What Is Cash Good For?
Can comparing anti-poverty programs to unconditional cash transfers improve development programming? Carson Christiano, CEGA Executive Director and DIL Managing Director, explores new findings from a portfolio of cash benchmarking research five years in the making.
In 2016, a research team led by CEGA Faculty Co-Director Edward Miguel set out to study the spillover effects of cash transfers on local communities. Their initial findings focused greater global attention on cash as a tool to fight poverty. Just this month, the team published their groundbreaking results, finding that cash dramatically improved the lives of recipients and non-recipients, while increasing local economic health by a factor of 2.5.
In recent years, the use of unconditional cash transfers as a tool for alleviating poverty has dramatically increased, thanks in part to the efforts of CEGA partner GiveDirectly as well as the challenges wrought by COVID-19. A growing body of evidence suggests that, despite fears, unconditional cash transfers are an effective tool for improving a range of development outcomes.
As interest in unconditional cash transfers has grown, more funders have begun to ask, can my development project compete at a dollar-for-dollar value with cash transfers and still achieve the same (or greater) impact? Enter the Development Impact Lab.
The Development Impact Lab
The Development Impact Lab (DIL) is a global consortium of universities and partners headquartered at the University of California Berkeley, where it is co-managed by CEGA and the Blum Center for Developing Economies. Launched in cooperation with the USAID Higher Education Solutions Network in 2012, DIL has made significant contributions to international development through its innovative Development Engineering approach.
One of DIL’s greatest legacies is its work to explore the potential of cash transfers as a benchmark for measuring the effectiveness of more “traditional” anti-poverty programs, such as those that deliver workforce training or nutrition interventions to beneficiaries. Cash is considered a useful benchmark because is has very low administrative costs (especially when delivered via mobile money) and tend to perform well on household-level outcomes like earnings, educational attainment, and nutrition. DIL has sought to help answer for donors and decision makers whether and when it is better to give people cash or deploy a development program.
This October, DIL released findings from a portfolio of six large-scale, randomized evaluations that—when taken in the context of the broader cash transfer literature—begin to provide a framework for answering this important question. In partnership with USAID, GiveDirectly, Innovations for Poverty Action (IPA), the Education Development Center (EDC), Catholic Relief Services (CRS), and committed local partners, the evaluations assessed the impacts of unconditional cash transfers in Liberia, Malawi, Rwanda, and the Democratic Republic of Congo.
In a recent panel for USAID researchers and staff, the DIL researchers discussed their findings, noting the potential value and limitations of cash benchmarking. Meanwhile, colleagues at USAID and elsewhere are thinking hard about how to operationalize the approach by employing tools like meta-analysis and Bayesian statistics. Ultimately, with more high-quality evidence and insights to work with, development practitioners could begin to simulate the effects of unconditional cash transfers within a given context and geography. They could then use the results as a reference for understanding the effectiveness of their own projects, without needing to run a (costly and time-consuming) head-to-head comparison.
How useful is cash as a benchmark?
While varied across context and design, the DIL cash benchmarking studies produced three important takeaways for development actors.
Cash has relatively consistent impacts on certain outcomes, including short-term consumption and productive assets. The researchers’ findings were mostly consistent with the existing cash transfer literature. “[Cash] is a great way to drive short-term individual improvements on economic indicators that economists tend to focus on,” reported Craig McIntosh, CEGA affiliated professor and researcher at UC San Diego.
However, the studies reinforced the need for more research to understand how cash impacts a broader set of development outcomes. For example, how cash affects specific outcomes, like food security, and why its impacts may differ across different countries and groups remains unclear. Shilpa Aggarwal, researcher at the Indian School of Business, noted that, “There are different pathways through which specific outcomes can occur.”
Importantly, the DIL studies lay the groundwork for synthesized cash benchmarks. Researchers agreed that the new evidence, when taken together with the existing cash literature, can help funders and implementers synthesize cash outcomes, essentially developing a profile of what to expect from a given type and size of cash transfer for certain groups of people.
This profile, which would naturally become more reliable as the evidence base grew, could serve as a benchmark for traditional aid and inform consequential funding and programming decisions. McIntosh added, “USAID could sit down and say, ‘For a given spend, and a given beneficiary group, these are the outcomes we could expect if we did cash transfers.’”
Yet, using cash as a benchmark has important limitations. Like the use of cash transfers for anti-poverty programming more broadly, cash benchmarking is not a silver bullet. It needs to align with internal motivations. Various program dimensions and donor priorities can affect the accuracy of cash as a benchmark, including the design of a project, its participants, and the desired outcomes. “Where outcomes, timing, and resources line-up well, these are the cases where benchmarking will lead to less ambiguity,” noted Andrew Zeitlin of Georgetown University.
The importance of high-quality cost data
“One of the important contributions of the DIL [cash benchmarking] studies is that they recognize costing as a primary research endeavor… elevating the discussion of cost-effectiveness and, in particular, measuring program cost and cost per impact,” said Liz Brown, CEGA Staff Scientist.
With better evidence, development actors have increasingly sought more guidance on how to use cash benchmarking effectively. Including high-quality cost data and cost-effectiveness analysis in future impact evaluations of programs implemented by USAID and others will ultimately enable donors to apply cash benchmarking to more sectors and contexts — far beyond the six programs evaluated by DIL. And, investing in more head-to-head evaluations of traditional aid programs against cash, when feasible, will further increase the value of cash benchmarking as a measurement tool.
Most importantly, the DIL studies provide a road map for the sector, helping to establish clear signposts for development actors on how to compare projects to cash and when cash may be more attractive. As aid budgets come under increasing scrutiny and donors seek to maximize impact at scale, cash benchmarking may end up being some of the smartest money out there.
For more information about Unconditional Cash Transfers:
