SPECIAL ARTICLE Do Inequality and Prices Affect Comparisons in Living Standards? The Indian Evidence
Ankita Mishra, Ranjan Ray
Notwithstanding the wide recognition that movements in relative prices and inequality affect welfare comparisons between households, there is a dearth of literature incorporating them in welfare analysis. Changes in relative prices open up a divergence between nominal and real expenditure inequalities that will persist even if all nominal expenditures are deflated by a household invariant price index. In a period of rising inequality, a comparison of summary welfare measures based on mean or median expenditure, even if inflation adjusted, will give us a misleading picture of changes in living standards. This paper provides evidence for India in the post-reform period. The results confirm that in a period of significant relative price changes and rising inequality, the omission of these factors leads to overstatement of the welfare gains. Another observation is that relatively affluent states do not do all that well if we focus on inequality or on their food consumption alone.
The research for this paper was funded by an Australian Research Council Discovery Grant (DP 0773489). The authors thank the NSSO for access to the unit record data sets used in this study, and acknowledge comments on an earlier version at the conference on “Growth and Development” held at ISI Delhi in December 2009. The usual disclaimer applies.
Ankita Mishra (ankita.mishra@buseco.monash.edu.au) and Ranjan Ray (Ranjan.Ray@buseco.monash.edu.au) are at the Economics Department, Monash University, Melbourne.
1 Introduction
P
There are other ways in which prices can affect welfare comparisons. In the context of India, the issue of spatial consumer price indices becomes relevant in welfare comparisons between households in different regions. Following the pioneering work of Bhattacharya and Chatterjee (1971) on the National Sample Survey (NSS) data, there is now a rapidly expanding literature on regional price differentials (see, for example, Aten and Menezes 2002; Coondoo et al 2004, 2009). There has, however, not been much attempt to integrate this work on regional price differentials with welfare analysis. This study uses the methodology proposed by Sen (1976) to compare and rank Indian states on the basis of real consumption levels taking note of regional prices.
In the rapidly expanding literature on the analysis of household welfare in India, not much attention has been paid to the central role played by prices in the gains and losses experienced by households. The present study departs from this tradition by providing quantitative evidence on how movements in prices, especially relative prices, affect the welfare rankings and our evaluation of gains and losses. We do so by examining the sensitivity of the state rankings on alternative welfare criteria with prices playing a significant role in each case. Moreover, we test for the statistical significance of the price effect on real consumption and also of its change over time. The calculations provide useful insights on regional differences in welfare during a period that has attracted much interest.
Another point of departure of this study is the explicit incorporation of inequality in the welfare comparisons. We do so through the adoption of the inequality adjusted measure of real income
Economic & Political Weekly
EPW
| proposed by Sen (1976). In a period of rising inequality, any in | and miscellaneous. While the consumer price index for agricul |
| crease in real income or expenditure will need to be moderated | tural labourers (CPI-AL) for these major commodity groupings |
| by adjusting for the inequality increase; otherwise we will over | was used as rural prices, the consumer price index for industrial |
| state the welfare improvement. Similarly, in cross country wel | workers (CPI-IW) was used as urban prices. |
| fare comparisons, one will overstate the welfare superiority of a | The magnitude and sign of the difference between the ine |
| high income country over a low income country if the former is | qualities in real and nominal expenditure per adult equivalent, |
| also the one with high inequality. The interaction of the move | i e, between the inequalities in (xˇht /moh ) and (xht/moh ), will |
| ment in relative prices with inequality changes has strong effects | depend on the movement in relative prices. In case of no change |
| on welfare comparisons as we illustrate with the Indian evidence. | in relative prices between current year t and base year 0, the two |
| The period considered is of considerable policy importance | inequalities will coincide in both years, not just in the base year. |
| since it covers what is referred to as fi rst and second generation | The real and nominal inequality indices, which are defi ned over |
| reforms. The state disaggregated results allow us to explore the | real and nominal expenditure per adult equivalent, are given by |
| robustness of the states’ welfare rankings between the alterna- | It R and It N, respectively (It R – Tt N) > o implies that the relative price |
| tive welfare criteria, namely, poverty, inequality and real con | movement has been regressive or inequality increasing, while the |
| sumption. We discuss these aspects in more detail below. Before | reverse is indicated if (It R – Tt N) < o. Previous studies using this |
| we do that, let us describe some of the key concepts and the data | methodology found the relative price changes to have been re |
| sets used in this study. | gressive in the UK during the 1970s (Muellbauer 1974; Ray 1985) |
| and in Australia during the late 1990s (Nicholas et al 2008). | |
| 2 Key Concepts and Methodology | The study then moves on to a comparison of real expenditure |
| Let us distinguish between nominal and real household expendi | between states using the household specific price indices de |
| tures. The former is the current value of a household’s expenditure | scribed above. The expenditure comparisons between states are |
| on various items, while the latter is the value of that expenditure | based on their median real expenditures. While such comparisons |
| after adjusting for inflation. Instead of calculating real expenditure | are of interest, they are somewhat limited by the omission of ine |
| as simply the nominal expenditure divided by a household invari | quality in the welfare comparisons. We therefore use the welfare |
| ant price index, we define it as follows. Let us denote the nominal | measure due to Sen (1976) that builds in distributional considera |
| expenditure by xht. It is obtained from the expenditure function, | tions. The Sen measure of welfare is given by |
| ch(ut, pt, zh), as the minimum expenditure required by a household | So = xg (1 – Gs) |
| h in time t with a demographic profile given by zh to obtain a utility | where xg is the mean expenditure (per adult equivalent) in state s |
| level ut, which is a measure of its level of living in time t. Following | in the Indian union and Gs is that state’s Gini inequality index. |
| Muellbauer (1974: 42), we define real expenditure of household | The issue of price sensitivity is investigated by (a) comparing the |
| h in year t, namely, xˇht as the minimum expenditure needed to | state rankings in nominal (x g N) and real (x g R) terms, and (b) using |
