Government Expenditure and Economic Growth by liaoqinmei

VIEWS: 29 PAGES: 13

									       Government Expenditure and Economic Growth:
             Evidence from China 1952-2000
Abstract
Building on Ram‟s (1986) growth accounting model, attempts are made to untangle
the mysterious relationship between government expenditure and economic growth
using China‟s time series data for the period 1952-2000. Based on the understanding
of the theoretical model, we justify that it is generally appropriate for any economy
including China, and our empirical results confirm Ram‟s general findings that
government size (expenditure) has an overall positive impact on economic growth.
However, there are critiques and comments both to the theoretical framework and to
the empirical application from an econometric perspective, which leaves us with the
precaution to interpret the results and tasks to further improve the static model. (This
paper together with its follow-ups build up part of the author‟s PhD thesis, contact the
author for more details)

I. INTRODUCTION

The relationship between government expenditure and economic growth has long
been a subject of analysis and debate. The analysis and debate are essentially about
the role of government in the national economic growth. There are mainly two views
regarding the relationship. On the one hand, within Keynesian macroeconomic
framework, the standard effective demand theory suggests that government
expenditure, seen as an exogenous factor, can be used as an important policy
instrument to stimulate economic growth. On the other hand, the „law of the
expanding state role‟, postulated by Adolph Wagner in 1890, suggests that
government expenditure is an endogenous factor or an outcome, not a cause of
economic development.

There have been abundant empirical studies within the former framework, however,
they have produced mixed or contradictory results as to the role of government
expenditure in promoting economic growth. For example, Landau (1983), based on a
cross-country study for 96 countries, concluded with a significantly negative relation
between the share of government consumption in GDP and the growth rate of per
capita GDP; Kormendi and Meguire (1985) found no significant relation between
average growth rates of real GDP and average growth rates or levels of the share of
government consumption in GDP, and Grier and Tullock (1989) extended the
Kormendi and Meguire form of analysis to 115 countries, and showed a significantly
negative relationship between the growth rate of real GDP and the ratio of
government consumption to GDP. Different to them, Ram (1986) applied
international comparable data of Summer and Heston (1984) for 115 countries to his
theoretical model and found that the overall effect of government size on economic
growth is significantly positive.

These previous empirical studies are primarily based on cross-sectional analysis for
developed countries, and most of them lack rigorous theoretical model. In this study
we examine the expenditure-growth nexus by using a set of time series data from
Chinese economy in the period 1952-2000 and resting our theoretical considerations
on Ram‟s (1986) model. The reason for focusing this study on China is also that it
has achieved impressive growth rate of gross domestic product, especially since the
start of its market reforms in 1978, with an annual growth rate of GDP of 9.5 percent.
Therefore, it would be interesting to know how government expenditure, among other
determinants, has contributed to economic growth in China for the period 1952-2000
as a whole.

The rest of the paper is organized as follows. Section II discusses the theoretical
framework used and summarizes Ram‟s empirical results. In Section III, we propose
some arguments as to Ram‟ model. Application of the model to China is presented in
Section IV. Some concluding remarks are offered in Section V.

II. RAM’S THEORETICAL FRAMEWORK AND EMPIRICAL RESULTS

One of the most influential and earliest researches on the relationship between
government expenditure and economic growth is by Ram (1986), who develops a
two-sector production theoretical framework and employs international comparable
data on output, investment, and government size, from Summers and Heston (1984)
for a large sample of 115 countries for the period from 1960 to 1980. Like most of
previous studies, cross-section analysis has been carried out in his study. In addition,
a time-series analysis has also been made with an attempt to reinforce the former.
The main results obtained by him are that (1) government size has a positive overall
impact on economic growth and (2) it also shows a positive externality effect on
economic development.

