See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/347933552

On the Role of Remittance and FDI in Economic Development of Nepal

Article · January 2010

CITATIONS READS
2 168

1 author:

Shashi Chaudhary
British University Vietnam
16 PUBLICATIONS   71 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Global Value Chains and international integration View project

Tourism Economics View project

All content following this page was uploaded by Shashi Chaudhary on 27 December 2020.

The user has requested enhancement of the downloaded file.

PACE - The Journal of Business and Management Vol. 2, Issue No.1 January 2010 pp. 1-10

On the Role of Remittance and FDI in Economic Development of Nepal

- Shashi K. Chaudhary

Abstract
In this paper cointegration and error-correction models have been used to analyze the role of
remittance and FDI in economic development of Nepal. The cointegration analysis provides
existence of long-run relation among the variables. Remittance inflows and FDI both are
marginally significant and bear positive association with real GDP in the long-run. One percent
rise in remittance inflow changes GDP by 0.012 percent while that of FDI changes the real GDP
by 0.002 percent in the long-run. In the short-run, the impact of remittance inflow is 0.12 and
that of FDI is 0.013 on real GDP. In comparison, their influences are stronger in the short-run
than long-run. There exists bidirectional causality between real GDP and real remittance inflow.
There is also causality from GDP to FDI but the reverse does not hold true. In conclusion, the
role of remittance on the real GDP is found quite dominant over that of FDI in both short-run
and long –run. The impact of FDI has been found almost constant and does not change in the
short-run.
Keywords: cointegration, error correction model, economic growth, FDI, remittance
JEL Classification Code: O11, C22, E62

1. INTRODUCTION

Because of political instability and unfavorable climate for agriculture, the economic sector of
Nepal has been passing through the critical phase of low–level equilibrium trap circumscribed by
poverty, high inflation and tremendous imports. Real sectors are hard hit by the unionized strike,
blockade and increasing load-shedding causing rise in the cost of production. This has decreased
the export competitiveness of the Nepalese product in the international market. In such a gloomy
condition, the remittance has become a crucial component for Nepalese economy. The GDP has
also accounted remittance as one of the major sources of national income of the country. The
remittance has contributed much to maintain Nepal’s BOP position favorable. Not only this, it
has become one of the prime sources of foreign currency earning and mitigating ever growing
imports bills for Nepal. The expansion of banking and financial sector has been possible due to


Assistant Professor in Economics, School of Business, Pokhara University. saw_sea@outlook.com
presence of remittance in Nepal. The high inflow of remittance for Bangladesh since 1996 have
allowed foreign exchange reserves to increase and provided the confidence to float her currency.
It is quite noticeable that Bangladesh has attained an economic growth rate of more than 6
percent1 since 2004 onward mainly by increasing the flow of remittance and maintaining strong
production sector. Thus, remittance as a major source of foreign exchange earnings can improve
a country's creditworthiness and enhance its access to international capital markets.
FDI is considered beneficial in view of its contribution to technological transfers, enhancement
of managerial capability and new opportunities for market access. It includes the transfer of
intangible assets such as trademark, technology and business management as well as the
authorization given to the investor to control the investment. Increase in FDI is seen as leading
factor to increase exports by creating international markets through new marketing and
organizational skills. Therefore, it is not unusual for economists to emphasize the importance of
FDI in fueling economic growth. In fact, since the early 1950s, FDI has been recognized as the
most crucial factor in enhancing economic development and ensuring a reasonable standard of
living for countries which have been the recipient of FDI. South Korea, Singapore, and Taiwan
have been examples of nations outside the OECD countries that have benefitted greatly from
FDI. In recent years, China and India have made remarkable progress in attracting FDI and in
realizing technological and economic successes.
In the light of the discussion mentioned above, it would surely be concluded that remittance and
FDI can play the role of lifeline and up lifter for the developing economies like Nepal. Of course,
Nepalese economy has been passing through serious structural constraints accompanied by
political turmoil and extremely limited internal resources in the recent decade, yet she has
enormous economic potentialities in terms of water resources, tourism, human resources and
biodiversity. Thus, in the situation of extremely limited internal resources, FDI is crucial to
sustain development activities and create employment opportunities.

