Do larger cash transfers improve longer-run outcomes more cost-effectively than smaller cash transfers?

By Florence Kondylis (Senior Economist and Program Manager, DIME), John Loeser (Economist, DIME), and Jonas Heirman (Evaluation Officer, World Food Programme).

Our results discipline a debate on “big push” interventions

What are the common perceptions of “big push” cash interventions?

“Big push” interventions are commonly proposed to generate significant, sustained increases in household, community, and national income. At the household level, two approaches to increasing intervention size may — in theory — enable households to escape poverty traps and produce persistent decreases in poverty:

1. When households are in a “scarcity poverty trap,” increasing the intensity (cash transfer size) of interventions can push households over a poverty threshold.
2. Alternatively, when households face “frictional poverty traps” (facing many obstacles), increasing the scope (adding complementary interventions to create multi-faceted programs) of interventions can enable households to overcome multiple constraints.

Evaluating increasing intervention intensity and scope requires measures of the impact of these approaches on cost-effectiveness. However, evidence on the longer-term persistence of cost-effectiveness is limited for both approaches.

Addressing the Evidence gap

In a recent Working Paper, we compile 38 estimates of the impacts of temporary cash transfers on household consumption from 14 countries. These estimates were from 17 randomized control trials (RCTs) of either (i) temporary unconditional cash transfers, or (ii) multi-faceted graduation programs with complementary interventions that included temporary unconditional transfers, targeting the ultra poor (TUP). On average, the transfers were administered over eight months for unconditional cash transfers, and over sixteen months for multi-faceted graduation programs.

We included these classes of programs to evaluate two approaches to increasing the size of cash transfers — increasing their intensity (small cash transfers or large cash transfers) and increasing their scope (cash transfers or multi-faceted graduation programs).

The RCTs in our sample all collected information on annual household consumption, cash transfer size, and program cost (in USD purchasing power parity). To measure cost-effectiveness, we focused on the effects of cash transfers on consumption.

Impacts of temporary unconditional cash transfers

• Surprisingly, our results suggest more persistent impacts of smaller unconditional cash transfers. This suggests that smaller cash transfers persistently push more households out of poverty per transfer unit.
• Unconditional cash transfers increased annual household consumption by 0.35 per unit of transfer. Over three years, this implies that increases in household consumption are larger than the initial value of transfers. Consistent with intervention uniformity, these estimates are remarkably consistent across contexts.
• Larger unconditional cash transfers had smaller impacts on consumption per unit of transfer over both the shorter and longer-term.
• The impacts of unconditional cash transfers were more persistent in developing countries than in wealthier countries.
• Lastly, the cumulative impacts of unconditional cash transfers on consumption over the first three years were larger than the size of transfers — providing strong evidence that unconditional cash transfers are cost-effective.

Impacts of complementary interventions that target the ultra poor

• Complementary interventions increased impacts on consumption.
• In our sample, complementary interventions were relatively expensive to implement. Therefore, the average complementary intervention was 5–43 percent less cost-effective at increasing consumption than the average unconditional cash transfer at the average evaluated time horizon (1.5 years for UCT and 2.6 years for TUP).
• Figure 1 shows that the reversal in sign of the relative effect of TUP on consumption per unit of transfer and per unit of cost are driven by the much higher implementation costs of TUP, and that differences in implementation costs across TUP programs are as important as differences in transfer costs between UCT and TUP.
• However, we also find evidence that the average impacts of TUP program complementary interventions mask important variation across contexts. Specifically, we find evidence of variation in the cost-effectiveness of complementary interventions on increasing household consumption.
• Lastly, the relative cost-effectiveness of complementary interventions grew over time — the impacts surpassed those of unconditional cash transfers after 3.4–7.7 years.
• However, only four of the twenty complementary intervention estimates in our sample were more than three years since the last transfer, highlighting the need for more long-run estimates.

Figure 1: Estimates of Impacts of UCT and TUP Interventions on Consumption Per Unit of Transfer or Per Unit of Cost

Recommendations

Our findings highlight the importance of context-specific estimates of the long-run impacts of complementary interventions that target the ultra poor to inform policy. It is worth noting that beyond cost-effectiveness, there are justifications for increasing intervention size to consider. For example, increasing transfer size or providing multi-faceted programs to the poorest households can be powerful tools for poverty reduction (despite the variance in cost-effectiveness).

However, our results also question the necessity of “big push” interventions to reduce poverty, specifically:

• Small (total transfer value under $1000 USD PPP/household) temporary cash transfers provide a strong benchmark for scalable, cost-effective poverty reduction across diverse contexts.
• The presence of poverty traps alone does not justify increasing intervention size.
• Instead, the distribution of poverty thresholds conditional on targeting is crucial.

Acknowledgements

This meta-analysis was completed during the preparation stages of WFP’s Cash-Based Transfers and Gender Impact Evaluation Window and the Climate and Resilience Impact Evaluation Window. It is based on data compiled to complete the Cash-Based Transfers and Gender Impact Evaluation Window pre-analysis plan.

References

Banerjee, A., Duflo, E., & Sharma, G. 2020. Long-Term Effects of the Targeting the Ultra Poor Program. NBER Working Paper Series (28074).

Banerjee, A., Duflo, E., Goldberg, N., Karlan, D., Osei, R., Parienté, W., Shapiro, J., Thuysbaert, B., & Udry, C. 2015. A Multi-faceted Program Causes Lasting Progress for the Very Poor: Evidence from Six Countries. Science, 348(6236).

Ghatak, M. 2015. Theories of Poverty Traps and Anti-Poverty Policies. The World Bank Economic Review, 29(suppl_1), S77–S105.

Kraay, A. & McKenzie, D. 2014. Do Poverty Traps Exist? Assessing the Evidence. Journal of Economic Perspectives, 28(3), 127–148.6.

Read the working paper, including the full results and a detailed discussion of the findings.

If you would like to receive updates on new blogs, events or impact evaluation publications, please fill in this form to receive regular updates or contact oev.impactevaluation@wfp.org for more information.

This blog is also published on the World Bank’s Let’s Talk Development blog.