Some Basic Issues Underlying Basic Income

1. Introduction

The notion of welfare support in the form of a basic income has been in the air for a while now. A number of economists, in recent times, have discussed the desirability and feasibility of basic income provisioning in the Indian context. A particularly valuable source of contributions to the subject can be found in the web portal Ideas for India , which features essays by several scholars, including—among others—Abhijit Banerjee (2016), Pranab Bardhan (2016), Maitreesh Ghatak (2016), Vijay Joshi (2016), Karl Moene and Debraj Ray (2016), Debraj Ray (2016), and T. N. Srinivasan (2016). A major study of the prospects for basic income policy in India is a book written by S. Davala, R. Jhabvala, S. K. Mehta and G. Standing (2015).

The basic income idea can be traced back at least to Thomas Paine, the great English-born American revolutionary and champion of the Rights of Man . In that book he had already advanced the case for a number of social welfare measures far before his (and indeed our!) times, including old age pension, unemployment relief, and assistance to the poor in the matter of life-cycle events such as births, marriages and funerals. In the book Agrarian Justice , published in 1795, he proposed the notion of a ‘citizen’s dividend’, to be made available, as a matter of right, to all citizens of the US.

It is remarkable that the idea of a basic income as a citizen’s entitlement should have found favour with thinkers across the ideological spectrum. Thus, in the 1960s, for instance, we find that both conservative economists like Milton Friedman (1962) and relatively radical economists like James Tobin (1966) have espoused the cause of some form of basic income provisioning. In more recent times, one of the most passionate and tireless advocates of the basic income idea has been the Belgian philosopher-economist Philippe Van Parijs (see, for instance, Van Parijs, 2000).

It is remarkable that the idea of a basic income as a citizen’s entitlement found favour across the ideological spectrum.

International scholarship on basic income policies is undertaken systematically by the Basic Income European Network (BIEN), and BIEN’s Newsletters are a signally valuable source of information and opinion on the subject. The preceding quick survey is a guide, for anyone interested in the subject of basic income, to some of the salient literature dealing with it.

The present essay will be an attempt to address a few basic issues underlying the idea of basic income. The essay is meant for those who may not have a professional economist’s understanding of the subject (though it seems to be fair to suggest that professional economists themselves do not have a necessarily coherent and common understanding of it!). In writing this piece, I shall draw considerably on an earlier paper written by my former colleague D Jayaraj and myself, and published in the journal Basic Income Studies (Jayaraj and Subramanian, 2017). The present account, I should add, will be mediated by my own, possibly idiosyncratic, interpretations of and views on the subject, and will be presented largely in the form of a series of Questions and Answers.

2. What is the basic idea behind Basic Income provisioning?

There are different ‘models’ of basic income support available in the literature, but a common minimum understanding underlying these seems to be the notion of citizens’ entitlement to some form of assistance in the form of cash transfers from the state. Two specific ‘models’ that are worth considering are those subsumed in the notions of a Universal Basic Income (UBI) and a Basic Income Guarantee (BIG).

A UBI is a scheme which is committed to providing a fixed income transfer of a given amount to every citizen of a country. The scheme is, first of all, universal : that is, it is not targeted to some specific demographic component of the population, such as only the poor, or the aged, or the physically handicapped. Secondly, it is an unconditional cash transfer, unlike, say, the welfare measure of a midday meal scheme in schools, where assistance is in the form of an in-kind transfer (food) that is conditional on the recipient being an enrolled school student. It is also unlike wages paid on a public works programme, where the transfers, though in the form of cash, are conditional on work done by the recipient; and similarly unlike food-grain transferred at a subsidised price to consumers which, while it may (or may not) be universal in scope, is an in-kind transfer.

A variant of UBI is a BIG scheme, which is committed to ensuring that every citizen is entitled to some basic minimum income. Notice that a BIG does not undertake transfer of a fixed amount of income to each person: it only ensures that every person will achieve command over a certain minimum income. Suppose, for specificity, that this minimum income is pegged at Rs.100. Then, under a BIG scheme, a complete destitute with no income at all will be entitled to a cash transfer of Rs.100; a person already earning an income of Rs.10 will be entitled to a cash transfer of Rs.90;…and persons already earning Rs.100 or more will be entitled to no transfers at all.

A BIG scheme, thus, is in the nature of a ‘top-up’, targeted at those that are currently earning less than the stipulated minimum income (which could be some threshold level of income such as the ‘poverty line’, or some fraction of it, or the minimum wage, or some such cut-off). At the time that this is being written (January 31, 2019), it appears—from such information as is available on the subject from the media—that the basic income scheme which the Congress Party has announced, ahead of the 2019 General Elections, is in the nature of a BIG, not a UBI.

We shall return, at a later stage, to issues relating to ‘targeting versus universalism’ that are thrown up by alternative basic income schemes such as BIG and UBI. Before that, there is a case for enquiring into the need for basic income provisioning in India.

