Highlights 2021

WP 21/18

In “A note of caution in interpreting cross-country correlations of COVID-19 vaccination and infection rates”, Francesco Bartolucci, Franco Peracchi, and Daniele Terlizzese show that the finding in Subramanian and Kumar (2021) of no discernible linear relationship between COVID-19 vaccination rates and new infection rates in a cross-section of countries is misleading, because it ignores the substantial degree of heterogeneity across countries. The latter reflects large differences in the stages of the infection and the vaccination process, healthcare systems, people’s attitude towards vaccination, etc. In the presence of such heterogeneity simple correlations are hardly interpretable. This is a well-known phenomenon, sometimes referred to as the Simpon’s paradox. Exploiting longitudinal data, they show that the estimated linear relationship becomes negative and statistically significant when controlling for time-invariant differences across countries.

WP 21/17

Information is crucial for job search. Workers need to know the full set of relevant information in order to find their best matches, including the wage that each job offers and the odds of getting hired. However, assuming that workers have full access to this information is unrealistic. Only a small fraction of job postings contain explicit information regarding wages. On the other extreme, assuming that workers do not have any information and randomly search for jobs is also unrealistic. In “Partially Directed Search in the Labor Market”, Liangjie Wu provides an equilibrium theory where workers have partial information in their job search, to understand the labor market implications of limited information inherent in the job search process. Workers pay a cost to direct job search that is proportional to the divergence between the chosen search strategy and a benchmark random search strategy. With this cost, workers apply to every job with a positive probability, but apply to high-payoff jobs with higher probabilities. In a wage posting model with partially directed search, employers have monopsony power: firms extract a markdown due to the cost of directing search. Efficiency of the market equilibrium depends on whether the markdowns are equally distributed across firms. Inefficiency arises when search cost is intermediate, which has new implications on policy remedies to monopsony.

WP 21/16

The level of private savings is a key indicator of macroeconomic performance. As such, in the last 20 years there has been an intense debate about the evolution of individual and aggregate savings. While some countries, such as China, appear to be experiencing saving gluts, large western economies appear to have embarked in dearths of savings. In “Saving Rates and Savings Ratios,” Guillermo Ordoñez and Facundo Piguillem proposed and alternative indicator that can reconcile apparently puzzling observations. The flow of savings as a fraction of disposable income saving rate and the stock of savings as a fraction of total wealth savings ratio are tightly connected. They use a standard dynamic model to show that they may move in opposite directions when financial and/or human capital change dramatically. Making this link theoretically explicit provides an internally consistent measure of savings ratios based on saving rates and other publicly available data. They implement this measure for the four largest economies: U.S., China, Germany and Japan, and identify periods in which saving rates and savings ratios have moved in opposite directions. They find that those departures are not explained by capital gains, but instead by changes in the value of human capital.

WP 21/15

While the underrepresentation of female college majors in STEM has long been recognized, the presence of similar or even larger gaps in Economics has instead been surfacing much more recently, despite their repercussions for women's occupational and earnings prospects and overall economic efficiency. In “Adams and Eves: The Gender Gap in Economics Majors”, Graziella Bertocchi, Luca Bonacini and Marina Murat investigate the gender gap in Economics among bachelor's and master's graduates in Italy between 2010 and 2019. They establish the following. First, being female exerts a negative impact on the choice to major in Economics, with only 73 women graduating in Economics for every 100 men at the bachelor level. The mathematical content of high school curricula is the key driver of the effect and the gap persists at the master level. Second, within a full menu of major choices, Economics displays the largest gap, followed by STEM and then Business Economics. Third, decomposition analyses expose a unique role for the math background in driving the Economics gender gap relative to other fields. Lastly, a triple difference analysis of a high school reform shows that an increase in the math content of traditionally low math curricula caused an increase in the Economics gender gap among treated students.

