Turbulence and the employment experience of older workers
Abstract
EconStor is a publication server for scholarly economic literature, provided as a non-commercial public service by the ZBW.
Full text
Lalé, Etienne Article Turbulence and the employment experience of older workers Quantitative Economics Provided in Cooperation with: The Econometric Society Suggested Citation: Lalé, Etienne (2018) : Turbulence and the employment experience of older workers, Quantitative Economics, ISSN 1759-7331, The Econometric Society, New Haven, CT, Vol. 9, Iss. 2, pp. 735-784, https://doi.org/10.3982/QE557 This Version is available at: https://hdl.handle.net/10419/217114 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by-nc/4.0/
Quantitative Economics 9 (2018), 735–784 1759-7331/20180735 Turbulence and the employment experience of older workers Etienne Lalé Department of Economics, Université du Québec à Montréal, CIRANO, and IZA This paper provides new interpretations of the effects of rising economic turbulence—an increase in the rate of skill depreciation upon job loss—and its interaction with labor market institutions. We have three main results, based on a life-cycle model with labor market frictions and labor force participation decisions. First, rising economic turbulence during the 1970s and 1980s accounts for the decline in employment among older workers in the United States. Second, the interaction between turbulence and institutions explains most of the reduction in labor force participation among older workers in Europe over this period, but ultimately explains little of the rise in unemployment. Third, only a small share of the increase in unemployment can be attributed to the early retirement policies that were implemented in Europe from the 1970s up until the early 1990s. Our analysis indicates that incorporating an operative labor supply choice can pose serious challenges to theories aiming to explain the European unemployment problem. Keywords. Job search, job loss, turbulence, European unemployment, labor force participation. JEL classification. E24, J21, J64. 1. Introduction The outbreak and persistence of high European unemployment since the 1970s compared with the dynamism of the U.S. labor market has sparked a large body of research over the past few decades. In his appraisal of this literature, Blanchard (2006) reached mixed conclusions about the results obtained so far. On the positive side, there are convergent findings pointing to the interaction between shocks and labor market institutions as a key explanation of the transatlantic employment gap. Meanwhile, on the negative side, data accumulated over time highlight the heterogeneity of situations and of trajectories across workers. This poses a challenge to virtually any explanation of the U.S.–Europe employment gap, that it should be simultaneously consistent with the heterogeneous employment patterns found in disaggregated data. The recent literature emphasizes the life cycle as one such major source of heterogeneity (Ljungqvist and Sargent Etienne Lalé: [email protected] This paper supersedes an earlier version circulated under the title “Skill Obsolescence, Discouraged Workers, and the Aggregate Labor Market.” A supplemental file with additional information is available at the web address: http://etiennelale.weebly.com/uploads/1/0/1/6/101665630/lssupplement.pdf. I am grateful to Pierre Cahuc, Grégory Jolivet, Dirk Krueger, Fabien Postel-Vinay, Karl Schmedders, Hélène Turon, Etienne Wasmer, and seminar participants at numerous institutions for useful suggestions on this project. I would also like to thank four anonymous referees for their constructive comments that helped to improve the paper. All errors are my own. ©2018 The Author. Licensed under the Creative Commons Attribution-NonCommercial License 4.0. Available at http://qeconomics.org.https://doi.org/10.3982/QE557
736 Etienne Lalé Quantitative Economics 9 (2018) (2008), Chéron, Hairault, and Langot (2009), Prescott, Rogerson, and Wallenius (2009), and Kitao, Ljungqvist, and Sargent (2017)). A related issue, which has received little attention to date, is that the contributions of unemployment and labor force participation to employment differences change over the life cycle. Hence, in addition to having the correct life-cycle implications for the identities of the nonemployed, a proper account of transatlantic employment experiences should also be consistent with the role played by those different margins of nonemployment. This paper takes a step in this direction, providing an analysis of the employment experience of older workers on both sides of the Atlantic. We develop a life-cycle model with a frictional labor market and an operative labor supply margin, wherein shocks interact with institutions in ways that deteriorate employment. We use the model to offer new interpretations of the employment effects of shocks and institutions, and the interactions between the two. First, we account for secular changes in the U.S. employment rate of male workers. Usually these changes are overshadowed by the attention to the unemployment rate, which has remained stable in the U.S. in the long run.1Second, we study the decline in European employment rates, and in doing so we clarify whether shocks and institutions explain the upward trend in unemployment, the downward trend in labor force participation, or a combination of the two. Third, we draw attention to one specific labor market institution that has changed over time, namely programs aimed at fostering early retirement.2These programs have been used in Europe to reduce labor force participation before normal retirement age, often with a “lump-oflabor” view of the relationship between older worker employment and unemployment among younger workers. The model enables us to quantify the implications of this relationship. Key facts of interest for the paper are depicted in Figure 1. The solid line shows the employment rate of older male workers in the three largest countries in continental Europe (France, Germany, Italy) and the U.S.3The dashed line shows an alternative employment rate, which has been calculated by holding the unemployment rate of older workers constant. As can be seen, employment among older workers has fallen secularly, and this decline is predominantly explained by labor force participation, that is, the dashed line closely tracks the solid line. The other salient fact in Figure 1is that the dynamics of older worker employment are qualitatively similar in the U.S. and Europe and differ only quantitatively. We complement these facts in three ways in Section 2.First, within each country these changes have a sizable impact on the aggregate employment rate. Second, across countries labor force participation accounts for a large fraction of 1This holds for the unemployment rate of men as well as for the unemployment rate of both men and women. Another reason why changes in the employment rate of U.S. men tend to be overlooked is that the aggregate employment-to-population ratio has remained stable as a consequence of the increase in female employment. In this paper, we focus on understanding the specific dynamics of male employment. We think the secular employment experience of women deserves a study in its own right, given the stark contrast with the employment experience of men. 2We provide an overview of the main trends in early retirement policies in Section 6of the paper. 3The facts shown in Figure 1hold true for a larger set of European countries. We present similar time series for Spain, Portugal, Norway, and Sweden and can be found on the journal website: http://qeconomics. org/supp/557/supplement.pdf.
