DETERMINANTS OF SWAZILAND MANUFACTURING OUTPUT – SUPPLY AND DEMAND APPROACHES A TECHNICAL PAPER SUBMITTED IN FULFILMENT OF MEFMI FELLOWSHIP PROGRAMME: BY: PATRICK NDZINISA 1 THE CENTRAL BANK OF SWAZILAND, RESEARCH DEPARTMENT MAY, 2007 1 Patrick Ndzinisa, Central Bank of Swaziland, Box 546 Mbabane, Swaziland, Tel: (268) 408 2204, Fax: (268) 404 0038, E-mail: [email protected]
37
Embed
DETERMINANTS OF SWAZILAND MANUFACTURING OUTPUT – …mefmi.org/mefmifellows/wp-content/uploads/2016/10/DETERMINANTS-OF...coefficient for foreign demand, proxied by manufactured good
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
DETERMINANTS OF SWAZILAND MANUFACTURING OUTPUT – SUPPLY AND
DEMAND APPROACHES
A TECHNICAL PAPER SUBMITTED IN FULFILMENT OF MEFMI FELLOWSHIP PROGRAMME:
BY:
PATRICK NDZINISA1
THE CENTRAL BANK OF SWAZILAND, RESEARCH
DEPARTMENT
MAY, 2007
1 Patrick Ndzinisa, Central Bank of Swaziland, Box 546 Mbabane, Swaziland, Tel: (268) 408 2204, Fax: (268) 404 0038, E-mail: [email protected]
ii
TABLE OF CONTENTS
Acknowledgements…………………………………………………..…………(iii)
Abstract…………………………………………………………………….….….(iv)
1. INTRODUCTION……………………………………………………………….…...1 2. OVERVIEW OF THE SWAZILAND ECONOMIC STRUCTURE……………...3
2.1 Manufacturing Sector……………………………………………………...3
2.2 Agricultural Sector…………………………………………………….……4
2.3 Other Sector…………………………………………………………..…….5 3. LITERATURE REVIEW…………………………………………….……………...7 4. DATA DESCRIPTION AND SOURCES………………………………………..11 5. METHODOLOGY…………………………………….………………………...…12
5.1 Specification of the equations……………………………………….…..12
5.2 Stationarity…………………………………………………………….…..13
5.3 Cointegration and Error Correction Model (ECM)…………………….14
I am greatly indebted to my mentor Theo Janse van Rensburg. He acted as a
furnace, helping to refine rough ideas and to focus the study. Without his
guidance and patience, this paper would not have been a success.
I am also grateful to MEFMI and the Central Bank of Swaziland, not only for
their financial assistance without which I would not have been able to attend
the relevant courses, but also for providing guidance and support. I am
particularly grateful for the generous time that the Central Bank of Swaziland
has granted me to complete this paper.
I am also indebted to the Research Staff of the Central Bank of Swaziland,
particularly Andreas Dlamini and Vusi Khumalo who assisted me with various
data issues - particularly during my visit to the South African National
Treasury in April 2007.
Lastly, but not least, I am thankful to my family for their understanding and
moral support throughout the study, but in particular during the period of the
MEFMI Customized Training Programme (CTP).
iv
ABSTRACT
The paper analyzed the determinants of the manufacturing output in Swaziland
from a supply and demand perspective. The results from the supply equation
confirmed the relative importance of both human- and physical capital. The
study also confirmed the strong linkage between agricultural- and
manufacturing output, indicating an elasticity of manufacturing production
with regard to agricultural output of +0.43
Meanwhile the demand equation also emphasized the importance of both
foreign and domestic demand in determining manufacturing output. The
coefficient for foreign demand, proxied by manufactured good exports,
signified the reliance of the Swazi economy, particularly manufacturing output,
on export for growth. It has been also observed that whilst FDI positively and
significantly affects manufacturing output it does so in the short-run. The
absence of long run FDI benefits is surprising, but may reflect unique
Swaziland circumstances, where FDI in the 1980’s was largely in response to
sanctions imposed on its neighbor, South Africa.
