India. Craig Jeffrey

2014: 81–3).

      This, ultimately mixed picture of recent trends, is borne out also in research by the World Bank economists Amber Narayan and Rinku Murgai (2016). They too refer to Tendulkar-based poverty estimates, and report that the pace of poverty reduction accelerated after 2004–05 – and then increased even further in the two years after 2009–10. This accelerated rate has been found surprising, however, given the deceleration of the economy in this period (as discussed in chapter 2), and it has also been pointed out that 2009–10 was a year of particularly severe drought – so we should probably not build too much upon the apparently sharp increase in the rate of poverty reduction after 2010 (see ‘Editorial’ 2013). Prosperity has been rising, Narayan and Murgai find, but the share of the bottom 40 per cent of the population in total consumption has fallen, with the rate of growth in this bottom group lagging behind the rate of growth of consumption for the population as a whole (in line with Himanshu and Sen’s observations regarding agricultural and casual labourers). The authors comment that poverty reduction would have been greater had consumption growth in the lower half of the population matched or exceeded average gains for the population. This finding recalls those of Petia Topolova for the period from 1993–94 to 2004–05, when there was also a shift in the distribution of income against the officially poor (in contrast to what had happened in the 1980s). Topolova concluded that ‘Distribution neutral growth would have generated a poverty decline in rural India that was 22 per cent higher; in urban areas, the decline in poverty would have been 76 per cent higher’ (2008: 8). Increasing inequality does matter, in relation to poverty reduction.

      Even more significantly, Narayan and Murgai also report a lot of churning, with a fairly high rate of downward as well as upward mobility (as Dang and Lanjouw 2015 also show), and they find that though the ‘middle class’, who are neither poor nor vulnerable to poverty, has increased as a share of the population (from 23 to 34 per cent), it is still the case that ‘Those who are in the vulnerable group struggle to rise into the middle class, despite the opportunities afforded by rapid economic growth’ (2016: 7). The point here is pithily put by a former civil servant, now professor of public policy, Anirudh Krishna, who writes:

Official statistics tend to depict a rosy-picture of fast declining poverty [exactly as the studies we have reviewed seem, at first, to show], suggesting a situation in which people are only moving up – and out of poverty. But India has one of the highest rates of poverty creation internationally. Between three and five per cent of the country’s population falls into poverty each year, adding to the large number of people already poor … Vulnerability to shocks is pervasive. Many who have moved out of poverty continue feeling its gravitational attraction (2017: 22).

      Krishna points out that, according to the findings of a fairly large body of micro-studies, large numbers of people experience big fluctuations in their economic circumstances, moving into and out of poverty. Research undertaken in different parts of India has shown between 55 and 88 per cent of all households as having experienced poverty for the entire year or for shorter periods (Krishna 2017: 99). Krishna provides a summary of data from studies of movement into and out of poverty in four states, and from three all-India studies. The most recent of these (for the period 1993–2005) shows 4 per cent more households moved into than out of poverty (2017, table 4.1, p. 102). And data from Krishna’s own studies in four states and two cities showed that more than 80 per cent of households that had succeeded in ‘escaping’ from poverty remained stuck within ‘the zone of poverty’. The pathway out of poverty most commonly followed – migration to the city – has inbuilt limitations because secure, formal jobs are very rarely open to poorly educated, low-skilled migrants from rural areas. The activities in which employment has been growing include, especially, construction, in which there are very few ‘good jobs’. Also, the numbers of those employed in transport, as drivers, have been growing relatively fast, as have the numbers employed in retail sales, in catering, as maids, and in personal services. Very few of these jobs will offer either much security or more than a rather basic income. In line with Krishna’s findings, Narayan and Murgai, and Dang and Lanjouw, find that higher levels of education, urban residence, being engaged in wage work, and belonging to social groups other than Dalit, adivasi or OBC, are positively associated with higher-than-average chances of upward mobility – suggesting that the objectives of ‘inclusive growth’ have been at least somewhat disappointed. We say more about this later.

Krishna makes the very important point that ‘The repeated observation of pervasive poverty creation suggests that something is deeply wrong with a policy that has a lot to say about raising people above the poverty line but contains nothing at all about preventing future poverty’ (2017: 103). His own extensive research into poverty dynamics in India has shown that though no single type of event, or particular household characteristic, is responsible for pushing people into poverty, ill-health has been found everywhere to be significant. He reports, ‘Between 3.5 per cent and 6.6 per cent of households in rural areas, and between 2.5 per cent and 5 per cent of urban households, averaging 5 per cent of the entire population, fall below the poverty line each year on account of poor health and the high cost of qualified medical care’ (2017: 105). So fixing health care, making it more accessible to poor people, and more effective, is the critically important preventive measure that will save people from being pitched into poverty. Yet public expenditure on health in India remains very low, at about 1 per cent of GDP, comparing very unfavourably with comparator countries – not that more funding alone will solve the problems of the health care system. Krishna, again, comments, ‘In Bangladesh, Vietnam and Thailand, countries whose recent growth record is no better than India’s, people spend less, incurring fewer debts, for obtaining medical treatment’ (2017: 115). It is to be hoped that a massive new health insurance scheme – known as Ayushman Bharat – announced by the Finance Minister in the Budget of 2018 will serve to correct India’s very poor record in regard to the provision of health care.

      Given Krishna’s well-reasoned arguments, the picture of health in India that comes from recent official studies is quite depressing. A massive report, entitled India: Health of the Nation’s States, was published by the Government of India in November 2017, accompanied by paper published in the medical journal The Lancet (2017), based on studies of a very large number of disease conditions and injuries and risk factors over a 26-year period (Bhuyan 2017). Health is a state responsibility in India, and though some funding, as well as policy and legislation, is provided by central government, states have wide scope in implementation. The largest proportion of health spending comes from the states’ budgets. Over time, each of the states has developed its own health story, and there are very large inequalities. Nine states – Assam, Bihar, Chhattisgarh, Jharkhand, Madhya Pradesh, Odisha, Rajasthan, Uttarakhand and Uttar Pradesh – have been classified as ‘high focus’, implying that they need special attention. Together they account for disproportionate shares of infant, child and maternal deaths, even while failing to spend the funds that they have allocated for healthcare. In some cases, even when these states have increased spending, outcomes have not improved by as much as might have been expected, because of inadequacies in the existing health infrastructure and personnel (Rao 2017).

The disparities between states in regard to health care, which belie the stated intentions of planners and policy makers to achieve ‘inclusive growth’, are reproduced in comparisons of poverty reduction. Both Himanshu and Sen, and Narayan and Murgai find higher rates of poverty decline in initially richer states, so that there is increasing divergence between all the major states. There is an increased concentration of those who are most vulnerable to poverty in the states that are classified as ‘low income’ (Assam, Bihar, Chhattisgarh, Jharkhand, Madhya Pradesh, Odisha, Rajasthan and Uttar Pradesh). Given their large populations and their poverty rates, UP, Bihar and MP together account now for 44.2 per cent of all India’s poor people (Narayan and Murgai 2016: fig. 11a). Except in the case of Odisha, poorer states (notably Assam, Bihar, Chhattisgarh and Jharkhand) have shown only marginal decline, or even some increase in the incidence of poverty (Himanshu and Sen 2014: 80). And simplistic arguments about the positive relationship between economic growth and poverty reduction are belied by the observation that the states that have