Quantitative Economics 9 (2018) Turbulence and older workers 737 Figure 1. Actual and counterfactual employment rates of older male workers. Notes:Datafrom the OECD labor force database for male workers aged 55 to 64; Germany refers to West Germany prior to 1991; see Appendix A.1 for details. In each plot, the solid line is the actual employment rate while the dotted line shows the counterfactual series that holds the unemployment rate fixed to its mean value over the sample period. the differences in aggregate male employment. Third, the separation between the two nonemployment margins matters because the odds of regaining employment from unemployment rather than from nonparticipation are much higher at older ages. We draw on various sources to construct a model that relates to the trends shown in Figure 1. The first one of those is the notion of economic turbulence proposed by Ljungqvist and Sargent (1998, 2008). Rising economic turbulence refers to an increase in the rate of skill depreciation upon job loss. This phenomenon captures the microeconomic effects of changes in the macro-environment, such as restructuring from manufacturing to the service industry or new information technologies. Thus it can aptly describe the type of shocks that have the potential to shift the steady-state equilibrium of the labor market. Next, as in the canonical framework of Mortensen and Pissarides (1994) our model features match productivity shocks that generate job destruction. Job creation is also endogenous. There is a single matching function, and hence
738 Etienne Lalé Quantitative Economics 9 (2018) firms cannot direct their vacancies toward specific groups of workers, such as, for example, younger workers. But the probability of being hired is not uniform across workers; it varies strongly with their individual characteristics, namely age, human capital, and welfare benefits. Last, the model embodies idiosyncratic, autocorrelated shocks to the value of being out of the labor force. Garibaldi and Wasmer (2005) used a similar assumption to generate endogenous movements in labor force participation, albeit in a much simpler setting. To our knowledge, the model we propose is the first to depart from a two-state abstraction (employment/nonemployment) to discuss the effects of the interaction between shocks and institutions. The analysis proceeds with a series of numerical experiments based on two calibrated models. Following Ljungqvist and Sargent (1998, 2008), we study the U.S. employment experience through the lens of a laissez-faire economy. We use this economy to measure changes in the degree of economic turbulence, by matching the 1970s–1980s increase in U.S. earnings instability highlighted by Gottschalk and Moffitt (1994, 2009). We also use a welfare state economy to describe labor markets in Europe, focusing on changes in policies that provided incentives toward early retirement before the 1990s. Most parameters (e.g., preferences, human capital) are common across the laissez-faire and welfare state economies, and are informed by the behavior of the U.S. labor market at the onset of the 1970s. Two idiosyncratic technology parameters (in addition to the welfare-state policy parameters) capture U.S.–Europe differences in unemployment and labor force participation in the initial steady state. The crux of our analysis is the evolution of equilibrium allocations in the U.S. and Europe, respectively, as we move away from the 1970s up until the 1990s. The first set of experiments analyzes the effects of the measured change in economic turbulence. We find that this process explains the decline in employment among older workers in the U.S., as it accounts quantitatively for the long-run reduction of their labor force participation. Over this period, we also find that rising economic turbulence explains the European decrease of a twice larger magnitude in labor force participation among older workers. The interaction between shocks and institutions per se accounts for about half of this effect.4Last, rising economic turbulence explains little of the increase in unemployment that coincided with the aforementioned changes in Europe. The main economic forces driving these results are as follows. First, workers whose skills depreciate upon job loss have poorer employment prospects. In a laissez-faire economy with only employment and unemployment states, these workers would “bite thebullet”andreturntoemploymentatlowerwages(Ljungqvist and Sargent (1998, 2008)). The additional option of moving to nonparticipation mitigates this effect in our model, and thereby explains the evolution of employment in the U.S. Welfare benefits and stringent employment protection amplify the employability problem of workers 4If we remove the difference in technology parameters between the two economies, we find that the decrease in labor force participation in the laissez-faire economy is almost 50 percent higher than under the baseline. The remaining gap is explained by the interaction between shocks and the labor market policies of the welfare state economy.
Quantitative Economics 9 (2018) Turbulence and older workers 739 whose skills depreciate in Europe. They become detached from the labor market and drop into nonparticipation instead of staying in the unemployment pool. Second, older workers are over-represented among workers moving to nonparticipation. Skill depreciation falls more heavily on older workers because they have accumulated more human capital. But a perhaps more fundamental reason is the “horizon effect” analyzed by Chéron, Hairault, and Langot (2009, 2013). From an employer’s perspective, the returns to hiring a worker close to retirement are lower because of the expected shorter duration of the match. From a worker’s perspective, the returns to staying in the labor force are lower because of the expected shorter duration of job search. These forces coalesce to make older workers choose nonparticipation over the other labor market states. The second set of experiments considers shifts in early retirement policies as an additional source of employment changes over time.5Our welfare-state economy is actually too stylized to model explicitly the numerous policies that provide exit routes to retirement.6Meanwhile, the model allows us to explore several polar cases to get a quantitative sense of the nature and magnitude of the effects of those policies. We find, first of all, that the policies considered have a perverse impact on aggregate employment. While leading to an almost one-for-one substitution between nonparticipation and unemployment among older workers, they contribute to unemployment at younger ages. This speaks strongly against the once popular idea that early retirement could be helpful to “make room for the young.”7Second, in quantitative terms early retirement policies generate little additional unemployment. In particular, although the implementation of these policies coincided with the 1970s–1980s increase in turbulence and continued at least until the early 1990s, this trend cannot reconcile our welfare state economy with the outbreak of high European unemployment. As noted in the opening sentence, there is a vast literature on employment differences between the U.S. and Europe.8Within this body of research our study is more 5Much of the literature considers the interaction between time-varying shocks and time-invariant institutions; see Blanchard and Wolfers (2000). This view is not undisputed, however. Nickell, Nunziata, and Ochel (2005) pointed out that some institutions have evolved in response to shocks in ways that sometimes aggravated the initial impact of the shocks. Early retirement policies seem to fit this description well: in addition to reducing unemployment numbers directly, several reforms were enacted with the objective of releasing jobs for the young in the new era of high unemployment (Ben Salem, Blanchet, Bozio, and Roger 2010). 6Consider for instance early retirement benefits and disability insurance benefits—two oft-cited examples of policies toward early retirement. To correctly analyze the effects of financial incentives, we would need a model where agents have a finite intertemporal elasticity of substitution and have access to savings. And to study disability insurance benefits, we would need a model that includes medical expenditures, health status, and health shocks. 7Over time, this idea has clearly (albeit slowly) lost ground. Starting in the early 1990s, new reforms were enacted in Europe in an attempt to reverse the trend and increase labor force participation at older ages; see Section 6. Running parallel to this trend, the idea that the efficient policy response is to raise the retirement age has gradually gained support. Hairault (2010) demonstrated this point both empirically and through the lens of a quantitative model. 8See, among others, Bertola and Ichino (1995), Marimon and Zilibotti (1999), Mortensen and Pissarides (1999), den Haan, Haefke, and Ramey (2005), Hornstein, Krusell, and Violante (2007), and the contributions by Ljungqvist and Sargent referenced in the paper.