3. Why Basic Income in India Now?

Official statistics on money-metric poverty suggest that we have a nicely declining trend of headcount ratios of poverty, the headcount ratio being simply the proportion of the population with consumption expenditure levels below the official poverty line. There is also a view that growth in per capita income has succeeded in trickling down successfully to the poor, so that growth has been not only instrumental in alleviating poverty but has delivered this outcome without being seriously inequitable. The view of economists such as Montek Ahluwalia (2011), Jagdish Bhagwati (2011), Surjit Bhalla (2011), and T. N. Srinivasan (2017) seems to be that inequality in the distribution of consumption expenditure has not registered alarming increases, at least not in the rural areas of the country which, by virtue of their relative share in aggregate population, might be expected to predominantly influence aggregate (rural-cum-urban) trends in inequality. 

I believe it would be unreasonable to deny a secular declining trend in poverty, but is the decline as dramatic, and the level as low, as it is made out to be by official statistics? I would argue not, for a number of reasons. For one thing, the official poverty line is pitched far too low in relation to any reasonable norm of ‘adequacy’, and this contributes not only to the lowness of the headcount ratio but (from both theoretical and empirical perspectives) to declines that are steeper than would hold for higher poverty lines. (‘Multidimensional’ indices of poverty tend to display very much higher levels of poverty than uni-dimensional [money-metric] indices do, as is discernible in the work of Jayaraj and Subramanian, 2009, among others.)

There are also other radically logical inadequacies in the methodology by which official poverty lines are derived, which have been discussed elsewhere by Utsa Patnaik (2004, 2007), the present author (Subramanian, 2005), and Sanjay Reddy (2007). It would take us far beyond the scope of this essay to deal in any sort of detail with these issues, and the best that can be done here is to refer the interested reader to some of the literature on the subject, as I have just done. Suffice to mention here that plausible variants of the official methodology can yield trends in the headcount ratio that are (a) less steeply declining than official estimates, or (b) are seen to describe alternating up-and-down movements over time, or (c) are actually increasing (as in Utsa Patnaik’s work)!

Insofar as trends in inequality are concerned, the relatively non-alarmist picture thrown up by the Indian data on the distribution of consumption expenditure is largely because of the particular way in which inequality is measured. The standard ‘work-horse’ measure of inequality is the so-called Gini coefficient which is what in the literature is called a relative measure of inequality. The predominant practice in the measurement literature, it may be added, is to employ relative measures such as the Gini coefficient.

A relative measure of inequality is one which remains unchanged when all incomes in a population are changed by the same proportion. Imagine now a distribution x described by the numbers (10,20)—by which is simply meant that we have a two-person society in which the poorer person has an income of Rs.10 and the richer person an income of Rs.20. Suppose now that as a consequence of growth in incomes, the distribution x is transformed into the distribution y described by the numbers (20,40)—by which is simply meant that each person’s income has doubled. Notice now that the ratio of the poorer person’s income to the richer one’s is the same, at 1/2, in both the distributions x and y . In relative terms, inequality in the distribution y is identical to inequality in the distribution x .

But suppose now that we were to take an absolute view of inequality. Notice that the gap in the incomes of the poorer and richer persons is Rs.10 in distribution x , but has risen to Rs.20 in distribution y . An absolute measure of inequality would require the measure to remain unchanged when each person’s income increases (or decreases) by the same amount , rather than by the same proportion . The economist Serge-Christophe Kolm (1976a, b) has pointed out that in the presence of income-growth, relative measures of inequality tend to display a ‘rightist’ bias while absolute measures tend to display a ‘leftist’ bias.

 A basic income intervention is a small and partial policy measure for addressing the twin scourges of poverty and inequality.

There is then a case for centrist or intermediate measures, which are neither relative nor absolute, but display the property that inequality increases with equal proportionate increases in all incomes, and decreases with equal absolute increases in all incomes. It turns out that centrist measures of inequality display a strongly increasing over-time trend of inequality in the distribution of consumption expenditure in India, and an even more explosively increasing trend in the distribution of household wealth (Anand and Thampi, 2016; Jayaraj and Subramanian, 2013, 2018).

The twin scourges of poverty and inequality continue to reside with us, and a basic income intervention is a small and partial policy measure for addressing these phenomena which together—for both intrinsic and instrumental reasons—must be regarded as a moral and political outrage that cannot continue to remain unaddressed. The proposition is elaborated on in the following section.

4. What does the ‘Quintile Income Statistic’ Reveal?

As we have seen above, official Indian statistics on money-metric poverty involving the headcount ratio can be misleading, and are subject to manipulation through the simple expedient of manipulating the ‘poverty line’. An alternative and very simple view of income poverty is available in an indicator due to Kaushik Basu (2001, 2006) which he has called the quintile income statistic (Q). The quintile income statistic takes a view of income poverty in which command over a reasonable level of income is seen as a desirable end in itself. The statistic Q is simply the average income of the income-poorest quintile (20 per cent of the population). It is a direct and unconditional indicator of income-poverty which focuses attention on the absolute lowness (or otherwise) of the poorest segments of a population. It is what a philosopher might call an income-poverty measure simpliciter .

From the periodic surveys of consumption expenditure carried out by the National Sample Survey Office (NSSO), we have information on the distribution of consumption expenditure (though data for the distribution of income in India are not available on a systematic basis). From calculations that have been provided in Jayaraj and Subramanian (2017), it turns out that the average consumption expenditure of the poorest 20 per cent of the population—the quintile income statistic Q—has registered an annual compound rate of growth of just 2.18 per cent in rural India and of 2.03 per cent in urban India over the 28-year period 1983 to 2011-12.

