Poverty and Income Inequalities in Developing Countries

1. Poverty and Income Inequalities in Developing Countries –Dr. Quazi Mesbahuddin Ahmed Managing Director Palli Karma-Sahayak Foundation (PKSF) and Former Member, Bangladesh Planning Commission           National Academy for Planning and Development March 05, 2013

2. Issues Covered • Poverty defined  Simple approach versus multi-dimensional approach • Measurement of Poverty in Bangladesh  Different methods of poverty measure in Bangladesh • Income inequality in developing countries  Calculating income inequality in Bangladesh

3. What is Poverty? 11/2/2014 3 A simple definition of poverty is lack of income. But in practice, poverty may mean lack of many things. A multilateral institution defines poverty as “pronounced deprivation in well-being.” The conventional view links well-being primarily to command over commodities, so the poor are those who do not have enough income or consumption to put them above some adequate minimum threshold. This view sees poverty largely in monetary terms. This of course begs the questions of what is meant by well-being and of what is the reference point against which to measure deprivation.

4. What is Poverty? • Different approaches of linking poverty to well-being: • One approach is to think of well-being as the command over commodities in general, so people are better off if they have a greater command over resources. • The main focus is on whether households or individuals have enough resources to meet their needs. Typically, poverty is then measured by comparing individuals’ income or consumption with some defined threshold below which they are considered to be poor. This is the most conventional view (i.e. to view poverty largely in monetary terms, as said before) and is the starting point for most analyses of poverty. Contd.

5. What is Poverty? • Different approaches of linking poverty to well-being: • A second approach to well-being (and hence poverty) is to ask whether people are able to obtain a specific type of consumption good: Do they have enough food? Or shelter? Or health care? Or education? • In this view, the analyst goes beyond the more traditional monetary measures of poverty: Nutritional poverty might be measured by examining whether children are stunted or wasted; and educational poverty might be measured by asking whether people are literate or how much formal schooling they have received. Contd.

6. What is Poverty? • Different approaches of linking poverty to well-being: • The broadest approach to well-being is the one articulated by Amartya Sen (1987), who argues that well-being comes from a capability to function in society. Thus, poverty arises when people lack key capabilities, and so have inadequate income or education, or poor health, or insecurity, or low self-confidence, or a sense of powerlessness, or the absence of rights such as freedom of speech. • Viewed in this way, poverty is a multidimensional phenomenon and less amenable to simple solutions. For instance, while higher average incomes will certainly help reduce poverty, these may need to be accompanied by measures to empower the poor, or insure them against risks, or to address specific weaknesses such as inadequate availability of schools or a corrupt health service. Contd.

7. Poverty, inequality and vulnerability • Poverty is related to, but distinct from, inequality and vulnerability. • Inequality focuses on the distribution of attributes, such as income or consumption, across the whole population. In the context of poverty analysis, inequality requires examination if one thinks that the welfare of individuals depends on their economic position relative to others in a society. • Vulnerability is defined as the risk of falling into poverty in the future, even if the person is not poor now; it is often associated with the effects of “shocks” such as a drought, a drop in farm prices, or a financial crisis. Vulnerability is a key dimension of well-being since it affects individuals’ behavior in terms of investment, production patterns, and coping strategies, and in terms of the perceptions of their own situations. Contd.

8. Why We Measure Poverty? • It takes time, energy, and money to measure poverty, since it can only be done properly by gathering survey data directly from households. Why, then, do we need to go to the trouble of measuring poverty? At least four good reasons come to mind: • Keeping Poor People on the Agenda • Targeting Domestic and Worldwide Interventions • Monitoring and Evaluating Projects and Policy Interventions • Evaluating the Effectiveness of Institutions

9. Why We Measure Poverty? • Keeping Poor People on the Agenda Perhaps the strongest justification is that “credible measure of poverty can be a powerful instrument for focusing the attention of policy makers on the living conditions of the poor.” Put another way, it is easy to ignore the poor if they are statistically invisible. The measurement of poverty is necessary if it is to appear on the political and economic agenda. Contd.

10. Why We Measure Poverty? • Targeting Domestic and Worldwide Interventions • A second reason for measuring poverty is to target interventions. Clearly, one cannot help poor people without knowing who they are. This is the purpose of a poverty profile, which sets out the major facts on poverty (as well as inequality). • One can then examines the pattern of poverty to see how it varies by geography (for example, by region, urban/rural), by community characteristics (for example, in communities with and without a school), and by household characteristics (for example, by education of household head, by size of household). • A well-presented poverty profile is invaluable because the most important operational use of the poverty profile is to support efforts to target development resources toward poorer areas. • A good poverty profile also makes employment targeting possible. Contd.

