Docstoc

The Effectiveness of Monetary Policy in China

Document Sample
The Effectiveness of Monetary Policy in China Powered By Docstoc
					             The Effectiveness of Monetary Policy in China
                                  During 1993-2004

                                         Min Gong

                            Center for Macroeconomic Research

                                          Wenpu Li

                            Center for Macroeconomic Research




Abstract: This paper investigates the effectiveness of monetary policy in China. We estimate
a VAR(2) model, using quarterly data for the period 1993: 1- 2004: 4 to investigate how the
change in money supply impacted GDP and RPI. Our empirical results indicate that (i) there
exists a limited effect of monetary policy shocks on output over the sample period, and (ii)
the change in monetary supply is not associated with the change of price index in China
during this period. Furthermore, we attempt to evaluate the causes of the impact of monetary
policy on output and the price index from the perspective of a transition economy. We hold
that the method by which monetary policy is implemented during the transition period
strongly influences the effectiveness of policy. It can be proved that administrative measures
have been necessary, and may in fact be more effective than monetary policy instruments in
China. In addition, it is necessary to pay special attention to maintaining an appropriate and
adequate increase in money supply in order to advance the sustained, rapid and sound
development of the national economy.

Keywords: effectiveness of monetary policy, vector autoregression model, transition
economy
1. Introduction

Since the economic reform programme started in the late 1970s, the Chinese economy has
experienced a number of episodes of pronounced economic growth, interrupted by generally
sharp but short-lived periods of slowdown. Given that the People’s Bank of China (PBC) has
a history of announcing monetary targets (generally in terms of an aggregate such as M2),
one can easily conclude that the conduct of monetary policy is central to understanding the
behavior of output and inflation in China before 1996.

Contractionary monetary policy, combined with the reform of the financial system,
successfully cooled down the rapidly rising inflation during the period 1993-1996 in China at
a low sacrifice ratio of GDP. In fact, before 1996, the effects of monetary policy aimed at
curbing inflation by decreasing the money supply and credit contraction proved to be
remarkable. In contrast, though the PBC continuously lowered the reserve ratio and interest
rates, expansionary monetary policy did not significantly contribute to rises in the GDP
growth rate and inflation during the period 1997-2003. As a result, the Chinese economy
slipped into periods of deflation. Specifically, the retail price index (RPI) declined at an
average rate of 1.2 per cent per annum, compared with an average rise of 9.7 per cent in
1979-96, while annual GDP growth slowed to 7.1 per cent in 1999 from 9.6 per cent in 1996.
Growth subsequently rebounded, reaching 8.4 per cent in 2000, and then fell back to 7.2 per
cent in 2001.

These developments have led analysts to wonder whether and to what extent monetary policy
can impact the real economy. There are two opposing opinions on the effectiveness of
monetary policy: one is that the effects of monetary policy are faint because of a lack of
money transmission mechanisms in China. The other opinion states that monetary policy can
not only affect output in the short run, but also influence price adjustment in the long run. In
this paper, we apply a vector autoregressive model (VAR) to evaluate the validity of
monetary policy during the period 1993-2004 in China within one econometric framework.
Together in analyzing the characteristics of the Chinese economy as a transition economy and
investigating the mechanisms by which monetary policy affects output and price index, we
hold that (i) monetary policy shocks have a limited effect on output over the sample period,
and (ii) deflation in China has been primarily ascribed to aggregate supply factors in recent
years.

The paper is organized as follows. In Section 2 we provide a brief overview of China’s
overall economic performance, including the practice and role of monetary policy since the
late 1990s. We survey the literature on the effectiveness of monetary policy with an emphasis
on the China’s experiences. We present two criticisms to be tested: (i) the effect of monetary
policy shocks on output does exist, but only to a limited extent from 1996-2004, and (ii) the
change in monetary supply is unrelated to price index changes during the sample period in
China. In Section 3 we apply a standard VAR framework to explore the empirical relationship
between money supply, price level and economic growth during the period 1993-2004. A
description of the data together with model estimates and their interpretation is also provided

                                               1
in this section. In Section 4 we analyze the factors that affect the role of monetary policy in
the economic transitional process. We hold that the effectiveness of monetary policy depends
on the methods by which policy is implemented, and relies on the extent to which the money
supply impacts investment. Section 5 concludes.

