Causality Analysis Real Wages and Opportunities

in Indonesia with a Dynamic Model Approach

Murtala, Teuku Roli Ilhamsyah Putra, Chalirafi, Eddy Gunawan, and Irham Iskandar

Faculty of Economics and Business Universitas Malikussaleh, Indonesia

Abstract. Wages are one of the deciding indicators in employment opportunities,

wage improvements in the economic development process are absolutely

necessary. With an increase in wages, it means an increase in income, an increase

in purchasing power and an increase in people's welfare. Increased income of the

community will have an impact on increasing demand for goods and services,

then in turn at a macro level will encourage companies to develop. This study

aims to determine the effect of real wages on employment opportunities in

Indonesia by using the Vector Auto Regression (VAR) approach. The data used

in this study are time series data during the 1980-2015 period. The analytical

method used in this study is the Vector Auto Regression (VAR) model.

Estimation results show that employment opportunities and stationary real wages

are at different rates using Philips-Perron. The results of the analysis did not occur

co-integration in the long run. Real wages have a negative effect on employment

opportunities in Indonesia. This condition illustrates that during the study period

the increase in real wages that occurred in Indonesia was not followed by a

balanced increase in productivity, so that an increase in real wages led to a

reduction in demand for labor.

Keywords: Real wage causality relationship ꞏ Job opportunities

1 Introductions

The employment problem in Indonesia is very large and complex. Great because it

requires millions of workers. Complex because labor problems affect simultaneously

by many interrelated factors, taking a pattern that is not easy to formulate (Tobing,

2006). Demographic factors influence the number and composition of the workforce.

Indonesia has been quite successful in reducing the number of births and deaths to a

full extent. However, this reverses the impact on working population growth which is

much faster than overall population growth (Ananta, 2003).

There are several main problems in the labor field that discuss about Indonesia,

namely the matter of employment opportunities, relatively high levels of labor force

growth, low wage rates, and labor productivity, which are compared between regions,

both in terms of wages and labor rates. This problem is not only related to the region,

but also a national problem (Hasibuan, 2007: 32).

The creation of employment opportunities is done by growing the business world

through various policies including in the fields of production, monetary, fiscal,

distribution, prices and wages, exports - imports, and in the field of employment. Thus,

190

Murtala, ., Putra, T., Chaliraﬁ, ., Gunawan, E. and Iskandar, I.

Causality Analysis Real Wages and Opportunities in Indonesia with a Dynamic Model Approach.

DOI: 10.5220/0009847600002900

In Proceedings of the 20th Malaysia Indonesia International Conference on Economics, Management and Accounting (MIICEMA 2019), pages 190-199

ISBN: 978-989-758-582-1; ISSN: 2655-9064

every decision in the field of employment and employment opportunities taken in

general, always has a political dimension. Labor issues also address wages and social

security issues, setting minimum wages, employment requirements, labor protection,

dispute agreements, freedom of association and industrial relations, and international

relations and cooperation. All of them contain economic, social and political

dimensions. In other words, this employment problem has multiple dimensions, is

complex and very complex.

The development of labor in Indonesia for the period 1990-2015 can be seen in

Figure 1:

Source: Department of Population and Labor Mobility of Indonesia, (data processed, 2016)

Fig. 1. Development of Manpower in Indonesia Period 1990-2015.

From Figure 1 illustrates the development of labor in Indonesia, shows. In 1990 the

workforce in Indonesia was 89,794,936 people, then in 1991 it increased by 0.53

percent or to 90,273,568 people. In 1992 the number of workers in Indonesia increased

again to 91,230,888 people, this situation continued to increase each year until 2002 the

workforce in Indonesia reached 104,631,968.00 people. In 2003 it decreased by 11.94

percent or 92,142,064.00 people. Then in 2004 the workforce in Indonesia again

decreased by 85,268,624.00 people or 7.46 percent. The development of the number of

workers in Indonesia until 2005 has increased again and in 2006 has decreased again,

however the number of workers in Indonesia over time with increasing population

conditions and improving economic conditions, the number of workers in Indonesia is

also increasingly increased, namely in 2015 reaching 135,541,032 inhabitants.

