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

net/publication/363394358

Relationship between Unemployment Rate and Economic Growth in Nepal:

An Econometric Estimation

Article in Turkish Journal of Agriculture - Food Science and Technology · September 2022
DOI: 10.24925/turjaf.v10i8.1586-1593.5358

CITATIONS READS

2 3,039

3 authors:

Rabin Thapa Shiva Chandra Dhakal

Agriculture Information and Training Center (AITC) AFU
32 PUBLICATIONS 130 CITATIONS 116 PUBLICATIONS 418 CITATIONS

SEE PROFILE SEE PROFILE

Bikash Gurung
Agriculture and Forestry University
9 PUBLICATIONS 15 CITATIONS

SEE PROFILE

All content following this page was uploaded by Rabin Thapa on 09 September 2022.

The user has requested enhancement of the downloaded file.

 Turkish Journal of Agriculture - Food Science and Technology, 10(8): 1586-1593, 2022
DOI: https://doi.org/10.24925/turjaf.v10i8.1586-1593.5358

Turkish Journal of Agriculture - Food Science and Technology

Available online, ISSN: 2148-127X │www.agrifoodscience.com │ Turkish Science and Technology Publishing (TURSTEP)

Relationship between Unemployment Rate and Economic Growth in Nepal: An

Econometric Estimation
Rabin Thapa1,a,*, Shiva Chandra Dhakal1,b, Bikash Gurung1,c
1
Department of Agricultural Economics and Agribusiness Management, Faculty of Agriculture, Agriculture and Forestry University, Nepal
*
Corresponding author
ARTICLE INFO ABSTRACT
Economic growth and employment are taken as the two major objectives of macroeconomic policy
Research Article schema in both the developing as well as developed countries. This study aims to determine the
relationship between the unemployment rate and GDP growth rate of Nepal using the time-series
data from 1991-2020. Difference version approach, dynamic version approach and Granger
Received : 01/07/2022 causality test were used to study the relationship between the macroeconomic variables. The
Accepted : 26/08/2022 difference version and dynamic version approach showed non-significant results for the regression
of unemployment rate on economic growth rate which implies that the unemployment rate is not
significantly affected by the economic growth rate of Nepal. The Okun’s coefficient in the
difference version was -0.057 and in the dynamic version approach it was -0.058. Thus, the Okun’s
Keywords: law doesn’t seem to hold in case of Nepal’s economy. The Granger causality test also revealed that
GDP the change in the unemployment rate is not a predictive variable of the GDP growth rate and vice
Okun’s law versa. The non-significant relationship between the macroeconomic variables might thus be
Policy affected by other factors. Thus, it can be suggested to the government and policymakers to
Time-series analysis recuperate the ways of solving the unemployment by formulating economic policies that are more
Unemployment directed towards structural and labor market transformation. Tax reduction policy, increase in
government spending, skill enhancement programs and youth employment subsidy can also be
suggested.

a rabinth007@gmail.com https://orcid.org/0000-0003-3625-2587 b scdhakal@afu.edu.np https://orcid.org/0000-0002-2801-8937

c gbikash518@gmail.com https://orcid.org/0000-0002-5419-6155

This work is licensed under Creative Commons Attribution 4.0 International License

Introduction
Unemployment and poor economic growth rate are one 20 years. The accessible data show that the remittances
of the burning issues in today’s world among the major inflows is about 30 percent of Nepal’s gross domestic
economic problems (Kori Yahia, 2010). The escalating products (GDP), on average, with some variations (Paudel
number of unemployed people is a highly debated issue and Bhusal, 2021). Remarkably, the inflows of remittances
around the globe, even in the developed countries. Since have increased since 2002 and it has become an
they provide a clear image of a country’s economic indispensable part of national economy revealing the fact
development, the two macroeconomic variables i.e., that a large portion of youths are employed in foreign
economic growth/ Gross Domestic Product (GDP) growth countries due to unemployment in Nepal (Paudel and
and unemployment are taken as the two major objectives Bhusal, 2021). To address the paucity of employment, the
of macroeconomic policy schema in both the developing as Government of Nepal has emphasized employment
well as developed countries (Clement, 2018; Dahal and generation in its different plans, policies and programs viz
Rai, 2019). Karnali One Family One Employment Program (2006),
Unemployment, with its economic and social effects is Youth Self-employment Program (2008) and Prime
one of the most irresistible constraints faced by Nepalese Minister Employment Program (2019), etc. Although huge
policy makers (Sharma, 2019). The high rates of investments have been made and such ambitious programs
unemployment lead to a deficit in the labor market, had been introduced, the current GDP growth rate and
intensifying the poverty incidence and poor standards of unemployment rate of Nepal are -2.088% and 4.44%,
living (World Bank, 2012; Chowdhury and Hossain, respectively (Figure 1). These figures reveal the significance
2014). Furthermore, the workers’ remittance has been a of the unemployment problem in Nepal.
customary phenomenon of Nepalese economy for almost
1586
 Thapa et al./ Turkish Journal of Agriculture - Food Science and Technology, 10(8): 1586-1593, 2022

