Samuel Johnson, a profound literary critic and essayist, wrote, “Poverty is a great enemy to human happiness; it certainly destroys liberty, and it makes some virtues impracticable, and others extremely difficult.” In sharp contrast, conventional measures of poverty in terms of income are limited and narrowly focused on scarcity of resources to eke out a bare subsistence. But there is much more to poverty than a bare subsistence, as emphasised by Johnson and others.
Nobel Laurate Amartya Sen pioneered a rich, innovative and broader perspective on well-being, focusing on capabilities and functionings. While capabilities are abilities to do this or that in a free and fair environment, functionings reflect achievements. An ability to live a healthy life, for example, is not necessarily related to affluence as it could result in obesity and vulnerability to non-communicable diseases. Achievements such as being healthy, on the other hand, require a nourishing diet and physical exercise. Professor Sen has, however, resisted aggregation of concepts such as capabilities into an overall measure of well-being as he believes that each capability is important in itself.
The MPI story
Unfortunately, the United Nations Development Programme (UNDP) seized upon capabilities to construct an overall measure of human development with uniform weights of the three components: health, education and standard of living and their sub-indices. Following this methodology, NITI Aayog and the UNDP released recently a National Multidimensional Poverty Index/MPI: A Progress Review 2023, also replicated in the UNDP Report, Making Our Future: New Directions for Human Development in Asia and the Pacific, released on November 7, 2023 . Hence, these reports suffer from the same flaws as the UNDP human development index: aggregation with uniform weighting. But, the MPI story is further distorted, as elaborated on below.
Astonishingly, the MPI 2023 estimates show a near-halving of India’s national MPI value and a decline from 24.85% to 14.96% between 2015-16 and 2019-21. This reduction of 9.89 percentage points implies that about 135.5 million people have exited poverty between 2015-16 and 2019-21. Besides, the intensity of poverty, which measures the average deprivation among the people living in multidimensional poverty, reduced from 47.14% to 44.39%.
But these estimates — especially the rapid reductions in MPI — cannot be taken at face value for various reasons. Indeed, these are misleading and ill-informed. First, the MPI relies upon National Family Health Survey (NFHS) 4 and 5, which are not detailed enough for its estimation. Moreover, NFHS 5 is blocked as its estimate of open defecation contradicted exaggerated official claim of its complete elimination. In fact, an eminent demographer, who led NFHS 5 was suspended. Intriguingly, while the survey was blocked for its alleged unreliability, NITI Aayog and the UNDP had no qualms about using it. Ideally, NFHS 4 and 5 should have been combined with the 75th Round of the NSS on household consumption expenditure. Unfortunately, this was abandoned too, as leaked poverty estimates indicated a rise.
What casts further doubts is the havoc caused by the COVID-19 pandemic in 2020-21. Millions lost their livelihoods, thousands died in reverse migration and from a lack of access to vaccines and medical care. In fact, as a consequence of this epidemic, there was a huge economic shock from which the Indian economy has been struggling to recover. To illustrate, GDP growth has declined from 8% in 2015-16 to 3.78 % in 2019-20 and slumped -6.60 in 2020-21, as also per capita income. Not just bare subsistence turned into a daunting challenge for millions but, equally seriously, public funding for maintenance and expansion of health and education and social safety nets suffered an irreparable blow.
Focus on covariates
Our recent analysis focuses on covariates of the MPI that include per capita state income, its square, share of criminals among State MPs, share of urban population, and health and education expenditure and unobserved state fixed effects (e.g., how progressive a State is). If we compare elasticities of MPI with respect to each covariate (i.e., proportionate change in MPI due to a proportionate change in a covariate such as State per capita income), the largest reduction in MPI is due to higher State per capita income. But since income decreased drastically, MPI spiked. The next in order of importance is urban location. A 1% increase in urban location results in a 0.90% increase in MPI. This is not surprising as rural-urban migration is associated with growth of slums and sub-human living conditions. However, reverse migration during COVID-19 may explain why the effect on MPI is less than proportionate. Both health care and education expenditure are associated with lower MPI — the elasticity of the latter is higher (in absolute value), implying that a 1% increase in the latter reduces MPI more than the same increase in the former. As State-level estimates suggest a decline in educational expenditure, a rise in MPI is likely. Although State-level health expenditure rose to combat COVID-19, it fell far short of what was needed. If the share of Members of Parliament with criminal cases in total State MPs exceeded 20%, the higher was the MPI. This is not surprising as criminal Members of the Legislative Assembly and MPs are notoriously corrupt and siphon-off funds allocated for social safety nets and area development programmes. Indeed, what is alarming is their rising share — 24% of the winners in the Lok Sabha election in 2004 had a criminal background; it rose to 30% in the 2009 general election, 34% in the 2014 election, and 43% in the 2019 election.
If we go by our estimates of MPI, the reduction between 2015 and 2019-21 is considerably lower than the official estimate: 4.7 percentage points compared with 9.89 percentage points. Our selective review of MPI estimates shows that poverty rose in India’s most populous State, Uttar Pradesh, by over seven percentage points. Of the States that went to the elections in November (Chhattisgarh, Madhya Pradesh, Mizoram, Rajasthan and Telangana), we find that the MPI fell in Chhattisgarh (by over six percentage points), in Rajasthan (by two percentage points) and, most strikingly, in Madhya Pradesh (by about eight percentage points).
Comment | A $5 trillion economy, but for whom?
In conclusion, not only does the MPI exaggerate the NDA’s success in fighting deprivation but also perhaps more seriously obfuscates conventional measures of it which may unravel a contradictory story of poverty.
Radhika Aggarwal is a doctoral student at Faculty of Management Studies, University of Delhi. Vani S. Kulkarni is Research Affiliate at the Population Studies Centre, University of Pennsylvania, U.S. Raghav Gaiha is Research Affiliate at the Population Studies Centre, University of Pennsylvania, U.S. The views are personal
