UNIT 4 POVERTY AND INEQUALITY: POLICY IMPLICATIONS

Transcription

1 UNIT 4 POVERTY AND INEQUALITY: POLICY IMPLICATIONS Structure 4.0 Objectives 4.1 Introduction 4.2 The Concept of Poverty 4.3 Measurement of Poverty Income Indicators of Poverty Indicators Covering Income and Non-income Dimensions of Poverty 4.4 Dimensions of Poverty in India Income Poverty Indicators Indicators Covering Income and Non-income Dimensions of Poverty 4.5 Concept of Inequality 4.6 Measurement of Inequality Measures of Inequality of Income Indicators Covering Inequality in Non-income Aspects of Life 4.7 Levels of Inequality Levels of Inequality in Income and Consumption Levels of Inequality in Non-income Aspects of Life 4.8 Policy Implications 4.9 Let Us Sum Up 4.10 Exercises 4.11 Key Words 4.12 Some Useful Books 4.13 Answers or Hints to Check Your Progress Exercises 4.0 OBJECTIVES After reading this unit, you shall be able to: define the concept of poverty; state different income and non-income indicators of poverty; assess the dimensions of poverty in India; define the concept of inequality; state the different methods of measurement of inequality of income; explain the levels of inequality; and state the policy implications of poverty and inequalities. 4.1 INTRODUCTION As seen in unit 2 earlier, the Indian economy has expanded and diversified considerably since the advent of planning. The Gross Domestic Product (GDP) grew at the average annual rate (compound rate) of 3.2 per cent in the 1960s and 1970s. The economy moved to a higher growth path thereafter to 5.8 per cent per annum in the 1980s and 5.7 per cent per annum in the 1990s. This momentum of growth has more or less been maintained during the first half of the current decade. The share of the agricultural sector in total GDP has declined from about 55 per cent to about one-fourth and that of the 50

2 services sector has increased from about 30 per cent to over 50 per cent over the fifty-year period. Has such economic expansion and diversification led to the socio-economic well-being of the people of the country? This does not seem to be the case for sizeable sections of the people. About one-fourth of the Indian population of over a billion is poor. A significant proportion of labour force remains unemployed, there are gross inequalities in distribution of income. In the present unit and the following unit 5, we address ourselves to these issues of economic development. Poverty and Inequality: Policy Implications 4. 2 THE CONCEPT OF POVERTY Poverty is a multidimensional concept. Poverty may be defined as a state of lack of access to the basic needs of income, food, shelter, education, health services, safe drinking water and sanitation that help an individual lead a decent, normal and effective existence. Indeed, the list of basic and other needs may vary, depending upon the society in question and what, in its view, constitutes normal and effective existence. The next question that comes up is about how lack of access is understood. Is a total lack of access to food (hunger), etc., necessary to be considered as being in a state of poverty? Or is there a minimum desirable level of access, only below which an individual will be considered as being poor? These questions lead one to the realm of estimating the incidence of poverty or the number who are poor and the related quantitative indices of poverty or deprivation and measures of human development. We shall deliberate upon these issues in the next section. 4.3 MEASUREMENT OF POVERTY Income Indicators of Poverty The first step in estimating the incidence of poverty is to define a poverty line. The Task Force on Projections of Minimum Needs and Effective Consumption Demand of the Planning Commission (1979), used an average energy (nutritional energy) requirement norm to define the poverty line. Since calorie is the unit of energy, the norm used was in terms of calories. The Task Force estimated the average daily per capita requirements for rural and urban areas by using the specific calorie allowances recommended by the Nutritional Expert Group (1968) for population groups of different age groups, sex and activity. In this manner, the Task Force attempted to capture in the average norms factors such as age, sex and occupational differences in the daily calorie requirement of the population. The calorie norms, thus, derived were rounded off to 2,400 calories per capita per day for rural areas and 2,100 calories per capita per day for urban areas. The monetary equivalents of these norms were obtained by using: (i) data on the quantity and value of items of household consumption and (ii) the calorie content of the items of food consumed by population groups belonging to different per capita expenditure classes with appropriate conversion factors. The Task Force, thus, estimated, on the basis of the observed consumer behaviour in , that, on an average, a consumer expenditure of Rs per capita per month was associated with a calorie intake of 2,400 calories per capita per day in rural areas and Rs per capita per month with a calorie intake of 2,100 per capita per day in urban areas. In other words, the poverty line was defined as the per capita expenditure level at which the average per capita per day calorie intake is 2,400 calories for rural areas and 2,100 calories for urban areas. This poverty line serves as a cut-off line for separating the poor from 51

