Home > Archives (2006 on) > 2020 > Universal Basic Income in India Too? A Dialogue | Pulin B Nayak and Mark (...)

Mainstream, VOL LVIII No 50, New Delhi, November 28, 2020

Universal Basic Income in India Too? A Dialogue | Pulin B Nayak and Mark Lindley

Saturday 28 November 2020

by Pulin B Nayak and Mark Lindley *


Pulin (Formerly, Professor of Economics, DSE) praises a lecture by Mark (Professor of Eminence, MGMU) on UBI but says India cannot seriously consider implementing UBI, because certain other government tasks — including the availability of basic health facilities to all, effective implementation of the public distribution system, and better public education — need to be taken care of first. He sees Mark’s notion of using steep Pigouvian taxes to fund a partial UBI as interesting — not least from the point of view of ecological considerations — but he is uncertain how substantial its revenue-garnering power would be. Mark responds and they discuss. According to Pulin, the current discussion of UBI in India could prove worthwhile if it leads to a way being found to direct the hand of public policy to eliminating the widespread destitution in India and to sustaining whatever is left of Mother Nature’s bounty to India.

o o

This dialogue follows up on a lecture Mark gave in March 2020. (The text is on the Web at Pulin argues that India is not ready to seriously consider implementing UBI, because certain other governmental tasks — including the availability of basic health facilities to all, effective implementation of the public distribution system (PDS), and better public education — need to be taken care of first. He sees Mark’s notion of using steep “Pigouvian” taxes to fund a partial UBI as a useful idea to explore — and not least from the point view of ecological considerations — but he is not sure how substantial its revenue garnering power would be in relation to the other major taxes. Mark responds and they discuss.

Pulin was Director of Delhi School of Economics and had taught for many years there an advanced level course on public economics. He writes often for EPW, and he has edited a four-volume anthology, published by Routledge, of historically notable essays on the economic development of India.

Mark is an ecological economist. His publications have included a book and an EPW article on J.C. Kumarappa as a Gandhian economist; an article, published by the Gokhale Institute, on modern economic theory as merely a “would-be science”; an article in the centenary issue of the Indian Journal of Economics, on the “Economic Man” premise of neoclassical theory; a book placing Mahama Gandhi’s political work in the history of crowd psychology, and a book, published by the University of Kerala, about money in the 20th and 21st centuries.

This is their second joint publication. The first one was a review, in the journal of the German Society for Critical Philosophy, of Herman Daly’s and Joshua Farley’s textbook, Ecological Economics: Principles and Applications.

Arvind Subramanian, as the Chief Economic Adviser to the Government of India, said in his Economic Survey 2016-17 that the time had come to discuss UBI with a view toward implementing it one day. Pulin feels that the time has indeed come to discuss it, but that the discussion will, alas, show mainly that it would be unfeasible here in India. Mark doesn’t presume to know what India ought to do, but his general attitude toward UBI is more hopeful than Arvind Subramanian’s. (This is due in part to the influence of Jan-Otto Andersson, a Finnish mentor and co-author of Mark’s.)

Here is the dialogue.

For several years it has been a kind of a conundrum for me that certain eminent Indian economists such as Pranab Bardhan and Vijay Joshi have put their intellectual weight behind the notion of UBI for India. They have resided in the USA and the UK for more than 40 years, but they place a good deal of their intellectual focus on India; and Indian economists and media often look to them for advice. Likewise in the case of two younger scholars, Maitreesh Ghatak at the London School of Economics, and Arvind Subramanian of the Peterson Institute for International Economics in Washington DC. Dr Subramanian had a successful stint as the Chief Economic Adviser to the Government of India during 2014-18.

The intellectual origins of the notion of UBI go back for several centuries and may be traced to the writings of some major social and political philosophers in England (Thomas More), France (the Marquis de Condorcet) and the US (Thomas Paine). Among the proponents of the idea in England in the first half of the 20th century were Bertrand Russell and C H Douglas, the originator of the notion of social credit. The Indian scholars mentioned above are making a case for its application to India in the proximate future. I have great regard for the body of work of each of them but am not in agreement with them when they suggest that India should go in for a UBI.

