Journal of Economic Impact 3 (3) 2021.

221-226

Available Online
Journal of Economic Impact
ISSN: 2664-9764 (Online), 2664-9756 (Print)
http://www.scienceimpactpub.com/jei

FACTORS AFFECTING LABOR PRODUCTIVITY: AN EMPIRICAL EVIDENCE FROM PAKISTAN

Ghulam Sarwar a,*, Muhammad Fayyaz Sheikh b, Iqra Rabnawaz c

a Noon Business School, University of Sargodha, Sargodha, Pakistan

b Lyallpur Business School, Government College University Faisalabad, Pakistan
c School Education Department, Sargodha, Government of Punjab, Pakistan

ARTICLE INFO ABSTRACT

Article history Labor productivity is important as it is the major factor determining nations' living
Received: October 04, 2021 standards. This study analyzes the factors affecting labor productivity in Pakistan using
Revised: December 22, 2021 time series data. ARDL model is applied for estimation of the long run relationship of
Accepted: December 25, 2021 variables for the period 1981-2018. Data have been taken from the Handbook of Statistics
of State Bank of Pakistan and various economic surveys of Pakistan. The findings show that
wages, human capital investment, labor force participation, and inflation significantly
Keywords affect labor productivity. The results indicate that wage rate has a positive effect on labor
Labor productivity
productivity, and human capital investment also is positively related to labor productivity.
Wage rate
At the same time, labor force participation and inflation are negatively related to labor
Human capital investment
Inflation rate
productivity. These findings imply that labor productivity can be raised by increasing the
wage rate and investing more in human capital. Results are consistent with efficiency wage
theory and human capital theory.
* Email: [email protected]
https://doi.org/10.52223/jei3032112
© The Author(s) 2021.
This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

INTRODUCTION
Labor productivity has become a dominant factor that not only markets. Higher productivity not only indicated the availability of
affects the competitiveness of the industry but also affects cheaper amount of goods and services in the economy which is
profitability of firms due to increasing competition in beneficial for domestic consumer but also encourage the foreign
international markets. Productivity can be formalized as the investor to setup their business in the country due to low per unit
capability of the firms to produce goods and services with a cost and higher profit. Expansion of labor productivity is essential
given level of inputs which include land, labor, capital, etc. (Patra for, raising quality of life and improvement of workers’ welfare
& Nayak, 2012). It is very important to improve the efficiency of since raising labor productivity can lead to higher wages and
factor of production in order to increase the economic growth of more investment in human resources (Heshmati & Rashidghalam,
a country. Higher productivity leads to low per unit cost, 2018).
produce higher quality product and higher capability of the Due to rapid globalization, unprecedented development
firms to compete in the international market, and growth in the occurs in the field of information and technology since the
export of the country (Papadogonas & Voulgaris, 2005). Though 1980s. Several developing countries which include China and
there are different aspects of productivity which include labor India get benefits from these modern developments and gain
productivity, capital productivity and total factor productivity, high economic growth. Unfortunately, Pakistan has failed to
but labor productivity is very important because labor is a take advantage from these advancements. Average economic
dominant and active factor used in production process. growth rate during decade of 1980s was 6.5%. Whereas, it
The importance of labor productivity is best elaborated by the remained relatively low in last three decades between 4-5%.
Nobel Laureate Paul Krugman “Productivity isn’t everything, but Labor productivity, grow at 4.2% per annum in 1980s.
in the long run it is almost everything. A country’s ability to Whereas, during 1990s, it decreased to 1.8% and
improve its standard of living over time depends almost entirely continuously dropped to 1.3% during 2000-2015. More
on its ability to raise output per worker”. Labor productivity is an importantly, its growth rate was just 1% during 2007-2015.
influential factor in the competitiveness for both national Labor productivity grew at 2.3% per annum 1980-2015
economies and individual enterprises. Due to rising competition, (Amjad & Awais, 2016).
labor productivity has become influential factor which affect the Dua and Garg (2019) studied determinants of labor
profitability of firm in domestic market as well as in international productivity in countries of the Asia‐Pacific region using panel

