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

net/publication/366824540

UNEMPLOYMENT AND OUTPUT DYNAMICS IN NIGERIA

Article · January 2023

CITATION READS

1 199

3 authors, including:

Aminu Umaru Mukhtar W Lawan

Central Bank of Nigeria Modibbo Adama University of Technology
39 PUBLICATIONS 497 CITATIONS 5 PUBLICATIONS 10 CITATIONS

SEE PROFILE SEE PROFILE

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

Sectoral nexus View project

Globalization and economic growth View project

All content following this page was uploaded by Aminu Umaru on 03 January 2023.

The user has requested enhancement of the downloaded file.

MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 19

UNEMPLOYMENT AND OUTPUT DYNAMICS IN NIGERIA

BY
AMINU UMARU1,
MUKHTAR WAKIL LAWAN2, and ABUBAKAR, MUHAMMED
BELLO3
1,2,3
Lecturers, Department of Economics
School of Management and Information Technology,
Modibbo Adama University of Technology, Yola, Adamawa State, Nigeria
E-Mail: [email protected], [email protected],
[email protected]

Abstract
This paper examines unemployment and output dynamic sin Nigeria through the use of unit root test,
cointegration and error correction techniques after which the ordinary least square technique was
employed. Data for the paper was data sourced from CBN statistical bulletin and National Bureau of
Statistics annual report for the period 1986-2014, the paper started by estimating the linear Okun’s
model using the transitory and permanent components of the unemployment and output series. Results
from the linear estimates suggest that the short-run relationship between output and unemployment is
negative particularly in agriculture, building, construction and trade, but that the long-run
relationship in all the sectors is positive. This result therefore supported the non-linearity hypothesis
in the dynamic relationship between unemployment and output. The paper therefore estimated a non-
linear Okun ‘s type model using the Generalised Method of Moments (GMM). The results confirmed
that the dynamic relationship between unemployment and output in Nigeria is non-linear and hump-
shaped. The threshold rate of unemployment was 16.76%, this implies that at unemployment rates
below the of 16.76%, the relationship is positive, and becomes negative at higher unemployment
rates. The paper therefore established that when unemployment rate is below the 16.76% threshold,
the economy experiences jobless growth. It was also found that output increases in the agriculture,
building, construction and trade sectors have the potential for employment generation but industrial
and service sectors all of which have no potential for employment generation. The paper therefore
recommended that policies that aim at raising output in these sectors would be more efficient in
reducing unemployment in Nigeria.
Key words: Unemployment, output, Okun’s law, dynamics, sectors

1.0 INTRODUCTION
Nigeria as a nation is endowed with huge human and natural resources. The country is richly
endowed with various mineral types all over the country. Huge amount of income is
generated annually from petroleum products. According to Musa (2010) more than 40 types
of solid minerals have been identified in over 500 locations in Nigeria. Despite Nigerian
huge human and natural resources the country remained largely underdeveloped. There are
numerous socio-economic challenges facing the country. The Nigerian economy has been
witnessing recovery which is immediately followed by recession and depression.

This situation in Nigeria is becoming disturbing. There are various macroeconomic policies
put in place by the government of Nigeria over the years but have been unable to achieve
sustained reduction in unemployment and sustained growth cannot be achieved. The poor
performance of the economy has engendered the need to manage the economy effectively.
The essence of macroeconomic management underlines the need for the existence of
government as a vital agent for economic development. However, it appears that government
intervention in Nigeria has not been able to solve the ills in the economy.
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 20

The perpetual economic crisis, with the attendant problems of high inflationary pressure, high
and unstable exchange rate, and high debt profile and deficit balance of payment is difficult
to explain. Against a high rate of unemployment and underutilization of resources, a large
public sector, low wages and poor working conditions, there has been an also persistent high
inflation rate in Nigeria. Also, unemployment and underemployment is prominent feature of
the Nigerian economy. Consequently, this undermines the full potentials of labour-surplus
economy where the labour surpluses have not been fully exploited.

Nigerian economy provided jobs for most Nigerian and absorbed a considerable number of
imported labour in the 1960s and early 1970s. The wage rate in the country compared
favourably with international standards and there was relative industrial stability in most of
the years. With the oil boom of the late 1970s, coupled with mass migration of people,
especially the active population (youth), to the urban centers seeking for white cola jobs.
Following the economic downturn in the early 1980s, the problems of unemployment
increased, this precipitated the introduction of the Structural Adjustment Programme (SAP).
The rapid depreciation of the naira exchange rate since 1986 and the inability of most
industries to obtain adequate raw materials required to sustain their output levels fuelled
inflation. There was rapid depreciation of the naira this situation caused sharp rise in the
general price level, leading to a considerable decline in real wages and increased poverty. The
low wages contributed to a weakening of the purchasing power of wage earners and declining
aggregate demand. Consequently, industries started to accumulate unintended inventories.
The growth of the Nigerian economy has been poor since the introduction of SAP (Structural
Adjustment Programme) in 1986. The continuous economic crisis reflected in high rates of
unemployment, high inflation rates, high level of corruption, exchange rates distortions, debt
overhang, to mention a few. Unemployment as a phenomena had eaten deep into the fabric of
the Nigerian economy over the years.

The trends in unemployment and output growth rates in Nigeria from 1986-2014 have been
puzzling. The data obtained from the Central Bank of Nigeria (CBN), 2014 Statistical
bulletin revealed that by 1986 economic growth rate stood at 3.1 percent, in 1987 the value
became negative -0.69 implying retrogression and was the least ever achieved for the period
under review; the highest economic growth rates achieved was 11.36 in 1990 after which the
rates had been abysmally lower until in 2003 when the growth rates hits 10.2 percent; from
2003 economic growth rate has been less than 10 percent, in 2012 the growth rate recorded
was 6.58. The trend in economic growth has been fluctuating over the years under review.

The trends in unemployment in Nigeria from 1986-2014 was also puzzling. The trend
revealed that by 1986 unemployment rate was 5.3 percent.

Based on the CBN data output and unemployment grow simultaneously over the years which
revealed that there is evidence of jobless growth in Nigeria. It is assumed theoretically that
when output is growing unemployment will reduce as well and if unemployment is increasing
output level will reduce. But this is not the case in Nigeria because as unemployment
increases output level also increases (Aminu, Manu & Salihu (2013). This study seeks to
examine the causes of such anti-theoretical behavior in Nigeria with the aim of finding
solution to the problem.

Studies conducted by (Garba, 2010, and Olowononi and Audu (2012), had examined the
nature and causes of unemployment in Nigeria and found disturbing trends. There are very
few studies which have been undertaken regarding unemployment and output dynamics in
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 21

Nigeria. Some of the existing studies employed descriptive statistics (see Olowononi and
Audu, 2012). Aminu and Anono, (2012), Bakare, (2012) and Rafindadi, (2012) conducted
similar studies and their findings were controversial especially in the area of the impact of
unemployment on the economic growth in Nigerian. A study conducted by Bakare (2012)
found negative relationship between unemployment and growth, Rafindadi (2012) also found
negative non-linear relationship between unemployment and output growth, Another study
was also conducted in the same vein in China by Chang-Shuai Li and ZI-Juan Liu (2012) on
unemployment rate, economic growth and inflation. The results revealed that unemployment
impacted negatively on growth while inflation impacted positively on growth in China. while
Aminu, Manu and Salihu (2013) found positive relationship between unemployment and
economic growth in Nigeria. The puzzling trends of economic growth rate and
unemployment rate in Nigeria and the controversial results obtained in the empirical results
provide the need to examine the relationship between unemployment and output in Nigeria.

2.0 Review of theoretical literature

This section is devoted to the review of existing literature on past studies on unemployment
and output dynamics in Nigeria.

According to Jhingan (2001) Economists distinguish between various types and theories of
unemployment; they include cyclical or Keynesian unemployment, frictional unemployment,
structural unemployment and classical unemployment. Some additional types of
unemployment that are occasionally mentioned are seasonal unemployment, hardcore
unemployment, and hidden unemployment; The U.S. Bureau for Labour Statistics measures
six types of unemployment. Though there were several definitions of voluntary and
involuntary unemployment in the economics literature. According Todaro (1992) Voluntary
unemployment was attributed to the individual’s decisions. Involuntary unemployment exists
because of the socio-economic environment (including the market structure, government
intervention, and the level of aggregate demand) in which individuals operate. In these terms,
most of frictional unemployment is voluntary, since it reflects individual search behaviour.
Voluntary unemployment includes workers who reject low wage jobs which is common in
Nigeria whereas involuntary unemployment includes workers fired due to an economic crisis,
industrial decline, company bankruptcy, or organizational restructuring. On the other hand,
cyclical, structural and classical unemployment are largely involuntary in nature. However,
the existence of structural unemployment may reflects choices made by the unemployed in
the past, while classical (natural) unemployment may result from the legislative and
economic choices made by labour unions or political parties.