The paper thus argues that in the light of Swaziland’s manufacturing
dependence on the agricultural sector, the country should institute policy
measures to enlarge and diversify its economic base; there is a need to invest
in human capital in order to complement the increase in physical capital; the
country should pursue vigorous strategies through SIPA in order to promote
FDIs with long term benefits, hence manufacturing output growth. Economic
policy should encourage a competitive environment. In this regard sound
monetary and fiscal policies aimed at inflation stability will play an important
role.
1
DETERMINANTS OF SWAZILAND MANUFACTURING OUTPUT –
SUPPLY AND DEMAND APPROACHES
1. INTRODUCTION
Swaziland is a tiny land-locked country with small domestic markets. The
economy of Swaziland is heavily dependent on exports, largely based on
agricultural production and agro-processing manufacturing industries.
Although agricultural production constitutes less than 10 per cent of GDP, the
sector plays a significant role in determining the country’s economic growth
and development through its many forward- and backward linkages with other
sectors. Although other sectors (with 2004 GDP share in brackets) such as
manufacturing (34.8%), government services (16.6%), wholesale, retail,
hotels and restaurants (12.8%) also play an important role, variations in
overall output is largely related to changes in agricultural output and its
linkages to other sectors, in particular manufacturing. For instance, the recent
sluggishness in real economic growth is primarily attributed to reduced
agricultural production and the consequent slowdown in agro-processing in
the manufacturing sector. Given the revised estimated population growth rate
of 2 percent2, the slow economic growth has failed to achieve the desired
improvement in living standards. In fact the standard of living as measured by
per capita income has been falling since 19913. This is contrary to the
government’s objective of alleviating poverty in the country.
During the 1980s Swaziland recorded high economic growth rates, driven by
an influx of foreign direct investment (FDI) arising from sanctions imposed on
South Africa, which propelled the relocations of enterprises into Swaziland.
The high levels of foreign direct investment caused an economic upturn in the
2 Given government’s vision to reduce poverty rate by 50 percent by 2015 coupled with the assumed population growth rate of 2.75 percent, a minimum annual economic growth rate of 5 percent is required in order also to accommodate the resultant growth in labor force. 3 GDP per capita are published in the Central Bank of Swaziland’s quarterly review bulletin
2
manufacturing sector, which became the main growth engine, which in turn
encouraged rapid growth in supporting sectors such as construction as well
as generating additional revenue which permitted the consequent expansion
of government services. Apart from the inflows into the manufacturing sector,
the growth performance was also aided by more conventional external stimuli,
such as improved export prices for sugar, reinforced by the real depreciation
of the lilangeni.
However, since the 1990s the pace of economic growth has been falling
substantially when compared to rates achieved during the 1980s (see graph 1
in appendix B). GDP growth averaged 6.7 percent during the 1980s before
declining to an average of 3.2 percent in the 1990s and to around 2.6 percent
during the period 2000 to 2004. Given the economy’s dependence on agro-
processed goods for growth, the observed decline in the growth rate
symbolises a similar trend in the manufacturing and agricultural sectors.
These sectors have been adversely affected by, among other factors, the
slowdown in investment inflows due to stiff competition for foreign investment
in the region, particularly as the political and business environment improved
in the region. This was aggravated by fluctuations in international commodity
prices and the exchange rate, as well as unfavourable weather conditions and
the consequent closure of some major companies.
Chart 1 in appendix B compares Swaziland’s economic growth with selected
SADC countries. Whilst a majority of the SADC countries recorded economic
growth in excess of 3 percent, Swaziland’s economic growth was hovering
below that level. Countries like Botswana and Mozambique continue to
experience strong GDP growth due to the continued expansion of the mining
sector in Botswana and the significant expansion of the manufacturing and
telecommunications sectors in Mozambique.
3
Given the direct linkage between the manufacturing sector and the
agricultural sector in Swaziland, any growth strategy should take these
linkages into account. This study econometrically confirms the above
observations and also indicates that the elasticity of manufacturing production
with regard to agricultural output is +0.43. The main objective of the study
therefore is to identify factors that drive manufacturing output both from a
supply and demand perspective.
