The claim that social protection is a luxury good--with a national income elasticity exceeding unity--has as been influential. The paper tests the "luxury good hypothesis" using newly-assembled data on social protection spending across countries since 1995, treating the pandemic period...

The safety nets in high-income countries before 1900 and in low-income countries today were based on savings and aid from extended family, friends, charities, churches, and small amounts from local governments. Mutual societies and eventually insurance companies offered insurance against lost...

Can participation in safety net programs have long-lasting negative effects across multiple generations? Prior work shows a 1993 Dutch disability insurance reform which tightened requirements and lowered benefits for participants resulted in better outcomes for their children. We study the third...

The inflow of refugees and their subsequent integration can be an important challenge for both the refugees themselves and the host society. Policy interventions can improve the lives and economic success of refugees and of their communities. In this paper, we review the socioeconomic...

We study fraud in the unemployment insurance (UI) system using a dataset of 35 million debit card transactions. We apply machine learning techniques to cluster cards corresponding to varying levels of suspicious or potentially fraudulent activity. We then conduct a difference-in-differences...

Two million Mexicans were granted lawful permanent residency in the U.S. under the Immigration Reform and Control Act of 1986 (IRCA). We find that occupation and program use variables in a prominent proxy for legal status poorly detect this event. A decade after legalization, the share of...