“The Effect of Job Flexibility on Female Labor Market Outcomes: Estimates from a Search and Bargaining Model” (with A. Moro)

IZA Discussion Paper 4829, 2010; Journal of Econometrics, Revise and Resubmit

Abstract:

This paper develops and estimates a searchmodel of the labor market where jobs are characterized by wages and work-hoursflexibility. Flexibility is valued by workers, and is costly to provide foremployers. The model generates observed wage distributions directly related tothe preference for flexibility parameters: the higher the preference forflexibility, the wider is the support of the wage distribution at flexible jobsand the larger is the discontinuity between the wage distribution at flexibleand non-flexible jobs. Estimation results show that more than one third ofwomen place positive value to flexibility, with women with a college degreevaluing flexibility more than women with a high school degree. Counterfactualexperiments show that flexibility has a substantial impact on the wagedistribution but not on the unemployment rate. We comment on the implicationsof our approach for gender differential in wages and schooling.

JEL:J30; C5

Keywords: Search Models; Work-Hours Flexibility; Structural Estimation.

“Household Search or Individual Search: Does it Matter? Evidence from Lifetime Inequality Estimates.” (with J. Mabli)

Work in Progress, 2010

Abstract:

The standard study of earnings inequality builds inequality indexesbased on a cross-section of individual earnings. While this information isuseful, it only offers an incomplete picture of the extent and the impact ofinequality. First, it ignores that individuals may move between differentpoints of the cross-sectional earnings distribution over the course of theirlife. Second, it ignores that most individuals live in households that poolresources.

Inthis paper we attempt to remove these two limitations by developing andestimating a household search model of the labor market. This procedure allowsus to build lifetime inequality measures taking into account earnings mobilityacross the cross-sectional distribution over time and to construct inequalitymeasures at the household level and not only at the individual level. Inallowing for household search we also provide a contribution to the labormarket search literature since the overwhelming majority of estimated searchmodels ignores the existence of the household as a crucial unit of labor marketdecisions.

Weestimate the model on the Survey of Income and Program Participation (SIPP)since it provides accurate information about the determinants of the income ofindividual households in the US.We find that while cross-sectional household earning inequality increasedsignificantly over the 1990s and early 2000s, there was a slower increase inlifetime household inequality over the same period. This difference is morepronounced when we look at individuals and interestingly it is differentbetween men and women: lifetime inequality increased more than cross-sectionalinequality for wives while the opposite was true for husbands.

"Sources of Earnings Inequality: Estimates from an On-the-Job Search Model of the U.S. Labor Market" (with M. Leonardi)

European Economic Review, 54(6): 832-854, 2010

Abstract:

Since the early 1980s the labor market inthe United Stateshas seen a substantial increase in earnings dispersion. We study the issue bydeveloping an on-the-job search model of the US labor market that allows forwage and employment mobility as a result of optimal individual behavior. Weestimate its structural parameters on PSID data at different points in time toclarify the sources of the evolution of earnings inequality and instabilitybetween 1987 and 1996. This procedure allows to: compute lifetime measure ofinequality on top of the usual cross-sectional measure of inequality andprovide conterfactual experiments that evaluate the contribution of differentparameters to changes over time by taking into account some equilibriumeffects. We find that the increase in lifetime inequality and incross-sectional inequality have been generated by different sources and thatthese sources are different by skills: changes in the wage offer distributionare the main determinant of the increase in inequality for skilled workerswhile both mobility changes and wage offer distribution changes are needed toexplain changes for the unskilled.

Keywords: Earnings Inequality; EarningsInstability; On-the-job search; Structural estimation

JEL Nos: J21, J31

"Gender Discrimination Estimation in a Search Model with Matching and Bargaining"

International Economic Review, 51(3): 745-783, August, 2010

Abstract:

Gender wage differentials, conditional onobserved productivity characteristics, have been considered a possible indicationof prejudice against women in the labor market. However, there is no conclusiveevidence on whether these differentials are due to labor market discriminationor to unobserved productivity differences. The objective of this paper is topropose a solution for this identification problem by developing and estimatinga search model of the labor market with matching, bargaining and employers'taste discrimination. In equilibrium all types of employers wage discriminatewomen: prejudiced employers because of preference and unprejudiced employersbecause of spillover effects that worsen the bargaining position of women.Estimation is performed by maximum likelihood on Current Population Survey datafor the year 1995. Results indicate that the productivity of women is 6.5%lower than the productivity of men and that about half of the employers areprejudiced against women. Three policy experiments are implemented using theestimated parameters: an equal pay policy, an affirmative action policy and awage differential decomposition that takes into account equilibrium effects.

JEL:C51; J7; J64

Keywords: Gender Differentials; Discrimination; Search Models;Maximum Likelihood Estimation; Structural Estimation; Affirmative Action.

