Welfare Analysis Of Controlling Indoor Air Pollution (The Case Of Urban Ethiopia)

WELFARE ANALYSIS OF CONTROLLING INDOOR AIR POLLUTION (THE CASE OF URBAN ETHIOPIA) By: Wondwossen Tsegaye Aselet E-mail: wendsentsgay@yahoo.com Content Page Abstract CHAPTER ONE Introduction 1.1Bakground………………………………………………………………………….4 1.2 Statement of the Problem………………………………………………………….5 1.3 Significance of the Study………………………………………………………….6 1.4 Objectives of the study…………………………………………………………….7 CHAPTER TWO Literature Review 2.1Theoretical Literature………………………………………………………………9 2.2 Empirical literature……… …………………………………………………….15 CHAPTER THREE Methodology of the study 3.1 Model Specification…………………………………………………………… 18 3.2 Data collection………………………………………………………………… 22 CHAPTER FOUR Empirical Results………………………………………………………………….23 CHAPTER FIVE Summary and Conclusion……………………………………………………… 25 References Abstract The main reason for the failure of many policy measures in developing countries is the fact that interaction of different social as well as economic activities are not taken in to consideration before they are being implemented It is important that governments in these countries should focus on addressing welfare, environmental and social aspect of their policy measures before implementing them Thus, mainly this paper focuses on introducing welfare analysis by taking a single policy measure, which helps to control indoor air pollution This study by using the marginal tax reform approach attempts to analyze the welfare effect of indoor air pollution control in urban Ethiopia No study has been done regarding this issue, thus the study tries to show one way of analyzing welfare on the implementation of pollution-controlling-policy instrument Hence, the 2000 urban household socio-economic data collected by the Department of Economics of the Addis Ababa University is used in order to identify goods that are heavily concentrated in the consumption of poor households Accordingly, the study found that introducing or increasing tax on commodities which are the main sources of air pollution (kerosene, charcoal) would affect the welfare of the urban households as the consumption of these commodities are more concentrated on poor than rich households Differently, the study has found that a tax increase or introduction on commodities such as fuel wood and Cigarettes would not have a significant welfare loss for the urban society CHAPTER ONE I Introduction 1.1 Background There are three broad sources of air pollution from human activities; stationary or point, mobile, and indoor air polution (Kathuria, V 2001) Industries, power plants, etc are the causes of stationary air pollution, where as indoor air pollution refers to pollutions from open fires for cooking and heating It is mostly a problem in developing countries, and its effect become intense in rural areas Mobile or Vehicular air pollution is particularly a serious problem in urban areas Since most economic activities or transactions are concentrated in urban Ethiopia, the air pollution that resulted from these three sources is a serious problem However, it seems that, in Ethiopia, compared to other developed countries, there is no serious problem related to air pollution But this is only because of the fact that no body gives a serious attention to the effects of air pollution Moreover, there is no study or reliable data regarding the effects of air pollution on the health condition of the people living in the country In now days, in Addis, vehicles are increasing alarmingly Ethiopians (mostly students and trainees), who have been in other western countries, have a legal- right to import any cars from the country where they have been Following this legal- right most of them import second hand cars (old cars) It is known that these vehicles contribute much for air pollution than the new ones However from the government side there is no measure taken to regulate or decrease the importation of these polluting vehicles Indoor air pollution, which is mostly ignored by the people affected by and government as well has a serious effect on the health of children as well as residents in the house In urban areas of Ethiopia, where large part of the population is using fuel wood, charcoals, and kerosene for cooking purposes in their daily life, it is unreasonable to ignore the air pollution that results from these burnings Almost more than 85 % of the people in urban areas of the country use fuel wood, charcoals and kerosene for cooking as well as for other daily activities Thus the contribution of this process to the environment pollution should not be treated as insignificant Nevertheless, controlling the air pollution resulting from these sources is also a problem in connection with their effect on the living condition of the people For instance, taxing or charging these commodities is one method of controlling the air pollution by discouraging the consumption of these commodities But, since the consumption of these commodities may be more concentrated on poor households than the rich ones the effect of taxing or charging the commodities should be analyzed first I.2 Statement of the Problem Policies that affect consumer and producer prices have an impact on welfare According to Duclos et al (2000), most governments use indirect