Divergence Analysis of Economic Development in West and East Regions of Indonesia (Cumulative Causation Approach)

Amna Suresti^*, Syafrizal⁺, Syofyardi⁺

⁺Study Program of Development Planning, Graduate Program of Andalas University, Padang, Indonesia
^*Graduated Student in 1998

Abstract
The economic growth difference between Indonesian West Region (Kawasan Barat Indonesia, KBI) and Indonesian East Region (Kawasan Timur Indonesia, KTI) was analyzed to know whether the characteristics of the development process in both regions is convergent or divergent toward an equilibrium growth and how to solve any divergence among the provinces of the regions. The analysis of the characteristics trend of the economic growth used the cumulative causation model. This approach has a principle that labor productivity growth is influenced by output growth affected by wage growth. However, this wage growth is influenced by the labor productivity growth. Using the regression analysis on data of gross regional domestic product (PDRB) of KBI and KTI for 1983-1996, the convergence and divergence grade could be obtained. The result suggests that the economic growth of KBI tends more convergent than of KTI due to the higher labor productivity growth. A recommendation for solving the divergence in KTI is that its labor productivity growth should be increased, e.g., by improving its labor quality and developing of economic growth regions in each province of KTI.
 

Keywords
Economic growth, cumulative causation, productivity, output, wage, KBI, KTI
 

Introduction
A development of a country is to increase prosperity of its whole population. However, in many developing countries, an interregional economic disparity is a serious problem. Economic growth is a process of per capita income increase evaluated for a period and can be used as an indicator to identify an economic development progress. A development disparity caused by a interregional growth difference can be determined by a economic growth model. Studies by economists of the economic growth theory lead to exploitation of natural resources and investments. However, in our study the growth model uses productivity growth. 

Developed regions like capital or big cities commonly have a much higher productivity growth than underdeveloped regions, so interregional disparities could occur. The problem is, whether the development process in KBI and KTI is convergent or divergent toward the equilibrium growth and if there is any difference, how to solve it. This study aim is to analyze of economics growth model of KBI and KTI by the cumulative causation model in order to propose a solution of the development disparities.
 

Growth Model
The cumulative causation model (Richardson, [6]) has a principle that labor productivity growth is influenced by output growth affected by wage growth. However, this wage growth is influenced by the labor productivity growth. This relationship is described with equations: 
 
 

r = a + b × y	(1)
w = c - d × r	(2)
y = e - f × w	(3)

where r is productivity growth of KBI, KTI and Indonesia, y is output growth based on PDRB of KBI and KTI, and PDB of Indonesia, w is wage growth of KBI, KTI and Indonesia, and b, d, f are coefficients and a, c e are constants obtained from a regression. To determine whether the growth model tends to a convergent or divergent condition, the constants called Verdoorn coefficients are calculated as followed:
 

g = b × d × f	(4)
h = e + f (a × d - c)	(5)

If g>1 and h<0, it is identified that the economic growth will be far from the equilibrium growth (divergent) and if g<1 and h>0, it will reach the equilibrium growth (convergent). The regression analysis is conducted using an SPSS program with time series data of KBI, KTI and Indonesia for the period of 1983-1996. The equilibrium growth (y_e) is determined using:
 
 

(6)

The factor influencing on the economical growth is productivity. This relationship can be written as 
 

y=a +b × r

(7)

where a and b are coefficients obtained from a regression analysis.

This study uses the PDRB, PDB, population and average monthly wage/salary. The data are obtained from SUSENAS (National Economic Census), SAKERNAS (National Labor Survey) and Statistics Yearbook of Indonesia published by BPS (Statistics Center). The definitions used in the research are:

1. Labor productivity growth: gross regional domestic product (PDRB or PDB) divided by number of labor.
2. Labor is population of 10 years old and above as described in SUSENAS of 1983-1996.
3. The defined wage is real wage calculated using the product of nominal wage growth with consumer price index of 1993. The nominal wage is wage or average salary of labor/employee per month in each province.
4. The defined regions are provinces in KBI and KTI. The provinces in KBI are all provinces in Sumatra and Java, and the others belong to KTI.
5. Disparity is an undistributed development process or a difference of interregional convergence degree. 
6. The PDRB is with and without oil and gas (Migas).

The cumulative causation model doesn’t consider that a convergence occurs in interregional prosperity difference of a country, even though the country is already developed. It considers that the convergence in development process is caused by the government involvement. 

As described by Dixon and Thirwall (1974), every country will usually experience the Verdoorn effect. This effect causes that developed country grows fast due to a positive relationship between productivity and output, while underdeveloped country will slowly grow due to its low output. This will cause a wider disparity.