For employers, wages are seen as a cost burden thus encouraging employers to act

rationally, namely by setting wages equal to the marginal value of products of labor.

However, the government's policy which requires employers to take into account the

Minimum Physical Needs of workers in setting wages has caused the average wage rate

of workers to increase, because now the wages received by some workers are higher

than the marginal product value it produces.

This Regional Minimum Wage Policy often brings criticism from the business

community. Because in reality the productivity level of most workers is still below the

minimum wage level, the establishment of the minimum wage is seen as an increase in

production costs. The way out by entrepreneurs to ensure their business continuity is to

increase the selling price of products or reduce the amount of labor used. Both of these

alternatives have consequences for labor. The decision of employers to increase the

selling price of products will lead to inflation, and that means that the real wages

received by workers do not increase. On the other hand, if the solution taken by the

employer is to lay off some workers, then that means there is unemployment.

The minimum wage was first introduced in Indonesia in the early 1970s, and in the

later stages of the 1980s along with changes in the labor market, once a minimum wage

became important. In the first half of the 1990s the government more than doubled the

0,00

50.000.000,00

100.000.000,00

150.000.000,00

1990

1994

1998

2002

2012

Tahun

Tenaga Kerja

Series1

Years

Labo

Causality Analysis Real Wages and Opportunities in Indonesia with a Dynamic Model Approach

191

real minimum wage. In the second half of the 1990s nominally the minimum wage

continued to increase but in real terms the increase was small. Even in 1998 the real

value of the minimum wage fell quite large because of high inflation in that year due to

the economic crisis that hit Indonesia.

The development of real wages in the 1980-2014 period in Indonesia is seen in

Figure 2.

Source: BPS Indonesia, 2014

Fig. 2. Development of Real Wages in Indonesia Period 1980-2015.

From Figure 2 illustrates that the development of real wages in Indonesia shows a

progressively increasing development. Starting from 1980 the real wage in Indonesia

was Rp. 20,000 \ months until 1990 real wages in Indonesia increased to Rp. 110,000\

month, then in 1990 it increased to Rp. 120,000 \ month. In 1992 real wages in

Indonesia again increased to Rp.140,000 \ month, this situation continued to increase

each year until 2009 real wages in Indonesia reached Rp. 1,200,000 \ month. This

increase continued until 2015. As inflation continued to increase, real wages also

increased. Where the real wage is the lowest monthly wage with work time 7 hours a

day or 40 hours a week for the 6 day work system and 8 hours or 40 hours a week for

the work system 5 days a week.

This study aims to find out how much influence the real wage has on employment

opportunities in Indonesia. Job opportunities can be interpreted as an employment or

all types of jobs available where workers to make a living to meet their needs.

Djojohadikusumo (2000: 27) defines employment opportunities as positions that

arise both inside and outside the company as a result of investment and population

growth and the labor force on the one hand will affect the problem of unemployment

and expansion of employment opportunities. In addition, employment opportunities can

also be interpreted as the number of residents who work or those who have already

gotten a job. The more people who work the more extensive job opportunities (Esmara,

2002: 134).

Job opportunities contain an understanding of business opportunities or

opportunities available to work as a result of an economic activity, thus employment

opportunities include jobs that have been filled and employment opportunities can also

be interpreted as participation in development (Sagir, 2004: 52). Based on the

definitions of the two descriptions above, it can be concluded that employment

opportunities are the number of people who participate in development by doing a job

and receive the results of the development.

The definition of employment opportunities or groups of people who work

according to the population census in 1980 and 1990 are those who during the week

before enumeration did a job with the intention of earning income with working time

of at least one hour. This group also includes those who during the week before

500.000

1.000.000

1.500.000

2.000.000

2.500.000

3.000.000

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

2004

2006

2008

2010

201

UM R

Tahun

Series 1

Years

Labo

MIICEMA 2019 - Malaysia Indonesia International Conference on Economics Management and Accounting

192

enumeration did not do work or worked for less than an hour, but had a permanent job,

but while they did not work because of illness, strike and others (BPS, 1999: 32).