WP 21/14

In “The Geography of Investor Attention”, Marco Pagano, together with Stefano Mengoli e Pierpaolo Pattitoni, documents that retail investors pay over twice as much attention to local companies than non-local ones, based on Google search data, and that news volume and volatility amplify this attention gap. Attention appears causally related to perceived proximity: first, acquisition by a nonlocal company is associated with less attention by locals, and more by nonlocals close to the acquirer; second, COVID-19 travel restrictions correlate with a drop in relative attention to nonlocal companies, especially in locations with fewer flights after the outbreak. Finally, local attention predicts volatility, bid-ask spreads and nonlocal attention, not viceversa. These findings are consistent with investors having an information-processing advantage about local firms relative to nonlocal investors, as opposed to the featuring behavioral biases in favor of local investments. They are also consistent with the findings of other recent papers documenting that investors' attention predicts stocks' risk-adjusted returns, and that this relationship is much stronger for local investors' attention.

WP 21/13

In “Loan Guarantees, Bank Lending and Credit Risk Reallocation”, Marco Pagano and Andrea Polo, together with Carlo Altavilla, Andrew Ellul and Thomas Vlassopoulos, investigate whether government credit guarantee schemes, extensively used at the onset of the Covid19 pandemic, led to substitution of non-guaranteed with guaranteed credit rather than fully adding to the supply of lending. They study this issue using a unique euro-area credit register data, matched with supervisory bank data, and establish two main findings. First, guaranteed loans were mostly extended to small but comparatively creditworthy firms in sectors severely affected by the pandemic, borrowing from large, liquid and well-capitalized banks. Second, guaranteed loans partially substitute pre-existing non-guaranteed debt. For firms borrowing from multiple banks, the substitution mainly arises from the lending behavior of the bank extending guaranteed loans. Substitution was highest for funding granted to riskier and smaller firms in sectors more affected by the pandemic, and borrowing from larger and stronger banks. Overall, the evidence indicates that government guarantees contributed to the continued extension of credit to relatively creditworthy firms hit by the pandemic, but also benefited banks’ balance sheets to some extent.

WP 21/11

In “Leave the Door Open? Prison Conditions and Recidivism”, Daniele Terlizzese and Giovanni Mastrobuoni, estimate the causal effect on recidivism of replacing time served in a closed-cell prison with time served in an open-cell one, using exogenous variation driven by nearby prisons’ overcrowding. Switching regimes for a year reduces recidivism by around 6 percentage points. The effects are largest for inmates with low levels of education and are weak for violent and hardened criminals.

WP 21/10

In "Cash: a blessing or a curse?" Fernando Alvarez, David Argente, Rafael Jimenez and Francesco Lippi estimate the response of criminal and black-economy activities to changes in the availability of cash as means of payment. Their analysis is based on two quasi-natural experiments that encouraged the use of debit cards and facilitated the use of ATMs in Mexico. They show that restricting cash usage has social benefits, linked to the reduction of some criminal activities, but also costs, because it distorts the individual choices regarding the means of payment. They find that the latter outweighs the former.

WP 21/09

In "Empirical Investigation of a Sufficient Statistic for Monetary Shocks" Fernando Alvarez, Andrea Ferrara, Erwan Gautier, Hervé Le Bihan and Francesco Lippi explore the empirical validity of the following theoretical claim: in a broad class of sticky price models the non-neutrality of nominal shocks is captured by a sufficient statistic computed from the distribution of price changes. They test this claim using micro data for a large number of firms representative of the French economy, and find evidence that is aligned with the theory. They also show that several moments of the distribution, not suggested by the theory, do not correlate with the neutrality of monetary shocks.

