ISSN (Print) - 0012-9976 | ISSN (Online) - 2349-8846

A+| A| A-

Economic Growth and Human Development in Indian States

This paper evaluates relative performance of 15 major Indian states on human development, and examines the two-way nexus between economic growth and human development. The estimates of cross-sectional growth regressions provide strong evidence of regional convergence in human development despite considerable divergence in real per capita income, indicating that the poor states that have failed to catch up with the rich ones in terms of per capita income have managed to catch up in terms of human development. The classification of the states based on their performance on HD and EG reveals that while only four states have been in the virtuous cycle category, as many as seven states have been in vicious cycle. The results suggest that the sequencing of policy should be such that the HD-induced growth process has to be strengthened for lifting the states from the vicious to virtuous cycle category.

Economic Growth and Human Development in Indian States

This paper evaluates relative performance of 15 major Indian states on human development, and examines the two-way nexus between economic growth and human development. The estimates of cross-sectional growth regressions provide strong evidence of regional convergence in human development despite considerable divergence in real per capita income, indicating that the poor states that have failed to catch up with the rich ones in terms of per capita income have managed to catch up in terms of human development. The classification of the states based on their performance on HD and EG reveals that while only four states have been in the virtuous cycle category, as many as seven states have been in vicious cycle. The results suggest that the sequencing of policy should be such that the HD-induced growth process has to be strengthened for lifting the states from the vicious to virtuous cycle category.

MADHUSUDAN GHOSH

I Introduction

S
ince the early-1990s, there has been a notable shift in the focus of development planning from mere economic growth to enhancement of human well-being. Sen (1985, 1987), Dasgupta (1993) and United Nations Development Programme [UNDP 1990] argue for viewing development as the process of enhancing people’s capabilities for improving quality of life. Mere economic growth in production of goods and services and the consequent growth in per capita income may not necessarily lead to an improvement in human well-being, which is broadly conceived to include not only consumption of goods and services, but also the accessibility of people to the basic necessities required for leading a productive and socially meaningful life. Human development is defined as a process of enlarging people’s choices to lead a long and healthy life, to acquire knowledge and be educated, and to have access to resources needed for a decent level of living. The Human Development Index (HDI), which is used as a summary measure of three dimensions of human well-being, viz, a long and healthy life, education, and a decent standard of living, is usually constructed by combining three indices, viz, life expectancy index, education index and income index. The per capita real income is usually considered as a means of good living and a catch-all variable capturing those aspects of well-being not well represented by life expectancy or literacy. Although the conventional measure of well-being such as per capita income does not capture the wider aspects of well-being as represented by HDI, it is a predominant means in advancing human development. Naturally, a close relation between economic growth (measured as an increase in per capita income) and human development is expected. While economic growth (EG) provides resources to achieve improvements in human development (HD), improvements in human capital can play a significant role in achieving EG. The two-way link between EG and HD can be easily conceived of, drawing inferences from theoretical growth literature. However, very few attempts have been made for investigating these links

empirically [see, for example, Ramirez et al 1998; Ranis and Stewart 2000, 2001]. Ramirez et al (1998) examine the relationships with the help of cross-country (developing countries) data for the period 1970-92, and find that there exists a strong positive relationship in both directions. Based on the results, they conclude that while ideally both EG and HD should be promoted simultaneously, HD should be given priority when a choice is needed due to resource or any other constraints. Ranis and Stewart (2001) extend this analysis to the experience of Latin American countries for the period 1960-92. Their results reinforce the earlier conclusion that a balanced approach to EG and HD has to be adopted. It is also imperative to emphasise on HD from the very outset of any development programme, as policies that emphasise economic growth alone do not ensure sustainable HD. Evaluating the performance of some selected countries from Africa, Asia and Latin America on HD during the period 1960-95, Ranis and Stewart (2000) have identified the best performing countries and the conditions for making success in HD. Based on general reasoning and empirical evidence, they have suggested the conditions for sustainability of success in HD.

There exist very few studies examining the two-way link between EG and HD on the basis of development experience of different regions within a country. This type of study is useful to learn about the performance of a country at the national and regional levels. This is more so particularly in the context of a developing economy like India. The Indian government has been concerned about how to improve the standard of living of its people. Since the results of an exercise on the two-way link between economic growth and human development have natural policy implications, it seems worthwhile to make an attempt to examine the relationships on the basis of the performance of Indian states. This enables us to suggest policies for improving human development.

There is another reason for studying the performance of Indian states on HD. There are concerns for regional inequality in development in general and in the standard of living in particular. This concern has always been expressed in government’s policies and planning since independence. Since inter-regional disparity in per capita state domestic product (SDP) does not necessarily imply the same degree of regional disparity in the standard of living or human development, special attention has been focused on the issue of inter-regional disparity in human development. A few empirical studies dealing with inter-regional disparity in human development are, however, available in the existing literature [see, for example, Kurian 2000; Singh et al 2003; Dholakia 2003]. While Singh et al (2003) find no evidence of absolute or conditional divergence in human development across 14 major states in India, Dholakia (2003) observes that while per capita income does not show any significant trend in regional disparity, the overall indices of human development show a clear and highly significant declining trend during 1981-2001. A review of the studies in the existing literature on HD indicates the need for undertaking further research to shed more light on the issue to guide development strategy at the national and regional levels.