According to Jhingan (2001) Frictional unemployment is the time period between jobs when
a worker is searching for or transitioning from one job to another. It is sometimes called
search unemployment and can be voluntary based on the circumstances of the unemployed
individual. Frictional unemployment is always present in an economy, so the level of
involuntary unemployment is properly the unemployment rate minus the rate of frictional
unemployment, which means that increases or decreases in unemployment are normally
under-represented in the simple statistics.

According to Adebayo (1992) Cyclical or Keynesian unemployment, also known as

deficient-demand unemployment, occurs when there is not enough aggregate demand in the
economy to provide jobs for everyone who wants to work.
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 22

Structural unemployment occurs when a labour market is unable to provide jobs for everyone
who wants one because there is a mismatch between the skills of the unemployed workers
and the skills needed for the available jobs. Structural unemployment is hard to separate
empirically from frictional unemployment, except to say that it lasts longer.

Technological unemployment is due to the replacement of workers by machines, might be

counted as structural. Alternatively, technological unemployment might refer to the way in
which steady increases in labour productivity mean that fewer workers are needed to produce
the same level of output every year.

He said that hidden or covered unemployment is the unemployment of potential workers that
is not reflected in official unemployment statistics, due to the way the statistics are collected.
In many countries only those who have no work but are actively looking for work (and
qualifying for social security benefits) are counted as unemployed.

The Okun’s Law

According to Okun’s (1962) as unemployment fall by 1%, output has to grow by 3%,
implying that output and unemployment do not move one-for-one. There are a number of
reasons why output and unemployment don’t move one-for-one. The measured
unemployment can be expected to be less volatile than output for a number of reasons. First,
in the face transitory fall in output, firms would prefer to allow workers to vary hours worked
rather than lay-off some. This is partly because it is costly to hire and train another set of
workers when normal orders return. In addition, as output begins to rise and firms begin to
employ, the participation rate may rise as workers who have lost hope and left the labour
market return (i.e., as the hysteresis effects dissipate). In addition, prevalence of
underemployment would also require large increase in output for employment to rise.

Secondly, firms in countries that have stronger labour market regulations or strong unions
would find it difficult to lay-off workers in the face of transitory fall in output. Output in this
case will be more volatile than unemployment, and such countries will tend to have larger
Okun’s coefficient. In general therefore, the greater the labour market flexibility, the smaller
the Okun’s coefficient will be. The size of this coefficient would vary over time because the
relationship between output and unemployment depends on laws, technology, demographics
and preferences (see Blanchard, 2006).

Output and Unemployment: Non-linearity

Malley and Molana (2007), postulated that the labour market is imperfectly competitive, and
that firms reward workers effort by paying them efficiency wage benefit. They also assume
that there are two states of workers efforts: State zero is characterised by “low effort”,
therefore workers are not putting their best, their efficiency, and productivity and output is
below optimal level. State on is characterized by “high effort”. In this state workers are
putting their best, their efficiency and productivity is optimal, hence, there will be no room
for improvements.

Malley and Molana (2007) stated that from State zero, when unemployment rate rises in the
economy, either because of the efficiency wage premium giving to workers, or due to fear of
losing their jobs, workers will begin to raise their productivity by putting more effort, this
will lead to increase in output in the face of rising unemployment. To them if the net increase
in output as a result of increase in per-worker efficiency is greater than the output loss due to
the rising unemployment, then increase in output will be attributed to rising unemployment.
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 23

This is possible because as output decreases firms will disengage the least efficient workers,
therefore, output loss may be less than output gains as a result of increase in workers
efficiency. Finally, Malley and Molana (2007) concluded the link between unemployment
and output negative.

Review of empirical literature

Here empirical literature on the relationship between unemployment and output were
reviewed. Some studies found a negative relationship between unemployment and output; the
like of Bakere (2012) conducted a study on stabilization policy, unemployment crises and
economic growth in Nigeria. He used OLS and found that the relationship between inflation,
unemployment and economic growth in Nigeria were negative. In the same vein Rafindadi
(2012) conducted a study on the relationship between output and unemployment dynamics in
Nigeria; and used OLS and Threshold model. He found a negative nonlinear relationship
between output and unemployment. Chang-Shuai Li and ZI-Juan Liu (2012) also conducted a
study on the relationship between Chinese unemployment rate, economic growth and
inflation; they used Granger causality test, unit root, cointegration, VAR and VEC model.
The study revealed that unemployment relate to growth negatively while inflation affected
growth positively in China. The study also revealed that there was no causation existed
between unemployment and inflation, but there was causation existing between
unemployment and growth, while two-way causation was found existing between inflation
and growth.

However some studies found positive relationship between unemployment and output
especially in Nigeria. Studies such as Arewa and Nwankanma (2013) investigated the
potential real GDP relationship and growth process of Nigerian economy: An empirical re-
evaluation of Okun’s law. They used VAR mechanism and found a positive relationship
between output gab and unemployment gab in Nigeria. Another study by Aminu, Manu and
Salihu (2013) investigated the impact of unemployment and inflation on economic growth in
Nigeria. They used OLS, Augmented Dickey-Fuller technique, Granger causality and
Johansen cointegration technique and found that both unemployment and inflation affected
economic growth positively in Nigeria. In the same vein Mahmoud (2013) conducted a study
on the impact of inflation and unemployment on Jordanian GDP. He employed simple
correlation coefficient and ANOVA and found that inflation impacted positively on Jordanian
GDP, while unemployment impacted negatively on Jordanian GDP. This controversy may be
attributed to the dominant manifestation of unemployment in Nigeria.

3.0 Methodology
This paper used multi-dimensional econometric procedure in estimating the effect of
unemployment and output dynamics in Nigeria. The Ordinary Least Squares (OLS)
techniques, and double log were employed to obtain the coefficients of the equation, the
double log technique was used in estimating the elasticities of various sectors on
unemployment, Augmented Dickey-Fuller tests was employed to test the unit root properties
of the series, after which Johansen cointegration test was employed to test the presence of
long-run relationship between output of various sectors of the Nigerian economy and
unemployment.

Models specification
This paper adopted model of growth by Aminu and Anono (2012) and incorporate
unemployment. The models estimated were specified as follows:-
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 24

LINEAR FORM
Model I: Response of unemployment to Aggregate GDP (Jobless growth)
UN = f(RGDP) ------------------------------------------------------------------------------------ (1)
Where RGDP is the gross domestic product and UN is unemployment rate. The model was
specified as follows
UN = α0+ α1RGDP + µ1 ------------------------------------------------------------------------ (2)

A piori expectation
It is expected that α0> 0 and α1< 0.

Model II:The Non-Linear Model: The Threshold model

To examine the empirical relationship between Unemployment and Output in Nigeria and
determine whether the relationship between Unemployment and Output in Nigeria has the
postulated threshold effect, and is hump-shaped, this paper adopted Malley and Molana
(2007) to estimate a non-linear model of the relationship between de-trended output and
unemployment using the Generalised Method of Moments (GMM). The model is specified
as:
𝑅𝐺𝐷𝑃 = 𝛽0 + 𝛽1 𝑈𝑁𝑡 + 𝛽1 𝑈𝑁𝑡2 + 𝜀𝑡 ……………………………………………. (3)
Where RGDP is the de-trended output, UN is the unemployment rate. As shown below, the
alternating sign of the two parameters, 𝛽1> 0 and 𝛽2< 0 will suggest that the relationship
between output and unemployment is hump-shaped: positive in UN, but negative in UN2. The
threshold level of unemployment (UNo) below which the relationship between
unemployment and output is positive can be estimated using the following expression (see
Malley and Molana, 2007)
𝛽
𝑈𝑁0 = − 𝛽1 ……………………………….. (4)
2

Model III: Relative Response of Unemployment to Sectoral Output

UN = f(AGRIC,IND,BUDC, SERV, WHOLS) ----------------------------------------- (5)
𝑈𝑁 = 𝛽 + 𝛽1𝐴𝐺𝑅 + 𝛽2𝐼𝑁𝐷 + 𝛽3𝐵𝑈𝐶 + 𝛽4𝑆𝐸𝑅 + 𝛽5𝑇𝑅𝐴 + 𝜇 − − − (6)
The paper estimated the linear model (Equation 3) using the permanent and transitory
components of sectoral output (Table 2). This allows us to; first, compare the relative
employment potential of the five sectors, after which the estimates of the Okun’s Coefficient
for the five sectors of the economy are also reported.