WP 21/08

In “Weighted-average least squares (WALS): Confidence and prediction intervals”, Giuseppe De Luca, Jan Magnus, and Franco Peracchi extend results from their recent paper in the Journal of Econometrics to inference for linear regression models based on weighted-average least squares (WALS), a frequentist model averaging approach with a Bayesian flavor. They concentrate on inference about a single focus parameter, interpreted as the causal effect of a policy or intervention, in the presence of a potentially large number of auxiliary parameters representing the nuisance component of the model. In their Monte Carlo simulations they compare the performance of WALS with that of several competing estimators, including the unrestricted least-squares estimator (with all auxiliary regressors) and the restricted least-squares estimator (with no auxiliary regressors), two post-selection estimators based on alternative model selection criteria (the Akaike and Bayesian information criteria), various versions of frequentist model averaging estimators (Mallows and jackknife), and one version of a popular shrinkage estimator (the adaptive LASSO). They discuss confidence intervals for the focus parameter and prediction intervals for the outcome of interest, and conclude that the WALS approach leads to superior confidence and prediction intervals, but only if one applies a bias correction.

WP 21/07

What indicators should be used to monitor the COVID-19 epidemic and the effects of policy interventions, such as lockdowns or other nonpharmaceutical interventions? In "Assessing Alternative Indicators for Covid-19 Policy Evaluation, with a Counterfactual for Sweden”, Giancarlo Spagnolo, Chiara Latour, and Franco Peracchi argue that a single indicator is likely to miss key aspects of the epidemic. Hence, concentrating attention on a single "best'' indicator, as often done in the literature and in policymaking, may be misleading. However, if several alternative indicators all point in the same direction and produce roughly similar results, this provides a much stronger basis for assessing the effects of a policy. To illustrate their argument, they compare different indicators of the spread and consequences of the COVID-19 pandemic, developing a novel method to adjust daily COVID-19 deaths to match weekly excess mortality. Focusing on Sweden, the only country that has good data and did not impose a lockdown, they construct counterfactuals for what would have happened if it had imposed a lockdown, using a synthetic control method. Correcting for data problems and optimizing the synthetic control for each indicator considered, they find stronger effects than previously estimated. Most importantly, studying the ratio of positives to the number of tests, they find that a lockdown would have had sizable effects already after one week. The 3–4 weeks delay highlighted in previous studies appears mainly driven by the large changes in testing frequency that occurred in Sweden during the period considered.

WP 21/06

The evidence on the demographics of COVID-19 fatalities points to an overrepresentation of minorities and an underrepresentation of women, but analyses at the intersection between race and gender have thus far been lacking, due to data availability. In “COVID-19, race, and gender”, Graziella Bertocchi and Arcangelo Dimico use individual-level, race-disaggregated, and georeferenced death data to jointly investigate the racial and gendered impact of COVID-19, its timing, and its determinants. The data are collected by the Medical Examiner of Cook County, IL, the US county that contains the City of Chicago. Through an event study approach they establish that Blacks individuals are affected earlier and more harshly and that the effect is driven by Black women. Rather than comorbidity or aging, the Black female bias is associated with poverty and channeled by occupational segregation in the health care and transportation sectors and by commuting on public transport. Living arrangements and lack of health insurance are instead found uninfluential. The Black female bias is spatially concentrated in neighborhoods that were subject to historical redlining.

WP 21/05

Is bank disintermediation the future of entrepreneurial finance? Regulators around the globe have recently enacted policies to promote new “alternative” funding options for SMEs, for example by facilitating access to public markets. In “Asymmetric Information and Corporate Lending: Evidence from SMEs Bond Markets”, Luana Zaccaria, together with Alessandra Iannamorelli, Stefano Nobili, and Antonio Scalia, examines the rise of public debt markets for SMEs and the role played by information frictions in the choice between market vs bank borrowing. Despite the informational advantage that banks have over capital markets, firms with better privately observed credit quality are more likely to issue bonds. Thus, the evidence suggests positive rather than adverse selection in corporate bond markets. This is consistent with a model where banks offer more flexibility than markets during financial distress and firms use market lending to signal credit quality to outside stakeholders.