| obtain current year utility, ut at base year price po. In other words: | the alternative methodology suggested by Sen (1976) by valuing |
| xˇht – c (po, ut, zh) | the consumption quantities using the prices of all the states including its own and comparing their rankings between alterna- |
| In per adult equivalent terms, nominal and real expenditures | tive sets of price confi gurations. |
| are given by, respectively, (xht /mht) and (x ~ ht/mht ), where mht is | |
| the adult equivalent scale of household h in period t, and is de | 3 Data Sets and Poverty Rankings |
| pendent on the household’s size and composition. In the base year, | This study uses the detailed information contained in the unit |
| 0, nominal and real expenditures will coincide, but will diverge | records from the 50th (1993-94), 55th (1999-2000) and 61st |
| in the comparison years depending on household preferences | (2004-05) rounds of the NSS. These three surveys cover a reason |
| and movement in relative prices. The implicit price defl ator | ably long time interval (1993-2004) to make the comparisons |
| obtained by dividing nominal by real expenditures will be house | meaningful. The price information was obtained from published |
| hold specific, takes note of preferences, and will vary with house | price series put out by the Government of India and the Reserve |
| hold size and composition. | Bank of India (RBI). The state-specific poverty lines are made |
| The equivalence scales and the expression for real expenditure | available by the Planning Commission.1 |
| require specification of the expenditure function, ch(u,p,z), and | Table 1 reports the price series of the four groups of items used |
| knowledge of the functional parameters. We do so through the | in the demand estimation. The all-India price indices were obtained |
| assumption that the cost function takes the Quadratic Almost | as population share weighted average of the state price indices. |
| Ideal (QAI) form proposed in Banks et al (1997) and the equiva- | Fuel and light and the composite item, miscellaneous, recorded |
| lence scale takes the Cobb Douglas form proposed in Ray (1983) and used in the distributional study on United Kingdom (UK) | Table 1: Price Indices for Rural and Urban Samples with 50th Round as Base Period Commodity Group Rural Urban |
| data in Ray (1985). The resultant demographic demand system is | 50th 55th 61st 50th 55th 61st |
| derived and estimated on all-India data separately for rural and urban India in Mishra and Ray (2009). The demand systems were estimated on the following four-item breakdown of household | Food group 1.000 1.414 1.508 1.000 1.655 1.869 Fuel and light group 1.000 1.485 1.912 1.000 1.689 2.609 Clothing, bedding and footwear 1.000 1.366 1.628 1.000 1.536 1.732 Miscellaneous group 1.000 1.551 1.832 1.000 1.684 2.111 |
| expenditure: food, fuel and light, clothing, bedding and footwear, | Source: Handbook of Statistics on the Indian Economy, RBI; and author’s calculation. |
| 40 | January 15, 2011 vol xlvi no 3 Economic & Political Weekly |
EPWSPECIAL ARTICLE
the larger price increases over this period. There was a signifi -from being based on a minimal consumption basket to that of relacant realignment of relative prices in both rural and urban areas. tive deprivation.2 Since this paper is not primarily about poverty
Table 2 provides estimates of poverty rates calculated at the state rates, we abstract from the debate on the “correct poverty line” or and all-India levels using the NSS data sets from the three large method to use in poverty calculations that has dominated the povsample rounds. For the 50th round, we used the offi cial poverty erty debate, especially in the Indian context (see, for example, Ray lines at the state and all-India levels. For the subsequent rounds, and Lancaster 2005; Reddy and Pogge 2007; Subramanian 2005, we updated the state-specific poverty lines in line with the in-2009). Table 2 presents the picture on relative affl uence of the difcrease in state-specific median per capita nominal expenditures. ferent states. Bihar and Orissa, at one extreme, and Punjab, at the In doing so, we incorporate both the notion of absolute poverty other, constitute the poverty spectrum. The picture has not underlining the official poverty line in the 50th round, and the notion changed much over this period. These poverty rates are somewhat of relative poverty implicit in the upward revision of the poverty lower than those quoted in the policy debates, for example, the line in the same proportion as the increase in state-specifi c median poverty rate of 27.5% in 2004, let alone the much higher poverty nominal expenditures. The switch from absolute to relative poverty rates that will eventuate if the government accepts the recommenreflects the fact that when a country develops and moves towards dations of the Tendulkar Committee on poverty measurement. developed country status, it merits a shift in the notion of poverty While much of the analysis was based on total expenditure Table 2: State-wise Relative Poverty Rates (per capita, in %) over the four items in the demand estimation, the comparison of
Relative Poverty Ratea real consumption levels between states and their rankings, using
States Rural Urban 50th 55th 61st 50th 55th 61st
Table 4: State-wise Real Expenditure Inequalitya (Per Adult Equivalent) in Rural Areas
Andhra Pradesh 11.1 8.8 9.4 30.4 32.0 30.5
States Real Expenditure Inequality Rankingb of States Based on Assamb 35.5 32.9 17.2 5.0 6.8 1.8 Real Expenditure Inequality 50th Round 55th Round 61st Round 50th Round 55th Round 61st Round
Biharb 48.2 47.6 34.0 23.8 29.8 18.8
Andhra Pradesh 0.220 0.226 0.294 9 7 13
Gujarat 16.6 15.6 13.3 21.2 19.5 18.2
Assam 0.181 0.204 0.229 1 1 3
Jammu and Kashmirc 6.2 2.4 4.2 7.1 6.0 4.3
Bihar 0.206 0.210 0.226 5 4 2
Karnataka 22.4 19.4 10.2 29.4 25.9 22.7
Gujarat 0.203 0.223 0.244 4 6 4 Kerala 20.4 18.1 19.3 19.2 22.2 25.1
Jammu and Kashmir 0.186 0.208 0.219 2 3 1
Madhya Pradeshb 30.2 28.7 18.0 37.2 37.2 35.6
Karnataka 0.214 0.234 0.251 6 10 6 Maharashtra 27.4 23.9 17.7 25.4 22.4 24.7
Kerala 0.250 0.257 0.435 14 14 15 Orissa 39.5 41.4 19.2 32.2 39.3 25.7
Madhya Pradesh 0.241 0.232 0.267 13 9 9 Punjabb 14.9 12.3 9.4 7.4 5.3 2.5
Maharashtra 0.240 0.249 0.271 12 13 11 Rajasthan 21.0 15.3 13.3 20.6 15.9 14.3
Orissa 0.220 0.231 0.259 8 8 8 Tamil Nadu 25.5 25.6 8.3 33.1 33.6 22.3
Punjab 0.250 0.240 0.281 15 11 12 Uttar Pradeshb 32.9 30.2 18.6 24.9 23.7 24.8
Rajasthan 0.219 0.221 0.247 7 5 5 West Bengal 32.9 31.9 22.3 13.4 11.6 10.7
Tamil Nadu 0.240 0.257 0.321 11 15 14 All-India 28.9 28.3 16.9 23.6 24.2 21.0
Uttar Pradesh 0.233 0.244 0.256 10 12 7
a The poverty lines for 55th and 61st rounds were obtained by increasing the 50th round poverty line in the same proportion as the state-wise median expenditures. West Bengal 0.200 0.206 0.270 3 2 10 b Assam includes Manipur, Meghalaya and Tripura; Punjab includes Haryana, Himachal Pradesh
All-India 0.237 0.250 0.283
and Delhi; Uttar Pradesh, Madhya Pradesh and Bihar include Uttarakhand, Chhattisgarh and a Measured by Gini coefficient.
Jharkhand respectively since their inception (here only for 61st round).
b The state with the lowest Gini coefficient (least unequal) is ranked 1 and highest Gini (most c For Jammu and Kashmir, the state poverty line was not available for 50th and 55th rounds, so unequal) is ranked 15.
the all India poverty line is used to calculate poverty rates for these two rounds.