2. Overview of China’s Economic Performance and the Practice of Monetary Policy

Since the economic reform programme started in the late 1970s, the Chinese economy has
consistently maintained rapid economic growth. Meanwhile, the Chinese economy has
experienced a number of episodes of pronounced economic growth, interrupted by generally
sharp but short-lived periods of slowdown. Figure 1 suggests a clear demarcation between the
eras of inflation (1990-1994) and disinflation (1994-2003) in both inflation and output growth
performance. In 1994, China's inflation roared to its highest ever level since China was
established in 1949. The 1994 retail price index reached 21.7 per cent and the consumer
price index reached 24.1 per cent. In response to the upsurge in inflation, the Chinese
government adopted a tight monetary policy to curb inflation. The People's Bank of China
(PBC) adopted a variety of monetary policies to exercise strict control over the total number
of currency credits. Combined with the reform of the financial system, the Chinese
government successfully cooled down the rapidly rising inflation in 1993-96 with a low
sacrifice ratio. The consumer price index (CPI) fell to 8.3 per cent in 1996 from 24.1 per cent
in 1994, while annual GDP growth slowed to 9.6 per cent from 12.6 per cent during the same
period. By 1996, China has succeeded in a soft landing. In fact, these cyclical swings have
been associated with macroeconomic policy (monetary policy) implementation at different
periods before 1996. The effects of monetary policy aimed at controlling inflation by the
decrease in money supply and credit contraction are considered highly remarkable (Zhou
1998). Overall, throughout more than 20 years of economic reform, China has experienced
rapid growth without suffering damaging inflation. Tight monetary policy played a significant
role in curtailing inflation.

Beyond 1996, macroeconomic conditions in China worsened, with a shortage economy being
replaced by a surplus economy, a phenomenon unprecedented since reforms began. The
consumer price index (CPI) rose 2.8 per cent in 1997, but fell 0.8 per cent in 1998 and further
decreased to 1.4 per cent in 1999 before it rose again by 0.4 per cent in 2000 and 0.7 per cent
in 2001. Since 1998, the retail price index (RPI) had been continuously sliding downwards at
an average rate of 1.2 per cent per annum, compared with an average rise of 9.7 per cent in
1979-96, while annual GDP growth slowed to 7.1 per cent in 1999 from 9.6 per cent in 1996.
Owing to the accelerating slowdown of the economy the Chinese government finally changed
its monetary policy stance from contractionary to expansionary. Since the latter part of 1997
up until now, an expansionary monetary policy has been in place to fight against deflation.
Although the PBC lowered the deposit reserve ratio twice, reduced interest rates on seven
occasions, and expanded the open market operations, aggregate demand and the price level
have failed to show any significant improvement. As a result, the Chinese economy slipped
into a period of disinflation (Yu 2004a).



                                               2
Compared with the significant role that monetary policy played in curtailing inflation before
1996, it seems that the deflationary pressure experienced throughout the late 1990s was an
indication of the eroding effectiveness of monetary policy. At least, monetary policy in China
has asymmetric effects over the business cycle,that is, the effects of monetary policy are
greater in a boom than in a recession. These developments have led analysts to wonder
whether and to what extent monetary policy could impact the real economy.

           Finger 1 Relationship between GDP Growth and Inflation in China
                                                                                    30
                                               24.1
                                      21.7                                          20


                                                            8.3                     10
                         16
                                                      6.1
                         14                                                         0
                         12             12.6
                                                                                    -10
                         10                                 9.6

                          8
                          6
                          4
                              90 91 92 93 94 95 96 97 98 99 00 01 02 03

                          the growth rate of GDP(lefe scale)      RPI inflation(right scale)
                          CPI inflation(right scale)

                         Sources: China Statistical Yearbook, 2004.

There is no consensual opinion among Chinese economists about the effectiveness of
monetary policy. In fact, there are two contradictory opinions: one is that the effects of
monetary policy are faint because the lack of a monetary transmission mechanism in China.
Yu(2004a) points out, because the PBC could no longer effectively control the money base,
and because the money multiplier has become increasingly unstable, the monetary
transmission mechanism failed to work well, which in turn resulted in a loss of monetary
policy effectiveness. Ai(2002) argue that, the decline in money velocity is attributed to the
increase in the degree of monetization, acceleration in the process of financial modernization,
and the rise in saving rates. Fei(2003) focus on the channels of circulation for money
increments, and found that a portion of the monetary supply increase seeped into a stock
market “funnel”, or banking system “black hole”, which are responsible for the weakening
effects of monetary policy.

On the other hand, a huge amount of empirical models have been developed to explain the
effectiveness of monetary policy in recent years. Some empirical research among these have
shown that monetary policy can not only affect output in the short run, but also influence
price development in the long run in China. For example, based on the estimate results from
VAR including real GDP, CPI, money supply (M2), and the real interest rate, Zhao(2002)
showed that changes in M2 influenced output in the short term, and may be attributable to the
price movements. Xie(2004) found evidence that there exists long-run monetary neutrality in
China, and the margins of price fluctuation are proportionally associated with changes in
money supply. In this regard, Liu(2002) also provides an empirical analysis of the long-run
stable relationship between CPI and money supply.

                                                            3
Nevertheless, neither of the two contrary opinions mentioned above is in line with the
Chinese economic performance and the practice of monetary policy from the late 1990s up
until the early 2000s. Firstly, to refer back to the period’s tendency towards deflation in
concert with aggressive fiscal policy, the PBC adjusted its methods of monetary policy
implementation in 1998. Since then, the increased money supply was used to finance
enlarged government expenditures in public works investment, as a means of stopping a
decline in the growth rate of investment from 1993. As a result, the economic growth was
maintained at a slightly lower than an average rate of 8 per cent per annum (1998-2002)
(Table 1). This outcome demonstrates the validity of monetary policy in the late of 1990s. We
conclude that monetary expansion, combined with a proactive fiscal policy, stimulated
significantly surges in investment.