3 Framework of Indian Economy 52 the non-poor if the distribution of population with per capita expenditure below the level defined by the poverty line is counted as poor. The proportion of the poor to the total population is the Poverty Ratio (PR) or the Head Count Ratio (HCR). PR or HCR measures the incidence of poverty. It is, thus, defined as: number of people below poverty line Incidenceof Poverty = PR = HCR = 100 total population It is expressed as a percentage. This is useful for comparing the poverty situation in two areas like the rural and urban areas or different States or the situation in an area in the year 2005 compared to, say, The computation of the poverty line for the base year ( ) has been done with prices of items in the base year. This is, therefore, updated for changes in prices over time. This is, then, used with the distribution of population by different per capita consumer expenditure classes available from time to time from periodic surveys of the NSSO on household consumption to arrive at estimates of the number of the poor and the poverty ratio for subsequent years. The measurement of poverty described above, namely, the poverty ratio or the head count ratio, is simply the proportion of the number of people below the poverty line in the population. This ratio, however, does not make any distinction within the broad category of the poor on the basis of their actual levels of consumption and deprivation. Consequently, the poverty ratio fails to capture the depth and severity of poverty in an adequate manner. A measure developed for this purpose is the Poverty Gap (PG) Index. The PG Index calculates the total shortfall of consumption below the poverty line, per capita of the total population. This is, then, expressed as a percentage of the poverty line. It can also be calculated as: (Povertyline Per capita consumption of the poor) PG Index = Poverty Ratio 100 Povertyline More comprehensive measures of the severity of poverty are the Squared Poverty Gap (SGP) and the Sen Index (SI). We shall not go into the formulae for these measures except to observe that: (i) SGP is not PG x PG, (ii) it possesses the properties of both the Poverty Ratio and the Poverty Gap Index and (iii) in addition, it also captures the extent of variation in the levels of consumption of the poor. It is, however, sensitive to measurement errors at the bottom of the per capita expenditure scale. The Sen Index takes note of the shortfall in average consumption of the poor from the poverty line as well as the inequality in consumption among the poor Indicators Covering Income and Non-income Dimensions of Poverty Poverty Ratios (PR) and measures related to PR provide a composite picture of people whose per capita consumption expenditure is below the level of per capita consumption expenditure corresponding to the basket of commodities constituting the desired minimum. These do not, however, provide a complete picture of the extent of deprivation or, alternatively, the state of well-being of the population. These are rooted in calorie consumption and do not say anything about several other factors that shape living standards, like: (a) the health status of the population like longevity, overall mortality, infant mortality, maternal mortality (mortality of women arising from child birth and related causes) and morbidity (prevalence of diseases) and in general, access to health services, (b) the nutritional status, (c) the educational status and

4 (d) the living environment like housing, access to safe drinking water and sanitation as also pollution-free air and water resources. Attempts have been made to capture these aspects in alternative measures of poverty. Let us have a brief look at them. a) The Human Development Index (HDI) and the Human Poverty Index (HPI) As discussed in unit 3, the UNDP has been preparing Human Development Reports (HDRs) and making estimates of the Human Development Index (HDI) for different countries since The HDI incorporates three most critical and socially useful choices, viz., i) the choice to lead a long and healthy life; ii) the choice to acquire knowledge; and iii) to have access to the resources needed for a decent level of living. The countries are ranked in order of the value of the HDI. HDR also presents estimates of Human Poverty Index (HPI). This being a measure of deprivation, HPI makes use of the following for the three areas of choices referred to above: i) Proportion of population not expected to survive beyond 40 years; ii) Adult illiteracy rate; and iii) (a) Percentage of population without sustainable access to an improved water source; and (b) Percentage of children aged 5 or below who are underweight for their age. The National Human Development Report, 2001 prepared by the Indian Planning Commission follows the framework of human development adopted in the UNDP HDR. It presents estimates of HDR for 1981 and 1991 for the country and different States and Union Territories. It also gives estimates of HDI for 2001 for the country and for 15 major States. Estimates of HDI for the other States and Union Territories could not be prepared due to lack of comparable data for these States for b) Gender-related Development Index (GDI) or Gender Equality Index (GEI) The Human Development Index (HDI) that we have discussed so far is based on indicators reflecting economic, educational and health attainments of the population. It does not, however, reflect gender-based disparities in such attainments. Gender-based discrimination is prevalent in every society developed or not to a lesser or greater degree. Such discrimination results in a higher incidence of poverty in the female population than in the male population, in whatever manner we measure poverty. Gender-related Development Index (GDI) or Gender Equality Index (GEI) seeks to reflect gender disparity in human development. This will help to focus attention on aspects of development planning that fail to reduce gender discrimination. This index is estimated as a proportion of economic, educational and health attainments of females to that of males. The common set of variables for which the attainments of females and males are compared is the same set that is used in estimating HDI. c) Capability Poverty Measure (CPM) UNDP HDR 1996 had also developed a Capability Poverty Measure (CPM) for different countries. Three indicators, (i) the percentage of children under 5 Poverty and Inequality: Policy Implications 53

5 Framework of Indian Economy who are underweight (ii) the percentage of births unattended by trained health personnel and (iii) the percentage of women aged 15 years and above who are illiterate, were used for computing CPM. Check Your Progress 1 Note: i) Space is given below each question for your answer. ii) Check your answer(s) with those given at the end of the unit. 1) In what way is the PG Index more useful in assessing the poverty situation? 2) What are the indicators which take note of income as well as non-income aspects of poverty? 3) What are the indicators on which Human Development Index and the Human Poverty Index estimated by the NHDR, 2001 of the Planning Commission are based? 4) What are the indicators of Capability Poverty Measure (CPM) prepared by NCAER? 4.4 DIMENSIONS OF POVERTY IN INDIA Income Poverty Indicators The incidence of poverty has declined from 56.4 per cent in to 27.1 per cent in in rural areas and from 49 per cent to 23.6 per cent during the same period in urban areas (See Table 4.1). The decline is from about 55 per cent to 26 per cent for the country as a whole. While the number of the rural poor declined, the rise in the number of the urban poor had been more than compensating for it till Thereafter, the number of urban poor has also declined. As of 2000, a little over one-fourth of the population of the country is poor. These 260 million people do not have the purchasing power