Table 3: State-wise Nominal Expenditure Inequalitya (Per Adult Equivalent) in Rural Areas Table 5: State-wise Nominal Expenditure Inequalitya (Per Adult Equivalent) in Urban Areas
States Nominal Expenditure Inequality Rankingb of States Based on States Nominal Expenditure Inequality Rankingb of States Based on Nominal Expenditure Inequality Nominal Expenditure Inequality 50th Round 55th Round 61st Round 50th Round 55th Round 61st Round 50th Round 55th Round 61st Round 50th Round 55th Round 61st Round
Andhra Pradesh 0.220 0.228 0.304 9 7 13 Andhra Pradesh 0.256 0.275 0.414 7 5 11
Assam 0.181 0.205 0.237 1 1 3 Assam 0.251 0.269 0.303 6 3 3
Bihar 0.206 0.211 0.231 5 4 2 Bihar 0.257 0.297 0.318 8 11 4
Gujarat 0.203 0.225 0.251 4 6 4 Gujarat 0.228 0.275 0.344 2 6 5
Jammu and Kashmir 0.186 0.211 0.226 2 3 1 Jammu and Kashmir 0.239 0.220 0.265 3 1 1
Karnataka 0.214 0.235 0.258 6 10 6 Karnataka 0.248 0.272 0.482 5 4 14
Kerala 0.250 0.260 0.467 14 14 15 Kerala 0.288 0.290 0.412 14 9 10
Madhya Pradesh 0.241 0.234 0.274 13 9 9 Madhya Pradesh 0.243 0.287 0.380 4 8 7
Maharashtra 0.240 0.251 0.280 12 13 10 Maharashtra 0.272 0.306 0.425 10 13 12
Orissa 0.220 0.232 0.264 8 8 7 Orissa 0.270 0.276 0.380 9 7 8
Punjab 0.250 0.243 0.296 15 11 12 Punjab 0.291 0.299 0.529 15 12 15
Rajasthan 0.219 0.223 0.255 7 5 5 Rajasthan 0.226 0.267 0.276 1 2 2
Tamil Nadu 0.240 0.260 0.340 11 15 14 Tamil Nadu 0.288 0.330 0.438 13 15 13
Uttar Pradesh 0.233 0.246 0.265 10 12 8 Uttar Pradesh 0.272 0.291 0.401 11 10 9
West Bengal 0.200 0.208 0.280 3 2 11 West Bengal 0.275 0.321 0.358 12 14 6
All-India 0.237 0.252 0.294 All-India 0.276 0.303 0.413 a Measured by Gini coefficient. a Measured by Gini coefficient. b The state with the lowest Gini coefficient (least unequal) is ranked 1 and highest Gini (most b The state with the lowest Gini coefficient (least unequal) is ranked 1 and highest Gini (most unequal) is ranked 15. unequal) is ranked 15.
Economic & Political Weekly
EPW
the procedure due to Sen (1976), considered food items only since the second half, i e, between 1999-2000 and 2004-05. Second, one requires quantity and unit value information at the state the urban inequalities exceed their rural counterpart with the level that is not available for non-food items in the NSS data sets. gap increasing sharply in the second half. Third, there has been
some shift in the inequality rankings of the states over this period. 4 Prices and Expenditure Inequality: State-wise Evidence For example, rural West Bengal has slipped but urban West Bengal The state-wise rural inequalities in per adult equivalent household has improved, while Bihar showed improvement in both sectors. expenditures, in nominal and real terms, are presented in Tables Some of the more affluent states such as Punjab and Maharashtra 3 and 4 (p 41), respectively. The corresponding urban expendi-are much lower ranked on inequality than some of the poorer ture inequalities are reported in Tables 5 (p 41) and 6. These ta-states such as Bihar. Fourth, the movement in relative prices has bles also report the all-India estimates and inequality ranking of been mildly progressive with the nominal expenditure inequalities the states, with those with higher inequality ranked lower. Let us exceeding the real inequality estimates. The state rankings are recall from our earlier discussion that nominal and real inequali-robust between the nominal and real expenditure inequalities. ties will coincide in the base round (50th round), but will diverge in the comparison rounds (55th and 61st rounds). The following 5 Real Expenditure Comparisons between States principal features emerge from these tables. First, there has been Tables 7 and 8 present the real (i e, constant utility, price adjusted) an increase in inequality, in both nominal and real terms, in all expenditure (per adult equivalent) of the median household in the states and across both rural and urban sectors of the Indian rural and urban areas, respectively.3 The median household was economy during this period with the increase accelerating during obtained for each state from the distribution of real expendi
tures (xhR) in that state, with the real expenditures calculated for
Table 6: State-wise Real Expenditure Inequalitya (Per Adult Equivalent) in Urban Areas
States Real Expenditure Inequality Rankingb of States Based on each household from nominal expenditures in conjunction with Real Expenditure Inequality
the QAI demand parameter estimates as explained earlier. Note
50th Round 55th Round 61st Round 50th Round 55th Round 61st Round
Andhra Pradesh 0.256 0.272 0.408 7 6 11 that these expenditure figures are over 30 days and include four Assam 0.251 0.266 0.300 6 3 3 items used in the demand estimation. The ranking of states Bihar 0.257 0.293 0.315 8 11 4 based on median real expenditure is consistent with that based
Gujarat 0.228 0.272 0.340 2 5 5
on the poverty rates reported in Table 2. For example, Punjab
Jammu and Kashmir 0.239 0.217 0.262 3 1 1
and Jammu and Kashmir are at the upper end, and Bihar and
Karnataka 0.248 0.269 0.473 5 4 14
Orissa at the lower end of both the poverty and the real expendi-
Kerala 0.288 0.287 0.407 14 9 10
ture spectrum. The picture is generally quite similar between
Madhya Pradesh 0.243 0.284 0.375 4 8 7
0.272 0.419 10 12 the rural and urban areas. There has been a very marginal
Maharashtra 0.302 13 Orissa 0.270 0.273 0.376 9 7 8 change in the rankings over this period. In rural areas, for
Punjab 0.291 0.295 0.514 15 12 15 example, Andhra Pradesh has improved, but Karnataka has Rajasthan 0.226 0.263 0.274 1 2 2 slipped, with the other states retaining their spots. While both Tamil Nadu 0.288 0.322 0.431 13 15 13
the state and the all-India figures show improvements in the
Uttar Pradesh 0.272 0.287 0.396 11 10 9
median real expenditure levels, the progress has been quite
West Bengal 0.275 0.313 0.354 12 14 6
uneven between regions. For example, between the neighbour
All-India 0.276 0.299 0.406
ing states of Kerala and Karnataka in the rural areas, Kerala
a Measured by Gini coefficient. b The state with the lowest Gini coefficient (least unequal) is ranked 1 and highest Gini (most
registers a much more impressive gain reaching the top rank in
unequal) is ranked 15.
