26
Dec

Unconditional Cash Transfers – What does research say?

By Vishnu Prasad, IFMR Finance Foundation

As the government plans to transfer 29 benefits (pensions, scholarships, fuel subsidies etc.) through the direct cash transfer scheme, we look at what research tells us about unconditional cash transfers.

1. Unconditional cash transfers could lead to positive outcomes

The South African Old Age Pension scheme is a means tested, unconditional cash transfer scheme for women above 60 and men above 65 years of age. Edmonds (2006) finds that these pension transfers are associated with a decline in hours worked per day of a child living with an elder. A complementary rise in school attendance rates is also seen among these children. In a similar study, Edmonds and Schady (2008) find that children in households that receive Bono de Desarrollo Humano (BDH) payments, an unconditional cash transfer scheme in Ecuador reduce their time spent in economic employment by 78% and unpaid economic activity inside their home by 32%.

Case, Hosegood and Lund (2005) examine the South African means-tested, unconditional child grant that is provided to the child’s parents or primary care giver. The study finds that the average child for whom the grant is obtained have parents who are less educated, less likely to be employed and own fewer assets and luxury items. Using longitudinal data, the study finds that there is a significant, positive association between receipt of the transfer and school enrolment.

2. The outcomes are highly positive when the beneficiary is a woman

Duflo (2003) finds that women receiving pensions in South Africa significantly improves the health and nutritional status of their granddaughters. Duflo’s estimates suggest that the pensions received by women helps their granddaughters to bridge the entire gap in height for age scores with American children1. There was no discernible effect when male beneficiaries received the pension.

In an evaluation of the same scheme, Case and Deaton (1998) find that households headed by a woman beneficiary spend a lot less of their pension on alcohol and tobacco.

Yanez-Pagans(2008) finds that a woman receiving old age pension in Bolivia increases schooling expenditures of the household by 56% to 91% (depending on the cultural background of the recipient). There was no statistically significant increase in schooling expenditure among households with a male beneficiary.

3. Conditional transfers create desired outcomes; denying benefits to households that do not meet them could be detrimental

Schady and Araujo (2006) examine the effect of the BDH program in Ecuador on school enrolment. BDH was initially conceived as a conditional cash transfer and was advertised in the media as a program linked explicitly to school enrolment. The conditions were never monitored in practice and the scheme did not penalise households for not meeting the conditions making it, in effect, an unconditional transfer scheme. However, a quarter of the respondents in the study continued to believe that school enrolment was a program requirement. The study estimates that school enrolments were about four times as large when the households believed the program to be conditional.

Baird, Macintosh and Özler (2011) describe the results of a randomized control trial (RCT) in Malawi that provided cash transfers to households with school-age girls. One treatment arm of the RCT received condition transfers, another arm received unconditional transfers and a control group received no transfers. The results focused on two sets of outcomes- schooling & human capital formation and marriage & fertility. Although, households receiving unconditional cash transfers exhibited a modest improvement in school enrolment, it was just 43% as large as the improvement in the conditional transfer arm. Girls in the conditional arm also scored better at English tests. However, the conditional transfers had little impact on reducing the likelihood of teenage pregnancies or marriages. The unconditional transfer arm was far more effective in delaying child bearing and marriage (by 27% and 44% respectively). Surprisingly, these impacts were almost entirely seen among girls who had dropped out of school after the start of the intervention.

These two studies provide some evidence to show that conditional transfers are better at creating desired outcomes. However, as the Malawi RCT results show, denying benefits to those who fail to meet desired conditions could be detrimental.

4. Cash transfers could reduce leakages

Dutta, Howes and Murgai (2010) find that social pension schemes (old age pension and widow pension) in Karnataka and Rajasthan have much lower levels of leakage than the public distribution system (PDS). The authors argue that one of the possible explanations of this could be low levels of discretion in the cash transfer process. A beneficiary might have to pay a bribe to get her name on the list but once she is on the list there is little scope for diverting funds. In the PDS, a beneficiary has to face the discretionary challenges of signing up and also persuade the shopkeeper to sell her the food.

5. Electronic transfers are cheaper, beneficiaries under-use their accounts

Bold, Porteous and Rotman (2012) examine cash transfer schemes in Brazil, South Africa, Mexico and Colombia2. The study finds that the move from cash to electronic payments via mainstream financial accounts need not be more expensive. Secondly, recipients in all four countries preferred the convenience of electronic payments over previous procedures where cash was distributed at a particular time or place. However, there is no evidence to show that beneficiaries use their accounts for anything more than withdrawal of transfers. The study attributes this to widespread confusion among beneficiaries about the functionality and existence of the account.

6. Replacing food subsidies with cash transfers need not compromise food security

Gangopadhyay, Lensink and Yadav (2012) describe the results of an RCT conducted in Delhi that provided a cash transfer instead of PDS support3. The study focused on the impact of cash transfers on six parameters- food security, change in consumption pattern, indebtedness, savings, sanitation and change in consumption of alcohol. The study finds that there is no significant difference in per capita calorie consumption between households that received transfers and households that received PDS support. There is also evidence to show that households receiving transfers shift consumption from cereals to non-cereals. The study also finds no support for the argument that unconditional transfers will shift consumption to ‘bads’ like alcohol. There was no significant difference in the per capita expenditure on alcohol between households.