That around 240 million individuals should have had such a low standard of living in 2011-12 is telling.

At the end of these modest growth performances, the quintile statistic Q is just around Rs.3,000 per month in 2011-12 (at current prices) for rural India, and around Rs. 4,200 in urban India for a family of five—considerably lower than even the modest official (Rangarajan Committee ¹ ) rural and urban poverty lines of Rs, 4,860 and Rs.7,035 respectively for a family of five. That around 240 million individuals should have had such a low standard of living in 2011-12 is telling. Furthermore, it turns out that the ratio of the average consumption of the richest 100 million of the rural population to the average consumption of the poorest 100 million has risen from 4.85 to 5.89 between 1983 and 2011-12, while the corresponding figures for the richest and poorest 50 million of the urban population are 3.72 and 6.22.

Briefly, percentile statistics reveal high levels of money-metric poverty and increasing trends for money-metric inequality, confirming the urgent need—as articulated at the end of Section 3—for some sort of shoring-up of the incomes of the poor through a scheme of redistribution from the more to the less affluent sections of the population. Redistribution is inevitable—unless basic income provisioning is to be made available from the unlikely descent of ‘manna from heaven’, or from a re-deployment of resources from certain extant welfare schemes to a basic income programme. Some such re-deployment seems to be at the heart of the basic income scheme outlined in the Government of India’s 2017 Economic Survey , as evident from this remark, which seems to be anxious to placate the well-off sections of the population: ‘It is important that UBI is not framed as a transfer payment from the rich to the poor’ (Government of India, 2017, p.175). Which brings us, somewhat naturally, to the next question.

5. Should Basic Income Provisioning Substitute or Supplement Other Welfare Schemes?

Promising, or ushering in, a basic income scheme with a good deal of fanfare, but largely only as a substitute for other extant welfare programmes, is likely (and rightly so) to be perceived as something of a gimmick. (A most instructive opinion in this regard is the commentary by David Piachaud, 2018, on the chapter on UBI in the Economic Survey of 2017 ² .)

This is not to deny that welfare programmes such as public works schemes, the public distribution system aimed at providing food security, and supplemental nutrition schemes such as midday meals in schools are fraught with corruption, leakage, inefficiency, waste and—where self-selection is not a built-in feature—targeting errors involving both wrongful inclusion and wrongful exclusion. But it is to suggest that when things go wrong, there is a case for trying to rectify them, or for learning from instances where they have not gone so wrong, rather than abandoning or replacing them with other schemes which may not, after all, be much less vulnerable to the same, or different, sorts of problems.

Cash transfers—usually both conditional and targeted—are often seen as markedly superior to in-kind assistance such as access to subsidised commodities in fair-price and ration shops, or supplementary feeding programmes. Basic income support, interpreted as various kinds of cash transfers, is then frequently endorsed not as an add-on to welfare schemes relating to nutrition, health, education and employment, but as a replacement for these. Not surprisingly, this is a cause—as Jean Dreze has put it (Dreze, 2017)—for the fear that ‘… there is a real danger of UBI becoming a Trojan horse for the dismantling of hard-won entitlements of the underprivileged.’

A more sensible view of the matter is that what kind of intervention is desirable is best viewed on a case-by-case basis, one which allows not just for the co-existence, but also mutual reinforcement, of different strategies. This point of view is very well articulated in a number of important treatments of the subject, including those by Chaudhari and Somanathan (2011) and Kotwal, Murugkar and Ramaswami (2011), which, while broadly in favour of cash transfers, also appealingly emphasize the virtues of gradualism, ‘piggy-backing’, and learning-from-doing.

A more explicitly non-conflictual view of basic income strategies vis-à-vis more conventional welfare measures is suggested in the work of Louise Haagh, when she says (Haagh, 2017a, p.3):

‘… there is no principled or indeed practical reason to view basic income as in conflict with more complex welfare systems that – such as in Nordic states – pursue human development more intentionally.’

Or again, in Haagh (2017b, p.1), we find: ‘[basic income] is not a radical alternative, but a natural extension of an established tradition.’ Sudha Narayanan (2011) supplies content and context to this proposition by spelling out those circumstances in which one may be inclined to favour in-kind provisioning, those in which ‘cash-assisted in-kind’ transfers may be appropriate, and those in which conditional cash-transfers are likely to be helpful.

Employing basic income as essentially a substitute for existing and ‘hard-won welfare entitlements’ would thus be a cynical fraud on legitimate expectations of a minimally sensitive state-mediated intervention in alleviating the rigours of inequality and poverty which have been reviewed, however briefly, in Section 4. This is the more so in a time such as the present one—of grave agrarian distress; of severe hardship occasioned to the informal sector in both the rural and the urban economies following on the disastrously ill-judged (or heedless) policy of demonetisation; and of a near-unprecedented level of unemployment (6.1 per cent) which has at last been confirmed—even as I write this—by the Business Standard citing the NSSO’s findings from its 2017-18 survey of employment and unemployment. (The Niti Ayog’s chief has since clarified that the Business Standard was citing only a draft report: one supposes we should wait with bated breath for the final, official report of the survey.)

Assuming that any government which comes to power after the next elections will be serious about introducing a basic income programme over and above existing welfare commitments, how may such a programme be implemented? One possible means to the end, in the form of what is known as a Negative Income-tax Schedule, is outlined in what follows, by resorting to simple illustrative examples designed to highlight some of the complexities of implementation.

