Figure 1: Feedback Loops at Macroeconomic Level

Economic growth

Human capital Income poverty formation reduction

In the context of feedback loops, IHDR 2011 assesses the performance across multiple dimensions of human development across states and across all social and religious groups. The main thrust of the report is analysing whether certain sections of India’s socially stratifi ed society that suffer from multiple deprivations have begun to share the benefits of development. The historically marginalised sections of society are largely concentra ted in the states that appear at the bottom end in human development rankings. Bihar, Chhattisgarh, Jhar khand, Madhya Pradesh, Orissa, Rajasthan and Uttar Pradesh together account for 47% of all SCs, 52% of all STs and 43% of Muslims in the country. In addition, SCs and STs account for more than 40% of the population in Chhattisgarh, Jharkhand, Madhya Pradesh and Orissa, while it is 32% in Rajasthan, 26% in U ttar Pradesh and 20% in Bihar (largely due to the SC population in the last two). Compared to these, SCs account for 20%, STs 9% and Muslims 13% of the country’s p opulation as a whole (NSS 2007-08).

The analysis is spread across eight chapters that examine the multiple dimensions of development – employment, poverty, nutrition, health, education, support infrastructure (such as housing, access to electricity, telephones and road connectivity), disability and child labour. In addition, an HDI is constructed for the states.

2 HDI of States

The HDI is a composite index consisting of three indicators – consumption expenditure (as a proxy for income), education and health. The income index is calculated using real per capita consumption expenditure4 adjusted for inequality. Life expectancy at birth is used to construct the health index. And the education index is constructed as a weighted average of literacy seven years and above and the mean years of schooling adjusted for out-of-school children.5

The IHDR 2011 estimates HDI for the beginning of the decade and for the year 2007-08.6 The top five ranks in HDI in both years are occupied by Kerala, Delhi, Himachal Pradesh, Goa and Punjab. States that perform better on health and education outcomes are also the states with higher HDI and thus higher per capita income, which is reflected in improved human functioning. However, at the other end of the spectrum are states such as Chhattisgarh, Orissa, Bihar, Madhya Pradesh, Jharkhand, Uttar Pradesh and Rajasthan. These states have over time shown tremendous improvement in their HDI and its component indices over time, leading to a convergence in HDI across states.7 The coefficient of variation of the HDI for states in 2000 was 0.313 and this had fallen sharply to 0.235 in 2008.

India’s HDI registered an improvement of 21% in the eight-year period 1999-2000 to 2007-08. This improvement was largely driven by achievements in the education sector – the education index improved by 28.5%. The pace of improvement in the health sector was lower (13%). Also, because the income index is constructed using monthly per capita consumption expenditure from the National Sample Survey (NSS), which is lower than both consumption estimates and income, the income index computed using the net

domestic product per capita results in an increase in the HDI of more than 30% and not 21% over the eight-year period. It, however, needs to be qualified here that the arguments in the IHDR 2011 are not merely based on these percentage changes, but on an analysis of each individual dimension of human development.

An issue was recently raised over whether the HDI of low income states is actually converging with that of India’s. Chakraborty (2012) argues that the HDI of Bihar (a low per capita income state) increased by 0.075 points while the n ational average increased by 0.080 points. So, he concludes, the IHDR 2011 argument does not hold. But is it not commendable that despite a low base, the improvement for Bihar is almost similar as that for all India? What is significant is that the absolute improvements in health and education indices for low per capita income states such as Chhattisgarh, Jharkhand, Madhya Pradesh and Orissa have been better than for all India, with their gaps with the all-India average narrowing over time.8

In six of the low HDI states – Bihar, Andhra Pradesh, Chhattisgarh, Madhya Pradesh, Orissa and Assam – the improvement in HDI (in absolute terms) is considerably more than the national average. If we look at absolute changes in HDI over the decade, our conclusion that the poorer states are catching up with the national average is strengthened. For instance, in Uttarakhand, the increase in HDI has been 0.151 points between 19992000 and 2007-08 compared to the n ational average of 0.080 points. Other relatively poor states that have seen an improvement in HDI greater than the all-India average are Assam (0.108 points), Jharkhand (0.108 points), Madhya Pradesh

(0.090 points) and Orissa (0.087 points). Chhattisgarh with an improvement of

0.080 points has performed as well as the national average in terms of HDI. However, among the relatively poor states, the increase in HDI in Bihar (0.075 points) and Uttar Pradesh (0.064 points) was less than the national average ( Table 2, p 61). But the relative improvement (that is, percentage change) in HDI is greater in Bihar than the national a verage. As Table 2 shows, the percentage change in HDI is greater for the majority

vol xlviI no 14

NOTES

of low per capita income states than the HDI improvement for India as a whole.