WP 21/04

Many financial contracts are structured as asset-sale contracts that embed a repurchase option. Examples of such contracts include repurchase agreements and collateralized debt and date back to ancient pawning arrangements. Why do so many financial contracts take the form of repos? Why don’t borrowers simply sell the assets and buy them back later? In “Repurchase Options in the Market for Lemons”, Saki Bigio and Liyan Shi rationalize the extensive use of repo contracts as a way to resolve the lemons problem. They argue that, in any asset market where asymmetric information is prevalent, repos emerge as a natural market response to adverse selection. Specifically, they study a competitive asset market with asymmetric information. Gains from trade emerge from a liquidity need, but private information about asset quality prevents the full realization of trade. They obtain a unique equilibrium, which features a pooling repo contract and full participation among borrowers. The equilibrium repo contract resolves adverse selection: the embedded repurchase option prevents the market unraveling that occurs in asset-sale markets. However, the contract is inefficient due to cream skimming. Competition to attract high-quality borrowers through the terms of the repurchase option inefficiently lowers liquidity. The equilibrium contract has a closed form and is portable to many applications.

WP 21/03

Noncompete employment contracts, agreements that prohibit employees from joining competing firms for some duration, are prevalent in the U.S. labor market. The anticompetitive effects of such contracts are concerning: restricted labor mobility precludes reallocation of workers to more productive employment and inhibits the entry of new firms. Employers, conversely, argue that noncompete contracts offer the protection they need to carry out investments. In “The Macro Impact of Noncompete Contracts”, Liyan Shi studies the aggregate impact of noncompete employment contracts, assessing the trade-off between restricting worker mobility and encouraging firm investment. She develops an on-the-job search model in which firms and workers sign dynamic wage contracts with noncompete clauses and firms invest in their worker’s general human capital. The incumbent employers use noncompete clauses to enforce buyout payments when their workers depart, ultimately extracting rent from future employers. The model implies that this rent extraction is socially excessive and restrictions on these clauses can improve efficiency. She quantitatively evaluates the model in the managerial labor market, using a novel dataset of executive employment contracts. She finds that the optimal restriction on noncompete duration is close to a ban.

WP 21/02

Rising international migration flows have sparked a heated debate on the effects of immigrants in host societies. Concerns over the effects of immigration are often linked to religion. Although the religious groups that trigger natives’ hostile reactions may differ across time and space, the current animosity is not a new phenomenon. In fact, between 1850 and 1920, when more than 30 million Europeans moved to the United States during the Age of Mass Migration, Catholic immigrants led to similar, hostile reactions. In "Faith and Assimilation: Italian Immigrants in the US", Stefano Gagliarducci and Marco Tabellini focus on the arrival of Italian Catholic churches in the US between 1900 and 1920, combining newly collected Catholic directories on the presence of Italian churches across years and counties with the full count US Census of Population. They find that Italian churches reduced the social assimilation of Italian immigrants, lowering intermarriage rates and increasing ethnic residential segregation. This result does not seem to be explained by either lower effort exerted by immigrants to "fit in'' the American society or increased desire to vertically transmit national culture. Instead, they provide evidence for other two, non-mutually exclusive, mechanisms. First, Italian churches raised the frequency of interactions among fellow Italians, likely generating peer effects and reducing contact with other groups. Second, they increased the salience of the immigrant community among natives, thereby triggering backlash and discrimination.

WP 21/01

In most households, “money chores” such as financial planning and investing are often allocated to the male spouse. In "From Patriarchy to Partnership: Gender Equality and Household Finance", Luigi Guiso and Luana Zaccaria ask whether compliance with this gender norm can induce sub-optimal financial choices, affecting households' welfare.They estimate a novel measure of gender norms on intra-household financial decision making by leveraging dramatic variation across Italian cohorts and regions in the gender of the spouse in charge of household finances that occurred over the last 30 years. Using these estimates to identify the effects of gender parity on household financial decisions, they find that more egalitarian norms increase household participation in financial markets, equity holdings and asset diversification. Egalitarian couples earn higher returns on investments which can raise wealth at retirement up to 15% compared to couples that strictly comply with patriarchal norms. This evidence suggests that gender roles in household financial management can have large economic costs. Consistent with this view, they show that patriarchal norms began receding in the early 1990s, when a pension reform made it too costly to comply with traditional roles.