The rest of the paper is structured as follows. Section 2 of the paper gives an
overview of the structure of the Swazi economy with emphases on the
manufacturing- and agricultural sectors. In section 3 a literature review on the
determinants of economic growth with particularly emphasis on the
manufacturing sector is presented. Section 4 gives an overview of the data
and the sources, whilst the estimation methodology is discussed in section 5.
In section 6 the estimation results is presented, followed by conclusions and
policy implications in the final section of the paper.
2. OVERVIEW OF THE SWAZILAND ECONOMIC STRUCTURE
2.1 MANUFACTURING SECTOR
Most of the industries are foreign-owned, export-oriented and have strong
backward and forward linkages with agriculture. Manufacturing industries
range from small factories engaged in light industry to large ones endowed
with the latest technology and producing highly sophisticated goods which,
given the small size of the domestic market, are destined mainly for the
export market. The major export commodities produced are wood products,
soft drink concentrates, canned fruits, sugar, and mineral products. The
sector’s share of overall output declined slightly from 35.8 percent in 2000 to
34.8 percent in 2004. The sector’s contribution to employment creation has
declined in recent years as firms raised their capital intensity in the production
process. However, new medium and large scale enterprises, which tend to be
more labour intensive has ventured into yarn production and wheat milling, as
4
well as the manufacturing of items such as refrigerators and knitwear.
Although there has been a move towards manufacturing diversification, this
trend needs to be intensified in order to create significantly more formal sector
employment.
The major constraints to further development of the manufacturing sector
include the small size of the domestic market and the relatively narrow
resource base. Inadequate infrastructure also continues to present a major
barrier to output growth. In order to improve the investment climate,
Government has given high priority to enhancing the availability of physical
infrastructure in transport and communications, and also to increase the
number of skilled personnel in management and engineering. Government is
also pursuing a wage policy that matches real remuneration growth with
increases in labour productivity.
However, future prospects of the manufacturing sector depends on the
country’s ability to retain the preferential trade treatment under the introduced
African Growth and Opportunity Act/Trade and Development Act of 2000,
which introduced a new co-operation agreement between the United States
and eligible sub-Saharan countries. Not only does this legislation represents a
solid, meaningful, and significant opportunity, but is also likely to result in
substantial new trade and investment flows between the US and Africa.
2.2 AGRICULTURAL SECTOR
The agricultural sector, which consists of the traditional and modern sub-
sectors, plays a pivotal role in Swaziland’s economy. Despite the declining
volumes of output, the agricultural sector remains indispensable for the
majority of Swazi people who continue to derive their livelihood and income
by engaging in this sector’s activities, which include the production of maize,
cotton, sugar, fruits, vegetables, citrus and livestock. Moreover, the
agricultural sector plays an important role in providing substantial support to
5
the manufacturing sector, in terms of intermediate inputs required by the
Rgdp = the real domestic GDP a proxy for domestic demand
13
Rmanexp = the real manufacturing export a proxy for foreign demand
Reer = the real effective exchange rate of the lilangeni against trading
partners’ currencies a proxy for relative price
Rmanfimp = real manufacturing imports
Rpr = the real prime interest rate
Rfdi_m = real manufacturing foreign direct investment
ε = the error term.
The above equation specifications was adjusted to reflect the appropriate
single equation estimation techniques (see equation 5.3.1 below), using E-
Views software. The estimated equations were subjected to a battery of
econometric tests to ensure that an efficient and correctly specified model
was estimated.
5.2 STATIONARITY
Time series data tend to exhibit a stochastic or deterministic trend with the
mean, variance and covariance changing over time, and thereby rendering
the series non-stationary. The first step is thus testing for stationarity for each
individual data series before estimating the equations. The null hypothesis of
non-stationarity of the variables is tested against the alternative hypothesis of
stationarity using the augmented Dickey-Fuller (ADF)-(Dickey-Fuller (1979)).
The ADF test establishes the data generating process (DGP) from the
following:
Pure random walk
Random with drift/constant
Random walk with drift and time trend
The ADF test for stationary is based on the following equation:
14
tit
n
iitt YYTY
21 ………………………………………….5.2.1
Where Y is the series tested for stationarity, i ,,, are parameters and εt
is white noise. The ADT test can be summarised as below.