6 needed to meet the specific standard of calorie intake with some margin for non-food consumption needs. About three-fourths of these are in the countryside. About one-half of the population of Orissa and about two-fifths of the population of Bihar and Madhya Pradesh do not have even this minimum level of purchasing power (Table 4.2). Rural as well as urban poverty is most severe in Orissa, with almost of its rural population and twofifths of its urban population being poor (Table 4.2). Year Table 4.1: Poverty Ratio and the Number of Poor Poverty Ratio (per cent) No. of Poor (million) Rural Urban Combined Rural Urban Combined (1) (2) (3) (4) (5) (6) (7) Source: 1) Report of the Tenth Five Year Plan ( ) Steering Committee on Urban Development, Urban Housing and Urban Poverty, 2001, Planning Commission. 2) National Human Development Report, 2001, Planning Commission. Table 4.2: Poverty in India: Poverty and Inequality: Policy Implications Sl. No. Rural Urban Total State No. Poor Poverty No. Poor Poverty No. Poor Poverty Ratio Ratio Ratio (Million) (Per cent) (Million) (Per cent) (Million) (Per cent) (1) (2) (3) (4) (5) (6) (7) (8) Major States 1. Andhra Pr Assam Bihar Gujarat Haryana Himachal Pr Karnataka Kerala Madhya Pr Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal

7 Framework of Indian Economy Other States 17. Arunachal Pr Goa Jammu & Ka Manipur Meghalaya Mizoram Nagaland Sikkim Tripura A & N Islands Chandigarh D & N Ha Delhi Pondicherry Lakshadweep India Source: National Human Development Report, 2001, Planning Commission. Incidence of Poverty among the Scheduled Castes and Tribes. Poverty ratios for these two groups for and are shown in Table 4.3. Roughly, two-fifths of the Scheduled Castes and Scheduled Tribes are poor, although there has been some reduction in the incidence of poverty in these groups of population over the six-year period ( to ). But incidence of poverty among Scheduled Tribes residing in rural areas is still very high. About one-half of these people are poor. Table 4.3: Incidence of Poverty among Scheduled Castes and Tribes ( & ) Group R U R U General Population Scheduled Castes Scheduled Tribes Source: Tenth Five Year Plan Vol. II, Planning Commission. 56 We referred earlier to the inadequacy of the poverty ratio in measuring the depth and severity of poverty. The poverty ratio does not give any idea of the distribution of the poor by levels of consumption. It does not say anything about, for instance, the number of the poor whose consumption levels are just about half of the poverty line consumption level. We have seen earlier how the Poverty Gap (PG) Index is able to take note of these aspects and show the magnitude of the effort required to raise the levels of consumption of all the poor below to the consumption level signified by the poverty line. We also noted that the Squared Poverty Gap (SPG) and the Sen Index (SI) are more comprehensive measures that reflect the severity of poverty. The position in India in the decade of 1990s, as reflected by the three measures, is shown in Table 4.4. We note that while the Poverty Ratio, the PG Index and the Sen Index have declined in 2000 to about 70 per cent of their levels in 1994, the

8 SPG Index has come down more sharply to about two-thirds its level in The depth and severity of poverty has declined as fast as incidence of poverty. Poverty and Inequality: Policy Implications Table 4.4: Some Alternative Indices of Poverty & Poverty Index Rural Urban Combined Rural Urban Combined (1) (2) (3) (4) (5) (6) (7) PG Index SPG Sen Index Source: Sundaram, K., Keynote Address in the 38 th Econometric Conference, January, Session of the Indian Indicators Covering Income and Non-income Dimensions of Poverty We have so far looked at levels of poverty as reflected in the poverty ratios. We noted earlier that poverty ratios do not reflect certain aspects of the living standards like health, education and nutrition and that the Human Development Index (HDI), the Human Poverty Index (HPI) and the Capability Poverty Ratio (CPM) are some of these attempts. Let us know what do these say about levels of living? a) Human Development Index (HDI) and the Human Poverty Index (HPI) India s rank on the basis of HDI is 127 out of 177 countries, according to the UNDP Human Development Report (UNDP HDR), HDI for India has increased by about 40 per cent between 1975 and Norway is at the top of the table of country-wise HDI values and Sierra Leone at the bottom. HDI for the developing countries as a whole is well above that for India. Pakistan s rank is 165 and Sri Lanka s is 34. HPI, 2001 for India is less than that for our neighbours except Sri Lanka, Myanmar and Mauritius, according to UNDP HDR, India s rank on the basis of HPI according to the UNDP HDR 2004 is 53 out of about 100 countries. Mauritius, Sri Lanka and Myanmar are relatively better off than India. Inter-State Variations Estimates of HDI for 2001 and those of HPI for 1991 for different States and Union Territories are shown in Table 4.5. Let us first look at the extent of deprivation across the country as revealed by HPI. Note that new areas get added to the list of high poverty areas given earlier on the basis of poverty ratio. The entire Northeastern Region except Mizoram, Rajasthan, Madhya Pradesh, Uttar Pradesh, Bihar, Orissa and Dadra and Nagar Haveli have a relatively high level of poverty as denoted by HPI. States and Union Territories with a relatively low level of human development are generally seen to have a relatively high incidence of poverty (HPI). This inverse relationship is seen to be very strong if we take into consideration only the 16 major States listed. 57