2.1 Theoretical model

The brief development of Ram‟s theoretical framework is summarized below. The
economy is assumed to have two sectors, government sector and private sector. Each
sector‟s output is a function of the factors allocated to the sector. In addition, the
output of the private sector depends on the level of output produced by the
government sector. This formulation represents the beneficial effects of government
sector on the private sector. The production functions for the two sectors are then
written as
                G = G(Lg, Kg),                                          (1.1)
                P = P(Lp, Kp, G),                                       (1.2)
Where Lg, Lp, Kg, Kp are sectoral inputs, and the output of government sector G enters
into the production function and thus affects output of the private sector. The total
factor inputs (L and K) are then given as
                L = Lg + Lp                                             (1.3)
                K = Kg + Kp                                             (1.4)
Suppose marginal factor productivity in the two sectors (i.e., their partial derivatives
GL, PL, GK and PK) differs and the relative productivity for both factors are identical.
This assumption can be denoted as
        GL/ PL= GK /PK=1+δ                                              (1.5)
So the sign of δ indicates which sector has the higher marginal factor productivity.

Differentiating equations (1.1) and (1.2), we get
               
        G  GL L g  GK I g                                          (1.6)
                                         
            P  PL L p  PK I p  PG G                                                     (1.7)
                                                                                      
where Ig and Ip are respective sectoral gross investment, L g and L p are sectoral
changes in labour force, and PG represents the marginal externality effect of the
government sector on the private output.

Using Y to denote the total output in the economy which is the sum of outputs in the
two sectors, and Y = G + P, it follows that
                         
       Y  G P                                                    (1.8)
Assuming the elasticity PG(G/P), is a constant parameter and manipulating the above
equations, the following equation for aggregate output growth rate can be
approximately obtained:1

                                                      G           
            Y / Y   ( I / Y )   ( L/ L)  (       ) (G/ G)   (G/ G)                          (R5)
                                               1       Y

For easy comparison, this equation is marked by (R5) to reflect the fact that it is the
same as equation (5) in Ram (1986a). In this equation, α is the marginal product of K
in the private sector, β is the elasticity of private output with respect to labour force
L,2 θ is the elasticity of private output with respect to G, and it equals PG(G/P) being
assumed as a constant parameter, and variable I is the total investment. Hence, θ
reflects the externality effect of government output on the private sector, and thus on
the economy as a whole.

                                               
A special case of (R5) is when                     equals θ, then (R5) becomes
                                           1 
                                                 
            Y / Y   ( I / Y )   ( L/ L)   (G/ G )                                              (R6)

Here then, θ not only gives the externality effect of government, but also the total
                       
effect given                    .
                   1 

In contrast to assuming PG(G/P) constant, if we assume PG is constant, then (R5) can
be rearranged and rewritten as

                                                        G 
             Y / Y   ( I / Y )   ( L/ L)  (      PG ) (G/ G)                                   (R7)
                                                1        Y

Comparing equations (R5)-(R7), according to Ram, yields the following points: (1)
                  G 
the coefficient of (G/ G ) in (R5) is smaller than that in (R7); (2) the overall effect
                  Y
of government on the private sector can be directly obtained from the estimation of
               G 
coefficient of (G/ G ) in (R7), while, unlike in (R5), neither the externality effect
               Y

1
    The approximation only applies to the parameter βwhose original value is given in Section 1.3.
2
    Again, the interpretation of the coefficientβis further discussed in section 1.3.
nor the factor productivity differential can be obtained in (R7); and (3) collinearity
          
                    G 
between G/ G and (G/ G ) might make regression coefficients less significant or
                    Y
even insignificant even if both variables are important.

2.2 Summary of Ram’s empirical results

Apart from equations (R5)-(R7) which are the basis of Ram‟s empirical work, he also
introduces and includes the equation below in his empirical analysis.
                         
       Y   ( I / Y )   L  (G / Y )                                     (R8)
His motivation for inclusion of this equation is to facilitate comparison between his
study with those by Rubinson (1977) and Landau (1983) who use the ratio of
government output to gross domestic product, G/Y, as an independent variable in their
model estimation.