II Objective and Database

We evaluate the relative performance of Indian states on human development, and examine whether regional disparity in it has increased or decreased over time. We also investigate the twoway link between EG and HD (viz, growth-induced HD and HDinduced EG) and explore the possibilities of vicious cycle, virtuous cycle, lopsided-EG and lopsided-HD, and then classify the states into these categories. Based on the results, we suggest policies for achieving sustainable improvement in HD for lifting the states from the vicious to virtuous cycle category. The data used here relate to 15 major states, viz, Andhra Pradesh (AP), Assam (AS), Bihar (BI), Gujarat (GUJ), Haryana (HAR), Karnataka (KAR), Kerala (KER), Madhya Pradesh (MP), Maharashtra (MAH), Orissa (ORI), Punjab (PUN), Rajasthan (RAJ), Tamil Nadu (TN), Uttar Pradesh (UP) and West Bengal (WB). The data set was compiled from National Human Development Report 2001 [GoI 2002], National Accounts Statistics [GoI 2003] and India Database [Chandhok and the Policy Group 1990]. The rest of the paper is organised as follows. Section III evaluates the performance of the states on HD and examines regional convergence in various indicators of HD. While Section IV investigates the two-way link between EG and HD, Section V explores the possibilities of vicious cycle, virtuous cycle, lopsided-EG and lopsided-HD, and classifies the states into these categories on the basis of their actual performance on HD and EG. Section VI summarises the main findings and draws policy conclusions.

III Performance of States on HD

In this section, we evaluate the relative performance of the states on various dimensions of HD and examine which states achieved sustained improvement in HD and which ones fell back over the period 1981-2001. The simplest way to do this is to rank the states’ performance on HD at different time points, and then to study the direction of change in the ranking during the period. Moreover, since our interest is to evaluate the achievement of the states in HD improvement, we have examined the performance of the states in HD-shortfall reduction (measured by the magnitude of change over the period in the percentage of deficiency of a state’s level of performance relative to the best performing state). It measures their ability to catch up with the best performing state by reducing shortfall in the level of HD relative to the best performing (or frontier) state.

Table 1 presents the data relating to achievements of the states in HDI and other dimensions of human well-being at different time points. We have ranked the states on the basis of their achievements in HDI, literacy rate (LR) and expectation of life at birth (ELB). Judged by the achievements in the chosen indicators of HD, Kerala is found to be the best performing state during 1981-2001. Bihar appears to be the worst performing state in terms of HDI during the same period. So far as LR is concerned, Rajasthan was the worst performing state in 1981 and Bihar in 1991 and 2001. Regarding ELB, UP was the worst performing state in 1981 and MP in 1991 and 2001. There has been a wide interstate variation in the performance of HDI. The estimated value of HDI varies from 0.237 to 0.50 in 1981, 0.308 to 0.591 in 1991 and 0.367 to 0.638 in 2001. Among the better-off states viz, Kerala, Punjab, TN, Maharashtra and Haryana had a HDI above 0.50, and the worse-off states like Bihar, Assam, UP and MP had a HDI less than 0.40 in 2001. Again, although seven states (viz, Bihar, Haryana, Kerala, Orissa, Punjab, UP and WB) could manage to maintain their relative position, and three states (viz, MP, Rajasthan and TN) managed to improve, the remaining five states (viz, AP, Assam, Gujarat, Karnataka and Maharashtra) experienced deterioration in 2001 relative to 1981. Considerable interstate variation has also been observed in the performance of LR and ELB. The LR varies from 30.11 to 81.56 in 1981,

38.48 to 89.81 in 1991 and 47.53 to 90.92 in 2001. Similarly, ELB varies from 50 to 68.4 during 1981-85, 54.7 to 72.9 during 1991-95, and 55.2 to 73.1 during 1992-96. The performance of all the states on all the indicators of human well-being has, however, improved over the period.

Evaluating the relative performance of the states in HD-shortfall reduction, we find that Rajasthan, TN, WB, MP, UP, Bihar and Orissa were more successful than the other states in reducing the shortfall in HDI relative to the best performing state (Kerala). The states were able to improve their score on HDI by more than 50 per cent by reducing the shortfall in it by more than 9.0 percentage points during 1981-2001 (Table 1). Improvement in HDI in these states has, however, taken place from a relatively lower level of HDI in the base year (1981). This finding may be construed to be an indication that the poorer states have made strides in improving their relative standard of living.

Convergence in HD

In order to get a clear idea about the nature of change in the degree of regional inequality in HD, we have considered various measures of convergence (σ, and absolute and conditional βconvergence) suggested by Barro and Sala-i-Martin (1992, 1995) and Sala-i-Martin (1996), and examined to what extent the selected indicators of human development alternative to per capita income are converging or diverging across the states over time. Since human development is measured in terms of HDI involving three dimensions of human well-being, viz, (1) education and knowledge, (2) health and longevity, and (3) a decent standard of living, we have investigated interstate variations in HDI as well as in its three components. While the first component of HDI is represented by the LR and the second one by ELB, the third is represented by per capita real income (PCI) measured in terms of per capita net SDP at constant prices. Since the extent of regional inequality in HD indicators (viz, HDI, LR and ELB)

Economic and Political Weekly July 29, 2006

may not necessarily be the same as in economic development indicator (viz, PCI), we shall compare the outcomes of the HD indicators with those of the economic development indicator.

The simplest way to examine σ-convergence is to estimate the coefficient of variation (CV) of the indicators across states. The estimated CVs of the indicators are presented in Table 1. It can be seen that that while the CVs of HDI, LR and ELB have consistently declined, the CV of PCI has consistently increased over time. Thus, while in terms of overall HDI, LR and ELB, the regional disparity has been consistently falling, there is clear evidence of an increasing trend in the regional disparity in PCI during 1981-2001. This is confirmed by the results of absolute and conditional β-convergence of the variables.

We have undertaken the test for absolute β-convergence by estimating the following equation.

[ln(Xi,t) – ln (Xi,t–τ)]/T = α + β ln(Xi,t–τ) + εi,t (1)

where [ln (Xi,t )- ln (Xi,t-τ )]/T is the ith state’s annual average growth rate of a variable between the period t and t–τ, and ln(Xi,t ) and ln(Xi,t–τ ) are the natural logarithms of a variable at time t and t–τ, respectively. T is the length of the time period. If the coefficient on initial level of a variable bears a statistically significant negative sign, i e, if β<0, then we say that there exists absolute β-convergence. Rejecting the null hypothesis of β=0 against the alternative of β<0 implies a negative correlation between the initial level of a variable and its subsequent growth. This signifies that relatively poor performing states improve the most and thus catch up with the rich ones.