740 Etienne Lalé Quantitative Economics 9 (2018) directly related to Ljungqvist and Sargent (2008), Chéron, Hairault, and Langot (2009), and Kitao, Ljungqvist, and Sargent (2017). These papers analyze the age structure of the U.S.–Europe employment-nonemployment gap through the lenses of heterogeneousagent life-cycle models. We add to this research by explicitly separating unemployment from nonparticipation. We analyze these margins empirically, and then within a quantitative model with endogenous worker transitions between the three labor market states (employment, unemployment, nonparticipation). We set up this model in general equilibrium for two reasons. First, this simplifies the calibration process because more variables are determined endogenously. For instance, wages (and hence the effects of skills on earnings) are endogenous to the model. Second, we use this framework to study the aggregate effects of policies that interact with labor force participation choices. While these policies are targeted at older workers, the effects may spill over on workers in other age groups. This paper contributes more broadly to research that aims at developing macromodels with labor market frictions and a labor force participation margin. Some examples of this strand of literature include Garibaldi and Wasmer (2005), Pries and Rogerson (2009), Shimer (2013), Krusell, Mukoyama, Rogerson, and ¸Sahin (2011, 2017), and Mankart and Oikonomou (2017). In contrast to these papers, we propose a model with many layers of worker heterogeneity (welfare benefits, skills, taste for leisure, age) so as to study the relationship between economic turbulence and labor market institutions. Our model yields a rich set of implications regarding the relationship between observable characteristics (either aggregate or individual) and worker flows across employment, unemployment, and nonparticipation. Thus, it offers a relevant theoretical framework to analyze why these worker flows are so different across countries (Elsby, Hobijn, and ¸Sahin (2013)), and why they are so volatile over the life cycle (Choi, Janiak, and Villena-Roldán (2015)). The rest of the paper is organized as follows. Section 2documents the empirical facts of interest for the paper. Section 3presents the model economy used to interpret these facts. We calibrate the model in Section 4and discuss important model-generated outcomes in Section 5. The main results are presented in Section 6: it contains two sets of numerical experiments that study the effects of rising economic turbulence and of the changes in early retirement policies. Section 7concludes. 2. Some facts The main facts on U.S.–Europe unemployment differences are well known and thoroughly documented in the literature. In a nutshell, while unemployment in the U.S. has been stable over the past decades, it increased in Europe at the end of the 1970s and has remained persistently high since then as a result of low job-finding rates (Layard, Nickell, and Jackman (2005), Machin and Manning (1999), Blanchard (2006), Rogerson and Shimer (2011)). The goal of this section is to present several additional facts that have, until now, been overshadowed somewhat by the emphasis on studying the unemployment rate. These facts relate to the behavior of labor force participation during the working life cycle and its contribution to aggregate employment differences over time and across countries.
Quantitative Economics 9 (2018) Turbulence and older workers 741 Trends and age heterogeneity Long-run changes in employment are not uniformly spread across age groups. Instead, they are concentrated both on younger (aged 15 to 24) and older (aged 55 to 64)workers. In particular, in recent decades the aggregate employment rates of male workers in the U.S. and Europe have been dragged down by the decline of employment among older workers. To make this observation precise, we begin with a simple identity. We let ei at,ui at, and pi at denote respectively the employment, unemployment, and labor force participation rates of workers of age ain country iat time t. Also, we denote by ωi at the population share of these workers. The aggregate employment rate, ei t, is the following weighted average: ei t= a ωi atei at= a ωi at1−ui atpi at(1) The first two columns in Table 1show that the employment rate of older workers is slightly below the aggregate rate in most countries. The second set of columns reports that both rates have decreased since the late 1960s or early 1970s. Aggregate male employment has fallen by 164pp. on average across European countries and by 885 perTable 1. Changes in male employment rates in the U.S. and Europe. ei t0ei t1−ei t0 All Older All Older ωi at0 ei at0 ei t0 France 857730−191−388152 335 Germany 916835−212−364175 330 Italy 828629−160−208140 230 Norway 810830−516 −137183 406 Portugal 888817−184−226141 199 Spain 872827−240−3371 49 219 Sweden 860867−137−222187 270 United States 853827−885 −175150 294 Note: Data from the OECD labor force statistics database for male workers; Germany refers to West Germany prior to 1991; see Appendix A.1 for details. ei t0(resp. ei t1) denotes the employment rate at the beginning (resp. end) of the sample period for workers in all age groups (column “All”) and for older workers (column “Older”) country i.ωi at0 ei at0 ei t0 is the beginning-of-period employment share of older workers in country i. The numbers in boldface give the contribution of changes in employment among older workers to changes in aggregate employment. All entries are expressed in percentage points.
742 Etienne Lalé Quantitative Economics 9 (2018) centage points in the U.S. As can be seen, the decline has been much larger for older workers: it is about twice the aggregate decrease in several countries, including the U.S. The numbers in boldface give the contribution of those changes to the fall in aggregate employment.9On average in Europe, the decrease of older worker employment explains 352percent of the decrease in aggregate employment. The corresponding figure for the U.S. is 294percent. These numbers are substantial because the share of older workers in aggregate employment at the beginning of the sample period (last column in Table 1)isonly182percent in Europe and 150percent in the U.S. In other words, the fall in aggregate employment is disproportionately concentrated on older workers. Causes of low employment Lower employment can be caused by higher unemployment, lower labor force participation, or a combination of the two. The counterfactual series in Figure 1illustrate that, in what concerns employment among older workers, labor force participation plays a predominant role in these dynamics in each country.10 Here, we add two important observations. First, labor force participation accounts for a substantial part of cross-country differences in aggregate employment. Second, participation among older workers contributes a large share of those cross-country differences. Consider, again, equation (1) and denote by ei tthe difference in aggregate employment between country iand some baseline country jadjusted for demographic differences (using at ≡ωi at+ωj at 2). ei tcan be decomposed into differences coming from, respectively, unemployment (ui t) and labor force participation (pi t). That is, aej at −ei atat ei t = aui at −uj atpi at +pj at 2at ui t + apj at −pi at1−ui at +1−uj at 2at pi t (2) Further, we can measure the contribution of each age group athrough each of the two nonemployment margins.11 These contributions are reported in boldface in Table 2, 9Following equation (1), the contribution of age group ais the ratio between ωi at0+ωi at1 2(ei at1−ei at0)and ei t1−ei t0. 10To verify this observation, consider decomposing the variations of employment within age group a using: Var(log(ei at )) =Cov(log(ei at )log(1−ui at ))+Cov(log(ei at )log(pi at )). For older workers, the variance contribution of labor force participation is typically between 80 and 95 percent, whereas for prime-age workers there is a more even split between unemployment and labor force participation. 11For instance, the numbers reported in boldface in the rightmost column of Table 2are given by the ratio between (pj 55−64t −pi 55−64t )1−ui 55−64t +1−uj 55−64t 255−64t and ei t.
