Poverty & Inequality

Basic income in a poor country

  • Blog Post Date 26 September, 2016
  • Perspectives
  • Print Page
Author Image

Pranab Bardhan

University of California, Berkeley


Pranab Bardhan, Professor of Economics, University of California, Berkeley argues that even though universal basic income is being considered unaffordable in some developed countries, it may well be feasible and desirable in a poor to medium-income country partly on account of low poverty thresholds and existing social safety nets that are threadbare and costly to administer.

Tweet using #basicincome

The (by now old) idea of recasting the welfare State by giving everyone - rich or poor - an unconditional basic income seems to have begun to catch the public imagination both on the left and the right. The left regards it as a simple and potentially comprehensive antidote to poverty; the right finds in it a way to demolish the complex welfare bureaucracy and yet to meet some social transfer obligations without harming incentives much. It also provides an assurance for the dreaded future when robots may replace many of us in the workplace. Even though on 5 June 2016 Swiss voters rejected the idea in a referendum, Canada, Holland, and Finland are reported to be actively flirting with it.

In the US and the UK, some prominent economists have dismissed the idea as simply unaffordable. In the US, for example, a handout of US$10,000 to everyone - which is less than the existing official poverty threshold for a single person - has been estimated to exhaust almost all of the tax revenue collected by the federal government today. So the fiscal arithmetic does not seem to make sense.

But does it make sense for a poor to medium-income country? It turns out, surprisingly, that it may very well be feasible there, apart from being desirable. This is partly because the poverty threshold is low, and because the existing social safety nets are often threadbare and costly to administer.

Take the case of India. The official poverty line is very low and yet about one-fifth of the population lives below it. Under the current anti-poverty programmes you are eligible for relief if you hold a BPL (below-poverty-line) card. Yet surveys (for example, the recent one by India Human Development Survey (IHDS)) show that about half of the Indian poor do not have the card and about one-third of the non-poor have it. In many programmes targeted at the poor, through a process of political and administrative collusion and connivance, benefits continue to leak out to non-targeted better-off people, while many of the intended beneficiaries are left out. In a country where the large majority of workers are in the informal sector, often self-employed, without benefits and without account-keeping or any income data, means-testing is often impossible. In general, the process of identifying the poor by some official criterion is costly, corrupt, complicated and controversial. An unconditional and universal basic income (UBI) can cut through much of this mess.

Fiscal feasibility

But how is one going to pay for it? Will it involve a large increase in the burden for taxpayers? Not necessarily. Let us show some back-of-the-envelope calculations for India.

Suppose we decide to fix the unconditional basic income at an inflation-indexed Rs. 10,000 at 2014-15 prices (this is about three-quarters of the official poverty line that year) to be paid to each person per year in India's the-then population of 1.25 billion. (At the moment to keep calculations simple, I am ignoring the adult-child difference in needs; besides, once you introduce this difference, an official gets to have power over eligibility of families to the total amounts, as many do not have birth certificates to indicate age.) This comes to about 10%of the Indian GDP (Gross Domestic Product) that year. We know from past estimates (these estimates are now being updated) of economists associated with the National Institute of Public Finance and Policy (NIPFP) in Delhi that every year the Indian government gives out about 14% of GDP in (implicit or explicit) subsidies for the central and state governments together. Roughly about two-thirds of these subsidies are described as “non-merit”, that is mostly going to the better-off sections of the population; so that amounts to about 9% of GDP. This deliberately excludes subsidies for health, education, nutrition, environment, rural and urban development programmes, etc. (so the significant leakages from these programmes to the non-target better-off groups are not included in the above-mentioned 9%). Similarly, all of food subsidies are included in the list of “merit” subsidies, even though it is arguable that subsidies in the form of support prices for agricultural products largely go to richer farmers, just as it is arguable that subsidies to higher education largely go to the better-off sections.

On top of the 9% of GDP in the form of regressive subsidies there is a category of “revenues foregone” in the Central budget (mostly in the form of tax holidays and exemptions) largely for firms or companies - this comes to about 6% of GDP (no one has yet estimated the revenues foregone in the state budgets). There are data and interpretation problems for the tax exemption figures; there are grounds to believe that they are overestimates of the potential revenue source, as they, for example, include necessary exemptions from customs duty for imports used in processing for re-exports. Yet if even half of the 6% is added to the earlier 9% for the regressive subsidies, the total easily exceeds the 10% of GDP required for the Rs. 10,000 per person basic income. So, if the government so wants, it can pay out that proposed basic income to everyone, rich or poor, without any fresh taxation, if only the subsidies and tax exemptions to the better-off are discontinued.

If the government does not have the political courage to discontinue them to the full extent, to that extent one will, of course, need either some more taxes (for example, by carrying out some simple reforms on the administrative process of the absurdly low-value assessments and property tax collections in India where the real estate sector has been booming off and on for many years) or a lower basic income payout. In any case a reasonable minimum basic income is very much within the realm of fiscal feasibility. With the biometric identification programme Aadhaarand the bank account programme Jan Dhansoon reaching the overwhelming majority of the people, the administrative feasibility problem for a universal grant should be smoothened out over time. It should be administered by the central government (with the Finance Commission making appropriate adjustments to the fiscal transfers to the states, if part of its funding is to come from state subsidies).


Let us now comment a bit on its desirability. The basic income scheme, while it can replace some egregiously dysfunctional current welfare policies, should not be thought of as a substitute for some other key welfare policies (like public education and healthcare or pre-school child nutrition programmes or the employment guarantee programme on public works). This is important because even with cash in hand the poor, for example, may not by themselves spend on health or child nutrition to the socially desirable extent.