Table2 Summary of the ADF Tests
Model Hypotheses Test Statistic
Random walk with drift and time trend
tit
n
iitt YYTY
21 H0: ρ=0, H1: ρ<0 τ τ
H0: ρ= λ=0, H1: ρ ≠0 and λ ≠0, Φ3
ρ =0 and λ ≠0, ρ≠ 0 and λ =0
Random walk with drift
tit
n
iitt YYY
21 H0: ρ=0, H1: ρ<0 τμ
H0: ρ= μ=0, H1: ρ ≠0 and μ ≠0, Φ1
ρ =0 and μ ≠0, ρ≠ 0 and μ =0
Pure random walk
tit
n
iitt YYY
21 H0: ρ=0, H1: ρ<0 τ
5.3 COINTEGRATION AND ERROR CORRECTION MODEL (ECM)
The consequence of working with non-stationary data series in the estimation
process is that this may yield a meaningless or spurious result, that is, there
is danger of obtaining apparently significant regression results from unrelated
data.
If non-stationary time series are used in regression models there is a need to
test further for cointegration amongst the series. Testing whether y and x
(which have been both generated by I(1) time series processes) are
15
cointegrated an Engle and Granger (1987) two-step procedure is widely used.
Johansen (1988) also proposed a general framework for testing
cointegration4. The Engle and Granger test for cointegration is often referred
to as the residual based test which the study uses based on the assumption
that there is only one cointegrating vector in the equations. The presence of a
cointegrating relationship allows us not only to estimate the long-run
relationship but also to further analyze the short-run dynamics and how
adjustment to equilibrium is achieved. Therefore according to the Granger
representation theorem, the existence of a stable long-run relationship
between the variables enables us to estimate an ECM.
Error correction models are based on the behavioral assumption that two or
more time series exhibit an equilibrium relationship that determines both
short- and long-run behavior. ECMs are useful as they reconcile the short-
and long-run behavior of the variables by shedding light on the speed or rate
of adjustment towards long-run stable equilibrium. Two different econometric
methodologies can be used in the construction of the ECM namely the
generalized one-step procedure5, and Engle and Granger two-step
procedure. The single-equation generalized error correction model (GECM)
has proven to be both theoretically appealing and also statistically superior to
the two-step estimator by Engle and Granger (1987) in many cases (Suzana
De Boef, 2000); hence the study uses the one-step method.
Given the long-run relationship between Y and X as ttt XY 21 the
GECM is estimated in one step as follows:
4The Johansen test is preferred when there are more than two time series variables involved because it can determine the number of cointegrating vectors 5The one-step error correction model, which was popularized by Davidson et al. (1978), is a transformation of an autoregressive distributed lag (ADL) model (Banerjee et al. 1993). Unlike the Engle and Granger Two-Step Method where the error term incorporated in the ECM is derived from a long-run equation, the one-step method estimate the error correction coefficient directly from a single equation containing both long- and short-run variables, rather than deriving it from alternative specifications (Suzanna De Boef 2000)
Note: *, ** and *** denotes significance at 10%, 5% and 1% respectively for t values. However, for the F statistics i.e. for Φ3 and Φ1: *, ** and ***represent significance at 95%, 97.5% and 99% confidence intervals respectively. Otherwise not significant hence the null hypotheses that the variable has a unit root cannot be rejected.
The stationarity test results indicate that all the variables, with the exception
of the real prime are integrated of order one, i.e. I(1). The residual terms for
the demand and supply equations (i.e. RESID_RMANDEM and
RESID_RMANSUP) are stationary, implying a cointegrating relationship in
each of the two estimated equations.
18
6.2 DIAGNOSTIC TESTS RESULTS
The various diagnostic tests results are reflected in the tables below.