221
 Journal of Economic Impact 3 (3) 2021. 221-226

data for the period 1980–2014. The study suggests that from 2009 to 2012. The study revealed that human capital had
foreign direct investment and trade openness have a a positive and significant influence on labor efficiency. Their
significant positive effect on developing countries while trade finding also revealed that education and health played an
openness only affects productivity of developed countries of important role to enhance the labor productivity in Malaysia.
Asia‐Pacific. Moreover, the share of agriculture in GDP affects Fleisher et al. (2011) tried to examine the role of education on
significantly productivity in developing countries only, and worker productivity in China using data of firms for the period
capital deepening has a lower effect for developed countries 1998-2000. Results of the study indicated that the marginal
than developing ones. Samargandi (2018) analyzed productivity of more educated workers increases by large
determinants of labor productivity of MENA countries using margin as compared to less educated. Ismail (2009)
panel data over the period 1980-2014. The study found that investigated the influence of human capital on output growth
human capital, financial development, trade openness and and labor productivity in the manufacturing and service
capital stock are positively related to labor productivity. sectors in Malaysia. Results of the study showed that human
Whereas, compensation is negatively related to labor capital and capital labor ratio had a positive association with
productivity. Asghar et al. (2017) tried to study the role of output growth and labor productivity.
human capital in labor productivity in district Lahore. Cross Given the trends in labor productivity, the objective of this
sectional data has been used and data was collected from study is to find out the determinants of labor productivity in
manufacturing, labor of trading and services sector. The study Pakistan.
indicated that education had a positive association with labor
efficiency. Human capital which includes educational training METHODOLOGY
and skill, increases the labor productivity which ultimately This study explored the determinants of labor productivity in
enhance the firm’s performance. Pakistan. The following empirical model is utilized to
Heshmati and Rashidgalan (2018) analyzed the labor investigate the determinants of labor productivity.
productivity in the manufacturing sector of Kenya. They used 𝐿𝐿𝑃 = 𝛽1 + 𝛽2 𝐿𝑊𝐺 + 𝛽3 𝐿𝐻𝐶 + 𝛽4 𝐿𝐼𝑁𝐹 + 𝛽5 𝐿𝐹 + 𝜇 (1)
cross sectional data and revealed that wage rate, education, All the variables are taken in logarithms form. LLP is log of
training and capital intensity were main determinants and labor productivity; LWG is log of wages; LHC is log of human
positively influenced the labor productivity. A large share of capital investment; LINF is log of Inflation; LF is labor force
female in labor force led to decrease the labor productivity. participation.
Amjad and Awais (2016) focused on the productivity
performance of Pakistan from 1980 to 2015. This study tried Variables Description
to explain those factors which reduce the total factor Labor Productivity: Labor productivity can be defined as
productivity and labor productivity. The results showed that, output per worker. In other words, it measures the efficiency
over these 35 years, the contribution of physical capital and of the country that how inputs are used to produce goods and
education remains modest and there has been a decreasing services. Ratio of real GDP to total employment is used as a
trend in TFP growth. Lack of sustainable growth and measure of labor productivity.
decreasing trend of investment is the major cause of low Wages: Real minimum wage is used as a proxy for real wages
contribution of total factor productivity. Islam et al. (2015) due to the lack of time series data on average real wages. Real
investigated the relationship between labor productivity and wages and labor productivity are closely related to each other.
wage rate in Tanzania and also tried to find out the Higher wages motivate the workers to improve their skills and
determinants of wage rate. The study was based on time series efficiency (Jain, 2019). Number of previous studies showed a
data from 1966 to 2010. Results of the study showed that positive association between these variables.
experience, education and location were the key determinants Human Capital Investment: Human capital is also a key factor
and positively influenced the wage rate. The key finding of the that raises the labor productivity. Human capital theory is
study was that wage rate influenced by other factors rather based on the assumption that education, training and
than workers’ productivity in Tanzania. Granger causality test knowledge increase the marginal physical product of labor.
clarified that real wage only granger caused the workers Real government expenditure on education is used as a proxy
efficiency and positively influenced the productivity of the for human capital investment. A positive relationship between
workers. education and labor productivity is expected.
Yildirim (2015) tried to estimate the relationship among labor Inflation: GDP-deflator is used as a proxy for inflation because
productivity, inflation and wage rate in manufacturing sector it measures the prices of all goods and services produced by
of Turkey over the period from 1988 to 2012. There existed a labor in the country. A negative relationship is expected
strong relationship between inflation and labor productivity. between inflation and labor productivity.
The study suggests that inflation adversely affected the labor Labor Force Participation: Labor force participation is
productivity because it decreased the purchasing power of the measured as the ratio of numbers of the employed labor force
workers and also affected the firm’s investment plans. to total population. Negative relationship is expected between
Inflation had a greater influence on productivity as compared labor force participation and labor productivity.
to wage rate. Moreover, there was a weak association between Macro level annual time series data is used from 1981 to 2018
wage rate and labor productivity. for estimation of the empirical model. Data have been taken
Arshad and Malik (2015) discussed the effects of human from the Handbook of Statistics of State Bank of Pakistan and
capital on labor productivity in Malaysia by using panel data various economic surveys of Pakistan. This study first tests the