9 Framework of Indian Economy Table 4.5: Human Poverty Indices (HPI) and Human Development Indices (HDI) for States & Union Territories, 1991 & 2001 Sl. No. States & Union Territories HPI1991 % Rural HPI 1991 % Urban HPI 1991 % Total HDI 1991 (1) (2) (3) (4) (5) (6) (7) 1. Chandigarh Ne 2. Delhi Ne HDI Kerala Goa Ne 5. A & N Islands Ne 6. Pondicherry Ne 7. Mizoram Ne 8. Daman & Diu Ne 9. Manipur Ne 10. Lakshadweep Ne 11. Nagaland Ne 12. Punjab Himachal Pr Ne 14. Tamil Nadu Maharashtra Haryana Gujarat Sikkim Ne 19. Karnataka West Bengal Jammu & Kashmir Ne 22. Tripura Ne INDIA Andhra Pradesh Meghalaya Ne 25. Dadra & Nagar Hv Ne 26. Assam Rajasthan Orissa Madhya Pradesh Arunachal Pradesh Ne 31. Uttar Pradesh Bihar Ne: not estimated. Source: National Human Development Report, 2001, Planning Commission. 58

10 b) Gender-related Development Index (GDI) or Gender Equality Index (GEI) UNDP HDR 2003 estimates GDI 2001 for India as 57.4 per cent and India s rank with reference to this index as 103 out of 175 countries. In other words, the attainment of women in the three dimensions covered by HDI is only about three-fifths of those of men. The attainment of women in human development dimensions covered by NHDR HDI (income, health and education) is only about two-thirds of that of men. Gender equality was the highest in Kerala (87.2 per cent) followed by Manipur (80.2 per cent), Meghalaya (79.9 per cent), Himachal Pradesh (78.3 per cent) and Nagaland (78.3 per cent) in the 1980s. It was the highest in Himachal Pradesh (85.8 per cent) in the 1990s and the least in Bihar (46.9 per cent). Estimates of GEI showed that women were generally better off in Southern India than in the Indo-Gangetic Plain, especially in Bihar and UP. Poverty and Inequality: Policy Implications c) Capability Poverty Measure (CPM) NHDR 2001 provides the basic data required for making estimates somewhat similar to the estimates of CPM. These estimates are given in Table 4.6. The extent of inadequate physical growth among women and children is alarming. Table 4.6: Some Measures Relevant to Capability Poverty: (Percentage) Measure Children who are underweight for age Children who are underweight for height Children who are not tall enough for their age INDIA Highest Value among States & Union Territories (Madhya Pradesh) (Orissa) (Uttar Pradesh) Women with BMI less than 18.5 kg./m 2 (Orissa) Lowest Value among States & Union Territories 20.6 (Sikkim) 5.3 (Haryana) 18.1 (Goa) 10.7 (Arunachal Pradesh) Source: National Human Development Report, 2001, Planning Commission. We have considered several measures of poverty and estimates of the levels of poverty based on these. One set of measures is income based. The most important among these is the poverty ratio, which gives the incidence of poverty with reference to the (lack of the) purchasing power required to afford the minimum desirable standard of calorie consumption. Another set attempts to capture the extent of deprivation that the poor suffer in the matter of access to food, safe drinking water, sanitation, medical attention, shelter, education and health and nutrition to ensure longevity. The third consists of those that measure levels of human development in the population, as this is one of the important instruments for tackling poverty. A part of this set is also the one that looks at gender disparity in human development. The fourth is based on the lack of the capability to attain a specified minimum desirable standard of living. And we find that the inability to access the basic needs of living and the low levels of human development and capability usually form part of a vicious circle. Only a development policy that is capable of cutting this vicious circle can effectively solve the problem of poverty. Let us now turn our attention to the problem of inequality. 59

11 Framework of Indian Economy Check Your Progress 2 Note: i) Space is given below each question for your answer. ii) Check your answer(s) with those given at the end of the unit. 1) What is hunger ratio? 2) Describe the trends with reference to incidence of poverty. 3) What is the incidence of poverty among scheduled castes and tribes? 4.5 CONCEPT OF INEQUALITY While the concept of poverty is rooted in the lack of access or a low level of access to food, nutrition, shelter, education and other services, inequality is related to unequal access or different degrees of access of different individuals or groups of individuals to these opportunities, services and benefits. Inequality is, thus, a more general concept than poverty. It looks at the relative levels of access of different groups to development opportunities and benefits. The different levels of access in the concept of inequality also include the low level of access below which people are considered poor. In fact, the low level of access or the limit (like for example, the calorie limit for consumption) that may be set for defining poverty will itself include a number of lower levels of access MEASUREMENT OF INEQUALITY Measures of Inequality of Income a) Lorenz Curve The most simple way to represent inequalities is called the Lorenz Curve. To draw a Lorenz Curve, we take the cumulative percentages of the population and their corresponding shares in the total income of all individuals. On x- axis, we represent shares in population and on y-axis the corresponding shares in total income. The resultant graph called the Lorenz Curve (LC). 60