Secondly, regarding the price level since 1998, China's price level has been continuously
going down, though the annual growth rate of M2 reached 17.6 per cent at one point, an
average rate of approximately 14.2 per cent per annum was realized over the period
(1998-2002) (Table 1). It is hard to find a remarkable relationship between the change in
money supply and price movements. As we know, the Chinese economy has experienced
structural changes in the transition period. The structural changes include a rise in the degree
of marketization and an inflow of FDI, which has improved the level of potential output,
resulting in a faster increase in aggregate supply than the expansion in aggregate demand (Yu
2004a; Fan 2003; Woo 2003). These factors may in part account for a weakening relationship
between monetary supply and price adjustment. Therefore, we hold that within China the
deflation experienced in recent years has been primarily driven by contemporary aggregate
supply factors; prices have been declining as supply exceeds demand.

Table 1 Growth Rate of GDP, Investment, and Money Supply in China: 1996-2003 (%)
                    Investment in       Money &                     Currency in
           GDP                                      Money(M1)
                     fixed assets   quasi-money(M2)               circulation(M0)
  1996      9.6          14.8              25.3       18.9              11.6
  1997      8.8           8.8              19.6       22.1              15.6
  1998      7.8          13.9              14.8       11.9              10.1
  1999      7.1           5.1              14.7       17.7              20.1
  2000      8.0          10.3              12.3       16.0               8.9
  2001      7.5          13.0              17.6       12.7               7.1
  2002      8.0          16.9              16.8       16.8              10.1
  2003     10.0          27.7              19.6       18.7              14.3
Sources: China Statistical Yearbook, 2004.

In this paper, we put forward two criticisms to be tested: (i) the effects of monetary policy
shocks on output do exist, but only in a limited sense since the late 1990s, and (ii) the change
in monetary supply is unrelated to changes in the price index in the period of 1996-2004 in
China.




                                               4
3. Empirical Analysis

To empirically examine our two criticisms, an appealing approach is to estimate a vector
auto-regression (VAR) model, which will assist in the analysis of how the changes in money
supply affect output and the price level, and the contribution of monetary factors to the output
fluctuation and price movements. In this paper we adopt the usual approach of estimating a
trivariate VAR model to explore the joint behavior of output, the price, and money stock in
China from 1993:1 to 2004:4. The impulse responses generated by the estimated VAR can be
used to find out qualitatively whether shocks of monetary policy affect output movements,
and influence price adjustment. The variance decomposition results generated by VAR will
provide quantitative evidence, and thereby determine the importance of monetary factors.

Data and sample limitations constrain model complexity under the current circumstances. We
use quarterly data with the sample period 1993:1- 2004:4 to estimate the VAR model,
consisting of real GDP, inflation and money supply (M1 or M2). Money supply measures are
used to proxy the direct effects of monetary policy since institutional considerations imply
that an interest rate instrument is unsuitable in China’s case. We simply point out that the
PBC relies primarily on the control of monetary aggregates and less so on an interest rate
instrument, although the PBC has not fully disclosed its intentions regarding such strategies.
In fact, the PBC does not have an interest rate target and the several key interest rates are for
the most part administratively determined.

The data has been collected from the National Bureau of Statistics (NBS, formerly known as
the State Statistics Bureau), the China Statistical Yearbook, the People’s Bank of China
website, and the Chinese Economic Information Net website (http://db.cei.gov.cn).

Because statistical administration in China is still developing, only nominal GDP was
available on a quarterly series, while no GDP deflator on a quarterly series is currently
published. Instead CPI and RPI at monthly rates are available. Therefore, we converted the
monthly CPI data series to a quarterly data series, and computed real GDP by dividing
nominal GDP by CPI. Additional seasonal adjustments were performed to remove seasonal
disturbances in real GDP levels. In addition, the money stock is measured using both M1 and
M2. Figure 2 plots the performance of GDP, M1, M2 and the price index over the sample
period.

Unit root tests indicate that output, money supply, and prices are I(0), see Table 2 for details.
Therefore, we follow the standard procedure to order the monetary policy instruments M1 or
M2 (in logs) last in the VAR model, after GDP (deflated with the consumer price index and in
logs) and RPI (in logs). The M1 series is sample frequency is quarterly, running from 1993:1
to 2004:4, whereas the M2 series is runs from 1994:1 to 2004:4.




                                                5
            Figure 2 GDP, M1, M2 and Price Index in China: 1993:1 – 2004:4
                      11.0                                                                                6.5


                      10.5                                                                                6.0


                      10.0                                                                                5.5


                        9.5                                                                               5.0


                        9.0                                                                               4.5


                        8.5                                                                               4.0
                              93    94    95    96   97    98       99    00   01    02    03    04             93    94    95    96    97       98    99    00    01    02    03    04



                               (a) Nominal GDP (in log)                                                               (b) Real GDP (in log)
               11.5                                                                                       12.5


               11.0
                                                                                                          12.0

               10.5
                                                                                                          11.5
               10.0

                                                                                                          11.0
                9.5


                9.0                                                                                       10.5
                      93      94    95    96    97    98       99    00       01    02    03    04               94    95    96        97    98        99    00     01        02    03     04