222
 Journal of Economic Impact 3 (3) 2021. 221-226

stationarity of all variables using Augmented Dickey and Fuller ∆𝐿𝐿𝑃 = 𝛼0 + ∑𝑛𝑖=1 𝛽1 ∆𝐿𝐿𝑃𝑡−1 + ∑𝑛𝑖=0 𝛽2 ∆𝐿𝑊𝐺𝑡−1 +
(1981) and Phillip and Perron (1988) tests. After testing the ∑𝑛𝑖=0 𝛽3 𝐿𝐻𝐶𝑡−1 + ∑𝑛𝑖=0 𝛽4 ∆𝐼𝑁𝐹𝑡−1 + ∑𝑛𝑖=0 𝛽5 ∆𝐿𝐹𝑡−1 +
stationarity, Autoregressive Distributed Lag Model (ARDL 𝜆1 𝐸𝐶𝑀𝑡−1 + 𝑢𝑡 (4)
hereafter) model is used for empirical model estimation. The study also tests for the serial correlation, heterosckedasticity
ARDL approach (Pesaran et al., 2001) is a standard least square and normality utilizing usual econometric tests.
method that consists of lag terms of dependent and independent
variables. ARDL model can be used whether underlying RESULTS AND DISCUSSION
variables are integrated on level i.e. I(0) or integrated of order Unit Root Tests
one i.e. I(1) or a combination of both but neither of variable The results of the Augmented Dickey-Fuller (ADF hereafter)
should be integrated of order two i.e. I(2). This technique can test and Phillip-Perron (PP hereafter) tests are presented in
also be used for small sample size which give unbiased and Tables 1 and 2, respectively. The stationarity is tested under
efficient results. ARDL is employed in three steps. two models, namely, only intercept and intercept with the
Step 1: The ARDL-ECM model for bound testing is written as trend for the variables of the study. Moreover, numbers in
follow; parentheses show number of optimum lags (k) chosen by
∆𝐿𝐿𝑃 =∝0 + ∑𝑛𝑖=1 𝛽1𝑖 ∆𝐿𝐿𝑃𝑡−1 + ∑𝑛𝑖=0 𝛽2𝑖 ∆𝐿𝑊𝐺𝑡−1 + Schwarz Information Criterion. The estimates of ADF test
∑𝑛𝑖=0 𝛽3𝑖 ∆𝐿𝐻𝐶𝑡−1 + ∑𝑛𝑖=0 𝛽4𝑖 ∆𝐿𝐼𝑁𝐹𝑡−1 + ∑𝑛𝑖=0 𝛽5𝑖 ∆𝐿𝐹𝑡−1 + show that all variables are stationary at the first difference I(1)
𝛽6 𝐿𝐿𝑃𝑡−1 + 𝛽7 𝐿𝑊𝐺𝑡−1 + 𝛽8 𝐿𝐻𝐶𝑡−1 + 𝛽9 𝐿𝐼𝑁𝐹𝑡−1 + 𝛽10 𝐿𝐹𝑡−1 + with both models except inflation which is also stationary at
𝜀𝑡 (2) level with only intercept with trend model but stationary at 1st
Where ∆ is difference operators while 𝜀𝑡 is white noise error difference with intercept model. The finding of PP test, in Table
term. In presence of long run relationship between variables, 2, verifies that all variables are stationary at first difference
i.e., cointegration, following ARDL model is estimated. with intercept and intercept with trend models. No variable is
𝐿𝐿𝑃 = 𝛼0 + ∑𝑛𝑖=1 𝛽1 𝐿𝐿𝑃𝑡−1 + ∑𝑛𝑖=0 𝛽2 𝐿𝑊𝐺𝑡−1 + integrated of order two, i.e., I(2); therefore, ARDL is used for
∑𝑛𝑖=0 𝛽3 𝐿𝐻𝐶𝑡−1 + ∑𝑛𝑖=0 𝛽4 𝐼𝑁𝐹𝑡−1 + ∑𝑛𝑖=0 𝛽5 𝐿𝐹𝑡−1 + 𝑒𝑡 (3) estimation of long run relationship of variables of empirical
Whereas, short run relationship is estimated as follows; model of the study.