11. Why We Measure Poverty? • Monitoring and Evaluating Projects and Policy Interventions • The third reason for measuring poverty is to be able to predict the effects of, and then evaluate, policies and programs designed to help poor people. Policies that look good on paper (new opportunities for microcredit for the poor, for instance) may in practice not work as well as expected. • To judge the effects, one would ideally like to monitor the effects of a policy on poor people and evaluate the outcomes in comparison with a control group. Contd.

12. Why We Measure Poverty? • Monitoring and Evaluating Projects and Policy Interventions • Information on poverty is also helpful in understanding the politics of many government policies • By collecting information on households and their economic status, one can assess who uses public services and who gains from government subsidies. If programs are cut or there is retrenchment of the public sector, poverty data provide information on the effects of these plans. The identification of the gainers and losers gives an understanding about who will support, or oppose, a given policy. Contd.

13. Why We Measure Poverty? • Evaluating the Effectiveness of Institutions • The fourth reason for measuring poverty is to help evaluate institutions. • One cannot tell if a government is doing a good job of combating poverty unless there is solid information on poverty. This does not only apply to governments. For example, the World Bank says that “Our dream is a world free of poverty,” and its first mission statement is “to fight poverty with passion and professionalism for lasting results.” The institution’s success in pursuing this goal can only be judged if there are adequate measures of poverty. Contd.

14. Why We Measure Poverty? • Evaluating the Effectiveness of Institutions • When evaluating projects, policies, and instruments, a primary concern is with poverty comparisons between times, households or even countries. In this context, one wants to know whether poverty has fallen (a qualitative measure) and by how much (a quantitative measure). Such comparisons are surprisingly difficult to do well—often they are not robust—and require close attention to issues of measurement.

15. Poverty is multi-dimensional As indicated above, we can identify poverty as multi dimensional and there can be • Income poverty (lack of ‘only' income or consumption) • Human poverty • Human dignity-based poverty (This concept takes into account aspects which make for a dignified human living including, broadly, social, cultural, economic and environmental opportunities and processes, underpinned by equity, ethics, and morality) • Empowerment-based poverty (Empowerment is implies enhancing the capacity of individuals/groups to make choices and to transform those choices into actions and outcomes)

16. Multi-dimensional vs the simple approach The problem of defining poverty, therefore, is a problem of finding a suitable synthesis of the many different dimensions of poverty. In other words, deprivation in one or more of the above dimensions may result in poverty. However, there is no composite index of poverty. To keep things simple, economists restrict themselves to the most widely used and generally acceptable poverty measures which have also been used in Bangladesh: • The direct calorie intake method • The food-energy intake method • The cost of basic needs method

17. DCI method of calculating poverty • A person whose daily calorie intake is less than 2122 k.cal per day is considered to be absolute poor. • A person having less than 80 percent of the above calorie requirement is called a hard-core poor. • The Bangladesh Bureau of Statistics has been compiling poverty data since 1973-74. The data based on DCI method was produced until 1991-92 at different intervals. • In another sense, the DCI method gives the food-share version of poverty line: • Cost of food-energy requirement • Food-share of ‘poor’ 11/2/2014 17

18. The food-energy intake (FEI) method • In the mid-1990s, another method (FEI) was introduced to determine the poverty line where a regression model is used to determine the poverty. • The functional form of the model is: • lny = a + bx + e where, • y = per capita monthly expenditure • (food and non-food) • x = per capita per-day calorie intake • e = disturbance term • FEI method involves finding expenditure or income at which food-energy requirements are met on average. 11/2/2014 18

19. The Cost of Basic Needs (CBN) Method • Poverty measures should be representative and consistent. The direct calorie intake method may not be representative, while the food energy intake method may not be consistent. Different members of the household would require different levels of calorie while for the same reason cost or income for meeting the average energy-requirement of the household would vary across households. • The cost of basic needs (CBN) method, as described below, is expected to be both consistent and representative. Since 1991-92, the Bangladesh Bureau of Statistics (BBS) uses CBN method for calculating poverty in Bangladesh. An upper poverty line and a lower poverty line are calculated. • As prices of some goods and services may vary between geographical areas, poverty lines are estimated at a disaggregated level. Specifically, the country was divided into 14 different geographic areas (9 urban and 5 rural), and poverty lines for each was calculated, with each area denoted by k. 11/2/2014 19

