ISSN (Online) - 2349-8846
-A A +A

A Reply from the Mahatma

David Piachaud ( is Professor Emeritus of Social Policy at the London School of Economics.

One year ago in January 2017 the chief economic adviser to the Government of India, Arvind Subramanian, in the Economic Survey, 2016–17 included a chapter entitled “Universal Basic Income: A Conversation With and Within the Mahatma.” While there continues to be significant publicity and debate about Universal Basic Income, there has been no focused reply to Subramanian’s conversation from the perspective that M K Gandhi might have taken. Here is a response as the Mahatma might have replied.

Dear Professor Subramanian,

I was honoured to read in your Economic Survey, 2016–17, a chapter concerning a universal basic income (UBI), which was in the form of a conversation with me. To be engaged in conversation by someone as distinguished as your good self is indeed unusual; the more so because I have not been involved in conversations with the Indian government on subjects such as industrialisation, urbanisation, materialism or nuclear weapons, on which I might have had strong views.

You begin:

This chapter examines UBI in the form of a conversation with the Mahatma, and indeed a conversation that the Mahatma would have had with himself had such a proposal been put to him. (para 9.2)

Yet, in it I see little conversation with me, and my beliefs get little mention. That is why, one year after your chapter was published, I write to question parts of your conversation and its conclusions on what I would have thought.

So, what is the UBI about which you converse with me?

I was, as you know, a lawyer trained at the London Bar. The law requires a certain precision, as did negotiation with the British Raj. Perhaps precision is not required of economists. Unfortunately, much of your conversation with me leads to confusion, not clarity.

You start by stating that

A universal basic income is, like many rights, unconditional and universal: it requires that every person should have a right to a basic income to cover their needs, just by virtue of being citizens. (para 9.3)

In practice, the possibility of a UBI rests on developments in financial inclusion that I had never dreamed of. Many politicians promise “jam tomorrow” but now it is claimed that India can have JAM today. This JAM depends on financial inclusion through Jan Dhan accounts, on complete population registration through the Aadhaar system, and on everyone having access to mobile phones. (How things have changed! When I lived in Wardha I had a special phone box so that I could talk on crackly lines to all parts of India, no doubt with the British listening to my conversations.)

Let me start by considering your case for and against UBI. In favour of UBI you list: poverty and vulnerability reduction, choice, better targeting of the poor, insurance against shocks, improvement of financial inclusion, psychological benefits and administrative efficiency (Table 1,
p 175). You write that UBI would reduce poverty and vulnerability “in one fell swoop.” Let me consider poverty and vulnerability in turn.

Reducing Poverty

How far UBI would reduce poverty depends on the level of the UBI and on how much of total UBI spending would go to the poor.

If a UBI is to be sufficient to cover people’s needs, then the starting point must be a definition of needs and its conversion into a poverty level. Then I would expect the cost of providing this to be estimated and ways of meeting this cost to be set out. I have looked in vain for such a straightforward approach.

Instead of starting with a definition of needs you propose a “Bang-for-Buck UBI” (para 9.46 and Appendix 4). This is the level that would achieve the greatest marginal reduction of poverty relative to its fiscal cost: a very different concept. It is far below the poverty level used by the Government of India. Thus at the outset you have ruled out a full UBI that would meet needs at the poverty level and have confined your discussion to a far lower “add-on” or “top-up” UBI.

You claim as an argument in favour of UBI that it would lead to better targeting of poor; you write:

As all individuals are targeted, exclusion error (poor being left out) is zero though inclusion error (rich gaining access to the scheme) is 60%. (p 176)

My first comment is on your language. To write “as all individuals are targeted” is strange: if all are targeted, then no one is targeted. You mean, I think, that all the poor would be included, in contrast to existing subsidy schemes which fail to reach many of the poor. Can you be so sure that all the poor would actually be included? Are all those living on the streets or in the remotest areas likely to be registered, have bank accounts and mobile phones?

Most seriously you pay little attention to the inclusion error, with 60% of the UBI going to those who are not poor. This proportion is far higher than for most existing subsidy schemes. To criticise existing schemes, as you do, and suggest they might be ended in order to pay for UBI when they are, in effect, better targeted at the poor than a UBI, is puzzling. Clearly, there is the need to ensure that any mechanism to help the poor actually reaches the poor, and existing systems clearly need to be improved, but accepting an inclusion error of 60% is a very expensive and inefficient way of reaching the poor.


As well as reducing poverty, you state that vulnerability would be reduced in one fell swoop. But vulnerability depends more on services and on human capital than on current money income. You state as an argument in favour that:

A UBI treats beneficiaries as agents and entrusts citizens with the responsibility of using welfare spending as they see best; this may not be the case with in-kind transfers. (p 175)

But to suggest that a UBI can deliver adequate universal basic services is to ignore the need for collective provision of many basic services. The very rich may be able to afford private water supplies, education and healthcare but those on average or lower incomes cannot. If the poor are to benefit from economic growth there need to be both effective public services and improved social protection: through subsidies to basic foods and school meals, employment guarantees and pensions. Hoping that a UBI can achieve all that sounds like an impossible dream.