The convergence equation was estimated by the Ordinary Least Squares (OLS) method using the data for 15 major states. It may be noted that although for PCI, annual time-series data over a long period are available, most of the data on HD indicators are available for discrete years with a gap of 10 years (i e, for 1981, 1991 and 2001). Due to this reason, the convergence equation was estimated with 10 years lag in the explanatory variables. The estimated results for HDI, LR, ELB and PCI are presented in Table 2. It can be seen that the coefficients on initial level of HDI, LR and ELB are negative and statistically significant in all the cases, suggesting that there has been a strong tendency of convergence in these indicators of human development. The estimated rate of convergence in the HD indicators varies from

1.25 per cent to 4.23 per cent per annum. On the other hand, the coefficient on initial level of PCI is found to be significantly positive in all the time periods considered. This indicates that there has been a strong tendency of divergence in PCI across the states. A comparative study of the results between the preand post-reform periods provides further insights. Although the rate of convergence in HDI has declined marginally from 1.65 per cent in the 1980s to 1.48 per cent in the 1990s, the rate of convergence in LR has increased considerably from 1.25 per cent to 4.23 per cent, and that in ELB from 1.64 per cent to 1.88 per cent. On the other hand, the rate of divergence in PCI has increased from 1.2 per cent to 2.0 per cent during the same period. Thus, while the HD indicators exhibit significant convergence, PCI displays significant divergence across the states during all the time periods considered. This indicates that the gap between rich and poor states observed for PCI has been reduced for the HD indicators. Despite significant divergence in PCI, the high rate of convergence in LR and ELB made it possible to achieve sustained reductions in the regional disparity of HDI. This is confirmed by the results of an estimated conditional convergencetype regression presented below.

[ln(HDI)t – ln(HDI)t–τ]/T = – 0.295 – 0.0555 ln(HDI)t–τ (–2.40) (–4.192)*

+ 0.0159 ln(LR)t-τ + 0.0478 ln(ELB)t-τ (2.339)** (1.967)** ⎯R2 = 0.658; F = 19.623; N = 30. (Figures in parenthesis are t-statistics. * and ** indicate significant at 1 and 5 per cent level respectively).

It can be seen that the conditioning variables (LR and ELB) have significantly positive effects on the annual average growth rate of HDI, the coefficients on the variables being positive and statistically significant. The inclusion of LR and ELB as conditioning variables in the convergence equation has led to an increase in the rate of convergence in HDI from 1.46 per cent (absolute convergence, see Table 2) to 5.55 per cent (conditional convergence in the above equation) per year. This seems to have

Table 1: Performance of States on Human Development

State Human Development Index (HDI) HDI Shortfall Literacy Rate (LR) (Per Cent) Expectation of Life at Birth Per Capita SDP at 1980-81
Reduction (ELB) (Years) Prices (PCI) (Rupees)
1981 1991 2001 (1981-2001) 1981 1991 2001 1981-85 1991-95 1992-96 1980-81 1990-91 2000-01
AP 0.298(9) 0.377(9) 0.416(10) 5.6 35.66 44.09 61.11 58.4 61.8 62.0 1380 2060 3069
Assam 0.272(10) 0.348(10) 0.386(14) 6.1 42.05 52.89 64.28 51.9 55.7 56.2 1284 1544 1670
Bihar 0.237(15) 0.308(15) 0.367(15) 10.1 32.05 38.48 47.53 52.9 59.3 59.4 917 1197 1225
Gujarat 0.36(4) 0.431(6) 0.479(6) 3.1 52.21 61.29 66.43 57.6 61.0 61.4 1940 2641 4257
Haryana 0.36(5) 0.443(5) 0.509(5) 7.8 43.88 55.85 68.59 60.3 63.4 63.8 2370 3509 4485
Karnataka 0.346(6) 0.412(7) 0.478(7) 5.7 46.21 56.04 67.04 60.7 62.5 62.9 1520 2039 3590
Kerala 0.50(1) 0.591(1) 0.638(1) 81.56 89.81 90.92 68.4 72.9 73.1 1508 1815 2778
MP 0.245(14) 0.328(13) 0.394(12) 12.8 36.63 44.2 64.08 51.6 54.7 55.2 1358 1693 2084
Maharashtra 0.363(3) 0.452(4) 0.523(4) 9.4 55.83 64.87 77.27 60.7 64.8 65.2 2435 3483 5283
Orissa 0.267(11) 0.345(12) 0.404(11) 9.9 40.97 49.09 63.61 53.0 56.5 56.9 1314 1383 1917
Punjab 0.411(2) 0.475(2) 0.537(2) 2.0 48.17 58.51 69.95 63.1 67.2 67.4 2674 3730 4897
Rajasthan 0.256(12) 0.347(11) 0.424(9) 15.3 30.11 38.55 61.03 53.5 59.1 59.5 1222 1942 2412
Tamil Nadu 0.343(7) 0.466(3) 0.531(3) 14.7 54.39 62.66 75.47 56.9 63.3 63.7 1498 2237 3643
U P 0.255(13) 0.314(14) 0.388(13) 9.8 33.35 41.6 57.36 50.0 56.8 57.2 1278 1652 1852
WB 0.305(8) 0.404(8) 0.472(8) 13.0 48.65 57.7 69.22 57.4 62.1 62.4 1773 2145 3745
India 0.302 0.381 0.472 13.6 43.57 52.21 65.2 55.5 60.3 60.7 1630 2223 3234
CV (per cent) 22.5 19.02 16.29 28.35 24.18 14.45 8.84 7.78 7.63 31.09 36.01 40.55

Notes: Figures in parenthesis are ranks of the states. The ranks in LR and ELB are not shown to avoid clumsiness. CV is coefficient of variation among the states. Kerala is the best performing state in terms of HDI, LR and ELB.