Table 8: State-wise Median Real Expenditure (Per Adult Equivalent) in Urban Areas (Rs/month) Table 7: State-wise Median Real Expenditure (Per Adult Equivalent) in Rural Areas (Rs/month)
States Median Real Expenditure Rankinga of States Based on States Median Real Expenditure Rankinga of States Based on Real Expenditure Real Expenditure 50th Round 55th Round 61st Round 50th Round 55th Round 61st Round 50th Round 55th Round 61st Round 50th Round 55th Round 61st Round
Andhra Pradesh 390.829 461.107 437.850 14 11 12
Andhra Pradesh 291.090 316.661 413.677 13 12 9
Assam 461.339 543.228 498.239 6 5 8Assam 353.490 376.543 450.269 6 6 6 Bihar 408.159 426.278 414.555 12 14 13
Bihar 276.613 306.042 367.490 14 13 12
Gujarat 371.873 405.242 457.806 5 5 5 Gujarat 483.018 538.988 563.873 4 6 5
Jammu and Kashmir 422.067 516.212 580.762 2 2 3 Jammu and Kashmir 536.557 637.678 626.662 2 1 1 Karnataka 308.488 351.785 363.612 9 9 13 Karnataka 426.194 502.495 458.390 10 8 10
Kerala 393.629 488.490 640.864 3 3 1 Kerala 416.313 551.057 583.678 11 4 3
Madhya Pradesh 298.804 297.348 345.464 11 14 14 Madhya Pradesh 428.496 441.737 406.814 9 13 14
Maharashtra 305.043 347.144 397.518 10 10 10 Maharashtra 503.953 560.075 567.088 3 3 4
Orissa 269.263 275.638 300.911 15 15 15 Orissa 392.492 423.664 360.665 13 15 15
Punjab 434.831 519.415 584.036 1 1 2 Punjab 571.926 624.457 615.349 1 2 2
Rajasthan 380.338 440.060 463.974 4 4 4 Rajasthan 477.994 525.917 506.172 5 7 6
Tamil Nadu 295.846 338.648 397.367 12 11 11 Tamil Nadu 374.614 465.146 461.686 15 10 9
Uttar Pradesh 328.545 355.437 427.266 8 8 7 Uttar Pradesh 428.755 448.087 453.143 8 12 11
West Bengal 332.618 364.047 425.977 7 7 8 West Bengal 440.090 485.681 501.561 7 9 7
All-India 331.392 365.427 435.694 All-India 455.829 517.474 510.561 a The state with the highest value of median nominal expenditure is ranked 1 and the lowest is ranked 15. a The state with the highest value of median real expenditure is ranked 1 and the lowest is ranked 15.
42 January 15, 2011 vol xlvi no 3
EPW
SPECIAL ARTICLE
the 61st round, in relation to Karnataka which slipped from 9 to 13 over this period.
The state rankings based on the real consumption fi gures at median are incomplete as they do not take account of intra-state inequalities. To do so, we calculate the welfare measure based on inequality adjusted real consumption due to Sen (1976) described earlier. To focus attention, we have done the calculations only for the 61st round. The 15×15 matrices presented for rural and urban areas in Tables 9 and 10, respectively, considered the four food items – cereals, pulses, oil and spices. The column shows a state’s real (per equivalent) mean expenditure, evaluated at the various state prices as we move down that column. The (i,j)th element in the matrix shows the inequality adjusted real food consumption of state j if everyone in that state was facing state i’s prices. The diagonal elements report the real expenditure values at the state’s own prices. The large variation in the values as we move down the column reflects the large regional variation in prices across states in India that led to early interest in regional price differentials in India (see Bhattacharya and Chatterjee 1971; more recently, Coondoo et al 2004; Coondoo et al 2009).
Table 9: Real Consumption Comparisons between States –Rural, 61st Round (Rs/month)
These matrices point to a ranking of states based on their inequality adjusted food consumption level which is different from that based on only real consumption level over all items and not adjusted for inequality. This is evident from Table 11 which
| States | Rural | Urban | ||||
|---|---|---|---|---|---|---|
| Beats | Beaten By | Beats | Beaten By | |||
| Andhra Pradesh | 11 | 3 | 10 | 4 | ||
| Assam | 3 | 11 | 7 | 7 | ||
| Bihar | 12 | 2 | 14 | 0 | ||
| Gujarat | 7 | 7 | 4 | 10 | ||
| Jammu and Kashmir | 14 | 0 | 13 | 1 | ||
| Karnataka | 0 | 14 | 5 | 9 | ||
| Kerala | 0 | 14 | 0 | 14 | ||
| Madhya Pradesh | 8 | 6 | 8 | 6 | ||
| Maharashtra | 5 | 9 | 3 | 11 | ||
| Orissa | 2 | 12 | 6 | 8 | ||
| Punjab | 2 | 12 | 1 | 13 | ||
| Rajasthan | 9 | 5 | 11 | 3 | ||
| Tamil Nadu | 6 | 8 | 6 | 8 | ||
| Uttar Pradesh | 11 | 3 | 9 | 5 | ||
| West Bengal | 5 | 9 | 1 | 13 |
Inequality Adjusted Real Consumption ĺ
| Andhra Pradesh | 216.208 | 201.299 | 223.790 | 202.344 | 229.152 | 187.237 180.940 | 209.566 | 201.206 197.434 197.447 213.670 | 201.223 | 227.761 | 202.119 | ||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Assam | 278.342 | 239.723 | 273.558 | 249.080 | 279.853 | 236.214 241.851 | 255.436 | 247.225 238.975 239.026 265.908 | 265.828 | 276.667 | 247.182 | ||
| Bihar | 225.814 | 201.634 | 226.567 | 209.716 | 234.082 | 193.218 192.121 | 211.844 | 206.561 197.759 199.823 218.738 | 212.113 | 229.777 | 205.759 | ||
| Gujarat | 241.828 | 197.908 | 230.411 | 214.040 | 236.685 | 203.636 213.503 | 215.448 | 211.599 198.541 202.253 225.511 | 234.187 | 233.477 | 207.674 | ||
| Jammu and Kashmir | 258.916 | 232.312 | 260.861 | 237.330 | 267.661 | 221.537 220.892 | 243.630 | 235.149 229.133 228.647 251.838 | 243.676 | 264.082 | 236.455 | ||
| Karnataka | 215.043 | 186.035 | 212.138 | 196.766 | 218.474 | 183.404 185.074 | 198.736 | 194.382 184.049 187.484 205.177 | 204.412 | 215.743 | 191.508 | ||
| Kerala | 269.483 | 252.972 | 279.422 | 253.308 | 287.242 | 232.893 225.180 | 261.039 | 250.826 246.999 245.692 267.604 | 249.592 | 283.276 | 253.869 | ||
| Madhya Pradesh | 210.266 | 175.780 | 202.676 | 188.795 | 208.882 | 177.652 183.925 | 189.487 | 186.158 175.051 | 178.211 | 197.823 | 201.966 | 205.353 | 183.260 |
| Maharashtra | 236.777 | 195.887 | 227.289 | 208.885 | 232.556 | 199.732 208.644 | 212.452 | 207.223 196.799 198.954 222.048 | 228.973 | 230.182 | 204.652 | ||
| Orissa | 218.983 | 191.303 | 216.711 | 204.187 | 225.116 | 186.921 187.059 | 202.977 | 200.217 187.421 192.338 209.585 | 206.603 | 220.351 | 196.696 | ||