However, the real issue lies elsewhere. Everyone knows that the HDI is an extremely crude indicator. The team leader of IHDR 2011 recall Mahbub ul

Table 2: Human Development Index (1999-2000 and 2007-08)

Haq saying to Amartya Sen, “Construct for me an index which is as crude as gross national product per capita”. Both Sen and Haq were clear that they were constructing a crude index, but a simple one that had the saving grace of being multidimensional (incomes, health, eduthat is the essence of the capability approach. Individual human functioning matters. That is, the probability of my dying prematurely, the probability of my remaining illiterate or even merely with primary education, the probability of my remaining undernourished are central to the realisation of my human potential. Every chapter in the IHDR 2011 clearly demonstrates that the states of India are converging on important indicators of human functioning and that the indicators of SCs, STs and Muslims – the most marginalised of Indian society – are converging with the national average (on which more later).9

Tables 3 and 4 show that this trend of convergence is also evident in the health and education indices. Chhattisgarh

(0.075 points), Uttar Pradesh (0.075), Orissa (0.074), Madhya Pradesh and R ajasthan (0.067 points each) saw an improvement in the health index greater than the national average of 0.066 points over 2000 and 2008. Jharkhand’s improvement in the health index was the same as that of all-India. Among the relatively poorer states, Bihar’s improvement in the health index was lower than the national average (0.058 points). H owever, Bihar’s difference with the national average Bihar has fallen over time.

Table 3: Health Index (2000 and 2008)

Table 4 shows that the relatively poor (with low per capita income) states (which are also educationally backward) such as Uttarakhand (0.267 points), Jharkhand (0.213), Chhattisgarh (0.161), Madhya Pradesh (0.157), Bihar (0.137), and Orissa (0.126) saw improvements in the education index higher than the national average (0.126 points) between 1999-2000 and 2007-08. These states

Table 4: Education Index (1999-2000 and 2007-08)

State Education Education Change in % Index Index Education Change 2007-08 1999-2000 Index

also managed to narrow their gaps with the all-India education index and had percentage improvements higher than

as a means of assessing the welfare of human beings. It is remarkable that some academics fail to understand the main argument of the IHDR 2011, which is clearly elaborated in its eight chapters and goes well beyond using the HDI as a the poorer states, Uttar Pradesh (0.121) and Rajasthan (0.114) had lower improvements in the absolute value of their education index than the national average. But due to their relatively low base, the pace of improvement in terms of per

Using the HDI alone as a measure of the improvement in the human development levels of Indian states would understate the complexity of the notion of human d evelopment, amounting to a crudely

Bihar 0.563 0.506 0.058 11.412

n ational average.

3 SC, ST Human Well-being

One of the main fi ndings of IHDR 2011 is that there is a convergence on most Muslims (Figures 2 to 7). SCs, STs and Muslims are converging with the all-India average on most health, education and income indicators such as infant mortality rate (IMR), under-five mortality rate, total fertility rate (TFR), contraception prevalence rate, child immunisation rate, literacy rate and unemployment rate, while nutrition and sanitation are two major exceptions where these groups are diverging from the national average.

EPW

(NFHS). The IMR across social and religious groups shows that a higher number of Muslim infants lived beyond age one compared to the national average (in 2005-06). SCs and STs have also been converging with the all-India average over time. The fall in the IMR was sharpest among STs, declining by 22 points from 84 per 1,000 live births in 1998-99 to 62 per 1,000 in 2005-06, compared to a fall from 67.6 to 57 per 1998-99 to 2005-06, while the percentage of Indian women with BMI less than 18.5 declined from 36% to 33%, the percentage of SC and ST women with BMI less than

18.5 remained unchanged – 41% and 46% respectively. And the incidence of malnutrition among Muslim women increased

Figure 3: Total Fertility Rate: SCs and Muslims Converging with National Average

3.06

2.85

2.68

3

TFR

1998-99 2005-06

determines and is highly correlated with

Figure 2: Infant Mortality Rate: SCs, STs

All India SCs STs Muslims

90

that mortality rates are better indicators

80

for assessing overall well-being and ine-

Thus, all these groups showed a diver

70

quality than income estimates. He argues

83 66.4 58.8 52.4

IMR

62.1

60

that mortality information refl ects the dence of female malnutrition (Figure 4).