Sources: National Human Development Report 2001 [Government of India 2002]; National Accounts Statistics [Government of India 2003].

played a significant role in achieving regional convergence in HDI. This suggests that the government policies aiming at balanced regional human development should be directed towards achieving improved regional performance on LR and ELB.

Observing the importance of literacy and life expectancy at birth in achieving higher rate of growth and convergence in HDI, we make an attempt to examine if any policy variable has been responsible for the observed regional convergence in HDI, LR and ELB. This is performed by estimating a conditional β-convergence equation, which includes social sector expenditure (SSE) as the conditioning variable, and a dummy variable (Di,t) for 1991 to see if there has been any shift in the structure of the relationship in that year. Thus, the conditional β-convergence equation is specified as:

[ln(Xi,t) – ln (Xi,t–τ)]/T = αi + β ln(Xi,t–τ)

+ θ ln (SSEi,t–τ) + δDi,t + εi,t (2)

The equation was estimated by the OLS method using the data for a panel of 15 major states corresponding to three time points. The estimated results for HDI, LR and ELB are reported in Table 3. It reveals that social sector expenditure (including expenditure on education, health, water supply and sanitation, urban development, information, and welfare and labour) has significantly positive effects on annual average growth rates of HDI, LR and ELB. Moreover, the evidence of a significantly negative coefficient on initial level of the variables implies that there has been a strong tendency of convergence in these measures of human well-being. The structure of the equation for all the variables appears to have changed in 1991, the coefficient on the dummy variable being statistically significant. These results are sufficient to indicate that higher growth rate and lower regional disparity in the indicators of HD can be achieved by increasing public investment in social sectors.

On the whole, our analysis clearly shows that there has been a strong tendency of convergence in the HD indicators. The overall HDI and its two components (LR and ELB) for the states show a clear and significant declining trend in regional disparity during 1981-2001. The period, however, shows a clear evidence of increasing trend in regional inequality in PCI. This suggests that it might be possible that the poor states that failed to catch up with the rich ones in terms of per capita income have managed to catch up in terms of other indicators of human well-being. The results of conditional convergence reveal that LR and ELB have played a significant role in reducing regional disparity in HD. Moreover, the social sector expenditure has been an important factor in attaining higher rates of growth and convergence in the HD indicators. These results have natural implications for policy. The government policies for improving human wellbeing, and achieving balanced regional development should give emphasis on literacy and life expectancy at birth. Higher growth rate and lower regional inequality in various dimensions of human well-being can be achieved by increasing public investment in education, health, water supply, sanitation, etc.

IV Link between EG and HD

Using the conceptual framework based on Ramirez et al (1998), this section examines the relationships between EG and HD. Needless to say, human development is the central objective of human activity and economic growth is potentially a very important instrument for achieving it. At the same time, achievements in HD themselves can make a significant contribution to EG. Hence, we may visualise two distinct causal relationships between EG and HD – one runs from EG to HD (growth-induced HD) and the other runs from HD to EG (HD-induced EG). Drawing mainly on Ramirez et al (1998), we first outline the conceptual framework providing the formal basis of the two-way link between EG and HD, and then estimate the relationships with the Indian data.

EG-induced HD

The EG measured in terms of an increase in per capita income contributes to HD mainly through household and government spending on various activities relating to human well-being. Households’ propensity to spend on HD-related items, e g, food, education, health, etc, depends on the level and distribution of income among households. Given the distribution of income, the higher economic growth and consequently higher per capita income leads to an increase in households’ ability to spend larger

Table 2: Evidence on Absolute
βββββ
-Convergence

Estimated Equation: (1)

Dependent Variable Period Constant Coefficient R2
on Initial Level
Growth in HDI 1981-1991 0.0042 -0.0165* 0.463
(0.115) (-3.347)
Growth in HDI 1991-2001 0.0007 -0.0148* 0.486
(0.170) (-3.507)
Growth in HDI 1981-2001 0.0018 -0.0146* 0.649
(0.525) (-4.921)
Growth in LR 1981-1991 0.066 -0.0125* 0.643
(6.739) (-4.842)
Growth in LR 1991-2001 0.190 -0.0423* 0.734
(6.761) (-5.984)
Growth in LR 1981-2001 0.115 -0.0251* 0.809
(9.013) (-7.433)
Growth in ELB 1981/85-1991/95 0.0735 -0.0164** 0.283
(2.518) (-2.267)
Growth in ELB 1991/95-1992/96 0.0833 -0.0188* 0.426
(3.333) (-3.105)
Growth in ELB 1981/85-1992/96 0.0723 -0.0161** 0.336
(2.848) (-2.564)
Growth in PCI 1980/81-1989/90 -0.061 0.012*** 0.156
(-1.087) (1.552)
Growth in PCI 1990/91-2001/02 -0.124 0.020*** 0.162
(-1.257) (1.585)
Growth in PCI 1980/81-2001/02 -0.091 0.016** 0.244
(1.571) (2.050)

Notes: Figures in parenthesis are t-statistics. *, ** and *** respectively denote significant at 1, 5 and 10 per cent level. Number of observation (N) = 15.

Table 3: Conditional
βββββ
-Convergence

Estimated Equation: (2)

Independent Conditional Convergence of Variable HDI LR ELB

Constant -0.0081 0.1041 0.0483 (-0.452) (4.064) (2.990) -0.0169* -0.0268* -0.0125*

In Xi, t–τ

(-4.696) (-5.285) (-2.784) In (SSE)i, t–τ 0.0035** 0.0048*** 0.0028**

(1.747) (1.598) (1.781) -0.0054* 0.0078* -0.0062*

Di, t

(-3.633) (3.282) (-8.748)

R2 0.711 0.516 0.805 F-statistics 24.83 11.32 41.15

Notes: Figures in parenthesis are t-statistics. *, ** and *** denote significant at 1, 5 and 10 per cent level respectively. N = 30.

