$170 billion needed to end poverty, Stanford researchers find

Stanford economic researchers estimate that ending poverty worldwide would cost $170 billion per year, establishing a concrete number for future policies.

The study, published in December by researchers across several universities, could help inform future measures to reduce poverty, especially in lower and middle-income countries. Lead author Roshni Sahoo, a third-year Ph.D. student at Stanford, used policy learning, a research method that proxies survey data, to produce this estimation.

“Our work quantifies how much cash you would need to allocate to people in a particular country for a year to get everyone to the poverty line,” Sahoo said. “You can think of this work as a blueprint for how to develop an actionable plan that allocates cash to achieve a poverty reduction objective.”

Co-authors Paul Niehaus and Joshua Blumenstock helped guide Sahoo’s research. Niehaus is an economist at UC San Diego, and Blumenstock is a public policy director at UC Berkeley.

“Niehaus and I had spent some time trying to figure out whether it would be possible to estimate the cost of ending poverty via direct transfers when we realized that there were some difficult statistical learning problems that we couldn’t solve ourselves,” Blumenstock said.

Sahoo was able to solve this issue using modern statistical learning based on high-quality survey data. 

According to Sahoo, estimating how much money is needed to end global poverty is difficult because the term’s definition differs based on context. Sahoo’s research approached this problem by using policy learning algorithms that ran on information drawn from World Bank surveys across 23 countries.

According to the World Bank’s definition, individuals earning less than $3 per day live under the global poverty line. “We set it up as our goal to minimize some poverty measure subject to a budget constraint of how many resources the policy makers had to allocate,” Sahoo said. “We turned this into a statistical learning problem.”

The study focused on lower and middle-income countries that suffer more severely from extreme poverty. According to Blumenstock, the total cost of eradicating poverty would likely be 0.3% of global gross domestic product, a relatively small figure. “Annual expenditures on alcoholic beverages account for 2.2% of global GDP, and annual expenditures on cosmetics account for 0.6% of global GDP,” he said.

For Sahoo, this research aims to help “policymakers who are interested in setting development assistance targets for reaching poverty reduction goals.” Sahoo noted that a tangible cost of poverty reduction is vital in setting perspective for donors to nonprofit organizations and government policymakers alike.

The study’s researchers said survey analytics through policy learning allow for more accurate and precise data that lawmakers can trust. “The goal of policy learning is to enable people to formulate goals and then deploy data such as to achieve these goals as efficiently as possible,” said Stefan Wager ’11 Ph.D. ’16, associate professor of operations, information and technology at Stanford’s Graduate School of Business and co-author of the study. 

While the study relied on up-to-date data, “robustness to temporal shifts could be a concern,” according to Sahoo, as populations or economies change over time.

“I hope that this paper can help people see that extreme poverty persists not because it is prohibitively expensive to address but because we lack the institutional and political will to do so,” Blumenstock said.

The post $170 billion needed to end poverty, Stanford researchers find appeared first on The Stanford Daily.