Highlights 2020

WP 20/20

Mutually beneficial trades often rely on both trust and trustworthiness. In exchanges where no history of behavior is observable, however, where does trust come from? Recent evidence suggests that the level of affinity parties in an exchange feel for each other positively affects trustworthiness and can, therefore, affect trust. In a new working paper, “Trust, Affinity, and Information” Luigi Guiso and Alexey Makarin provide the theoretical framework for thinking about the relationship between trust and affinity and empirically test the resulting theoretical predictions. Specifically, the authors propose a simple model that predicts a positive relationship between trust beliefs, affinity, and trustworthiness and a negative relationship between the trust dispersion and affinity level. The model also suggests that trust should be slower to update after a shock to trustworthiness when affinity is high. The authors then test the model’s predictions using data from two unrelated datasets—a proprietary survey of Italian entrepreneurs and an extensive international survey (Eurobarometer). Using international trade data, they show that, in line with the model, adverse shocks to trustworthiness cause a reallocation of trade from low-affinity to high-affinity partners, especially so in trust-intensive industries.

WP 20/19

In "Risk Sharing within the Firm: A Primer", Marco Pagano provides an overview of what we know and what we don't regarding risk sharing between firms and employees. He starts by asking why such insurance is provided within the firm and what determines its boundaries. He identifies four main constraining factors: availability of a public safety net, moral hazard on the employees’ side, moral hazard on the firms’ side, and workers’ wage bargaining power. These factors explain three empirical regularities: (i) family firms provide more employment insurance than nonfamily firms; (ii) the former pay lower real wages, and (iii) firms provide less employment insurance where public unemployment benefits are more generous. His paper also explores the connection between risk sharing and firms’ capital structure: greater leverage calls for high wages to compensate employees for greater job risk; nevertheless, firms may want to lever up strategically in order to offset the bargaining power of labor unions. Hence, the distributional conflict between shareholders and workers may limit risk sharing within the firm. By contrast, bondholders and workers are not necessarily in conflict, as both are harmed by firms’ risk-taking. In principle, firms may also insure employees against uncertainty about their own talent, but their capacity to do so is constrained by workers’ inability to commit to their employer: in the presence of labor market competition, high-talent employees will leave unless paid in line with their high productivity, making uncertainty about talent uninsurable. He concludes by showing that risk sharing within firms has declined steadily in the last three decades, and by discussing the financial, competitive, technological and institutional developments that may have conjured this outcome.

WP 20/18

Discussion on the disproportionate impact of COVID-19 on African Americans has been at center stage since the outbreak of the epidemic in the United States. In “COVID-19, Race, and Redlining” Graziella Bertocchi, together with Arcangelo Dimico, provides first evidence that race does affect COVID-19 outcomes. They base their analysis on an extraordinarily detailed and so far unexplored source of information on daily deaths from COVID-19 collected by the Medical Examiner's Officer of Cook County, Illinois, the county that hosts the City of Chicago. The data confirm that blacks are overrepresented in terms of COVID-19 related deaths since - as of June 16, 2020 - they are dying at a rate 1.3 times higher that their population share. To understand the reasons behind the higher vulnerability of blacks, the authors exploit information on the georeferenced home address of the deceased and combine the spatial distribution of mortality with the 1930s redlining maps for the Chicago area. They find that, after the outbreak of the epidemic, historically lower-graded neighborhoods display a sharper increase in mortality, which is driven by blacks. Thus, the discriminatory lending practices of the 1930s exert a persistence influence by decreasing the resilience of the black population to the shock represented by the COVID-19 epidemic. This influence is channeled through socioeconomic status and household composition.

WP 20/17

In "The Macroeconomics of Sticky Prices with Generalized Hazard Functions" Francesco Lippi, together with Fernando Alvarez and Aleksei Oskolkov, gives a thorough characterization of a large class of sticky-price models where the firm’s price setting behavior is described by a generalized hazard function. Such a function, originally proposed by the works of caballero and Engel, provides a tractable description of the firm’s price setting behavior and allows for a vast variety of empirical hazards to be fitted. Two main results are established. First, it is shown how to identify all the primitives of the model using the distribution of price changes or the distribution of spell durations. Second, the authors derive a sufficient statistic for the aggregate effect of a monetary shock: given an arbitrary generalized hazard function, the cumulative impulse response to a once-and-for-all monetary shock is given by the ratio of the kurtosis of the steady-state distribution of price changes over the frequency of price adjustment times six.