taxes; mainly VAT, to raise tax revenues thus affecting consumer and producer prices An increase or a decrease in the tax of a good will have an effect on producer or consumer prices and this in turn would entails a decline or an improvement in the welfare of an individual Policy analysts must routinely evaluate the impact of such polices on social welfare Marginal social cost of revenue from the tax on a particular good depends on the rate at which household welfare falls as the tax rate increases, and on the rate at which public revenue rises (Ahmad and Stern 1984) If many people consume more of a particular good and the less responsive is government revenue to a tax increase, then policy makers should focus on decreasing the tax rate on that good If any reform of indirect taxes is to be welfare improving, it requires detailed empirical analysis In Ethiopia, where per capita income is one of the lowest in the world, indirect tax reform should not only focus on targeting higher revenue, but the tasks of tax reform should also include analyzing whether the directions of the reform would not aggravate the poverty of the people In almost all developing countries, the livelihood of the society is dependent on products that are obtained from the environment Based on this fact, it could be argued that there is a need to control for the rapid environmental deprivation, such as reduction in air quality, deforestation, soil degradation and noise pollution Nevertheless, In these countries, where most part of the population is living under the poverty line, environmental assets protection only may not be the crucial thing that government should focus on, rather government or policy makers should also give due emphasis for the improvement of the living condition of the society In this regard, the study attempts to demonstrate empirically the possible welfare impact of introducing tax on indoor pollution sources (fire wood, kerosene and charcoal) Hence, the main task of this paper is to pave a way for making welfare analysis before any tax policy instrument is used for controlling negative environmental impacts Thus, the paper, specifically, focuses on making welfare analysis on an increase or decrease of tax or charge on indoor air pollution sources (fuel wood, kerosene and charcoal) by calculating the distribution characteristics of these commodities The study attempts to measure the distributional characteristics, which reassure how heavily the consumption of each good is concentrated on the poor This measure helps to answer which item should have to have a tax increase or a tax decrease on the ground of distributional equity and poverty alleviation it also helps to identify goods that are good candidate for introducing a new tax 1.3 Significance of the Study Ahmad and Stern (1991) have pointed out that the main purpose of taxation is to raise resources to finance government expenditure They underline the problem of tax design as one of finding a way of raising resources in a manner which is administratively and politically feasible and which promotes equity and efficiency In a similar vein, a recent publication by the IMF (cited in Islam 2003) notes that taxation creates distortions and the main objective of tax policy is to design a system that raises enough revenue while minimizing the level of associated distortions Ethiopia is currently implementing a Tax Reform Program, which includes the replacement of single stage sales tax by Value Added Tax (VAT) The indirect tax reform mainly focuses on raising more revenue, facilitating the economy and modernizing the tax administration However, little is known about the likely effects of the tax reform on the welfare of the society Despite its significance, no study has been done in analyzing the welfare effect of a tax reform in Ethiopia In Ethiopia, particularly in urban areas, air pollution is increasing surprisingly Though there is no reliable figure of the effects of air pollution on health, it would not be difficult to see that air pollution is a serious problem in urban areas In Ethiopia, where Per Capital Income of the population is below the thresholds that people should have for a living, it seems ridiculous to discus environmental issues This is because of the fact that no body gives attention to the effects of pollution, rather they prefer to focus on how to improve their living conditions No study has been conducted regarding the welfare effects of air pollution in urban areas Hence the paper is a pioneer by attempting to investigate whether applying policy instruments, particularly; tax policy to decrease air pollution in the urban Ethiopia would lead to welfare gain This study by using marginal tax reform approach introduces for the first time how welfare analysis should be done before policy instruments are applied for the purpose of curbing environmental damage This would help especially the governments in developing countries to check first the effect of the policy on the welfare of the society and tries to compare whether the gain from the implementation of the environmental policy does compensate the welfare loss Hence, this study by showing the need to make welfare analysis before implementing any policy for the purpose of controlling environmental hazards, addresses the