Education, public water supplies, sanitation, and preventive health services are all in-kind transfers designed to enhance human capital and reduce vulnerability. But none of these would be boosted by a UBI. Indeed, if a UBI were financed in part by cutting back these human services and relying more on individual agency, there can be little doubt that it would be the poorest who would suffer. Rolling back the state and forcing individuals to be more self-sufficient is precisely the goal of far-right advocates of a UBI in the United States.

Certainly, UBI would not directly reduce vulnerability since it is not improving human capital—education or health—nor is it doing anything to redistribute land or other physical or financial capital that might widen social opportunities.

The Case against UBI

As a case against UBI you list: conspicuous spending, moral hazard, gender disparity induced by cash, implementation problems, fiscal cost of exiting from UBI, political economy of universality, and exposure to market risks (Table 1, p 175). Perhaps the most important of these is the political economy of a UBI.

When you consider “The Way Forward” you write:

the irresistible force of even as powerful an idea as UBI will run into the immovable object of a resistant, pesky reality … There exist, when translating the idea into reality, tensions that tug in opposing directions. (paras 9.44 and 9.57)

Let me consider how you propose to resolve the tension between the idea and the reality.

One way you suggest is to make a distinction between “de jure universality” and “de facto universality.” What is meant by the latter is the exclusion of the “obviously rich” from UBI since “transferring even some money to the well-off may be difficult” (para 9.59). One possible method is excluding the “non-deserving” on the basis of ownership of an automobile or air conditioner or a bank balance over a certain size; of course if this were laid down in law it would be “de jure non-universality.” Three other methods are suggested: a “give it up” scheme of voluntary non-receipt of UBI that relied on social pressures; “naming and shaming” the rich who choose to receive UBI; and “self-targeting” whereby the rich would not find it worth their time to claim UBI. Such methods might reduce the UBI going to the rich but this would depend on their decisions, not on any systematic rules. How can this be justice under the law? Your notion of “de facto universality” is not convincing.

One route to limit the cost that is discussed would be “to phase in a UBI for certain vulnerable groups—widows, pregnant mothers, the old and the infirm— first” (para 9.67). Since poverty is more prevalent in these groups than among the population generally, benefits focused on these groups would be more effective in tackling poverty. But benefits for these groups, which exist in many countries and for some limited groups in India, are “categorical” or “contingency-related”—they are not a Universal Basic Income and it only causes confusion to call them that.

You also suggest that a UBI might only go to women (para 9.66). This too might be more effective in tackling poverty than a genuinely Universal Basic Income, but nothing could disguise the fact that it would be a “Women’s Benefit,” not a real UBI. It would certainly be a social revolution to transfer a substantial proportion of income from men to women. But you do not discuss whether India is now ready and willing to take such a measure to redress centuries of female repression, nor how it might be made ready.

The final possibility you discuss is to offer a UBI in place of existing entitlements to other subsidies: for food, for kerosene, for mid-day meals, for the Mahatma Gandhi National Rural Employment Guarantee Scheme (MGNREGS) (which bears my name, and tempts me to pride although I strive to resist temptations), and many other subsidies. You show in your conversation, in considerable detail, that many of the subsidies intended for the poorest do not reach them because of administrative inefficiency, corruption or other reasons. If a UBI could be sure to reach them then this might indeed be administratively less costly and more effective. Yet, two questions arise. First, are the subsidies good in themselves? Second, how certain is it that a UBI would reach everyone?

Subsidies like that for mid-day meals are of benefit not only directly to children who receive the meals but they may also improve children’s educational outcomes—a hungry child learns less—and long-term income prospects, as well as reducing child labour and the bondage that afflicts many children. The employment guarantee system operates when there is agricultural drought or depression and meets, with appropriate conditions or tests of need, the specific and devastating problems that confront rural employment.

So what do you do when the “irresistible force” of the idea of UBI meets “pesky reality?” You modify the idea of UBI and, in effect, consider something that is not what you define as a Universal Basic Income or suggest ending some other subsidies which have clear and strong rationales. I do not think that is what I would have done. If the universality intrinsic to Universal Basic Income is desirable then surely it should not be undermined. If universality is not a fair and good principle then the case for a genuinely UBI does not stand up.

You discuss another aspect of vulnerability in relation to social evils such as alcohol and tobacco. I was, as you know, concerned about “temptation goods” and you discuss whether additional income from a UBI might be spent on wasteful activities. Your evidence shows that “social evils” form a smaller share of consumption as overall consumption increases. But where there was an individual problem, the remedy I advocated was renunciation, not limiting the incomes of the poorest. If anyone should have their income restricted to limit temptation and vice, it would surely be the very rich: I have heard that some super rich can fly from their palaces in helicopters and even have a choice of a helipad.