3.4 Diagnostic Test

The diagnostic tests which this paper employed were Augmented Dickey-Fuller (ADF), and
Johansen cointegretion test.
Augmented Dickey-Fuller Test.
The model of unit root is specified as follows
Δ𝑅𝐺𝐷𝑃𝑡 = ϕ𝑅𝐺𝐷𝑃t-1 + 𝜖𝑡 − − − − − − − 3.6.1𝑎
Δ𝑈𝑁𝑡 = ϕ𝑈𝑁t-1 + 𝜖𝑡 − − − − − − − − − 3.6.1𝑏
Δ𝐴𝐺𝑅𝑡 = ϕ𝐴𝐺𝑅t-1 + 𝜖𝑡 − − − − − − − − − 3.6.1𝑐
Δ𝐼𝑁𝐷𝑡 = ϕ𝐼𝑁𝐷t-1 + 𝜖𝑡 − − − − − − − − − 3.6.1𝑑
Δ𝐵𝑈𝐶𝑡 = ϕ𝐵𝑈𝐶t-1 + 𝜖𝑡 − − − − − − − − − 3.6.1𝑒
Δ𝑆𝐸𝑅𝑡 = ϕ𝑆𝐸𝑅t-1 + 𝜖𝑡 − − − − − − − − − 3.6.1𝑓
Δ𝑇𝑅𝐴𝑡 = ϕ𝑇𝑅𝐴t-1 + 𝜖𝑡 − − − − − − − − − 3.6.1𝑔

Decision Rule: The null hypothesis ϕ = 1, i.e. a unit root exist in RGDP, UN, AGR, IND,
BUC, SER and TRA (RGDP, UN, AGR, IND, BUC, SER and TRA are non-stationary) but
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 25

when ϕ < 1, i.e. a unit root does not exist in RGDP, UN, AGR, IND, BUC, SER and TRA
(RGDP, UN, AGR, IND, BUC, SER and TRA are stationary) . The decision rule as to
whether to accept the null hypothesis or not is that ADF statistics should be less than critical
t-value at certain percent level, and hence unit root exist; but if ADF statistics is greater than
the critical t-value at certain percent, then the null hypothesis is reject, hence, there is no unit
root and RGDP is stationary. This is similar to all the variables of the model.

4.1 Results and discussion

Here data for the paper were presented and analyzed.

TABLE 4.1 DESCRIPTIVE STATISTICS

VARIABLES UN AGR IND BUC SER TRA
Mean 2.048461 7.289200 7.573656 4.136784 6.075014 6.312318
Median 2.468100 7.496364 8.021227 4.078554 6.153520 6.438743
Maximum 3.173878 9.640847 9.682359 7.851066 10.42614 9.661679
Minimum 0.587787 3.687128 4.122284 1.313724 2.911263 2.386007
Std. Dev. 0.855060 1.835138 1.725423 1.777894 2.020747 2.025710
Skewness -0.184056 -0.495713 -0.546828 0.007325 0.019416 -0.348243
Kurtosis 1.484084 2.004635 2.080440 2.035149 2.074576 2.051540

Jarque-Bera 2.940489 2.384860 2.467022 1.125142 1.036651 1.673141

Probability 0.229869 0.303483 0.291268 0.569742 0.595517 0.433194

Sum 59.40537 211.3868 219.6360 119.9667 176.1754 183.0572

Sum Sq. Dev. 20.47155 94.29644 83.35839 88.50540 114.3358 114.8980

Observations 29 29 29 29 29 29
Source: E-views 7

Table 4.1 shows that unemployment, agricultural output, industrial output, contribution of
building and construction to GDP, output of service sector and output of trade and commerce
in Nigerian were normally distributed as indicated by their Jarque-Bera values of 2.940489,
2.384860, 2.467022, 1.125142, 1.036651, and 1.673141 respectively and the probability
values of 0.229869, 0.303483, 0.291268, 0.569742, 0.595517 and 0.433194 respectively. The
null hypothesis that the data is normally distributed is accepted at 5 per cent significant level.
This implies that the data can be use for estimation.

4.2 EMPIRICAL RESULTS AND DISCUSSION

Table 4.2.: ADF Unit Root Results
Variables Level 5% Critical Value First Difference 5% Critical Value Remark
GDP -1.6611 -3.5742 -3.6116 -2.9678 1(1)
UNEMPL -1.6563 -3,5684 -4.6938 -2.9678 1(I)
GDP_AGRIC 1.3726 -1.9525 -25.7093 -2.9678 1(1)
GDP_IND 0.9274 -1.9525 -29.2177 -2.9678 1(1)
GDP_BUC -2.0859 -3.5684 -9.8655 -2.9678 1(1)
GDP_TRA 4.0789 -1.9529 -1.9488 -1.9534 1(1)
GDP_SER -2.3983 -2.9640 -14.9369 -1.9529 1(1)

4.2. Discussion of unit root test results

The results of unit root test were contained in table 4.2.The results of ADF revealed that all
the variables of the model were stationary at 1percent as indicated by their probability values.
The result further indicated that output of various sector of the economy and unemployment
were stationary at first difference 1(1). The ADF statistics for all the variables are less than
the critical values in negative direction.
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 26

4.3: Discussion of cointegration results

The Johansen cointegration test results contain in appendix (ii) confirmed the existence of
long-run relationship between unemployment and output of various sectors of the economy
(economic growth), as indicated by the TRACE-Statistic. The TRACE-statistics results
revealed that there was 3 cointegrating equation at 5 per cent level. The Max- Eigen-statistics
results revealed that there was 1cointegrating equation at 5 per cent level. Overall, these
results are in agreement with similar study on Nigeria conducted by Aminu and Anono
(2012) on the long run relationship between unemployment, and economic growth in Nigeria.

Table 4.3: Error Correction Model (ECM) Parsimonious Results

Dependent Variable: D(UN)
Method: Least Squares
Sample (adjusted): 1989 2014
Included observations: 26 after adjustments

Variable Coefficient Std. Error t-Statistic Prob.

C 0.824419 0.211063 3.906043 0.0011

D(UN(-1)) -0.241918 0.217010 -1.114780 0.2805
D(AGR(-2)) -0.788867 0.574673 -1.372722 0.1877
D(IND(-1)) 0.300116 0.234971 1.277249 0.2187
D(IND(-2)) -0.297544 0.213647 -1.392690 0.1817
D(BUC(-1)) -3.436898 0.932745 -3.684713 0.0018
D(SER(-1)) 1.253258 1.239258 1.011297 0.3261
D(TRA(-1)) -0.752856 0.896324 -0.839937 0.4126
ECM(-1) -0.705852 0.194757 -3.624263 0.0021

R-squared 0.679825 Mean dependent var 0.054918

Adjusted R-squared 0.529155 S.D. dependent var 0.407327
S.E. of regression 0.279501 Akaike info criterion 0.555799
Sum squared resid 1.328049 Schwarz criterion 0.991294
Log likelihood 1.774608 Hannan-Quinn criter. 0.681206
F-statistic 4.512003 Durbin-Watson stat 2.367236
Prob(F-statistic) 0.004364

4.3: Discussion of ECM results

The coefficient of ECM in Table 4.3 shows the speed of adjustment to the deviation in the
long run equilibrium. The negative value of the coefficient of ECM implies that there is a
long run relationship between unemployment and the output of various sectors of the
Nigerian economy. The value of the coefficient of ECM (-0.705852) shows that the model
will adjust by 70.59 percent to go back to equilibrium in the long run. Precisely, -0.705852
implied that when there is a state of disequilibrium between unemployment and the output of
various sectors will be brought back to equilibrium in at least two years’ time. The R2 value
of 0.6798 shows that 67.98 per cent variation in unemployment was explained by output of
various sectors of the Nigerian economy; this further implies that the model of this study is fit
to explain the relationship between unemployment and the output of various sectors. The
Durbin Watson statistics value of (2.367236) shows the absence of autocorrelation; hence the
model is non-spurious and can be used for policy purpose.
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 27