Ram‟s cross-section results have shown that (1) the estimated coefficient of
G 
  (G/ G ) in equation (R5) is not significantly different from zero at any reasonable
Y
                                                       
significant level; (2) The estimated coefficients of G/ G are positive and statistically
significant at least at the 5 percent level, which can fairly lead to the conclusion that
the externality effect of the government on the economic growth is positive for
periods under study; (3) The overall effect of government size on economic growth,
                                                                 G 
represented by significant estimates for the coefficient of (G/ G ) in equation (R7),
                                                                 Y
is positive, and maybe very large; (4) the estimate for coefficient of G/Y in equation
(R8) is negative in every regression and, while not significant in the period 1960-70,
is statistically significant at least at the 5 percent level for period 1970-80 (Ram, p.196,
Table 1).

Although there are only 20 observations for each country, his time series evidence
based on equation (R6) and (R7) has revealed a broad judgment over the
government‟s effect on economic growth. Out of 115 cases, the externality effect of
government is positive in 100 cases, and the overall effect is positive in 98 cases.
Furthermore, over 55 to 60 percent of the above positive effects (both externality
effect and overall effect) are significant. Among 70 countries where regressions are
significant at least at the 10 percent level, the externality effect is positive in 66 cases
and the overall effect is positive in 65 cases, and nearly three-quarters of the
coefficients are significant (Ram, pp. 200-201, Table 3 and 4).

Resting on the above results from both cross-section and time-series analysis, Ram
finds it difficult not to conclude that government size generally affects economic
growth and performance in a favorable manner, largely through a positive externality
effect.

III. COMMENTS WITH RESPECT TO RAM’S FRAMEWORK

Ram‟s theoretical framework was developed under a few assumptions, one of which,
denoted by equation (1.5), is that the marginal product of each factor input in the
government sector bears a proportional relation to its counterpart in the private sector,
and the constant of proportionality is the same for both factor inputs. However, this
does not necessarily hold. In another words, the difference for the two sectors in the
marginal labour product is not necessarily equal to the difference in the marginal
capital product. And they can even take different signs. For example, in an economy
where there were very limited capital resources but abundant labour force available to
the private sector, while the private sector may have a relatively high marginal capital
productivity, it may have a lower marginal labour productivity compared with the
government sector. Although this is an assumption which Ram does not justify, the
theoretical framework manifested by equations (R5)-(R7) would not have been
obtained without it.

Another assumption Ram made was relating to the constancy of either θor PG.
However, it is not clear whether they can be regarded as constant over time.
Furthermore, the marginal product of the government sector, PG, may not be constant
due to possible diminishing returns to government size input in the private sector,
which is just PG itself.

The interpretation by Ram for the coefficient βis the elasticity of private output with
                                                                                 L
respect to labour force L. Yet it is not appropriate in a real sense since   PL .
                                                                                 Y
Only when the government (private) sector is relatively small (large) in the economy
can we approximately interpret β as the elasticity of private output with respect to
labour input.

Ram‟s framework ignores technical progress in the production function of both the
government sector and the private sector, as criticized by Dowrick (1993). Ram‟s
omission of technology growth leads to the absence of the term G/Y in his model
specification. By considering the technological progress implied in the production
function, Dowrick (1993) developed a general model which accommodates both the
Landau and Ram specifications.
In addition to the above comments, here we can further work out the interpretation of
                
coefficient (      +PG ) in (R7) which is not given by Ram. First consider the effect
              1 
of a marginal increase in capital in the government sector on total output in the
economy. The total increment to GDP brought about by such an increase would be
GK+PGGK. With constraints for production factors, an increase in capital in the
government sector implies a corresponding decrease of capital input in the private
sector, which reduces private sector output by PK. Then the final change to GDP by
such an increase is (GK+PGGK - PK). The difference ratio between the change to GDP
                                                            G  PG GK  PK
relative to the change to the government sector‟s output is K                  , which,
                                                                   GK
                                                                        
by simple manipulation, equals to (       +PG ). So, the coefficient (      +PG )
                                    1                               1 
measures the difference between the marginal contribution to GDP of production
factors in the government sector relative to the marginal contribution of these factors
to the government sector‟s output.
IV. APPLICATION OF RAM’S MODEL TO CHINA

Ram‟s model was developed based on a two-sector economy. Since any economy can
be regarded as a two-sector economy irrespective of the size of each sector, it does not
restrict itself to economies with different systems, namely market economy, central
planned economy or mixed economy although no command economies had been
included in her empirical analysis. Non-inclusion of command economies was mainly
due to data inavailability for these command economies of that time. So application
of Ram‟s model to Chinese economy justifies.