issue of streamlining economic development with environment 1.4 Objectives of the Study The main objectives of the paper are:  To see empirically the welfare effects of controlling indoor air pollution in urban Ethiopia Hence, the paper tries to investigate whether an attempt for controlling air pollution using taxation or charging the source may result in welfare loss of the people.This would be done by identifying goods whose consumption is more heavily concentrated on poor households  The paper attempts to introduce ways of analyzing welfare in an attempt of controlling negative environmental impacts, particularly air pollution  To identify or recommend other ways of controlling indoors air pollution, which may not result in serious welfare loss for the society CHAPTER TWO Literature Review II.1 Theoretical Literature 2.1.1 Welfare Measurements Most people would agree that, other things being equal, increased consumption of goods and services raises individuals’ levels of welfare There may be other factors other than the consumption of goods and services that affect welfare, but since these tend to be much more difficult to measure, economists usually restrict themselves to that portion of human welfare which is attributable to consumption In welfare economics the starting point for measuring welfare is the utility function (Glewwe and Twum 1998) It asserts that welfare rises as the consumption of various goods and services increases But, most consumption and household data are collected at the household level and thus require analysis using a household level utility function rather than individual Hence one should assume: i) A household’s utility is a function of the consumption of goods and services and the composition of household members The composition is needed to account for the fact that house holds with different compositions require different consumption levels to attain the same level of welfare (for example, house- holds with large number of members need more goods and services to attain the same welfare level as households with small household size ii) Each household possesses the same household utility function in order to compare the welfare of different households If they posses different welfare function it would be impossible or meaning less to compare welfare among different categories of people iii) Since one does not know the distribution of welfare with in the household, one has to assume that all household members enjoy the same level of welfare (Glewwe and Twum 1998) When government wants to impose an indirect tax on a commodity it may be better first to find out whether such tax may not significantly increase the price of the commodity and therefore substantially reduce the welfare of people Similar information may be needed with respect to imposition of indirect tax on commodities This necessitates a welfare measure that could help governments or policy makers in their decision on the policy measures Several welfare measures have shown great progress in the last twenty-five years Consumer Surplus: It is the difference of the total amount that an individual is willing to pay and the amount he actually pays Consumer surplus is used to measure individual welfare and social welfare based on its unweighted sum over the population According to Slesnick (1991), consumer surplus is the overwhelming choice as a welfare indicator, because it is an intuitive concept with modest data requirement Slesnick (1991) has shown that in the simplest case of a change in the price of a single good say, a commodity from P01 to P11, the change in consumers surplus simply is the negative of the change in the area under the demand curve P CSk= ∫ x1(t,p2 .pn,Mk,Ak) dt P Where Mk= the level of expenditure of the kth household P= (P1,P2, Pn) is a vector of prices Ak= is a vector of demographic characteristics Xi (P1,Mk,Ak) = the demand function for good i The use of Marshallian consumer surplus as a measure of welfare change is problematic for the following reasons i) It works only for one price change If several price changes or price and income changes, we need a more potent approach for estimating the welfare implication ii) Even for the case of a single price change Marshallian consumer surplus, as a measure of welfare change is problematic on account of maintaining the marginal utility of income as constant 10 In their empirical analysis they also found that for moderate level of inequality aversion, e = 1, λi for cereals is greater than for all commodities except fuel and light and that for clothing has dropped to the third lowest position For highest level of inequality aversion, say e = 5, λi for fuel and light and cereals ranking highest and clothing dropped to the second lowest position Thus they concluded that for the value judgment represented by e = 1, increasing the tax on clothing by an amount sufficient to raise revenue and decreasing taxes on cereals by a corresponding amount increases social welfare Creedy (1998) has examined indirect tax reform in Australia using the method developed by Ahmad and Stern (1984) He calculated the marginal revenue cost for 16 commodity categories He found a change in the ranking as inequality aversion (e), which indicates higher social value for poor household, is increased from a zero