The diagnostics tests were conducted on the ECM specification of the supply
and demand equations and the results are depicted in tables 6.2.1 and 6.2.2
respectively. The results indicate that the errors of the two equations are
normally distributed and that there is no serial correlation in either of the two
equations at the 5 percent level. Similarly the ARCH test results indicate that
there is no presence of heteroscedasticity in the estimated equations. Finally
both the Chow Forecast test and the visual test (recursive coefficients)6
stability tests indicated that the parameters of the models are stable.
6.3 ESTIMATION RESULTS AND INTERPRETATIONS
The estimation results of the supply and demand equations are presented in
appendix C and D respectively. The findings of the study are in line with
earlier empirical studies.
With respect to the recent empirical literature, the supply equation results
confirm the relative importance of both human and physical capital in
determining manufacturing output in Swaziland. Assuming a constant returns
6 For the visual test results see appendix H and I
19
to scale, Cobb Douglas production specification7, the estimation results
indicate that the long run elasticities of manufacturing output with respect to
physical capital is +0.31 and with respect to labour is +0.69. This confirms the
labour intensity of manufacturing production. However, the strongest long-run
elasticity of +1.24 is with respect to educational spending8 - thereby reflecting
the importance of education in raising technological progress. The importance
of agro-based industries in Swaziland is reflected in the strong long run
elasticity of (+0.43) with respect to the agricultural sector.
In order to better understand the transmission mechanisms from the various
explanatory variables to manufacturing output, a series of impulse responses
was also conducted. The impulse response results indicate that although the
bulk of the adjustment to new long run equilibrium may take place over the
short term, if often takes many years for the adjustment to be fully completed
(See appendix F for the impulse response results).
The estimation results for manufacturing output from a demand perspective
are depicted in appendix D. The results is consistent with empirical findings
from other studies in that manufacturing output is determined by domestic
and foreign demand, relative prices, manufacturing imports, interest rates and
foreign direct investment. The respective coefficients were found to be
statistically significant and the signs of the coefficients were in line with
economic theory. However, the magnitude of the long run coefficient with
regard to real GDP seems rather large at 1.98 as it implies that manufacturing
output will increase by 1.98 percent for a one percent rise in real GDP, over
7 In the Cobb-Douglass function the elasticity of substitution between capital and labor is 1 for all values of capital and labor 8 The long run coefficients are calculated by dividing each of the ECM coefficients by the adjustment coefficient and the long run elasticities is also presented in appendices C and D
20
the long term. This strong coefficient is puzzling given that the bulk of
Swaziland manufactured production is destined for export markets9.
Foreign demand was proxied by manufactured exports. The high long-run
elasticity of +0.86 implies that a one percent increase in manufactured
exports should raise manufacturing output by 0.86 percent over the long run.
This re-emphasises the dependence of the Swazi economy (particularly the
manufacturing sector) on exports for growth. The real effective exchange rate
and the manufacturing imports variables have the expected signs and are
statistically significant. The estimated equation indicates that manufacturing
output is highly sensitive to exchange rate developments, i.e. a one percent
decrease in the real effective exchange rate will increase manufacturing
output by 1.1 percent over the long run.
As expected, the elasticities with regard to manufacturing imports and the real
prime lending rate were found to be negative. The (-0.59) long-run elasticity of
manufacturing imports implies that a one percent increase in manufacturing
imports should lower manufacturing output by 0.59 percent over the long run.
The estimation results also indicate that a one percent increase in the real
prime lending rate will lower manufacturing output by 2.72 percent over the
long-run, reflecting the sensitivity of manufacturing output to changes in the
user cost of capital in Swaziland.
The study also found that although foreign direct investment has a positive
impact on manufacturing output, it does so only in the short run, but not over
the long run. Given the historical context of FDI in Swaziland, this is not a
complete surprise. Prior to the 1990s, the country was perceived as a safe
haven for foreign investment in the light of economic sanctions imposed on
South Africa and consequently the manufacturing sector experienced high
9 The strong coefficient may be as a result of multicollinearity. However, literature indicates that this is an important explanatory variable, which should not be omitted as otherwise it may lead to the equation being misspecified.
21
growth rates. However, it would seem that the bulk of the FDI was largely for
political reasons, often only to circumvent SA sanctions, thereby not
contributing to long term sustainable growth in the Swaziland manufacturing
sector.