20. Steps in CBN approach • The CBN approach entails three main steps. First, a food bundle is chosen. This bundle is selected in such a way that it provides minimal nutritional requirement corresponding to 2122 k.cal per day per person, the same threshold used to identify the absolute poor with the direct caloric intake method. • The bundle consists of eleven items: rice, wheat, pulse, milk, oil, meat, fresh-water fish, potato, other vegetables, sugar and fruits. • The prices of the food items in this bundle are then calculated. • Prices for each item in the bundle were estimated for the 14 geographic areas. • To capture the price paid by the poor for each food item, regressions were used to control for the impact of household characteristics, e.g. total consumption, education, and occupation on the quality of the food consumed (better off households buy more expensive food than the poor). 11/2/2014 20

21. Steps in CBN approach • Denoting the required quantities in the food bundle to meet the caloric requirement by (F1…FN), where Fj is the required per capita quantity of food item j, food poverty lines were computed as Zkf = ΣPjkFj. In the equation, nutritional needs are the same for all areas, but the prices for each item are area-specific, subscript k referring to area k. • Non food-allowances • The second step involves computing two allowances for non-food consumption. • First, take the average amount spent on non-food items by those households whose total consumption is equal to their food poverty line Zkf. • These households spend less on food than the food poverty line. Hence, what they spend on non-food items must be devoted to bare essentials. • Denoting total per-capita consumption of household i by yi and food per-capita consumption by xi , the lower allowances for non-food consumption were estimated as : ZLkn = E[yi - xi| yi = Zkf ] 11/2/2014 21

22. Steps in CBN approach • The “upper” allowances for non-food consumption are estimated by taking the amount spent on non-food items by those households whose food expenditure is equal to the food poverty line (these households do meet their food requirements). These upper non-food allowances can be expressed as : ZUkn = E[yi - xi | xi = Zkf] • The third step in estimation of the poverty lines consists simply in adding to the food poverty lines the lower non-food allowances and the upper non-food allowances to yield the lower and upper poverty lines for each of the 14 geographical areas: • ZLk = Zkf + ZLkn where ZLkn = E [yi - xi | yi = Zkf ] • ZUk = Zkf + ZUkn where ZUkn = E [yi - xi | xi = Zkf ] • Thus in each area, the estimates of cost of basic food needs are the same with the lower and upper poverty lines. The difference between them arises because of differences in non-food allowances non-food consumption. • ZLk incorporates a minimal allowance for non-food goods while ZUk makes a more generous allowance in this regards. 11/2/2014 22

23. Upper Poverty line Lower Poverty line 1995 2000 2005 2011 1995 2000 2005 2011 National 51.0 49.8 40.0 31.5 34.4 33.7 25.1 17.6 Urban 29.4 36.6 28.4 21.3 13.7 19.1 14.6 7.7 Rural 55.2 53.0 43.8 35.2 38.5 37.4 28.6 21.1 Trends in CBN Head-Count ratio • Head Count Rate (HCR) provides the estimate on the percentage of people living below the poverty line. In CBN method, it is a process of counting the poor on the consumption expenditure threshold and expressed in percentage term. 11/2/2014 23

24. Income Inequality Concept of Income Inequality • The concept of poverty is culturally not physiologically determined (People in a developing country would be happy to live a bit below the US poverty line. Again, the poverty line in USA a hundred years ago was different than it is now). Poverty is thus a relative concept and it seems to be a problem of disparities in income. • The poor are so poor because the rich are so rich. • Nothing in the market mechanism prevents large differences in incomes. On the contrary, it tends to breed inequality. The system of the market mechanism is to reward those who are successful in operating efficient enterprises and to punish those who are unable or unwilling to operate efficiently in the market. • Some economists point to the fact that both inequality and the social tolerance of inequality have substantially increased in almost every society in the post-World War II period (Inequality in our age, Azizur Rahman Khan, Professor of economics, University of California, mimeographed). • In South Asian countries including in Bangladesh, inequality of income has similarly increased during the past decades. 11/2/2014 24

25. Depicting Income Distributions Income distributions are normally calculated in deciles as it is shown below for Bangladesh for 2005 and 2010 Preliminary Report on Household Income & Expenditure Survey- 2010, Bangladesh Bureau of Statistics, June 2011 11/2/2014 25