| Punjab | 264.331 | 194.729 | 236.760 | 222.149 | 242.462 | 218.218 243.025 | 220.881 | 219.724 200.433 | 206.427 | 236.228 | 264.318 | 239.141 | 212.204 |
| Rajasthan | 213.629 | 173.967 | 202.931 | 192.154 | 209.947 | 180.227 187.739 | 190.203 | 188.963 173.495 | 179.788 | 198.232 | 206.483 | 206.462 | 182.922 |
| Tamil Nadu | 204.215 | 187.224 | 209.574 | 195.502 | 216.628 | 177.282 170.063 | 197.136 | 192.879 182.301 187.752 199.459 | 190.080 | 214.828 | 188.920 | ||
| Uttar Pradesh | 221.555 | 180.282 | 210.390 | 197.404 | 216.861 | 186.653 195.387 | 196.974 | 194.656 180.470 185.547 205.801 | 214.577 | 213.641 | 189.569 | ||
| West Bengal | 252.814 | 225.882 | 254.501 | 231.968 | 260.933 | 216.653 215.601 | 238.079 | 230.109 223.058 223.821 245.197 | 238.363 | 258.333 | 230.002 |
| Andhra Pradesh 233.948 Assam 270.083 Bihar 234.159 Gujarat 261.172 Jammu and Kashmir 263.861 Karnataka 256.757 Kerala 264.784 Madhya Pradesh 222.256 Maharashtra 280.720 Orissa 233.406 Punjab 268.875 Rajasthan 218.551 Tamil Nadu 233.223 Uttar Pradesh 234.062 West Bengal 273.500 Economic & Political Weekly | 232.636 260.250 204.853 254.220 209.089 249.778 287.878 233.802 280.430 237.857 221.447 252.726 206.806 248.026 207.003 237.108 275.397 222.999 267.053 229.133 247.808 283.612 229.529 276.827 232.926 248.802 280.773 221.315 273.301 227.807 260.848 292.480 235.859 287.945 235.740 199.485 233.280 192.657 227.130 195.059 257.002 297.249 237.808 287.536 246.468 217.594 250.119 207.506 245.882 206.066 232.029 275.852 226.488 266.164 233.999 196.043 229.303 192.254 224.542 191.815 231.152 259.293 207.124 254.212 208.594 208.822 244.923 202.473 238.230 205.280 258.737 295.581 236.354 287.422 242.122 January 15, 2011 vol xlvi no 3 | 183.167 217.485 186.158 212.439 211.205 204.078 207.246 180.345 228.362 185.876 222.824 176.609 181.850 190.339 218.560 | 233.905 259.124 228.223 247.352 255.309 251.741 263.943 210.538 266.310 226.495 247.789 207.702 233.874 221.042 265.642 | 205.349 226.465 205.991 236.904 230.316 246.984 228.497 267.406 204.007 217.031 202.295 233.878 219.150 236.001 217.360 256.942 226.626 243.990 225.216 262.646 220.416 243.889 220.785 257.937 234.514 253.200 233.866 268.150 188.821 198.525 186.279 217.879 234.307 255.734 233.168 276.801 204.050 213.628 201.524 232.068 220.783 234.190 217.667 260.248 187.922 194.431 184.748 214.493 207.205 224.652 207.083 236.036 198.327 208.202 195.544 228.952 234.493 254.953 233.744 272.387 | 215.046 239.573 200.490 255.901 263.055 223.061 219.100 231.882 196.892 249.770 250.793 212.846 248.254 259.521 219.926 238.938 257.059 216.478 243.279 269.318 227.686 213.099 213.071 180.946 267.911 270.418 229.084 219.470 229.693 195.299 262.689 250.086 212.932 209.025 209.890 178.871 214.233 239.287 200.441 225.043 223.614 189.844 257.099 270.644 227.930 43 |
|---|
EPW| States | Cereal Consumption of | Rankinga of States Based on | ||||
|---|---|---|---|---|---|---|
| a Median Household (in Kgs/Month) | Cereals Consumption | |||||
| 50th Round | 55th Round | 61st Round | 50th Round | 55th Round | 61st Round |
| Andhra Pradesh | 17.464 | 16.929 | 15.347 | 10 | 9 | 9 |
|---|---|---|---|---|---|---|
| Assamb | 17.754 | 18.187 | 17.994 | 8 | 6 | 4 |
| Biharb | 20.000 | 19.498 | 18.709 | 5 | 3 | 1 |
| Gujarat | 14.700 | 13.733 | 12.984 | 15 | 15 | 15 |
| Jammu and Kashmir | 20.633 | 17.398 | 17.082 | 4 | 8 | 7 |
| Karnataka | 17.718 | 15.310 | 13.749 | 9 | 11 | 12 |
| Kerala | 15.147 | 13.793 | 13.016 | 14 | 14 | 14 |
| Madhya Pradeshb | 19.575 | 17.940 | 16.880 | 7 | 7 | 8 |
| Maharashtra | 15.634 | 15.479 | 13.620 | 13 | 10 | 13 |
| Orissa | 21.333 | 20.375 | 18.016 | 1 | 1 | 3 |
| Punjabb | 16.401 | 15.228 | 14.491 | 11 | 12 | 10 |
| Rajasthan | 20.819 | 20.202 | 17.816 | 3 | 2 | 5 |
| Tamil Nadu | 15.737 | 13.906 | 13.890 | 12 | 13 | 11 |
| Uttar Pradeshb | 19.648 | 19.381 | 18.342 | 6 | 4 | 2 |
| West Bengal | 20.846 | 18.655 | 17.268 | 2 | 5 | 6 |
| All-India | 18.226 | 17.352 | 16.277 |
a The state with the highest value of cereals consumption is ranked 1 and the lowest is ranked 15. b Assam includes Manipur, Meghalaya and Tripura; Punjab includes Haryana, Himachal Pradesh and Delhi; Uttar Pradesh, Madhya Pradesh and Bihar include Uttarakhand, Chhattisgarh and Jharkhand since their inception (here only for 61st round).
| States | Cereal Consumption of | Rankinga of States Based on | ||||
|---|---|---|---|---|---|---|
| a Median Household (in kg/month) | Cereals Consumption | |||||
| 50th Round | 55th Round | 61st Round | 50th Round | 55th Round | 61st Round |
| Andhra Pradesh | 15.201 | 14.486 | 13.886 | 6 | 8 | 8 |
|---|---|---|---|---|---|---|
| Assamb | 16.519 | 16.968 | 17.214 | 3 | 3 | 2 |
| Biharb | 18.766 | 17.560 | 17.510 | 1 | 2 | 1 |
| Gujarat | 12.000 | 11.018 | 10.903 | 15 | 15 | 15 |
| Jammu and Kashmir | 15.141 | 14.986 | 15.467 | 8 | 5 | 5 |
| Karnataka | 14.450 | 13.029 | 12.455 | 10 | 10 | 10 |
| Kerala | 13.084 | 12.484 | 11.328 | 12 | 11 | 13 |
| Madhya Pradeshb | 15.167 | 14.977 | 14.923 | 7 | 6 | 7 |
| Maharashtra | 12.199 | 12.181 | 11.051 | 14 | 12 | 14 |
| Orissa | 17.398 | 17.983 | 16.550 | 2 | 1 | 3 |
| Punjabb | 12.427 | 12.051 | 12.002 | 13 | 14 | 12 |
| Rajasthan | 15.393 | 14.724 | 15.429 | 5 | 7 | 6 |
| Tamil Nadu | 13.482 | 12.097 | 12.104 | 11 | 13 | 11 |
| Uttar Pradeshb | 15.576 | 15.192 | 15.641 | 4 | 4 | 4 |
| West Bengal | 15.004 | 13.599 | 12.902 | 9 | 9 | 9 |
| All-India | 14.431 | 13.824 | 13.501 |
a The state with the highest value of cereals consumption is ranked 1 and the lowest is ranked 15. b Assam includes Manipur, Meghalaya and Tripura; Punjab includes Haryana, Himachal Pradesh and Delhi; Uttar Pradesh, Madhya Pradesh and Bihar include Uttarakhand, Chhattisgarh and Jharkhand since their inception (here only for 61st round).