Other dangers are inefficiency and corruption. You wrote: “the success of the UBI hinges on the success of JAM” (para 9.77). Can we really assume that all the very poorest will be financially included? You also warn that:

Given the amount of cash that will flow through the system under the UBI and the fungible nature of money, one can imagine a perverse equilibrium where the UBI results in greater capture by corrupt actors. (para 9.77)

Corruption would not end simply because UBI is in principle a simple scheme: the larger the sums of money involved the larger the temptation to maladministration and corruption.

Social Values

One thing you do not discuss at all are the social values and ideals that should guide a just society. Here Hindu teaching has much to contribute. The ideal life consists of four asramas or stages: first, brahmacharya, the period of education; second, garhasthya, the life of the active worker; third, vanaprastha, retreat from active life; and, last, sannyasa, the life of the hermit withdrawn from the world. Different stages have different needs and different responsibilities. Yet, UBI makes no such distinctions and treats all ages and conditions the same. This differs from the social security systems that have developed in most countries which have, for example, benefits for the elderly and disabled people which are not available to all, and impose conditions on eligibility to decide for example who is unemployed and needing social security support, just as MGNREGS has rules and conditions. As K M Sen put it in Hinduism (Penguin Books, 1961: 25):

The Hindu ethic … is one of considerable complexity. Depending on the temperament of the person concerned and the stage of his career, his duties may differ. There are of course some universal values, like truthfulness, kindness, and love, which are considered to be everyone’s duty, but man’s more specific pursuits are supposed to be relative to his age and temperament. So renunciation and what is sometimes called “other-worldliness” are not the only Hindu values. Active material service is as much a part of Hindu life as contemplation and spirituality. Even the approach to the Supreme may be either through jnana (knowledge), or through karma (work), or with the help of bhakti (devotion).

The idea of selfless work set out in the Gita, with work treated as a duty and not undertaken for the fruits it may bring, was a guiding principle that I strove to pursue.

Social security systems that take account of work—who is expected to work, who cannot work because of age or infirmity or lack of opportunity, who needs extra income because of inadequate income from work or extra needs—may be much fairer and more in accord with the stages of life than a UBI system that is only about money.

Work is needed for the welfare of the society and many now lack work. Providing jobs is a serious problem that may be becoming more severe. A UBI does not, however, offer any solution to that problem.


It is good to have been engaged in a “conversation,” but I regret you have not based your argument on clear principles.

You start by quoting something I said: “Whenever you are in doubt apply the following test … ask yourself if the step you contemplate … will lead to Swaraj for the hungry and spiritually starving millions.” The American political philosopher John Rawls has written almost exactly the same thing in his Theory of Justice (Belknap 1971). I might have hoped that you would apply this test to UBI and ask whether it is in reality the best way to aid the hungry and relieve poverty.

When asked for advice by many people, when I did not have the time or the knowledge to offer good advice, I sometimes replied “Be True.” In your conversation with me you do not always seem true. You are not clear and consistent on what UBI means.

Is it to provide for the needs of all and be set at the poverty level, as your definition suggests? No, it is set at the level that achieves the biggest “bang for buck” in its marginal effect on poverty. Is it really universal? Sometimes yes, and sometimes no. Is it a supplement to other social security, subsidy and employment systems, or a replacement? Sometimes one, sometimes the other.

There are many choices to be made in finding the best way to reduce poverty, but I do not see them clearly set out and compared.

You conclude your conversation:

One can easily imagine the Mahatma as fair mediator, deliberating and examining both sides of the argument carefully … on balance he may have given the go-ahead to the UBI. Or so one might tentatively infer. (paras 9.79–80)

I count no less than four hesitant qualifications: “on balance,” “may,” “might,” “tentatively.” In my life I often had doubts but I also tried to be decisive. Therefore, I wish that a conversation with me could have led to a more decisive conclusion.

Being true to the interests of the poorest requires social benefits that give priority to the poorest and public services of a high standard that increase opportunities for all. I therefore conclude three things.

First, that there are proven benefits from improving the incomes and opportunities of the poorest in both rich and poor countries. Second, that basic income is only one of a set of possibilities in a spectrum of distributional policies including social security policies, social protection, food distribution schemes, and employment policies. Third, that in relation to poverty, a full UBI for the foreseeable future seems distinctly inferior to other possibilities in terms of its fairness, its economic efficiency, and its political feasibility.

There is, I agree, much that can be learned from the debate on UBI. But I cannot agree that I “may have given the go-ahead to the UBI.”

Your obedient servant,

M K Gandhi.

Updated On : 29th Jan, 2018


(-) Hide

EPW looks forward to your comments. Please note that comments are moderated as per our comments policy. They may take some time to appear. A comment, if suitable, may be selected for publication in the Letters pages of EPW.

Back to Top