Regression results of Equation 1, 2 and 3 could determine weather the development process in both regions had a tendency to a convergence or divergence. The model was analyzed with a simultaneous regression and the empirical results with t_test and correlation coefficient (R²) are shown as followed:
 

A. With Oil and Gas (Migas)
KBI
r=-1.8319 + 0.7975 y       (t_test=10.170, R²=0.9039)
w=12.9922 – 2.1491 r       (t_test=3.004, R²=0.4572) 
y=1.7768 – 0.2411 w        (t_test=2.784, R²=0.4133)

KTI 
r=-3.9124 + 1.2793 y       (t_test=15.387, R²=0.9556)
w=15.4745 – 0.8830 r       (t_test=5.088, R²=0.7018) 
y=-4.1541 – 0.5345 w       (t_test=3.620, R²=0.5436)

Indonesia
r=-2.8807 + 0.9937 y       (t_test=11.343, R²=0.9212)
w=13.8309 – 1.6353 r       (t_test=3.722, R²=0.5575) 
y=0.6991 – 0.2997 w        (t_test=3.072, R²=0.4618)
 

B. Without Oil and Gas (Migas)
KBI
r=-3.0056 + 0.9878 y       (t_test=12.149, R²=0.9306)
w=17.6109 – 0.7501 r       (t_test=1.775, R²=0.2227) 
y=0.6414 – 0.3135 w       (t_test=1.969, R²=0.2605)

KTI
r=-2.2809 + 0.8299 y       (t_test=13.262, R²=0.9554)
w=13.5376 – 1.1358 r       (t_test=4.429, R²=0,4407) 
y=0.7798 – 0.3620 w       (t_test=3.420, R²=0.5153)

Indonesia
r=-3.0920 + 1.0316 y       (t_test=4.683, R²=0.9215)
w=15.2016 – 1.1427 r       (t_test=3.386, R²=0.5103) 
y=-1.0035 – 0.4034 w       (t_test=3.097, R²=0.4656)

The interregional growth difference is smaller than equilibrium growth, where the initial growth (y_o) of 5.75% in KBI, 5.00% in KTI and 6% in whole Indonesia (without Migas), while that with Migas is also smaller than the equilibrium growth with the initial growth of 5.9% in KBI, 3.6% in KTI and 5.6% in whole Indonesia. This suggests that both regions have increasing economic growths. The situation shows that the economic development in KBI or KTI is generally getting better caused by the regional development policy referring to underdeveloped region like developments of structure and infrastructure, distribution of investment allocation, and other policies to increase labor opportunities.

Although KBI, KTI and Indonesia have the same regional economic growth model, which is convergent toward the equilibrium growth, but the convergence degrees of the regions are different. If g value is low, the growth model tends to be more convergent. The g value of KBI is the lowest compared to that of KTI and Indonesia, while that of KTI is the highest at 0.603. This suggests that although the growth of KTI is convergent, but its regional economic growth difference is relatively larger than of KBI and Indonesia. This illustrates that the development in KTI is not well-distributed due to many underdeveloped provinces. Besides, the equilibrium growth degree of KTI is lower than that of KBI.

Figure 1 shows the equation slopes of y_t+1 of KBI, KTI and Indonesia. The slope for KBI is very small that means a well development distribution, while of KTI is high so its development distribution is very low. For whole Indonesia, the slope is almost the same to of KTI that suggests a lowly distributed development. 
 
 

As described above, the output growth is influenced by productivity growth (y=a +b× r). Tabel 2 shows the empirical result for KBI and KTI with and without Migas. This table explains that the productivity growth influence on output growth is high in KBI and KTI. But KTI has a lower output growth than KBI, so the policy to decrease the interregional disparity is increasing labor productivity and output growths in KTI. Therefore, the government has an important role to increase capability of regional goverments in KTI in order to be more active in the economical development. 

Conclusion and Recommendation

Based on the empirical result of the cumulative causation model, the regional economic growth in whole Indonesia is convergent toward the equilibrium growth as well as in KBI and KTI. The factors influencing on the economic growth of these regions should be the labor productivity growth. But the economic growth of KBI is more convergent than of KTI and Indonesia. This suggests that the economic development of provinces in KBI is more distributed than in KTI. 

The proposed recommendation is to give the highest priority of the economic development to underdeveloped regions with increasing the labor productivity growth by increasing skill and education of the labor. The other ways are creating economic growth regions in KTI based on their potential resources and to make investors interested to go to the East.
 

References

1. Biro Pusat Statistik. 1982. Upah Rata-Rata Buruh yang Bekerja di Sektor Perhotelan, Angkutan Darat, Industri Pengolahan dan Pertambangan. UPSU, 0662, Jakarta
2. ________________. 1996. Statistik Indonesia 
3. Budiono, Uppal. 1986. Regional Income Disparity in Indonesia. Majalah Ekonomi Keuangan, Edisi 30, No.3, Jakarta
4. Mubyarto. 1996. Ketimpangan Pembangunan Antar Wilayah dan Upaya Mengatasinya, Majalah Triwulanan Perencanaan Pembangunan, No. 10
5. Oskar. 1987. Perbedaan Pertumbuhan Ekonomi Regional di Indonesia. Undergraduate Thesis, FE-Unand, Padang
6. Richardson, H. W. 1980. Regional and Urban Economics. Penguin Books 

Syafrizal. 1997. Profil dan Permasalahan Wilayah Indonesia Bagian Barat. Prisma, No. 3

This article was published in:
Proceeding of 5^th Indonesian Student’s Scientific Meeting: p. 49-52, ISTECS Europe & PPI Prancis, Paris Oct. 2000. 
 
 

.