                                         (c) M1 (in log)                                                                          (d) M2 (in log)
                130
                                                                                                           130
                125

                120                                                                                        120

                115
                                                                                                           110
                110

                105
                                                                                                           100
                100

                   95                                                                                       90
                        93     94    95    96    97       98    99       00    01    02    03    04              93    94    95    96       97    98    99    00    01        02    03    04



                        (e) CPI                       (f) RPI
          Source: Chinese Economic Information Net website (http://db.cei.gov.cn)


                                                                              Table 2 Unit Root Tests
                      Test     Test                            ADF Test Critical
        variable                              Sample
                      type     equation                        Statistic      Value
        ln(GDP)       ADF      (C, T, 1)      1993.1-2004.4 -3.7026           -3.5045*
        ln(rGDP)      ADF      (C, T, 1)      1993.1-2004.4 -3.7216           -3.5045**
        Ln(M1)        ADF      (C, T, 1)      1993.1-2004.4 -4.0965           -3.5045**
        Ln(M2)        ADF      (C, T, 1)      1994.1-2004.4 -4.2740           -4.1896*
        lnRPI         ADF      (C, 0, 3)      1993.1-2004.4 -3.2050           -2.9241**
* denotes significance at 1 per cent level; ** denotes significance at 5 per cent level.

The impulse responses derived from our model estimation allow us to investigate the real
effects of monetary policy shocks. Figure 3 presents the impulse response functions of real
GDP to a monetary policy shock on a horizon of 16 quarters over the estimation period. It

                                                                                                      6
contains four graphs: (a) and (b) indicate the impulse response of real GDP to M1 innovation
and M2 innovation respectively, (c) and (d) plot the impulse response of real GDP (seasonally
adjusted) to M1 innovation and M2 innovation, respectively.

The results suggest that monetary policy is primarily responsible for Chinese macroeconomic
performances since the late 1990s. The increased money supply is a contributing factor to
sustained growth in the national economy. Interestingly, we find that the response of output is
sensitive to the chosen GDP data series. In the case of the seasonally adjusted GDP series
clearly M1 and M2 shocks are relatively large. Besides, monetary shocks have long-run
impacts on output. A one per centage point increase in M1 leads to an increase in real GDP of
roughly 0.05 per cent after two quarters, although this effect eroded down to less than 0.02
per cent after two years. By contrast, a one per centage point increase in M2 only leads to an
increase in real GDP after four quarters of roughly 0.005 per cent.

While impulse responses are useful, perhaps a more helpful technique comes from an
examination of variance decompositions (VD). Tables 3 presents VDs for the GDP case
where real GDP is seasonally adjusted. The left hand side of Table 3 shows the results from
the model with M1 while the right hand side provides results from the model with M2. We
find that M1 shocks can explain about 21.39 per cent of variation in output after four quarters,
and 24.91 per cent after two years, whereas M2 shocks can explain 2.46 per cent of variation
in output after four quarters, and 8.25 per cent after two years.

                      Figure 3 Impulse Responses of GDP to a Monetary Policy Shock
                                                                                       Response of LRGDP to One S.D. LM2 Innovation
               Response of LRGDP to One S.D. LM1 Innovation
                                                                                   0 . 0 8
       0.04

                                                                                   0 . 0 6
       0.02


       0.00                                                                        0 . 0 4


       -0.02                                                                       0 . 0 2


       -0.04                                                                       0 . 0 0


       -0.06                                                                      - 0 . 0 2


       -0.08                                                                      - 0 . 0 4
                  2       4       6       8       10    12    14    16                     2       4       6       8    10        12        14        16




 (a) response of real GDP to M1 innovation                                   (b) response of real GDP to M2 innovation
           Response of LRGDPSA to One S.D. LM1 Innovation                          Response of LRGDPSA to One S.D. LM2 Innovation
                                                                                 0 . 0 2 0
          0.08

                                                                                 0 . 0 1 5
          0.06

                                                                                 0 . 0 1 0
          0.04
                                                                                 0 . 0 0 5

          0.02
                                                                                 0 . 0 0 0

          0.00                                                                   - 0 . 0 0 5


         -0.02                                                                   - 0 . 0 1 0
                      2       4       6       8    10    12    14   16                    2    4       6       8       10    12        14        16




             (c) response of real GDP                                             (d) response of real GDP
      (seasonal adjustment) to M1 innovation                                    (seasonal adjustment) to M2 innovation
      Source: calculated by author.



                                                                         7
                        Table 3 Variance Decompositions for GDP
                   Real                                  Real
       Horizon                RPI      M1     Horizon           RPI                                       M2
                   GDP                                   GDP
          1         100        0        0        1        100     0                                         0
          2       61.2072 17.4287 21.3641        2     98.0929 0.1619                                    0.7452
          3       61.1601 18.9741 19.8658        3     97.2095 1.3420                                    1.4485
          4       58.5087 20.1005 21.3908        4     96.2504 1.2854                                    2.4643
          5       57.2381 20.6146 22.1473        5     94.9701 1.2869                                    3.7430
          6       56.0476 20.8208 23.1317        6     93.3129 1.4762                                    5.2109
          7       55.0992 20.8621 24.0388        7     91.4202 1.8309                                    6.7489
          8       54.2858 20.8047 24.9095        8     89.4841 2.2618                                    8.2542
      Source: calculated by author.