12 Lorenz Curves on Consumer Expenditure - Rural and Urban India: Lorenz Curve: Poverty and Inequality: Policy Implications Share of Consumption Population Shares eg. Line U R We notice that the points of the curve corresponding to eg. Line in the figure lie on a straight line from the origin in the southwest corner of the graph and moving to the northeast corner. The percentages of population and the corresponding shares of these groups in total income are equal at any point on this line. This straight line, therefore, represents the line of complete equality or the egalitarian line. The graph representing the actual shares, however, appears as a loop joining the two corners of the graph and lying below this line of equality. The extent to which the loop deviates from the line of equality represents the extent of inequality. The situation of extreme inequality is given by the X-axis from the origin to the point representing 100 per cent on the X-axis and the perpendicular line joining the 100 per cent point on the X-axis and the (last) point on the northeast corner of the line of equality corresponding to this point. This is the situation where all the incomes generated in the economy accrue to one individual. The Lorenz Ratio of Inequality or the Lorenz Concentration Ratio (LCR) is given by the ratio. (area enclosed by the line of equality and the LC) LCR = (area of the right angled triangle formed by the X-axis, the line of equality and the perpendicular line standing on the X-axis at the 100 per cent point) This can be arrived at by computing the area enclosed by LC, the X-axis and the perpendicular line referred to earlier and subtracting it from the area of the right angled triangle. The required ratio is then easily calculated. A comparison of the extent of inequalities (of income) between two economies or the trends in the degree of inequality in an economy over time is possible with the Lorenz Curve. Let us draw two such LCs, say for the rural and urban areas of India. Let one Lorenz Curve, say the curve for the rural areas lie completely within the other (the urban curve) all through the range of values. (This is the case in the LCs drawn as an illustration see Graph). That is, one LC dominates the other. This is called Lorenz domination. Clearly, the inequality in the society corresponding to the curve nearer to the line of equality is less severe than the inequality in the society corresponding to the outer curve. Problems of comparison, however, arise when the two (or more) Lorenz Curves intersect, when it is not possible to arrive at a conclusion regarding which curve depicts a worse inequality situation than the other. 61

13 Framework of Indian Economy Summary measures of inequality will be useful in such cases. To these we turn now. b) Summary Measures of Inequality The measures given below are ratios and, therefore, dimensionless, that is, free of scale. These measure relative inequality in distribution, that is, these will not be affected by any proportional change in all incomes. Let the incomes of P individuals in a society be Y 1, Y 2,. Y i,.., Y P, where Y i is the income of the i th individual in the population. The mean income (M) of all the individuals will be M = ( Y i ) P. The mean income M does not tell us anything about the extent of inequality in incomes in this society. Let us then look at other measures that can help us in this regard. Let the individuals be arranged in the increasing order of their incomes. The Median Income M e is the income of the middlemost person in the ranking of individuals in the increasing order of their incomes, that is, M e = Y (P + 1)/2, if Y k is the income of the individual whose rank is k. M e shows that the incomes of one-half of the population are below it. This is otherwise not a useful measure of inequality: it does not reduce to zero in the case of equality of all incomes. 1) Range of the incomes R = the difference between the highest individual income and the lowest individual income divided by the mean income M, that is, R = (Maximum of Y i Minimum of Y i )/M. R is zero when all incomes are equal and is equal to P when all the incomes accrue to a single person. 2) Relative Mean Deviation RMD = (1/PM) ( Y i M ), where x stands for the absolute value of x (see footnote) 1 and M the mean income. 3) Gini Ratio (GR) is defined as one half of the relative mean difference. 2 Thus, GR = ( Y i - Y j ) (2 P P M), where the summation is done first with reference to j = 1, 2,...P and then with reference to i = 1, 2,..P. 3 GR can also be expressed as follows: GR = 1 + (1/P) 2 [PY 1 + (P 1) Y Y P-1 + Y P ]/(P P M) where Y 1 Y 2 Y P. 4 (Individual incomes arranged according to increasing order of income) 62 1 For example ( - ) 3 = 3. 2 Mean difference is the arithmetic mean of the absolute differences between all pairs of incomes (P P pairs) and is divided by M to get the relative mean difference. It is equal to ( Y i Y j )/(P P M). 3 For example, Y i Y j for i= 1, 2 and j = 1, 2 will be Y 1 Y 1 + Y 1 Y 2 + Y 2 Y 1 + Y 2 Y 2. 4 less than or equal to.