All the data employed by the empirical analysis in this study are obtained from the
Statistical Yearbook of China (various issues) and are annual data over the period of
1952-2000. And they are transformed to real terms and in logarithmic form for easy
handle statistically.

Based on Ram‟s theoretical framework, we apply China‟s time series data to estimate
each of equations (5), (6) and (7). Like Ram‟s study, we also apply the data to
equation (8), which enables us to compare the results with those from Landau (1983).

Table 1 shows the main results from the regressions which correspond to equations
(R5) to (R8), and in each regression a constant term and a random stochastic
disturbance term with usually assumed properties are included.

         Table 1: Government expenditure and economic growth in China,
                               under Ram model

Explanatory Variables               Eq. (R5)          Eq. (R6)           Eq. (R7)          Eq. (R8)

INVESTR                             0.386***          0.405***           0.506***          0.987***
                                    (2.743)           (3.066)            (3.781)           (4.754)
DLPO                                3.338***          3.387***           3.791***          7.560***
                                    (3.286)           (3.386)            (3.672)           (5.417)
DLGEGER                             -0.284                               1.055***
                                    (-0.438)                             (7.352)
DLGE                                0.403**           0.321***
                                    (2.116)           (7.995)
GER                                                                                        0.129
                                                                                           (0.8651)
Adj. R-sq.                          0.7522         0.7568                0.7321            0.4058
                                          Diagnostic Tests
Serial correlation                  0.064          0.055                 0.032             0.346
Functional form                     0.052          0.070                 0.084             0.003
Normality                           0.476          0.536                 0.817             0.767
Heteroscedasticity                  0.391          0.442                 0.281             0.778

Notes: with t-statistics in parentheses; dependent variable: DLGDP; *** statistically significant at the
1% level; ** statistically significant at the 5% level; * statistically significant at the 10%; results of
diagnostic tests are shown as p-values.
In the regressions above, we use DLPO, the population growth rate to approximate
the growth rate of labour input in the production functions. Although not necessarily
a good proxy in some cases, employing DLPO has been justified by the relative
reliability and availability of demographic data with reliable time-series data on
labour force being seldom found, particularly for developing countries. The variable
INVESTR is the investment ratio which is approximated by the ratio of total fixed
capital formation to GDP. DLGE is the growth rate of government expenditure. The
variable DLGEGER is defined as the product of DLGE and GER- the ratio of
government expenditure to GDP.

In comparison to Table 1 in Ram, our results show that the coefficient of DLGEGER
in regression (R5) is insignificant at any reasonable significant level, which is a
                        G 
similar case to that of (G/ G ) in Ram‟s equation 5. Therefore, by Ram‟s argument,
                        Y
we can drop this variable, thus turning regression (R5) to (R6). 3 In Ram‟s equation 6,
                                                                                     
the same sign of δ and θ is implied, with θ, the coefficient of G , giving the
externality effect of government size on economic growth and δ reflecting the factor
productivity differential between public and private sectors. From the estimation of
regression (R6) in our analysis, we may conclude that, for China from 1952 to 2000,
the externality effect of government expenditure on economic growth on average is
positive (represented by the coefficient of DLGE, that is 0.321), and that the factor
productivity of the government sector is higher than the private sector.

The coefficient of DLGEGER in our regression (R7) indicates the total effect of
government expenditure on economic growth. Its point estimate is significantly
positive even at the 1% significance level, which further consolidates the conclusion
reached by Ram that the total impact of government expenditure on economic
development is significantly positive. However, unlike Ram‟s results, the estimate of
the coefficient of GER in our regression (R8) is positive, but statistically insignificant.