WP 20/16

In “Divided We Stay Home: Social Distancing and Ethnic Diversity”, Alexey Makarin, together with Georgy Egorov, Ruben Enikolopov, and Maria Petrova, investigates the role of ethnic divisions in community's adherence to voluntary social distancing during COVID-19. On the one hand, as a public good, voluntary social distancing should be more commonplace in more homogeneous and altruistic societies. However, for healthy people, observing social distancing has private benefits, too. If sick individuals are more likely to stay home, healthy ones have fewer incentives to do so, especially if the asymptomatic transmission is perceived to be unlikely. Theoretically, the authors show that this interplay may lead to a stricter observance of social distancing in more diverse and less altruistic societies. Empirically, they show that, consistent with the model, the reduction in mobility following the first local case of COVID-19 was stronger in Russian cities with higher ethnic fractionalization and cities with higher levels of xenophobia. For identification, the authors predict the timing of the first case using pre-existing patterns of internal migration to Moscow. Using SafeGraph data on mobility patterns, they further confirm that mobility reduction in the United States was also higher in counties with higher ethnic fractionalization. The paper's findings highlight the importance of strategic incentives of different population groups for the effectiveness of public policy.

WP 20/15

In “Bitter Sugar: Slavery and the Black Family” Graziella Bertocchi, together with Arcangelo Dimico, investigates the long-term determinants of the African American family structure, which is characterized by a high prevalence of families headed by single women. Their hypothesis is that this structure is rooted in the history of slavery. However, they show that female single headship among blacks is more likely to emerge in association not with slavery per se, but with slavery in sugarcane plantations. The extreme demographic and social conditions prevailing under sugar planting have affected family formation patterns well past the Abolition of slavery in 1865, exerting a persistence influence at least until 1940. The effect of sugar is attenuated starting with the 1920s in connection with the Great Migration of blacks to northern cities. Consistent with the spread of its influence through migration and intermarriage, by 1990 the effect of sugar on the black family is replaced by that of the black share in the population.

WP 20/14

In “The COVID-19 Shock and Equity Shortfall: Firm-level Evidence from Italy”, Marco Pagano, together with Elena Carletti, Tommaso Oliviero, Loriana Pelizzon and Marti Subramanyam, estimate the drop in profits and the equity shortfall triggered by the COVID-19 shock and the subsequent lockdown, using a representative sample of 80,972 Italian firms. They find that a 3-month lockdown entails an aggregate yearly drop in profits of €170 billion, with an implied equity erosion of €117 billion for the whole sample, and €31 billion for firms that become distressed, i.e., ended up with negative book value after the shock. As a consequence of these losses, about 17% of the sample firms, whose employees account for 8.8% of total employment in the sample (about 800 thousand employees), become distressed. Small and medium-sized enterprises (SMEs) are affected disproportionately, with 18.1% of small firms, and 14.3% of medium-sized ones becoming distressed, against 6.4% of large firms. The equity shortfall and the extent of distress are concentrated in the Manufacturing and Wholesale Trading sectors and in the North of Italy. Since many firms predicted to become distressed due to the shock had fragile balance sheets even prior to the COVID-19 shock, restoring their equity to their pre-crisis levels may not suffice to ensure their long-term solvency.

WP 20/13

In "Estimating the prevalence of the COVID-19 infection, with an application to Italy”, Franco Peracchi and Daniele Terlizzese show how to use the available data, in conjunction with minimal, easily interpretable and credible assumptions, to obtain a range of plausible values for the point prevalence of the COVID-19 infection – that is the fraction of people who are currently infected, and can therefore still transmit the infection. The estimates are provided for Italy as a whole and for each of its regions. The method presented offers a quick, albeit less precise, alternative to random testing of the population, which is costly, complicated to carry out, and prone to non- sampling error.

WP 20/12

In “Asocial Capital: Civic Culture and Social Distancing during COVID-19” Luigi Guiso, together with Ruben Durante and Giorgio Gulino, show that areas in Italy with stronger civic capital were quicker to internalize the negative externality of mobility on the spread of contagion and to comply with the restrictions imposed by the Government. They also extend the analysis to Germany, measuring the reduction in the number of deaths that would have been observed if all Italian regions had the highest civic capital.