Figure 1. Unemployment and GDP growth rate in Nepal (1991-2020),

Source: Authors compilation from World Bank (2021)

Despite the social and economic impacts associated Literature Review

with the high unemployment rate, some additional
concerns remain unsettled like -- Does the size of Various studies have been done on the determination of
unemployment follow an autonomous route independent relationship between economic growth and unemployment
from other macroeconomic variables? It is a widely in different countries using diverse approaches and
accepted view in economics that a higher GDP growth rate methods. Ozturk and Aktar (2009) used the Variance
increases employment (Suleiman et al., 2017; Dahal and Decomposition and Impulse response function analysis
Rai, 2019). This theoretical proposal is popularly referred and reported that GDP is not associated with
to as ‘Okun’s Law’. It is considered one of the pronounced unemployment rate in Turkey. However, Tunah (2010)
relationships in the macroeconomics theory and has been used the ADF test followed by Granger causality test for
found to be true in case of several countries and some determination of factors affecting unemployment and he
regions, primarily developed countries (Thaba et al., 2020). reported a significant relationship between unemployment
Although Okun (1962) proposed that there exists a growth and GDP. Kreishan (2011) estimated the linkage
negative relationship between GDP growth rate and between unemployment and economic growth in Jordan
unemployment rate, the answer to these questions is related using Okun’s law and reported that the Okun’s law cannot
to the econometric estimation of the relationship between be confirmed in the case of Jordan. Similarly, Mosikari
GDP growth rate and unemployment rate. Some studies in (2013) reported a non-significant relationship between
the past have determined the simple relationship using unemployment and GDP in South Africa. Phiri’s (2014)
correlation and linear regression between the two study of Southern African countries economy, covering the
macroeconomic variables, but our study goes beyond that 2000–2013 period, revealed a non-linear equilibrium
with the assessment of short- and long-term relationship of between economic growth and unemployment. A
the variables using difference and dynamic version momentum threshold autoregressive model was used for
this purpose. Padder and Mathavan (2021) in their study on
approach along with the estimation of causality between
Indian economy from 1990-2020, reported a non-
the variables.
significant relation between GDP and unemployment
Thus, an econometric estimation of the relationship
growth along with absence of Granger causality between
between the variables will permit analysts to conduct
the variables. Thaba et al. (2020) also reported similar
appropriate policies so as to curb the unemployment and
results in their studies of South Africa’s economy.
ensure a sustainable growth in the country. Furthermore,
Ting and Ling (2011) aimed to validate the Okun’s
the countries like Nepal have different economic relationship in the case of Malaysia’s economy using the
environments than the advanced and emerging economies. first difference and gap model and found a significant
Hence, the difference in logic of their labor market makes relationship between GDP and unemployment. Similarly,
it important to determine whether or not the Okun’s law Anderton et al. (2014) determined the Okun relationship
applies in the context of Nepal. for various GDP components of 17 Eurozone countries.