Once again a series of impulse responses were conducted on the demand
equation and the results are depicted in appendix G.
7. CONCLUSIONS AND POLICY IMPLICATIONS
Manufacturing output in Swaziland is affected by both supply and demand
factors. This study has econometrically established that both physical and
human capitals as well as agricultural output are major drivers of
manufacturing output from a supply perspective. On the demand side, factors
such as foreign and domestic demand, the level of the real effective
exchange rate, the real prime rate and foreign direct investment are important
drivers of Swazi manufacturing output.
The study clearly affirms the dependence of the manufacturing sector on
agricultural production for growth. Considering the substantial contribution of
agriculture to manufacturing output growth, it is imperative that if the country
put strategies aimed at enhancing manufacturing output such strategies
would also have to be aligned with the promotion of agriculture. However, the
country will also benefit from diversifying its economic base, as it would
reduce the inherent volatility associated with agricultural output, which is so
dependent on weather factors.
The estimated supply equation demonstrated that physical capital is a
necessary, but not sufficient condition for the development of the Swaziland
manufacturing sector. Given that the manufacturing sector is labour intensive
and that human capital plays a major role in manufacturing output it is
imperative that the country have a large pool of skilled people. There is thus a
22
need to invest in human capital in order to complement the increase in
physical capital.
The estimation results indicated that FDI inflows have had a positive short-run
impact on Swazi manufacturing output. It would thus seem that Swaziland
has not yet gained the full long term benefits associated with FDI inflows.
In view of the country’s limited control of both monetary and exchange rate
policies due to the domestic currency parity status with the SA rand, fiscal
policy is the only tool through which the country can affect the real effective
exchange rate by practicing sound fiscal policy aimed at stabilising inflation in
the country. This will ensure a competitive environment for the Swazi
manufacturing output in world markets. Furthermore policy makers need to
vigorously pursue policies of greater openness and integration in the world
economy which can be simultaneously pursued with appropriate domestic
economic and institutional reforms in the country.
23
APPENDIX A
Table 1: Sector Contribution to GDP at Factor Cost
Sector 2000 2001 2002 2003 2004
Agriculture 9.8 8.8 8.5 8.6 8.8
Mining 1.1 0.8 0.9 0.9 0.9
Manufacturing 35.8 35.6 35.0 35.2 34.8
Electricity & Water 3.1 3.2 3.2 3.6 3.2
Construction 6.2 7.0 7.7 6.8 6.7
Retail, Hotel & Restaurants 11.1 11.4 11.8 12.1 12.8
Transport & Communications 6.0 6.1 6.0 6.0 6.1
Banking, Insurance, & Real Estate 7.7 7.6 8.1 8.5 8.0
Government Services 16.6 16.6 16.2 16.4 16.6
Other** 2.6 2.9 2.5 2.1 2.1
GDP at Factor Cost 100.0 100.0 100.0 100.0 100.0
Source: Central Statistical Office Note: **Other includes forestry, owner-occupied dwellings and other services
Table 2 Variables and Sources Variables Sources