level to a positive level Gibson (1998) made empirical analysis on the distributional effect of indirect tax reform in Papua New Guinea He used nationally representative household survey data to identify the items that are consumed mainly by poor household in Papua New Guinea By using the distributional characteristics formula for a good, which is defined by Newberry (1995), he calculated the distributional characteristics for 87 goods and services The study showed that at both low and high level of inequality aversion, sago is the food whose consumption is most heavily skewed towards the poor households When the lowest level of inequality aversion (v = 0.5) is used in the calculation, among the items that have to be purchased and hence could be taxed, axes, bush knifes, and other garden tools, school fees, children's clothing, pots and pans, salt, rice, and finned fish have consumption most heavily skewed towards the poor Adult clothing joins the list when higher level of inequality aversion is used In our country, though much work has not been done regarding the welfare aspect of indirect tax reform, a paper by Munoz and Cho (2003) addressed the poverty and social impact of replacing Ethiopia's sales tax with a value-added tax (VAT) 16 In their study, the calculation of the distributional characteristics of exempted goods showed that health expenditure and public transport, which are exempted items, ranked lowest This showed that these items are not much consumed by the poor The goods disproportionably consumed by the poor were primarily basic food items and traditional fuel sources such as dung cakes and firewood From these results Munoz and Cho (2003) concluded that exempted goods and services not properly serve the purpose of alleviating poverty or improving inequality Marginal tax reform approach, after its empirical analysis pioneered by Ahmad and Stern (1984), has got acceptance from many researchers and policy makers for identifying welfare effects of indirect tax reforms The study by Munoz and Cho (2003) compared the value added tax with the replaced sales tax, in their impact on the poor However, by following the marginal tax reform approach, the current study tries to identify how a tax increase or decrease on environmentally bad items should be directed to improve the welfare of the society Vinish Kathuria(2001), attempted to investigate whether the enactment of policy instruments and efforts have led to commensurate fall in pollution in Delhi or not His results and analysis showed that the imposition has not resulted in concomitant improvement in air quality He found that enacting a policy instruments has not lead to any fall in the air pollution as none of the coefficients attain significance, though coefficients for all the parameter s have the correct sign CHAPTER THREE Methodology And Data Source 3.1 Model Specification 17 This study mainly follows the marginal tax reform approach which was initially developed by Feldstein (1972 cited in Ahmad and Stern 1984) and empirically applied first by Ahmad and Stern (1984) This helps to identify the possible directions of indirect tax or subsidy reforms on environmentally bad commodities that would improve the welfare of the society In this study, calculation of the distributional characteristics for 75 items is done This approach follows Deaton A (1997) theory of marginal tax reform Using this approach the paper attempts to identify whether a tax increase on a specific commodity, such as on fuel wood, kerosene and charcoals, to control air pollution could be effective policy instrument Its effectiveness depends on whether it will decrease pollution with out affecting the welfare of the society If it affects the welfare of the society by increasing the price of the commodities then its effectiveness is questionable Hence alternative policy instruments should be recommended This kind of analysis is necessary especially in developing countries, to asses the issue that a policy change would not lead to an increase in the cost of the living for poor households The social welfare function (w) is a function of the individual welfare value u, which, in turn is given by the value of the indirect utility functions that gives the maximum attainable welfare in terms of prices (p) and Outlays (yh) is: Uh = φ (yh, p) W = V(u1, u2,….uN) A simplistic model of price determination,(Deaton (1997)), assumes that the consumer price is the sum of a fixed (for example, world) price and the tax or subsidy, so that P i = pi o + R i Where Ri = tax or subsidy rate Government revenue (GR) is the sum over all goods and all households of tax payments: GR = M Σ i=1 H Σ Ri qih i=1 18 Where qih = amount of good i purchased by household h and there are m goods Hence revenue with respect to tax change is ∂R/∂ti i== 1Σ qir +i = Σ Σ Ri ∂qih/∂pi j=1 The effect of social welfare is ∂W/∂ti= j =Σ1 ηqih Where η = ∂v ∂φh = ∂w ∂uh ∂xh ∂xh ,is the marginal utility of income The cost - benefit ratio of tax is: λi = ∂W/∂ti = ∂R/∂ti ΣH ηqih Σ qi + Σ Σ ti ∂qih/∂Pi j=1 H j=1 λi is the social cost of raising one unit of government revenue by increasing the tax on good i This figure gives us whether an increase in the tax rate on environmentally bad commodities in order to decrease air pollution would result in welfare