6. How would a BIG ‘Negative Income Tax Schedule’ Work?

A negative income tax schedule is a system of transfers—or negative taxes—that are made available to the beneficiaries of an income support system. One such system, as we have noted, is a BIG, in which all citizens are assured of a basic income, which is sought to be achieved by ‘topping up’ the incomes of those who fall below the stipulated basic income level. An example of such a BIG is the sort of support system that is being contemplated by the Congress Party as part of its pre-election promise of welfare provisioning.

The precise details of the proposed programme are as yet unclear, but it appears, on the basis of media reports, that the idea is to specify a threshold level of income (the official poverty line, perhaps?), and to bridge the shortfalls in the incomes of the poor from the poverty line. Effectively, then, we are speaking of a BIG aimed at targeting the poor population. How might such a scheme work? A simple arithmetical example is provided below to explicate some basic features of such a programme.

Imagine that the country has just twenty individuals, indexed 1, 2, …,20, in ascending order of incomes. (In the cause of greater realism, each ‘individual’ may be interpreted as a vintile , a vintile referring to one-twentieth, or 5 per cent, of a population.) Imagine, for simplicity, that the total income of our hypothetical community is Rs.1,000, so that the average per capita income is Rs.50. Suppose, for illustrative purposes, that the poverty line is about two-third of the average income, say Rs.33. How the incomes of the poor will be topped up to meet the poverty line will depend upon the existing distribution of income. Let me, for specificity, assume that the distribution of income in our present example mimics the actual distribution of consumption expenditure, as it obtained in 2011-12, and which can be inferred from the National Sample Survey data on consumption for the year 2011-12.

The income distribution is then represented as the proportion of the total income earned by each individual (interpretable also as the average income of each vintile). Given a total income of Rs.1,000, these individual income shares can be converted into corresponding absolute income levels. (That is, if a given individual’s income share is 2 per cent, her income level will be Rs.20, which is 2 per cent of the total income of Rs.1,000). Table 1 below is constructed as follows.

Column 1 of the Table labels the individuals, in increasing order of their incomes, from 1 to 20. Column 2 indicates the income share of each individual. By applying each income share to the total income, one can derive the income-distribution of the population, as displayed in column 3. Given a poverty line of Rs.33, column 4 provides information on the short-fall of each person’s income from the poverty line: this difference (the poverty line minus the actual present income) will represent the transfer—or ‘negative tax’—which is due to each poor person under the BIG that has been designed to ensure that no person remains poor after the transfers. The transfers to the non-poor will, of course, be zero. Column 4 of Table 1, thus, is the appropriate transfer schedule (interpretable as a negative tax for the poor and zero for the non-poor) which the BIG scheme under consideration will aim to achieve.

Table 1: Transfer Schedule for a BIG designed to Eliminate Poverty

Note: The figures in the Table may display rounding-off errors.

Table 1 indicates that transfers will be available to the poorest eight individuals (that is, to the poorest 40 per cent of the population in our example). The total quantum of transfers—the size of the BIG programme—turns out to be Rs.73.60, or 7.36 per cent of the total (assumed) national income of Rs.1,000. The transfers, as we can see, are progressive: the poorer a poor person is, the larger the transfer she will be entitled to, while the non-poor receive no benefit at all. Before we return to a consideration of what else Table 1 reveals, we shall take a look at what a negative tax schedule would look like under a Universal Basic Income (UBI) scheme.

7. How would a UBI ‘Negative Income Tax Schedule’ Work?

Let us retain all the background assumptions underlying Table 1 but imagine, instead, that the BIG budget of Rs.73.60 considered in the previous section will now be available as a universal equal provisioning of Rs. 3.68 (Rs.73.60 divided among 20 individuals) to each person in our imagined community. In our earlier example, we have just, implicitly, assumed that the transfer budget will be financed without any resort to income taxation. As we shall see later, this assumption does no damage to certain important lessons to be learnt from Table 1, which will be reviewed in Section 8. But for these lessons to be learnt in the context of a UBI scheme, we shall need to assume that the transfer budget will be financed by an income tax. The simplest tax scheme one can think of is a proportional tax scheme. Notice that a transfer budget of Rs.73.60 works out to 7.36 per cent of the total national income of Rs.1000. In Table 2 below, we consider what happens when we uniformly tax each individual to the extent of 7.36 per cent of her income.

In Table 2, columns 1, 2, and 3 are identical to the corresponding columns in Table 1. Column 4 of Table 2 indicates the tax liability for each individual that would arise from taxing 7.36 per cent of her income. Column 5 presents the net effective transfer that would be available to each person if a universal basic income of Rs.3.68—adding up to a total transfer budget of Rs.73.60—were to be made available to each of the 20 persons in our hypothetical society.

Table 2 indicates that net (positive) transfers will be available to the poorest 14 individuals (that is, to the poorest 70 per cent of the population in our example), while the richest six persons (30 per cent of the population) will effectively be taxed on net. The aggregate net benefit to the recipients (given by the sum of the net transfers of the poorest 14 individuals), at about Rs.19.50, is precisely balanced by the aggregate net tax paid by the richest six individuals. The transfer schedule, again, is progressive—though the poorest of the poor, obviously, would benefit substantially more under a BIG scheme than a UBI scheme. 