Economic and Political Weekly July 29, 2006

amount of income on food and non-food items, which may lead to an improvement in HD.

There also exists an indirect link between EG and HD, operating through trickle down effects of EG on the incidence of poverty. When poverty incidence is low either because of high per capita income or because of more equal distribution of income, the households’ expenditure on HD-related items is expected to be high. The empirical evidence shows that higher per capita income is associated with lower poverty incidence due to trickle down of benefits of growth. In particular, it has been demonstrated that improved agricultural performance measured as an increase in agricultural production per head of rural population is associated with lower level of rural poverty in India [see, for example, Ahluwalia 1978, 1986; Saith 1981, Bardhan 1984; Ghosh 1996]. The evidence suggests that expenditure on HDrelated items is strongly affected by the rate of poverty reduction. When poor households earn an extra income, they increase their food expenditure significantly. The evidence also indicates that expenditure on child schooling and demand for health are significantly related to household income [Deolalikar 1993; Thomas et al 1990].

The economic growth and the consequent increase in government spending on HD-related activities may lead to an improvement in HD. The allocation of government resources to HDrelated activities is a function of total public sector expenditure, the proportion of this amount allocated to the HD sectors and the nature of allocation among these sectors. The variable which may be used for measuring the influence of government expenditure on HD is the HD-allocation ratio, defined as the proportion of total government expenditure spent on HD-related activities. The public expenditure on HD inputs, viz, education, health, sanitation, drinking water, etc, constitutes instruments for achieving improvement in various dimensions of human well-being. Given the distribution of income, a higher per capita income and a higher HD-allocation ratio are expected to lead to an improvement in HD.

There are empirical studies which provide evidence in favour of the causality running from EG to HD Morris and McAlpin (1982) observe a very high degree of correlation between per capita income and components of physical quality of life index (PQLI) across nations. In the Indian context, Panikar (1980) and Gopalan (1985) show that income, employment and economic development are preconditions for raising nutritional level of the population. Geeta Rani (1995) finds that economic development is an important factor for attaining high human development in India. There are other studies that explicitly recognise strong relationships between the indicators of economic and human development. For example, Kurian (2000) observes that developed states in India are characterised by better demographic and social development, higher per capita income, lower poverty and better infrastructural facilities. Zaidi and Salam (1998) find a high correlation between per capita income and literacy rate and enrolment in higher education. Using the data for the major Indian states, Dholakia (2003) finds a two-way causality between human and economic development.

HD-induced EG

So far as the causality running from HD to EG is concerned, there is a natural presumption that as people become more healthy, well nourished and educated, they contribute more to economic growth. The higher level of HD due to improved education, nutrition and health of people affects economic growth by enhancing their capabilities, efficiency and productivity. Evidence at the micro and macro levels suggests significant influence of HD on EG. At the micro level, numerous studies indicate that increases in earnings are associated with additional years of education, with the rate of return varying with the level of education [see, for example, Behrman and Deolalikar 1988; Schultz 1988, 1993; Strauss and Thomas 1995]. In agriculture, evidence shows that education has positive effects on productivity

Table 4: Effects of EG on HD

Estimated Equation: (3)

Independent Dependent Variable
Variable ln(HDI)t ln(LR)t ln(ELB)t
Constant -5.618 -0.867 2.352
(-8.596) (-0.971) (7.385)
In (PCI)t–5 0.376* 0.331* 0.141*
(6.993) (4.512) (5.398)
In (SSE)t 0.509* 0.662* 0.194*
(3.824) (3.636) (3.003)
D1 0.088** 0.034 0.019
(1.755) (0.501) (0.777)
D2 0.106** 0.143** -0.022
– R 2 (1.833) 0.734 (1.799) 0.606 (-0.786) 0.506
F-statistics 31.342 17.956 12.277

Notes: Figures in parenthesis are t-statistics. * and ** respectively indicate significant at 1 per cent and 5 per cent level. N = 45.

Table 5: Effects of HD on EG

Dependent Variable: ln (PCI)t+3. Estimated Equation: (4)

Independent Variable Eqn 4a Eqn 4b Eqn 4c

Constant 8.964 4.134 -3.851

(32.22) (4.64) (-1.61) ln(HDI)t 1.346*

(5.80) – – ln(LR)t – 0.864* –

(3.69) ln(ELB)t – – 2.785*

(4.72) D1 -0.047 0.103 0.061

(-0.405) (0.815) (0.508) D2 0.031 0.184 0.316*

(0.232) (1.202) (2.62)

R2 0.569 0.411 0.492 F-statistics 20.41 11.27 15.22

Notes: Figures in parenthesis are t-statistics. * indicates significant at 1 per cent level. N = 45.

Table 6: Classification of States’ Performance

State 1981 1991 2001

AP Vicious Vicious Vicious Assam Vicious Vicious Vicious Bihar Vicious Vicious Vicious Gujarat Virtuous Virtuous Virtuous Haryana Virtuous Virtuous Virtuous Karnataka Lopsided-HD Lopsided-HD/Virtuous Virtuous Kerala Lopsided-HD Lopsided-HD Lopsided-HD MP Vicious Vicious Vicious Maharashtra Virtuous Virtuous Virtuous Orissa Vicious Vicious Vicious Punjab Virtuous Virtuous Virtuous Rajasthan Vicious Vicious Vicious Tamil Nadu Lopsided-HD Virtuous Virtuous UP Vicious Vicious Vicious WB Lopsided-EG/Virtuous Virtuous Lopsided

EG/Virtuous

and efficiency among farmers using modern technologies [Chawdhuri 1979; Kalirajan and Shand 1985; Rosenzweig 1995; Foster and Rosenzweig 1995]. Moreover, improved health and nutrition have direct effects on labour productivity, especially among poorer households [Strauss 1986; Deolalikar 1988; Behrman 1993, 1996].