14 The alternative formula shows that GR involves a weighted sum of the incomes of individuals where the weights are the ranks of individuals when they are arranged in the increasing order of their incomes; but the poorest individual gets the biggest weight P, the next individual who is better off than the first gets the weight (P 1) and so on and finally, the richest individual gets the weight one. Other ratio measures can also be used. The distribution of income derived in the manner described under the section on Lorenz Curve can be used to compute simple ratio measures of inequality. For instance, we can compare the shares in total income of the richest (top) 10 per cent of the population and the poorest (bottom) 10 per cent of the population (Decile Ratio). Or we might look at the ratio of the share in total income of the top half (50 per cent of the population) of the income distribution to that of the bottom half. Clearly, the measures mentioned above will give different values for a given distribution of income. The important question to be considered is whether these measures would rank different distributions of income in the same way in terms of distributional inequality. These measures would give the same ranking for, say, two distributions of income when there is Lorenz domination, that is, where the two corresponding Lorenz curves are such that one lies completely within the other. In cases, where the two curves intersect, the choice of a suitable ratio has to be based on further considerations. For example, a small transfer of income from a rich person to a person poorer than him should be expected to result in a reduction in the inequality measure. This property is known as the Pigou-Dalton effect. How sensitive are the measures mentioned above to this effect? Indicators Covering Inequality in Non-income Aspects of Life Quality of life, as we noted earlier, has dimensions other than income, like access to basic needs such as shelter, safe drinking water and services like sanitation, electricity, education and health and employment opportunities. An assessment of the varying degrees of access of different individuals or households to one or more of these services and facilities is one way of analysing such inequalities. Another is to attempt a similar assessment of how the level of access to these services and facilities differ between and among population groups like males and females, the rural folk and urban residents, different areas like the various States, Union Territories and other administrative divisions, the remote, the hilly and the backward areas, the socially challenged groups like the Scheduled Castes/Tribes and the physically challenged. The inclusion of the first population group and the last two population groups in this list has to be specifically emphasised since the most important social dimensions that need to be built into analysis of inequality in levels of living are gender and social and physical disability. Discrimination based on these has to do with social and cultural attitudes and biases and these have to be dealt with through empowerment of the groups concerned and through efforts to change the social mindset. A third is to attempt, if possible, the first and the second type of analysis by income/consumption expenditure classes. This would, in a way, integrate the analysis of income inequality attempted in an earlier section with the larger area of levels of quality of life. Poverty and Inequality: Policy Implications 63

15 Framework of Indian Economy The GDI and its component indices considered in an earlier section are all indicators depicting gender inequality. The HDIs, GDIs and their component indices are available for different States and Union Territories and in some States at the district level also and can help analyse spatial inequality, that is, inequality among areas. This will help building location-specific needs into policy and policy implementation. The component indices of HDI and GDI are more specific and their availability for different vulnerable groups like the Scheduled Castes/Tribes, rural areas and women would help identifying problems needing attention. One can also compute these indicators for different income classes, areas and social classes to get an even better focus for policy action. It is possible for example to compile the distribution of school attendance, literacy, skills etc., by monthly per capita consumer expenditure classes (MPCE classes) from NSSO consumer expenditure surveys. Such a distribution by MPCE classes can also be extracted for the Scheduled Castes/Tribes and for women from these surveys. Lorenz curve/gr analysis can also be attempted for non-income aspects of quality of life. For example, quality attributes like proportion of educated persons (matriculates and above) to population in each income class can be analysed in the same way as income by income classes as in the preceding section (See the example worked out in the next section). Employment shares by income classes can be examined in a similar fashion. 4.7 LEVELS OF INEQUALITY Levels of Inequality in Income and Consumption Let us look at levels of inequality in income or consumption. Consumer expenditure of households is a good proxy for income, at least in the lower classes. A study of inequalities in levels of consumption will by itself be useful in an economy where agriculture, the unorganised sector, payment of wages in kind and the non-monetised sector still play an important role. Such an analysis will be able to pinpoint attention on specific areas of concern in the consumption pyramid. Let us, therefore, turn to levels of inequality in consumption. Sl. No. Table 4.7: Inequality in Consumption: 1983 to Inequality Measure 1. Share (%) in total consumption of the Rural Urban Rural Urban Rural Urban (a) Bottom 10% of pop (b) Top 10% of pop (c) Ratio of (b) to (a) (d) Bottom 20% of pop (e) Top 20% of pop (f) Ratio of (e) to (d) (g) Bottom half of pop (h) Top half of population (i) Ratio of (h) to (g) Gini Ratio Sources: National Human Development Report, 2001, Planning Commission. 64