Specifically, a request describes the maximum amount of labor employed by the

employer to be employed at each possible wage level within a certain period of time.

The demand for labor by employers can be seen in the short run (short run) and long

run (long run). In the short term employers are more likely to increase employment if

they want to increase production. In the long run, the tendency of companies to

substitute towards the use of relatively cheaper inputs. The amount of labor demanded,

both in the short term and in the long term has a negative relationship with wage levels.

According to Connell, (2009: 133) changes in labor demand can be influenced by

various factors, namely the demand for production (output) produced by labor, labor

production, the amount of labor available in the labor market, and the price of other

resources (substitute goods).

The terminology of wages means labor income received from the employer because

he is considered to have done work (Soepomo, 2002: 152). From the above

understanding, there are several things that cause given wages to someone. First: wages

are earned because of services that have been given by someone to the owner of the

object of work in accordance with the work (Kartasapoetra, 2007: 93). Second: wages

are earned because someone has done work that should be done or not doing work,

because of the dispensation for workers who are sick or unable for reasons that are

acceptable to both employers, this is contained in Government Regulation No. 8 of

1981 concerning wage protection (the drafting team of the Republic of Indonesia

Department of Labor, 1996: 21 Article 5 Paragraph (1) Letter a).

The National Wage Research Council, states that wages are an acceptance as a

reward from the employer to the recipient of work for a job or service that has and will

be carried out as a guarantee of a decent survival for humanity and the production is

expressed or valued in the form of money stipulated according to an agreement, law or

regulation and paid on the basis of an employment agreement between the employer

and the recipient of work (Handoko, 2005: 118).

According to Law No. 33 of 1947, concerning compensation payments to workers

who had an accident related to work relations. Article 7 Paragraphs (a) and (b) referred

to as wages are:

a. Each payment is in the form of money received by workers in exchange for

work.

b. Housing, food, food and clothing are useless, the value of which is estimated at

the general price in that place.

According to Prisono (2006: 79) wages or salaries can be seen as a reward or

repayment to workers for the production output that has been produced, while the

minimum wage is the lowest wage that has been calculated as a basis for the provision

of wages that should be sufficient to be used as a survival cost the worker and his family

according to their level of need (Kartasapoetra, 2007: 158).

In the economy there are various types of goods and services. From year to year

they experience price increases / changes that are not uniform. There is a high price

increase and there is a relatively slow increase in price. In addition, various types of

goods are very different interests in human life. Some are often bought by consumers,

such as food, clothing and rental homes. There are also those whose purchases are not

made too often such as buying houses and cars, or traveling abroad. This difference has

Causality Analysis Real Wages and Opportunities in Indonesia with a Dynamic Model Approach

193

a different effect on the welfare of the community if the prices of these items become

higher. The problems just described cause difficulties in trying to show the rate of

change in prices prevailing in an economy from year to year. This in turn makes it

difficult to calculate real wages from year to year.

Each country usually describes changes in prices in its economy by creating a price

index, which is an index that gives an idea of the average level of changes in prices over

time. One of these price indices is the Consumer Price Index (CPI). This price index can

be used to estimate the real wages of workers from year to year (Sukirno, 2012: 352).

Ferdinand (2011) said that the factors that significantly affected employment in the

Province of West Sumatra in 2005-2010 were government spending and the amount of

Gross Regional Domestic Product (GRDP) which had a positive effect. While real

wages have a negative effect on employment. Meanwhile Nazamuddin (1998) the

unemployment rate in a place is negatively correlated with the level of wages in that

place, so that an area or country where the wage level is relatively high, there is low

unemployment. Increasing the level of wages, thus should not have an impact on rising

unemployment.On the other hand Nindya and I Wayan (2014), that real wages have a

significant negative effect on employment. An increase in real wages will reduce labor

absorption. In contrast to research conducted by Setiya (2013) the real wage variable

has a positive and significant effect on employment with α 5%.

