How Corruption Distorts Welfare Economic Policy: A Case of the Cash Transfer Program to Compensate the Kerosene Subsidy Reduction in Indonesia

Karaniya Dharmasaputra
Master of Public Policy Program, the George Washington University, US
Autumn, December 1, 2005

Introduction

Why does the free market economy fail to replicate its success in many developing countries? This question has been haunting developmental economists for many years and the answer is still inconclusive. Nonetheless, it is clear enough today that the traditional economic models are insufficient and that they were built on assumptions that are too good to be true for the Third World context. In underdeveloped economies information is far from free and transaction costs are mostly high. In other words, institution matters.

Within this framework, as I will argue later, another determinant factor should also be taken into consideration: corruption. As evidence clearly shows how corruption hinders economic development, one who focuses his or her works on developing economies must be stunned seeing how economists place corruption merely as a marginal variable. I will try to show the magnitude of how corruption distorts the welfare economic policy by using a case of the cash transfer program in Indonesia.

In this paper, corruption is defined as "the behaviour of private individuals or public officials who deviate from set responsibilities and use their position of power in order to serve private ends and secure private gains" (Lebanon Anti-Corruption Initiative Report 1999 in UNODC website). Thus, the scope of corruption is wider than moral hazard.

The Reform Failures and the Role of Institutions

Evidence has shown how the reform toward a free market economy in Russia and Eastern Europe functioned poorly. Gerard Roland (2002) pointed out that even though those economic systems have been liberalized, success could not be guaranteed. The massive privatization in Russia only wound up in another inequality. Roland concludes that several other factors beside the presence of the free-market also determine the economic success. One of those is the preexisting robust social networks as shown by the pivotal roles of the Catholic Church and the trade unions in Poland’s economic reform. Meanwhile, Robert Nelson (2001) views the lack of social capital, as represented in dishonest and corrupt transactions, as the primary causes of the failure.

In that context, institutional economics offers a more compelling framework. Nobel laureate Douglas North (1994) challenges the standard paradigm saying that institutions do not matter and equilibrium was simply determined by the laws of supply and demand. He emphasizes how low transaction costs are very essential for a productive economy. Yet, most world economies, in fact, are imperfect and are characterized by high transaction cost, and “when it is costly to transact, then institutions matter” (p. 360). In North’s eyes it is not only competition that could reduce transaction costs, but also institutions. Coercive enforcement by a judicial system is one of its important features. As institution grows in a time function through the adaptive-efficiency framework rather than the allocative efficiency, the stagnating economies then can not be automatically transformed only by simply plugging in the free market model.

Parallel with North’s argument, Joseph E. Stiglitz (2001), another Noble laureate, challenges the second too-wonderful assumption of the competitive equilibrium paradigm, free information. He argues that in the real world of economy, where there is no perfect and free information, none of the traditional theorems are relevant. The economies with information imperfections would not ever achieve Pareto efficiency, even though the cost of obtaining the information is taken into account. Transparency, he emphasizes, is one of the determinant factors in economic development.

Institutional reforms therefore become a significant issue. And as listed by Besley and Burgess (2003), promoting accountability of the government is one of the six reforms needed for reducing poverty and inequality.

How Corruption Defects Welfare Economy

Corruption is indeed a clear and present danger for economic development. It blocks information, raises transaction costs, and hence distorts welfare economic policy substantially. A study of the World Bank Institute (2004) estimates more than $1 trillion is paid in bribes every year worldwide, in both developed and developing countries. This figure does not include embezzlement of public funds or theft of public assets and does not mention the losses in investment and the delays this causes in the development of poor countries. A model developed by Paulo Mauro (2005) concludes that a one standard deviation (2.38) improvement in corruption index is associated with over a 4 percentage point increase in its investment rate and over a 0.5 percentage point increase in the annual growth rate of per capita GDP.

Indonesia, the sixth most corrupt country in the world according to the 2005 Corruption Perception Index of Transparency International (CPI-TI), suffered a 30 percent loss, equivalent to Rp.118 trillion, of the total of its national budget in fiscal year 2003. According to the Indonesian Supreme Audit Agency, between 2001 and 2003, irregularities and leakages have affected some $60 billion of the administration and state enterprises’ budget (Bappenas, n.d.).

Corruption also adversely affects income distribution and increases poverty (Gupta et al., 2002). Roughly cross-examining the level of corruption with the poverty rate reveals an interesting insight. For this purpose, this paper uses CPI-TI as the measurement of countries’ corruption and PovcalNet-World Bank for the poverty rate assessment. The 2002 data is used since this is the latest data available in the PovcalNet database. Not all countries indexed by the CPI-TI are also found in the PovcalNet database. Then, only twenty countries found in the PovcalNet database are ranked by the CPI-TI within two groups: the ten most corrupt countries (Tanzania – Bangladesh) and the other ten least corrupt countries (Chile - Lithuania). The CPI indexes the countries from 0 to 10, with 0 representing the most corrupt country. The PovcalNet measures the poor-head percentage of the total population using the “$1 a day” poverty line, the default poverty line of $32.74 per month at 1993 Purchasing Power Parity.