Quantitative Economics 9 (2018) Turbulence and older workers 749 workers with skill level h. The social insurance system is financed through a flat-rate tax τraised on the product of job matches. Two-tier labor market It is important to note that government-mandated programs in the WS economy create a two-tier labor market. First, the employment protection tax Ω changes the outside option of the employer when bargaining with an incumbent worker versus when meeting a new worker. Second, on meeting an employer, a worker may be collecting a benefit payment bthat differs from the benefit associated to the new job (this occurs if the worker’s skill level has changed since his previous job). In both instances, there is an insider-outsider phenomenon at work in the WS economy. We use an index i∈{0+}to capture this phenomenon, with i=0indicating the initial employment period and i=+for the continuation periods of the job. 3.2 Bellman equations The behavior of workers and employers who populate the economy can be described byasystemofBellmanequations. 18 Denoting by vn,vu,vei, the value of being in nonparticipation, unemployment, and employment with i∈{0+}, respectively, and by vo(·)≡max{vn(·)vu(·)}the value of being out of work, workers’ decisions are governed by: vn(bhza)=za +γab+β a αaa h μohh ×1−snf(θ)vobhza+snf(θ) (8) ×maxve0ybhzavob hzadG0hydFz|z vu(bhza)=b+β a αaa h μohh ×1−f(θ)vob hza+f(θ) (9) ×maxve0ybhzavob hzadG0hydFz|z ve0(ybhza)=w0(ybhza)+β a αaa ×λ h μdhhvob(h) hza+(1−λ) h μehh(10) ×maxve+yhzavobhhzadGhy|ydFz|z 18The Bellman equations are written with a summation over hwith the understanding that h=0H. The summation over ais written with the understanding that a=a a +1and the additional convention that α(A A +1)=0. In doing so, we are able to write the Bellman equations for all ain {0A}.
750 Etienne Lalé Quantitative Economics 9 (2018) ve+(yhza)=w+(yhza)+β a αaa ×λ h μdhhvob(h)hza+(1−λ) h μehh(11) ×maxve+yhzavobhhzadGhy|ydFz|z In equations (10)and(11), w0(·)and w+(·)are the wages paid during employment when i=0and i=+, respectively. The wage-setting rule is provided below. Assuming that there is free entry of firms, employers’ values of having a filled job vf0and vf+are given by: vf0(ybhza)=(1−τ)y −w0(ybhza)+β a αaa ×−λΩ +(1−λ) h μehh(12) ×maxvf+yhza−ΩdGhy|ydFz|z vf+(yhza)=(1−τ)y −w+(yhza) +β a αaa−λΩ +(1−λ) h μehh(13) ×maxvf+yhza−ΩdGhy|ydFz|z The decision rules for match formation and match continuation derive from the “max” operator in the Bellman equations above. These decisions are privately efficient from the viewpoint of each worker-employer pair under the assumption that agents bargain over the match surplus. 3.3 Nash bargaining As is standard, wages are set by period-by-period Nash bargaining. ψ∈[01]denotes the bargaining power of workers. The two-tier wage schedule is given by: w0(ybhza) =argmax wve0(ybhza)−vo(bhza)ψvf0(ybhza)1−ψ(14) w+(yhza) =argmax wve+(yhza)−vob(h)hzaψvf+(yhza)+Ω1−ψ(15) We can use the first-order conditions associated with (14)and(15) to obtain the decision rules for match formation and match continuation, y0(bhza) and y+(hza).
Quantitative Economics 9 (2018) Turbulence and older workers 751 These are pinned down by: vf0 y0(bhza)bhza=0(16) vf+ y+(hza)hza=−Ω (17) 3.4 Participation margin Workers’ labor force participation choice is subsumed by a threshold z(bha) which satisfies: vnbh z(bha)a=vub h z(bha)a(18) By combining this definition with equations (8)and(9), it is straightforward to show that at z= z(bha) the gains and losses of nonparticipation (relative to unemployment) offset each other: z(bha)a =(1−γa)b +(1−sn)f (θ)β a αaa h μohh ×maxve0ybhza(19) −vobhza0dG0hydFz| z(bha) Equation (19) also highlights how individual participation decisions and aggregate labor market conditions are intertwined. That is, z(bha) depends on the aggregate jobfinding probability f(θ) only when the worker faces a discounted net present value of employment (measured by the term after f(θ)) that is greater than 0. 3.5 Aggregate conditions Labor market tightness θand the payroll tax τare pinned down by aggregate equilibrium conditions. To write these conditions, n(b h z a),u(b h z a),e0(ybhza),and e+(yhza)denote the measures of workers in nonparticipation, unemployment, and employment in i=0and i=+. Free entry Employers create new vacancies until the discounted net present value of doing so is exhausted. Vacancies and job seekers meet by the end of a model period. Therefore, the free entry condition is given by: η=βf(θ) θ bha a αaa h μohh ×maxvf0ybhza0dG0hydFz|zu(b hza) +snn(bhza) u+snndz (20) where u=bha u(b hza)dz and n=bha n(b hza)dz. On the right-hand side of the equation, u(bhza)+snn(bhza) u+snnis the probability of drawing a worker with state variables b,h,z,afrom the pool of job seekers.