At the macro level, the “new growth theories” assert that higher level of education of the workforce leads to higher overall productivity of capital because of its positive effects on innovation [Lucas 1988; Romer 1990]. A number of empirical studies [Barro 1991; Barro and Lee 1993] have shown the positive effects of education on EG. In the Indian context, a few empirical studies examine the influence of human capital base on the growth of output and total factor productivity. For example, Trivedi (2002) observes that the stock of educational capital, represented by the secondary school enrolment rate, has a significant positive impact on the steady-state level of per capita income. Dholakia (2003) obtains significant effects of HD indicators on economic development measured by per capita income.

Empirical Evidence

Against this background, we have examined the two-way causality between economic growth and human development using the data for a panel of 15 major Indian states. We have measured economic growth by average per capita SDP (PCI) and human development by HDI and two other indicators, viz, LR (representing educational human capital) and ELB (representing non-educational human capital). In the absence of continuous time series data on HD indicators, we have examined the twoway relationship in a cross-sectional setting, by testing causality using appropriate leads and lags in the dependent and independent variables, respectively. To study the causality running from EG to HD, we have examined the effects of average PCI over the preceding five years (t-5) on HD indicators in a year (t), specifying the relationship as: ln( HD) =α+βln( PCI ) +θln( SSE) +δ D +δ D +ε (3)

tt−5 t 11 22 t

And, to study the reverse causality, we have examined the effects of HD in a year (t) on the triennium average PCI of succeeding three years (t+3). The relationship is specified as:

(4) where HD is human development indicator. is average per capita income (per capita real SDP) over the five years preceding the period t; is average per capita income over the succeeding three years; ln is natural logarithm; SSE is the ratio of social sector expenditure to total government expenditure; D1 is a dummy variable for 1991 (D1 = 1 for 1991, but 0 for 1981 and 2001; D2 is a dummy variable for 2001 (D2 = 1 for 2001, but 0 for 1981 and 1991). The dummy variables are included in the equations to see if there has been a change in the structure of the relationships in 1991 and 2001.

The equations were estimated for three HD indicators (HDI, LR and ELB) by the OLS method with the pooled state-wise data corresponding to three time points: 1981, 1991 and 2001. The estimated results of equation (3) reported in Table 4 clearly show that economic growth measured by average PCI has significantly positive effect on all the HD indicators, the coefficient on PCI being positive and statistically significant in all the cases. It can also be seen that the social sector expenditure has contributed significantly to improvement in the HD variables. The coefficients of the dummy variables are found to be positive and significant for HDI, but not for ELB. This indicates that while the structure of the relationship for ELB has been stable, the same for HDI has undergone significant change over time. There has also been a change in the structure of the relationship for LR in 2001.

The estimated results of equation (4) measuring the influence of HD on EG are reported in Table 5. It can be seen that all the indicators of HD have significantly positive effects on EG, the coefficients on these variables being positive and statistically significant in all the cases. Moreover, the structure of the relationship for HDI and LR has been stable, although there has been a change in the case of ELB in 2001. On the whole, the results of our exercise demonstrate the presence of two-way causality between EG and HD in India.

V Classification of States’ Performance

The existence of two-way causality, which is strongly supported by both the conceptual framework and the empirical evidence, gives rise to the possibility of virtuous or vicious cycles of development, with good or bad performance on HD and EG reinforcing each other over time. There are also possibilities of lopsided development, with good performance in one dimension but not the other. The states may be on a mutually reinforcing upward spiral, with high levels of HD leading to high EG, and high EG in turn further promoting HD (virtuous cycle). On the other hand, weak HD may result in low EG and consequently poor progress towards HD improvement (vicious cycle). It may also be possible that high performance on HD may be achieved without good performance on EG (lopsided-HD) and good performance on EG may not be translated into good performance on HD (lopsided-EG). Thus, on the basis of actual performance on EG and HD, the states may be classified into four categories, viz, vicious and virtuous cycles, lopsided-EG (lopsided with strong EG but weak HD) and lopsided-HD (lopsided with strong HD but weak EG).

One way of classifying the states into four categories is to make a comparison of their performance on HD and EG with the average performance of the economy as a whole. To this end, we have presented the state-wise data on HDI for each year against the triennium average value of per capita SDP centring each year of HDI. Thus, the state-wise HDI data for 1981, 1991 and 2001 are plotted respectively against the average value of per capita SDP during 1980-83, 1990-93 and 2000-03 (see Figures 1A, 1B and 1C). The vertical and horizontal lines dividing the figures into four quadrants represent the average performance of the country as whole. While the north-east quadrant represents the virtuous cycle, and the south-west the vicious one, the southeast and the north-west, respectively, represent the lopsided-EG and lopsided-HD categories. The movements of the states across categories can be learned by observing their movements across quadrants over time. The classification of the states based on their actual achievements in HD and EG is reported in Table 6.

It can be seen that about 50 per cent of the states (7 out of 15), which were in the vicious cycle category in 1981, remained in that category throughout. These states include the BIMARU states (Bihar, MP, Rajasthan and UP), AP, Orissa and Assam. The states started with very low levels of HDI which possibly acted as constraint on growth potential, and their low per capita

Economic and Political Weekly July 29, 2006

HDI (1991) HDI (1981)

Figure 1A: Classification of Performance of States (1981)

KER

PUN GUJ

HAR MAH

KAR TN

WB AS ORI AP

BI RAJ MPUP

Average Per Capita SDP (1980-83)

Figure 1B: Classification of Performance of States (1991)

0.7

0.6

0.5

0.4

0.3

AP KER TN WBKAR BI ORI UP MP AS RAJ

1000 1500 2000 2500 3000 3500 4000

Average Per Capita SDP (1990-93)

Figure 1C: Classification of Performance of States (2001)

0.6

0.5

0.4

0.3

0.2

SDP prevented them from generating necessary resources for

500 1000 1500 2000 2500 3000

achieving improvements in HD. None of these states could manage to get out of the vicious cycle of development during 1981-2001.