In your lecture, I find some valuable new elements. Best of all is how you apply Keynes’s distinction between “absolute” and “relative” needs (i.e. when one has less than the neighbours have and is therefore disdained) to clarify Hayek’s position and to resolve in quasi-medical terms the difference between destitution and a less drastic degree of poverty. I appreciate also your clear distinction, which many writers fudge, between ‘negative poll tax’ (a flow of income to every adult citizen’s own bank account) and ‘negative income tax’ (i.e. to poor folks only, and that too perchance to the household rather than to the individual, and thus a battered wife who dislikes being battered might have a desperate monetary incentive to put up with it; this latter point is, as you acknowledge, in a text posted by Prof. Kasy at Harvard). I take pleasure in seeing how you link the main findings of ‘happiness economics’ with the basic neoclassical concept of abundance causing the marginal value to decline. And there is your insight — admired by Herman Daly as well — that UBI would raise the price of prostitution and thereby reduce its market supply. Daly has noted that “Reducing degrading occupations is as important as increasing the opportunity for worthy occupations. UBI can do both, but the first is often not recognized.”

I agree with Daly, in his preface to the published version of your lecture, that it ought to be widely read. Since, however, it puts the case for UBI on such a strong theoretical basis, I feel the time is ripe to speak up and to join with our good friend Jean Drèze — among others — in showing how unfeasible it is for India.

If I were to quiz you on the lecture, it would be for your suggesting that UBI might be applied in India even though you think the fiscal cost might amount to 10% of India’s GDP. You are a citizen of the US and are habituated to the human condition prevalent in contemporary times in Europe and North America, where a certain minimal material living standard is automatically regarded as the due of every citizen by way of ensuring each person’s basic human dignity. So I can see why you expect the same in any other country one might like to consider. But here in India, the conditions are so different that you have to reexamine your suppositions. The entire annual tax revenues of the central government amount to barely 10% of GDP. Government expenditure on education is (according to the World Bank) 3.8% of GDP. Defense expenditure is slightly more than 2% of GDP; the central government’s expenditure on the health of the citizenry is merely 1.3% of GDP. And as you know, the government has to spend for other things besides the education of the citizenry and national defense and the health of the citizens; it has to pay its legislators, its police etc., it pays for roads (so that, for instance, food can be brought to the cities), and so on. However, if the central government’s tax revenues remain at 10% of GDP, then for it to spend that share of GDP on supporting UBI would mean saying goodbye to funding anything else whatever with its tax revenues. It would be disastrous.

This argument is daunting. You know more about India and about fiscal matters than I do, and I want to hear more details. By the way, that 10%-of-GDP figure as the supposed fiscal cost of a complete UBI (i.e. enough to provide every citizen — even those who might be doing no work at all — with the wherewithal to live in decent poverty rather than in destitution) wasn’t even a back-of-the-envelope calculation, it was just a guess derived from Subramanian’s reckoning of something else (the fiscal cost of a negative-income-tax programme). And yet I don’t imagine that it exaggerates the likely cost of a complete UBI.

Please do, however, help me explain how certain kinds of unprecedentedly steep GST might be structured to finance or help finance a complete or partial UBI. (Partial UBI would be enough to help shield every citizen to a significant extent from destitution, but the citizen would have to somehow get other money to attain ‘decent poverty’ or perhaps affluence.) One of my notions is that ‘Pigouvian’ GST levies, i.e. on various kinds of activity destroying the ‘natural capital’ bequeathed by the Earth to Humankind — would fluctuate according to the government’s assessment of the current rate of loss of the nation’s ‘natural capital’ (which really ought to be called ‘natural endowment’).

Social-security payments in the US fluctuate according to the government’s current assessment of the cost of living. If fiscal payouts can thus fluctuate automatically in keeping with the government’s assessment of a particular kind of current condition, then I imagine that GST rates could also fluctuate automatically in keeping with an assessment of a particular kind of current condition.

We will find time to do more than one thing in this dialogue.
For clarity let us be alert to the difference between percentages of government budget and percentages of GDP. Let’s stay primarily with percentages of GDP, while recognizing that if, say, half of the GDP amount goes to the government in taxes, then a given percent of GDP will be equal to twice as high a percentage of those governmental revenues, whereas if only, say, a third of GDP is collected in taxes, then a given percent of GDP will be equal to three times as high a percentage of those revenues.