2. LITERATURE REVIEW

2.1 Relationship between Remittance and Economic Growth

Evidence on the increasing dependency on remittance is strengthen by the fact that after export
earnings, remittance inflows have become one of the largest and most stable sources of foreign

1 World Economic Outlook database 2009. IMF.

2
exchange for developing countries. Despite the significant size and stability of these inflows,
economists find it difficult to determine the effects of remittance on the growth of recipient
economies in the developing world. So far, empirical research on the relationship between
remittance and economic growth has yielded mixed results. Spatafora, Aggrawal and Cabugao
(2005:73-77) found no statistically significant relationship between remittance flows and per
capita income (quoted in Leon-Manlagnit, 2005: 4). Prior to this Chami, Fullenkamp and Jahjah
(2003: 77-78) found evidence of a significant negative relationship between the growth rates of
GDP and remittance. They argued that remittance tend to compensate recipients for bad
economic outcomes. So this creates incentives for recipients to be less productive and more
dependent on these inflows.
Conducting a comprehensive review of the theories on remittance for more than 30 years,
Rapoport and Docquier (2005:54-55) argued that remittance at the best could only be used to
overcome short-run liquidity constraints and had minimal long-term effects. They added
remittance were also purported to discourage labor supply and work effort among recipients
which resulted in increased dependency, lower productivity and thus delayed growth. Stark
(1991: 211-214) noted that because remittance were mainly in the form of cash, they were
fungible and could therefore be used to purchase both financial as well as physical assets. Such
assets, in turn, could be used in productive activities like farm investments and entrepreneurial
formation. In this way, remittance act more as a catalyst for growth rather than a direct input to
it. In the study about role of remittance in economic development of Nepal, Srivastava and
Chaudhary (2007:34) found that even less than eight percent of the total remittance was directed
towards the directly productive sectors in Nepal. Hereby concluding that in spite of its great
caliber, remittance was not flown in the right direction. They added that the coefficients of
growth of real remittance was found to be 0.002, 0.003 and 0.003 respectively showing the
marginal effect of the growth of remittance on the growth of real GDP, GNP and per capita real
income.
Therefore, in order to settle the issue of whether remittance contributes or not to long-run growth,
it is first important to determine how the remittance incomes are being used across different
remittance receiving economies. The pattern by which recipients allocate remittance between
consumption and saving will decide what policy should be employed to harness remittance as a
tool for growth.

3
2.2 Relationship between FDI and Economic Growth
According to neoclassical growth model, Foreign Direct Investments cause medium-term
temporary increases in economic growth in the countries where investments are made through
increasing the amount of investment and its efficiency. On the other hand, new endogenous
growth theories focus on the long-term growth as a function of technological processes.
Therefore; they claim that Foreign Direct Investments can continuously increase growth rate
through technology transfer and spillover effects. Ericsson and Irandoust (2001:122-132)
calculated the cause and effect relationships between FDI and economic growth by using the data
collected from four OECD countries (Denmark, Finland, Norway and Sweden) in 2001and failed
to find a causality relationship for Denmark and Finland. On the contrary, Makki and Somwaru
(2004:795-801) in their study found that FDI affects economic growth to a large extent together
with foreign trade, human capital and domestic capital.

3. METHODOLOGY

The present work follows a three step procedure. First, the stationarity of the data series have
been examined to determine the order of integration of the series. For this, Augmented Dickey
Fuller (ADF) test has been applied. In the second step, cointegration among the variables has
been tested using Engle – Granger2 and Johansen techniques. In the last step, the causality
dynamics among the variables has been tested through error correction model (ECM). All the
necessary computations have been made using Eviews 5.1 version.

3.1. Theoretical Background

The level of output in an economy is determined by the availability of factors of production. This
can be expressed as follows:
Y = f (K, L) (1)
where Y denotes the output level (real gross domestic product, GDPr), K denotes the amount of
capital, and L denotes the amount of labour. After adding remittance (Rm) and foreign direct
investment (FDI), equation (1) can be written as:
Y = f (K, L, Rm, FDI) (2)

2
It is a two step procedure. First, the time series properties of each variable are examined by unit root tests. Having
tested the stationarity of each time series, the cointegration regression between variables are estimated using the
OLS. The second step involves directly testing for the stationarity of error processes of the cointegration regression
estimated in first step.