1587
 Thapa et al./ Turkish Journal of Agriculture - Food Science and Technology, 10(8): 1586-1593, 2022

The results of panel regression showed that unemployment author developed two distinct models to study the
is linked to GDP. Furthermore, Apap and Gravino (2014) relationship between the variables – the difference version
used a regression analysis and found a significant negative and the dynamic version. In our study, we use both of these
relationship between output and unemployment. Based on models to analyze Nepal’s economy. For this purpose, the
the assessment of Banda (2016), a positive relationship time-series data were made stationary by suitable
between GDP and unemployment was observed in South differentiation with the help of Augmented Dicky Fuller
Africa. Hjazeen et al. (2021), using the Autoregressive (ADF) test for the purpose of using the analytical model.
distributed lag approach, determined a significant
relationship between the two macroeconomic variables in The Difference Version Approach
the economy of Jordan. The difference version model given by Okun (1962)
The evidence from multi-country studies by Pizzo was used to study the relation between unemployment rate
(2019) revealed that there has been concerns in quantifying and economic growth rate, which can be expressed
the relation between GDP and unemployment in the last empirically as:
years. The study reviewed the findings in developed and
developing countries and revealed that the Okun's (UNEMPt – UNEMPt-1) = α + β (GDPt - GDPt-1) + εt
coefficient seems to be generally higher in developed ΔUNEMPt = α + β ΔGDPt + εt (1)
economies, i.e., output would have a stronger effect on
unemployment in developed economies with respect to Where,
developing ones. Moreover, Okun’s coefficients are more ΔUNEMPt = unemployment rate in the t period
precisely estimated for developed than developing ΔGDPt = GDP growth in the t period
countries (Pizzo, 2019). For example, according to Ball et εt = Error term in the t period
al. (2016), the average Okun’s coefficient of -0.2 was
observed for the developing countries while the developed In simple words, this version refers to the regression of
countries had the value of -0.4. The R2 value usually ranges unemployment rate on GDP growth. The coefficient of
between 0.2 and 0.3 for the developing countries, being regression, β refers to the ‘Okun’s coefficient’. The
about a half of the developed economies (Ball et al., 2016). coefficient gives the rate of change in the unemployment
Farole et al. (2017) and Bartolucci et al. (2018) rate due to the GDP growth rate i.e., the real output. Okun
distinguished different countries in terms of their GDP per proposed the negative relation between the GDP growth
capita and they reported that unemployment seems to be and unemployment which implies that if β is not negative
more responsive to GDP in rich economies. Similarly, An et then it implies that the study opposes the Okun’s rule or if
al. (2017) consider the unemployment rate figures as the value is negative then there is a relationship between
unreliable in the developing countries. Furthermore, Lal et the GDP growth and unemployment as proposed by Okun.
al. (2010), determined the Okun’s coefficient, and checked The estimated elasticity gives a measure of the relationship
the validity of Okun’s law in some Asian countries using the among the variables i.e., a low coefficient implies poor
time series data from 1980-2006 in which the empirical correlation between them while high estimate of the slope
evidences revealed that Okun’s law interpretation may not coefficient indicates an existence of the relationship in par
be applicable and valid in some Asian developing countries. with Okun’s law.
Ball et al. (2016) analyzes some explanatory variables,
such as business and labor market adjustment, the The Dynamic Version Approach
contribution of services to GDP, the size of the shadow Okun (1962) came with a new approach after he
economy and a skill mismatch index as the elements of the realized the shortcomings of the difference version as the
unemployment rate. An et al. (2017) consider the poverty model failed to address the past output levels, which plays
rate, the informality rate, as well as business regulation and a strong role in affecting the current unemployment rate
the structural unemployment as the potential explanatory i.e., the lags of the variables weren’t addressed. In this
variables for unemployment rate. light, the dynamic version was brought by him, which
includes both the current and past GDP growth as the
Materials and Methods independent variable thus accounting for the variation in
the current unemployment rate (Knotek, 2007). The
Data Collection empirical representation of the dynamic approach is shown
In our study, we used the time-series data for GDP below:
growth and unemployment rate from 1991 to 2020 (30
years data). The total number of observations is hence 30. ΔUNEMPt = α + β1 ΔGDPt + β2 ΔGDPt-1 + β3
The variables associated with the econometric estimation ΔUNEMPt-1 + εt (2)
of the relationship between the economic growth rate and
unemployment rate was obtained from the World Bank Where,
open data. ΔGDPt-1 = first lag of GDP growth
ΔUNEMPt-1 = first lag of unemployment rate
Analytical Model
As mentioned earlier, Okun’s law (1962) was the first In this case, UNEMPt and GDPt denote the
to address the relationship between the output/GDP growth unemployment rate and GDP growth, respectively, and in
rate and the unemployment rate. Okun proposed that to the dynamic version, there are two lags one for the GDP
reduce 1% rate of unemployment, U.S. economy should growth and the other for the unemployment rate, which are
increase GDP by nearly 3% (Okun, 1970; Elshamy, 2013), the predictor variables for the variation brought in the
taking into account the period from 1947 to 1960. The current unemployment rate (Knotek, 2007).
1588
 Thapa et al./ Turkish Journal of Agriculture - Food Science and Technology, 10(8): 1586-1593, 2022