2 Materials and Methods

The data used in this study are time series data for the period 1980 - 2015. The analytical

method used in this study is multivariate vector autogression (VAR). Before arriving at

the VAR analysis there are several estimation steps that will be used in the analysis,

which consists of:

1. Test data stationarity and degree of integration

2. Determination of the lag length

3. Granger causality test

4. VAR estimation

5. Variance Decomposition

Where :

KK = Job Opportunity

UR = Real Wage

3 Results and Discussion

3.1 Stationary Test / Unit Root (Unit Root Test)

Stationarity testing is the initial stage before estimating the time series model. Time

series data that is directly analyzed will cause spurious in the results because in these

MIICEMA 2019 - Malaysia Indonesia International Conference on Economics Management and Accounting

194

variables often contain unit root. Therefore it is necessary to do a unit root test to see

the time series data stationarity. Unit root test is done by the Philips-Perron (PP)

method.

Table 1. Unit Root Test with Philips-Perron.

Variable

Probabilitas

2th Difference

KK 0.0000

UR 0.0000

Source: Data Processing Results, 2019

The hypotheses tested were H0: β1 = 0 (indicating the presence of a root unit or

non-stationary data) and H1: β1 ≠ 0. Here β1 is the PP value. If the PP probability value

is greater than the alpha testing level, then the hypothesis H0 which states that the data

contained in the root unit is rejected means that the time series data is stationary, and

vice versa. Based on the unit root test results from Table 1, it can be explained that at a

significance level of 5 percent, the real variable wage (UR) and employment (KK) has

a PP probability value smaller than the 5 percent alpha testing level on lsecound

different or I (0).

3.2 Determination of Optimal Lag

In determining the optimal lag the values of likelihood ratio (LR), final prediction error

(FPE), Akaike information criterion (AIC), Schwarz information criterion (SIC), and

Hannan-Quin criterion (HQ) are used. The optimal lag length chosen based on the

criteria above is shown in Table 2.

Table 2. Optimal Lag Length Based on Several Criteria.

L LR FPE

0 -1060.116

A3.10e+25

1 -944.4201 210.3561 3.56e+22

2 -936.7871 12.95305 2.87e+22

3 -928.7292 12.69731* 2.26e+22*

Source: Data Processing Results, 2019

Based on Table 2 the LR and FPE criteria choose lag order 3. Thus in this study the

optimal lag length that will be used is 3.

3.3 Cointegration Test

Cointegration test is a test of whether there is a long-term relationship between

independent and dependent variables, this test is a continuation of the unit root test

(Unit Root Test) and the degree of integration test (Integration Test). Cointegration test

results are seen by comparing the value of the trace test and max-eigen test with the

value of critical value. Cointegration test results can be seen in Table 3.

Causality Analysis Real Wages and Opportunities in Indonesia with a Dynamic Model Approach

195

Table 3. Cointegration Test Results.

pothesize

Trace 0.05

o. of CE(s) Ei

envalue Statistic

Critical

Value Prob.**

one * 0.627615 38.95965 25.872 0.00

At most 1 0.175327 6.361346 12.517 0.41

Trace test indicates 1 cointe

ratin

eqn(s) at the 0.05 level

* denotes re

ection of the h

pothesis at the 0.05 level

**MacKinnon-Hau

-Michelis (1999) p-values

Unrestricted Cointe

ration Rank Test (Maximum Ei

envalue)

Source: Data Processing Results, 2019

From Table 3 shows that of the two variables tested only one variable occurred

cointegration so that from the test results it was stated that the two variables tested did

not have cointegration.

3.4 Granger Causality Analysis

Causality analysis shows a causal relationship. To know the characteristics of causality

relationships, namely changes in a variable that is more influential on other variables,

it requires a Granger causality test. To find out the results of the causality test can be

seen in Table 4.