Regarding the impact of M1 and M2 shocks on price level, Figure 4 plots the impulse
response of RPI to a monetary policy shock at a horizon of 16 quarters over the estimation
period. Figure 4a indicates the impulse response of RPI to M1 innovation, and Figure 4b
presents the impulse response of RPI to M2 innovation. We find that monetary shocks have a
positive, albeit temporary impact on RPI.

                   Figure 4 Impulse Responses of RPI to a Monetary Policy Shock
                Response of LRPI to One S.D. LM1 InnovationResponse of LRPI to One S.D. LM2 Innovation
       0.006                                                          0 . 0 0 6


       0.004                                                          0 . 0 0 4

       0.002
                                                                      0 . 0 0 2
       0.000
                                                                      0 . 0 0 0
       -0.002

                                                                     - 0 . 0 0 2
       -0.004


       -0.006                                                        - 0 . 0 0 4
                  2    4     6     8    10    12    14    16                   2   4   6   8   10   12   14   16



      (a) response of RPI to M1 innovation                         (b) response of RPI to M2 innovation
     (with seasonal adjustment GDP series)                         (with seasonal adjustment GDP series)
      Source: calculated by author.

Meanwhile, Tables 4 presents VDs for the RPI case from the estimated VAR model where
real GDP is seasonally adjusted. The left hand side of Table 4 shows the results from the
model with M1 while the right hand side provides the results from the model with M2. We
find that money supply shocks can only explain less than 1 per cent of variation in RPI.

In the case of deflation, we cannot find evidence that money shocks can explain Chinese
price movements. These results suggest that the change in monetary supply is unrelated to the
change in price index during this period in China, denoting that inflation and deflation in
China are more likely driven by aggregate supply than exclusively, or even primarily, an
aggregate demand phenomenon in the late 1990s.




                                                               8
It is important to review the reasons that might explain why the price levels have fallen in
recent times. There is a subtle difference in opinions regarding whether deflation in China
was primarily driven by aggregate demand or supply factors, which has important broader
implications for our understanding of the role of monetary policy. By taking into account the
structural change factors, improvements in market-oriented reform, and the inflow of FDI, we
conclude that aggregate supply shocks have become, in relative terms, the more important
factor in movements of the price level during the economic transition to a more
market-oriented economy, or, in other words, China’s deflation since the late 1990s was
supply-driven.

                        Table 4 Variance Decompositions for RPI
               Real                                     Real
  Horizon                  RPI        M1     Horizon                      RPI         M2
               GDP                                      GDP
     1        9.7813     90.2188       0        1      0.9023           99.0977         0
     2       20.8976 78.2098        0.8927      2     19.4038           79.8115      0.7847
     3       27.9535 71.5108        0.5357      3     28.7806           70.3023      0.9171
     4       32.3532 67.2655        0.3813      4     35.1268           63.8984      0.9748
     5       35.2243 64.4418        0.3339      5     39.2736           59.7317      0.9948
     6       37.1530 62.5081        0.3389      6     42.0577           56.9427      0.9996
     7       38.4700 61.1607        0.3692      7     43.9185           55.0843      0.9972
     8       39.3753 60.2133        0.4114      8     45.1405           53.8678      0.9917
  Source: calculated by author.

In short, based on the empirical results above, our principal conclusions are as follows:

Firstly, we find that the increased money supply has played a limited role in maintaining a
steady growth rate in China since the late 1990s. In contrast, the impact of M1 shocks on real
GDP is larger than the impact of M2 shocks. The 23.9 per cent variation in output can be
explained by M1 shocks, whereas only a 2.46 per cent fluctuation in output can be explained
by M2 shocks. This finding implies that an appropriate and adequate growth rate in money
supply selected by the PBC is vitally important in achieving the policy objective. It is
difficult to maintain a steady GDP growth rate with decelerating growth in the money supply
in conducting expansionary monetary policy during a recession. The slowing down in the
growth rate of money supply may account for the fact that China’s GDP growth slid steadily
for four successive years from 1996 to 1999. Empirically our results are consistent with the
evidence found by Zhao(2003).

Secondly, our empirical results indicate the emergence of a weak and short-lived relationship
between the increase in monetary supply and the change in price index during the 1993 -
2004 period. Less than 1 per cent of variation in RPI can be explained by M1 shocks or M2
shocks. Therefore, we hold that deflation in China has been primarily driven by aggregate
supply factors in recent years. The second part of our empirical findings is again in line with
the evidence found in the research findings of Yu(2004a) and Fan(2003).


                                               9
4. Analysis of the Effect of Monetary Policy on Output and Price Index from the
Perspective of the Transition Economy

We attempt to evaluate the effect of monetary policy on output and the price index from the
perspective of the transition economy, and analyze the factors that affect the role of monetary
policy in the economic transitional process.