It is worth mentioning that for regressions (R5) – (R8) in table 1, the diagnostic tests
for serial correlation in (R5) and (R6) are marginal at the 5% significant level and that
in (R7) indicates rejection of the null hypothesis of no serial correlation. This
problem leaves us with the precaution to interpret the preliminary results and the tasks
to further improve the current static model. In addition, the diagnostic tests have also
shown a problem with functional form, especially for regression (R8).

Since our results are obtained based on time-series data, it would be more convincing
and appropriate to compare them directly with the evidence from Ram‟s time series
data although his data have a small number of observations.

With 115 countries included, his time series evidence, based on equations (R6) and
(R7), has revealed that the externality effect of government is positive in 100 out of
115 cases and the overall effect is positive in 98 cases. Furthermore, over 55 to 60
percent of the above positive effects (both externality effect and overall effect) are

                                                                                         
                                                                           G 
3
    However, Rao (1989) argues that, due to possible correlation between     (G/ G ) and G/ G , the statistically
                                                                           Y
insignificance of coefficient of any of them may not be sufficient to assume that its value is zero.
significant. Focusing on 70 countries where regressions are significant at least at the
10 percent level, he finds that the externality effect is positive in 66 cases and the
overall effect is positive in 65 cases, and nearly three-quarters of the coefficients are
significant.

What is, on average, the degree of the impact of government on economic growth for
these countries as a whole then? Based on his results, we can calculate the mean for
the 70 estimated coefficients of government variables in both regressions 6 and 7 for
these 70 individual countries. The averaged coefficient for θ in regression 6 for these
                                                     G 
70 countries is 0.254, and the average coefficient of (G/ G ) , i.e., the average
                                                     Y
                            
estimated value of (       +PG ), is 1.3054. This means that, on average, governments
                      1 
have exerted positively both an externality effect and an overall effect on economic
development. Comparing the average values with our results for China, we can notice
that they are quite close. Hence our results reinforce Ram‟s conclusion.
Our results above are obtained by using the government expenditure to approximate
government size, while Ram uses the government consumption to represent
government services. Table 2 reports the regression estimates of impact of
government consumption on economic growth as a result of replacing government
consumption by government expenditure in the regressions. Here, the variable DLGC
is defined as the growth rate of government consumption, GCR1, is the share of
government consumption in overall government expenditure and DLGCGCR1 is the
product of DLGC and GCR1.

         Table 2: Government consumption and economic growth in China,
                               under Ram model

Explanatory Variables                 Eq. (R5)           Eq. (R6)            Eq. (R7)           Eq. (R8)

INVESTR                               0.695***           0.685***            0.689***           0.959***
                                      (4.655)            (4.600)             (4.679)            (5.063)
DLPO                                  7.153***           7.184***            7.166***           8.003***
                                      (6.674)            (6.708)             (6.752)            (5.310)
DLGCGCR1                              5.483                                  3.069***
                                      (0.9798)                               (5.408)
DLGC                                  -0.320             0.399***
                                      (-0.4338)          (5.286)
GCR1                                                                                            0.899
                                                                                                (1.070)
Adj. R-sq.                            0.6332         0.6335                  0.6401             0.4111
                                            Diagnostic Tests
Serial correlation                    0.747          0.609                   0.697              0.566
Functional form                       0.024          0.013                   0.025              0.035
Normality                             0.683          0.602                   0.629              0.721
Heteroscedasticity                    0.957          0.796                   0.965              0.805
Notes: as before, with t-statistics in parentheses; dependent variable: DLGDP;
*** statistically significant at the 1% level; ** statistically significant at the 5% level; * statistically
significant at the 10%; results of diagnostic tests are shown as p-values.
In equation (R5), the coefficients of DLGCGCR1 and DLGC are both statistically
insignificant, which may be the result of colinearity between the two variables. When
either DLGCGCR1 or DLGC is excluded from the regression, which becomes down
to regression (R6) and (R7) respectively, the remaining variable DLGC or
DLGCGCR1 is significant, as we can see from the above table. All the coefficients in
the regressions (R6) and (R7) are statistically significant at 1% level. Similarly to the
case where government expenditure is used to approximate government, for
regression (R8) in table 2, the government consumption ratio variable, GCR1, is not
significant even at the 25% level.