57

50

nature of social and economic inequalities, This was also true of the percentage of

including gender bias and racial disparities. Sample registration system (SRS) estimates show that over the last decade (2000 to 2009), the IMR has dramatically fallen in rural and urban areas. It was 68 per 1,000 live births in 2000, which dropped to 50 in 2009. The greatest improvements were in Orissa (from 95 in 2000 to 65 in 2009), Jharkhand (from 70 in 2000 to 44 in 2009), Chhattisgarh (from 79 in 2000 to 54 in 2009) and Tamil Nadu (from 51 in 2000 to 28 in 2009), followed by Madhya Pradesh (from 87 in 2000 to 67 in 2009), Rajasthan (from 79 in 2000 to 59 in 2009) and Uttar Pradesh (from 83 in 2000 to 63 in 2009). So, in the relatively poor states the IMR improved by

40 1998-99 2005-06 All India SCs STs Muslims

the state of the public health systems (primary health centres or PHCs, subcentres, district hospitals) in the country. More importantly, from the perspective of human well-being, these indicators reflect the degree of autonomy women enjoy in household decisionmaking and how much time and energy they have to perform tasks other than just child-rearing.

Nine major states – Andhra Pradesh, Delhi, Himachal Pradesh, Karnataka, Kerala, Maharashtra, Punjab, Tamil Nadu, and West Bengal – achieved the women with anaemia. More than half of Indian women suffer from anaemia and there is an increasing trend of it among SC, ST and Muslim women.

Poor nutritional outcomes are often a result of constant exposure to waterborne diseases, which are due to lack of proper sanitation facilities and open defecation. Of the 1.1 billion people in the world practising open defecation, 638 million or 58% live in India (WHO and UNICEF 2010). Despite an increase in the number of toilets being constructed, open defecation remains one of the

Figure 4: BMI: SCs, STs and Muslims Diverging from National Average

50

46.3

% of women with BMI <

45

41.1

40 35.8 35 34.1

35.2

30

25

improvements in rural areas. In 2000, the IMR in rural areas was 74 per 1,000 live births and this fell to 55 in 2009. In urban India, the decline was from 44 per 1,000 live births in 2000 to 34 in 2009. There was a sharp reduction in the ruralurban gap in the poorer states – Andhra Pradesh, Arunachal Pradesh, Assam, Chhattisgarh, Jharkhand, Karnataka, Madhya Pradesh, Orissa and Uttarakhand. Estimates for the IMR across social and religious groups are available from National Family Health Surveys health system in these states improves, the population growth rate cannot be r educed. While the TFR has been converging in the case of SCs and Muslims (Figure 3), the ST TFR marginally increased from 1998-99 to 2005-06. This is surprising because the contraception prevalence rates for all groups – SCs, STs and Muslims – are converging with the national average. The decline in the TFR was the highest in the case of SCs.

A body mass index (BMI)10 less than 18.5 shows malnutrition among women. From

1998-99 2005-06 India Muslims SC ST

largest threats to health and nutritional s tatus, in addition to the safety of women and girls. There are also concerns over toilets built through subsidy programmes often being used for storage, bathing and washing. Even if a single household defecates in the open, it can be a source of diarrhoea in all neighbouring households.

Our analysis of the NSS data for IHDR 2011 shows that about half of Indian

EPW

NOTES

households lacked access to sanitation facilities in 2008-09. In Bihar, Chhattisgarh, Jharkhand, Madhya Pradesh, Orissa, Rajasthan and Uttarakhand more than 60% of households were without toilets. The situation was more dismal in rural areas, particularly in these states, where more than 75% of the households did not have toilet facili

49% of Indian households, as much as 65% of SC and 69% of ST households lacked toilet facilities and were diverging from the national average over time (Figure 5).

Figure 5: Sanitation: SCs and STs Diverging from National Average

80 78

beginning of the decade. However, compared to the all-India average of 66%, only 58% of SC and 38% of ST households reside in pucca houses (NSS 200809) and the gap with the national

Figure 6: Literacy Rate: SCs, STs and Muslims Converging from National Average

80 72

all other indicators. This clearly demonstrates that government policies to promote SC-ST access to these three services need to be more carefully designed.