752 Etienne Lalé Quantitative Economics 9 (2018) Balanced budget Finally, the balanced budget condition is given by: τ ha ye+(yhza)+ b e0(ybhza)dy dz = b b ha u(bh za) +γan(bhza)dz (21) On the left-hand side of the equation, τmultiplies total output produced by the economy. The right-hand side of the equation links the generosity of social insurance schemes to the population shares of benefit recipients. 3.6 Equilibrium Having described the Bellman equations and aggregate equilibrium conditions, we are in a position to give the following definition. Definition An equilibrium is a list of value functions vn(bhza),vu(bhza),ve0(y bh za),ve+(yhza),vf0(ybhza),vf+(yhza), a set of decision rules for match formation and match continuation y0(bhza), y+(hza) and for labor force participation z(bha), a list of wage functions w0(ybhza),w+(yhza), a distribution of workers across the state space of the economy n(bhza),u(bh za),e0(ybhza), e+(yhza), and a value for labor market tightness θand the tax rate τsuch that: 1. Optimal match formation and match continuation decisions: Given θ,τand the value functions vf0(ybhza),vf+(yhza), match formation and match continuation decisions y0(bhza), y+(hza)solve equations (16)and(17), respectively. 2. Optimal labor force participation decisions: Given θ,τ, and the value functions vn(bhza),vu(bhza), labor force participation decisions z(bha) solve equation (18). 3. Nash bargaining: Given θ,τ, and the value functions vn(bhza),vu(bhza), ve0(ybhza),ve+(yhza),vf0(ybhza),vf+(yhza), the wage functions w0(ybhza),w+(yhza)are given by equations (14)and(15), respectively. 4. Time-invariant distribution: Given θ, the decision rules z(bha), y0(bhza), y+(hza), and the exogenous laws of motion of y,b,h,z,a,themeasuresn(b hza), u(bh za),e0(ybhza),e+(yhza) are time-invariant and their sum adds up to one. 5. Free entry: Given the measures n(b hza) and u(bhza) and the value of match formation vf0(ybhza), labor market tightness θsolves the free entry condition (20). 6. Balanced budget: Given the measures n(b hza),u(bh za),e0(ybhza), e+(yhza),τsatisfies the balanced budget condition given by equation (21). The following assumptions complete the description of condition 4 (time-invariant distribution). New labor market entrants are out of work initially, they are entitled to collect
Quantitative Economics 9 (2018) Turbulence and older workers 753 the lowest level of benefits b(0)and draw a leisure value zfrom the distribution F(·|z) (z denotes the unconditional mean value of z). The latter assumption is mostly innocuous because workers do not derive any utility from leisure while they belong to age group a=0. 4. Calibration The calibration process is organized as follows. First, using data moments for the U.S., we specify and calibrate parameters that are common to the two setups in Section 4.1. This pins down values for 15 parameters. Second, we set values for parameters that are specific to each economy in Section 4.2. These fall into one of two categories: (i) government-mandated programs, which include parameters unique to the WS economy, and (ii) two technology parameters, namely aggregate matching efficiency and the volatility of productivity shocks. Government-mandated programs per se can explain only a part of the differences between the U.S. and Europe observed in the initial period, and so we need (ii) to capture the residual difference in labor market dynamics. The first two steps of the calibration target the steady state of economies observed in tranquil times. The working assumption is that the parameters that have been set up at this point are invariant across time. Before closing this section, we explain in Section 4.3 how we define economic turbulence and how we measure its changes from tranquil to turbulent times. 4.1 Common parameters In this subsection, we set up the values for 10 out of 15 parameters using external information. We calibrate the remaining five jointly with the parameters discussed in the next subsection. It is useful to note that the five variables are essentially preference parameters, which are held common across the LF and WS economies. Throughout the analysis, one model period is considered to be half a quarter. Demographics19 The working life of individuals is divided into the following periods. While in the age bracket 20–49, workers transit across six consecutive five-year-long age groups. The probability of remaining in each of these is 0975. The subsequent age bracket, 50–54, consists of five one-year-long age groups; the probability of remaining in these groups is 0875. By pooling together the age groups of the 50–54 bracket and the last five age groups of the 20–49 bracket, we obtain the model counterpart of the so-called category “prime-age workers.” The last age bracket, 55–64, is the direct counterpart of the category “older workers.” It contains 20 six-month-long age groups, as the probability of remaining in each of these is 0750. The obvious role of this partition is to make the policy functions more flexible with respect to age toward the end of the working life. Discount factor The discount factor βis 09951 to accord with an annual interest rate of 4percent. 19For the sake of space, the demographic probabilities (the α(a a)’s) are not reported in Table 3.
754 Etienne Lalé Quantitative Economics 9 (2018) Table 3. Parameter values (one model period is half a quarter). Parameter Description Value A. Preference parameters LF WS βDiscount factor 09951 ξ20−54 Transitory shocks, prime-age workers* 0477 ξ55−64 Transitory shocks, older workers* 0578 πPersistence of leisure utility* 0700 zsup Upper bound for leisure utility* 0138 B. Human capital and match productivity μeProbability of upgrading skills 0033 μoProbability of losing skills 0066 y0Mean of match productivity, lower skill 10 yHMean of match productivity, higher skill 20 ρPersistence of productivity 0974 σStandard deviation of idiosyncratic shocks* 0432 0291 C. Labor market frictions λProbability of exogenous job destruction 00166 κElasticity of job filling w.r.t. tightness 05 ψBargaining power of workers 05 ηVacancy posting cost 3016 MMatching efficiency* 0495 0648 snRelative matching efficiency in nonparticipation* 0240 D. Policy schemes WS ΩJob destruction tax* 6500 φUnemployment benefits replacement ratio* 0328 γaRelative generosity of early retirement schemes γ55−59 :00 γ60−64 :05 Note: Parameters marked with an asterisk (*) are calibrated to match the data moments reported in Table 4.Thepersistence of leisure utility (Panel A) is calibrated to match the life-cycle profile of the transition probability from nonparticipation to unemployment shown in Figure 3. Leisure shocks We have assumed that the valuation of leisure begins at 0and increases with age (see equation (3)). This specification is clearly not designed to explain nonparticipation among younger workers and, to some extent, among prime-age workers.20 To sidestep this problem and fit the data on labor force participation for prime-age workers, we consider a simple extension of the model. We assume that, in addition to voluntary transitions, there are also involuntary reasons prompting workers to move in and out of the labor force. Specifically, we replace the value of being out of work (formerly defined as vo=max{vn(·)vu(·)})by vo(bhza)=ξavn(bhza)+(1−ξa)maxvn(bhza)vu(bhza)(22) 20Our model builds on leisure shocks (or entry costs to the labor market; see footnote 12) to rationalize labor force participation choices. In our view, a model focused on younger workers would need a different driving force and link their labor force participation to schooling investment choices. The earlier version of this paper (IZA working paper #10061 (2016)) provides an informal discussion of some changes of the model along those lines.