Table 1. Augmented Dickey Fuller Test.

At level At 1st difference
Variables Intercept Trend & intercept Intercept trend & intercept

LLP -2.387(0) -2.355(0) -5.592*(0) -5.898*(0)

LWG 2.541(3) -2.386(0) -6.647*(0) -5.380*(2)
LHC -1.086(0) -2.842(0) -4.448*(0) -4.496*(0)
LINF 0.839(0) -4.440*(4) 3.841*(4) -3.544(4)
LF -0.957(0) -1.806(0) -4.413*(0) -4.312*(0)

Table 2. Phillip-Perron Test.

At level At 1st difference
Variables
Intercept Trend & intercept Intercept Trend & intercept
LLP -3.232 -2.345 -5.592* -6.026*
LWG -2.971 -2.225 -6.647* -19.468*
LHC -1.055 -3.029 -4.375* -4.444*
LINF 0.538 -2.057 3.840* -3.857*
LF -0.957 -1.931 -4.413* -4.312*

Autoregressive Distributed Lag Model (ARDL) labor productivity, wage rate, human capital investment,
For ARDL analysis, first we choose an appropriate model from inflation and labor force participation. F-statistic for bound
the alternatives. An Akaike information criterion (AIC) is used testing is presented in Table 3. The results of bound test of
for the selection of optimal lag length for the variables of the cointegration show the existence of cointegration among the
model. The model with the lowest AIC value is considered the variables because the F-Statistic value 5.66 is greater than the
best. ARDL (1, 0, 1, 1, 1) is the best model for this study given upper bound critical value. It neither exists below the lower
the data on variables. Next, bound testing requires to estimate bound (no cointegration) nor lies between the upper and
ARDL-ECM equation (4) to check the cointegration among lower bound (inconclusive zone).

223
 Journal of Economic Impact 3 (3) 2021. 221-226

Table 3. Bound Testing Cointegration.

Bound Critical Values
F-Statistics Value K Significant level Intercept and no trend
I(0) I(1)
10% 2.43 3.57
5.66 4 5% 2.85 4.01
1% 3.83 5.12

After finding the cointegration, the ARDL approach estimates raise the productivity of labor, they offer higher wages than
the long run relationship. The results of long run coefficients market clearing wage. Other reason to offer higher wages is to
are shown in Table 4. The coefficient of wages is positive and minimize the labor turnover, so that firms can retain skilled
statistically significant, which indicates that one percent and experience employees than hiring newly and unskilled
increase in wage rate leads to 0.06% rise in labor productivity. employees. This finding is consistent with the findings of Jain
It verifies the “Efficiency wage theory” which suggested that (2019), Dritsaki (2016), Tang (2014), Kumar et al. (2012),
higher wages influence labor productivity. When firms want to Narayan and Smyth (2009), and Yusof (2008).

Table 4. Estimated Long Run Coefficients.