4
In equation (2), it has been argued that the K and L coefficients are positively related to Y. For a
moment, if K and L are assumed to be given for an economy, then FDI itself can influence the
level of Y either through capital formation or changes in the level of technology. That’s why,
literature linking FDI to economic development is more consistent and this variable has been
expected to have a positive effect on the level of output. The effect of remittance on domestic
output (Y) cannot be confirmed as it has tendency to increase aggregate demand. So there are
possibilities for import to hike up as well. Thus, though its coefficient is expected to be positive,
if remittance recipients increase the demand for imported goods, its coefficient may become
negligible or negative too.

3.2. Nonstationary Series and Testing for Unit Root

The series {lnGDPr}, {lnRmr) and {lnFDIr) have been investigated for stationarity where GDPr
stands for gross domestic product, Rmr stands for remittance inflows and FDI stands for foreign
direct investment, all measured in real terms.
To distinguish a unit root, we can run the regression
m
∆yt = b0 +  bi y t − i + βt + λyt-1 +μt (3)
i =1

The regression includes enough lags of ∆yt so that μt contains no autocorrelation. If there is a
unit root, differencing yt should result in a white noise series.
m
∆∆yt = b0 +  bi  y t − i + λ∆yt-1 + μt (4)
i =1

3.3. Cointegration Test and Error Correction

If xt and yt are cointegrated, there is a long-run relationship between them. Furthermore, the
short-run dynamics can be described by the error correction model (ECM). This is known as the
Granger representation theorem.
If xt ~ I(1), yt ~ I(1) and zt = yt – βxt is I(0), then x and y are said to be cointegrated. The
Granger representation theorem says that in this case xt and yt may be considered to be generated
by ECMs of the form
∆xt = ρ1zt-1 + lagged (∆xt, ∆yt) + ε1t (5)
∆yt = ρ2zt-1 + lagged (∆xt, ∆yt) + ε2t (6)
where at least one of ρ1 and ρ2 is nonzero and ε1t and ε2t are white noise errors.

5
3.4. Testing for Causality
As per the Granger representation theorem, if xt ~ I(1) and yt ~ I(1), then the variables have the
error correction form
p p
∆xt = ρ1zt-1 +  1i  x t − i +   2i y t − i + ε1t (7)
i =1 i =1

p p
∆yt = ρ2zt-1+  1i  x t − i +   2i y t − i + ε2t (8)
i =1 i =1

where ∆ is first difference operator on variables, ε1t and ε2t are white noise disturbances. The
residual zt-1 is the error correction term which is the lagged residuals from the cointegration
relations. Its magnitude, ρ1 or ρ2 implies the deviation from long run equilibrium in period (t-1).
The independent variables are said to long run ‘cause’ the dependent variable if the error
correction term, (ρt-1) is significant based on t-test statistics or short run ‘cause’ if the coefficients
of the lagged independent variables are jointly significant based on F-test statistic in their first
differences.

3.5. Data
The time series data for GDP, FDI and remittance has been used in this analysis covering the
period of 1990/91 to 2008/09. The data have been extracted from the Economic Survey (2005/06,
2006/07, 2007/08 and 2008/09) of MOF, G/N and A Handbook of Government Finance Statistics
(2008), NRB. The nominal figures of GDP, FDI and remittance have been deflated by the GDP
deflator (base year 2000/01) to express them in real terms.

4. EMPIRICAL RESULTS

Table 1 presents the results of unit root tests obtained using the ADF test on individual series.
The lag length selection was set automatic based on SIC (Schwarz Information Criterion), max
lag = 6. The absolute calculated values are less than the corresponding McKinnon critical values
at levels of the variables. Hence, the null hypothesis cannot be rejected. However, the null
hypothesis can be rejected when the first difference of the variables are taken. Thus, their first
difference is found to be stationary and hence all the series are integrated of order one, I(1).

6
 Table 1: Results of Unit Root Tests
ADF statistic
Series Degree of Integration
Level First difference
lnGDPr -3.24 -5.61* I(1)
lnRmr -2.13 -3.65 I(1)
lnFDIr -3.66 -4.512** I(1)
*, ** & indicate 0.01, 0.0 5 and 0.06 level of significance based on Mackinnon’s critical values.
All the regressions with level and first difference have been done with trend and intercept criteria.