decline If this ratio is large, social welfare would be improved by decreasing the price of the good either because its right price is hurting those whose incomes are socially valuable, or because the taxation of the good is discretionary, or both Goods with low λi ratios are those that are candidate for a tax increase When all the ratios are the same there is no scope for beneficial reform By aggregating some of the variables in the social cost benefit ratio the distributional characteristic could be obtained The distributional characteristic (dih) of a good is defined by Newberry (1995) as a measure, which shows how heavily the consumption of each good is concentrated on the poor, which is derived from the marginal tax reform theory and is given by: 19 dih = Σβh qih β*Qi Where βh = The social weight qih = Consumption of the ith good by household β* = Average of the social weights Qi = The aggregate consumption of good i The distributional characteristic will be higher, the more heavily the consumption of good i is concentrated on the socially deserving (i.e those with high social marginal values of consumption, βh) The distributional characteristics is a scale neutral measure, which makes it useful for identifying goods that are candidate for having reduced rates of tax on equity grounds For this calculation, it will be possible to get information about consumption level of each good However the problem lies on how to calculate welfare weight (βh) Welfare weight (βh) According to Ahmad and Stern (1984), policy choices generally involve implicit or explicit trade-offs of gains or losses to different group In practice, estimation of distributional characteristics as well as marginal social cost benefit analysis requires some approximation to the welfare weights, rather than deriving them from estimated preferences The social weight applied to consumption by household h is given by the social marginal utility of consumption And hence it is approximated by: βh = (yh)-e Where Y = income, for which total expenditure can be taken as a better proxy The only uncertainty concerns the appropriate value for the inequality aversion parameter, e Different values of the coefficient e reflect different judgment about the desirability of making transfers to reduce income inequality “The source of a behavioral estimate of e is the information contained in the patterns of private transfers between households Inequality aversion that is equal to zero (e = 0) implies that there is no aversion to inequality on the part of people donating transfers”(Olivia and Gibson 2003) This means that a one birr addition (transfer) to a poor household taken from the 20 rich one doesn’t change the welfare of the two individuals That is the cost to the rich individual by losing one birr is equivalent to the benefit to the poor individual, and thus not alter the distribution of income Inequality aversion which is greater than zero (e > 0) implies that a one birr increase to a poor individual from the rich one would have a higher benefit to the poor than the cost to the rich individual In this case transfers make the distribution more equal, implying aversion to inequality on the part of donors (e > 0) To put it differently, for instance, when inequality aversion parameter is, say, e = 2, taking one extra birr (indirect tax) from a poor household is judged to cause four times as much social cost as does taking one extra birr in tax from a household with twice the consumption level In contrast if e = 0, taking one more birr in tax from the poorest household has the same social cost as taking another birr in tax from the richest household (Gibson 1998) Since calculating this values require greater data need the paper uses, four values of the inequality aversion parameter (e = 0.5, e = 1,e =2 and e= 5) These values are used in many similar studies such as Ahmad and Stern (1991) , Gibson (1998) and others 3.2 Sources of Data Urban household socio-economic survey data of year 2000 collected by the Department of Economics of Addis Ababa University in collaboration with Goteborg University is used for this study Urban households are chosen because of many different reasons, but it is mainly because of the fact that economic activity is relatively more concentrated (transactions are active) in urban areas and also the change in the tax reform will mostly affect the welfare of the urban population directly rather than the rural population The life of the rural population follows hand to mouth subsistence economy and hence may not actively participate in the economic activity of the country A total of 1472 urban households were surveyed Seven major urban centers- Addis Ababa, Awassa, Bahr Dar, Dessie, Dire Dawa, Jimma, and Mekele were selected The household survey contains information about demographic characteristics and income of the household In addition to this information about household characteristics 21 and income, households were asked about their last week expenditure and consumption of different food items from the market The survey also collects data on non-food expenditures over the past months and year Chapter Four Empirical Results By using the urban household survey the study has calculated the distributional characteristics for 75 items The distributional characteristics