تداوم تورم، انتظارات تورمی، و سیاست های پولی در چین

This paper constructs a quarterly series of GDP deflator inflation for China from 1979 to 2009 and tests for a structural break with an unknown change point in the dynamic inflation process. Empirical results suggest a significant structural change in inflation persistence. Employing a counterfactual simulation method, we show that the structural change is primarily attributed to better conduct of monetary policy and the resultant better anchored inflation expectations. This finding implies that the quiescence of inflation in China over the past decade could well be followed by a return to a high inflation era in the absence of a determined effort by the monetary authorities in managing inflation expectations. Therefore, the use of a preemptive monetary policy to anchor inflationary expectations and to keep inflation moderate is warranted in China.

Since 1979, China has witnessed remarkable changes in the pattern of its inflation process. In particular, the rate of inflation in China is now much lower and less volatile than it was in the 1980s and early 1990s. This change coincides with China's monetary policy reforms since the late 1990s. Significant examples of these reforms include establishments of interbank money markets and bond markets in 1995–1996 and commencement of open market operations in 1998, through which China has changed its monetary policy from direct credit quota control to indirect adjustment with both quantity- and price-based instruments. These shifts in monetary policy may have induced major structural changes in the dynamic models of the inflation process, as articulated in the famous Lucas critique (Lucas, 1976). In the context of the Chinese economy, these important changes in monetary policy may have induced structural instability in the dynamic process of inflation, and in particular the persistence of inflation. That is, the change in inflation dynamics could reflect the fundamental shift in monetary policy whereby the PBOC now systematically acts to stabilize inflation around a potential long-run target and has gained credibility with the public that it will continue to do so into the future. Surprisingly, relatively little effort seems to have been devoted to testing the empirical importance of the Lucas critique in terms of China's inflation dynamics and monetary policy. The existing literature focuses particularly on issues relevant in the United States, while little consensus has been achieved. For instance, Estrella & Fuhrer, 2003 and Rudebusch, 2005 assess the empirical importance of the Lucas critique in US monetary models. Their results suggest that the inflation process with backward-looking descriptions of expectation formation experiences no structural change even when monetary policy changed. However, recent studies by Kim & Kim, 2008, Zhang et al., 2008 and Zhang & Clovis, 2009a suggest a link between changes in inflation dynamics and monetary policy shifts, indicating the empirical importance of the Lucas critique. Notwithstanding these empirical controversies, standard monetary models with inflation dynamics for China assume that the monetary policy regime in China is fixed. For example, Hasan (1999) argues that there is a stable feedback relationship between inflation and monetary growth in China. Guerineau and Guillamont (2005) estimate a dynamic equation of consumer price variations for China over the period 1986–2002, assuming no structural change in monetary policy and the underlying model for inflation dynamics. Zhang (2009) attempts to specify and estimate a price-based (i.e. interest rate) monetary policy reaction function, whilst assuming that monetary policy in China has experienced no structural shift. Recently, Zhang and Clovis (2009b) noted possible changes in China's monetary policy regime, and the parallel change in inflation dynamics. The focus there, however, is on consumer price inflation, and no quantitative evidence linking the change in inflation process and the shift in monetary policy is provided. This paper considers a different and yet (hopefully) more important measure of price inflation in China, namely Gross Domestic Product (GDP) deflator inflation (denoted GDPDI), spanning a relatively long period of time, from 1979 to 2009. We construct a quarterly series of GDPDI by using published data from China's National Bureau of Statistics (NBS) for the growth rate of real GDP and the level of nominal GDP, in conjunction with the estimation method in Abeysinghe and Rajaguru (2004). We then test for structural change in the dynamic models of the inflation process, using structural break tests at unknown break point. In addition, to quantify the link between the structural break in inflation dynamics and monetary policy changes, we specify a multivariate dynamic model to capture dynamic interactions between the real economy, inflation, and monetary policy, and then work through counterfactual simulations. Our empirical results suggest that the structural change in the dynamic models of the inflation process is attributed to improvements in the conduct of China's monetary policy and the resultant better anchored inflation expectations. To this end, Section 2 describes the construction of the GDP deflator inflation series. Section 3 specifies alternative models for inflation dynamics and identifies possible structural break in the underlying models. Section 4 documents systematic changes in monetary policy and discusses the corresponding changes in inflation expectations. Section 5 then performs counterfactual simulations to quantify the contribution of monetary policy shifts to the change in inflation persistence. Section 6 concludes the paper.

In this paper, we have presented evidence of structural instability over time in dynamic equations for inflation in China. The significant shift in the inflation persistence parameter is shown to be the response of inflation expectations to the sequence of reforms in China's monetary policy regime that have taken place since the mid 1990s. This finding suggests that the Lucas critique is not a purely theoretical result, but rather a warning that underscores the importance of applying appropriate structural stability tests to dynamic models for inflation in China. The baseline finding in the present paper also implies that inflation in China is now less persistent and less responsive to inflationary shocks. However, since the structural change in the inflation process is attributed mainly to better monetary policy and the associated better inflation expectations, it is also possible for high inflation to strike back at China in the absence of a determined effort by the monetary authorities to continue to manage inflation expectations. The estimation results of the UCSV model for the underlying inflation series show that the importance of the trend component is still large enough to be economically meaningful, indicating a potential rise of inflation in coming periods. Therefore, systematic monetary policy improvements for managing market expectations of future inflation and hence short-run inflation dynamics are clearly still warranted in China. It should be noted that our search for the causes of the significant structural break in Chinese inflation dynamics is not intended to be exhaustive. Other factors may also influence inflation, and some of these may provide other possible explanations for the recent change in Chinese inflation dynamics. For example, increased globalization and competition may have lowered the sensitivity of domestic inflation to alternative shocks. Another factor possibly influencing inflation in China around the mid 1990s is changing exchange rate policy. Just prior to the identified structural break point in 1995, there was a large depreciation of the RMB against the US dollar, which marked a distinct shift in exchange rate regime in China. Therefore, it could be fruitful for future research to adopt a more structural framework incorporating, when tractability allows, all relevant factors pertaining to inflation dynamics in China. Studies toward this direction may provide more compelling results that may complement the present research.