Causality Test show the absence of autocorrelation and partial

A simple Granger causality model was used to further autocorrelation in the time-series data as the Q statistic
determine the relationship between the unemployment rate shows non-significant results (p>0.05) for any of the lags
and economic growth of Nepal. However, it should be for both the variables.
notable to state that the word ‘causality’ in the term Our time series data are in conformity with ADF test
‘Granger Causality’ doesn’t necessarily refer to movement for stationarity and Ljung-Box test for absence of
in one variable leads to a movement in the other, but rather autocorrelation or the presence of white-noise in the time
the word implies a chronological ordering of movements series data making it liable for using the analytical models.
of the series (Brooks, 2019).
The Difference Version Approach
UNEMPt=α0+∑𝑛𝑖=1 ∝𝑖 𝑈𝑁𝐸𝑀𝑃𝑡−𝑖 +∑𝑛𝑗=1 𝛽𝑗 𝐺𝐷𝑃𝑡−𝑗 +µ1t (3) The econometric estimation from the difference version
approach for studying the relationship of GDP growth and
GDPt=λ0+∑𝑛𝑖=1 𝜆𝑖 𝑈𝑁𝐸𝑀𝑃𝑡−𝑖 + ∑𝑛𝑗=1 𝛿𝑗 𝐺𝐷𝑃𝑡−𝑗 + µ2t (4) unemployment rate is presented in Table 5 below. From the
perusal of table 5, it can be seen that the coefficient of
It is assumed that the disturbances μ1t and μ2t are regression is negative, in line with the Okun’s law. The
uncorrelated in the above equations. Equation (3) and (4) coefficient, also known as Okun’s coefficient was
implies that the dependent variables UNEMPt and GDPt are observed to be -0.057 but it was statistically non-
determined by the lagged variables of UNEMPt and GDPt significant at 5%. This implies that the difference version
in both equations, respectively. It is assumed that GDPt and showed no significant relationship between the variables.
UNEMPt are made stationary before running the causality In other words, the Okun’s law is not applicable in case of
test. The terms μ1t, μ2t are the disturbances which satisfy Nepal’s economy.
the regularity assumptions of linear regression, and ‘n’
represent the number of optimal lags. Furthermore, Table 1. Descriptive statistics of the time-series data
UNEMPt does not determine GDPt if βi = 0 in (3), for any Unemploymen Output/GDP
i = 1, 2…, n. In other words, the past values of UNEMPt do Properties
t rate (Y) (X)
not provide any additional information on GDPt Mean 2.12 4.41
performance (Thaba et al., 2020). The test provides the
Maximum 4.44 8.97
direction of the relationship among the variables i.e., bi-
Minimum 1.33 -2.08
directional, unidirectional feedback or if there is no
Standard Deviation 0.68 2.27
causation between the two variables (Suleiman et al.,
2017). STATA Version 16.0 was used for the purpose of
data analysis.

Results and Discussion

Descriptive Statistics
Table 1 below shows the descriptive statistics of the
time series data. The average unemployment rate and
output growth were 2.12% and 4.41%, respectively. The
extreme values of both the variables are close to the mean,
depicting very little spread i.e., low variation of data. It can
be further confirmed by the low value of standard
deviations. Thus, it can be asserted that there doesn’t exist
any kind of high magnitude variations in the
macroeconomic variables examined.
(a)
Unit Root Test and Test for Autocorrelation
Augmented Dicky Fuller (ADF) test was used to
determine the stationarity property of the time series data.
The null hypothesis of the ADF test is that the data is not
stationary and if we are able to reject the null hypothesis
i.e., the ADF test is significant, we conclude that the data
is stationary.
In our study, the GDP growth rate was statistically
significant at 1% for the test of stationarity (Figure 2) and thus
there was no need for further differentiation while the
unemployment rate showed non-significant results upon ADF
test. Upon fourth differentiation, the unemployment rate was
significant at 1% and hence showed stationarity (Table 2).
Ljung-Box test was used to determine the presence of
autocorrelation in the time-series data. The results of the (b)
test (Q-statistic) for autocorrelation and partial Figure 2. Stationarity of (a) Unemployment rate (4th
autocorrelation of GDP growth rate and unemployment differentiation) and (b) GDP growth rate (No
rate are shown in Table 3 and 4, respectively. The results differentiation)
1589
 Thapa et al./ Turkish Journal of Agriculture - Food Science and Technology, 10(8): 1586-1593, 2022