Real added value of agriculture WDI Real added value of manufacturing WDI Number of people employed in manufacturing sector
MoEE
Real manufacturing capital stock CSO Real manufacturing exports WDI Lilangeni effective exchange rate CBS Real manufacturing exports CBS Real expenditure on education CSO
24
APPENDIX B
Graph 1. Real GDP and Employment Growth Rates
-10
-5
0
5
10
15
2019
80
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
2002
2004
Per
cen
tag
es
Real GDP Growth Rates Employment Growth Rates
Chart 1 GDP Growth Rates of Selected SADC Countries
0
2
4
6
8
10
Bo
tsw
ana
Les
oth
o
Mo
zam
biq
ue
Nam
ibia
So
uth
Afr
ica
Sw
azil
and
Per
cen
tag
es
2003
2004
2005
25
APPENDIX C SUPPLY EQUATION ESTIMATION RESULTS
Dependent Variable: DLOG(RMAN) Method: Least Squares Date: 05/04/07 Time: 14:40 Sample (adjusted): 1980 2004 Included observations: 25 after adjustments DLOG(RMAN) =C(1)*LOG(RMAN(-1)) +C(2)*LOG(EMAN(-1)) +(-C(1) -C(2))*LOG(RCAP(-1)) +C(3)*LOG(REDUC(-1)) +C(4)*LOG(AGR( -1)) +C(5) +C(6)*DLOG(RMAN(-1)) +C(7)*DLOG(REDUC)
R-squared 0.860469 Mean dependent var 0.062953 Adjusted R-squared 0.813958 S.D. dependent var 0.127987 S.E. of regression 0.055204 Akaike info criterion -2.724064 Sum squared resid 0.054855 Schwarz criterion -2.382778 Log likelihood 41.05079 Durbin-Watson stat 1.840826
Impulse response of 1% increase in real domestic prime interest rate
-3.0-2.5-2.0
-1.5-1.0-0.50.0
0.51.01.5
0 2 4 6 8 10 12 14 16 18 20 22 24
period in years
pe
rce
nta
ge
s
RPR_5-RPR_0 (RMAN_5/RMAN_0-1)*100
Impulse response of 1% increase in real domestic exchange rate to US Dollar
-1.5
-1.0
-0.5
0.0
0.5
1.0
1.5
0 2 4 6 8 10 12 14 16 18 20 22 24
period in years
pe
rce
nta
ge
s
(REER_4/REER_0-1)*100 (RMAN_4/RMAN_0-1)*100
Impulse response of 1% increase in real manufacturing FDI
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0 2 4 6 8 10 12 14 16 18 20 22 24
period in years
pe
rce
nta
ge
s
(RFDI_M_6/RFDI_M_0-1)*100 (RMAN_6/RMAN_0-1)*100
APPENDIX H
-1.6
-1.2
-0.8
-0.4
0.0
0.4
90 91 92 93 94 95 96 97 98 99 00 01 02 03 04
Recursive C(1) Estimates ± 2 S.E.
-0.8
-0.4
0.0
0.4
0.8
1.2
1.6
90 91 92 93 94 95 96 97 98 99 00 01 02 03 04
Recursive C(2) Estimates ± 2 S.E.
0.0
0.4
0.8
1.2
1.6
2.0
2.4
90 91 92 93 94 95 96 97 98 99 00 01 02 03 04
Recursive C(3) Estimates ± 2 S.E.
-1.2
-0.8
-0.4
0.0
0.4
0.8
90 91 92 93 94 95 96 97 98 99 00 01 02 03 04
Recursive C(4) Estimates ± 2 S.E.
-16
-12
-8
-4
0
4
90 91 92 93 94 95 96 97 98 99 00 01 02 03 04
Recursive C(5) Estimates ± 2 S.E.
-0.5
0.0
0.5
1.0
1.5
2.0
2.5
90 91 92 93 94 95 96 97 98 99 00 01 02 03 04
Recursive C(6) Estimates ± 2 S.E.
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
90 91 92 93 94 95 96 97 98 99 00 01 02 03 04
Recursive C(7) Estimates ± 2 S.E.
SUPPLY EQUATION
31
APPENDIX I
-.7
-.6
-.5
-.4
-.3
-.2
-.1
92 93 94 95 96 97 98 99 00 01 02 03 04
Recursive C(1) Estimates ± 2 S.E.
0.0
0.2
0.4
0.6
0.8
1.0
1.2
92 93 94 95 96 97 98 99 00 01 02 03 04
Recursive C(2) Estimates ± 2 S.E.
0.0
0.2
0.4
0.6
0.8
1.0
92 93 94 95 96 97 98 99 00 01 02 03 04
Recursive C(3) Estimates ± 2 S.E.
-.35
-.30
-.25
-.20
-.15
-.10
-.05
.00
92 93 94 95 96 97 98 99 00 01 02 03 04
Recursive C(4) Estimates ± 2 S.E.
-.8
-.6
-.4
-.2
.0
.2
.4
92 93 94 95 96 97 98 99 00 01 02 03 04
Recursive C(5) Estimates ± 2 S.E.