will help us to identify which items are consumed mainly by the poor Thus, according to the result obtained from own calculation (see appendix), charcoal, kerosene and fuel wood, rank th ,6th and 35 th respectively at a higher inequality aversion (e = 5) This indicates that these items (except fuel wood) are more heavily concentrated on the consumption of poor households and hence it would be inequitable to levy additional or introduce a new tax on these items Though there are other considerations such as health and environment effects that should be analyzed together with this, at least using this methodology, it could be clearly seen that using tax policy instrument for controlling pollution will result in welfare loss of the urban people in Ethiopia Despite the above result, fuel wood, ranks 35 at higher inequality aversion, seems more concentrated in the consumption of rich households 22 This indicates that introducing tax on this commodity would not affect the welfare of the urban society From the appendix, one can see that at a low inequality aversion (e = 0.5) modern medical service rank 12th in terms of its consumption being heavily concentrated on poor households, and ranks 33rd at a high inequality aversion (when distribution does mater) But traditional medicine (healers) rank 70 th at low inequality, and 10th at a high inequality aversion in terms of their consumption being heavily concentrated on poor This result implies that at a high inequality aversion, that is when distribution favors poor households, it is only traditional medicine (healer) that is relatively more concentrated on poor households’ consumption Therefore only traditional medicine (healers) deserves the privilege of being exempted from the newly implemented value added tax From this we could see that even if the negative health effect of kerosene and charcoal is higher on poor households the chance that poor households get modern medical service is much lower than the rich households Poor households prefer to go to traditional healers Thus, with respect to the welfare analysis and according to the result obtained from this study, it could be concluded that government should be careful when a tax is introduced on charcoal and increased on kerosene Further, government should also be selective whether to exempt modern medical service or traditional medicine and healers From the result it could be clearly seen that Value Added tax exemption of modern medical service does not favor equity, since poor households use traditional medical services than the modern ones This result is logical in the sense that as long as one of the main objectives of a tax policy is to maintain equitable income distribution, the good with the highest social cost of welfare should be exempted Excise taxes are levied on goods and services in order to discourage the consumption of luxuries items and goods whose consumption is not favored Cigarettes are the main sources of revenue from excise taxes, but are one of the commodities that could contribute for indoor air pollution Cigarettes ranks 63 rd and 50th at the lowest and highest inequality aversion respectively, in terms of their consumption being heavily concentrated on poor households Hence, this result justifies the fact that it is equitable to 23 increase tax on these items in addition to the need to tax them for the purpose of controlling indoor air pollution CHAPTER FIVE Summary And Conclusion The main aims of the currently implemented indirect tax reform, specifically the Value Added Tax, are to raise government revenue, to facilitate the economy and modernize the tax administration The current reform of indirect taxes is likely to have considerable effects on consumer welfare But due to small number of exempted items under VAT compared to sales tax, VAT seems to contribute to the increase of the cost of living for the poor According to Gibson (1998), indirect tax reform should be designed so as to avoid excessive increases in the cost of living for the poor In this study an attempt has been made to analyze empirically how indirect tax reform should be directed so that it could increase the welfare of the society The study applied marginal tax reform approach for analyzing the welfare effect of introducing or increasing tax on commodities that are the source of indoor air pollution (Kerosene, fuel wood, charcoal and others) Distributional characteristics of 75 goods and services has been calculated In the paper, household urban socio-economic data of 2000 collected by the Department of Economics of the Addis Ababa University has been used 24 The results obtained from the calculation of the distributional characteristics show that kerosene and charcoal are commodities, which have higher social cost ratio, hence from poverty alleviation and distribution point of view these commodities deserve a tax reduction as their consumption is concentrated more on poor households Further, modern medical services, which are exempted from the value added tax, have higher ratio of distributional characteristics, thus rank lower than traditional medical services This implies that since only traditional medicine is concentrated on the consumption of poor households, exempting modern medical service does not contribute to equity and