Bottom Up Sovereign Debt Preferences
by Alessandro Del Ponte and Matthew DiGiuseppe
PDF LINK
The public places an important constraint on a government’s ability to maintain sta- ble debt burdens and repay sovereign debt in times of crisis. Yet, scholars have only begun to examine how the public reasons about government debt. The nascent liter- ature finds that a mix of ego-tropic reasoning and top-down elite cues inform public attitudes on sovereign debt policy. In this paper, we propose a bottom-up explanation to complement existing theories. We argue that citizens’ preferences over sovereign debt are also driven by their attitudes toward private debt. Drawing from theories and methods from moral psychology, we propose that the link between private and public debt is similar to other folk economic beliefs that conflate how household and government budgets work. To test our argument, we conducted several preregistered studies. In Brazil and Italy, we find observational evidence that private debt attitudes are significantly correlated with citizens’ positions on public debt, and this is in part driven by their self-reported moral convictions. A survey experiment fielded in Italy finds that manipulating attitudes toward private debt changes support for public debt repayment.
Putting the Public in Public Debt
by Matthew DiGiuseppe, Jaroslaw Kantorowicz, Bastián González Bustamante & Alessia Aspide
PDF LINK
Models of public debt accumulation and consolidation often rely on implicit assumptions about citizen preferences regarding their attention to fiscal policy and the cleavages that form around policies. However, few studies directly examine public engagement with the issue. We address this gap using original open-ended survey data from three highly indebted countries: Italy, Japan, and Brazil. Leveraging generative Large Language Models (LLMs) to code open-ended responses, we assess the public's understanding of the consequences of rising debts and if it aligns with economic evaluations by comparing their responses to those of ``synthetic economists'' utilizing the domain knowledge of LLMs. We find that almost all citizens expect negative consequences from public debt increases but are less specific in their answers than synthetic economists. Next, we look for evidence that cleavages over debt policy fall on partisan and generational lines and find that age and partisanship are associated with small differences in preferred policies to reduce debt. However, the pattern is not consistent across countries. The analysis reveals that the public is not entirely ignorant of the consequences of public debt, and cleavages are not as dramatic as popular models may have assumed.
IMF Survival Instincts: Risk Exposure and Design of Loan Programs
by Kathleen Brown
PDF LINK
When does the International Monetary Fund (IMF) play hardball? While some governments in distress are asked to make costly reforms in exchange for a bailout, others access emergency funds with ease. Previous work has attributed this variety to borrowing government characteristics; instead, I propose that the IMF's overall risk exposure determines the deal that each new borrower receives. As the global lender of last resort the goal of the IMF is to preserve financial stability, but it must also ensure its own solvency and survival. I argue that when a greater share of outstanding loans is owed by high-risk borrowers, the IMF mandates stricter policy conditions to protect itself from default. Using a new index of the IMF's risk exposure and several indicators of the design of loan conditions, I demonstrate that the IMF changes its lending behavior to protect its own balance sheet. During periods of high-risk exposure, the IMF requires that its borrowers complete more policy conditions across a broader scope of policy areas, increasing Fund control over repayment. These findings illustrate how the IMF's goal of self-preservation shapes emergency sovereign lending and contributes to ongoing debates about how bureaucratic interests influence the policy outputs of international organizations.
Scaling Open-ended Survey Responses Using LLM-Paired Comparisons
by Matthew DiGiuseppe & Michael E. Flynn
PDF LINK
Survey researchers rely heavily on closed-ended questions to measure latent respondent characteristics like knowledge, policy positions, emotions, ideology, and various other traits. While closed-ended questions ease analysis and data collection, they necessarily limit the depth and variability of responses. Open-ended responses allow for greater depth and variability in responses, but are labor-intensive to code. Large Language Models (LLMs) can solve some of these problems, but existing approaches to using LLMs have a number of limitations. In this paper, we propose and test a pairwise comparison method to scale open-ended survey responses on a continuous scale. The approach relies on LLMs to make pairwise comparisons of statements that identify which statement ``wins'' and ``loses''. With this information, we employ a Bayesian Bradley-Terry model to recover a `score' on a the relevant latent dimension for each statement. This approach allows for finer discrimination between items, better measures of uncertainty, reduces anchoring bias, and is more flexible than methods relying on Maximum Likelihood Estimation techniques. We demonstrate the utility of this approach on an open-ended question probing knowledge of interest rates in the US economy. A comparison of 6 LLMs of various sizes reveals that pairwise comparisons show greater consistency than zero-shot 0-10 ratings with larger models ($>$ 9-billion parameters). Further, comparison of pairwise decisions are consistent with high-knowledge crowd source workers.
Information, Party Politics, and Public Support for Central Bank Indpendence
by Matthew DiGiuseppe, Carolina Garriga, Andreas Kern
PDF Link
Threats to the independence of central banks (CBI) are emerging in many countries after central banks have hiked interest rates. While existing literature has extensively documented conflicts between elected politicians and independent central bankers, underscoring the importance of maintaining the political independence of central banks, we know surprisingly little about what (if anything) the public thinks about CBI. We hypothesize that support for CBI is influenced by citizens' limited understanding of central bank governance and their beliefs about who will gain control over monetary policy if independence is reduced. We expect that, when informed that the President would gain more influence, respondents' support for CBI will increase. Further, we argue that support for CBI hinges on which party holds the presidency. When a co-partisan (out-partisan) is President, respondents should favor reduced (increased) independence. Our expectations are confirmed by a preregistered survey experiment and a pre-post-election test. Informing respondents that the Presidency will gain influence if CBI is reduced and, in separate tests, altering expectations of co-partisan presidential election victory alter attitudes on CBI. Further, attitudes toward CBI shift post-election conditional on partisanship. From a policy perspective, our findings indicate that CBI has an important institutional function as a check-and-balance on government power, safeguarding price and macro-financial stability during intense political disagreement along partisanship lines.