After confirming the existence of unit roots for the data series, the next step involves to test
whether the variables are cointegrated. This has been done by both the Engel-Granger and
Johansen techniques. The table 2 presents the test statistics for Engel-Granger technique while
table 3 presents the test statistics for Johansen technique.
The table 2 indicates that the estimated ADF statistics for the residuals are greater than their
corresponding critical values at one percent level except for model (2.2) which is significant at
five percent level. Thus, GDPr, Rmr and FDIr can be concluded to be cointegrated at five
percent level of significance. This finding is also confirmed by the CRDW statistic which is also
significant at one percent level. Thus, the stationarity of the residuals of cointegration equations
is confirmed and hence implies existence of a long-run association among variables.

Table 2: Results of Engle-Granger Cointegration Test

Model Cointegration equation CRDW ADF statistic for Residuals
2.1 lnGDPr = f(lnRmr, lnFDIr)3 0.989* -2.71*
2.2 lnRmr = f(lnGDPr, lnFDIr)4 0.999* -2.54**
2.3 lnFDIr = f(lnRmr, lnGDPr)5 1.66* -3.80*
* & ** indicate the statistical significance at the 0.01 and 0.05 levels of significance based on
Mackinnon’s critical values. The critical value of the CRDW statistic is 0.511 at the 0.01 level of
significance.

3
lnGDPr = 0.2340359172 lnRmr - 0.005392027142 lnFDIr + 10.57496388 R2 = 0.936
(14.45)* (-0.194) (35.38)*
4
lnRmr = 3.96899825 lnGDPr - 0.01917154936 lnFDIr - 40.96337862 R2 = 0.936
(14.45)* (-0.167) (-10.51)*
5
lnFDIr = -0.09183735229 lnRmr - 0.4380388258 lnGDPr + 13.84787782 R2 = 0.113
(-0.167) (-0.194) (0.583)

7
The existence of cointegration is also confirmed by Johansen methodology and confirms one
long-run cointegrating equation (table 3).

Table 3: Results of Johansen Cointegration Test

Trace Max-Eigen Value
Hypothesized
No. of CEs Statistic 0.05 Critical Value Statistic 0.05 Critical Value
None* 62.61 42.91 46.12 25.82
At most 1 16.48 25.87 13.36 19.38
At most 2 3.11 12.51 3.11 12.51
Trace test indicates 1 cointegrating equation at the 0.05 level.
Max-eigen value test also indicates 1 cointegrating equation at the 0.05 level.
* denotes rejection of the hypothesis at the 0.05 level.

Next is the Ganger causality test with error correction terms from the cointegration equations.
The empirical results of the estimated error-correction models are presented in Table 4. It
presents the results of both the long run Granger causality test based on a standard t-test statistics
of the error terms, (zt-1) lagged one period as well as the short run Granger causality test based on
a standard F-test statistics for the jointly significance of the coefficients of the explanatory
variables in their first differences.
Based on the t- test statistic, models (2.1) and (2.2) are only significant for the analysis of the
long-run equilibrium. However, the coefficients are greater than unity indicating explosive
nature of the models, hereby suggesting need of suitable actions to bring it back to the long-run
equilibrium path. Further, the growth of one year lag remittance and FDI has established a
positive relationship with long-run real GDP growth6, but their impact is very marginal. The
findings suggest that one percent change in remittance influence real GDP by 0.012 percent
while that of FDI influence by merely 0.002 percent. Moreover, the bidirectional causality exists
between one year lagged real GDP and real remittance inflow in the long-run. Since the F-
statistics of both the models are significant at five percent level, the findings are considered to be
reliable. On this basis, I have excluded the model (2.3).