Diagnostic tests in Table 2 do not suggest the existence of residual serial correlation
like in Table 1, however, the problem of functional form remains as shown by the p-
values for functional form in Table 2. This casts doubt on the validity of the linear
model we employ.

Whichever variable, government expenditure or consumption, we apply to measure
the government sector, the empirical results take a similar pattern in terms of the
effect of the government size on economic development or growth. It is not difficult
to conclude, although tentatively and with precaution, that over the period 1952 to
2000, government expenditure has played an important role in promoting Chinese
economic growth, and in general, the factor productivity in the government sector is
higher than in the private sector.

Landau (1983) found a negative relationship between the share of government
consumption expenditure in GDP and the growth of per capita GDP for a cross
section of 96 countries over several periods between 1961 and 1976. In his growth
regression equation, apart from other explanatory variables included as control
variables such as level of per capita GDP, total investment in education and energy
consumption, he uses the share of government consumption expenditure in GDP to
represent the government in the economy. Despite many differences between his
study and the one undertaken here both in regression and type of data employed, a
direct comparison might be worthwhile.

While Landau‟s tentative and weak conclusion of a negative relationship between
government expenditure and GDP per capita growth has been drawn, there is an
inconformity when the bottom (defined by income running from 3-13% of U.S. per
capita GDP) half sub-sample of 48 countries is empirically analyzed. That is, the
negative relationship between government expenditure and GDP per capita growth
does not hold for this group of countries in his study. Further to this, positive
coefficients of the term G/Y are evidenced although without significance even at the
20% level (Landau, p. 790).

Similar to the inconsistency in Landau‟s analysis, our estimated coefficient of GER in
equation (R8), as mentioned earlier, is positive but insignificant even at the 30% level.
So if we take for granted the Landau-type regression, our results based on it weaken
his conclusion of the proposed negative relationship, and at the same time confirm his
finding about the low-income bottom half 48 countries because China would have
been included in his bottom half group of countries if its data had been available.
However, according to Ram‟s arguments, Landau‟s results are apparently based on a
misspecification of the growth regression equation simply because the variables used
                                                   
                                                                G 
to analyze the growth effect of government are G/ G and/or (G/ G ) , rather than
                                                                Y
G/Y (Ram, p.197). Further, his study lacks a rigorous theoretical framework on which
the empirical analysis rests. From a statistical perspective, I would also argue that
regression (R8) does not better capture the growth impact of government expenditure
than any other regression considered above for two reasons. First, the time series
variable GER, suggested by the unit root test (result not produced to save space,
available from the author upon request) , appears to be a nonstationary process of
integrated order one, while the others involved in regressions (5) – (8) are all
stationary. Regressing a stationary on a nonstationary process would invalidate the
regression itself thus makes regression (R8) unattractive. The second reason is easy
to see simply by comparing the adjusted R-squares of these four regressions. The
adjusted R-square for (R8) is the lowest while those for (R5), (R6), and (R7) are fairly
close to each other and are much bigger than that for (8). In another words, the share
of government expenditure (or consumption) in GDP contributes less than other two
variables, namely, government expenditure (or consumption) growth rate DLGE (or
DLGC) or the variable DLGEGER (or DLGCGCR1), in explaining the growth rate of
GDP.

Landau (1983) was the first to systematically study the relationship between
government expenditure and economic growth. However, there are some deficiencies
in his study as discussed above. Ram (1986) overcomes some of the shortcomings by
developing a more rigorous theoretical model but with a few assumptions that are
easily questioned on their validity. These early studies initiate my interest in
analyzing the relationship between the government expenditure and economic growth
in China. But they are based on static theoretical models and only on the aggregate
government expenditure level. Therefore, dynamic models and the effects of
government component expenditures can be accommodated in further study. In
addition, the development of modern econometric techniques makes it possible for
such an empirical analysis to be more comprehensive and in depth.