Table 14: Real Consumption Based State Rankings at All State Prices-Rural, 61st Round
reports the number of states that a particular state dominates (i e, “beats”) and is dominated by (i e, “beaten”) at that state’s prices. Affl uent states such as Punjab and Gujarat have not done well while the poorer states such as Bihar and Uttar Pradesh are much higher ranked than one would expect of them from their poverty rankings (Table 2) or their real consumption rankings (Tables 7 and 8). The picture is similar between rural and urban areas. We should qualify this observation by noting that the matrices in Tables 9 and 10 are based on only expenditures on food items for which quantity and values are available at the household level in the NSS unit records. The importance of the omitted food items varies across states, so the picture portrayed by Tables 9-11 is, at best, a partial one. This point is further underlined in Tables 12 and 13 which report the quantity of monthly cereal consumption (per adult equivalent) at the state and all-India levels in rural and urban areas, respectively. Consistent with the evidence presented in Ray (2007), there has been a decline in cereals consumption over this period. While the decline has been uneven among states and between the rural and urban areas, the rankings have remained largely unchanged.
Tables 14 and 15 (p 45) present evidence on the price sensitivity of the welfare comparisons by reporting the combinations of (a,b), from the (i,j) elements of the matrices in Table 9 (rural) and Table 10 (urban), respectively, where ‘a’ shows the number of states that state j “beats”, and ‘b’ shows the number of states that j is “beaten by”, when evaluated at state i’s prices. These tables therefore present a more complete picture than does Table 11. The price sensitivity of the rankings is easily established by comparing the (a,b) combinations vertically, i e, those reported in the same column but between various rows, since the rows vary with respect to the state prices used in the evaluation. While the picture is generally robust in the sense that states that are ranked in the bottom half rarely jump to the top half, there is considerable movement in the (a,b) combinations within the two halves of the welfare rankings.
6 The All-India Picture
Table 16 (p 45) presents the inequality adjusted expenditure index, in nominal and real terms, in the three rounds, at all-India levels and their standard errors. These were calculated using the
| At Prices of Ļ Andhra Pradesh Assam Bihar Gujarat Jammu & Kashmir (a,b) : a = no of states that the state “beats”, b = no of states that it gets “beaten by”. Andhra Pradesh (11,3) (6,8) (12,2) (8,6) (14,0) Assam (13,1) (3,11) (11,3) (7,7) (14,0) Bihar (11,3) (4,10) (12,2) (7,7) (14,0) Gujarat (14,0) (0,14) (10,4) (7,7) (13,1) Jammu and Kashmir (11,3) (4,10) (12,2) (7,7) (14,0) Karnataka (12,2) (3,11) (11,3) (7,7) (14,0) Kerala (11,3) (6,8) (12,2) (7,7) (14,0) Madhya Pradesh (14,0) (1,13) (11,3) (7,7) (13,1) Maharashtra (14,0) (0,14) (10,4) (7,7) (13,1) Orissa (12,2) (3,11) (11,3) (8,6) (14,0) Punjab (14,0) (0,14) (9,5) (7,7) (11,3) Rajasthan (14,0) (1,13) (10,4) (8,6) (13,1) Tamil Nadu (11,3) (3,11) (12,2) (8,6) (14,0) Uttar Pradesh (14,0) (0,14) (10,4) (8,6) (13,1) West Bengal (11,3) (4,10) (12,2) (7,7) (14,0) 44 | Karnataka (1,13) (0,14) (1,13) (3,11) (1,13) (0,14) (1,13) (2,12) (3,11) (0,14) (4,10) (3,11) (1,13) (3,11) (1,13) | Kerala (0,14) (4,10) (0,14) (6,8) (0,14) (2,12) (0,14) (5,9) (6,8) (1,13) (12,2) (5,9) (0,14) (6,8) (0,14) | Madhya Pradesh Maharashtra Orissa Punjab Rajasthan (9,5) (4,10) (2,12) (3,11) (10,4) (8,6) (6,8) (1,13) (2,12) (10,4) (8,6) (6,8) (2,12) (3,11) (10,4) (8,6) (5,9) (1,13) (2,12) (9,5) (6,6) (5,9) (3,11) (2,12) (10,4) (6,6) (6,8) (1,13) (4,10) (10,4) (9,5) (5,9) (3,11) (2,12) (10,4) (8,6) (6,8) (0,14) (3,11) (9,5) (8,6) (5,9) (1,13) (2,12) (9,5) (7,7) (6,8) (2,12) (4,10) (10,4) (6,8) (5,9) (1,13) (2,12) (8,6) (7,7) (6,8) (0,14) (2,12) (9,5) (9,5) (7,7) (2,12) (4,10) (10,4) (7,7) (5,9) (1,13) (2,12) (9,5) (8,6) (6,8) (2,12) (3,11) (10,4) January 15, 2011 vol xlvi no 3 | Tamil Nadu Uttar Pradesh West Bengal (5,9) (13,1) (7,7) (9,5) (12,2) (5,9) (9,5) (13,1) (5,9) (12,2) (11,3) (4,10) (9,5) (13,1) (6,8) (9,5) (13,1) (5,9) (4,10) (13,1) (8,6) (10,4) (12,2) (4,10) (11,3) (12,2) (4,10) (9,5) (13,1) (5,9) (13,1) (10,4) (3,11) (12,2) (11,3) (4,10) (6,8) (13,1) (5,9) (12,2) (11,3) (4,10) (9,5) (13,1) (5,9) Economic & Political Weekly |
|---|
EPWSPECIAL ARTICLE
Table 15: Real Consumption Based State Rankings at All State Prices-Urban, 61st Round
At Prices of Ļ Andhra Pradesh Assam Bihar Gujarat Jammu & Kashmir Karnataka Kerala Madhya Pradesh Maharashtra Orissa Punjab Rajasthan Tamil Nadu Uttar Pradesh West Bengal (a,b) : a = no of states that the state “beats”, b = no of states that it gets “beaten by”.