Table 4.4: Estimates of the Linear Model and the Okun’s Law Coefficients for the
Various Sectors
∆𝑼𝒕 = 𝑪 + 𝜷∆𝑮𝑫𝑷𝒕 + 𝜺𝒕 Short-Run Long-Run
Okun’s Law Okun’s Law
Sectoral Output C 𝜷 Coefficient C 𝜷 Coefficient
(𝟏⁄𝜷) 𝟏⁄ )
𝜷
0.37 0.94 0.36 -0.01 -
RGDP 1.06 100
Agricultural 0.37 -0.51 -1.96 0.04 -0.01 -100
Industrial 0.37 0.64 1.56 0.15 -0.05 -20
Building and Constr 0.37 -1.86 -0.54 0.06 -0.02 -50
Services 0.37 1.01 0.99 0.13 -0.04 -25
Wholesale and Retail 0.37 -0.80 0.10 -0.03 -
-1.25 33.33
4.4. Relative Response of Unemployment to Sectoral Output
The estimated the linear model (Equation 3) using the permanent and transitory components
of sectoral output (Table 4.4). This allows us to, first, compare the relative employment
potential of the five sectors and, second, compare the short-run (columns 2-4) and long-run
(columns 5-7) behaviours of unemployment in response of output changes. Estimates of the
Okun’s Coefficient for the five sectors of the economy are also reported.

From Table 4.4, it can be observed that the short-run relationship between unemployment and
some sectoral output such as agricultural output, building and construction and trade and
commerce is negative, consistent with the theoretical expectation but the relationship between
unemployment and industrial output and service sector is positive and inconsistent with
theoretical expectation. This result implies that agricultural sector, building and construction
and trade and commerce are the only sector having the potential of reducing unemployment
in Nigeria while industrial sector and service sector have no potential for reducing
unemployment in Nigeria. The long-run relationship between unemployment and various
sectors of the Nigerian economy is however negative and consistent with theoretical
expectation. This paper revealed that in the long run all the five sectors of the Nigerian
economy have potential for reducing unemployment in Nigeria.

In terms of relative employment potential, larger β’s measure the elasticity of unemployment
with respect to respective sectoral output. In the short-run, the building and construction
appears to have greater unemployment-reducing potential follow by wholesale and retail
trade and the least employment- generating sector appears to be agriculture which is
attributed to the neglect of the sector and shifting to industrial and service that have no
potential for reducing unemployment in the country. The F-statistics value of 22.25767,
which measured the joint significance of the parameter estimates, was found statistically
significant at 1 per cent level as indicated by the corresponding probability value of
0.000000. This implies that all the variables of the model were statistically jointly and
significantly affected unemployment rate in Nigeria.

The R2 value of 0.8586 (85.86%) implied that 85.86per cent total variation in unemployment
is explained by the output of various sectors of the Nigerian economy. This indicated that
unemployment was significantly explained by the included variables and the model is fit in
explaining the relationship between the variables under consideration. Coincidently, the
goodness of fit of the regression remained very high after adjusting for the degree of freedom
as indicated by the adjusted R2 (R2 =0.8200 or 82.00%). The Durbin-Watson statistic 1.5078
in appendix i was observed to be higher than R2 0.8586 indicating that the model is non-
spurious (meaningful) therefore, it can be used for policy purpose. The Durbin-Watson
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 28

statistics 1.5078 shows that there was negligible presence of autocorrelation among the error
value because the value is tending toward 2.

This finding appears to contradict the general impression that agriculture and service sectors
have greater potential to generate employment than other sectors (e.g., Njoku 2011) and
(Rafindadi 2012) respectively. The present study shows that building and construction has the
relatively greater potential for reducing unemployment in Nigeria.

The implication of this finding is that if the agricultural, building and construction and
wholesale and retail trade were the main drivers of growth, the potential of jobless-growth
will be reduced.

4.5 Estimate Non-Linear Model: Evidence of Threshold Effect

The summary of GMM estimates of the non-linear model which is represented as:
RGDP = 1.3722 + 0.5968U - 0.0178U2
t-stat (1.0412) (2.0301) (2.4745)
Prob. Value of F-statistic = 0.0174, Prob. Value of j-statistic = 0.00000

The result above revealed that all the coefficients of the model have the expected signs and
are statistically significant as indicated by probability value of F-statistics of 0.0174. The J-
statistic for Hansen’s was used to test for validity of the overriding restriction and the
probability of j-statistic = 0.00000 suggested that the restrictions are valid at 5 per cent level
of significant.

The results therefore suggest that as output rises, unemployment will initially rise until a
certain threshold level of unemployment is reached when unemployment will begin to fall.
This evidence is also consistent with our earlier results from the linear model, which suggest
that the correlation between the long-run (permanent components of) output and
unemployment is positive, while the short-run correlation is negative. This is because, in the
long-run, unemployment is likely to be low as the economy is close to full employment. As
unemployment rises from such low levels, workers supply of higher effort will raise
productivity and the resultant output gain due to efficiency could be sufficiently large to
dominate the fall in output as a result of lay-offs.
The threshold level of unemployment below which the relationship is positive is calculated as
(see Malley and Molana, 2007):
𝛽 0.5968
U0= - 2𝛽1 = - 2(−0.0178) = 16.76
2

The result above shows the threshold rate of unemployment at which the relationship
between unemployment and output switches from positive to negative is about 16.76%. It is
the rate of unemployment below which the Okun’s law does not hold, but above which the
Okun’s law holds. This rate is close enough to the 10.59% found from the linear models of
the short-run and long-run relationship. The threshold range is therefore 10.59% - 16.76%
unemployment rate.
4.5 Dynamics Analysis using variance decomposition and impulse response
Impulse response function shows the effects of shocks on the adjustment path of the
variables. Forecast error variance decompositions measure the contribution of each type of
shock to the forecast error variance. Both computations are useful in assessing how shocks to
economic variables reverberate through a system.
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 29

Appendix iii shows the variance decomposition of all the variables of the model.
Unemployment responses 100 per cent to shock in itself but the response of other variables to
shock in unemployment is not significant and die out immediately. Unemployment response
more to shock in industrial sector follow by agricultural sector, service sector, trade and
commerce and lastly building and construction. The response of unemployment to shock in
agricultural sector is almost 27 per cent, to industrial sector is almost 32 per cent, to service
sector is almost 27 per cent, to shock in trade is almost 24 percent but to shock in building
and construction is almost 11 per cent. This implies that unemployment response more to
shock in itself follow by industrial sector.

5.0 Conclusion and recommendations

This study empirically investigates unemployment and output dynamics in Nigeria from 1986
to 2014. The data was sourced through CBN statistical bulletin and National Bureau of
Statistics annual report. The major findings of this study are that, first, the relationship
between unemployment and output shows a nonlinear behaviour. The relationship is positive
at low level of unemployment with a threshold level of about 16.76%, beyond which the
relationship changes from positive to negative. This implies that in the long-run when
unemployment rates is low, the relationship between unemployment and output will be
negative and rising unemployment rates will be followed by falling output as the employed
workers put less effort. At unemployment rates below the 16.76% threshold, output growth is
associated with rising unemployment rates. The policy implication of this finding is that
when the unemployment rate is below 16.76% positive government policies that are designed
on the basis of the traditional Okun’s law will have adverse effect on employment, instead of
the policies to create employment, they may end up increasing unemployment. This study
therefore recommended that policies that aim at reducing labour market imperfections would
be more effective in reducing unemployment when the rate of unemployment is below the
threshold (or period of output growth alongside increase in unemployment). Finally, given
the relative response of unemployment to changes in sectoral output, the findings of this
study suggest that the Agricultural sector, building and construction and wholesale and retail
trade are the only sectors having the greatest potential for reducing unemployment.
Employment policies can be more effective if they are directed at raising outputs in these
sectors.
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 30

REFERENCES
Adofu, I. (2010), ‘‘Accelerating Economic Growth in Nigeria, The role of Foreign Direct
Investment’’.Current Research Journal of Economic Theory.2(1),11-15.

Ajao, W. (2004)Neglect of technical, vocational education increases youth unemployment.

The Vanguard. 23 December, pp. 23.