Table 2. ADF test for GDP growth rate (no-differentiation) and unemployment rate after 4th differentiation
Variables Test statistic Z(t) 1% Critical Value 5% Critical Value 10% Critical Value
Unemployment rate -5.814*** -4.380 -3.600 -3.240
GDP growth rate -5.209*** -3.723 -2.989 -2.625
MacKinnon approximate p-value for Z(t) = 0.0000
Note: ‘***’,’**’ and ‘*’ represent significant at 1%, 5% and 10% level of significance, respectively and NS = non-significant, Source: Own elaboration

Table 3. Ljung-Box test for autocorrelation and partial autocorrelation in GDP growth rate
Lag AC PAC Q Prob>Q
1 -0.1189 -0.1637 0.46797 0.4939
2 -0.2827 -0.4202 3.2074 0.2012
3 -0.1328 -0.3267 3.8346 0.2799
4 0.220 0.1963 5.6222 0.2292
5 0.0482 0.2302 5.7113 0.3353
6 -0.1335 -0.0863 6.4246 0.3773
7 0.0219 0.0152 6.447 0.4889
8 -0.1432 -0.6227 7.3397 0.5005
9 0.1764 0.4105 8.762 0.4595
10 0.0295 -0.0413 8.8039 0.5508
11 -0.0741 0.0180 9.081 0.6144
12 -0.2096 -0.7373 11.423 0.4931
13 0.0194 -0.4009 11.44 0.5737

Table 4. Ljung-Box test for autocorrelation and partial autocorrelation in unemployment rate
Lag AC PAC Q Prob>Q
1 -0.2751 -0.7172 2.2036 0.1377
2 -0.0164 -0.9984 2.2117 0.3309
3 0.2156 0.4749 3.6832 0.2978
4 -0.1266 0.1038 4.2132 0.3779
5 -0.0343 -0.1925 4.2541 0.5134
6 0.1167 0.5213 4.7496 0.5763
7 -0.1682 -0.1779 5.8329 0.5594
8 0.0675 -0.0649 6.0174 0.6453
9 0.1330 2.1244 6.7749 0.6605
10 -0.2619 -4.3236 9.8951 0.4497
11 0.0841 -0.0166 10.238 0.5091

Table 5. Regression output of the difference version

Variables Coefficient Std. Error t-statistic P>|t|
Const. 0.3049 0.1623 1.88 0.073
GDP -0.057 0.0339 -1.68NS 0.105
Number of obs. 26
R-squared 0.1056
Adjusted R-squared 0.0684
F-statistics 2.83
Prob (F-statistic) 0.1052
Durbin-Watson stat 2.21
Note: ‘***’,’**’ and ‘*’ represent significant at 1%, 5% and 10% level of significance, respectively and NS = non-significant, Source: Own elaboration

The Dynamic Version Approach Causality Test

From the perusal of Table 6, it can be observed that the With the purpose to determine any possible causal links
coefficients of GDP growth rate (GDP) and lagged GDP between the unemployment rate and GDP growth, the
growth rate (DL1.GDP) are statistically non-significant at Granger causality test was used. The test is useful for
5% while the lagged unemployment rate statistical detection of the presence of any cause effect
(DL1.Unemployment) is statistically significant at 5%. interaction between the variables considered. The results of
This implies that, 1 unit increase in the unemployment rate the Granger causality test are shown in Table 7 below.
in time ‘t-1’ decreases the economic growth rate in time ‘t’ From the perusal of Table 6, it can be observed that
by 0.802 units. The dynamic model is able to explain only both the null hypothesis couldn’t be rejected at 5% level of
34.3% of the variation in the unemployment rate due to the significance. This means that there is no existence of
variation in the GDP growth rate and the lagged variables. feedback between the variables or there is absence of
causality between the unemployment rate and the GDP
1590
 Thapa et al./ Turkish Journal of Agriculture - Food Science and Technology, 10(8): 1586-1593, 2022