4.1 The Effect of Monetary Policy Shocks on Output Exists, but only in a Limited Sense

With regard to the impact of monetary policy on output, we hold that the methods by which
monetary policy is implemented are central to the effectiveness of the policy. Between 1996
and 1998, the broad money M2 grew at an average rate of 19.9 per cent and 17.6 per cent per
annum respectively. The GDP growth rates were 9.6 per cent, 8.8 per cent, 7.8 per cent,
respectively. Since the second half of 1998, while maintaining an accommodating monetary
policy, the government has shifted the emphasis of macroeconomic policy for revitalizing the
economy from monetary policy to fiscal policy. Meanwhile, the PBC adjusted the way by
which it conducts monetary policy, from purely quantitative expansion to cooperation with
proactive fiscal policy. The increased money supply financed enlarged government
expenditures in public works, resulting in increases in the growth rate of investment (Figure
6). After a fall to 5.1 per cent in 1999 from 61.8 per cent in 1993, the growth rate of total
investment in fixed assets ultimately started rising. The growth rate in GDP increased
gradually from 7.1 per cent in 1999. Between 1999 and 2002 China’s economic growth rate
stopped sliding and operated steadily between 7-8 per cent.

In fact, one of the most important features of the Chinese economy nowadays is the unstable
growth of investment. It is easy to see that over the past twenty years, growth rates of the
economy and that of investment have displayed a very close correlation, albeit with a lag of a
year or so (Figure 6). For example, in 1993, investment growth rate jumped to 61.8 per cent,
the highest ever since 1949, while the economic growth rate reached 13.1 per cent. In 1999
the investment growth rate edged up 5.1 per cent, while economic growth moved down to 7.2
per cent.
  Figure 6 Correlation of GDP Growth with Fixed Assets Investment Growth in China
                                                                                               80

                                                              61.8                             60

                                                                                               40
                            16
                                                                                               20
                            14                                                    5.1
                                                              13.1
                                                                                               0
                            12
                                                -7.2
                            10                                                                 -20

                            8
                            6                                                      7.2
                                          4.2
                            4
                                  1986 1988 1990 1992 1994 1996 1998 2000 2002

                                 the growth rate of GDP(left scale)
                                 the growth rate of total investment in fixed assets(right scale)


                         Sources: China Statistical Yearbook, 2004.

                                                             10
In recent years, efforts were made to promote the reform of the Chinese monetary policy
mechanism and the monetary policy instruments. The transition of monetary policy from
direct to indirect control has basically been realized. Open market operations have become
the principal tools for the routine operation of monetary policy. However, due to certain
characteristics of the transition economy, the indirect transmission mechanism which is
conducted by monetary instruments → operational targets → intermediate targets → ultimate
goals, failed to continually work well; another indirect transmission system which is linked
by the central bank → financial institutions (via financial markets) → enterprises and
households → national income also failed to operate successfully.

In the eras of inflation (before 1996, in 2003), the most direct and effective way to control
investment demand was to control the scale of credit. Contractionary monetary policy, by
means of banks’ credit restraint, was very effective. In the eras of disinflation (1996-2002),
the increased money supply was used to finance government expenditures in public works,
resulting in increases in the growth rate of investment. During the economic transition
process, due to the lack of an effective financial market, arising from the fact that there were
no profit-maximizing banks, expansionary monetary policy via pure quantitative expansion
or by interest rate reductions both proved ineffective.

As Yu(2004b) pointed out, firstly, Chinese banks have accumulated a huge amount of
non-performing loans (NPLs), while Chinese banks have increasingly commercialized their
operations and the government has tightened the regulation on and supervision over banks’
lending activities. Heavy penalties have been imposed on bank managers who are responsible
for the new NPLs. As a result, commercial banks have become increasingly concerned about
loan safety rather than profitability, thus they are very reluctant to lend to most enterprises,
fearing loan risks which may lead to NPLs. Secondly, many enterprises do not want to
borrow from the bank either. As a result of a high debt-equity ratio in contemporary times,
even enterprises with a relatively good credit rating are cautious of further borrowing and do
not wish to worsen their debt burden, and evidently those enterprises with low credit rating
cannot obtain loans anyway.

During the second half of 1998, owing to the slow-down of economic growth, the PBC
ceased using pure quantitative expansion as means of expansionary monetary policy. The
increased money supply was used to finance an increase in government expenditure in public
works, resulting in increases in the growth rate of investment. Meanwhile, a series of
macroeconomic policies aimed at encouraging banks’ to offer loans to medium (small) -sized
enterprises were pursued in succession. In the absence of a change in the way monetary
policy was implemented, the increase in money supply would have failed to affect investment
behavior.