Table 2: Transfer Schedule for a UBI designed to transfer Rs.3.68 to each person

Note: The figures in the Table may display rounding-off errors.

The main objective of the illustrative examples represented by Tables 1 and 2 is to highlight certain 'information problems' that are a feature of most transfer schemes, whether targeted or universal. To this issue we now turn.

8. What ‘information problems’ do the BIG and UBI Schemes of Tables 1 and 2 reveal?

Unless we are wary, the ethically attractive progressiveness of the transfer schedules featured in Column 4 of Table 1 and Column 5 of Table 2, could blind us to a deep practical problem that lies hidden behind these schedules. The progressiveness of these schedules means that we must deliver different net transfers to different individuals.

However, individuals in an economy do not go around carrying (believable!) labels of what their incomes are. At best, from sample surveys of the NSSO type we may acquire some knowledge of the general shape of an income or consumption distribution. That is to say, we may know that the poorest 5 per cent of the population account for 1.47 per cent of the total income, that the next poorest 5 per cent account for 1.87 per centof the total income,…, and that the richest 5 per cent account for 21.53 per cent of the total income. But we do not know the identity of the persons constituting the poorest 5 per cent or of the persons constituting the next poorest 5per cent, …, or of the persons constituting the richest 5 per cent of the population.

We are in danger of making errors of wrongful inclusion and wrongful exclusion.

One can expect that there will be a tendency for people to misrepresent (typically, understate) their true incomes in order to be eligible for programme benefits. Gathering the information on 'Who's Who’ in the income distribution would typically be prohibitively costly (even if one could avail oneself of such data as are available from mechanisms of ‘indirect targeting’ and the like). We are thus in danger of making errors of wrongful inclusion and wrongful exclusion. This is the problem, in a context where information is private and dispersed, of what is known as ‘ adverse selection ’.

If adverse selection is a problem of ‘hidden information’, then ‘ moral hazard ’ is a problem of ‘hidden action’, with which people in the insurance business are familiar: an agency that insures a car-owner against accidents must reckon with the possibility that the very fact of insurance could make the car-owner more careless or negligent in his driving. In general, moral hazard is the problem of provoking the very contingency which a measure is supposed to prevent.

In the context of the present discussion, an income provisioning scheme such as BIG is supposed to extricate the poor from their poverty. But a poor person who is some distance away from the poverty line may decide not to expend any effort in earning an income if he is assured of a ‘topping-up’ scheme that will guarantee him a basic poverty line level of income. In this reading of the situation, a BIG scheme could simply encourage a poor person to remain poor; or even cause a person who has worked at just escaping poverty to lapse into it since—in the presence of an income guarantee—he will not have to work his way out of his deprived status.

It is possible that the threat of moral hazard is overstated: just achieving a poverty line income is not necessarily the acme of ambition for a poor person. Rather, an income support may provide such a person with just the impetus needed to work hard and to transcend poverty.

An income support may provide a person with just the impetus needed to work hard and to transcend poverty.

But even if the ‘hidden action’ problem is judged to be an exaggerated fear, the ‘hidden information’ problem is a very real one. In the case of a UBI, as Table 2 shows, an income-tax-financed transfer cannot escape the requirement of precise targeting, because individual-specific taxes are variable with income. The problem is even more severe with a BIG, because the gross transfer itself—even if no income taxation is involved—is different for different individuals.

Income transfer schemes are thus vulnerable to the problem of what some basic income theorists have called ‘ cadasterability ’ (J. D. Wispelaere and L. Stirton, 2011). A cadaster is defined in the Free Dictionary as ‘a public register showing the details of ownership and value of land; made for the purpose of taxation.’ In the present context of income transfers, a cadaster could be interpreted as an elaborate economy-wide, individual-specific list of names of persons and their current incomes; and ‘cadasterability’ is the amenability of a population to the making of such a public register. The problem must indeed be regarded as an awesome one to address: but even if it is a monumental task, it is one that can be undertaken by summoning the Adhaar enumeration to one’s aid, instead of employing it for purposes of intrusive surveillance.

Nor does the difficulty stop with the making of such a list, even were it possible to do so. The purpose of the list is to convey the benefits promised by the income scheme to the names catalogued in the list. What are the logistical mechanics of such successful conveyance? This is the problem of what Wispelaere and Stirton (2011) call ‘ conduitability ’, which requires that the policy-maker finds a suitable conduit for transferring benefits to their assigned recipients. The issue, as these authors point out, is not only that mechanisms must be found which ensure that beneficiaries are not prevented from accessing their benefits, but that they are actively enabled to do so. How does one address conduitability in the context of income transfers? If we have a cheque for a beneficiary or a bill for a donor, how do we make sure that the cheque and the bill actually reach their intended addresses, and will thereafter be encashed or paid, as the case may be? Through bank accounts, postal savings accounts, electronic transfers? This immediately raises questions of the infrastructure required for ‘conduitability’: bank branches, especially in rural areas, post offices, digitisation.

The conduitability problem is not amenable to instantaneous resolution. Awareness of the problem, however, should place a policy-making body in a position of being prepared to plan for and provide the requisite infrastructure for delivery over the long haul. Hasty promises of targeted income transfers could therefore be a recipe for swift failure on the part of the state, and eventual disappointment and disillusionment for its citizens.

