We study the long-term effects of immigration on anti-immigrant sentiment and voting for nationalist parties. Drawing on a natural experiment in post-war Germany where the Allied occupation led to a discontinuous and quasi-exogenous distribution of forced migrants in one region, we examine how a large migrant inflow shapes local voting outcomes and electoral reactions to subsequent immigration between 1949 and 2021. Applying a spatial regression discontinuity design and combining historical migration records with panel data at the municipality level, our results reveal a weaker nationalist backlash against present-day immigration where more migrants settled in the late 1940s. To study the mechanisms, we conduct a geocoded survey with experimental elements and open-ended questions in the study region and find that both family history and local collective memory of immigrant integration contribute to explaining these findings.

• news coverage by FAZ

What are the distributional effects of placed-based policies? Drawing on household data from 2.4 million survey respondents in the European Union (EU), we show that income inequality within European regions is substantial, has widened since the 1990s and contributes more to overall inequality than cross-regional inequality. Using regression discontinuity and difference-in-differences designs, we find that the world’s largest place-based policy, the EU’s Cohesion Policy, increases disposable income for affluent households but barely affects low-income households in supported regions. Evidence on mechanisms demonstrates that place-based funds exacerbate intra-regional inequality by primarily boosting labor income for the highly skilled.

• basis for the European Parliament's parliamentary question E001467/2023

• coverage by The Economist, Süddeutsche Zeitung (SZ), Politico, Euractiv, Le Monde

How do powerful illiberal states affect the norms and institutions of the liberal international order (LIO)? We argue that they actively promote rival illiberal norms through the fora provided by liberal institutions and expect such illiberal norm promotion to be effective when supported by material tools of statecraft. To test this argument with respect to human rights norms, we introduce HR-RES, the most comprehensive dataset of decision-making on resolutions in the United Nations Human Rights Council (HRC). HR-RES covers granular information on all HRC resolutions from 2006 to 2023, including sponsors, topics, full texts, and votes cast. Using natural language processing and panel regressions, we find that China stands out in actively promoting illiberal norms while consistently challenging human rights friendly resolutions. Receiving aid and bailouts from China predicts voting alignment with China on resolutions with illiberal content; non-material tools of statecraft such as diplomatic and cultural initiatives are ineffective. Topic modeling suggests a gradual shift in resolution content consistent with illiberal influence over the HRC. We discuss the implications of these results for understanding the global rise of illiberalism, the geopolitics of human rights, and the resilience of the LIO.

International organizations are often approached as expertise-driven homogeneous technocracies. In this article, we argue that their bureaucracies are neither unitary nor politically neutral but staffed by individuals with ideological biases of their own, which can manifest in organizational output. Examining this argument in the context of International Monetary Fund (IMF) conditionality, we collect individual-level data on the careers of 835 IMF `mission chiefs' – staffers with primary responsibility for a specific member country – and match them to newly-coded data on more than 15,000 IMF conditions of the 1980-2016 period. Leveraging the appointment of the same mission chiefs to different countries throughout their careers, we find that individual staffers are able to influence the number, scope, and political content of IMF conditions according to their own ideological leaning. The results cast doubt on prominent arguments of international organizations as cohesive and impartial bureaucracies, instead highlighting the `micro-foundations' of their policy output.

Can debt moratoria help countries weather negative shocks? We exploit the Debt Service Suspension Initiative (DSSI) to study the bond market effects of deferring official debt repayments. Using daily data on sovereign bond spreads and synthetic control methods, we show that countries eligible for official debt relief experience a larger decline in borrowing costs compared to similar, ineligible countries. This decline is stronger for countries that receive a larger relief, suggesting that the effect works through liquidity provision. By contrast, the results do not support the concern that official debt relief could generate stigma on financial markets.