Figure 1 shows the result. Despite several intriguing outliers, in general the graph indicates a positive relationship between corruption and poverty—the less corruption, the less poverty. A more thorough study is certainly needed to examine this rough finding.

[Figure 1 is here]

Theory says that competitive economies leads to a Pareto-efficiency and that Pareto-efficient resource allocation can be reached by compensating variation. The compensating variation for a price change is the size of the budget change under the new conditions that is necessary to compensate the consumer loss and to restore the consumer to the initial utility level. In other words, the amount of compensation must equal the loss in consumer surplus (Friedman, 2002, and Weimer et al., 2005). Yet, as I will show in the following section, corruption would significantly threaten this theorem.

Evidence from the Indonesian Compensation Program

Forced by the skyrocketing international price of oil, on September 30, 2005, the Indonesian government announced a dramatic increase in domestic fuel prices. As of October 1, 2005, the fuel prices were raised by a weighted average 114 percent (World Bank, 2005). The highest increase was on household kerosene price which has risen by 186 percent to Rp.2,000 per liter, up from Rp.700 per liter.[1] This paper focuses on kerosene since in Indonesia, kerosene is a necessity for the poor, especially for cooking and lighting, and thus a dramatic increase will certainly hit their economic conditions.

Those price increases were a result of the government policy that substantially reduced the subsidy on the fuel price. In Indonesia, the fuel prices are administered by the government. As the oil world price hikes, the subsidy burden threatens the country’s monetary and fiscal sustainability. With the fuel price in a range of $60 to $70 per barrel, if the price adjustment was not made, the government fuel subsidy would have reached more than Rp.113.7 trillion or more, and the budget deficit would have expanded to as much as Rp.46.3 trillion or 1.7 percent of GDP (Menneg PPN, 2005; World Bank, 2005).

Realizing the dramatic increase will adversely affect the income of the poor, at the same time the government also launched a cash compensation program. With this unprecedented program, the government projects the compensation program would make the poor to be better off. The government provides a cash transfer of Rp.100,000 per quarter to each of 15.5 million poor households or to 62 million individuals; they who earn Rp.175 thousand per month or Rp.700 thousand per household. The total budget spent for this program is Rp.1.55 trillion per month (Menneg PPN, 2005). According to the 2002 National Socio Economic Survey, the poor consume kerosene 83.3 percent of their total fuel consumption. Thus, proportionately to this, the cash transfer fund to compensate only the kerosene price increase equals Rp.83.300 per month or Rp.1.29 trillion per month in total.

The other main objective of the compensation program is to create a better income distribution scheme. The previous fuel subsidy system heavily benefited the rich rather than the poor. As shown in Table 1, 75 percent of the kerosene subsidy in 2002 went to the middle and wealthy households. This implies that the kerosene subsidy was poorly targeted and tended to increase inequality.

[Table 1: Fuel Subsidy Distribution by Household Group and Figure 2: Impact of the Reduction of Kerosene Price Subsidy]

The impact of the kerosene subsidy and the cash compensation program is depicted on Figure 2. Since the supply price is regulated by the government, the Indonesian kerosene supply is almost perfectly elastic and hence its schedule could be assumed to be a horizontal line. The price increase shifts the supply schedule from S0 to S1.

On the demand side, according to the government’s estimate, every poor household consumes, on average, 30 liters kerosene per month (Menneg PPN, 2005). It means the aggregate monthly consumption of poor and near poor household is approximately 465 thousand kiloliters per month. A World Bank study, using the 1996, 1999 and 2002 Indonesian National Socio Economic Survey (Susenas) data, estimates the price elasticity of demand of the household kerosene in Indonesia is -.602.[2]

The demand function is defined through the slope estimate and the linear approximation methods (Boardman et al., 2001). Knowing the price elasticity of demand is -.602 and considering the analysis is dealing with a large change in kerosene price, from Rp.700 to Rp.2000 per liter, the quantity demanded after the October increase is computed by the Arc formula. The new quantity demanded resulting from the October price hike, holding all else constant, thus would be approximately 220.14 thousand kiloliters per month. Given the price and quantity data computed above, the function of the inverse demand curve equals: P = 3,168.8 - 0.0001 Q.

At this point, our main question emerges: Could the compensation program pay off the consumer loss? Compensating theory says that the Pareto efficient allocation is reached if and only if the consumer loss caused by the policy could be restored by the compensation provided (Friedman, 2002). Thus, the amount of compensation necessary to restore the consumer to the initial utility level must be at least equal to the loss in consumer surplus under the compensated demand curve as represented by the area BCED.

