What constitutes a good UBI?

The lack of "universality" becomes the biggest problem in any Universal Basic Income scheme since it raises the challenges of effective targeting

The idea of Universal Basic Income (UBI) seems to have made a permanent entry into the Indian policy lexicon. The NDA government took its first step with PM-KISAN and the Congress manifesto promised Nyuntam Aay Yojana (NYAY). Telangana and Odisha have had their own cash transfer programmes for a while and Andhra Pradesh, Sikkim, West Bengal and Jharkhand might have their variants soon.
 
UBI in its textbook avatar rests on three pillars — it is unconditional and universal, and involves a fixed amount of cash transfer. Developed eco­nomies including the US, Canada and Finland have run pilot programmes to understand how well a UBI will work. Their pri­ncipal concern is that of auto­mation creating a jobless society and their versions conform to the textbook model in that all households would receive the same cash transfer.
 
Developing countries, on the other hand, see UBI largely as an anti-poverty tool. Given resource constraints, their programmes are forced to jettison the first critical feature of UBI — that of universality. This includes Brazil’s much-discussed Bolsa Familia, China’s Dibao, and all of the schemes proposed or operational in India. These are partial Universal Basic Income schemes if one is willing to ignore the basic contradiction involved in the phrase.
 
Not surprisingly the lack of “universality” becomes the biggest problem since it raises the challenges of effective targeting. For a targeted UBI, the four key elements that determine its success are captured in the acronym GAM-CAP. These are effective governance (particularly for local administrative bodies), the optimal amount of transfer, the appropriate metric used to identify beneficiaries, and, most critically, the presence of adequate capacities for the supply of goods and services that households are likely to demand with the cash they receive.
 
Targeting becomes difficult in emerging markets (EMs) principally because of a large informal sector. Since hard data on family incomes is difficult to come by, federal or state governments need to rope in local bodies (the equivalent of our panchayats) to identify the poor. International experience shows that governance standards determine how well this works. Both China (Dibao) and Brazil (Bolsa Familia) implemented their basic income schemes through the support of local governments. However, in China, corruption at the municipal level resulted in rampant misuse while in Brazil, the Bolsa Familia programme was a success due to the effective participation of local bodies.
 
In cases where cash transfer is targeted at a specific sector, alternative metrics could be used instead of income thresholds. It is imperative to choose a metric that ensures that the right people get the transfers. India provides examples of how the wrong choice can lead to both inclusion and exclusion errors — that is of transfers reaching those who should not be entitled and excluding the deserving. Telangana’s Rythu Bandhu, which followed a massive exercise to collect and update its land ownership data before the implementation of the scheme, is a good example. By focusing exclusively on land records, it failed to exclude income-tax payers or government employees who held less than the threshold amount of land, resulting in a classic case of inclusion error. Further, by making land records the basis of the benefit transfer, the scheme ended up excluding tenant farmers, who are often the most vulnerable.
 
PM-KISAN also chooses land holding as the metric of identification and, apart from the daunting task of sifting through the land records (often undigitised), it risks the exclusion of the vast population in the farm sector that deserves transfers the most but do not own land.
 

Illustration by Ajay Mohanty

The amount of cash transfer also becomes critical. Give too large a sum and the incentive to cheat increases; give too small an amount and it barely makes a dent in living conditions. Under NYAY, the cash transfer proposed seems too large. At ~72,000 per annum, or roughly 58 per cent of the median household income for the population as a whole, the incentive to game the system is high and could lead to large inclusion errors. Those who believe that the use of BPL cards will do the trick might want to look at a simple case study done on the patients at the All Indian Institute of Medical Sciences (errors of inclusion and exclusion in income-based provisioning of public health care, Bajpai V et al, 2017, NCBI). Of the 374 study subjects, 69 per cent of the poor did not possess a BPL card.
 
The solution is to impose a self-selection filter making the transfer conditional. For instance, payments can be made in lieu of the work done. Thus, MGNREGA clearly scores over NYAY on this count. If, however, for some reason it has to be unconditional, the design of the UBI transfer amount has to ensure that the incentive to game is minimised.
 
The capacity issue (the CAP of our GAM-CAP) asks a simple question. If cash transfers create a demand for goods and services, is there enough capacity to supply them? In health and education, which are textbook instances of market failure, should the government spend its resources creating capacity or augmenting incomes, assuming that it has to make a choice? Is a UBI bound to be more successful in Brazil, which spends about 6 per cent of GDP of public money on education whereas India spends 2.8 per cent, and $607 per capita on public health, compared to $59 per capita in India?
 
Income transfers do not make people lazy. Nor do they encourage spending on sin goods like alcohol. There is a trove of case studies across countries that debunk these myths and show that it can make a significant difference to people’s lives. Bolsa Familia, initiated in 2004, led to a 20 per cent drop in inequality and a 28 per cent drop in poverty in Brazil. In Kenya, a $45 pay-out a month cut the number of days children went without adequate nutrition by 42 per cent and increased livestock holdings by 51 per cent.
 
In our case, there are two critical questions that need answers. Will the government have the courage to phase out subsidies to fund a UBI so that fiscal limits are not busted? Equally importantly, does the design of a cash transfer tick all the GAM-CAP boxes?
 
Barua is chief economist and Gupta is economist, HDFC Bank