Over the past decades, incomes converged across countries yet diverged within countries, changing the structure of global inequality. We provide theory and evidence suggesting that economic globalization promotes this dual global trend. A simple model predicts diminishing marginal gains of globalization across countries and increasing inequality within countries. Our empirical analysis uses global panel data at the country-decile-group level and exploits geographic diffusion of liberalization policies for identification. Globalization since 1970 has led to a) income convergence across countries, because marginal gains from globalizing diminish and ultimately disappear for highly globalized countries, and b) income divergence within countries, because gains are largest for the richest ten percent of national income distributions. 

• coverage by IMF, World Economic Forum, Mannheimer Morgen, Tagesspiegel, Bild der Wissenschaft, Research in Germany

World Development 176: 106501 (2024)

A vast literature evaluates the effectiveness of development aid and often reaches sobering conclusions. We argue that a key shortcoming of this literature is the focus on a concept of effectiveness---mostly economic growth---that does not match the kind of effectiveness that aid donors actually aim at. To determine actual donor motives, we first survey the literature on aid allocation and identify a large set of motives that is common to many donors. We then employ this set of donor motives for compiling a survey of the aid effectiveness literature by donor motive. We find that while aid has a moderate effect on economic development at best, aid is effective in achieving many of the effects primarily intended by its donors. We conclude by speculating that future research on aid effectiveness will be more likely to identify significant effects of aid when taking donor motives into account.

How do governments choose between bilateral and multilateral foreign policy? We argue that powerful governments can exploit their influence over multilateral organizations to hide contentious foreign policies from domestic audiences. Applying this argument to international financial flows, we formulate hypotheses on the choice between bilateral and multilateral financing for political purposes. We test them by examining how the United States incentivizes support for decisions on United Nations Security Council resolutions. Introducing a new dataset on 2,530 Security Council decisions between 1946 and 2015, the results show that temporary Security Council members receive more bilateral and multilateral financing only when they support the positions of the United States. The United States uses bilateral aid to incentivize the support of allies and uses its power over the World Bank and the International Monetary Fund to channel multilateral finance to countries that its domestic audience views more negatively. 

Policymakers in crisis countries often hesitate to enter IMF programs out of the fear that they trigger adverse reactions on financial markets. We explain why credit ratings and investor assessments are reliable measures of creditworthiness during crises, and examine how IMF programs affect them with three distinct identification strategies. The first strategy exploits the differential effect of changes in IMF liquidity on loan allocation as an instrument, the second uses the exact timing of program agreements, and the third provides text-based evidence from rating agency statements. When accounting for endogenous selection, we find that IMF programs help countries regain their creditworthiness. Even though IMF programs tend to result in economic contractions, the agreement on a program is perceived as a positive signal on financial markets. Our text-based analysis supports this signaling effect and suggests that the content of programs matters for how they are perceived.

Does the International Monetary Fund (IMF) increase inequality? To answer this question, this article introduces a new empirical strategy for determining the effects of IMF programs that exploits the heterogeneous effect of IMF liquidity on loan allocation based on a difference-in-differences logic. The results show that IMF programs increase income inequality. An analysis of decile-specific income data shows that this effect is driven by absolute income losses for the poor and not by income gains for the rich. The effect persists for up to 5 years, and is stronger for IMF programs in democracies, and when policy conditions, particularly those that demand social-spending cuts and labor-market reforms, are more extensive. These results suggest that IMF programs can constrain government responsiveness to domestic distributional preferences.

Much of the International Finance Corporation's (IFC) lending benefits private companies from rich countries and supports projects in middle-income countries. Large corporations such as Lidl or Mövenpick have received its loans for highly profitable investments. This contrasts to some extent with the IFC's official mandate, which is to finance poverty-reducing projects for which private capital is not available on reasonable terms. Investigating a potential driver of this mismatch, we argue that some governments can influence the allocation of IFC loans to the benefit of private companies in their countries. Using new data for more than 3000 IFC projects over the 1995–2015 period we show that (joint) IFC Board membership of countries where borrowing companies are based and of countries where the projects are implemented increases the likelihood that these countries receive IFC loans. This has implications for the debate on leveraging private-sector investments for development.