The problems under review can be mitigated, to an extent, by having a UBI which is not financed by income-taxation, but from some other source that does not entail arduous targeting. In terms of Table 2, the transfer to each individual will be a fixed and equal amount of Rs.3.68, an outcome that does not require targeting. It goes without saying that this outcome will be a good deal less progressive than a differentiated, income-dependent schedule of transfers would be. But that is the price which must be paid for mitigating the risk of adverse selection inherent in a targeted approach of income transfers. Unfortunately, such tradeoffs are unavoidable. Having said this, how may such a UBI for India be funded?

9. Towards a UBI for India: Some Ball-Park Estimates

In pointing to a possible way ahead, there is a case for providing at least some orders of magnitude of budgetary commitments that would be required, and how these might be met. (As ordinary mortals, we have that obligation—unlike other agencies which are in a position to promise an income transfer scheme for farmers costing Rs.75,000 crore after providing income-tax concessions to tax-payers!)

Suppose, for specificity, that we aim at an equal, per-person, universal transfer of Rs.10,000 per annum. This will mean a monthly allowance of Rs.833 per person per month, or of Rs.4,165 per month for every household of five members. This is a non-trivial sum for really poor households, even if it may amount to not much more than a relatively insignificant supplementary income for rich households. This being the case, it is conceivable that individuals at the upper end of the income spectrum—say the richest 20 per cent—may voluntarily renounce the assistance, leaving the remaining 80 per cent to an entitlement of Rs.12,000 per person per annum. We do not, of course, make any such assumption, but only point to the possibility of such an outcome.

Given a 2018 population of around 1.36 billion, the total UBI bill should come to Rs.13,600 billion for 2018 (= Rs.10,000 per person per year x 1.36 billion people)—or, assuming an exchange rate of Rs.68.50 rupees to a dollar—to $194 billion (= 13,600 billion ÷ 68.50). The estimated national income, or Gross Domestic Product (GDP) of India stands at $2948 billion ³ , so that the proposed UBI as a fraction of national income works out to 6.58 per cent (= [194 ÷ 2948] per cent). The numerator of this ratio may have to be revised upward over time for inflation and population growth, while the denominator may also be expected to increase over time for reasons of both population growth and growth in per capita income. I shall make the assumption that the UBI bill as a proportion of GDP will remain constant at 6.58 per cent per year over time. From where can one raise 6.58 per cent of the national income without disturbing existing welfare schemes? We explore a few indicative options in what follows.

(a)The central government offers a number of tax concessions or incentives on income tax to corporates, individuals and associates of individuals, apart from exemptions on customs and excise duties. These are commonly referred to as ‘revenue forgone’, and at least until 2013-14, information on the total annual quantum of revenue foregone was conveniently available in Annexure 15 of the Receipt Budget in each Union Budget of the central government. According to the calculations of Jayaraj and Subramanian (2017), the simple average of the ratios of revenue foregone to GDP over the nine years between 2005-06 and 2013-14 was of the order of 6.45 per cent.

...it should be possible to find funds from elsewhere in the budget—without compromising existing welfare schemes...

I am not able to easily access the figures for ‘revenue foregone’ in subsequent years, but even if these should happen to be zero, and given that the proportion of welfare spending and net capital expenditure in GDP has not displayed any noticeable increase in these years, it appears that it should be possible to find funds from elsewhere in the budget—without compromising existing welfare schemes or spending on capital formation—to account for 6.45 per cent of the national income. This fits in nicely with a required-UBI bill-to-GDP ratio of 6.58% we have derived in the previous paragraph.

(b) According to Barclay’s Hurun Index Rich List⁴ , in a single year from 2017 to 2018, the number of individuals in India with a net worth of Rs.1,000 crore or more rose from 617 to 831. The wealth (net worth) of these 831 richest entities stood at $719 billion (employing an exchange rate of Rs.68.51 to a U.S. dollar). Assuming that this wealth will grow at 10% per annum, it should be possible to cap these wealth holdings at their present level through an annual wealth tax of 10 per cent. That should yield, for 2018, a wealth tax revenue of $71.9 billion, which works out to a revenue-to-GDP ratio, for 2018, of 2.44 per cent (= [71.9/2948] per cent). But why stop with the 831 wealthiest entities? It would, I argue, be both fair and feasible to track the wealth of the richest 3,000 entities in the economy and to tax that wealth in the cause of an eminently reasonable rich-to-poor transfer. What we have, instead, is an abolition of the wealth tax in the Union Budget of 2016-17!

(c) What, next, of the extravagant election promises of the incumbent government in the matter of mopping up black money? We not only do not have any credible official estimate of the size of the unaccounted economy; but additionally the only policy measure unleashed to (purportedly) address this problem—demonetisation—has failed disastrously with respect to its stated purpose and in a number of other ways. That black money is an immense potential source of revenue to the state is clear from a report of the National Institute of Public Finance and Policy (NIPFP), New Delhi, which was submitted to the Finance Ministry in December 2013 and has still not been released in the public domain. According to a press leak reported in The Hindu in 2014 ⁵ , the NIPFP has estimated the black economy—greased substantially by the profits of higher education institutions, real estate transactions and mining incomes—to be around three-quarters the size of the country’s official GDP. Work done by Arun Kumar (2016) suggests that the black economy in 2012 was perhaps of the order of 62 per cent of GDP. Even assuming that 5 per cent of GDP is taxable income that annually evades the income-tax net, and that 30 per cent of this income ought to accrue to the state as tax, we are speaking of an additional annual tax revenue of 1.5 per cent of GDP.