Amin, A.A. (1998). Cameroon’s Fiscal Policy and Economic Growth.African Economic
Research Consortium’s Research Paper 85, Nairobi (November).

Aminu, U. and Anono, A.Z. (2012).An empirical Analysis of The Relationship between
Unemployment and Inflation in Nigeria from 1977-2009.Business Journal,
Economics and Review, Global Research Society. Pakistan. 1 (12), 42-61.

Aminu, U. and Manu, D. and Salihu, M. (2013). An Empirical Investigation into the Effect
of Unemployment and Inflation on the Growth in Nigerian.Interdisplinary Journal of
Research in Business. 2(12)

Anyanwu, J.C. (1997) Modern macroeconomic: theory and applications in

Nigeria.Onitsha.Joanee Educational publishers Ltd.

Arewa, A. and Nwakanma, P.C. (2013) Potential-Real GDP relationship and Growth process
of Nigerian Economy: An Enpirical Re-evaluation of Okun’s Law. European
Scientific Journal. 8(9) 25-33

Aschauer, D.A (1989) ‘‘Is Government Expenditure Productive?’’ Journal of Monetary

Economics (March).

Ayesha, W. and K. Rukhsana (2010) ‘‘The Impact of Inflation and Economic Growth on
Unemployment’’: Time series Evidence from Pakistan.Proceedings of 3rd
international Conference on Business Management (ISBN: 978-969-9368-07-3).

Bakare, A.S. (2012) Stabilization policy, Unemployment Crises and Economic Growth in
Nigeria.Universal Journal of Management and Social Sciences.2(4).

Balami, D.H (2006). Macroeconomic Theory and Practice.Salawe prints, Off Leventies,
Wulari, Maiduguri.

Brian, S. and V.R. Howard (2005) Modern Macroeconomic, Its Origins, Development and
Current State.Edward Elgar Publishing Limited. UK.

Chang-shuai, L. and .Zi-juan, L. (2012) The relationship among Chinese unemployment rate,
Economic growth and Inflation.Advances in Applied Economics and Finance, Vol.1,
No.1.World science Publishers.United states.

Dickey, D.A. and W.A. Fuller (1979), “Distribution of the Estimators for Autoregressive
Time Series with a Unit Root,” Journal of the American Statistical Association, 74,
427-431

Domar, E.D. (1948) ‘‘The problem of Capital accumulation’’.American Economic Review,

December.
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 31

Ekpo, A.H. (1995) ‘‘Public Expenditure and Economic Growth in Nigeria,1960-1992.’’ Final
Report, AERC, Nairobi, Kenya.

Emeka, C. (2013) ‘‘Nigeria’s economy; inflation and other economic

indicators’’[email protected].

Engle, R.F. and C.W.J. Granger (1987), “ Cointegration and Error Correction:
Representation, Estimation and Testing,”Econometrica, 55, 251-276

Evans, P. and Karas, G. (1994) ‘‘Are Government Activities Productive? Evidence from a
Pannel of U.S States.’’The Review of Economic and Statistics LXXVI (1), February.

Fakhri, H. (2011) Relationship between Inflation and Economic Growth in Azerbaijani

Economy: Is there any threshold effect? Asian Journal of Business and Management
Sciences 1 (1).

Faraji, K. and Kenani, M. (2011) ‘‘Impact of Inflation on Economic Growth: A Case study of
Tanzania’’.Asian Journal of Empirical Research, 3 (4):363-380.

Garba, A. S. (2010) Refocusing Education System towards Entrepreneurship Development in

Nigeria: a Tool for Poverty Eradication. European Journal of Social Sciences.15(1),
140-150.

Granger, C.W.J., (1969).Investigating causal relationships by econometric models and cross

spectral methods.Econometrica, 17(2): 424-438.

Harrod, R.F. (1948) ‘‘Towards a Dynamic Economics. London. Macmillan.

Innocent, C.O., Nkechinyere, C.U., Ebele, N.and Geraldine, N. (2012), ‘‘Economic Growth
and Foreign Direct Investment in Nigeria: An Empirical Investigation. Journal of
Economics and Sustainable Development.ISSN : 2222-1700, 3(13). International
Labour Organization (ILO, 2001), Labour Statistics Yearbook, Geneva.

Jhingan, M. L. (2001) Advanced Economic Theory (11th Ed). Delhi: Vrinda Publications (P)
LTD.

Jhingan, M.L. (2002) Monetary Economics, 5th Revised and Enlarged, (Ed); Delhi: Vrinda
Publications (P) Ltd.

Joao, R.F. and Francisco, F.G. (2001), Does High Inflation Affect Growth in the Long and
Short Run?.Journal of Applied Economics, 4(1), 89-105

Johansen, S. (1988) “Statistical analysis of co-integrating vectors,”Journal of Economic

Dynamics and Control, 12,231-254.

Krugman, P. (2009) “Growth and Unemployment in the USA” The New York Times,
2009/08/01

Lindbeck, A. (1992), Macroeconomic Theory and the Labour Market, European Economic
1eview, 36, 209-235
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 32

Mahmoud, A.J. (2013), ‘‘Impact of Inflation and Unemployment on Jordanian

GDP’’.Interdisciplinary Journal of Contemporary Research in Business.Vol.4(10).

Malley, J. and H. Molana (2007) The Relationship between Output and Unemployment with
Efficiency Wages. German Economic Review, Vereinfur Social politik, vol. 8, pages
561-577

Mohsin, S. K. and Abdelhak, S. S. (2001) ‘‘Threshold effects in the Relationship between

Inflation and Growth’’. IMF Staff Papers Vol. 48(1).

Munnell, A.H. (1990) ‘‘Why Has Productivity Growth Declined?Productivity and Public
Investment.’’New England Economic Review (January/March).

Musa, M. S. (2010) ‘‘Nigerias solid minerals Resources and Development’’ A lecture

presented to the National Defence College Course 19, Abuja.

Ndiyo, N.A. and Ebong, F.S. (2003), ‘‘The Challenges of Openness in Developing
Economics: Some Empirical Lessons From Nigeria’’, NES 2003, Annual Conference.

Njoku, A. and. Ihgba, O. A. (2011) “Unemployment and Nigerian Economic Growth (1985-
2009)” Proceedings of the 2011 International Conference on Teaching, Learning and
Change International Association for Teaching and Learning (IATEL).

Obadan, M.I. (1992), ‘‘Direct Investment in Nigeria:An Empirical Analysis’’,African

Review, Vol. XXV (1).

Odusola, F.A. (1996). ‘‘Military Expenditure and Economic Growth’’ The Nigerian Journal
of Economic and Social Studies 38 (1,2 and 3).

Ogiogio, G.O. (1995), ‘‘Government Expenditure and Economic Growth in Nigeria.’’

Journal of Economic Management 2(1), October.

Olowononi, G.D. and Audu, I. (2012) ‘‘Unemployment, National security and sustainable
Development in Nigeria’’.Being a paper presented at ABU 50 Humanities
International Conference on National security, Integration and sustainable
Development in Nigeria Assembly Hall, ABU, Zaria, 19th- 20th November, 2012.

Omeke, P.C. and Ugwunyi, C.U. (2010) Money, Price and Output: A Causality Test for
Nigeria. American Journal of Scientific Research ISSN 1456-223X, Issue 8, pp.78-87.
Euro Journals Publishing, Inc.

Okun, Arthur M. (1962) Potential GNP: Its Measurement and Significance. Proceedings of
the Business and Economics Section of the American Statistical Association,
Washington, DC: American Statistical Association, pp 98-104.

Popovic, G. and Popovic, J. (2009) Inflation and Unemployment in the EU: Comparative
Analysis of Phillips Regularity. UDK 336.748.12, 331.56

Rafindadi, A.S. (2012) Macroeconomic Policy, Output and Unemployment Dynamics in

Nigeria: Is There Evidence of Jobless Growth? Paper presented at the 53th Annual
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 33

conference of the Nigerian economic society on ‘‘Youth Employment and Poverty

Reduction in Nigeria’’. NICON Luxury Hotel, Abuja.

Ratner, J.B. (1983). ‘‘Government Capital and the Production Function for US Private
Output.’’ Economic Letters 13.

Romer, C.D. (1996), ‘‘Changes in Business Cycles: Evidence and Explanations’’, Journal of
Economic Perspectives, Spring.