growth rate. So, the study concluded that the change in the GDP and unemployment rate in the economy of Jordan.
unemployment rate is not a predictive variable of the GDP Furthermore, Banda (2016) estimated that the
growth rate and vice versa. unemployment and economic growth in South Africa has a
Similar results have been reported by Ozturk and positive relationship. The evidence from multi-country
Akhtar (2009) in their study in Turkey where they found studies by Pizzo (2019) revealed that the Okun's coefficient
GDP is not associated with unemployment rate. However, seems to be generally greater in economies of developed
Tunah (2010) used the ADF test followed by Granger nations, i.e., GDP would have a higher impact on
causality test for determination of factors affecting unemployment in developed economies with respect to
unemployment and he reported a significant relationship developing ones. Moreover, Okun’s coefficients are more
between unemployment growth and GDP. Kreishan (2011) precisely estimated for developed than developing
estimated the relationship between unemployment and countries (Pizzo, 2019). Furthermore, according to Ball et
output in Jordan and reported that the Okun’s law doesn’t al. (2016), the average Okun’s coefficient of -0.2 was
hold in case of Jordan. Similarly, Mosikari (2013) reported observed for the developing countries while the developed
a non-significant relationship between unemployment and countries had the value of -0.4. The R2 value normally
GDP in South Africa. Phiri’s (2014) study of Southern ranges between 0.2 and 0.3 for the developing countries,
African countries economy, covering the 2000–2013 being about 0.5 times to that in the economies of developed
period, revealed a non-linear equilibrium between nations (Ball et al., 2016).
economic growth and unemployment. Padder and The approaches used in our study showed that there is
Mathavan (2021) in their study on Indian economy from a non-significant relationship between the variables thus
1990-2020, reported a non-significant relation between resisting any link between the GDP growth rate and
GDP and unemployment growth along with absence of unemployment rate of Nepal. Furthermore, the causality
causality between the variables. Thaba et al. (2020) also test validated the absence of any short-run causality
reported similar results in their studies of South Africa’s connecting the GDP growth to the unemployment rate.
economy. Farole et al. (2017) and Bartolucci et al. (2018) This implies that the unemployment rate changes cannot be
reported that unemployment seems to be more reactive to solely explained by the GDP growth variation, but rather
GDP in case of rich economies. Similarly, An et al. (2017) other factors play a part in its determination. There might
consider the unemployment rate statistics as unreliable in be some potential explanatory variables, such as business
the developing countries. Furthermore, Lal et al. (2010) and labor market adjustment, the share of services in GDP,
determined the validity of Okun’s law in some Asian the size of the shadow economy, as well as a skill mismatch
countries and reported that the Okun’s law interpretation index that ascertain the unemployment rate (Ball et al.,
may not be germane in some Asian developing countries. 2016). Furthermore, the poverty rate, the informality rate,
In contrast with our results, Ting and Ling (2011), in as well as business regulation and the structural
their study of Malaysia’s economy using the first unemployment are also considered to be some impending
difference and gap model, found a significant relationship explanatory variables for the unemployment rate (An et al.,
between GDP and unemployment. Similarly, Anderton et 2017). The relocation of people from rural to urban leading
al. (2014) determined the Okun’s law for various GDP to increased labor force supply, social aspects like debasing
components of 17 Eurozone countries and reported that social status, geographical fixity, population growth, and
unemployment is receptive to the GDP. Apap and Gravino flawed education system, unexperienced, occupational
(2014) also revealed that there was a significant negative unfitness, and disease and disability might also be
relationship between output and unemployment. Hjazeen responsible factors for the unemployment rate (Chand et
et al. (2021) also reported a significant linkage between al., 2018).