Variables Coefficient Standard Error t-statistics
LWG 0.068836 0.026243 2.622997
LHC 0.190252 0.028514 6.672265
LINF -0.194216 0.053681 -3.617974
LF -1.029194 0.233913 -4.399907
C 4.749511 0.050193 94.62432

Human capital investment shows a positive relationship with productivity. It may be due to the surplus labor force in the
labor productivity. If there is one percent increase in the workforce, especially in the agriculture sector, that decreases
human capital investment, labor productivity will increase by the workers' efficiency.
0.190 percent. Dua and Garg (2019), Samargandi (2018), Islam The short run results are presented in Table 5. The coefficients
et al. (2015), and Aggrey et al. (2010) reported similar results. of wages and labor force participation shows statistical
Inflation rate has a negative impact on labor productivity. Its significance and signs are also according to economic theory.
coefficient represents that one percent increase the inflation Whereas the coefficients inflation and human capital
rate, reduce the labor productivity by 0.194 percent. Inflation investment have a correct sign but are not statistically
negatively affects labor efficiency because it raises the rental significant in the short run. The coefficient of ECM(-1) is equal
price of capital and firms substitute the labor for capital. to -0.80 suggesting a high speed of the adjustment back to
Ultimately productivity will decrease due to the entrance of equilibrium. This indicates that the 80% deviation from
new workers. Dritsaki (2016), Yildirim (2015), Tang (2014), equilibrium is corrected during a year.
Kim et al. (2013), Kumar at el. (2012) and Papapetrou (2003) Finally, diagnostic tests are applied to check the serial
have reported similar findings. The result shows the negative correlation, heteroskedasticity and normality. The estimates
and significant impact of labor force participation on labor of various test statistics are shown the Table 6.

Table 5. Estimate Error Correction Model.

Variable Coefficient Standard Error t-statistic
D(LWG) 0.055113 0.020836 2.645093
D(LHC) 0.041293 0.040966 1.007976
D(LINF) -0.046099 0.083231 -0.553867
D(LF) -1.421756 0.224677 -6.327996
ECM(-1) -0.800653 0.198095 -4.041768
Akaike Information Criterion= -7.377 R-Square= 0.988
Schwarz Criterion= -6.935 Adjusted R-Square= 0.981
Hannan-Quinn Criterion= -7.259 F-Statistic= 154.83
DW Statistic= 1.99 Prob(F-statistic) = 0.000

Breusch-Godfrey Serial Correlation LM Test has been applied to test is used to test the null hypothesis of no heteroskedasticity, it
the serial correlation. The test reports that p-value is greater than indicates the absence of heteroskedasticity in the model. To identify
the significant level 0.05 and we accept the null hypothesis, which normal distribution of the residuals, Jarque-Bera test is used. It
means that there is no autocorrelation in the model. White LM test shows that the residuals are normally distributed.

224
 Journal of Economic Impact 3 (3) 2021. 221-226

Table 6: Diagnostic Tests

Breusch-Godfrey Serial Correlation LM Test
Null Hypothesis: There is no Serial Correlation
F-Statistic 0.028161 P-Value 0.9723
LM-Statistic 0.103530 P-Value 0.9496
White’s Heteroskedasticity Test:
Null Hypothesis: There is no Heteroskedasticity
F-Statistic 0.737598 P-Value 0.6587
LM-Statistic 0.275767 P-Value 0.5610
Jarque-Bera Normality Test:
Null Hypothesis: Residuals are normally distributed
JB- Statistic 0.466377 P-Value 0.792004

CONCLUSIONS developed countries of Asia‐Pacific. Pacific Econ. Rev.