Three states (Kerala, Karnataka and Tamil Nadu) were in the lopsided-HD category in 1981. Tamil Nadu succeeded to move to virtuous cycle category in 1991, and remained in that category till 2001. Karnataka moved to the borderline between lopsided-HD and virtuous cycle category in 1991, and then moved marginally to virtuous cycle category in 2001. Early progress in HD seems to have helped them to take advantage of policy reforms for attaining high growth, which eventually assisted the movement towards virtuous cycle category. However, Kerala remained in the lopsided-HD category throughout. The state’s high achievement in HD without corresponding achievement in EG, by giving priority to public investment in social sectors has impressed many. Very recently, however, doubts have been raised about the sustainability of this achievement. Slow growth of the economy and deteriorating financial position of the state may not allow the government to maintain high level of social sector expenditure in the long run [Jeromi 2003].

0.30.40.50.60.71000150020002500300035004000450050005500Average Per Capita SDP (2000-03) KAR KER TN HAR PUN MAH WB GUJ BI UP AS ORI RAJ MP AP

Only four states (Gujarat, Maharashtra, Haryana and Punjab) were in the virtuous cycle category in 1981, and they were able to retain this position throughout. These states experienced relatively high growth rate of per capita SDP throughout the period. It seems that they were on a mutually reinforcing upward spiral, with high HD leading to high EG, and high EG in turn promoted further improvement in HD. West Bengal was on the borderline between lopsided-EG and virtuous cycle category in 1981, and moved to virtuous cycle category in 1991, but fell back to the borderline again in 2001. No state followed the lopsided-EG path, and no one could manage to move directly from the vicious to virtuous cycle category. It is also important to note that HD-lopsidedness was a temporary phenomenon for most of the states which followed this path. Two of the three states which followed this path succeeded to move to virtuous cycle category. This suggests that since the movement directly from the vicious to virtuous cycle category is virtually impossible, sequencing of policy should be such that HD is strengthened before a virtuous cycle can be attained. For lifting the states from the vicious to virtuous cycle via HD-lopsided category, HD-induced growth process has to be strengthened by allocating more resources to social sectors like education, health services, drinking water, etc. The HD-improving policies should be given priority in any economic reforms, as improvement in HD is likely to result in high economic growth, which would push the states from the vicious to virtuous cycle category eventually. On the other hand, economic growth may not be sustained if it is not accompanied or preceded by improvement in HD.

VI Summary and Conclusion

We have evaluated the relative performance of 15 major Indian states on human development during 1981-2001. We have addressed the issue of convergence and examined to what extent measures of human well-being (viz, HDI, LR and ELB) alternative to real per capita income are converging across the states. Estimating the cross-sectional growth regression, we have found strong evidence of regional convergence in all the measures of human well-being despite considerable divergence in per capita income. It might be possible that the poor states that failed to catch up with the rich ones in terms of per capita income have managed to catch up in terms of the indicators of HD. The social sector expenditure appears to have been an important factor in achieving regional convergence in HD through its positive effects on LR, ELB and HDI. We have also obtained ample evidence in favour of the two-way causality between EG and HD giving rise to the possibilities of virtuous cycle, vicious cycles, lopsided-EG and lopsided-HD categories of development. The classification of the states based on their actual performance on HD and EG reveals that while only four states have been in the virtuous cycle category, as many as seven states have been in the vicious cycle category during the whole period. Three states followed lopsided-HD growth path, two of them succeeded to move to virtuous cycle category. No state followed lopsided-EG path.

These results have important implications for policy. Based on the findings, it may be argued that since the movement directly from vicious to virtuous cycle category is far from reality, the sequencing of policy should be such that HD is strengthened before a virtuous cycle can be attained. The HD-induced growth

Economic and Political WeeklyJuly 29, 20063328

process has to be strengthened by allocating more resources to social sectors like education, health services, sanitation, drinking water, etc, with the ultimate objective of lifting the states from vicious to virtuous cycle category. Dholakia (2003) argues that the government need not be unduly concerned about regional disparity either in economic or in human development, and the central institutions “should pursue the national priority of achieving high economic growth. The other concerns are most likely to be addressed thereby”. Contrary to Dholakia (2003), our results, confirming the existence of two-way nexus between EG and HD and the superiority of lopsided-HD path in attaining virtuous cycle of development process, suggest that HDimproving programmes should be given priority in any economic reforms for achieving sustainable economic and human development. The states need not wait until they attain high level of economic growth before undertaking large investment for expansion of education and health services. Improved human development is most likely to ensure high economic growth, which would eventually move the states from the vicious to virtuous cycle of development.

EPW

Email: ghosh_m55@yahoo.co.in

References

Ahluwalia, M S (1978): ‘Rural Poverty and Agricultural Performance in India’, Journal of Development Studies, Vol 14, No 3, 298-323.

  • (1986): ‘Rural Poverty and Agricultural Production and Prices: A Reexamination’ in J W Mellor and G M Desai (eds), Agricultural Change and Rural Poverty: Variations on a Theme by Dharm Narain, Oxford University Press, Delhi.
  • Bardhan, P K (1984): Land, Labour and Rural Poverty, Oxford University Press, Delhi. Barro, R (1991): ‘Economic Growth in a Cross-section of Countries’, Quarterly Journal of Economics, Vol 106. Barro, R J and J W Lee (1993): ‘International Comparison of Educational Attainment’, Journal of Monetary Economics, Vol 32. Barro, R J and X Sala-i-Martin (1992): ‘Convergence’, Journal of Political Economy, Vol 100, No 2, 223-51.
  • (1995): Economic Growth, McGraw-Hill, New York.
  • Behrman, J R (1993): ‘The Economic Rationale for Investing in Nutrition in Developing Countries’, World Development, Vol 21.
  • (1996): ‘Impact of Health and Nutrition on Education’, World Bank Research Observer, Vol 11.
  • Behrman, J R and A B Deolalikar (1987): ‘Will Developing Country Nutrition Improve with Income? A Case Study for Rural South India’, Journal of Political Economy, Vol 95.
  • (1988): ‘Health and Nutrition’ in H B Chenery and T N Srinivasan (eds), Handbook of Development Economics, Vol 1, North-Holland, Amsterdam. Chandhok, H L and the Policy Group (1990): India Database: The Economy, Vol 1, Living Media India, New Delhi. Chaudhuri, D P (1979): Education, Innovations and Agricultural Development, Croom Helm, London. Dasgupta, P (1993): An Inquiry into Well-being and Destitution, Clarendon Press, Oxford.
  • Deolalikar, A B (1988): ‘Nutrition and Labour Productivity in Agriculture: Estimates for Rural South India’, Review of Economics and Statistics, Vol 70.
  • (1993): ‘Gender Differences in the Returns to Schooling and School Enrolment Rates in Indonesia’, Journal of Human Resources, Vol 28.
  • Dholakia, R H (2003): ‘Regional Disparity in Economic and Human