Quantitative Economics 9 (2018) Turbulence and older workers 755 ξais the age-dependent probability of a shock that forces a worker to spend one period in nonparticipation. Unlike shocks to the leisure component z, we think of the ξashocks as being transitory. For instance, such shocks could capture relocation to a new city, which would temporarily lower the arrival rate of job offers. We use only two values for ξato maintain parsimony, namely ξa∈{ξ20−54ξ55−64}. As just mentioned, zfollows a persistent stochastic process. Its Markov transition matrix is constructed as follows: with probability πthe value of zremains unchanged, while with probability 1−πanewvaluezis drawn from the uniform distribution over the support [0zsup]. Overall, the different shocks that generate worker transitions in and out of the labor force depend on four parameters: ξ20−54,ξ55−64,π,zsup. We target the following data moments, which are meant to capture the state of the U.S. labor market at the onset of the 1970s: (i) the labor force participation rate of prime-age workers is 95 percent, (ii) the participation rate of older workers is 80 percent, (iii) the unemployment rate of older workers is 35percent, (iv) the probability of transitioning from nonparticipation to unemployment falls by 10 percent per year between the ages of 55 and 64.Letusremark on targets (iii) and (iv). Regarding (iii), we target the unemployment rate of older workers, but alternatively we could target their unemployment-to-nonparticipation transition probability. The rationale behind target (iv) is that we need a data moment on the persistence of nonparticipation in order to disentangle the sources of movements in and out of the labor force. The calibration procedure yields ξ20−54 =0477,ξ55−64 =0578, π=0700,zsup =0138.21 The value of πimplies that zis resampled on average every 5 months. Production The unconditional means of the productivity process, the yh’s, are set to evenly partition the [12]interval. Thus, the match productivity of a worker who has reached the top of the skill ladder is, on average and unconditionally, twice higher than that of a new labor market entrant. It turns out that these values imply that wages increase by almost 75 percent from labor market entry to the mid-forties, in tune with the literature. Next, we draw on results from Chang and Kim (2006) to parametrize the persistence of match productivity, ρ. The authors use annual wage data to estimate the parameters of an autoregressive productivity process while controlling for selection into employment. The second panel of Table 1 in Chang and Kim (2006) shows that the annual persistence of idiosyncratic productivity is 0809 for men. This number implies ρ=08091/8=0974 since our model period is half a quarter. Exogenous job destruction We use data on the labor market history of displaced workers to parametrize λ, the probability of suffering an exogenous job destruction. In Appendix A.3, we document that workers with at least 1year of employment experience get displaced after spending on average 75years at the same job. Thus, we set λequal to 00166. 21The fact that ξ20−54 is lower than ξ55−64 might seem counterintuitive. However, this result does not mean that older workers experience more involuntary transitions out of the labor force. If a worker is better off in nonparticipation, then max{vn(bhza)vu(bhza)}=vn(bhza)and, therefore, the ξashock in equation (22) does not affect the worker. By this token, one needs a higher ξ55−64 to change labor force participation among older workers.
756 Etienne Lalé Quantitative Economics 9 (2018) Skill dynamics There are five grid points for the support of skills, {0H}.22 To construct the law of motion of h, we use the returns to human capital accumulation estimated by Kambourov and Manovskii (2009b).23 Denoting by xa worker’s job tenure, a regression of their estimates against a quadratic polynomial of xyields the following profiles: −00014 +00487x−00017x2for the OLS and −00003 +00287x−00010x2for the IV-GLS estimates. These profiles show that returns to tenure reach their peak after 14 to 15 years. Thus, we set the probability of upgrading skills, μe,to0033:giventhenumber of grid points H,ittakesaworker15 years on average to move from the lowest skill level to the highest one conditional on being employed continuously. For the probability of losing skills μo, we follow Ljungqvist and Sargent (1998, 2008) in assuming that depreciation of human capital when out of work is stochastically twice as fast as skill accumulation. Existing estimates of skill depreciation are quite disparate across studies, and the literature provides little additional guidance for choosing this parameter. We find, meanwhile, that the results are robust to varying μoby an order of magnitude. The main reason for this is that skill depreciation affects workers in the model most acutely when skills are destroyed immediately upon job loss, and less so when their skills deteriorate gradually during a spell of nonemployment. Matching and bargaining The elasticity of the job-filling probability with respect to labor market tightness, κ,issetto050 (Petrongolo and Pissarides (2001)). As is usual in the literature, we use the same parameter value for the workers’ share of the match surplus, ψ. In the next subsection, we set different aggregate matching efficiencies (M) for the LF and WS economies, but we do use a common value for the relative matching efficiency faced by nonparticipants, sn.24 We calibrate it to match the monthly transition rate from nonparticipation to employment of 640 percent tabulated by Krusell et al. (2011) (see panel “Men 21–65” in Table 3 of their paper). This yields sn=0240. We follow standard practice to pin down the vacancy posting cost, η.Wenormalize the value of labor market tightness to 1in the LF economy in tranquil time and use the free-entry condition to fix the parameter η. This yields η=3016. While this number may appear high, it is important to keep in mind that jobs in this model enable workers to accumulate skills and become more productive. In the steady-state equilibrium of the LF economy, output per worker is 2367. 4.2 Economy-specific parameters In the WS economy, we can, and do, calibrate the parameters for employment protection and unemployment insurance to match data targets. Due to a lack of good mapping between the model and data, we fix the value of the parameter that governs early retirement incentives, and discuss the effects of changing this value in subsequent sections. Last, we calibrate the remaining technology parameters. 22The results are robust to using a finer grid. We use five grid points to reduce computational costs. 23We use Table 2 from Kambourov and Manovskii (2009b). The authors report the returns at 2years, 5 years, and 8years of occupational tenure. Their OLS estimates are 00891,01995, and 02794, respectively. The corresponding numbers based on the IV-GLS estimation are 00539,01197, and 01680. 24One can think of snas a preference parameter insofar as sncould reflect the disutility of making search efforts.
Quantitative Economics 9 (2018) Turbulence and older workers 757 Government-mandated programs Boeri, Garibaldi, and Moen (2017) compile information on judicial discretion over severance payments in OECD countries. Table 1 of their study indicates that the cost of a fair economic dismissal for a worker with 20 years of job tenure amount to 101months of wages.25 Assuming that total dismissal costs are split in half between direct payments to the worker and payments to third parties,26 this yields a target of 507 months of wages for the job destruction tax, Ω.Weusetheaverage wage among workers with skill level h=Hto proxy the wage of high-tenure workers and obtain, through calibration, Ω=650. Next, we set up a target for the replacement ratio of unemployment insurance benefits, φ. Consider unemployment benefits in the U.S., which have a replacement ratio of 40 percent and last for 6months. Assuming a 45 percent (semi-quarterly) job-finding rate, the government would provide unemployed workers with the same expected payment using a replacement ratio of 31 percent and no time limit on the duration of benefits.27 Based on similar calculations, one can show that a 71 percent replacement ratio with infinite duration yields the equivalent of benefits with a replacement ratio of 75 percent and a duration of 36 months.28 We interpret the gap between 71 and 31 percent as capturing the difference in generosity of unemployment insurance benefits between the U.S. and Europe, and calibrate φto match a 40 percent replacement ratio in the WS economy. As previously mentioned, the mapping between the model and data is less clear in what concerns policies toward early retirement. First, there are multiple programs that provide these types of incentives, and it is beyond our scope to include them explicitly in the model. Second, what the model actually captures is the effect of those programs on the flow cost of nonparticipation (see equation (19)). Thus, our approach is to first fix the γa’s to reasonable values, and then study how they affect the results. We focus on γa∈{γ20−54γ55−59γ60−64}and set γ20−54 to 0throughout the analysis. γ55−59 =00and γ60−64 =05is our baseline specification up until Section 6.2. Other technology parameters For matching efficiency (M) and the standard deviation of productivity shocks (σ), we set up the following targets for the LF economy: (i) the unemployment rate among prime-age workers is 55percent, and (ii) their monthly separation rate during employment (i.e., transitions out of employment) is 25percent. Note 25Boeri, Garibaldi, and Moen (2017) report that the costs of a fair economic dismissal for a high-tenure worker amount to 74months of wages in France, 17 months in Germany, and 60months in Italy. We refer to the average of these three numbers. 26This 50:50 split is a compromise between the high uncertainty of legal procedures described in Boeri, Garibaldi, and Moen (2017) (which suggests a large deadweight loss) and the estimates of Garibaldi and Violante (2005) showing that direct payments to workers can account for up to two-thirds of total dismissal costs. 27Let fdenote the job-finding rate and denote by qthe per-period probability of exhausting benefits. An unemployed worker faces an expect payment of ∞ t=0βt(1−f)t(1−q)tb=1 1−β(1−f)(1−q) b(we ignore the life-cycle dimension here, which has a negligible impact on the calculations). With benefits of infinite duration, denoted as b∞, that payment becomes 1 1−β(1−f)b∞. Plugging f=045,q=025,b=040 into b∞=1−β(1−f) 1−β(1−f)(1−q) byields b∞=031. 28To see this, use f=045,q=00417,b=075 in the formula of footnote 27.Weusethesamevalueofthe job-finding rate fin these calculations because, as Kitao, Ljungqvist, and Sargent (2017) point out, there was little difference in unemployment duration between the U.S. and Europe in the 1970s.