"Intergenerational Mobility and Schooling Decisions"

IZA Discussion Paper 2876, 2007 (with D. Checchi)

Abstract:

Intergenerational mobility in income andeducation is affected by the influence of parents on children's school choices.Our focus is on the role played by different school systems in reducing ormagnifying the impact of parents on children's school choices and therefore onintergenerational mobility in general. We compare two apparently similareducational systems, Italyand Germany,to see how the common feature of the presence of separate track at SecondarySchool level may produce different impacts on children destinies. Using datafrom cross-country surveys (PISA 2003), we study the impact of parentaleducation on track choice, showing that the greater flexibility of the Italiansystem (where parents are free to choose the type of track) translates intogreater dependence from parental background. These effects are reinforced whenmoving to post-secondary education, where the aspiration to go to college isaffected not only by the school type but also (in the case of Italy only) byparental education. We then move to country-specific data sets to study actualcollege choices in nationally representative sample, and we find that parentaleducation still play an independent role on student choices.

"Prejudice and Gender Differentials in the U.S. Labor Market in the Last Twenty Years"

Journal of Econometrics, , 156, 190-200, 2010

Abstract:

Earnings differentials between men and womenhave experienced a stable convergence during the 1980s, following a processstarted in the late 1970s. However, in the 1990s the convergence has almoststopped. The first objective of the paper is to evaluate if discrimination,defined as explicit prejudice, may have a role in explaining this slowdown in the converge. The second objective is to assess whether theprediction of a decrease in the proportion of prejudiced employers implied bythe Becker's model of taste discrimination is taking place and if so at whatspeed. These objectives are achieved by developing and estimating a searchmodel of the labor market with matching, bargaining, employer's prejudice andworker's participation decisions. The results show that the proportion ofprejudiced employers is estimated to be decreasing at an increasing speed,going from about 69% in 1985 to about 32% in 2005. Therefore prejudice does notseem a relevant factor in explaining the slower convergence between male andfemale earnings in the 1990s. The results are consistent with the Becker'smodel of taste discrimination if one is willing to assume a very slowadjustment process.

''Returns to Education in the Economic Transition: A Systematic Assessment Using Comparable Data'' (with S. Paternostro and E. Tiongson)

Economics of Education Review, 27, 724-740, 2008

Abstract:

Thispaper studies a sample of economies in transition to verify the assertion thatreturns to schooling increase as an economy transitions to a marketenvironment. This claim has been difficult to assess in the past as theempirical evidence so far has covered only a few countries over short timeperiods. A number of studies find that returns to education increased from the“pre-transition” period to the “early transition” period; it is not clear whathas happened to the skills premium through the late 1990s, or the periodthereafter. We implement a more systematic analysis using data comparableacross countries and over time to estimate returns to schooling in eighttransition economies (Bulgaria,Czech Republic,Hungary, Latvia, Poland,Russia, Slovak Republicand Slovenia).The analysis covers the early transition period up to 2002; in the case of Hungary, wecapture the transition process more fully, beginning in the late 1980s.Compared to the existing literature, we also implement more comprehensiverobustness checks on the estimated returns, although at best we offer only anincomplete solution to the problem of endogeneity. We find that the evidence ofa rising trend in returns to schooling over the transition period is weak.There are, however, significant differences in returns across countries. Thesedifferentials have remained roughly constant over the last fifteen years. Wespeculate on the likely institutional and structural factors underpinning theseresults.

JEL: J31, P23

Keywords: returns toschooling, transition economies

"Taste Discrimination in a Search Model with Free-Entry"

Work in Progress, 2004

Abstract:

Becker's [1971] model of tastediscrimination has been very influential but also strongly criticized because,as Arrow [1973] put it, the ``model predicts the absence of the phenomenon itwas designed to explain.'' The issue is that only the least discriminatoryemployers survive in the long run, eventually leading to the disappearance ofdiscrimination. Recently, labor market models with search frictions havestarted to be used to address this problem. The objective of this paper is toshow if prejudiced and unprejudiced employers can both survive in a labormarket with search friction, matching, bargaining and free-entry of employers.Two types of modelling strategies may generate outcomes in which differenttypes of employers survive. In the first, employers are ex-ante homogenous andthe disutility from hiring women is match specific. Inthe second, workers direct search toward prejudiced or unprejudiced employers,then are randomly matched to employers within the specific type chosen.

''Productivity, Seniority, and Wages: new evidence from personnel data'' (with A. Ichino)

Labour Economics, 8, 359-387, 2001

Abstract:

Wages may be observed to increase with seniority becauseof firm-specific human capital accumulation or because of self-selection ofbetter workers in longer jobs. In both these cases the upward sloping wageprofile in cross sectional regressions would reflect higher productivity ofmore senior workers. If this were true, the observation of an effect ofseniority on wages would depend on the presence of controls for individualproductivity. In this paper we replicate, using personnel data from a largeItalian firm, the results of the pioneering work of Medoff and Abraham (1980and 1981) in which supervisors' evaluations were used as productivityindicators. Since the validity of supervisors' evaluations as measures ofproductivity has been widely criticised, we extend the work of Medoff andAbraham using different direct measures of productivity based on recordedabsenteeism and misconduct episodes. Both these indicators and supervisors'evaluation suggest that the observed effect of seniority on wages does notreflect a higher productivity of more senior workers. Only at the lowest levelsof the firm's hierarchy, the human capital theory contributes to explain theeffect of seniority on wages. At least at all other levels, the explanation ofthe observed upward sloping profile has to be based on theories in which wagesare deferred for incentive or insurance reasons.