European social democrats sometimes resist the basic income proposal, even when it is not to replace some existing welfare policies, because any weakening of their current social insurance programmes may undermine the worker solidarity in which they originated, or because they weigh the social value of work beyond the income it generates (whereas basic income is not conditional on any work for society), or they worry about the morale problem in a society divided into diligent workers and indolent grant-receivers.

I think these arguments are less forceful for poor countries. In any case most poor countries do not currently have social insurance for the overwhelming majority of the people who are in the large informal sector.

And as for the dignity and solidarity-enhancing value of work, any feasible basic income for countries like India in the foreseeable future will be so small that it will not significantly replace work. If anything the poor are often overworked in back-breaking oppressive work, and it will be better if they, particularly women, can work a little less. In particular, it can be a great relief for the stark livelihood uncertainties faced daily by the vast numbers of the self-employed and the marginalised casual and migrant workers, and will help them in seeking better jobs. It will also enhance their bargaining power against the traders, middlemen, contractors, creditors and landlords they encounter. For women it can boost their autonomy within the household, and for the self-employed poor like small producers and vendors it can relieve a part of their credit constraint. For the socially stigmatised workers in India, like scavengers and waste-carriers, basic income can provide an escape ladder, and induce society to mechanise as much as possible such unwanted filthy jobs.

A possible economic argument against giving out cash as opposed to goods and services in kind is that in remote places the supply of essential goods may not be there unless publicly provided, or more generally, monopoly suppliers may hike the prices to neutralise the extra income. The counter-argument to this is that even in remote places, over time, demand is likely to generate its own supply, and the goods the poor consume are often labour-intensive goods where the barriers to entry for firms are not large, so monopolies may be easy to break.

A more common argument against basic income in poor countries is that the poor (particularly the men in the household) will blow it up on alcohol and drugs and gambling. One can, of course, try to deposit the basic income for the household in the adult woman's bank account. But more importantly, there is actually not much evidence of misuse of the grant in the accumulating experimental evidence on this. There are now many experiments in different parts of the world on the use of unconditional cash grants, where most of the money is found to be spent on worthwhile goods and services. (Keeping in mind the few able-bodied but feckless people who may misuse the grant, the employment guarantee programme on public works is retained in our list as a fallback option).

Thus even if the currently-discussed basic income proposal turns out to be only a gleam in the eye of utopian socialists and libertarians, the scare stories of its expensiveness in rich countries should not deter trying it out in poor to middle-income countries. The low poverty threshold and the inequality in the current distribution of government subsidies in the latter countries provide an opportunity which makes such a proposal remarkably affordable, apart from its being relatively easy on the scarce administrative resources of such governments.

Political feasibility

Finally, how politically feasible is this proposal for countries like India? It is a long shot but worth trying. Many people who otherwise like the idea throw up their hands and say that the better-off in India will never give up the subsidies they enjoy (which incidentally come to several times our total anti-poverty budgets). My own position is that instead of reconciling ourselves to the massive dole we give every year to the rich, we should think in terms of mobilising public opinion and activate social movements on a platform like UBI which otherwise some people both on the left and the right find acceptable. The way to start is first to convince all the informal workers´ associations and welfare boards in India about the feasibility and desirability of basic income, as they will be the largest beneficiaries. Then these associations - maybe SEWA (Self-Employed Women's Association), which already supports the basic income idea - can take the leadership in this and negotiate with the organised-sector worker unions about how basic income is really an extension of the idea of pension, but for everybody. What do the trade unions gain from this negotiation? One, they get a much larger body of workers to be associated in their demands and bargains. Basic income can be a common bridge between the unionised and the vastly larger number of other workers, a divide which for many years has weakened the labour movement. Second, today one-third to 40% of workers in the organised sector are contract labourers deprived of most benefits. Unions have been demanding benefits for the latter for some time; their struggle will be strengthened if it now becomes part of a much larger movement.

Even if the workers, both formal and informal, get united on this, the strength of the opposition from business, rich farmers and the salaried class should not be underestimated, as in my scheme the latter groups will have to give up some of the subsidies they currently enjoy. Will rich farmers, for example, give up on minimum support prices and subsidies for fertiliser, water and electricity in exchange of each family (say, of five people) getting Rs. 50,000 as basic income? At least they may now have some difficulties in mobilising small farmers for their cause. (One reason I have kept the figure of basic income at a relatively high figure - amounting to 10% of GDP - is that a higher figure may encourage small farmers and other such people to rally behind it instead of the trickle-down subsidies they currently get that are mainly lapped up by the richer people.) No doubt the whole process will involve tough negotiations and give-and-take all through. At least one argument the rich usually give against welfare schemes for the poor that much of it is stolen or wasted will apply much less in the case of basic income.

One should have no illusion about the difficulties in the political-economy process for implementing basic income. But one thing going in its favour is that it attracts support from people in different parts of the political spectrum, which may someday generate a winning coalition.


  1. Aadhaar or Unique Identification number (UID) is a 12-digit individual identification number issued by the Unique Identification Authority of India (UIDAI) on behalf of the Government of India. It captures the biometric identity – ten finger prints, iris and photograph – of every resident, and serves as a proof of identity and address anywhere in India.
  2. Pradhan Mantri Jan dhan Yojana (PMJDY) is the Indian government's flagship financial inclusion scheme. It envisages universal access to banking facilities with at least one basic banking account for every household; financial literacy, access to credit insurance and pension facility.

This is a revised and enlarged version of an article that came out in Project Syndicate on 22 June 2016. The author's first exposition of the basic income idea in the Indian context was published in Economic and Political Weekly, 5 March 2011.

Further Reading

No comments yet
Join the conversation
Captcha Captcha Reload

Comments will be held for moderation. Your contact information will not be made public.

Related content

Sign up to our newsletter