The cash transfer program induces an income transfer among the poor households and thus shifts out the initial demand schedule, D0, and forms a compensated demand schedule, D1. The amount of the cash subsidy pushing out D0 is discounted by the kerosene price increase contribution to the inflation rate. The inflation rate in October 2005 (one month) is 8.7 percent per year (BPS, 2005). Yet, one could fairly suspect that this figure was overstated because of some other interfering factors such as Eid ul-Fitr, the Muslim holiday celebrated on November 3 that always significantly increases the inflation rate. Previous studies also conclude that the fuel price increases would only create a temporary shock on inflation (Ihsan et al., 2005; LPEM-UI, 2005).

Econometric evidence suggests that inflation rate may rise by 0.5 to 0.6 percent a year per 10 percent increase in administered fuel prices (World Bank Jakarta, 2005). Thus, the kerosene price increase, which contributed 49 percent of the total October fuel price increases, would approximately increase the monthly inflation rate by 0.2 to 0.25 percent. The mean value 0.225 percent is used in this analysis.

The net present value of the poor household’s income (after compensated by the government) would be Rp.781.542 per month (= {700,000 + 83,300} / {1 + .225%}). The increase of the real monthly income equals 11.65% (= {781.542 – 700.000} / 700,000). Thus, it could be assumed that the compensated demand function is Q = 596,844,501.4 (1 + 11.65%) -188,349.3 P or P = 3,494.2 - .00001 Q. The new quantity demanded hence would be 289.67 thousand kiloliters per month, the kerosene prices at point A equal Rp.3,538 and the quantity at point E would be 598.19 thousand kiloliters per month. Thus, the estimate of the consumer loss under the compensated demand schedule (BCED) is about Rp.577.1 billion per month. Given the total amount of the cash transfer, Rp.1.29 billion per month, the result would be a net consumer surplus of Rp.711.14 billion per month.

However, corruption threatens that result. Corruption suddenly occurred as soon as the compensation program started. The press reported many complaints by the recipients of how they were forced to give 10 – 50 percent of the aid fund to the local officials in advance as a precondition to obtain the program identification card (Kompas, 2005, October 5-6). Targeting and registration errors exacerbate this problem. Just before the program started Coordinating Minister of the Social and Welfare Affairs admitted that 5-10 percent of the identification cards of the compensation program are not reliable (Kompas, 2005, September 29). The government projected such errors and calculated that corruption would affect at most 28.4 percent of the program beneficiaries (Kompas, 2005, November 1).

Based on this estimate, the potential loss of the cash transfer program is about Rp.365.86 billion per month (= {28.4% * Rp.1.29 trillion}/{1+.225%}). The net consumer surplus then significantly drops to be only Rp.345.3 billion per month or 48.6% less than the initial level (without corruption).

This finding strengthens a study conducted by the Institute for Economic and Social Research, Faculty of Economics, University of Indonesia, or LPEM-UI (2005). LPEM-UI even concluded that the corruption in the compensation program would hit the poor harder than the impact of the previous fuel price increase in March 2005 (only for gasoline and diesel oil). Running a simulation, they concluded that 25 percent leakage of the compensation program to the poor would result in an increase in poverty by 0.55 percent. Meanwhile, the fuel price increase by itself is projected to increase the poverty by only 0.24 percent.

Conclusions

Institutional economics provides a more compelling framework in reforming developing economies. Yet, corruption should be taken into consideration due to its adverse impact in distorting access for information and increasing transaction costs, the two main assumptions of the competitive equilibrium model. Cross examining the level of corruption and poverty in 20 countries reveals an intriguing pattern that corruption positively corresponds to poverty level.

The compensation program in Indonesia shows the magnitude of the distorting effect of corruption to the welfare economic policy. The cash transfer program would result in a net consumer surplus and make the poor households better off even though they have to face a dramatic kerosene price increase. Yet, the current level of corruption and errors (28.4%) is significantly eroding the result and distorting the Indonesian government’s effort to redistribute the fuel subsidy from the high-income household to the poor and establish a better income distribution. ***

Readings:

Bappenas (The Indonesian National Development Planning Agency) (n.d.). Rencana Nasional Aksi Pemberantasan Korupsi 2004 – 2009 (The National Action Plan of Anti Corruption). Jakarta, Indonesia.

Besley, T., and Burgess, R. (2003). Halving global poverty. Journal of Economic Perspectives, 17(3), 3-22.

Boardman, A.E., Greenberg, D.H., Vining, A.R., and Weimer, D.L. (2001). Cost-benefit analysis: Concepts and practice (2nd ed.). New Jersey: Prentice Hall.