16 The household consumer expenditure surveys of the NSSO provide us with trends in levels of consumption of expenditure in the population by monthly per capita consumer expenditure (MPCE) classes. Estimates of the different measures of inequality made from the data collected in these surveys are shown in Table 4.7. A comparison of the share of the bottom 10 per cent (or 20 per cent or 50 per cent) of the population in total consumption with that of the top 10 per cent (or 20 per cent or 50 per cent) of the population brings out dramatically the extent of inequality in consumption. The inequality situation is worse in urban areas than in rural areas. This is so in all States and Union Territories. Inequality in consumption is declining, albeit slowly, in rural areas according to all measures of inequality. On the other hand, urban inequality shows no sign of any decline. Regional Variations Gini Ratios calculated for the rural and urban areas of different States and Union Territories are shown grouped by ranges of values of their Gini Ratios in in Table 4.8. Inequality in consumption levels is worse than in the rest of the country (GR greater than 0.3) in urban areas of all the four Southern States and adjoining Maharashtra and in urban areas of the region extending from Chandigarh to West Bengal and urban Assam. These are also areas where incidence of (urban) unemployment is high. Arunachal Pradesh, Dadra & Nagar Haveli, Pondicherry, the urban areas of the region extending from Gujarat to Himachal Pradesh, Goa, Orissa and Sikkim and rural areas of Tamil Nadu, Kerala and Maharashtra are only slightly better off in this regard, with a GR of 0.25 to Table 4.8: Gini Ratios of States and Union Territories Poverty and Inequality: Policy Implications Range of Value of Gini Ratio Rural Urban (1) (2) (3) More than 0.3 Nil Assam,West Bengal, Bihar, UttarPradesh, Madhya Pradesh, Delhi, Chandigarh, Maharashtra, Andhra Pradesh, Tamil Nadu, Karnataka, Kerala 0.25 to 0.30 Arunachal Pradesh, Chandigarh, Maharashtra, Dadra & Nagar Haveli, Tamil Nadu, Pondicherry, Kerala 0.20 to 0.25 Assam, Sikkim, West Bengal, Orissa, Bihar, Uttar Pradesh, Madhya Pradesh, Punjab, Himachal Pradesh, Haryana, Rajasthan, Daman & Diu, Gujarat, Goa, Karnataka, Andhra Pradesh, A & N Islands to 0.2 Tripura, Manipur, Meghalaya, Nagaland, Mizoram, Delhi, Jammu & Kashmir, Lakshadweep Arunachal Pradesh, Tripura, Sikkim, Orissa, Himachal Pradesh, Punjab, Haryana, Rajasthan, Dadra & Nagar Haveli, Gujarat, Goa, Pondicherry Meghalaya, Manipur, Nagaland, Mizoram, Jammu & Kashmir, Daman & Diu, A & N Islands, Lakshadweep Source: Gini Ratios are from the National Human Development Report, 2001 of the Planning Commission. Let us now turn to inequality in aspects of life other than income. Nil 65

17 Framework of Indian Economy Levels of Inequality in Non-Income Aspects of Life We look at levels of inequality in the matter of access to certain basic needs like employment, education, health and shelter, adopting one or more of the three approaches to assessing the inequality situation, listed in Sub-section a) Employment As brought out by the 55 th Round of the NSSO, the quality of employment of those in the bottom 25 per cent of the population group in terms of whatever parameter we may choose to specify leaves much to be desired. Their share in regular wage/salaried employment is only 11 per cent and in casual labour, it is about one-third in rural areas and, roughly, one-half in urban areas. Most of the employment opportunities that come their way are low paid, insecure, intermittent opportunities casual labour or self-employment. The levels of their earnings are obviously too low as the combined earnings of one or more such employed individuals in a household are hardly sufficient to raise the per capita monthly consumption level of the household to Rs. 300/- in rural areas and Rs. 425/- in urban areas at price levels. The NSSO notes that there is sharp increase in the proportion of regular wage/salaried workers and the sharp decrease in the proportion of casual labour, especially in urban areas, as we move from the bottom one-fourth of the population to the rest of the population. b) Other Aspects The rural-urban divide and the social divide in the matter of quality of housing, education, health, shelter and other related aspects and facilities are indeed striking. Regional variations abound but the picture of divide is common everywhere. 4.8 POLICY IMPLICATIONS What are the policy implications of the prevailing levels of inequality and poverty? Let us examine. a) Structure of GDP Growth The overall growth of GDP of an economy can be expressed in terms of the growth of GDP of the income classes. Let there be three income classes, the bottom 30 per cent of the population, the middle 40 per cent of the population and the top 30 per cent of the population. Let the rates of growth of overall GDP and GDP of the three income classes be r, r 1, r 2, and r 3 respectively and the shares of the three classes in total GDP in the initial year be y 1, y 2 and y 3 respectively. It can be shown that: 66 r = y 1 r 1 + y 2 r 2 + y 3 r 3 This is true of consumption also. If the income of each of the classes grows at the same rate during a period of time, overall GDP will also grow at the same rate. Further, the income shares of the income classes in total GDP at the end of that period will also remain the same as in the initial period. The income share of any income class will be higher than in the initial period only if the rate of growth of income in that class is higher than the overall rate of growth. However, the shares of all classes cannot increase simultaneously; some will increase while others will decrease. Thus, the share of the bottom 30 per cent class may increase while those of the other two classes may decrease. In such a case, income inequality will also come down. A development strategy that