Sodipe, O. A. and Oluwatobi, S. (2010) “Output and Unemployment Relationship: How

Applicable is the Okun‘s Law to Nigeria?” available at
http://covenantuniversity.acadernia.edu/OluyomiSodipe/Papers/ 154201 8/Output and
Unemployment_Relationship_How_Applicable_is_the Okuns_Law_to_Nigeria

Sawtelle, B. (2007). ‘Analyzing the Link between Real GDP and Employment: An Industry
Sector Approach.”Business Economics 42(4): 46-54.

Solow, R.M. (2002) “Neoclassical Growth model” in B.Snowdon and H.R Vane (eds);An
Encyclopedia of macroeconomics, Cheltenham, UK and Northampton, MA, USA:
Edward Elgar.

Stock, J.H and Watson, M.W. (1999), ‘‘Forecasting Inflation’ ’National Bureau of Economic
Research NBER Working Paper 7023, March.

Todaro, M. (1992) Economics for a Developing World 2nd Ed. England: Longman Group,
U.K. Limited.

Todaro, M.P. and Smith, S.C. (2011) Economic Development.11th.Ed.Pearson Education

Limited, England.

Vikesh, G. and Subrina, H. (2004) ‘‘Relationship Between Inflation and Economic Growth’’.
Working Paper, Reserve Bank of Fiji.Suva Fiji.

Williams, O. and Adedeji, O.S. (2004).‘’Inflation Dynamics in the Dominican Republic’’.

IMF Working Paper, WP/04/29, Western Hemisphere Department: Washington,
D.C.,February.
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 34

APPENDICES
Appendix i :Cointegration test results
Sample (adjusted): 1988 2014
Included observations: 27 after adjustments
Trend assumption: Linear deterministic trend
Series: UN AGR IND BUC SER TRA
Lags interval (in first differences): 1 to 1

Unrestricted Cointegration Rank Test (Trace)

Hypothesized Trace 0.05

No. of CE(s) Eigenvalue Statistic Critical Value Prob.**

None * 0.940934 154.8238 95.75366 0.0000

At most 1 * 0.671363 78.43800 69.81889 0.0087
At most 2 * 0.573953 48.39233 47.85613 0.0445
At most 3 0.448298 25.35580 29.79707 0.1491
At most 4 0.287137 9.297632 15.49471 0.3386
At most 5 0.005873 0.159048 3.841466 0.6900

Trace test indicates 3 cointegrating eqn(s) at the 0.05 level

* denotes rejection of the hypothesis at the 0.05 level
**MacKinnon-Haug-Michelis (1999) p-values

Unrestricted Cointegration Rank Test (Maximum Eigenvalue)

Hypothesized Max-Eigen 0.05

No. of CE(s) Eigenvalue Statistic Critical Value Prob.**

None * 0.940934 76.38580 40.07757 0.0000

At most 1 0.671363 30.04567 33.87687 0.1340
At most 2 0.573953 23.03654 27.58434 0.1719
At most 3 0.448298 16.05817 21.13162 0.2215
At most 4 0.287137 9.138583 14.26460 0.2748
At most 5 0.005873 0.159048 3.841466 0.6900

Max-eigenvalue test indicates 1 cointegrating eqn(s) at the 0.05 level

* denotes rejection of the hypothesis at the 0.05 level
**MacKinnon-Haug-Michelis (1999) p-values
Unrestricted Cointegrating Coefficients (normalized by b'*S11*b=I):

UN AGR IND BUC SER TRA

-0.329431 6.984097 -5.438953 17.50289 -10.93346 -6.175678
1.047585 3.112117 -7.197047 -4.977768 -0.547563 7.921301
4.739857 -1.669163 -2.739766 6.647245 -16.44679 12.71325
-0.355814 -11.10761 1.352704 -0.966373 1.898662 7.708496
0.106691 -4.215405 1.284605 8.475327 -13.02664 8.245202
-0.833197 -7.947801 -0.373515 -7.243112 10.39582 4.526491

Unrestricted Adjustment Coefficients (alpha):

D(UN) 0.093957 0.073115 -0.175027 -0.099116 0.033487

D(AGR) 0.043346 -0.015773 -0.012907 0.054611 -0.009637
D(IND) 0.150011 0.155331 -0.076216 0.077403 -0.048237
D(BUC) 0.055954 0.012509 0.081722 0.017238 0.089119
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 35

D(SER) 0.126339 0.015012 0.106207 -0.007193 0.127968

D(TRA) 0.096880 -0.023487 0.046716 0.025288 0.034602

1 Cointegrating Equation(s): Log likelihood 145.6360

Normalized cointegrating coefficients (standard error in parentheses)

UN AGR IND BUC SER TRA
1.000000 -21.20048 16.51014 -53.13066 33.18892 18.74650
(2.84116) (1.54228) (3.77855) (3.36664) (2.69297)

Adjustment coefficients (standard error in parentheses)

D(UN) -0.030952
(0.02265)
D(AGR) -0.014280
(0.00689)
D(IND) -0.049418
(0.02049)
D(BUC) -0.018433
(0.01739)
D(SER) -0.041620
(0.02229)
D(TRA) -0.031915
(0.00869)

2 Cointegrating Equation(s): Log likelihood 160.6588

Normalized cointegrating coefficients (standard error in parentheses)

UN AGR IND BUC SER TRA
1.000000 0.000000 -3.996592 -10.69766 3.620618 8.936177
(1.04708) (2.75693) (2.51997) (1.22138)
0.000000 1.000000 -0.967277 2.001511 -1.394700 -0.462740
(0.07392) (0.19462) (0.17789) (0.08622)

Adjustment coefficients (standard error in parentheses)

D(UN) 0.045642 0.883750
(0.07324) (0.50991)
D(AGR) -0.030803 0.253649
(0.02262) (0.15750)
D(IND) 0.113304 1.531101
(0.05598) (0.38976)
D(BUC) -0.005329 0.429719
(0.05788) (0.40302)
D(SER) -0.025894 0.929083
(0.07420) (0.51665)
D(TRA) -0.056520 0.603523
(0.02835) (0.19741)

3 Cointegrating Equation(s): Log likelihood 172.1771

Normalized cointegrating coefficients (standard error in parentheses)

UN AGR IND BUC SER TRA
1.000000 0.000000 0.000000 5.929085 -6.223906 0.604011
(1.34227) (1.22975) (0.46073)
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 36

0.000000 1.000000 0.000000 6.025604 -3.777324 -2.479336

(0.81547) (0.74712) (0.27991)
0.000000 0.000000 1.000000 4.160230 -2.463230 -2.084818
(0.92348) (0.84607) (0.31698)

Adjustment coefficients (standard error in parentheses)

D(UN) -0.783962 1.175899 -0.557709
(0.25906) (0.41671) (0.50199)
D(AGR) -0.091978 0.275192 -0.086882
(0.09918) (0.15954) (0.19219)
D(IND) -0.247951 1.658318 -1.725012
(0.23297) (0.37474) (0.45143)
D(BUC) 0.382020 0.293313 -0.618258
(0.23968) (0.38554) (0.46444)
D(SER) 0.477510 0.751807 -1.086174
(0.30664) (0.49324) (0.59419)
D(TRA) 0.164906 0.525547 -0.485876
(0.11428) (0.18383) (0.22145)

4 Cointegrating Equation(s): Log likelihood 180.2062

Normalized cointegrating coefficients (standard error in parentheses)

UN AGR IND BUC SER TRA
1.000000 0.000000 0.000000 0.000000 -2.526555 2.198748
(0.33135) (0.32597)
0.000000 1.000000 0.000000 0.000000 -0.019784 -0.858639
(0.17222) (0.16943)
0.000000 0.000000 1.000000 0.000000 0.131071 -0.965847
(0.18541) (0.18240)
0.000000 0.000000 0.000000 1.000000 -0.623596 -0.268968
(0.05373) (0.05286)

Adjustment coefficients (standard error in parentheses)

D(UN) -0.748695 2.276844 -0.691784 0.212908
(0.23487) (0.65419) (0.45856) (0.93389)
D(AGR) -0.111409 -0.331404 -0.013009 0.698630
(0.07845) (0.21850) (0.15316) (0.31192)
D(IND) -0.275492 0.798561 -1.620309 1.271001
(0.21693) (0.60423) (0.42354) (0.86256)
D(BUC) 0.375886 0.101843 -0.594940 1.443662
(0.23955) (0.66721) (0.46768) (0.95246)
D(SER) 0.480070 0.831708 -1.095904 2.849508
(0.30735) (0.85607) (0.60007) (1.22208)
D(TRA) 0.155908 0.244661 -0.451669 2.098688
(0.11104) (0.30927) (0.21679) (0.44150)