Table 6. Regression output of the dynamic version

Variables Coefficient Std. Error t-statistic P>|t|
GDP -0.058 0.031 -1.86NS 0.077
DL1.GDP 0.035 0.039 0.91NS 0.375
DL1.Unemployment -0.802 0.293 -2.73** 0.012
Const. 0.157 0.240 0.65 0.522
Number of obs. 25
R-squared 0.3430
Adjusted R-squared 0.2492
F-statistics 3.66
Prob (F-statistic) 0.0290
Durbin-Watson stat 1.99
Note: ‘***’,’**’ and ‘*’ represent significant at 1%, 5% and 10% level of significance, respectively and NS = non-significant, Source: Own elaboration

Table 7. Granger Causality test between unemployment rate and GDP growth rate
Null Hypothesis Observations Lags F-statistic Prob.
GDP doesn’t impact UNEMP 30 2 1.791NS 0.193
UNEMP doesn’t impact GDP 0.969NS 0.397
Note: ‘***’,’**’ and ‘*’ represent significant at 1%, 5% and 10% level of significance, respectively and NS = non-significant, Source: Own elaboration

1591
 Thapa et al./ Turkish Journal of Agriculture - Food Science and Technology, 10(8): 1586-1593, 2022

Based on the results of our study, it can be suggested Bartolucci F, Choudhry M, Marelli E, Signorelli M. 2018. GDP
that the government and policymakers should devise Dynamics and Unemployment Changes in Developed and
economic policies that are more directed towards structural Developing Countries. Applied Economics, 50(31): 3338-3356.
changes and labor market transformation to augment the Brooks C. 2019. Introductory Econometrics for Finance. In
Introductory Econometrics for Finance. Cambridge
existing policies. This is because unemployment is a major University Press. https://doi.org/10.1017/9781108524872
hindrance to the social development and results in wastage Chand K, Tiwari R, Phuyal M. 2018. Economic Growth and
of the efficient manpower in which the government has Unemployment Rate: An Empirical Study of Indian
financed. Other policies that the government can consider Economy. PRAGATI : Journal of Indian Economy, 4(02).
to reduce unemployment include a tax reduction. https://doi.org/10.17492/pragati.v4i02.11468
Furthermore, the government should invest in the Chowdhury M. S. R, Hossain M. T. 2014. Determinants of
educational sector as it creates employment in the long run Unemployment in Bangladesh: A Case Study. SSRN
and also developing the infrastructure in unison could be Electronic Journal, 4(3):16–20. https://doi.org/10.2139/
vital in solving the unemployment. Nepal also faces a ssrn.2402908
Clement S. 2018. The effect of unemployment on economic
major problem of structural unemployment. To increase growth in South Africa (1994-2016). Munich Personal RePEc
the employment opportunities, the government should Archive, 85305, 1–15.
embark skill development programs to abolish structural Dahal M. P, Rai H. 2019. Employment Intensity of Economic
unemployment. Other policies that could be adopted Growth: Evidence from Nepal. Economic Journal of
include employment subsidies and a reduction of trade Development Issues, 27(1): 34–47. https://doi.org/10.3126/
union powers. Nonetheless, introducing youth employment ejdi.v28i1-2.33195
subsidy can be a step in the right direction to help increase Elshamy H. 2013. The Relationship Between Unemployment and
employment. Output in Egypt. Procedia - Social and Behavioral Sciences,
81: 22–26. https://doi.org/10.1016/j.sbspro.2013.06.381
Farole T, Ferro E, Michel Gutierrez V. 2017. Job creation in the
Conclusion private sector, Jobs Working Paper No. 5, World Bank.
Hjazeen H, Seraj M, Ozdeser H. 2021. The nexus between the
The study aimed to determine the empirical relation economic growth and unemployment in Jordan. Future
between the economic growth rate and the unemployment Business Journal, 7(1). https://doi.org/10.1186/s43093-021-
rate. We used the difference version and dynamic version 00088-3
approach, both of which showed non-significant results Lal I, Muhammad S.D, Jalil M.A, Hussain A. 2010. Test of
implying the inability of economic growth rate to define the Okun's Law in some Asian countries Co-Integration
change in unemployment rate. The Okun’s coefficient in the Approach. European Journal of Scientific Research, 79-80.
difference version was -0.057 and in the dynamic version Knotek E. S. 2007. How useful is Okun’s law? Economic-Review
Federal Reserve Bank. Kansas City, 92(4): 73.