JEL: J30, J31

Keywords: Wage determination, Returnsto Seniority, Productivity.

"The Transmission Mechanism of Monetary Policy in Europe: Evidence from Banks' Balance Sheets" (with C.A. Favero and F. Giavazzi)

NBER Working Paper, W7231, July, 1999

Abstract:

Available studies on asymmetries in the monetarytransmission mechanism within Europe areinvariably based on macro-economic evidence: such evidence is abundant butoften contradictory. This paper takes a different route by using micro-economicdata. We use the information contained in the balance sheets of individualbanks (available from the BankScope database) to implement a case-study on theresponse of banks in France,Germany, Italy and Spain to a monetary tightening. Theepisode we study occurred during 1992, when monetary conditions were tightenedthroughout Europe. Evidence on such tighteningis provided by the uniform squeeze in liquidity, which affected all banks inour sample. We study the first link in the transmission chain by analysing theresponse of bank loans to the monetary tightening. Our experiment providesevidence on the importance of the "credit" channel in Europe, and thus on one possibly important source ofasymmetries in the monetary transmission mechanism. We do not find evidence ofa significant response of bank loans to the monetary tightening, which occurredduring 1992, in any of the four European countries we have considered. However,we find significant differences both across countries and across banks ofdifferent dimensions in the factors that allow them to shield the supply ofloans from the squeeze in liquidity.

JEL: E51, E52, G21

"Returns to Schooling in Italy: OLS, IV and Gender Differences"

Bocconi University Working Paper, 1, 1999

Abstract:

The conventional wisdom inthe literature is that OLS estimates of returns to schooling are biased andinconsistent due to endogeneity in the schooling variable (Griliches 1977). Arecent solution to this endogeneity problem has been found in identifyingexogenous sources of variation in schooling to build a new set of instrumentalvariables for years of education attained (Angrist and Krueger 1991; Card1998).

The aim of this paper isto apply new instrumental variables estimation to Italian data using the Bankof Italy data set Indagine sui bilancidella famiglie nell’anno 1991. IVestimation results suggest a downward bias in OLS estimates. Moreover theextent of the bias is asymmetric in regard to gender: OLS estimates indicatethe usual hierarchy - i.e. female returns higher than male returns (Lucifora1994; Blau-Kahn 1992) - whereas IV estimates suggest an opposite hierarchy inthe point estimates.

JEL: J 31

Keywords: human capital, returns to education, earnings.

In Italian:

"La Scelta della Scuola Secondaria in Italia"

Rivista di Politica Economica, 7/8, 2001

Abstract:

The objective of this paper is to evaluatethe determinants of Secondary Schooling choices for young Italians, focusing inparticular on the impact of family background. The problem is relevant becauseSecondary schooling in Italyis very important in terms of future performance in the labor market and toaccess highest level of education. This problem is assessed by estimating aMultinomial Logit Model on secondary schooling choice on Bank of ItalyHousehold Survey data for the year 1998. Thanks to some post-estimationcalculation is possible to obtain upper and lower bounds of the marginaleffects of relevant regressors on the probabilities to choose the variousalternatives. On the preferred specification the marginal effect of an increaseof parents' schooling is always positive on the probability of choosing a Liceowhile it is always negative on the probability of not acquiring additionalschooling. Effects on the other alternatives (i.e. on the choice of other typesof 5-years High School or on the choice of a 3-years High School)are mixed. These results point to a Secondary Schooling system that tends toperpetuate schooling stratification across generations.

''Dieci anni di contrattazione aziendale in Lombardia: 1986-1995'' (with D. Checchi)

Lavoro e Relazioni Industriali, 2, 1999

Abstract:

Dopo gli accordi del Luglio 1992 e 1993 il tema della struttura dellacontrattazione in Italia ha ricevuto molta attenzione come uno dei possibilistrumenti di controllo della dinamica salariale. Tuttavia è carente laconoscenza degli effettivi meccanismi che conducono alla stipulazione diaccordi tra le controparti. In questo lavoro ci proponiamo di utilizzare i datia nostra disposizione per indagare il processo negoziale tra impresa, sindacatoe lavoratori a livello aziendale. Per questo abbiamo scelto di sintetizzare levariabili relative al contesto e alle modalità di relazione tra le parti, perpoi cercare di predire i risultati del processo negoziale tramite questiindicatori. Interessante è valutare se questi indicatori possano costituire unadeterminante complementare o alternativa a quella in genere sottolineata inletteratura, tipicamente la dimensione d’impresa.