5 Cointegrating Equation(s): Log likelihood 184.7755

Normalized cointegrating coefficients (standard error in parentheses)

UN AGR IND BUC SER TRA
1.000000 0.000000 0.000000 0.000000 0.000000 -0.389227
(0.08672)
0.000000 1.000000 0.000000 0.000000 0.000000 -0.878904
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 37

(0.01602)
0.000000 0.000000 1.000000 0.000000 0.000000 -0.831590
(0.01818)
0.000000 0.000000 0.000000 1.000000 0.000000 -0.907723
(0.02262)
0.000000 0.000000 0.000000 0.000000 1.000000 -1.024309
(0.03286)
Adjustment coefficients (standard error in parentheses)
D(UN) -0.745123 2.135682 -0.648766 0.496723 1.186908
(0.23192) (0.67618) (0.45679) (1.00609) (1.12838)
D(AGR) -0.112437 -0.290782 -0.025389 0.616957 -0.023797
(0.07772) (0.22661) (0.15308) (0.33717) (0.37815)
D(IND) -0.280638 1.001899 -1.682275 0.862178 0.303644
(0.21016) (0.61273) (0.41393) (0.91169) (1.02251)
D(BUC) 0.385395 -0.273830 -0.480458 2.198975 -3.090874
(0.21785) (0.63514) (0.42907) (0.94503) (1.05990)
D(SER) 0.493723 0.292272 -0.931516 3.934077 -4.816950
(0.27201) (0.79306) (0.53575) (1.18000) (1.32344)
D(TRA) 0.159600 0.098801 -0.407220 2.391948 -2.217428
(0.10409) (0.30348) (0.20502) (0.45155) (0.50644)

Appendix ii: Variance Decomposition and Impulse Response

Varia
nce
Deco
mposi
tion of
UN:
Perio
d S.E. UN AGR IND BUC SER TRA

1 0.261329 100.0000 0.000000 0.000000 0.000000 0.000000 0.000000

2 0.265845 100.0000 1.11E-19 1.27E-20 1.13E-20 2.39E-21 3.90E-21
3 0.269225 100.0000 1.38E-19 1.25E-20 1.16E-20 5.49E-21 6.05E-21
4 0.269598 100.0000 1.37E-19 1.28E-20 1.16E-20 5.49E-21 6.03E-21
5 0.269719 100.0000 1.38E-19 1.28E-20 1.16E-20 5.55E-21 6.07E-21
6 0.269740 100.0000 1.38E-19 1.28E-20 1.16E-20 5.56E-21 6.07E-21
7 0.269745 100.0000 1.38E-19 1.28E-20 1.16E-20 5.56E-21 6.07E-21
8 0.269746 100.0000 1.38E-19 1.28E-20 1.16E-20 5.56E-21 6.07E-21
9 0.269746 100.0000 1.38E-19 1.28E-20 1.16E-20 5.56E-21 6.07E-21
10 0.269746 100.0000 1.38E-19 1.28E-20 1.16E-20 5.56E-21 6.07E-21

Varia
nce
Deco
mposi
tion of
AGR:
Perio
d S.E. UN AGR IND BUC SER TRA
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 38

1 2.37E-12 1.20E-05 99.99999 0.000000 0.000000 0.000000 0.000000

2 2.47E-12 8.116818 91.88318 5.07E-21 3.44E-20 8.25E-21 8.46E-21
3 2.77E-12 26.80651 73.19349 1.16E-20 2.82E-20 7.36E-21 7.56E-21
4 2.77E-12 27.00154 72.99846 1.45E-20 3.17E-20 8.90E-21 9.47E-21
5 2.78E-12 27.33006 72.66994 1.45E-20 3.17E-20 9.62E-21 1.00E-20
6 2.78E-12 27.35660 72.64340 1.45E-20 3.17E-20 9.62E-21 1.00E-20
7 2.78E-12 27.36731 72.63269 1.45E-20 3.17E-20 9.63E-21 1.00E-20
8 2.78E-12 27.36892 72.63108 1.45E-20 3.17E-20 9.63E-21 1.00E-20
9 2.78E-12 27.36935 72.63065 1.45E-20 3.17E-20 9.63E-21 1.00E-20
10 2.78E-12 27.36943 72.63057 1.45E-20 3.17E-20 9.63E-21 1.00E-20

Varia
nce
Deco
mposi
tion of
IND:
Perio
d S.E. UN AGR IND BUC SER TRA

1 5.75E-12 3.22E-07 18.99193 81.00807 0.000000 0.000000 0.000000

2 6.09E-12 10.96497 16.90947 72.12556 1.26E-19 3.13E-20 3.42E-20
3 6.98E-12 32.23120 12.87060 54.89820 1.22E-19 3.66E-20 3.66E-20
4 6.99E-12 32.43046 12.83276 54.73678 1.26E-19 3.84E-20 3.89E-20
5 7.01E-12 32.78407 12.76560 54.45033 1.26E-19 3.91E-20 3.93E-20
6 7.01E-12 32.81209 12.76028 54.42762 1.26E-19 3.90E-20 3.93E-20
7 7.01E-12 32.82356 12.75810 54.41834 1.26E-19 3.91E-20 3.93E-20
8 7.01E-12 32.82527 12.75778 54.41695 1.26E-19 3.91E-20 3.93E-20
9 7.01E-12 32.82573 12.75769 54.41658 1.26E-19 3.91E-20 3.93E-20
10 7.01E-12 32.82581 12.75767 54.41651 1.26E-19 3.91E-20 3.93E-20

Varia
nce
Deco
mposi
tion of
BUC:
Perio
d S.E. UN AGR IND BUC SER TRA

1 1.76E-12 9.52E-07 4.849285 80.04907 15.10164 0.000000 0.000000

2 1.84E-12 8.419751 4.440987 73.30914 13.83012 1.11E-20 1.36E-20
3 1.87E-12 11.20559 4.305894 71.07911 13.40941 4.29E-20 3.77E-20
4 1.87E-12 11.20699 4.305826 71.07799 13.40920 4.37E-20 3.85E-20
5 1.87E-12 11.24825 4.303825 71.04496 13.40297 4.38E-20 3.86E-20
6 1.87E-12 11.24935 4.303772 71.04408 13.40280 4.38E-20 3.86E-20
7 1.87E-12 11.25038 4.303722 71.04325 13.40265 4.38E-20 3.86E-20
8 1.87E-12 11.25049 4.303717 71.04317 13.40263 4.38E-20 3.86E-20
9 1.87E-12 11.25052 4.303715 71.04314 13.40263 4.38E-20 3.86E-20
10 1.87E-12 11.25053 4.303715 71.04313 13.40263 4.38E-20 3.86E-20
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 39

Varia
nce
Deco
mposi
tion of
SER:
Perio
d S.E. UN AGR IND BUC SER TRA

1 1.15E-12 7.55E-06 71.97970 10.87263 11.35451 5.793144 0.000000

2 1.16E-12 2.238010 70.36879 10.62930 11.10040 5.663493 1.90E-20
3 1.34E-12 26.57483 52.85122 7.983252 8.337071 4.253626 1.43E-20
4 1.35E-12 27.00287 52.54312 7.936712 8.288468 4.228829 1.58E-20
5 1.35E-12 27.44289 52.22639 7.888870 8.238506 4.203338 1.63E-20
6 1.35E-12 27.48547 52.19575 7.884241 8.233672 4.200872 1.63E-20
7 1.35E-12 27.50060 52.18486 7.882596 8.231953 4.199995 1.63E-20
8 1.35E-12 27.50304 52.18310 7.882331 8.231677 4.199854 1.63E-20
9 1.35E-12 27.50366 52.18266 7.882263 8.231606 4.199818 1.63E-20
10 1.35E-12 27.50378 52.18257 7.882250 8.231593 4.199811 1.63E-20

