should adopt separate financing or joint financing. There are two key differences between our model and their models, however. First, in their models, there is only one entrepreneur under both separate financing and joint financing. But in our model, we have two independent local governments under both debt centralization and debt decentralization. Second, the first difference implies that there is no soft budget constraint (SBC) syndrome in their models, but debt centralization could result in SBC of local governments in our model.

    Our work suggests that debt decentralization could help to harden the budget constraint of local governments. By debt decentralization, we mean that each local government issues its own bond product by itself, and is responsible for its own repayment obligation. Because debt decentralization eliminates the origin that gives rise to the SBC of local governments in our model, it could help to harden the budget constraint of local governments.

    Our work has important policy implications. On October 20, 2011, the central government of China gave the authorization to the local governments of Shanghai, GuangDong, ShenZhen, and ZheJiang to issue local government bonds by themselves as a trial, which is a milestone reform on the fiscal relationship between the central and local governments. This trial program of debt decentralization indicates that China is moving from debt centralization to debt decentralization. Our work implies that this "debt decentralization" reform is in the right direction because it could help to harden the budget constraint of Chinese local governments.

    The remainder of the paper is organized as follows. Section 2 reviews the relevant literature, which helps position our work in the relevant literature and make clear our contribution to the relevant literature. Section 3 presents our model and shows that debt centralization could result in SBC of local governments. This section also briefly discusses the consequence of SBC of local governments, i.e., the moral hazard problem of local governments. Section 4 provides some empirical evidence that motivates our research and supports our theory. We also discuss the policy implications of our work in this section. Finally, Section 5 briefly concludes the paper.

  2. LITERATURE REVIEW

    The concept of SBC was first introduced by Kornai (1979, 1980, and 1986) in late 1970s and early 1980s. Basically, the budget constraint of an organization (could be a firm, a local government, or any organization) is said to be soft if it expects to be bailed out in case of financial difficulties.

    Following the seminal work of Kornai, many formal theories have been developed to endogenize SBC. (1) These theories can be classified into two categories: (1) theories on the causes of SBC of enterprises; and (2) theories on the causes of SBC of local governments. Almost all of these theories focus on the causes of SBC of enterprises (e.g., Kornai, 1980; Shleifer and Vishny, 1994; Dewatripont and Maskin, 1995; Segal, 1998; Qian and Roland, 1998; Bai and Wang, 1998; Li, 1998; and Lin and Li, 2008).

    To the best of our knowledge, there are only three prior theoretical papers that focus on the causes of SBC of local governments, i.e., Wildasin (1997 and 2004) and Crivelli and Staal (2013). These three contributions show that positive externalities in the provision of local public goods may create incentives for the central government to rescue local governments in financial troubles. There are at least two points that can distinguish our theory from theirs. First, the origins that give rise to SBC of local governments are different. In their models, the SBC of local governments is caused by positive externalities in the provision of local public goods, whereas in our model, the SBC of local governments is caused by...

To continue reading

FREE SIGN UP