26. Depicting Income Distributions • The above table shows that the share of income in Bangladesh for the lower five deciles comprising 50 percent of people remains almost same in 2010 compare to that of 2005. • The percentage share of income of the lowest 5% households has slightly increased to 0.78% in HIES, 2010 from 0.77% in 2005. • The income share of top 5% households has significantly declined to 24.61% in 2010 compared to 26.93% in 2005. • The income share of the households belonging to decile-10 has also slightly decreased as compared to 2005. 11/2/2014 26

27. Depicting Income Distributions : The Lorenz curve Statisticians and economists use a convenient tool to portray data shown above graphically. The device, called a Lorenz curve, is shown below. To construct a Lorenz curve, a square is drawn whose vertical and horizontal dimensions both represent 100 percent. Then the percentage of households are recorded on the horizontal axis and the percentage of income that these households received on the vertical axis using the cumulative data from the table shown above. 11/2/2014 27

28. Depicting Income Distributions : The Lorenz curve • Four important properties of a Lorenz curve are: • It begins at the origin because zero families naturally have zero income. • It always ends at the upper-right corner of the square, since 100 percent of the nation’s families must necessarily receive all the nation’s income. • If income were distributed equally, the Lorenz curve would be a straight line connecting these two points (the thin solid line in the box). This is because, with everybody equal, the bottom 20 percent of the families would receive 20 percent of the income, the bottom 40 percent would receive 40 percent, and so on. • In a real economy, with significant income differences, the Lorenz curve will “sag” downward from this line of perfect equally. It is easy to see why this is so. If there is any inequality at all, the poorest 20 percent of families must get less than 20 percent of the income and 40 percent of families would received less than 40 percent of the income and so on. 11/2/2014 28

29. Depicting Income Distributions : Gini coefficient • The Gini coefficient (also known as the Gini index or Gini ratio) is a measure of statistical dispersion developed by the Italianstatistician and sociologistCorrado Gini. • The Gini coefficient measures the inequality among values of a frequency distribution (for example levels of income). A Gini coefficient of zero expresses perfect equality where all values are the same (for example, where everyone has an exactly equal income). A Gini coefficient of one (100 on the percentile scale) expresses maximal inequality among values (for example where only one person has all the income). 11/2/2014 29

30. Depicting Income Distributions : Gini coefficient • The graph shows that the Gini is equal to the area marked 'A' divided by the sum of the areas marked 'A' and 'B' (that is, Gini = A/(A+B)). It is also equal to 2*A, as A+B = 0.5 (since the axes scale from 0 to 1). • The Gini coefficient is usually defined mathematically based on the Lorenz curve, which plots the proportion of the total income of the population (y axis) that is cumulatively earned by the bottom x% of the population (see diagram). The line at 45 degrees thus represents perfect equality of incomes. The Gini coefficient can then be thought of as the ratio of the area that lies between the line of equality and the Lorenz curve (marked 'A' in the diagram) over the total area under the line of equality (marked 'A' and 'B' in the diagram); i.e., G=A/(A+B). • The Gini coefficient can range from 0 to 1; it is sometimes expressed as a percentage ranging between 0 and 100. More specifically, the upper bound of the Gini coefficient equals 1 only in populations of infinite size. 11/2/2014 30

31. Depicting Income Distributions : Gini coefficient • A low Gini coefficient indicates a more equal distribution, with 0 corresponding to complete equality, while higher Gini coefficients indicate more unequal distribution, with 1 corresponding to complete inequality. • The Gini index is defined as a ratio of the areas on the Lorenz curve diagram. If the area between the line of perfect equality and the Lorenz curve is A, and the area under the Lorenz curve is B, then the Gini index is A/(A+B). Since A+B = 0.5, the Gini index, G = A/(0.5) = 2A = 1-2B. If the Lorenz curve is represented by the function Y = L(X), the value of B can be found with integration and: 11/2/2014 31

32. 2010 2005 2000 1995/96 1991/92 1988/89 Total 0.458 0.467 0.451 0.432 0.388 0.379 Rural 0.430 0.428 0.393 0.384 0.364 0.368 Urban 0.452 0.497 0.497 0.444 0.398 0.381 Income Inequality in Bangladesh:growing deterioration • It was seen that income inequality in Bangladesh did not deteriorate during 2005 through 2010. But the previous decade had seen growing worsening of income distribution as shown in the table 11/2/2014 32

33. Thank You 11/2/2014 33