The Chinese economy has managed to find a way out of deflation and entered into a new
stage of economic growth in late 2002 and early 2003. Expansionary monetary policy was
responsible for this macroeconomic outcome. In 2003, GDP grew by 10 per cent, the highest
growth rate since 1997. The inflation rate in 2003 measured by CPI was 1.2 per cent, which

                                               11
signaled that the Chinese economy had finally shaken off the deflation that had plagued the
economy since the second half of 1997. However, in China, the banking system is not
independent, as the regional banks have to take directives to give grants not only from the
PBC but also from the strong local governments. In late 2003, this, together with the need for
local economic development, the requirements to improve local budget balances and, most
importantly, the low and tightly regulated interest rates coupled with low land prices and the
low labor wage rate, quickly led to an investment hunger among local governments and
private enterprises. Owing mostly to this “abnormally” high growth rate of fixed asset
investment, total investment in 2003 increased by 26.7 per cent over the previous year.

Correspondingly, the government decided to tighten the loop. Some tight policy
measurements, such as the decrease in money supply and the rise of interest rates, failed to
cool down the demand for investment again. To supplement the measurements to control
growth of credit and investment, administrative orders were issued to restrain construction
works and investment in real estate and property development. Therefore, in the transition
process, the effectiveness of monetary policy becomes increasingly dependent on ways by
which policy is implemented---that is, other than quantitative controls, by which means and
to what extent can we affect the investment.

Moreover, as for the effect of monetary policy on output, the second factor that we should
pay attention to is the structural changes of ownership formations in the Chinese economy.
Since the reform and opening up of China, the non-SOEs have developed remarkably.
Currently, the non-state-owned economy accounts for more than 70 per cent of added value
in China's industry, and more than 80 per cent of the economic growth rate. Additionally,
non-SOEs provide more than 95 per cent of new jobs. In contrast to SOEs, the non-SOEs are
more sensitive to interest rate changes. From May 1, 1996 to June 10, 1999, in the seven
rounds of rate cuts, the average deposits interest rates decreased by 5.73 per cent and that of
loans dropped by 6.47 per cent, thus the interest spread was narrowed to 0.71 per cent. As a
result, the growth rate of investment by the non-SOEs started rising in 2000. According to the
Table 5, we can see that the total investment rate by private enterprises rose to 12.2 per cent
in 2000 from 6.8 per cent in 1997, 15.3 per cent in 2001, and 20.1 per cent in 2002. The
increase in investment growth by the non-SOEs has greatly contributed to the recent steady
economic growth.

                 Table 5 Total Investment in Fixed Assets by Ownership
                   State-owned Collective-owned       Individuals    Other Types of
           Total
                      Units            Units           Economy         Ownership
   1990     2.4         6.3            -7.1                -3
   1991    23.9        24.4            31.7               18.1
   1992    44.4        48.1            94.8                3.3
   1993    61.8        44.1            70.5               20.8
   1994    30.4        21.3            19.1               33.5            99.4
   1995    17.5        13.3            19.2               29.9            21.3
   1996    14.8        10.6            11.3               25.4            23.7

                                              12
   1997     8.8          9                5.5                 6.8                13
   1998    13.9        17.4               8.9                 9.2               11.6
   1999     5.1         3.8               3.5                12.1                5.3
   2000    10.3         3.5              10.7                12.2               28.5
   2001     13          6.7               9.9                15.3               28.9
   2002    16.9         7.2              13.4                20.1               36.2
   2003    27.7        14.7              33.8                18.4               50.0
   Sources: China Statistical Yearbook, 2004.

In fact, if we go a step further, moving to the arguments about the efficacy of China's
monetary policy, they further reveal the conflicts between the market-oriented reform of the
macro-control mechanism of monetary policy and the traditional economic system. It implies
that China needs not only to utilize market-based instruments of monetary policy actively, but
also to create the institutional conditions and the microeconomic foundations vigorously so
that these instruments can function effectively in order to establish the market-based
transmission mechanism of monetary policy (Wang 2001).

In western countries, market economies are highly developed and their monetary policies are
faced mainly with problems relating to aggregates. But in China, which is undergoing the
transition process from a planned economy to a market economy, the monetary policy is
faced with not only the problems of aggregate, but also the problems inherent in the
economic system and economic structure.

4.2 Emergence of a Weak Relationship between the Change in Monetary Supply and the
Change in Price Index

Since 1998, China's price level had been continuously following a downward trend, though
the annual growth rate of M2 reached 17.6 per cent, an average rate of approximately 14.2
per cent per annum can be calculated (1998-2002) (Table 1). With regards to a weak
relationship between the change in monetary supply and the change in price index, we hold
that disinflation in China has been primarily driven by aggregate supply factors in recent
years.

As mentioned above, the main features of Chinese economic growth are investment-driven.
The most important difference between an increase in current consumption and current
investment is that the latter will lead to an increase in supply and hence a reduction in
inflationary pressure in the future. Meanwhile, the Chinese economy has experienced
structural changes in the transition period. The structural changes, market-oriented reforms,
and huge amount of FDI flushed into China have improved the level of potential output,
resulting in a faster increase in aggregate supply than the expansion in aggregate demand
(shown in Figure 7).

For example, China’s increasing receptivity to FDI is fostering significant
technology-induced efficiency gains, which moderate upward pressure on prices. Besides that,

                                              13
productivity and technological changes lead to an expansion of aggregate supply relative to
aggregate demand thereby leading to lower prices while economic activity expanded.