We exploit the degree of discretion embedded in the World Bank-IMF Debt Sustainability Framework (DSF) to understand the decision-making process of international financial institutions. The unique, internal dataset we use covers the universe of debt sustainability analyses conducted between December 2006 and January 2015 for low- income countries. These data allow us to identify cases where the risk rating implied by the application of the DSF's mechanical rules was overridden to assign a different official rating. Our results show that both political interests and bureaucratic incentives influence the decision to intervene in the mechanical decision-making process. Countries that are politically aligned with the institutions' major shareholders are more likely to receive an improved rating; especially in election years and when the mechanical assessment is not clear-cut. These results suggest that the room for discretion international financial institutions have can be a channel for informal governance and a source of biased decision-making.

• coverage by Handelsblatt

We study the social and distributional implications of the euro area (EA) sovereign debt crisis in countries that received external financial assistance for macroeconomic adjustment. Combining granular data from about 1 million survey interviews with households across euro area states carried out between 2004 and 2019, we show how disposable income and its components evolved across different social groups before, during and after the EA crisis and the associated adjustment programmes. Using difference-in-difference and triple-difference methods, we compare distributional outcomes in countries with large financial assistance programs (Greece and Portugal) to euro area countries with similar pre-crisis trajectories and macroeconomic fundamentals. The results show that incomes for most social groups in these countries declined during the crisis, but income losses were not uniform and differed across groups in terms of gender, age, and income level. Overall, public transfers mitigated increases in inequality of market incomes, yet both the timing and the magnitude of changes in market income and public transfers varied across income groups. While reforms of public transfer policies in the early macroeconomic adjustment period placed a greater burden on poorer income groups than on richer ones, reforms in later stages reduced inequality by supporting poorer households, especially in the case of Greece. 

Expressions of political discontent are often interpreted as responses to perceived threats to people’s social status or social identity. While the literature finds associations between status threat and political discontent, evidence is scarce on whether this is due to a causal effect. In this paper, we leverage the relegation of local football teams as an exogenous shock to social identity in order to study the effect of identity shocks on political behavior. In many European countries, including Germany, local football teams are an important part of people’s social identity. The decline of such teams can thus constitute a powerful threat to this identity. We use Google search data to identify municipalities affected by the relegation of a popular football team and match these data to a panel of municipality-level federal election results since 1994. Using cross-sectional regressions, panel regressions with two-dimensional fixed effects, and a regression discontinuity design, we find that citizens in municipalities affected by such shocks are less likely to support established mainstream parties and more likely to support fringe parties on the left and on the right. We thus demonstrate that football can affect political outcomes through its effect on social identity. More generally, we provide evidence for a link between social status threat and electoral outcomes. 

The level of global income inequality is staggering despite a significant decline over recent years. While the awareness of global redistribution schemes is high among elites, political leaders, and international institutions, we know little about the popularity of such transfers among the broader population in rich countries. This paper examines support and opposition to global redistribution through the lens of moral values. We propose that individual-level heterogeneity in moral values accounts for differences in the support of redistributive measures at the global level. Empirically, we focus on the context of the unequal distribution of COVID-19 vaccines at the early stages of the pandemic, which closely resembled global income inequality. We run a large-scale experiment with N = 10,115 observations in Germany, investigating public support for the redistribution of COVID-19 vaccines from the perspective of people in a rich country. To test the moral boundaries of support for global redistribution, we experimentally manipulate the framing of the benefits of vaccine transfers, varying between a universalist and a communitarian (less universalist) framing. In line with a universalism framework, altruism moderates the effect of the universalistic treatment but not the effect of treatments emphasizing in-group benefits. People who donate their survey incentives to a charity more strongly respond more to the universalistic treatment than people who do not donate.