BPS (The National Statistics Agency) (2005, November 1). Perkembangan indeks harga konsumen/inflasi (Progress of consumer price index/inflation). Berita Resmi: 52,VIII. Jakarta, Indonesia.

Friedman, L.S. (2002). The microeconomics of public policy analysis. New Jersey: Princeton University Press.

Weimer, D.L., and Vining, A.R. (2005). Policy analysis: Concepts and practice (4th ed.). New Jersey: Pearson Prentice Hall.

Gupta, S., Davoodi, H., and Alonso-Terme, R. (1998, May). Does corruption affect income equality and poverty? Washington, D.C.: International Monetary Fund.

Hope, E., and Singh, B. (1995, March). Energy price increases in developing countries: Case studies of Colombia, Ghana, Indonesia, Malaysia, Turkey, and Zimbabwe. Washington, D.C.: World Bank, Public Economic Division.

LPEM-UI (Institute for Economic and Social Research, Faculty of Economics, University of Indonesia) (2005). Final report: Monitoring economic impact of March 2005 BBM price increase. Jakarta.

Ikhsan, M. (2005, October 3). Kebijakan penyesuaian BBM: Mengapa perlu dilakukan? (Why the fuel adjustment policy is needed?). Pacific Link website. Retrieved October 4, 2005 from: http://kolom.pacific.net.id/ind/lain-lain/mohamad_ikhsan/kebijakan_penyesuaian_bbm:_mengapa_perlu_dilakukan.html

Ikhsan, M., Sulistyo, M.H., and Usman, T.D. (2005, March 21). Kajian dampak kenaikan harga BBM 2005 Terhadap Kemiskinan (A study of the impact of the fuel price increase in 2005 toward poverty). Jakarta: LPEM-UI.

Kompas Daily (2005, September 29). 5-10 persen kartu kompensasi BBM salah sasaran (5-10 percent the fuel compensation card are not valid). P.1

Kompas Daily (2005, October 5). Protes warga miskin berlanjut: Dana kompensasi BBM dipungli (Protests of the poor continue: Fuel compensation fund corrupted), p. 24.

Kompas Daily (2005, October 6). PNS golongan III punya kartu miskin (Civil servant category III obtains poverty identification card).

Menneg PPN (the Indonesian Minister of National Development Planning) (2005, September 25). Subsidi langsung tunai kepada rumah tangga miskin (The cash transfer program for the poor households). Presentation. Bappenas: Jakarta.

Manion, M. (2004). Corruption by design: Building clean government in mainland China and Hong Kong. Cambridge and London: Harvard University Press.

Mauro, P. (2005). The effects of corruption on growth and public expenditure. In Heidenheimer, A.J., and Johnston, M. (Eds.). Political corruption: Concepts & contexts (3rd ed.). New Brunswick and London: Transaction Publisher.

Nelson, R.H. (2001). Economics as religion: From Samuelson to Chicago and beyond. Pennsylvania: Pennsylvania State University Press.

North, D. (1994). Economic performance through time. American Economic Review, 84(3), 359-368.

PovcalNet website. Retrieved November 12, 2005, from: http://iresearch.worldbank.org/PovcalNet/jsp/index.jspl.

Stiglitz, J.E. (2001). Information and the change in the paradigm in economics. The American Economic Review, 92(3), 460-501.

Transparency International (TI) (2002, August 28). Transparency International corruption perceptions index 2002. Retrieved on October 23, 2005, from: http://www.transparency.org/pressreleases_archive/2002/2002.08.28.cpi.en.html

UNODC (United Nations Office On Drugs and Crime). Why a Global Programme against Corruption? Retrieved on November 2, 2005, from: http://www.unodc.org/unodc/en/corruption.html

World Bank (2004, December 8). World Bank denounces the inertia of rich countries in fighting corruption. Press release. Washington, D.C.

World Bank – Indonesia Office. (October 3, 2005). Indonesia: Preliminary description and assessment of the fuel price increase and compensation programs (paper draft). Jakarta.

World Bank (n.d.).Government convenes donors for mid-term CGI meeting. Retrieved on October 27, 2005, from: http://web.worldbank.org/WBSITE/EXTERNAL/COUNTRIES/EASTASIAPACIFICEXT/INDONESIAEXTN/0,,contentMDK:20668606~menuPK:224605~pagePK:141137~piPK:141127~theSitePK

-------------

[1] Based on the Indonesian Presidential Decree No.55/2005, the gasoline price was also increased by 88 percent to Rp.4,500 per liter, and the automotive diesel prices by 105 percent to Rp.4,300 a liter.

[2] The author obtained the data from Director of Poverty Reduction, the National Development Planning Agency (Bappenas), Pungky Sumarno, through an email correspondence on November 10, 2005. The coefficient is referred to Molyneaux, J. (2002). Sectoral elasticity in Indonesia: energy. World Bank Office of Jakarta.