6
The long-run relation can be represented as lnGDPrt-1= 12.19 + 0.012 lnRmrt-1+ 0.002 lnFDIrt-1 + 0.05t + zt-1.

8
 Table 4: Vector Error Correction Estimates
Dependent Variables
Independent Variables ∆lnGDPrt (Model 2.1) ∆lnRmrt (Model 2.2) ∆lnFDIrt (Model 2.3)
1.18 -36.25 -32.08
ECM (zt-1)
(1.55)*** (-4.50)* (-0.94)
-1.92 27.16 31.61
∆lnGDPrt-1
(-2.38)** (3.17)* (0.87)
-0.75 8.00 20.15
∆lnGDPrt-2
(-1.85)** (1.85)** (1.11)
0.12 -1.57 -0.94
∆lnRmrt-1
(3.50)* (-4.1)* (-0.58)
0.07 -0.93 -2.14
∆lnRmrt-2
(1.78)** (-2.04)** (-1.12)
0.013 -0.21 -0.49
∆lnFDIrt-1
(2.01)** (-3.09)* (-1.70)***
0.01 -0.12 -0.84
∆lnFDIrt-2
(1.65)*** (-1.43)*** (-2.36)**
0.17 -1.28 -2.51
Constant
(3.11)* (-2.22) (-1.03)
R-square 0.834 0.818 0.58
F-statistic 5.78** 5.14** 1.62

where zt-1 = (lnGDPrt-1 - 0.012 lnRmrt-1- 0.002 lnFDIrt-1 – 0.05t - 12.19), is the error correction term.
Figures in parentheses indicate t-statistic. *, ** & *** indicate statistically significant at the 0.01, 0.05
and 0.10 levels respectively.

Let us now examine the short run results, which can also be inferred from table 4. As mentioned
above that the F-statistics for the models (2.1) and (2.2) are significant, it means the lagged
independent variables cause the dependent variable in the short-run. The coefficients show that
impact of remittance in significant and positive in the growth of real GDP. One percent change
in remittance would bring 0.12 percent change in real GDP but this influence is decreasing over
time. The FDI is similarly associated with the real GDP. The intensity of FDI is 0.013 in the first
and second lags both, hereby indicating a constant contribution to real GDP (model 2.1).
In model (2.2), the change in real GDP has positive influence on remittance inflow and
association is quite strong. Further, the remittance and FDI are negatively associated. One
percent change in FDI would reduce the remittance inflow by 0.21 percent in the first lag while
reduces by 0.12 percent in the second lag.

9
 5. CONCLUSION

This paper has examined the role of remittance and FDI with perspective to economic
development of Nepal through cointegration and error-correction models. The cointegration
analysis provides existence of long-run relation among the variables. The remittance inflow and
FDI, both have positive impact on growth of the real GDP, though intensities are marginal. There
exists bidirectional causality between real GDP and real remittance inflow. There is causality
from GDP to FDI but the reverse is not valid. In the analysis, the role of remittance on the real
GDP is found quite dominant over that of FDI in both short-run and long –run. The impact of
FDI has been found almost constant and does not change over time in the short-run.

REFERENCES

▪ Afsar, M. (2007). The Causality Relationship between Economic Growth and Foreign Direct
Investment in Turkey. Retrieved on November, 21, 2009 from Business and Economics
Society International; website: www.besiweb.com/BESIP07.
▪ Chami, R., Fullenkamp, C., & Jahjah, S. (2005). Are Immigrant Remittance Flows a Source
of Capital for Development? IMF Staff Papers, 52(1): 55-81.
▪ Ericsson, J., & Irandoust, M. (2001). On the Causality Between foreign Direct Investment
and Output: A Comparative Study. The International Trade Journal, 15: 122-132.
▪ Leon-Manlagnit, P.J.F. (2006). The Role of Remittance in the Growth of Developing
Recipient Economies. An unpublished Master’s thesis, Simon Fraser University.
▪ Makki, S. S., & Somwaru, A. (2004). Impact of Foreign Direct Investment and Trade on
Economic Growth: Evidence from Developing Countries. American Journal of Agricultural
Economics, 86 (3): 795-801.
▪ MOF (2006, 2007, 2008, 2009). Economic Survey. Kathmandu: G/N.
▪ NRB (2008). A Handbook of Government Finance Statistics. Kathmandu: Research
Department, Government Finance Division.
▪ Rapoport, H., & Docquier, F. (2005). The Economics of Migrants’ Remittance. Discussion
Paper No. 1531/March 2005. Institute for the Study of Labor.
▪ Srivastava, N. L., & Chaudhary, S. K. (2007). Role of Remittance in Economic Development
of Nepal. The Journal of Nepalese Business Studies, 4(1): 28-37.
▪ Stark, O. (1991). The Migration of Labor. Oxford: Basil Blackwell Ltd.



10

View publication stats