Portal de Periódicos da CAPES

The role of monetary aggregates in Chinese monetary policy implementation

2011; Routledge; Volume: 16; Issue: 3 Linguagem: Inglês

10.1080/13547860.2011.589633

1469-9648

Yuanquan Chen, Richard A. Werner,

Monetary Policy and Economic Impact

Abstract Monetary targeting has been abandoned in deregulated and liberalized financial systems. Theoretically, this could imply that emerging markets that have not yet deregulated financial markets could employ monetarist policies. We analyse the case of China, in order to explore whether monetary targeting was in theory a possible policy framework for the central bank, and to glean policy lessons for emerging markets. Employing Svensson's criteria for the selection of intermediate targets, we find that a measure such as M1 fulfils the criteria and can serve as an intermediate target. However, it is also found that the relatively small error between monetary target and actual variables may be due to alternative monetary policy procedures, in particular, the use of 'window guidance' credit controls. Next, we test the relevance of the McCallum rule in China, which appears more relevant than the Taylor rule. In particular, we find that the actual movement of M1 fits the McCallum rule reasonably well, even during the high inflation period from 1992 to 1994. This suggests that before the official adoption of M1 as the intermediate target in 1994, the People's Bank of China may have already been 'practising' its use by implicitly following the McCallum rule. It is also found that monetary policy was too loose during 1992–1994 and a little too tight during 1998–2002. We conclude that an analysis of the traditional monetary aggregates is insufficient, and research on the role of credit aggregates would appear to be more promising. Meanwhile, policy lessons from our study include that central banks, even in emerging markets that maintain relatively regulated and 'repressed' financial markets, cannot rely too much on quantitative monetary aggregates, if traditionally defined. Keywords: Chinamonetary aggregatesmonetary policyintermediate targetsMcCallum ruleSvensson criteriaJEL classifications: E42E52E58 Acknowledgements Richard A. Werner would like to acknowledge the source of all wisdom (Jeremiah 33:3). Helpful comments from Costas Lapavitsas are gratefully acknowledged. Kostas Voutsinas provided outstanding research assistance. All errors are ours. Notes 1. The amount of money circulating in the 'underground economy' is negligible compared with bank loans; thus, it is unlikely to significantly affect the calculation of monetary aggregates in China (Li 2005). 2. The section so far has drawn on People's Bank of China (1984 The People's Bank of China. 1984. The temporary rules of the People's Bank of China on credit capital management [in Chinese], Beijing: The People's Bank of China. [Google Scholar]), Zhang et al. (1990 Zhang, P., Kuang, S. and Jiang, L. 1990. The credit plan management [in Chinese], Kaifeng: Henan University Publishing House. [Google Scholar]), Shang (1995 Shang, F. 1995. Monetary credit policy and system in China [in Chinese], Beijing: China Financial Publishing House. [Google Scholar]), Cui (2005 Cui, J. 2005. The research on the efficiency of Chinese monetary policy [in Chinese], Beijing: China Financial Publishing House. [Google Scholar]), Wang (2003 Wang, H. 2003. The analysis on the efficiency of monetary policy in China [in Chinese], Beijing: China Renmin University Press. [Google Scholar]) and Zhang (2005 Zhang, H. 2005. Financial institution and monetary transmission mechanism [in Chinese], Beijing: China Financial Publishing House. [Google Scholar]). On window guidance in historical perspective, see Werner (2002a Werner, R. A. 2002a. A reconsideration of the rational for bank-centered economic systems and the effectiveness of directed credit policies in the light of Japanese evidence. The Japanese economy, 30: 3–45. [Taylor & Francis Online] , [Google Scholar], 2003 Werner, R. A. 2003. Princes of the Yen, Armonk, NY: M.E. Sharpe. [Google Scholar]). On its relationship to other monetary policy tools in Japan, see Werner (2002b Werner, R. A. 2002b. Monetary policy implementation in Japan: what they say versus what they do. Asian economic journal, 16: 111–151. [Crossref] , [Google Scholar]). 3. The quarterly constant value of GDP and GDP deflator are not available, so the quarterly CPI is used to measure the price level and calculate RGDP from NGDP. The sample period is from 1987Q1 to 2006Q4. 4. For a detailed survey on views of money and output, see Walsh (2003 Walsh, C. E. 2003. Monetary theory and policy , 2nd ed, Cambridge, MA: MIT Press. [Google Scholar]), Chapters 1 and 5. 5. The unit root tests indicate that the yearly data are all I(1) processes. 6. The Chow breakpoint test was employed in 1994 (in which year the PBC officially adopted the monetary aggregates as the intermediate target) and 1998 (in which year the PBC officially abandoned the mandatory credit plan). The results indicate that there might not be a structural change in these two years. 7. For the detailed requirements, see the website of State Administration of Foreign Exchange at http://www.safe.gov.cn. 8. Open market operations in China began in 1996. In 1997, the PBC injected a monetary base of 70.1 billion and this number increased to 192 billion Yuan in 1999. By the end of 2005, the outstanding amount of central bank bills stood at 2.0662 trillion Yuan. In that year, the PBC cumulatively withdrew base money of 3.5924 trillion Yuan and injected 2.2076 trillion Yuan; in net terms, it withdrew base money of 1.3848 trillion Yuan from the market, more than twice the amount of 2004 (China Monetary Policy Reports, Quarter Four, p. 11). 9. Some economists try to explain this situation from the banks' viewpoint, and offer as reasons financial repression and the incentive to reduce non-performing-loan ratio (Li and Li 2004 Li, J. and Li, J. 2004. The reform of state-owned commercial banks: analysis from a macro angle of view [in Chinese], Beijing: Economic Science Press. [Google Scholar]). 10. The actual figures can be obtained from the authors. The authors also assume different nominal output targets, but this does not significantly change the results.