-1.2
-1.0
-0.8
-0.6
-0.4
-0.2
0.0
92 93 94 95 96 97 98 99 00 01 02 03 04
Recursive C(6) Estimates ± 2 S.E.
-.3
-.2
-.1
.0
.1
.2
.3
.4
92 93 94 95 96 97 98 99 00 01 02 03 04
Recursive C(7) Estimates ± 2 S.E.
.0
.1
.2
.3
.4
.5
92 93 94 95 96 97 98 99 00 01 02 03 04
Recursive C(8) Estimates ± 2 S.E.
DEMAND EQUATION
32
REFERENCES Cancelo M, Guisan MC, Frias I. (2001) “Supply and Demand of
Manufacturing Output in OECD Countries 1970 – 95”, Applied Econometrics and International Development. AEEADE. Vol. 1-1(2001).
Central Bank of Swaziland. Various. Quarterly Review. Central Bank of
Swaziland Publications, Mbabane, Swaziland. Central Statistical Office (Swaziland). Various reports. Annual Statistical
Bulletin, Central Statistical Office Publications, Mbabane, Swaziland. Chenery, H. (1986) “growth and Transformation in H. Chenery, S. Robinson
and Syrquin eds Industrialisation and growth, Washington DC: World Bank.
Edwin Dewan and Shajehan Hussein. (2001) “Determinants of Economic
growth” (Panel Data Approach), Working Paper, Reserve Bank of Fiji, Suva, Fiji
International Monetary Fund, (United States of America). Various reports.
“International Financial Statistics” Washington DC: IMF. Krishma K.L. (2004). “Patterns and Determinants of Economic Growth In
Indian States”, Working Paper No. 144, Indian Council for Research on International Economic Relations, New Delhi-110 003, India
Lucas, R. (1988) “On the Mechanics of Economic Development”, Journal of
Monetary Economics, 22: 3-42. Mahadea* D. and Mkocha** A, “Determinants of Small-Scale Manufacturing
Enterprise Performance in Tanzania”, *School of Economics and Finance/**Iringa University, Tanzania
Manhal M. Shotar and Waleed Hmedat, (2003) “Imports of Intermidiate
Goods and Growth in Manufacturing Industries in Jordan: An Empirical Investigation”, Journal of the College of Commerce and Economics, No.2, pp. 222-256.
Miao Grace Wang, (2003) “Manufacturing FDI and Economic Growth:
Evidence from Asian Economies”, JEL-codes: F21, F43. Otani, I. and D. Villanueva. (1991) “Theoretical Aspects of Growth in
Developing Countries: External Debt Dynamics and the Role of Human Capital”, IMF Staff Papers, 36:307-42.
33
Rabelo, S. (1991) “Long Run Policy Analysis and Long Run Growth”, Journal of Political Economy, 99: 500-21.
Renelt, D. (1991) “Economic Growth: A Review of the Theoretical and
Empirical Literature”, World Bank Working Paper No. 678, Washington DC: World Bank.
Roberto Alvarez E. (2002) “Determinants of Firm Export Performance in Less
Developed Country”, JEL Classification: F10, D21, L60. Romer, P. (1986) “Increasing Returns and Long run Growth”, Journal of
Political Economy, 99: 500-21. Sergio Lodde, (1999). “Education and Growth: Some Disaggregate Evidence
from the Italian Regions”, CRENos, University of Cagliari Sichei M, “Regional Course in Macroeconomic Modelling Held at Kenya
School of Monetary Studies” Nairobi Kenya, April, 2006. Solow, R. (1956) “A Contribution to the Theory of Economic growth”.
Quarterly Journal of Economics, 70: 65-94. South African Reserve Bank (SARB). Various reports. Quarterly Bulletin.
SARB Publications, Pretoria, South Africa. Suzanna De Boef, (2000) “Modelling Equilibrium Relations: Error Correction
Models with Strongly Autoregressive Data” The Pennsylvania State University, Department of Political Science, 107 Burrowes Building, University Park, PA 16802.
Swan, T. (1956) “ Economic Growth and Capital Accumulation”, Economic