758 Etienne Lalé Quantitative Economics 9 (2018) that at this stage we have used the unemployment rates of both prime-age workers and older workers in the U.S. as calibration targets. In the WS economy, we search for the parameter values of Mand σthat (iii) minimize the unemployment rate and (iv) fit a labor force participation rate of older workers at 65 percent. Again, the data moments we target are representative of the state of the U.S. and Europe in the early 1970s. We obtain M=0495 and σ=0432 in the LF economy, and M=0648 and σ=0291 in the WS economy. Not surprisingly, in the WS economy with its costly governmentmandated programs, there are fewer incentives for firms to post vacancies and the model thus needs a higher matching efficiency to rationalize low unemployment rates. This economy attributes the lower separation rates in Europe to a mix of employment protection and a less volatile productivity process. Table 3provides a summary of the parametric specification and calibration of the model. In Table 4, the first column reports the nine targeted moments discussed in Sections 4.1 and 4.2. The second column of that table shows that the model performs well at matching the targets. We make additional important connections between the model and data in the next subsection and Section 5. 4.3 Economic turbulence We use Ljungqvist and Sargent’s (1998) construct to specify the stochastic process of skill loss when a worker is exogenously separated from his job. For each skill level h,the μd(hh)’s are drawn from the left half of the Normal distribution with mean h,truncated at hand normalized to integrate to 1over {0h}. Notice that μd(hh)=0for any h>h, and that the probabilities of moving to a lower skill level depend on a single parameter, namely the dispersion of the underlying Normal distribution. A higher degree of economic turbulence refers to an increase of this parameter. Next, we use the LF economy as a tool to estimate the degree of economic turbulence. Bertola and Ichino (1995), Ljungqvist and Sargent (1998), Kambourov and Table 4. Assessment of the model fit. Description Target Model A. LF economy Unemployment rate, 25–54 550 545 Transition rate from Eto (U N),25–54 250 238 Participation rate, 25–54 950951 Unemployment rate, 55–64 350 384 Participation rate, 55–64 800799 Transition rate from Nto E,20–64 640 683 B. WS economy Participation rate, 55–64 650659 Job destruction tax 507 504 Unemployment benefits replacement ratio 400410 Note: The following abbreviations are used: E:employment;U:unemployment;N: nonparticipation. All entries are expressed in percentage points. In Panel B, the job destruction tax is expressed as a fraction of the monthly wage of high-tenure workers. The replacement ratio is the ratio between average unemployment benefits and the average wage.
Quantitative Economics 9 (2018) Turbulence and older workers 765 Table 6. LF and WS economies in turbulent times. Degree of economic turbulence 000 020 040 060 080 100 A. Aggregate outcomes Tax rate τ255 274 304 319 326 329 Net output LF 1822 1757 1676 1641 1626 1618 WS 1539 1468 1379 1340 1323 1316 Average skill level LF 1671 1597 1505 1465 1448 1441 WS 1671 1595 1502 1461 1444 1436 Average wage LF 2031 1977 1914 1888 1877 1869 WS 1575 1516 1447 1417 1404 1398 B. Prime-age workers Unemployment rate LF 545 569 612 635 645 660 WS 490 520 571 598 610 625 Job-finding rate (Uto E)LF365350327318314307 WS 272254229218214208 Separation rate (Eto (UN))LF238 240 245 248 249 251 WS 164 164 166 167 167 168 Participation rate LF 9519499469449439 42 WS 956953948946945943 C. Older workers Unemployment rate LF 384 387 390 391 389 390 WS 379 403 446 474 486 504 Job-finding rate (Uto E)LF356340316306302296 WS 190160120103960 867 Separation rate (Eto (UN))LF336 346 361 368 371 375 WS 235 236 238 239 240 240 Participation rate LF 799776740723716706 WS 659619560534522508 Transition to nonparticipation (N) From employment (E)LF276 289 309 318 323 418 WS 200 204 210 213 215 217 From unemployment (U)LF228239258266269275 WS 347373410427434443 Note: The following abbreviations are used: E:employment;U:unemployment;N: nonparticipation. The tax rates in Panel A and the entries in Panels B and C are expressed in percentage points. Job-finding, separation rates (Panel B) and transitions to nonparticipation (Panel C) are monthly transition probabilities. Age-specific outcomes Panels B and C of Table 6report the consequences of turbulent times on employment among prime-age and older workers, respectively. More details are provided in Panel C in order to explain changes in the labor force participation rates of older workers. Prime-age workers experience a slight decrease in employment, by 215 percent in the LF economy and 276 percent in the WS economy. As can be seen in Panel B, the rates of labor force participation remain almost unchanged, so that the bulk of employment
766 Etienne Lalé Quantitative Economics 9 (2018) changes is driven by an increase in unemployment. The job-finding rate is the main variable explaining changes in the unemployment rate of prime-age workers. Notice that job-finding rates depend on three elements: labor market tightness, the decision rule for match formation, and the cross-sectional distribution of workers. It can be shown that the decrease in vacancies (and thus labor market tightness) accounts for the behavior of the job-finding rate among prime-age workers; see Appendix B.1.Conversely, shifts in the cross-sectional distribution lead to a compositional change that explain the increase in separation rates (employed workers have lower skills), but this plays little role in the dynamics of unemployment in Panel B. The employment impact of turbulence is much more significant for older workers. Panel C indicates that their employment rates decreases by 117percent in the LF economy, from 768to 678percent. The effect is twice as large in the WS economy: older worker employment rate decreases by 239percent, from 634to 482percent. Their unemployment rate remains almost unchanged in the LF economy, while it increases by one-third in the WS economy. But the main effect is the decrease in labor force participation across the two economies. We highlight below that those changes (namely, −116 percent in the LF economy and −229percent in the WS economy) are quantitatively consistent with the data. There are two mechanisms driving this effect. First, as shown by equation (19), the decrease in labor market tightness lowers the opportunity costs of being in nonparticipation relative to unemployment. Second, the cross-sectional distribution of the economy shifts toward older workers with a lower conditional probability of match formation, leading to a higher probability of moving to nonparticipation. The last rows of Panel C confirm that participation decreases because both employed and unemployed workers drop from the labor force earlier. Older workers face a severe employability problem for two reasons. First, the process of building up human capital implies that age is correlated with a higher skill level (meaning relatively larger skill losses in turbulent