OK. I recall fondly how, when our friend Jean Drèze was initially touting his carefully designed MGNREGA, he said that if it were ever fully implemented in every village, then the fiscal cost might be something like one percent of GDP. People in civil society thought, “One percent — maybe we can afford that.” If he had said “ten percent of the central government’s annual tax revenues”, people would have thought “It’s too much.”

I mentioned a moment ago that in India the central government’s tax revenues amount to barely 10% of the GDP. Here are some broader and more comprehensive data for the fiscal year 2017-18. The data pertain to expenditure and revenue of all the three levels of government, viz., central (or federal), state and local.

- Total Government Expenditure as % of GDP Total Government Revenue as % of GDP
Finland        53.4              52.8
Sweden       48.4              49.2
UK          38.2              36.8
US          35               29
India         26.1              19.7

The percentage of the GDP amount which government is able to get a hold of is well less than half, in India, of what it is in Finland and Sweden. The corresponding conditions in the UK and US are somewhere in between. The nature of the political economy in Finland and Sweden is such that the citizens of those countries are comfortable about their democratically elected governments mobilising half of the money that they, the citizens, have earned, and to spend them in ways that are dictated by their political decision- making processes.

(The data show also, for India and for the US, a big excess of government expenditures vis à vis government revenues.)

Equally important is whether the citizens are earning enough money to be able to afford, in reality, to pay the kind of steep increase in taxation which you have contemplated. The answer is yes for the US, but no for India. As per IMF data for 2019, per-capita annual income average in the US (i.e. the total of everyone’s incomes that year, divided by the number of people) was more than $65,000, whereas the equivalent figure for India (in rupees of course) would fetch less than $2,200 in currency exchange. The analogous figure for China was slightly more than $10,000.

I don’t think India should aspire to the per-capita income level of the US. That would surely be too wasteful of natural resources. The $10,000 level would, I think, be a sensible aspiration. However, in addition to the need for India’s average to rise thus more than fourfold above the equivalent in rupees to $2,200, it would be vital to have a more equitable distribution of income within the nation, since India has one of the most skewed distributions in the world — so much so that many of the poor are destitute. UBI would, if affordable, rectify that. But it’s not affordable. We have to find a feasible way.

As for conditions in India meanwhile, I fear that if we move in the direction of UBI as envisaged according to the prevailing mores in the intellectual circles of the West, the government’s vital duty of providing for education and health — which government in India has recently been soft-pedalling — would be given up for good. And there could be a likewise dire impact on India’s elaborate public distribution system (PDS), across the length and breadth of the country, to give subsidized grain to the poor. The PDS was introduced during the Second World War as a war-time rationing measure. In 1997 the government launched the Targeted Public Distribution System (TPDS) with the focus on the poor. Under the new slogan of ’direct benefit transfer (DBT)’ many state governments have sadly allowed recently the efficacy of their PDS programmes to decline. 

In the past decade a number of pilot studies have examined the relative merits of the DBT and the PDS schemes, and it is far from certain that the former is any way decidedly superior. In a pilot cash-transfer project launched by the Jharkhand government in the Negri block of Ranchi district in 2017, it was revealed that the beneficiaries could not access banks and customer service centres easily to collect their money. With every transfer, the beneficiary had to visit the bank to confirm credit of the DBT amount. This was a difficult process because the bank branches were far away. The study showed that 94% of the households preferred the old system of PDS over the DBT.

Some states, Kerala and Tamilnadu for instance, have always had efficient PDS programmes. The laggard states like Bihar, Jharkhand and Chhatisgarh need to understand and emulate the best practices elsewhere. But this ultimately calls for political will as well as effective bureaucratic implementation.

Once India would be able to take care of education and nutrition and health-care in a socialized manner — with no expansion of top-end luxurious private hospitals, and certainly with no government subsidies for them, but rather with plenty of decent, functional, public hospitals serving the poor as well as the rich — then we could turn our attention to UBI, if funds permit. It seems clear to me that that moment is a long way off, and so I feel that detailed discussion of how UBI might be implemented in India is not really worthwhile in the immediate run.