         Figure 7 Relationship Between price level and AD-AS changes in China

                        P




                                                                Y
To summarize, because the driving force for high growth in China was investment, together
with market-oriented reforms and the inflow of FDI, the growth rate of aggregate supply is
exceeding that of aggregate demand, and therefore, prices have been declining. As a result,
increases in aggregate demand stemming from expansionary monetary policy since the late
1990s did not necessary lead to high inflation (Yu 2004a; Fan 2003; Woo 2003).

5. Conclusions and Policy Implications

The aim of this paper is to evaluate the effectiveness of monetary policy in China. We
estimate a VAR (2) model on quarterly data for the period 1993:1- 2004:4 to investigate how
the change in money supply affected GDP and RPI. We can derive two major policy
implications from the model. One is that the way to implement monetary policy is of great
importance to the effectiveness of policy in the transition period. It has been proven that
administrative measures were necessary, and may even be more effective than monetary
policy instruments in China. In addition, it is imperative to pay special attention to
maintaining an appropriate and adequate growth rate in the money supply in order to advance
the sustained, rapid and sound development of the national economy. Another important
conclusion is that monetary policy may not be as effective in supporting price stabilization.
Deflation in China has been primarily driven by aggregate supply factors in recent years.

In addition, given that China’s economic reform still has a long way to go, the conflicts
between the market-oriented reform of macro-control mechanisms of monetary policy and
the traditional economic system may weaken the effectiveness of monetary policy. On one
hand, in order to achieve policy objectives, administrative orders may be more effective than
market-based instruments of monetary policy in China. On the other hand, it implies that
China needs not only to utilize market-based instruments of monetary policy actively, but
also to actively create the institutional conditions and the microeconomic foundations so that
these instruments can function effectively in order to establish the market-based transmission
mechanism of monetary policy.

                                              14
Finally, as yet the official interest rate cannot act as intermediate target of monetary policy in
China. Due to the imperfections of financial markets and interest rate control, interest rates
can neither reflect demand and supply for funds, nor vary with market demand and supply. It
is in fact a rather exogenous variable. Meanwhile, the interest rate is less relevant to the state
of economic activities. Currently, SOEs and state-wholly-owned commercial banks
predominate China's economy. As far property right features are concerned, they are not
genuine business firms and genuine commercial banks - they aim at dual targets of operation
and pay attention to both economic benefits and social benefits, so their lending and
borrowing behaviors are not completely promoted by market forces. From this point of view,
the reform of SOEs and state-wholly-owned commercial banks is a prerequisite of
marketization of interest rates (Wang 2001).




                                                15
References
Ai H. D. and N. Fang (2002), “Empirical Analysis of the Velocity of Money in China”, World
  Economy, 8, 53-59.
Fan G. (2003), “Deflation, Effective disinflation, and Economic Fluctuation”, Economic
  Research, 7,3-9.
Fei P. and P. Xiong (2003), “Seeping Effect in the Channel of Circulation of Money
  Increment in China”, Economic Research, 8, 21-27.
Liu W., S. R. Li, and S. Y. Li (2002), “Monetary Expansion, Economic Growth and System
  Innovation of Capital Market”, Economic Research, 1,27-32.
Oppers, Erik, 1997, “Macroeconomic Cycles in China”, IMF Working Paper: WP/97/135.
Wang Ruifang, 2001, “Effectiveness of Monetary Policy in Post-Reform China: Some
  Empirical Evidence”, A Presentation at the 3rd International Conference on the Chinese
  Economy, Organised by CERDI-IDREC in Clermont-Ferrand, France.
Wang S. P. and Z. N. Li (2004), “The Cointegration Analysis of China’s Monetary Demand
  and the Monetary Policy Proposals”, Economic Research, 7, 9-17.
Wang Y. and M. Ma (2001), “Reforming the Transmission Mechanism of Monetary Policy
  in China” , 6, China & World Economy, http://www.iwep.org.cn/wec/english/e46.htm.
Woo, W. T., 2003, “The Travails of Current Macroeconomic and Exchanger Rate
  Management in China: The Complications of Switching to a New Growth Engine”,
  http://ideas.repec.org/p/wpa/wuwpdc/0310001.html.
Xie P. (2004), “Analysis of Chinese Monetary Policy: 1998-2002”, Financial Research, 8,
  1-20.
Yu Y. D. (2004a), “How to Analyze the Current Macroeconomic Performance in China”,
  http://www.iwep.org.cn, 9 August.
Yu Y. D. (2004b), “China’s Macroeconomic Situation Since 2003”, China & World Economy,
  12, 3-20, http://www.iwep.org.cn/wec/english/e62.htm.
Zhou S. P., J. Wang and H. B. Wang (1998), “The Effect of Macroeconomic Policy
  Cooperation on the “soft-landing””, Economic Research, 2, 20-29.
Zhao X. D., F. Chen, and T. M. Gao (2002), “Empirical Analysis of the Effect of Monetary
  Policy Instruments in China”, Quantitative Economy & Technical Economy Research, 7,
  103-106.




                                           16

				
DOCUMENT INFO
Shared By:
Categories:
Tags:
Stats:
views:0
posted:10/29/2012
language:English
pages:17