Embedded Files
Working Papers
Working Papers
Effect on volume of exports
with Pedro Moreira
How do exporters respond to changes in the cost of credit? We exploit an exogenous shift in banking regulation that increased the cost of financing for exports in the European Union. Using customs, firm-level, and credit registry data on Portuguese firms, we find that a rise in financing costs reduces exports by 8 percent through the intensive margin, while firm entry into new destinations declines. Within firms, exporters adjust their product mix, skewing it toward goods with lower working capital requirements. We provide direct evidence of the mechanism: loan rates for exporters increase, and loan amounts decline by 7 percent. We also provide evidence of credit rationing. A multi-sector Ricardian model shows that these disruptions lead to a depreciation in terms of trade, reducing welfare in both the EU and its trading partners.
[PDF] [Cite]
[PDF] [Cite]
Effect on Boeing orders by country group
We study the effectiveness of U.S. industrial policy by examining a sharp disruption in the Export-Import Bank’s ability to support exporters. In 2015, EXIM lost the quorum required to approve large loan guarantees, reducing total aid. We focus on Boeing, EXIM’s largest beneficiary, and use detailed aircraft-level data to measure how the lapse affected orders. On average, demand for Boeing aircraft did not decline—despite 12% of sales having been EXIM-financed. However, airlines based in developing countries and those with low liquidity significantly reduced their purchases, consistent with binding credit constraints. By contrast, carriers in developed countries show no response. Since EXIM’s stated objective is to maximize U.S. exports, aid should be directed toward customers whose demand is most sensitive to credit terms. Under this criterion, the pre-shock allocation appears inefficient: reallocating aid toward credit-constrained buyers could have raised Boeing orders by 16%.
[PDF] [Cite]
[PDF] [Cite]
Effect on relative wage of women - primary earners vs. non-primary earners
Intra-Household Decisions and Labor Market Outcomes - Evidence from Shared Parental Leave (2024)
with Andrea Ferrara and Menaka Hampole.Intra-Household Decisions and Labor Market Outcomes - Evidence from Shared Parental Leave (2024)
This paper examines the impact of intra-household decisions over the split of childcare duties on labor market outcomes. We study the introduction of shared parental leave in Portugal, which allows parents to decide on the allocation of leave days. Using a model of the household, we show that introducing shared parental leave leads to an increase in women’s wages, as they are allocated a lower of childcare duties when compared with the allocation before shared parental leave is introduced. Moreover, this wage increase should be more pronounced for high-productivity women. Using a novel data set which combines household data with matched employer-employee data, we find that the monthly wages of women increase by 1 percent relative to the wages of men. We also find that most of this increase is driven by women which are the primary earners in their household. Our results suggest that the effectiveness of childcare policies in mitigating gender inequality in the labor market may be determined by intra-household decisions.
[PDF] [Cite]
[PDF] [Cite]
Median distance between borrower and lender
Does Distance Still Matter? Evidence from 30 Years of Mortgage Lending (2025)
with Menaka Hampole and Adam Jorring.Does Distance Still Matter? Evidence from 30 Years of Mortgage Lending (2025)
Mortgage lending by banks is still predominantly a local business. Local branches have a significant advantage in attracting and originating mortgages. More than 25% of borrowers are less than two miles from their branch, and more than 50% are within 10 miles. Distance is associated with lower credit quality, and the additional risk is priced: higher distance is associated with a lower probability of approval, a higher interest rate spread, and a higher ex-post likelihood of delinquency. We document business cycle patterns as the distance rises in booms and falls in busts but find that the sensitivity of distance for loan approval and pricing has persisted over the past 30 years despite the dramatic rise in securitization and online banking. Using a quantitative spatial model of the mortgage market, we explore the economic mechanisms driving our results. Through the lens of the model, we find that decreasing search costs lowers the match quality of borrower search, which counteracts the effects of increasing screening technology. Our findings highlight that local bank branches are still highly relevant despite the "fintech’’ revolution.