Varia
nce
Deco
mposi
tion of
TRA:
Perio
d S.E. UN AGR IND BUC SER TRA

1 1.83E-12 3.14E-05 0.002315 50.15493 33.72595 4.799595 11.31718

2 2.04E-12 19.41796 0.001866 40.41587 27.17707 3.867613 9.119614
3 2.11E-12 24.76361 0.001742 37.73477 25.37420 3.611043 8.514638
4 2.11E-12 24.76666 0.001742 37.73324 25.37317 3.610897 8.514293
5 2.11E-12 24.84242 0.001740 37.69524 25.34762 3.607261 8.505719
6 2.11E-12 24.84438 0.001740 37.69426 25.34696 3.607166 8.505497
7 2.11E-12 24.84627 0.001740 37.69331 25.34632 3.607076 8.505283
8 2.11E-12 24.84646 0.001740 37.69321 25.34626 3.607067 8.505262
9 2.11E-12 24.84653 0.001740 37.69318 25.34623 3.607064 8.505254
10 2.11E-12 24.84654 0.001740 37.69318 25.34623 3.607063 8.505253

Chole
sky
Orderi
ng:
UN
AGR
IND
BUC
SER
TRA
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 40

IMPULSE RESPONSE

Resp
onse
of
UN:
Perio
d UN AGR IND BUC SER TRA

1 0.261329 0.000000 0.000000 0.000000 0.000000 0.000000

2 0.048792 8.85E-12 3.00E-12 2.83E-12 1.30E-12 -1.66E-12
3 0.042529 -4.63E-12 3.11E-13 -6.27E-13 -1.51E-12 1.28E-12
4 0.014180 2.67E-13 4.41E-13 2.44E-13 -1.16E-13 2.60E-14
5 0.008086 -5.42E-13 1.22E-13 -3.45E-14 -2.15E-13 1.68E-13
6 0.003323 -6.71E-14 7.93E-14 2.48E-14 -5.51E-14 3.47E-14
7 0.001655 -8.19E-14 3.04E-14 2.19E-16 -3.78E-14 2.80E-14
8 0.000734 -2.39E-14 1.58E-14 3.21E-15 -1.41E-14 9.66E-15
9 0.000349 -1.49E-14 6.84E-15 6.28E-16 -7.47E-15 5.38E-15
10 0.000159 -5.84E-15 3.30E-15 5.28E-16 -3.20E-15 2.24E-15

Resp
onse
of
AGR:
Perio
d UN AGR IND BUC SER TRA

1 8.19E-16 2.37E-12 0.000000 0.000000 0.000000 0.000000

2 -7.03E-13 -6.33E-23 -1.76E-23 -4.58E-23 -2.24E-23 2.27E-23
3 1.25E-12 -2.45E-23 -2.41E-23 -7.69E-24 -7.80E-24 7.90E-24
4 1.43E-13 5.91E-23 1.50E-23 1.66E-23 1.09E-23 -1.22E-23
5 1.86E-13 -2.51E-23 4.51E-25 -3.96E-24 -7.67E-24 6.66E-24
6 5.30E-14 2.87E-24 2.00E-24 1.38E-24 -3.47E-26 -3.15E-25
7 3.37E-14 -2.68E-24 4.31E-25 -2.49E-25 -9.87E-25 7.93E-25
8 1.31E-14 -1.34E-25 3.36E-25 1.30E-25 -1.89E-25 1.08E-25
9 6.75E-15 -3.68E-25 1.18E-25 -7.49E-27 -1.61E-25 1.22E-25
10 2.93E-15 -8.58E-26 6.50E-26 1.53E-26 -5.43E-26 3.65E-26

Resp
onse
of
IND:
Perio
d UN AGR IND BUC SER TRA

1 3.26E-16 2.51E-12 5.17E-12 0.000000 0.000000 0.000000

2 -2.02E-12 -3.87E-22 -3.60E-24 -2.16E-22 -1.08E-22 1.13E-22
3 3.41E-12 2.15E-22 3.92E-23 1.14E-22 7.88E-23 -7.19E-23
4 3.79E-13 1.64E-22 4.11E-23 4.58E-23 3.05E-23 -3.39E-23
5 5.07E-13 -6.92E-23 1.10E-24 -1.10E-23 -2.10E-23 1.83E-23
6 1.43E-13 8.06E-24 5.46E-24 3.81E-24 -2.59E-26 -9.20E-25
MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 41

7 9.16E-14 -7.35E-24 1.16E-24 -6.92E-25 -2.70E-24 2.17E-24

8 3.54E-14 -3.41E-25 9.14E-25 3.58E-25 -5.07E-25 2.87E-25
9 1.83E-14 -1.00E-24 3.19E-25 -2.16E-26 -4.39E-25 3.31E-25
10 7.95E-15 -2.31E-25 1.77E-25 4.18E-26 -1.47E-25 9.86E-26

Resp
onse
of
BUC:
Perio
d UN AGR IND BUC SER TRA

1 1.72E-16 -3.88E-13 1.58E-12 6.85E-13 0.000000 0.000000

2 -5.34E-13 -9.81E-23 6.01E-24 -4.25E-23 -1.94E-23 2.15E-23
3 3.26E-13 7.78E-23 1.81E-23 4.30E-23 3.35E-23 -2.93E-23
4 -7.43E-15 2.39E-23 4.25E-24 5.89E-24 5.21E-24 -5.31E-24
5 4.03E-14 -8.09E-24 -3.96E-25 -1.52E-24 -2.23E-24 2.03E-24
6 6.58E-15 1.54E-24 4.70E-25 4.69E-25 2.50E-25 -3.01E-25
7 6.38E-15 -7.47E-25 3.70E-26 -1.07E-25 -2.38E-25 2.03E-25
8 2.03E-15 5.80E-26 6.70E-26 4.00E-26 -1.25E-26 -6.25E-28
9 1.20E-15 -8.47E-26 1.72E-26 -6.21E-27 -3.28E-26 2.58E-26
10 4.83E-16 -8.39E-27 1.18E-26 3.95E-27 -7.71E-27 4.74E-27

Resp
onse
of
SER:
Perio
d UN AGR IND BUC SER TRA

1 3.16E-16 -9.77E-13 -3.80E-13 3.88E-13 2.77E-13 0.000000

2 1.74E-13 3.76E-23 2.70E-24 2.96E-23 1.69E-23 -1.61E-23
3 -6.71E-13 -4.60E-24 5.25E-24 -2.68E-24 2.17E-25 1.84E-25
4 -1.03E-13 -2.69E-23 -7.86E-24 -8.02E-24 -4.50E-24 5.32E-24
5 -1.05E-13 1.26E-23 -5.41E-25 1.85E-24 3.99E-24 -3.42E-24
6 -3.28E-14 -1.08E-24 -1.11E-24 -6.81E-25 1.70E-25 4.20E-26
7 -1.95E-14 1.41E-24 -2.76E-25 1.09E-25 5.42E-25 -4.29E-25
8 -7.84E-15 1.26E-25 -1.93E-25 -6.66E-26 1.23E-25 -7.47E-26
9 -3.96E-15 2.04E-25 -7.13E-26 1.55E-27 9.23E-26 -6.88E-26
10 -1.74E-15 5.43E-26 -3.80E-26 -8.23E-27 3.30E-26 -2.24E-26

Resp
onse
of
TRA:
Perio
d UN AGR IND BUC SER TRA

1 -1.03E-15 -8.81E-15 -1.30E-12 -1.06E-12 -4.01E-13 6.16E-13

2 8.99E-13 1.14E-22 -6.31E-25 4.16E-23 1.59E-23 -1.89E-23
3 -5.44E-13 -7.48E-23 -1.60E-23 -4.16E-23 -3.10E-23 2.70E-23
 MAUTECH Journal of Economic Studies: Vol. 1 No. 1 2017 42

4 1.34E-14 -4.00E-23 -7.09E-24 -9.85E-24 -8.74E-24 8.91E-24

5 -6.70E-14 1.35E-23 6.72E-25 2.55E-24 3.72E-24 -3.39E-24
6 -1.08E-14 -2.59E-24 -7.82E-25 -7.84E-25 -4.23E-25 5.07E-25
7 -1.06E-14 1.25E-24 -6.00E-26 1.79E-25 3.97E-25 -3.38E-25
8 -3.36E-15 -9.89E-26 -1.11E-25 -6.68E-26 1.99E-26 1.71E-27
9 -1.98E-15 1.41E-25 -2.84E-26 1.05E-26 5.44E-26 -4.29E-26
10 -7.99E-16 1.37E-26 -1.95E-26 -6.59E-27 1.27E-26 -7.80E-27

Chole
sky
Orderi
ng:
UN
AGR
IND
BUC
SER
TRA

View publication stats