(d) Joseph Stiglitz (2006) has written about the need to reform the international global reserve system, so that poorer economies are not obliged to maintain large reserves of foreign exchange when some part of these reserves could be fruitfully employed for development schemes and the management of climate change. Large forex reserves are necessitated by, among other things, the risk emanating from volatile capital flows. Gurbachan Singh (2019) has recently written about the feasibility of mitigating this risk through the levying of a tax on sudden and large influxes of foreign capital, which could release a part of these reserves for investment, the return from which can be used for welfare spending. Indeed, he suggests that with an appropriate scheme of taxation it should be possible to draw down a half of our foreign exchange reserves of $396.68 billion. These, invested for a return of 7 per cent per annum, should yield some thing like one-half of 1 per cent of GDP, which can be used as a resource for funding a UBI.

10. Summary and Conclusion

This essay suggests that the levels and trends of poverty and inequality in this country are such as to cause these phenomena to be regarded as both morally reprehensible and politically unsustainable. Some reparation, in the form of redistributive policy aimed at funding a Universal Basic Income, must be considered to be an overdue measure of amelioration, justice and entitlement. In working out the details of such an income support scheme, the following cautionary stances are indicated.

• One, due note should be taken of what have been described, earlier in this essay, as the problems of ‘cadasterability’ and ‘conduitability’ which typically preside over any income transfer scheme.
• Two, the provision for income support must be neither minuscule nor extravagant, neither an insignificant alleviation disguised and advertised as a major intervention, nor an irresponsible strain on budgetary viability.
• Third, the state should display some genuine commitment to the notion that a mature society must expect the rich to make a few reasonable sacrifices to remedy the undeserved want of their poor compatriots; and such an attitude should make it possible to raise the resources necessary for a modest Universal Basic Income Scheme, while preserving other extant welfare programmes and improving them: the ambition is a feasible one.
• Fourthly and finally, there is a case for making promises that are modest, honest, feasible, gradualist in their unfolding, and transparent in their execution. We are speaking of a welfare measure which is necessary and difficult, but do-able with patience and sincerity.

The best advice to follow may be that contained in a haiku due to the 17 ^th /18 ^th century Japanese poet, Issa, in which he says:

O snail: Climb Mount Fuji— But slowly, slowly!

[ S. Subramanian  is a former  Indian Council of Social Science Research  (ICSSR) National Fellow and a former professor of the  Madras Institute of Development Studies . Subramanian is an elected fellow of the  Human Development and Capability Association  (HDCA). He has worked extensively on measurement and other aspects of poverty, inequality, and demography, and on topics in collective choice theory, welfare economics, and development economics. In 2015, he was appointed as a member of the Advisory Board of the World Bank's Commission on Global Poverty under the Chairmanship of Sir Anthony Atkinson. He lives and works in Chennai. He can be contacted at  [email protected] ]

Endnotes:

[All URLs were last accessed on February 6, 2019]

1. See Government of India (2014). Return To text.

2. Government of India (2017). Return to Text.

3. See Wikipedia: " The Economy of India "; data based largely on CIA World Fact Book . [http://en.wikipedia.org/wiki/Economy_of_India Return to Text].

4. Hurun Report. 2018. " Mukesh Ambani tops the Barclays Hurun India Rich List 2018 for the 7 year in a row ",  September 25. [http://www.hurun.net/EN/Article/Details?num=C9F27DA1DA1C]. Return to Text.

5. See Puja Mehra (2014). Return to Text.

References:

[All URLs were last accessed on February 6, 2019]

Ahluwalia, M. S. 2011 . “Prospects and Policy Challenges in the Twelfth Plan”, Economic and Political Weekly , Vol. 46, Issue. 21, pp. 88-105.

Anand, I. and Thampi, A. 2016 . “Recent Trends in Wealth Inequality in India”, Economic andPolitical Weekly , Vol. 51, Issue. 50, pp. 59-67.

Kumar, A. 2016 . “Estimation of the Size of the Black Economy in India, 1996-2012”, Economic and Political Weekly , LI (48), pp. 36-42.

Bhagwati, J. 2011 . “Indian Reforms: Yesterday and Today”, In: P. S. Mehta and B. Chatterjee (eds) Growth and Poverty: The Great Debate. Jaipur: Cuts International.

Bhalla, S. S. 2011 . “Inclusion and Growth in India: Some Facts, some Conclusions”, LSE Asia Research Centre Working Paper 39.

Bardhan, P. 2016. " Basic income in a poor country ", Ideas for India , September 26 . [http://www.ideasforindia.in/article.aspx?article_id=1694].

Basu, K. 2001 . "On the Goals of Development". In G.M. Meier and J.E.Stiglitz (eds), Frontiers of Development Economics: The Future in Perspective , Oxford University Press: New York.

Chaudhuri, A. R. and Somanathan, E. 2011 . "Impact of biometric identification-based transfers", Economic and Political Weekly , May 21, Vol. 46, Issue. 21, pp. 77–80.