WP 20/11

In “Interlocking Directorates and Competition in Banking”, Fabiano Schivardi, together with Guglielmo Barone and Enrico Sette, investigates the effects on loan rates of a quasi-experimental change in the Italian legislation which forbids interlocking directorates between banks. The change part of a law passed at the peak of the sovereign debt crisis under the mounting pressure of financial markets and European Institutions and was completely unexpected. Using a difference-in-differences approach and exploit multiple banking relationships to control for unobserved heterogeneity, the authors show that the reform decreased rates charged by previously interlocked banks to common customers by between 10-30 basis points. The effect is stronger if the firm had a weaker bargaining power vis-a-vis the interlocked banks. Consistent with the assumption that interlocking directorates facilitate collusion, interest rates on loans from interlocked banks become more dispersed after the reform. These results indicate that prohibiting IDs can have pro-competitive effects. These findings can therefore inform the policy debate on the (seldom enforced) existing ban in the US and on its possible adoption at the EU level, where IDs are not specifically regulated but rather managed by the general competition law.

WP 20/10

In “Are Executives in Short Supply? Evidence from Deaths' Events”, Fabiano Schivardi and Julinen Sauvagnat use exhaustive administrative data on Italian social security records to construct measures of local labor market tightness for executives that vary by industry and location. Death events have a substantial negative and long-lasting impact on firm performance, but only in thin markets. Following an executive’s death, ROA drops by 1.8 percentage points, against a sample mean of 2%. The effect lasts for three years. These results are robust to both changes in the specification setting -- using a continuous rather than dichotomous indicator of density -- and in the measure of performance -- using productivity instead of ROA. Death events are followed by an increase in the separation rate for the other executives, in particular for those with a college degree. Consistent with the hypothesis that the drop in performance is due to executives' short supply, the authors show that after a death event executives wages in other firms increase, but only in thin markets.

WP 20/09

In “The Macroeconomics of Hedging Income Shares” Adriana Grasso, Juan Passadore and Facundo Piguillem analyze the implications of changing capital and labor income shares for the financial markets. They study a neoclassical growth model with aggregate shocks that affect income shares and financial frictions that prevent firms from fully insuring idiosyncratic risk. They show theoretically that accumulation of safe assets by firms and risky assets by households emerges naturally as a tool to insure income shares' risk. They also show how aggregate risk sharing is distorted by the combination of idiosyncratic risk and moving shares. Finally, they calibrate the model to the U.S. economy and find that low rates, rising capital shares, and accumulation of safe assets by firms and risky assets by households can be quantitatively rationalized by persistent shocks to the labor share.

WP 20/08

In "Disaster Resilience and Asset Prices", Marco Pagano, together with Christian Wagner and Josef Zechner, investigate whether security markets price the effect of social distancing on firms’ risk. The authors make three distinct contributions. First, they investigate whether the COVID-19 outbreak triggered a different stock return response depending on companies' resilience to social distancing, which is the most severe constraint imposed on firms' operations. On this score, they find that more resilient companies greatly outperformed less resilient ones, even after controlling for all conventional measures of risk premia. Second, they explore whether similar cross-sectional return differentials already emerged before the COVID-19 outbreak. Indeed this is the case: in the 2014-19 interval, the cumulative return differential between more and less pandemic-resilient firms is of about the same magnitude as during the outbreak, i.e. between late February and early April 2020. This can be interpreted as evidence of learning by investors, i.e., of growing awareness of the potential threat posed by pandemics well in advance of its materialization. Finally, the authors use option price data to infer whether, since the COVID-19 outbreak, investors price pandemic risk over different horizons, and find that they do: even on a 2-year horizon, stocks of more pandemic-resilient firms are expected to yield significantly lower returns than less resilient ones, reflecting lower exposure to disaster risk. Hence, going forward, markets appear to price exposure to a new risk factor, namely, pandemic risk.

WP 20/07

In "Careers in Finance", Marco Pagano, together with Andrew Ellul and Annalisa Scognamiglio, assess the attractiveness of employment in the finance industry relative to manufacturing and high tech, and how it has changed over time. Employees in finance are known to earn higher wages and returns to talent than non-finance workers since the 1990s, suggesting that finance may have attracted talent at the expense of other industries. However, the allocation of talent is likely to respond to differences in career paths across industries ( i.e. in the resulting value and riskiness of human capital), not in wages at a given date. The authors analyze the careers of 9,964 individuals from 1980 to 2017 based on their resumes, and find that they tend to remain in the same industry for most of their working lives, consistently with them choosing occupations based on comparisons of entire career paths. Comparing various aspects of careers –- levels, slopes, PDV and risk of pay profiles -– finance as a whole appears to offer a career premium compared to manufacturing and high tech, through higher and steeper pay profiles. This however masks significant diversity within finance: while asset managers enjoy a large career premium and no commensurate career risks, the opposite applies to banking and insurance employees. Furthermore, relative to manufacturing, the asset management career premium has risen for cohorts entering soon before and during the financial crisis, even after controlling for career risk, while the high-tech career premium has become commensurately large for the latest cohorts.