It would be good, I agree, to have Pigouvian levies nudging Indian citizens and businesses away from destroying our natural heritage of ores, fresh water etc. But, just because having a certain kind of tax would be inherently good doesn’t mean that to have vast heaps of it would be good. One problem is that Pigouvian levies are mainly in the nature of indirect commodity taxes (the consumer doesn’t pay the government directly; the tax is tucked into the price of the commodity), and such taxes are regressive: a poor person gets soaked for a bigger portion of his or her income than a rich one does. Direct taxes, income taxes for example, can be readily designed to be progressive (whereby the more income you have in a given year, the higher a percentage of it you have to hand over to the government). Various rates of indirect GST on a broad variety of goods and services might be designed cleverly to impart some degree of progressivity to them. If a certain kind of good is needed just as much by a poor person as by a rich one, then it could be taxed at a low rate, whereas goods that are consumed relatively more by the rich and the super-rich could attract steep GST rates.

Yes, that’s part of what I have in mind — for instance, very steep GST on airplane fares; quite steep on petrol for automobiles; steeper on a heavy car (when you buy or rent it) than on a light one, but relatively moderate on motorcycles and on their petrol. With regard to an electric appliance or a lamp, the GST could depend on the watts (i.e. the rate at which the product uses up electric current). With regard to products made from different kinds of metal, the GST could depend on (1) how rare, in the crust of Planet Earth, are the ores for those various kinds of metal, on (2) how much fuel has to be burnt up in order to produce from, say, a pound of the ore an ounce of the metal pure enough to make the product from it, and on (3) how harmful to people are the waste-products spewed out into the nation’s land and water and air.

I think you are making some extremely useful suggestions here. But the key in all this would be in careful designing of the tax and then its actual implementation. I think that in the Indian political context, the possibility of taxation along such lines adding up to an amount equivalent to, say, 5% of GDP would be slight. But maybe 1% or 2%? Who knows?

In that case, progressive income tax would be a more feasible (as well as a more direct) way to fund a government scheme for mitigating excessive inequalities of income among the citizens.

In the last few decades, a lot of academic writing on “optimum income taxation” has been produced in the wake of some influential mathematical modelling by James Mirrlees (to whom the Bank of Sweden awarded in 1996 its “Nobel Prize” in economic sciences). The general thrust of this theorizing has been to say that the ideal way to tax income is to have as wide a tax base as possible — i.e. to minimize exemptions and allowable deductions — and to set moderate rates. In the last few decades, most of the mature capitalist countries have lowered drastically the income-tax rates on their richest citizens from more than 90% to somewhere around 45% to 50%. The personal-income-tax schedule in India has a three-rate structure (after an exemption-limit at the bottom, i.e. exempting the poor) with the rates being 10%, 20% and, for the richest people, 30%. This low top rate is one reason why the Indian government’s experience at revenue garnering from income taxes has been remarkably poor. Another reason is that the culture of paying taxes as an honest citizen is, after seven decades of Independence, a relatively alien concept to several sections of the population. Less than 2% of the citizens pay any income tax at all, and it’s believed that a lot of small enterprises, self-employed folks and professionals are effectively preventing the government from knowing how much income they actually have. Government would have to spend more money ferreting out the truth about their incomes and forcing them to pay up than would be added to fiscal income as a result of the effort.

If these problems could be mitigated — if the top income tax rate could be set at, say, 40% or 45% (rather than merely 30%) and if more members of the Great Indian Middle Class could feel motivated to pay their proper shares — then some of the additional revenues might be directed at a partial UBI along with revenues from Pigouvian taxation. And in saying this I don’t mean to imply that income tax and GST must forever be the only kinds of taxation paid by our richest citizens. We ought to have a modest tax — no more, say, than half of one percent — on financial transactions. (Such proposals have been mooted for more than 50 years, but vigorously blocked by the banking ‘industry’.) I also feel that the time has now come to seriously consider a progressive tax on both inheritance and wealth. India earlier had the inheritance tax, or estate duty, but it was withdrawn. The implementation of the wealth tax is not as systematic and sincere as it ought to be. In his most recent work, Capital and Ideology (2020), Thomas Piketty makes a strong case for a progressive form of the latter two taxes. We must recall here that India at present has one of the most unequal distributions of income and wealth, and as such, the principle of vertical equity would strongly support a progressive tax on both income and wealth.