References

  1. Bold, Chris, Porteous, David and Rotaman, Sarah, 2012. “Social Cash Transfers and Financial Inclusion: Evidence from four countries,” CGAP Focus Note No. 77
  2. Case, Anne and Angus Deaton, 1998. “Large cash transfers to the elderly in South Africa,” Economic Journal 108(450), 1330-61
  3. Case, Anne, Hosegood, Victoria & Lund , Frances , 2005. “The reach and impact of Child Support Grants: evidence from KwaZulu-Natal,” Development Southern Africa Vol. 22, No. 4.
  4. Duflo, Esther, “Grandmothers and Granddaughters: Old-Age Pensions and Intrahousehold Allocation in South Africa,” World Bank Economic Review, Vol. 17, no. 1, (June 2003): 1-25
  5. Dutta, Puja, Howes, Stephen and Murgai, Rinku, 2010. “Small but Effective: India’s Targeted Unconditional cash transfers,” ASARC Working Paper18
  6. Edmonds, Eric V. & Schady, Norbert, 2008. “Poverty alleviation and child labor,” Policy Research Working Paper Series 4702, The World Bank.
  7. Edmonds, Eric V. 2006. “Child Labor and Schooling Responses to Anticipated Income in South Africa, ” Journal of Development Economics, Vol. 81, pp. 386-414
  8. Gangopadhyay, Shubhashis, Lensink, Robert and Yadav, Bhupesh, “Cash or Food Security through the Public Distribution System? Evidence from a Randomized Controlled Trial in Delhi, India” (October 1, 2012). Available at SSRN: http://ssrn.com/abstract=2186408 or http://dx.doi.org/10.2139/ssrn.2186408
  9. Sarah Baird & Craig McIntosh & Berk Özler, 2011. “Cash or Condition? Evidence from a Cash Transfer Experiment,” The Quarterly Journal of Economics, Oxford University Press, vol. 126(4), pages 1709-1753.
  10. Schady, Nobert and Aruajo, Maria Caridad, 2006. “Cash transfers, conditions, school enrollment, and child work : evidence from a randomized experiment in Ecuador,” World Bank Policy Research Working Paper Series 3930
  11. Yanez-Pagans, Monica, 2008. “Culture and Human Capital Investments: Evidence of an Unconditional Cash Transfer Program in Bolivia,” IZA Discussion Paper No. 3678

  1. About 1.20 standard deviations in height for age z scores
  2. All transfer schemes except the one in South Africa are conditional
  3. Households selected for the program were divided into three groups- 1. Households that received cash transfer and opened a bank account, 2. Households that opened bank accounts but did not receive transfers and 3. Households that neither opened bank accounts nor received transfers

  • @rahulrg

    Thanks for the very informative post, particularly for pointing out the India-specific papers.

    Another excellent survey on cash transfers is here: http://www.givewell.org/international/technical/programs/cash-transfers.

    In the face of all the evidence, I don’t think anyone can reasonably argue against the position that cash transfers will make us better off than where we are today (at-least, in theory). However, in the interest of being fair, perhaps there should be stronger research and debate on some of the more ambiguous effects of cash transfers as well? Not to be cynical, but I worry that if a lot of the literature is strongly positive, one needs to cautiously ask if an entire field (in this case, developmental economics) is producing results that are somewhat self-serving. Academia has its own hype cycles.

    Here are a couple of points:

    – Consumption vs. Investment : Several studies (refer the above link) have shown that the marginal propensity to consume is very high. In other words, the transfer is often seen as a permanent effect (as opposed to a pick-me-up). While one would expect that the very poorest households would show a high propensity to consume, it is not clear that this is optimal utilization of the cash transfer to a wealthier household. Interestingly enough, studies also seem to show that the ROIs on invested capital are very good.

    – Effects on willingness to work: Never been a big believer in the Laffer curve, but the question still needs to be asked and answered rigorously.

    Note that we haven’t even started to talk about the public choice aspects of the actual cash transfer implementation on the ground. Is there any literature on this?

    Thanks again for this post.

    • Vishnu Prasad

      Sir, thank you very much for your comments and the very informative link.
      I agree that some of the more ambiguous effects of the cash transfers will become clearer as the program unfolds. Most of the studies mentioned in the survey are from Africa and South America and the scheme needs to be rigorously tested in the Indian context in order to understand its complete effects (including some of the points you have raised). I haven’t come across any paper that talks about the public choice aspects of cash transfers yet.

  • Informative post. There is always a danger of evaluating such a shift in terms of addressing every problem and then rejecting it. To illustrate:
    i. Cash transfers will reduce leakage. At least govt knows max resources go to beneficiaries.
    ii. through electronic transfers, its creating bank account and financial history. Necessary for financial inclusion – will enable them to access bank finance – a huge challenge IFMR is well aware of.

    The transfer won’t magically create social equality and responsible behavior – any more than exists already. It will, however, gradually inculcate a sense of planning a bit since the money will come in with certainty over time and there is discretion. Which is plenty to achieve.

    • Vishnu Prasad

      Sir, thank you very much for your comments.