fair resource distribution objective of a tax system REFERENCES Ahmad, E, and Stern N H (1984) The Theory of Tax Reform and Indian Indirect Taxes Journal of Public Economics 28, 259-298 Ahmad E and Stern N H (1991) The Theory and Practice of Tax Reform in Developing Countries Cambridge University Press, Cambridge Atkinson,A.B and Stiglitz,J.E (1980) Lectures on Public Economics McGraw-Hill International Editions-Economic series Belew, F (2001) Fiscal Reform Paper Presented for the 11 th Annual Conference on the Ethiopian Economy Ethiopian Economic Association, Nazareth Bibi, S and Ducles, J (2003) Poverty-Decreasing Indirect Tax Reforms: Evidence from Tunisia Retrieved from www.wider.unu.edu/conference/conference /conference2003.2 Bird M R and Oldman O (1967) Reading on Taxation in Developing Countries The Johns Hopkins Press, Baltimore Creedy, J (1998) Measuring Welfare Changes and Tax Burdens Edward Elgar Cheltenham, UK 25 Creedy, J (1999) Modeling Indirect Taxes and Tax Reforms Edward Elgar Cheltenham, UK Cropper,Maureen L(1981) Measuring the Benefits from Reduced Morbidity: American Economic Review vol 71,no 2,pp235-240 Deaton A (1977) Equity, Efficiency, and the Structure of Indirect Taxation Journal of Public Economics, vol.105, pp 299-312 Deaton, A (1997) The Analysis of Household Surveys: a Micro econometric Approach to Development Policy The Johns Hopkins University Press, London Duclos, J., Makdissi, P and Wodon, Q (2002) Socially Efficient Tax Reforms Retrieved from www.eco.ulaval.ca/w3/recherche/cahiers/2002/0201 Due, J.F (1981) Government Finance and Economic Analysis Second edition, Long man Economic Series, Homewood Gerking,S.and Stanley,L.(1986) An Economic Analysis of Air Pollution and Health: Review of Economics and Statistics,68(1):115-121 Gibson, J (1998) Indirect Tax Reform and the Poor in Papua New Guinea Pacific Economic Bulletin, vol 13, No Glewwe,P and Twum, B.A.(1998) The Distribution of Welfare in Ghana, 1987-88 Living Standards Measurement Study Working Paper No.75 Islam A (2003) Issues in Tax reform Asia-Pacific Development Journal, Vol.8 and No John,M and Smolensky,E (1994) Modern Public Finance Harvard University Press, Cambridge, Massachusetts and London Joseph P Pinto and Lester D Grant(1999), Approaches to Monitoring of Air Pollutants and Evaluation of Health Impacts Produced by Biomass Burning National Centre for Environmental Assessment-WHO Background Papers, 147 Junghun K (2003) The Welfare Effect of Property Transaction Tax Korean Institute of Public Finance Retrived from www.gemini.econ.umd.edu/cgi-bin/conference/ Kathuria,V (2001) Vehicular Pollution Controls in Delhi, India-Are the efforts enough? The Beijer Institute, International institute of Ecological Economics 26 King, M.A (1983) Welfare Analysis of Tax Reforms Using Household Data Journal of Public Economics, Vol 21 pp 183-214 Mulat, T (1979) The Revenue Effectiveness of Ethiopian Coffee Export Taxation Public Finance, Vol xxxiv, No Munoz,S and Cho, S.W (2003) Social Impact of a Tax Reform The Case of Ethiopia, IMF working paper, wp 03/232 Musgrave,R.(1967) Fiscal Systems New Haven and London, Yale University Press Newberry, D and Stern, N.H (1987) The Theory of Taxation for Developing Countries Oxford University Press, New York Newberry, D (1995) The Distributional Impact of Price Changes in Hungary and the United Kingdom The Economic Journal, vol 105(3): 847-863 Olivia S and Gibson J (2003) Equity and Efficiency Aspects of Indirect Tax and Subsidy Reform in Indonesia Retrieved from URL: www.nzae.org.nz/files/ %2358-OLIVIA-GIBSON Ostro, Bart (1992) Estimating the Health and Economic Effects of Particulate Matter in Jakarta: A Preliminary Assessment, Paper Presented at the Fourth Annual Meeting of the International Society for Environmental Epidemiology, 26-29 August Cuernavaca, Mexico Profeta, P (2003) Tax Systems and Tax Reforms in Europe: Public Finance and Political Economy Working paper No 197 197/2003, Societal Italiana Dieconomic Pubblica Pudney, M Markvov, N and Ackrill, R (1990) Indirect Tax Reform in Bulgaria Public Sector Economics Research Center, University of Leicester Sadoulet, E and de Janvry, A (1995) Quantitative Development Policy Analysis The Johns Hopkins University Press, Baltimore and London Schenk, A (2002) On the Implementation of the Ethiopian Value Added Tax Seminar on Proposed Ethiopian Value Added Tax, May , Ministry of Revenue Seifu, M (1998) Performance of The Ethiopian Sales and Excise Tax Addis Ababa University, School of Graduate Studies M.Sc Thesis Slesnick, D.T (1991) Empirical Approaches to the Measurement of Welfare Journal of Economic Literature, vol 37, pp 2108-2165 27 Yeshanew, T (2001) Household Consumption Behavior in Ethiopia An Urban-Rural Comparison using Microeconomic Data M.Sc thesis, Addis Ababa University Zelalem, Y (1998) Revenue Productivity of the Tax System in Ethiopia M.Sc Thesis, Addis Ababa University Appendix -The distributional characteristics of goods and services (Di) Items E=0 Pepper (Berbere) 0.014421 Abish (Local spice) 0.002636 Jinger (Jinjible) 0.001573 Other Spices 0.00394 Milk 0.012927 Cheese 0.005023 Butter 0.051098 Beef 0.071751 Mutton 0.075671 Chicken 0.005891 Eggs 0.007153 Fish 0.001953 Bread (Local bread) 0.000119 Bread 0.03452 Furno Duket( Wheat flour) 0.010911 