Table 4. Granger Causality Test Lag 8.

ull H

pothesis: Obs F-Statistic Prob.

UR does not

Gran

er Cause KK 34 6.59548 0.0044

KK does not Granger

Cause UR 0.67572 0.5166

Source: Data Processing Results, 2019

From the results of the Granger causality test in Table 4 it can be explained, UR has

a one-way causality with KK, meaning that real wages are suitable for use as

independent variables and employment opportunities as dependent variables.

3.5 Vector Autoregression Estimates Analysis

As previously explained that there are two research variables that are stationary in the

second difference so that all variables can be said to be integrated in degree 2 or I (2).

Furthermore, the estimated VAR will be performed on the second difference with

MIICEMA 2019 - Malaysia Indonesia International Conference on Economics Management and Accounting

196

endogenous variables DDKK and Cholesky Ordering DDUR → DDKK as follows

(statistics t [......]): Equation ............................ 1

Vector Autore

ression Estimates

DDKK DDUR

DDKK(-1) 1.396190 -0.000506

(0.20576) (0.00302)

[ 6.78556] [-0.16743]

DDKK(-2) -0.516275 -0.005122

(0.33416) (0.00490)

[-1.54500] [-1.04453]

DDKK(-3) -0.021707 0.003464

(0.19158) (0.00281)

[-0.11331] [ 1.23210]

DDUR(-1) 17.16994 0.621151

(14.6278) (0.21467)

[ 1.17379] [ 2.89346]

DDUR(-2) -18.91750 0.084022

(19.8429) (0.29121)

[-0.95337] [ 0.28853]

DDUR(-3) 4.582129 0.627216

(17.5102) (0.25698)

[ 0.26168] [ 2.44077]

C 12297440 188244.6

(6513920) (95596.6)

[ 1.88787] [ 1.96916]

R-square

0.951448 0.995176

Adj. R-

square

0.939309 0.993970

F-statistic 78.38511 825.2033

Source: Data Processing Results, 2019

From equation 1 partially, the real wage has a negative effect on employment in

Indonesia in the first period, as well as in the second period, the real wage still has a

negative effect on employment opportunities in Indonesia, meaning that if there is a

shock to an increase in the real wage, employment opportunities will decrease and vice

versa. This is due to the increase in real wages that occur not followed by an increase

in balanced productivity, so that an increase in real wages encourages a reduction in

demand for labor.

Causality Analysis Real Wages and Opportunities in Indonesia with a Dynamic Model Approach

197

R square value of 0.95 shows that the real wage is able to explain employment

opportunities by 95 percent and the remaining 5 percent is explained by other variables

outside this research model.

3.6 Analysis of Variance Decomposition

Variance decomposition provides estimates of how much a variable contributes to

changes in the variable itself and other variables in the coming periods.

Table 5. Variance Decomposition.

Variance Decomposition of DDKK:

Perio

S.E. DDKK DDUR

1 2838612. 100.0000 0.000000

2 4847961. 97.86369 2.136308

3 6289431. 97.66674 2.333265

4 7073693. 97.50407 2.495928

5 7451087. 96.53752 3.462478

6 7609198. 95.24495 4.755048

7 7672046. 93.88854 6.111461

8 7755015. 92.16676 7.833236

9 7904401. 90.14906 9.850943

10 8125281. 88.03926 11.96074

Source: Data Processing Results, 2019

Table 5 shows that in the first period 100 percent of the variability of employment

is explained by the average employment itself, while the real variable is zero percent.

In the second period the variability of employment opportunities is explained by the

average employment opportunity itself of 97.83 percent, while from the real wage

variable of 2.13 percent. In the third period the variability of employment opportunities

is explained by the average employment opportunity itself of 97.66 percent, while that

of the real wage variable is 2.33 percent. Up to ten percent of the variability of

employment opportunities is explained by the average employment opportunity of

88.03 percent, while from the real wage variable of 11.96 percent.