Andhra Pradesh (10,4) (8,6) (14,0) (2,12) (13,1) (5,9) (0,14) (9,5) (3,11) (7,7) (4,10) (11,3) (6,8) (12,2) (1,13)
Assam (12,2) (7,7) (14,0) (4,10) (13,1) (5,9) (0,14) (9,5) (3,11) (6,8) (12,2) (11,3) (8,6) (10,4) (1,13)
Bihar (12,2) (8,6) (14,0) (4,10) (13,1) (5,9) (0,14) (9,5) (3,11) (6,8) (12,2) (11,3) (7,7) (10,4) (1,13)
Gujarat (12,2) (7,7) (14,0) (4,10) (13,1) (5,9) (0,14) (8,6) (3,11) (6,8) (12,2) (11,3) (9,5) (10,4) (1,13)
Jammu and Kashmir (12,2) (7,7) (14,0) (4,10) (13,1) (5,9) (0,14) (9,5) (3,11) (6,8) (12,2) (11,3) (8,6) (10,4) (1,13)
Karnataka (10,4) (8,6) (14,0) (4,10) (13,1) (5,9) (0,14) (9,5) (2,12) (7,7) (3,11) (12,2) (6,8) (11,3) (1,13)
Kerala (10,4) (8,6) (14,0) (9,5) (13,1) (4,10) (0,14) (9,5) (3,11) (7,7) (12,2) (11,3) (6,8) (12,2) (1,13)
Madhya Pradesh (12,2) (7,7) (14,0) (4,10) (13,1) (5,9) (0,14) (8,6) (3,11) (6,8) (12,2) (11,3) (10,4) (9,5) (1,13)
Maharashtra (12,2) (7,7) (14,0) (4,10) (13,1) (5,9) (0,14) (8,6) (3,11) (6,8) (12,2) (11,3) (9,5) (10,4) (1,13)
Orissa (12,2) (7,7) (14,0) (9,5) (13,1) (4,10) (0,14) (9,5) (3,11) (6,8) (12,2) (11,3) (8,6) (10,4) (1,13)
Punjab (13,1) (9,5) (14,0) (4,10) (12,2) (6,8) (3,11) (8,6) (2,12) (7,7) (1,13) (10,4) (11,3) (9,5) (0,14)
Rajasthan (12,2) (7,7) (14,0) (9,5) (13,1) (4,10) (0,14) (8,6) (3,11) (6,8) (12,2) (11,3) (9,5) (10,4) (1,13)
Tamil Nadu (9,5) (8,6) (14,0) (11,3) (13,1) (5,9) (0,14) (10,4) (4,10) (7,7) (12,2) (11,3) (6,8) (12,2) (1,13)
Uttar Pradesh (12,2) (7,7) (14,0) (4,10) (13,1) (5,9) (1,13) (8,6) (3,11) (6,8) (12,2) (11,3) (10,4) (9,5) (0,14)
West Bengal (12,2) (8,6) (14,0) (4,10) (13,1) (5,9) (0,14) (9,5) (3,11) (6,8) (12,2) (11,3) (7,7) (10,4) (1,13)
NSS Rounds Rural Urban Nominal Expenditure Real Expenditure Nominal Expenditure Real Expenditure
| NSS 50 | 259.9079 | 259.9079 | 361.6901 | 361.6901 |
|---|---|---|---|---|
| SEb | 0.5451 | 0.5451 | 1.2311 | 1.2311 |
| 95% UB | 260.9764 | 260.9764 | 364.1030 | 364.1030 |
| 95% LB | 258.8394 | 258.8394 | 359.2772 | 359.2772 |
| NSS 55 | 407.8796 | 286.4917 | 587.5809 | 411.3852 |
| SEb | 0.8557 | 0.6137 | 2.2148 | 1.6429 |
| 95% UB | 409.5568 | 287.6944 | 591.9218 | 414.6052 |
| 95% LB | 406.2024 | 285.2889 | 583.2401 | 408.1652 |
| NSS 61 | 655.4721 | 411.6605 | 882.6691 | 474.4904 |
| SEb | 2.0557 | 1.2691 | 6.8239 | 3.7777 |
| 95% UB | 659.5012 | 414.1479 | 896.0436 | 481.8947 |
| 95% LB | 651.4429 | 409.1731 | 869.2945 | 467.0862 |
a Mean expenditure discounted by inequality. b SE = Bootstrap standard error of the estimate; UB = Upper bound, LB= Lower bound.
bootstrap methods outlined in Mills and Zandvakili (1997) and Biewen (2002). The estimations were based on repeated replications of sub-samples of 10,000 observations. Let us recall that the price effects kick in only in the 55th and 61st rounds when there is a divergence between the nominal and real magnitudes. There has been a general improvement in living standards and this improvement is statistically significant as seen from the upward shift in the non-overlapping confidence intervals at 95% confi dence level. Note, however, that the improvement in nominal terms overstates the real improvement in living standards. The difference between the nominal and real magnitudes of the expenditure index is due to the two ways that prices enter the latter, namely, in the price adjustment made in the calculation of utility compensated real expenditures, and the difference between the nominal and real expenditure inequality magnitudes. These results further underline the importance of incorporating the price and inequality movements in welfare comparisons.
Further evidence on the all-India picture on welfare change is contained in Table 17 which estimates the change in magnitude of the expenditure index between successive rounds and over the period as a whole. While there was an improvement in living standards in rural areas in both nominal and real terms, the picture is quite different in the urban areas. The large increase in urban inequality between the 55th and 61st rounds swamped the
Economic & Political Weekly
EPW
| NSS Rounds | Rural | Urban | ||
|---|---|---|---|---|
| Nominal Expenditure | Real Expenditure | Nominal Expenditure | Real Expenditure | |
| NSS 55-NSS 50 | 93.6864 | 22.9587 | 223.6037 | 50.9691 |
| SEb | 0.2254 | 0.2105 | 0.5139 | 0.5075 |
| 95% UB | 94.1283 | 23.3713 | 224.6110 | 51.9639 |
| 95% LB | 93.2445 | 22.5460 | 222.5964 | 49.9742 |
| NSS 61-NSS 55 | 260.0843 | 109.7569 | 13.6611 | -86.4239 |
| SEb | 0.8257 | 0.4330 | 1.0867 | 0.5816 |
| 95% UB | 261.7028 | 110.6057 | 15.7912 | -87.5639 |
| 95% LB | 258.4657 | 108.9082 | 11.5310 | -85.2839 |
| NSS 61-NSS 50 | 353.7706 | 132.7156 | 237.2648 | -35.4548 |
| SEb | 0.7418 | 0.4981 | 0.9127 | 0.5160 |
| 95% UB | 355.2247 | 133.6921 | 239.0539 | -34.44339 |
| 95% LB | 352.3166 | 131.7392 | 235.4757 | -36.4663 |
a Mean expenditure discounted by inequality. b SE = Bootstrap standard error of the estimate; UB = Upper bound, LB= Lower bound
gain in median consumption and led to a decline in urban living standards, as revealed by the direct bootstrap estimates of the magnitude of change in the inequality adjusted expenditure measure. This decline undid the gains in the earlier period so that the period as a whole recorded a decline in the measure for urban India.