approach it was -0.058. Furthermore, Granger causality test Kori Yahia A. 2010. Bayesian Linear Estimation of Okun
also showed the absence of causal relationship between the Coefficient for Romania: Sensitivity to Priors. The Romanian
variables. This implies that, there is not enough evidence Economic Journal, 38: 53–65.
from the study to provide a link between the variables and Kreishan F. 2011. Economic Growth and Unemployment: An
the unemployment rate is thus affected by other factors. It Empirical Analysis. Journal of Social Science, 7(2): 228–231.
can be suggested to the government and policymakers to Mosikari T.J. 2013. The effect of unemployment rate on gross
improve the ways of dealing with unemployment problems domestic product: Case of south africa. Mediterranean
by employing economic policies that are more oriented Journal of Social Sciences, 4(6): 429–434.
towards structural changes and labor market transformation https://doi.org/10.5901/mjss.2013.v4n6p429
Okun A. 1962. Potential GNP:Its Measurement and Significance.
to supplement the existing policies. Furthermore, tax Proceedings of the Business and Economics Statistics
reduction policy, increase in government spending, skill Section. American Statistical Association.
development programs and youth employment subsidies can Okun A. 1970. “Potential GNP: Its Measurement and
be suggested. Significance.” In: The Political Economy of Prosperity. New
York: Norton, pp. 132-145.
References Ozturk L., Aktar I. 2009. “Can Unemployment be Cured by
Economic Growth and Foreign Direct Investment in
An Z, Ghazi T, Prieto N.G. 2017. Growth and jobs in developing Turkey?” International Research Journal of Finance and
economies: Trends and cycles, IMF Working Paper Economics, 27:1450-2887.
WP/17/257. Padder A.H, Mathavan B. 2021. The Relationship between
Anderton R, Aranki T, Bonthuis B, Jarvis V. 2014. Unemployment and Economic Growth in India: Granger
Disaggregating Okun’s Law Decomposing the Impact of the Causality Approach. Natural Volatiles and Essential Oils,
Expenditure Components of GDP on Euro Area 8(04): 1265–1271. https://doi.org/10.32368/fjes.20211701
Unemployment. European Central Bank Working Paper Ser., Paudel R.C, Bhusal T. P. 2021. Role of Workers ’ Remittances in
1747. Retrieved from: www.ecb.europa.eu Export Performance of Nepal : Gravity Modelling Approach.
Apap W, Gravino D. 2014. Okun’s Law in Malta: Lessons Learnt Nepal Rastra Bank, 33(1): 1–13.
from a Sectoral Perspective. Min. Fin. Work. Paper Ser., 01. Phiri A. (2014). Nonlinear co-integration between unemployment
Retrieved from: www.mfin.gov.mt and economic growth in South African Management Global
Ball L, Furceri D, Leigh D, Loungani P. 2016. Does One Law Fit Transformation, 12(4): 303.
All? Cross Country Evidence on Okun’s Law, mimeo. Pizzo A. 2019. Empirical Studies on Okun’s Law in Latin
Banda H. 2016. The impact of economic growth on America and the Caribbean. International Labour
unemployment in South Africa: 1994–2012. Investment Organization, 3(7): 27–43.
Management and Financial Innovations, 13(2): 246–255. Sharma S. 2019. The Rhetoric of Unemployment, Racism and
Brooks C. (2002). Introductory econometrics for finance. Sexism in Nepal. International Journal of Recent Academic
Cambridge University Press. Research, 1(07): 336–339.
1592
 Thapa et al./ Turkish Journal of Agriculture - Food Science and Technology, 10(8): 1586-1593, 2022

Suleiman S, Kassim S, Hemed I. 2017. Unemployment and Ting N.Y, Ling L.S. 2011. Okun’s law in Malaysia: an
Economic Growth in Tanzania. Journal of Economics, autoregressive distributed lag (ARDL) approach with
Management and Trade, 20(2): 1–8. Hodrick-Prescott (HP) filter. Journal of Global Business
https://doi.org/10.9734/jemt/2017/38500 Economy, 2(1): 95–103.
Thaba T. K, Belete A, Hlongwane J. J, Ledwaba L.J. 2020. World Bank. 2012. World Development Report 2013: Jobs.
Econometric estimation of the relationship between the Washington,DC: World Bank.
unemployment rate and economic growth in Limpopo World Bank. 2021. GDP growth (annual %).
Province, South Africa. Journal of Agribusiness and Rural https://data.worldbank.org/indicator/NY.GDP.MKTP.KD.Z
Development, 58(4). G
https://doi.org/10.17306/j.jard.2020.01291

1593

View publication stats