4 Conclusion

a. Based on the estimation results of the two variables analyzed, namely employment

opportunities and stationary real wages at different securities using Philips-Perron.

b. Of the two variables analyzed, it turns out that cointegration does not occur in the

long run.

c. Real wages affect employment opportunities by 95 percent and the remaining 5

percent are explained by other variables outside this research model.

MIICEMA 2019 - Malaysia Indonesia International Conference on Economics Management and Accounting

198

d. Real wages have a negative effect on employment opportunities in Indonesia. This

is due to an increase in real wages that does not occur followed by an increase in

balanced productivity, so that an increase in real wages drives a reduction in demand

for labor.

5 Recommendation

a. It is expected that the government is able to maintain the minimum wage of labor

every year by considering the ability of the company so that there is no termination

of employment (PHK), so that the power consumption of workers who receive

wages with the standard Wage Rill does not experience a decline.

b. Then the government is also expected to stabilize inflation, by issuing regulations

such as imposing sanctions on traders who sell goods at high prices.

c. Considering the very limited ability of large and medium industrial sub-sectors to

absorb labor, the government should focus more attention on developing small and

household industries, especially in the form of facilities to obtain capital and at the

same time guidance and guidance in increasing efficiency.

References

Ananta, Aris (2003). Ciri Demografis, Kualitas Penduduk dan Pembangunan Ekonomi, Lembaga

Demografi FEUI, Jakarta

Badan Pusat Statistik, (2000). Penjelasan Perhitungan Penduduk Miskin Provinsi Daerah

Istimewa Aceh.

________, (1999). Pendapatan Regional Propinsi di Indonesia. Jakarta.

Hasibuan, Nurimansyah, (2007). Analisis Statistik Industri Besar dan Sedang, BPS PPEM.

Universitas Sriwijaya, Jakarta.

Handoko, (2005). Kebijakan Dalam Upah Minimum Regional. Edisi ketiga, jilid 4, Erlangga,

Jakarta.

Ferdinan Hery (2011) Pengaruh Pengeluaran Pemerintah, PDRB, Dan Upah Riil Terhadap

Penyerapan Tenaga Kerja di Sumatera Barat. Littri Vol.7 No. 4: 101-112. Olah

Kartasapoetra. G (2007). Hukum Perburuhan di Indonesia Berdasarkan Pancasila, Cet.1, PT.

Bina Aksara. Jakarta.

Connell, Campbell R And Stanley L Brne (2009). Contemporary Labor Economics, Second Edit

ion Me Graw-Hill, US'A.

Nazamuddin.(1998). Jurnal Ekonomi dan Bisnis, Vol. 1. No. 1. Mei 1998. Fakultas Ekonomi,

Universitas Syiah Kuala. Banda Aceh.

Nindya Eka Sobita dan I Wayan Suparta (2014). Pertumbuhan Ekonomi Dan Penyerapan Tenaga

Kerja di Provinsi Lampung. JEP-Vol. 3, N0 2, Juli 2014:141-166.

Prisono (2006). Penentuan Upah Minimum Regional. PT. Bumi Aksara, Jakarta.

Sagir. (2004). Ekonomi Industri, Berdikari Student Studi Club Union. Jakarta.

Setiya Luthfi Priambodo (2013). Pengaruh Pertumbuhan Ekonomi, Upah Riil Dan Investasi

Terhadap Penyerapan Tenaga Kerja Di Kota Semarang. Jurnal Ekonomi Pembangunan. Vol

3 No 2: 61-75

Soepomo, Imam. (2002). Pengantar Hukum Perburuhan, PT. Ikhtiar Mandiri Abadi. Jakarta.

Sukirno, Sadono (2012). Makro Ekonomi Teori Pengantar, Edisi ke duabelas. Jakarta: PT. Raja

Grafindo Persada.

Causality Analysis Real Wages and Opportunities in Indonesia with a Dynamic Model Approach

199