Table 18 provides direct (and further) evidence of the importance of the price movements in the welfare comparisons by reporting the bootstrap estimates of the difference between the nominal and real magni-
Table 18: Bootstrap Estimate of Difference between tudes of the expenditure-Nominal and Real Values of the Inequality Adjusted
Expenditure Index
based welfare measure
Welfare Index in the two comparison Rural Urban
NSS 55 93.6863 172.6349
rounds, 55th and 61st. In
95% CI [94.1282,93.2444] [172.7238,172.5461]
each sector of the Indian
NSS 61 244.0136 272.7199
economy, the nominal
95% CI [244.6513,243.3759] [273.7202,271.7196]
magnitudes overstate the real improvements in living standards, with this overstatement being larger in the urban areas.
7 Concluding Remarks
Notwithstanding the wide recognition that movement in prices, especially relative prices, and changes in inequality will affect welfare comparisons between households over time and between
45 regions, there does not seem to be much of a literature incorporating them in welfare analysis. For example, changes in relative prices will open up divergence between nominal and real expenditure inequalities that will persist even if all the nominal expenditures are deflated by a household invariant price index. Again, in a period of rising inequality, a comparison of summary measures of welfare based on mean or median expenditure, even if inflation adjusted, will give a misleading picture of changes in living standards. The simultaneous effect of relative price and inequality changes, if ignored, may lead to a distorted picture of changes in living standards. The distortion gets worse if there is interaction between the price and inequality effects with changes in relative prices affecting the poorer households with limited substitution ability more strongly than the affl uent ones.
This paper provides evidence on this issue for India during a period that has attracted much interest. We do so through the use of a welfare measure that incorporates relative prices and
Notes
1 Further details are available on request. 2 It is customary to use the concept of relative poverty line based on, for example, the half median in cross country poverty comparisons involving developed countries similar to the cross state comparisons attempted here (see, for example, Buhmann et al 1988; Lancaster et al 1999). 3 Since the ranking of states was virtually identical between the nominal and real expenditures, we have not reported the former, but will make them available on request.
References
Aten, Bettina and Tatiane Menezes (2002): “Poverty Price levels: An Application to Brazilian Metropolitan Areas”, World Bank, ICP Conference, Washington DC, 11-15 March.
Banks, J, R Blundell and A Lewbel (1997): “Quadratic Engel Curves and Consumer Demand”, Review of Economics and Statistics, 79(4), 527-39.
Bhattacharya, N and G S Chatterjee (1971): “Consumer Prices and Per Capita Household Consumption in Rural India”, Economic & Political Weekly, 6(44), 2249-53.
Biewen, M (2002): “Bootstrap Inference for Inequality, Mobility and Poverty Measurement”, Journal of Econometrics, 108, 317 -42.
Buhmann, B, L Rainwater, G Schmaus and T Smeeding (1988): “Equivalence Scales, Well-Being, Inequality, and Poverty: Sensitivity Estimates across Ten Countries Using the Luxembourg Income Study (LIS) Database”, Review of Income and Wealth, 34, 115-42.
Coondoo, D, A Majumder and S Chattopadhyay (2009): “Estimating Consumer Price Indices through Engel Curve Analysis”, forthcoming in Review of Income and Wealth.
Coondoo, D, A Majumder and R Ray (2004): “A Method of Calculating Regional Consumer Price Differentials with Illustrative Evidence from India”, Review of Income and Wealth, 50 (1), 51-67.
Lancaster, G, R Ray and M R Valenzuela (1999): “A Cross-Country Study of Household Poverty and Inequality on Unit Record Household Budget Data”, Economic Development and Cultural Change, 48(1), 177-208.
Mills, J and S Zandvakili (1997): “Statistical Inference via Bootstrapping for Measures of Inequality”, Journal of Applied Econometrics, 12, 133-50.
Mishra, A and R Ray (2009): “Prices, Inequality and Poverty: Methodology and Indian Evidence”,
Monash University
inequality. The results confirm that in a period of relative price changes and rising inequality, the omission of these factors will lead to a large overstatement of the gains in real living standards. The importance of relative prices, not just the overall price level, is also established from the sensitivity of the state rankings on median real consumption to the state prices at which the expenditure values are calculated. The latter result points to the importance of regional price differentials in welfare comparisons between households residing in different states. Other notable results include the feature that states that are widely recognised as affluent on aggregate consumption levels or poverty rates do not do as well if we focus on inequality or their food consumption. These results, while not denying that there have been overall welfare gains in India during the period of economic reforms and beyond, warn against the danger of overstating the gains by ignoring relative price changes and inequality movements.
Discussion paper, 24-09,
available at http://www.buseco.monash.edu.au/ eco/research/papers/2009/index.html.
Muellbauer, J (1974): “Prices and Inequality: The United Kingdom Experience”, Economic Journal, 84, 32-55.
Nicholas, A, R Ray and M R Valenzuela (2008): “Evaluating the Distributional Implications of Price Movements: Methodology, Application and Australian Evidence”, forthcoming in The Economic Record.
Ray, R (1983): “Measuring the Costs of Children: An Alternative Approach”, Journal of Public Economics, 22, 89-102.
NEW EPW 5-Year CD-ROM 2004-08 on a Single Disk
The digital versions of Economic and Political Weekly for 2004, 2005, 2006, 2007 and 2008 are now available on a single disk. The CD-ROM contains the complete text of 261 issues published from 2004 to 2008 and comes equipped with a powerful search, tools to help organise research and utilities to make your browsing experience productive.The contents of the CD-ROM are organised as in the print edition, with articles laid out in individual sections in each issue.
With its easy-to-use features, the CD-ROM will be a convenient resource for social scientists, researchers and executives in government and non-government organisations, social and political activists, students, corporate and public sector executives and journalists.
Price for 5 year CD-ROM (in INDIA) Individuals - Rs 1500 Institutions - Rs 2500
To order the CD-ROM send a bank draft payable at Mumbai in favour of Economic and Political Weekly. The CD can also be purchased on-line using a credit card through a secure payment gateway at epw.in
Any queries please email: circulation@epw.in
Circulation Manager,
Economic and Political Weekly
320-321, A to Z Industrial Estate, Ganpatrao Kadam Marg, Lower Parel, Mumbai 400 013, India
January 15, 2011 vol xlvi no 3
EPW
India in the 1990s”, Development and Change, 38 (2), 321-44.
Ray, R and G Lancaster (2005): “On Setting the Poverty Line Based on Estimated Nutrient Prices: Condition of Socially Disadvantaged Groups during the Reform Period”, Economic & Political Weekly, 40(1), 46-56.
Reddy, Sanjay and Thomas Pogge (2007): “How Not to Count the Poor” in Sudhir Anand and Joseph Stiglitz (ed.), Measuring Global Poverty (Oxford: OUP).
Sen, A (1976): “Real National Income”, Review of Economic Studies, 43 (1), 19-39.
Subramanian, S (2005): “Unravelling a Conceptual Muddle”, Economic & Political Weekly, 40 (1), 57-66.
– (2009): “How Many Poor in the World? A Critique of Ravallion’s Reply”, Economic & Political Weekly, 44 (5), 67-71.
NEW 