18 aims at reduction in income inequality would include steps to ensure that the rate of growth of income in the bottom 30 per cent of the population is significantly higher than the rate of growth of overall GDP so that its share of GDP goes up. Alternative exercises with the Plan model carried out while formulating the Fifth Five Year Plan, for instance, postulated that by way of reduction in the inequality in expenditure distribution the rate of growth of consumption in the bottom few deciles of the population would be much higher than the growth rate of aggregate consumption during the Plan period. The subsequent Five Year Plans included specific targets of reduction in the poverty ratio in the relevant target years generally on the basis of a similar analysis. The Plans naturally, therefore, included policies and programmes designed to achieve these targets as also for realising objectives in non-income aspects of poverty and inequality like, for example, improvement of urban slums shelter and basic services for the urban poor. Notwithstanding these and the overall rates of growth of GDP as also the considerable sectoral diversification of GDP achieved over the last five decades, there have been shortfalls in the realisation of such targets. Indicators of non-income dimensions of poverty and levels of inequality in non income aspects of life show that quality of life is far from satisfactory not only for the poor but also for sizable sections of society above the poverty line. We may not be in a position to give a complete policy prescription in this Unit. However, we shall highlight some problems and policy issues in areas relevant to this Unit. The first is, of course, the restructuring of GDP growth in favour of the bottom three or four deciles so that the shares of these deciles in total GDP rises substantially as early as possible. Besides reorienting the structure of the production in line with this objective, action is called for in a number of other fronts. Some are dealt with below. b) Employment i) Employment Orientation of GDP Growth The rate of growth of employment has lagged behind the rate of growth of the labour force during the 1980s and the 1990s and as a result, unemployment has increased, from 20 million in 1994 to 27 million in The rising rate of growth of the economy during the period (6.7 per cent per annum on the average) has hardly helped in generating employment, as the employment intensity of GDP growth has declined substantially during the 1990s. Growth of GDP by one per cent during the period resulted in employment growth of one half of one per cent. During the period , however, the growth of GDP by one per cent led to employment growth of only about one-sixth of one per cent. It is, therefore, necessary to ensure that the structure of GDP growth should become more employment oriented than before. Globalisation facilitates easier access than before to the latest technology and international capital. These should be accessed only if the implications of such a step for employment in the short run as well as in the long run are clearly seen to be favourable. Quality of employment [see the sub-section (a)] should be raised through: (a) stricter enforcement of labour laws relating to minimum wages and social protection to workers and (b) provision of loans and technical and marketing support to the selfemployed. Poverty and Inequality: Policy Implications 67

19 Framework of Indian Economy 68 ii) Poverty Alleviation Programmes and Employment Schemes Poverty alleviation programmes and a variety of employment and selfemployment promotion programmes for the non-poor have been implemented for about 25 years now. The impact of these programmes on poverty and unemployment and, therefore, on inequalities has not been as much as expected. The factors responsible for such a state of affairs have to be tackled. Personnel implementing the programmes, especially at the grass-roots level, have to be trained properly. The routine approach being adopted by the official machinery to the implementation of the programmes, much to the detriment of their objectives, has to be changed, attitudinal changes in it have to brought about and it should be suitably reoriented to serve the goal of poverty reduction. Other impediments to the programmes which need to be dealt with firmly are: (a) the nexus between the official machinery and the rich and the powerful that runs counter to the interests of the poor, (b) corruption, and (c) the lack of a proactive cooperation from agencies whose inputs are crucial to the success of the programmes. The involvement of Panchayati Raj Institutions (PRIs) and Urban Local Bodies (ULBs) is important for better implementation of the programmes but these bodies are currently not endowed with the necessary powers and administrative support. PRIs and ULBs should, therefore, be strengthened in this regard and involved in the implementation of these programmes. The credit system is somewhat reluctant to provide credit to the poor and the unemployed or the self employed in view of their past experience with such lending programmes and a trustworthy linkage has to be built up between that system and the beneficiaries of the new set of programmes. The processes of identification of the poor that are liable to be adversely affected by factors like: (a) the way the society is organised, (b) the lack of voice for the poor and (c) corruption, (d) lax implementation of programmes on the ground, (e) use of contractors despite instructions to the contrary, and (f) failure to utilise funds and other resources allocated for the programmes, lead to leakages in the programmes. These can be prevented only through: (i) a tightening up of the monitoring of programme implementation, (ii) organisation of the poor through social mobilisation and awareness creation in collaboration with NGOs who have successfully done such work, (iii) rooting out corruption and (iv) bringing about transparency in programme implementation. Finally, it is necessary to take a total view of poverty and reorient the planning process suitably to strike at the root causes of poverty. Most of these steps apply equally well to the problem of generation of productive employment opportunities on a scale sufficient to liquidate unemployment and underemployment. iii) The National Rural Employment Guarantee Act, 2005 A debate on the question of making the Right to Work a Fundamental Right enshrined in the Constitution and the need for a national employment guarantee scheme had been going on in the country for a long time. At last, things have moved towards this goal. The National Advisory Council (NAC) prepared a draft Employment Guarantee Act (EGA) and based on the draft, Government came out with the National Rural Employment Guarantee Bill. The same has been passed by the Parliament and has become the Act i.e., The National Rural Employment Guarantee Act, This Act provides the enhancement of livelihood security of the households in rural areas of the country by providing at least one hundred days of guaranteed wage employment in every financial year to every household whose adult members volunteer to do unskilled manual work.