After an extraordinarily comprehensive examination of inequality regimes in various historical periods and various parts of the world, Piketty advocates a ‘Universal Basic Endowment’ for all individuals worldwide, financed by a progressive taxation of wealth whereby fiscal income would amount to 50% of the monetary value of production and the governments would provide not only UBI but also a one-time payment of ‘inheritance’ to every individual at the age of 25. It is an inspiring vision, but I don’t see it happening soon.

In that case the question would be “How much of a partial UBI could be funded?”. (It would still be ‘universal’ in the sense that each individual adult citizen would get it, but it would be ‘partial’ in the sense that a destitute citizen’s income would be brought only part-way up to the level of ‘decent poverty’; he or she would still have to beg, borrow, steal or, let’s hope, labour.) My lecture includes some details about the substantial experimentation, in Madhya Pradesh in 2011-2012, with partial UBI at some 20%-30% of the rate income which could have lifted the lucky individuals entirely out destitution (if in fact they were destitute). By and large they did not waste the money but used it in opportune ways to improve their families’ health, education and enterprise. They worked harder than did their neighbors who had not received the ‘no strings-attached’ monetary and psychological boost. (Some details are available at I doubt that a UBI at only a tenth of the rate needed to boost a destitute citizen up to decent poverty would have such a good psychological effect.

Jean Drèze believes, on the other hand, that for government to pay somehow to each poor household the full difference between its earnings and the poverty-line level of income “would mean that anyone who earns less than the minimum income might as well give up working, since the government is supposed to fill the gap anyway”.

I am of a different opinion about that. I think a lot of people do a lot of useful work because they like doing useful work. Richard Layard in the 2011 edition of his book Happiness: Lessons from a New Science has described a neat experiment (in the USA in 1986) in which a university psychologist “gave puzzles to two groups of students. One group he paid for each correct solution, the other he did not. After time was up, both groups were allowed go on working. The unpaid group did much more extra work — owing to their intrinsic interest in the exercise. But for the group that had been paid, the external motivation [the cash] had reduced the internal motivation that would [presumably] otherwise have existed. They did [only] half as much additional work on the puzzles.” I am impressed by such evidence and I think it is not rare. On the basis, however, of your argument that complete UBI would anyway be fiscally foolhardy here in India, could you nevertheless envisage government raising enough revenue to pay for, say, a 25% partial UBI?

There would be, first of all, the possibility, and I would say the feasibility, of increasing the income-tax rate on our richest citizens.

If the demand for the commodities subjected to Pigouvian levies is high and ‘inelastic’ — that is, if people buy a lot of them and would go on buying just as much as they did before the Pigouvian levies on them were imposed — then, in principle, very high rates of Pigouvian taxes might be levied to raise handsome amounts of tax revenues, and these revenues might in turn be used to fund a partial UBI. (But of course we shouldn’t think in terms of applying this principle so sharply as to usher in a tide of violent opposition. We have trouble already with sand mining.)

In regard, however, to commodities for which demand is ‘elastic’ (i.e., people do buy them but would be willing to get on with a good deal less of them), or if the demand is relatively ‘flat’ (i.e. people would do entirely without them if they became expensive), then the potential for raising large revenues via Pigouvian levies would be limited. It is ultimately an empirical issue.

Yes. It seems to me that in a lot of countries nowadays, two of the social and indeed cultural issues are: (1) To what extent will the citizens allow the government to become more authoritarian in order to deal with the environmental disasters caused historically and presently by capitalist-style affluence? The citizens of mainland China obviously differ in this regard from those of the Western countries. And (2) what kind of governmental authoritarianism will the citizens support? Will they support a fascist type whereby the government stirs up popular hatred against scapegoats, or will a “Hobbesian” authoritarianism prevail — citizens consenting in order to prevent a bellum omnium contra omnes (a war of each against all)?

I like your distinction between those two brands of governmental authoritarianism. Our Gandhian cultural heritage, as well as the sheer moral weight of the Indian Constitution authored by Babasaheb Ambedkar, may together help us steer away from copying the Nazi brand of the 1930s and the current Chinese brand.