Davala, S., et al. 2015 . "Basic income: A transformative policy for India", London and New York: Bloomsbury Academic.

Dreze, J. 2017 . " Decoding universal basic income for India ", Ideas for India, January 20 . [http://www.ideasforindia.in/article.aspx?article_id=1763].

Friedman, M. 1962 . "Capitalism and Freedom", University of Chicago Press .

Ghatak, M. 2016 . " Is India ready for a universal basic income scheme? " Ideas for India, September 28. [http://www.ideasforindia.in/article.aspx?article_id=1696].

Government of India. 2014 . “Report of the Expert Group to Review the Methodology for Measurement of Poverty ” Planning Commission, Government of India: New Delhi.

Government of India. 2017 . " Economic survey 2016–17 " , Ministry of Finance, Department of Economic Affairs, Economic Division. [http://indiabudget.gov.in/es2016-17/echapter.pdf].

Haagh, L. 2017a . "Basic income’s radical role. Social Europe".

Haagh, L. 2017b . " Basic income should be seen as a democratic right: We should consider basic income a democratic right rather than a solution to unemployment ". [https://medium.com/rsa-journal/basic-income-should-be-seen-as-a-democratic-right-b249ab6078b0] (First Published in Royal Society of Arts Journal, Issue 4, 2016–17).

Jayaraj, D. and Subramanian, S. 2009 . "A Chakravarty-D’Ambrosio View of Multidimensional Deprivation: Some Estimates for India", Economic and Political Weekly , Vol. 45, Issue. 6, pp. 53-65.

Jayaraj, D. and Subramanian, S. 2013 . "The Evolution of Consumption and Wealth Inequality in India: A Quantitative assessment", Journal of Globalization and Development , Vol. 4, Issue. 2, pp. 253-281.

Jayaraj, D. and Subramanian, S. 2018 . "The Distribution of Household Assets in India: 1991-92 to 2012-13", Indian Journal of Human Development , Vol. 12, Issue. 2, pp. 181-203.

Joshi, V. 2016 . "Universal basic income for India", Ideas for India .

Kolm, S. Ch. 1976a . "Unequal Inequalities I", Journal of Economic Theory , Vol. 12, Issue. 3, pp. 416-454.

Kolm, S. Ch. 1976b . "Unequal Inequalities II", Journal of Economic Theory , Vol. 13, Issue. 1, pp. 82-111.

Mehra, P. 2014 .  " Black Economy Now Amounts to 75% of GDP ", The Hindu , August 4. [https://www.thehindu.com/news/national/black-economy-now-amounts-to-75-of-gdp/article6278286.ece].

Moene, K. O. and Ray, D. 2016 . " The universal basic share and social incentives ", Ideas for India. Website . [http://www.ideasforindia.in/article.aspx?article_id=1698].

Narayanan, S. 2011 . "A case for reframing the cash transfer debate in India", Economic and Political Weekly , XLVI(21): pp. 41–48.

Parijs, P. V. 2000 . " A basic income for all ", Boston Review: A Political and Literary Forum . [http://bostonreview.net/archives/BR25.5/vanparijs.html].

Patnaik, U. 2004 . "The Republic of Hunger", Social Scientist , Vol. 32, Issue. 9-10, pp. 9-35. [Text of a Public Lecture on the occasion of the 50 Birthday of Safdar Hashmi on April 10, 2004, New Delhi.]

Patnaik, U. 2007 . "Neoliberalism and Rural Poverty in India", Economic and PoliticalWeekly , Vol. 48, Issue. 32, pp. 3132-3150.

Piachaud, D. 2018 . "A Reply from the Mahatma", Economic and Political Weekly , Vol. 53, Issue. 4.

Ray, D. 2016 . " The universal basic share ", Ideas for India , September 29. [www.ideasforindia.in/article.aspx?article_id=1697].

Reddy, S. G. 2007 . "The Great Indian Poverty Debate", Development , Vol. 50, Issue. 2, pp. 166-171.

Singh, G. 2019 . " How the Minimum Income Guarantee Programme Can be Financed ", The Wire, January 31. [https://thewire.in/economy/minimum-income-guarantee-programme-financing].

Srinivasan, T. N. 2016 . " Minimum standard of living for all Indians: Perspective ", Ideas for India, September 30 . [http://www.ideasforindia.in/article.aspx?article_id=1699].

Srinivasan, T. N. 2017 . “Planning, Poverty and Political Economy of Reforms: A Tribute to Suresh D. Tendulkar”, in K. L. Krishna, V. Pandit, K. Sundaram and P. Dua (Eds.): Perspectives on Development and Economic Policy in India in Honour of Suresh D. Tendulkar , Springer: New Delhi.

Stiglitz, J. E. 2006 . " Towards a New Global Reserve System ". [https://aric.adb.org/grs/papers/Stiglitz.pdf].

Subramanian, S. 2005 . "Unravelling a Conceptual Muddle: India’s Poverty Statistics in the Light of Basic Demand Theory", Economic and Political Weekly , Vol. 40, Issue. 1, pp. 57-66.

Tobin, J. 1966 . " The case for an income guarantee ", The Public Interest , 4  (Summer 1966): pp. 31–41. [https://www.nationalaffairs.com/public_interest/detail/the-case-for-an-income-guarantee].

Download PDF [558 KB]