This study explored the determinants of labor productivity in 24, 686–704.
Pakistan. After testing the stationarity of data, the ARDL model is Fleisher, B.M., Hu, Y., Li, H., Kim, S., 2011. Economic transition,
used for estimation. Results of the study indicate that real wage higher education and worker productivity in China. J.
positivity affects labor productivity because a higher wage rate Dev. Econ. 94, 86–94.
encourages the workers. Human capital investment is also Heshmati, A., Rashidghalam, M., 2018. Labour productivity in
positively related to labor productivity as more educated workers Kenyan manufacturing and service industries, in:
can perform better by adopting the new innovation and Determinants of Economic Growth in Africa. Springer,
technology as compared to uneducated workers. Inflation has a pp. 259–286.
negative effect on labor productivity as it increases the rental Islam, R., Kinyondo, A., Nganga, J., 2015. Real wages and labour
price of capital. Moreover, labor force participation negatively productivity in Tanzania: How do they link? J. African
affects labor productivity due to the prevalence of unskilled labor Stud. Dev. 7, 81–98.
force in the production process. The findings have many Ismail, R., 2009. The impact of human capital attainment on
important policy implications. Most importantly, labor output and labor productivity of Malay firms. J. Int.
productivity can be increased by raising real wages because Manag. Stud. 4, 221–230.
higher wages raise labor productivity, reduce worker turnover, Jain, H., 2019. Wage–productivity relationship in Indian
attract more skilled and educated labor, and increase the feelings manufacturing industries: Evidences from state-
of loyalty among workers. Moreover, the impact of human capital is level panel data. Margin J. Appl. Econ. Res. 13, 277–
undesirably low in spite of an improvement in education over the 305.
years, suggesting that human capital is still not at the level when it can Kim, S., Lim, H., Park, D., 2013. Does productivity growth lower
be a major factor in increasing labor productivity in Pakistan. inflation in Korea? Appl. Econ. 45, 2183–2190.
Therefore, the government should increase the investment in Kumar, S., Webber, D.J., Perry, G., 2012. Real wages, inflation
human capital in order to enhance labor productivity. and labour productivity in Australia. Appl. Econ. 44,
2945–2954.
REFERENCES Narayan, P., Smyth, R., 2009. The effect of inflation and real
Aggrey, N., Eliab, L., Joseph, S., 2010. Human capital and labor wages on productivity: new evidence from a panel of G7
productivity in East African manufacturing firms. Curr. countries. Appl. Econ. 41, 1285–1291.
Res. J. Econ. Theory 2, 48–54. Papadogonas, T., Voulgaris, F., 2005. Labor productivity
Amjad, R., Awais, N., 2016. Pakistan’s Productivity growth in Greek manufacturing firms. Oper. Res. 5, 459–
Performance and TFP Trends, 1980–2015: Cause for 472.
Real Concern. Lahore J. Econ. 21, 33-63. Papapetrou, E., 2003. The Inflation and Productivity
Arshad, M.N.M., Malik, A.Z., 2015. Quality of human capital and Relationship in Poland. J. Econ. Bus. 2003, 11–33.
labor productivity: a case of Malaysia. Int. J. Econ. Manag. Patra, S., Nayak, S.R., 2012. A theoretical study on the
Account. 23, 37-55. relationship between wages and labor productivity in
Asghar, N., Danish, M.H., Rehman, H., 2017. Human Capital and industries. Int. J. Econ. Res. 3, 157–163.
Labour Productivity A Case Study of District Lahore. Pesaran, M.H., Shin, Y., Smith, R.J., 2001. Bounds testing
JPUHS. 30, 163-179. approaches to the analysis of level relationships. J. Appl.
Dickey, D.A., Fuller, W.A., 1981. Likelihood Ratio Statistics for Econom. 16, 289–326.
Autoregressive Time Series with a Unit Root. Phillips, P.C.B., Perron, P., 1988. Testing for a unit root in time
Econometrica 49, 1057–1072. series regression. Biometrika 75, 335–346.
Dritsaki, C., 2016. Real wages, inflation, and labor productivity: Samargandi, N., 2018. Determinants of labor productivity in
Evidences from Bulgaria and Romania. J. Econ. Financ. MENA countries. Emerg. Mark. Financ. Trade 54, 1063–
Stud. 4, 24–36. 1081.
Dua, P., Garg, N.K., 2019. Determinants of labour Tang, C.F., 2014. The effect of real wages and inflation on labour
productivity: Comparison between developing and productivity in Malaysia. Int. Rev. Appl. Econ. 28, 311–322.

225
 Journal of Economic Impact 3 (3) 2021. 221-226

Yildirim, Z., 2015. Relationships among labour productivity, Yusof, S.A., 2008. The long-run and dynamic behaviors of
real wages and inflation in Turkey. Econ. Res. wages, productivity and employment in Malaysia. J. Econ.
Istraživanja 28, 85–103. Stud. 35, 249–262.

Publisher’s note: Science Impact Publishers remain neutral with regard to jurisdictional claims in published maps and institutional
affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and
indicate if changes were made. The images or other third-party material in this article are included in the article’s Creative Commons
license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license
and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission
directly from the copyright holder. To view a copy of this license, visit https://creativecommons.org/licenses/by/4.0/.

226