    Development in India’, Economic and Political Weekly, Vol 38, No 39,

    September 27, 4166-72.

    Foster, A D and M R Rosenzweig (1995): ‘Learning by Doing and Learning from Others: Human Capital and Technical Change in Agriculture’, Journal of Political Economy, Vol 103.

    Geeta Rani, P (1995): ‘Human Development Index in India: A District Profile’, Arthavijnana, Vol XLI, No 1, 9-30.

    Ghosh, M (1996): ‘Agricultural Development and Rural Poverty in India’, Indian Journal of Agricultural Economics, Vol 51, No 3, 374-80.

    Gopalan, C (1985): ‘The Mother and Child in India’, Economic and Political Weekly, Vol 20, No 4, January 26, 9-30.

    GoI (2002): National Human Development Report 2001,Planning Commission, Government of India, New Delhi.

    – (2003): National Accounts Statistics, Central Statistical Organisation, Department of Statistics, Ministry of Planning, Government of India (Download from website: http://www.mospi.nic.in.)

    Jeromi, P D (2003): ‘What Ails Kerala’s Economy: A Sectoral Exploration’, Economic and Political Weekly, Vol 38, No 16, April 19, 1584-1600.

    Kalirajan, K P and R T Shand (1985): ‘Types of Education and Agricultural Productivity: A Quantitative Analysis of Tamil Nadu Rice Farming’, Journal of Development Studies, Vol 21, No 2, 232-43.

    Kurian, N J (2000): ‘Widening Regional Disparities in India: Some Indicators’, Economic and Political Weekly, Vol 35, No 7, February 12, 538-50.

    Lucas, R E (1988): ‘On the Mechanics of Economic Development’, Journal of Monetary Economics, Vol 22.

    Morris, D and Michelle, B McAlpin (1982): Measuring the Condition of India’s Poor, Promilla and Co, New Delhi.

    Panikar, P G K (1980): ‘Inter-regional Variation in Calorie Intake’, Economic and Political Weekly, Vol 15, No 41-43, October, 1803-14.

    Ramirez, A, G Ranis and F Stewart (1998): Economic Growth and Human Development, Working Paper No 18, Queen Elizabeth House, Oxford.

    Ranis, G and F Stewart (2000): Strategies for Success in Human Development, Working Paper No 32, Queen Elizabeth House, Oxford.

    – (2001): Growth and Human Development: Comparative Latin American Experience, Discussion Paper No 826, Economic Growth Centre, Yale University, New Haven.

    Romer, P M (1990): ‘Endogenous Technological Change’, Journal of Political Economy, Vol 98, No 5 (Part 2), 71-102.

    Rosenzweig, M R (1995): ‘Why Are There Returns in Schooling?’, American Economic Review, Vol 85, No 2.

    Saith, A (1981), ‘Production, Prices and Poverty in Rural India’, Journal of Development Studies, Vol 17, No 2.

    Sala-i-Martin, X X (1996): ‘The Classical Approach to Convergence Analysis’, The Economic Journal, Vol 106, No 437, 1019-36.

    Schultz, T P (1988): ‘Education Investments and Returns’ in H B Chenery and T N Srinivasan (eds), Handbook of Development Economics, Vol 1, North-Holland, Amsterdam.

  • (1993): ‘Investments in the Schooling and Health of Women and Men: Quantities and Returns’, Journal of Human Resources, Vol 28.
  • Sen, A K (1985): Commodities and Capabilities, North-Holland, Amsterdam.
  • (1987): Standard of Living, Cambridge University Press, New York.
  • Singh, N, L Bhandari, A Chen and A Khare (2003): ‘Regional Inequality in India: A Fresh Look’, Economic and Political Weekly, Vol 38, No 11, March 15, 1069-73.

    Strauss, J (1986): ‘Does Better Nutrition Raise Farm Productivity?’, Journal of Political Economy, Vol 94, No 2, 297-320.

    Stauss, J and D Thomas (1995): ‘Human Resources: Empirical Modelling of Household and Family Decisions’ in J R Behrman and T N Srinivasan (eds), Handbook of Development Economics, Vol 3, North-Holland, Amsterdam.

    Thomas, D, J Stauss and M H Henriques (1990): ‘Child Survival, Height-for-Age and Household Characteristics in Brazil’, Journal of Development Economics, Vol 33.

    Trivedi, K (2002): Educational Human Capital and Levels of Income: Evidence from States in India, 1965-92, Discussion Paper No 97, Department of Economics, Nuffield College, University of Oxford, Oxford.

    UNDP (1990): Human Development Report 1990, Oxford University Press, New York.

    Zaidi, N and Md A Salam (1998): ‘Human Development in India: An Interstate Comparison’, Indian Journal of Economics, Vol LXXVIII, No 311, 447-60.

    Dear Reader,

    To continue reading, become a subscriber.

    Explore our attractive subscription offers.

    Click here

    Back to Top