As for ‘elastic’ vs ‘inelastic’ commodities, I think electric current is a special case in India. Everyone agrees that the destitute should be able to get more of it than they now do. But meanwhile the fact that the price is subsidized (i.e. the market price is less than the electricity actually costs to deliver) has meant, in effect, that a lot of people waste it by running air-conditioners in an immoderate way (making the room uncomfortably cool instead of tolerably tepid), by leaving the lights and TV on when they’re not needed, etc. So I have had this notion of government replacing the subsidy with a Pigouvian levy but meanwhile providing enough of a partial UBI to enable the destitute and the poorest of the non-destitute to buy enough current to meet their real needs. There would thus be an empirically and fluidly linked pair of issues — how much to take for the levy and how much to give for an adequate partial UBI. The destitute and the poorest of the non-destitute would come out ahead on balance. For people at a somewhat higher level of private incomes, the balance would be even: the UBI amount which the government gives automatically to them would match exactly the total amount of the new Pigouvian levies which it is taking away from them. Citizens at the higher levels of affluence would pay more in Pigouvian levies than they get back via the UBI, but they could mitigate the loss to some extent by using electric current more prudently. In my lecture I speculated that a political party offering such a scheme might manage to get elected by a coalition of poor citizens together with some rich ones who are worried about environmental degradation and about the possibility of destructive social turmoil due to excessive monetary inequality.

I understand your logic. Let me offer three comments. First, the more prudent the affluent citizens become in their uses of electric current as a result of the Pigouvian levies on it, the less revenue would, prima facie, flow to the government via those levies, and so then some other kinds of taxation would have played a larger role in funding your UBI, and maybe this wouldn’t be feasible. Secondly, what you have described is not only quite an elaborate kind of scheme, but also a radically new kind, and I think that repairing an old elaborate scheme which we’re quite familiar with might well be better than scrapping too much of it. None of our discussion has caused me to budge from my view that the greater immediate need now in India is for the government to invest more in public health-care and public education and PDS.

An interesting innovation since 2014 has been the Corporate Social Responsibility (CSR) law. It is applicable to Indian corporations that have a net worth of Rs 500 crore or more, or an annual turnover of Rs 1000 crore or more, or a net profit of Rs 5 crore or more. The corporation is required to set aside, for CSR activities, 2% of its average net profits of the previous three years; it has set up a “CSR Committee” comprised of three or more corporation-directors, one of whom must be from a different corporation only; the money has to be spent on socially desirable activities, which may include initiatives in the areas of education or health; and, the details of the CSR activities have to be included in the corporation’s annual board report. There have been serious — not to say sordid — defects in the implementation of the law. They need to be mended, and I would like to see the funding increased from 2% to 3% of the company’s net profit. However, the money is not a tax collected by the Central Board of Direct Taxes (CBDT) and so could not possibly be, according to the law as written, steered into funding UBI.

An equally interesting triad of innovations has very recently — on July 3rd — been proposed by Chakravarthi Rangarajan and S. Mahendra Dev in The Indian Express. They propose that government “give cash transfers to all women above the age of 20 years, ...expand the number of days provided under MGNREGA and ... have a national employment guarantee scheme in urban areas”. The article — accessible at HYPERLINK  — includes fiscal reckonings and mentions that a psychologically effective way of reducing the cost might be to oblige each woman wishing to get the handouts to sign a statement to the effect that her total income is less than 6,000 rupees per month.

I have learned a lot from this dialogue.

And I as well. Even though I regard UBI as unfeasible in India in the proximate future, I find compelling your idea of the government giving money to destitutes as a method of helping the ones who aren’t senile find their way up to ‘decent poverty’ or even better. Sir William Beveridge in his famous report of 1942 on Social Insurance and Allied Services recommended that government take an active role in fighting “the five giants: want, disease, ignorance, squalour and idleness” in the UK. It was done there and it must be done here in India. If some formulation can be worked out — such as suggested now by Rangarajan and Mahendra Dev — to specially direct the benevolent hand of public policy to eliminating the destitution which is so appallingly widespread here in India, then I would say that the current discussion of UBI in India would have been worthwhile. And all the more so if mitigating extreme monetary inequality can be combined, as you have suggested, with an effort to sustain whatever is left of Mother Nature’s bounty to this ancient land.

(Authors: Pulin B Nayak (Formerly, Professor of Economics, Delhi School of Economics)
and Mark Lindley (Professor of Eminence, Mahatma Gandhi Mission University, Aurangabad)

Notice: The print edition of Mainstream Weekly is now discontinued & only an online edition is appearing. No subscriptions are being accepted