4 Muslims’ Social Indicators

The prime minister’s high-level committee, the Sachar Committee on the Status

70

64.8

Literacy rate

ties (IHDR 2011). This was refl ected in

the poor access of SC and ST households ried out an analysis of the social, eco

60

59.1

60.5

54.7

nomic and educational status of Mus

50

47.1

to

sanitation facilities. Comparedlims in India and came to the conclusion

to

40

2001 2007-08 All India SCs STs Muslims

average has been widening over time. According to NSS estimates, there was an overall increase in households with electricity connections for domestic use from 64% in 2002 to 75% in 2008-09. Even the poorer states saw a huge inthat Muslims were doing much worse than the rest of the population on most social indicators. The IHDR 2011 has undertaken an analysis to examine how the socio-economic indicators of the Muslim community have evolved over the past decade.

Table 5 shows a series of indicators

74.9 70

% households withno toilet facility

60

59.8

50 49.2

40

We saw earlier that improvements in the HDI have been largely guided by improvements on the education front. With an all-round increase in literacy across the country, there was a decline in the interstate variation in literacy rates. both in rural and urban India, showing that the poorer and the educationally higher than the all-India average, SCs and STs are converging with it (Figure 7).

To summarise, but for nutrition, sanitation and pucca houses, SCs and STs are converging with the national average on

Figure 7: Access to Electricity: SCs and STs Converging with All-India Average

covers a 10-year period beyond what was captured in the Sachar Committee report. The average for Muslim underweight children (48.3%) as per the NFHS 2 (1998-99) and the Sachar Committee r eport was worse than the all-India average (47%). However, over eight years

after 1999, there was a drop in the per

80

% households withelectricity for domestic use

75

backward states are catching up with

the national average. The rate of in

crease in literacy was higher in rural

a reas, implying that rural and urban

I ndia is converging. SCs, STs and Muscentage of Muslim underweight children

by 6.5 percentage points compared to a

70

66

64

60

61

52

drop of 4.5 percentage points in the

50 47

national average. As a result, the Muslim

40

2002 2008-09

average for underweight children was

lims are catching up with the all-India literacy rate with the sharpest increase among STs. The literacy rate increased by 13 percentage points among STs from 2001 to 2007-08, compared to 7 percentage points in all India (Figure 6).

Determinants such as housing conditions, access to electricity, telephones and road connectivity are crucial inputs in the development process. Supporting human development, infrastructure is a source of positive externalities and a stimulant to economic growth. There have been improvements in housing conditions over the past decade. Twothirds of Indians now reside in pucca houses compared to less than 50% at the

Economic & Political Weekly

EPW

Table 6: Convergence on Most Human Development Indicators among SCs, STs and Muslims IHDR 2011 does not refer to the fi rst National Indicators Convergence Divergence Human Development Report 2011, which is not the case, as seen in Chapter 2 of IHDR 2011.

Per capita consumption expenditure SCs STs, Muslims

Unemployment rate SCs, Muslims STs

reduction is a questionable one”. The IHDR Child labour rate SCs, STs, Muslims 2011 argues that economic growth is a neces-Health and nutrition sary condition for the reduction of income pov-Female malnutrition (Body Mass Index < 18.5) SCs, STs, OBCs, Muslims erty, but not a sufficient one. Investment in human capital is required if the income poor

Women with anaemia SCs, STs, OBCs, Muslims

Infant mortality Rate SCs, STs, OBCs, Muslims

Under five mortality rate SCs, STs, Muslims

Total fertility rate SC, Muslims STs, OBCs 3 Incidence of poverty based on Tendulkar pov-Contraception prevalence rate SCs, STs, Muslims erty line estimates.

Child immunisation STs, Muslims SCs 4 It is because of the lack of state-wise data on income per capita that the income index is com-

Toilet facility OBCs SCs, STs

main reason why the IHDR 2011 estimate of the Literacy SCs, STs, Muslims HDI is different from the global HDR 2011. Support infrastructure5 Global HDRs use mean years of schooling and Pucca houses OBCs STs expected years of schooling to construct the

Electricity for domestic use SCs, STs, OBCs

april 7, 2012