[PDF] [Cite]
[PDF] [Cite]
Impulse response of likelihood of conflict to an increase in military spending
We revisit the theory of critical deterrence—the idea that military spending may reduce the risk of conflict by increasing its expected costs. To test this theory, we assemble a new cross-national dataset combining information on armed conflicts and defense spending over a 75-year period. We find that increases in military spending have no effect on short-run conflict risk—contrary to concerns that buildups may provoke escalation—but that they do lead to a small and persistent decline in conflict over the long run. Although the effect of spending on conflict is modest, its effect on the cost of war is large: higher military spending raises the scale of battlefield casualties. The effect of military spending on conflict is more pronounced for democracies, which are less likely to use force offensively, and is concentrated in internal conflicts involving nonstate actors. The deterrent power of military spending is strongest in ethnically polarized societies—precisely where the threat of civil war is most acute. Last, using dyadic data, we find that higher military spending causes a decrease in bilateral interstate conflict, and that this effect may be larger than the one we document for intrastate conflict using panel data. Taken together, our results point to a low pass-through from the cost of war to its incidence: even sharp increases in military spending yield only modest gains in deterrence.
[PDF] [Cite]
[PDF] [Cite]
Effect of conflict on real output
The Economic Consequences of War (2025)
with Efraim Benmelech.The Economic Consequences of War (2025)
[VoxEU Column] [NPR Planet Money] [Kellogg Insight]
We provide systematic evidence on the macroeconomic consequences of war using a new dataset covering 115 conflicts and 145 countries over the past 75 years. We document three main findings. First, conflict generates large and persistent real effects: real GDP falls by 13% on average, with no recovery even after a decade, while investment collapses as financial frictions reduce domestic credit. The drop in real activity is more pronounced for intrastate conflicts than it is for interstate conflicts. Second, government finances deteriorate as revenues contract and expenditures remain stable, thus raising primary deficits. Real government debt also declines, and governments shift 1.2% of GDP toward short maturities. Third, governments rely heavily on inflation to finance their deficits: the price level and money supply both rise by nearly 50%, eroding debt and generating seigniorage but also depressing investment and raising the cost of imported capital goods.
[PDF] [Cite]
[PDF] [Cite]
Effect of subsidy on capital-to-intangible ratio
How do financial frictions shape firms’ investment in intangible capital? We show that financial constraints distort firms’ input choices, leading to systematic underinvestment in intangible assets. Exploiting an investment subsidy in Portugal that lowered the cost of both physical and intangible capital while keeping their relative price unchanged, we find that treated firms reduced their capital-to-intangible ratio by 12 percent, with larger effects for financially constrained firms. The distribution of treatment effects declines sharply across percentiles - firms with higher initial capital-to-intangible ratios adjust the most - revealing the signature of a binding financial wedge. Going beyond average effects, we recover the entire cross-sectional distribution of wedges between the marginal rate of technical substitution and the price ratio. The recovered distribution corresponds to the least distorted economy consistent with the data, showing that only a small share of firms are unconstrained while roughly one-quarter face wedges exceeding twenty percent - providing a direct quantitative map of financial distortions in production.
[PDF] [Cite]
[PDF] [Cite]
Work in Progress
Work in Progress
We provide the first global, long-run evidence on how war reshapes democratic institutions. Using data on all conflicts since 1948, we show that the onset of conflict causes a large and persistent decline in democracy: institutions weaken immediately, continue to erode for nearly a decade, and do not recover. Yet this deterioration is highly selective. It appears only in first-time conflicts, intrastate wars, highly fractionalized societies, and conflicts that governments win. The decline operates through political channels - media censorship, judicial purges, curtailed civil liberties, irregular leadership turnover, and constitutional suspensions - rather than through any functional requirement of war-making. Autocratization does not increase the probability of victory, and institutional instability reduces it. Taken together, the findings show that war does not require autocracy; it enables executives to expand their authority and implement institutional changes that would be difficult to enact in peacetime.
Page updated
Google Sites
Report abuse