WP 20/06

In "Aggregate Risk or Aggregate Uncertainty? Evidence from UK Households" Claudio Michelacci and Luigi Paciello show that UK households with preferences for higher inflation and higher interest rates have lower expected inflation. They show that the relationship between preferences and expectations tightens after major economic events such as the failure of Lehman Brothers or the Brexit referendum. If households had treated uncertainty as measurable risk, consumption and output would have been around 1 percent higher both during the Great Recession and in recent years.

WP 20/05

In “A Simple Planning Problem for COVID-19 Lockdown” Francesco Lippi, together with Fernando Alvarez and David Argente finds that, conditional on an initial 1% fraction of infected agents and no cure for the disease, the optimal policy requires a near-total economic lockdown (with 75% of activities shut down) for about 5 weeks, followed by a fast relaxation. The quantitative result are useful to gauge what parameters of the problem are important in shaping the intensity and duration of the optimal lockdown policy.

WP 20/04

In “The Optimal COVID-19 Quarantine and Testing Policies” Facundo Piguillem and Liyan Shi analyze whether the intensity and the duration of the implemented quarantines to contain the spread of COVID-19 make sense when one considers the high costs in lost production. They find that the observed policies are very close to a simple welfare maximization problem of a planner who tries to stop the diffusion of the disease. These extreme measures seem optimal in spite of the high output cost that it may have in the short run, and for various curvatures of the welfare function. The desire for cost smoothing reduces the intensity of the optimal lockdown while extending it for longer, but it still amounts to reducing economic activity by a half. They then analyze the possibility of either complementing or substituting the lockdown policy with massive random testing. They argue that, under some preliminary estimates of the cost of testing, testing could generate sizeable welfare gains and almost eliminate the need for indiscriminate quarantines.

WP 20/03

In "Sampling properties of the Bayesian posterior mean with an application to WALS estimation”, Giuseppe De Luca, Jan Magnus and Franco Peracchi explore the finite sample properties of weighted average least squares (WALS). In a frequentist setting, the sampling accuracy of this and other classes of learning methods that rely on Bayesian ideas is typically assessed using the variance of the posterior distribution of the parameters of interest given the data. This may be permissible when the sample size is large because, under the conditions of the Bernstein–von Mises theorem, the posterior variance agrees asymptotically with the frequentist variance. In finite samples, however, things are less clear. To explore this issue, the Authors first consider the frequentist properties (bias and variance) of the posterior mean in the important case of the normal location model, which consists of a single observation on a univariate Gaussian distribution with unknown mean and known variance. Based on these results, they derive new estimators of the frequentist bias and variance of the WALS estimator in finite the proposed estimators by a Monte Carlo experiment with design derived from a real data application about the effect of abortion on crime rates.

WP 20/02

In “K-Returns to Education”, Luigi Guiso (with Fagereng, A., Holm, M., and L. Pistaferri) analyses whether formal general education pays off in capital markets as it does in the labor market. Using a compulsory school reform in Norway to obtain exogenous variation in years of schooling, the authors show that, while education predicts returns to wealth in OLS estimates, it has no casual effect in IV regressions or when unobserved heterogeneity is taken care of using a twins design. General education predicts returns to wealth only because it is correlated with ability and risk tolerance, and the latter seem to be the relevant drivers of heterogeneity in individual returns on capital. This is at odds with the evidence on labor earnings, where general education has a casual effect on returns. Where does this asymmetry come from? According to the authors, one possibility is that general education matters for labor earnings because it signals ability, which is relevant in the labor market, but clearly irrelevant for returns on self-managed wealth.

WP 20/01

In “The Insurance Role of the Firm” Luigi Guiso, with Luigi Pistaferri, reviews the recent literature on the risk sharing role of the firm and provides a framework for studying risk sharing between workers and firm owners vis-à-vis firms specific shocks of different nature. The authors show how this framework can be taken to the data to provide estimates of the extent of insurance within the firm. Their estimates from a large number of Western countries strongly support the view that in capitalist economies the firm is a large, albeit far from complete, wage insurance instrument. They quantify the welfare benefits of firm-provided wage insurance, show evidence on how workers react to firms shocks passed through wages, and discuss the future role of the firm as a wage insurance provider.