Pasta (Spaghetti) 0.017492 Enset (Popular food in the southern Ethiopia) 0.001926 Salt 0.007998 Cooking oil 0.036647 Potatoes 0.015865 Tomatoes 0.011164 Carrot 0.006506 Gomen (Cabbage) 0.005447 Keyser (Red Potatoes) 0.001218 Karia (Green Pepper) 0.0047 Oranges 0.006264 Rank 29 56 64 52 32 48 45 42 60 76 11 E=0.5 0.002636 0.000682 0.000687 0.00115 0.002305 0.00064 0.005085 0.00779 0.00866 0.000938 0.001784 0.000388 3.62E-05 0.006969 35 24 61 39 26 34 43 46 66 49 44 Rank 28 57 56 45 33 58 16 47 38 65 76 E=1 0.000827 0.000319 0.000469 0.000613 0.000635 0.000138 0.000778 0.00127 0.00128 0.000215 0.000635 0.000119 1.48E-05 0.002183 Rank 33 52 48 42 40 64 34 23 24 58 40 66 75 12 E=2 5.89E-05 4.33E-05 0.000106 0.000138 2.25E-05 3.92E-06 1.08E-05 1.5E-05 1.6E-05 5.43E-06 2.84E-05 5.44E-06 9.27E-07 0.000107 Rank 36 40 31 25 50 70 59 53 54 65 45 64 75 30 E=5 4.79E-08 6.73E-09 6.26E-08 3.99E-07 7.61E-11 7.73E-12 8.76E-10 8.17E-13 8.18E-13 8.85E-12 3.26E-11 2.9E-11 4.09E-12 3.22E-09 0.001238 43 0.003366 23 0.000299 0.000909 53 30 1.08E-05 2.51E-05 58 47 4.33E-11 62 1.61E-11 66 0.00047 0.002189 0.005668 0.004712 0.003784 0.001722 0.002219 0.000584 0.002353 0.001539 0.000178 0.001373 0.001433 0.002183 0.001934 0.000936 0.001294 0.000395 0.001902 0.000541 60 20 19 12 15 28 22 49 16 44 1.35E-05 0.000297 6.08E-05 0.000182 0.000211 0.00013 0.000176 7.25E-05 0.000573 2.39E-05 56 12 34 17 16 28 19 32 48 2.04E-09 1.43E-07 5.71E-09 2.83E-09 1.18E-08 4.52E-08 4.77E-08 7.59E-08 8.52E-07 4.47E-11 64 36 12 18 20 39 35 60 32 40 28 Rank 29 43 28 18 58 69 51 74 75 68 63 64 72 45 47 23 44 46 39 31 30 26 15 61 Banana 0.005332 Papaya 0.001165 Pineapple 2.74E-05 Avocado 0.00028 Tella (Local beer) 0.002447 Tejj (Local drink) 0.001492 Araki (Local drink) 182458.4 Soft Drinks 0.012879 Beer 0.004688 Other liquor 0.000253 Coffee bean 0.035869 Tea leaves 0.009119 Chat (Stimulant leaves) 0.007348 Cigarettes 0.001672 Sugar 0.039062 Honey 0.008243 Injera (pancake like bread) 0.18257 Clothes /Shoes for men 0.020494 Clothes /Shoes for women 0.01763 Clothes /Shoes for boys 0.01967 Clothes /Shoes for girls 0.018411 Sheets, Towels, Blanket 0.003383 Kitchen Equipments 0.002544 Furniture 0.00898 Lamp/Torch 0.000276 Matches 0.002396 Batteries 0.00056 Candles 0.001119 Bulbs 0.001011 Charcoal 0.016164 Fuel wood 0.017575 Kerosene 0.022535 Laundry Soap 0.014527 Toilet Soap 0.001906 Other Personal care 0.004495 Building Material 0.014261 Housing maintenance 0.019775 Ceremonial expenses 0.080685 Contribution to Iddir (association) 0.020212 Donation to religious service 0.003072 Taxes (excluding income tax) 0.007579 Other contribution 0.003176 Modern Medical service 0.032098 Traditional Medicine and Healers 0.000735 School fees 0.030591 47 67 77 73 58 65 33 50 75 10 36 41 63 38 17 0.001427 0.00031 7.41E-06 4.39E-05 0.000564 0.000254 302096.3 0.00207 0.000317 7.1E-05 0.007797 0.003132 0.001049 0.000296 0.008485 0.000938 0.043943 0.002462 42 68 77 75 62 70 37 67 74 25 46 69 48 31 0.000538 0.000121 2.47E-06 1.23E-05 0.000222 6.67E-05 404875.1 0.000538 5.93E-05 2.72E-05 0.002512 0.001885 0.000235 8.4E-05 0.002776 0.000168 0.018514 0.000479 45 65 77 76 57 71 45 72 74 10 17 56 69 61 47 2.89E-05 7.7E-06 7.27E-08 5.05E-07 1.65E-05 2.29E-06 2751855 1.79E-05 2.35E-06 1.5E-06 0.000133 0.000376 6.27E-06 4.76E-06 0.000148 6.66E-06 0.002126 1.21E-05 44 61 77 76 52 73 51 72 74 27 63 68 24 62 57 9.94E-10 49 9.84E-11 57 77 2.72E-13 76 2.01E-10 54 5.14E-12 70 2.94E+11 1.88E-11 65 5.57E-11 59 4.63E-12 71 2.62E-08 34 4.96E-07 16 1.11E-11 67 8.9E-10 50 1.13E-08 40 3.41E-08 32 1.61E-06 12 1.64E-10 55 22 20 21 53 57 37 74 59 71 68 69 25 23 15 28 62 51 30 19 0.002632 0.002799 0.002882 0.000577 0.000525 0.000621 0.000223 0.002222 0.000374 0.000906 0.000841 0.005137 0.004946 0.006371 0.005488 0.000857 0.001448 0.000809 0.001208 0.006484 29 27 26 61 63 59 71 34 66 49 52 15 17 10 14 51 41 53 44 0.00064 0.000669 0.000721 0.000146 0.000242 0.000101 0.000273 0.003284 0.000359 0.001191 0.00105 0.002397 0.002013 0.002573 0.003047 0.000609 0.000746 9.88E-05 0.000163 0.001358 39 37 36 63 55 67 54 50 25 26 11 14 43 35 68 62 21 2.31E-05 2.57E-05 3.34E-05 4.14E-06 4.6E-05 4.89E-06 0.000173 0.003216 0.000136 0.001007 0.0006 0.000309 0.000152 0.000234 0.000444 0.000154 0.000115 4.81E-06 7.93E-06 6.08E-05 49 46 43 69 39 66 20 26 11 23 15 22 29 67 60 35 1.01E-10 1.21E-09 8.71E-08 3.23E-12 6.96E-08 3.29E-10 2.3E-05 0.000724 4.31E-06 4.02E-05 6.06E-06 3.15E-06 1.6E-08 2.24E-05 3.3E-07 1.7E-07 1.58E-06 5.67E-10 1.33E-08 9.45E-09 18 0.006169 11 0.002746 0.000272 13 3.14E-07 21 55 0.000891 50 0.000666 38 0.000334 10 0.000288 40 54 12 0.000776 54 0.000726 55 0.003703 21 0.00018 0.000351 0.000919 59 51 29 1.43E-05 6.92E-05 5.62E-05 55 33 38 1.42E-08 36 4.64E-07 17 2.92E-08 33 70 13 0.00017 72 0.003592 22 7.92E-05 0.000901 70 32 3.49E-05 5.74E-05 42 37 1.76E-06 11 1.02E-08 41 29 56 48 25 73 27 53 35 20 22 13 52 38 42 Other education 0.014218 31 0.002594 30 0.000943 27 0.00016 Water bill 0.015528 27 0.005583 13 0.003071 0.000437 Electricity 0.047361 0.009947 0.00343 0.000259 Telephone bill 0.020584 16 0.003311 24 0.000904 31 4.28E-05 Rent 0.025404 14 0.004423 19 0.00175 18 0.000179 Source: Calculation based on urban socio-economic data, 2000,of Addis Ababa University 21 14 41 18 2.48E-06 3.88E-07 1.5E-06 1.34E-07 1.37E-08 30 10 19 14 24 37