V. CONCLUDING REMARKS

This paper examines the relationship between government expenditure and economic
growth in China for the period of 1952 to 2000. The primary aim of the article was to
investigate empirically this relationship within the Keynesian framework. Based on
the theoretical model proposed by Ram (1986), one of the earliest and most influential
papers in this area, Chinese macroeconomic data are applied. The empirical results
coming through the application generally support Ram‟s findings that government
expenditure has an overall positive impact on economic growth and that it also shows
a positive externality effect on economic development. It is not surprising given that
Chinese economy is still in the early stages of development and government has
played a central role in the economy. Nevertheless, several points have been raised
relating to both Ram‟s framework and empirical analysis. This motivates further
improvement to the static model and more convincing results.
REFERENCES

Dowrick, S., 1993, “Government Consumption: Its Effects on Productivity, Growth
and Investment”, in The Growth of the Public Sector, edited by Gemmell, N., Ddward
Elgar, pp. 136-50

Grier, K. and G. Tullock, 1989, An empirical analysis of cross-national economic
growth, 1951-1980, Journal of Monetary Economics 24, 259-276

Kormendi, R.C., and Megurie, P.G., 1985, Macroeconomic determinants of growth:
cross-country evidence, Journal of Monetary Economics 16, 141-63

Landau, Daniel, 1983, Government expenditure and economic growth: a cross-
country study, Southern Economic Journal 49, 783-92.

Ram, R., 1986, Government size and economic growth: a new framework and some
evidence from cross-section and time-series data, The American Economic Review 76,
191-203

Robinson, R., 1977, “Dependency, Government Revenue, and Economic Growth,
1955-70”, Studies in Comparative International Development, Summer, 12, pp. 3-28.

Rao, V., 1989, “Government Size and Economic Growth: A New Framework and
Some Evidence from Cross-Section and Time-Series Data: Comment”, The American
Economic Review 79, 272-280
                         Curriculum Vitae – Xiangjie Wu

Personal Details
Name:                          Xiangjie Wu
        Gender:                Male
Date of Birth:         22 January 1965                                 Marital
Status: Married
Nationality:           Chinese
        State of Health:       Excellent
Email Address:         xiangjie_wu@hotmail.com                 Telephone:
        07789172898
Contact Address:       34 Linton House, 2A Wellington Road, Manchester M14 6EQ,
U. K.

Education
University                                         Course
              Date
Shandong University, China                  BSc in Applied Mathematics
      1982-1986
The University of Manchester, UK            M.A. in Economics
      1995-1996
The University of Manchester, UK            PhD in Economics
      2001-

Professional Experience
Deputy Director, Education Centre, Ministry of Finance, 1997-2001
     Responsible for the training of senior officials within China‟s public finance
         sector, namely, formulate annual training plan based on training assessment,
         implement the plan by delivering core courses, overseeing and evaluating the
         courses entrusted to other training institutions.
     Lectured on quantitative methods applied to decision-making and policy
         analysis in more than ten training courses
                    Translated working reports annually and served as an interpreter
         on several important international occasions including mainly overseas
         training courses
Section Chief, Department of Education, Ministry of Finance, 1990-1997
     Administered the joint self-taught programme with Ministry of Education
         designed for nationwide financial administrators who wish to pursue
         Bachelor degree in Finance, Taxation or Accounting
     Implemented A-10 subproject under TCC-Ⅲof the World Bank Project, which
         aims to enhance the training capacity of China‟s public finance sector
     Translated six World-Bank‟s country study reports into Chinese published by
         China Financial and Economic Publishing House, and served as an
         interpreter on five overseas training courses sponsored by World Bank
         Institute, and many other international exchange and cooperative
         programmes
     Held editorial position with the China Financial Education Quarterly
Lecturer, Department of Economic Management, Central University of Finance and
Economics, 1986-1990
     Taught Statistics, Economic Control Theory, and Basic Econometrics for 3rd
       and 4th year undergraduates
     Finished eight chapters in two published books

Research field
Econometric methods; Macroeconomics; Fiscal studies

								
To top