times) and more generous welfare benefits. Second, the “horizon effect” (Chéron, Hairault, and Langot (2009, 2013)) implies that the returns to hiring older workers are lower. More on the mechanisms In turbulent times, labor force attachment among employed workers decreases. The model enables us to formalize this idea and quantify its implications. That is, we can use it to compute the share of employed workers who would choose nonparticipation over unemployment if they were not employed. We find that the workers account for 444percent of employment at age 60 in the LF economy under tranquil times, and that this number rises to 560percent in turbulent times. In the WS economy, the corresponding figure for workers aged 60 in tranquil times is 744percent. This figure is so large that it increases “only” to 787percent in turbulent times. Put differently, in the WS economy workers become less attached to the labor force at younger ages. For instance, 543percent of employed workers would prefer nonparticipation over unemployment at age 58 in turbulent times. A direct implication of these observations is that wages should be less responsive to aggregate economic conditions at older ages.34 34Hairault, Langot, and Zylberberg (2015) proposed a life-cycle employment model where older workers may prefer retirement over unemployment conditional on being out of work. The authors show that, if this
Quantitative Economics 9 (2018) Turbulence and older workers 767 We can also use our model to ask how labor force attachment among nonemployed workers contributes and responds to aggregate outcomes. In experiments not reported here, we addressed two such questions. First, by self-selecting themselves out of the labor force, do workers contribute significant improvements to the quality of the pool of job seekers? The answer is a clear “no.” Holding the surplus value of firms vf0(ybhza) fixed to its initial value while using the distributions n(b hza) and u(bhza) from turbulent times to calculate the returns to posting a vacancy, we found that labor market tightness (θ) was only marginally lower than in the equilibrium under tranquil times. The second question is: how much of the decline in labor force participation is driven by aggregate conditions measured by f(θ)? To answer this, we performed a partial-equilibrium exercise, shifting the job-finding probability from its value in tranquil times to its value in turbulent times. Labor force participation among older workers decreased by only 1to 2pp. in the LF economy and around 3pp. in the WS economy. In sum, the bulk of changes in labor force participation in Panel C of Table 6 comes from shifts in the cross section of workers, rather than shifts in the policy functions. Taking stock We now examine the levels and trends observed in the data through the lens of the model. In Table 7, the set of rows titled “data” reports the relevant empirical moments measured at the beginning and end of the period considered, followed by their change measured in percentage terms. Panel A refers to the U.S. and Panel B displays the average of France, Germany, and Italy. The first remarks concern the ability of the model to accurately describe labor force participation and explain its evolution from the early 1970s to the late 1980s. The LF economy matches the U.S. levels well, and effectively links the bulk of changes in labor force participation among older workers to the increase in economic turbulence. It predicts a decrease of 116percent while the actual decrease is 135percent. So, the model explains the decline in employment among older workers in the U.S. as the main driving force behind this dynamic is the change in their labor force participation. Similarly, the WS economy provides a good quantitative account of the behavior of labor force participation among older workers in Europe. The table shows that, through the lens of this economy, the increase in turbulence leads to a fall in participation by 229percent versus 241percent in the data. Meanwhile, as we discuss below, it cannot rationalize the important changes that accompanied this dynamic. The explanatory power of the model is lower along the other dimensions. In line with the U.S. data, the LF economy exhibits little change in labor force participation among prime-age workers, but it predicts little change in the unemployment rates of so happens, then the search externality vanishes for older employed workers because their Nash-bargained wage is independent of labor market tightness. The “unattached employed workers” in our model are in similar, but not identical, positions. While they prefer nonparticipation over unemployment given their current state variables, they may still experience a negative shock to leisure utility (z) or positive productivity shocks (yor h) that reverse this ordering.
768 Etienne Lalé Quantitative Economics 9 (2018) Table 7. Employment changes: LF and WS economies versus data. Early Late % 1970s 1980s Change A. United States Unemployment rate, 25–54 Data 359 553 +541 LF 545 660 +211 Participation rate, 25–54 Data 949935−150 LF 951942−095 Unemployment rate, 55–64 Data 310 435 +405 LF 384 390 +156 Participation rate, 55–64 Data 780675−135 LF 799706−116 B. Europe Unemployment rate, 25–54 Data 151 474 +213 WS 490 625 +276 Participation rate, 25–54 Data 970939−320 WS 956943−136 Unemployment rate, 55–64 Data 255 564 +121 WS 379 504 +330 Participation rate, 55–64 Data 677514−241 WS 659508−229 Note: Panels A and B: data from the OECD labor force database for male workers; see Appendix A.1 for details. The early 1970s (resp. late 1980s) refer to the mean value over the years 1970–1974 (resp. 1986–1990). In Panel B, Europe refers to the (unweighted) average of statistics for France, Germany, and Italy. All entries in both panels are expressed in percentage points. prime-age and older workers.35 The fit of the WS economy with respect to European unemployment rates is also less satisfactory. This is not surprising for the levels of unemployment in the early 1970s since the calibration of the WS economy does not target these moments.36 But the model does miss by a significant margin the outbreak of high European unemployment that occurred at the end of the 1970s. It predicts only between 10 and 30 percent of those changes, depending on the demographic group considered. In light of these results, it seems that rising economic turbulence cannot explain high unemployment if one accounts for the endogenous labor supply decisions of workers. Here, we make two additional comments. First, we have thus far analyzed the consequences of rising economic turbulence in two economies which differ with respect 35The data moments in the first column of Table 7are calculated over the years 1970 to 1974. Thus, they are slightly different from the data moments of the calibration, which capture the state of the U.S. labor market at the onset of the 1970s (see Panel A of Table 4). In particular, the unemployment rates in Table 7 are lower because of the recovery period after the 1970 U.S. recession. Note that this makes the relative change (last column of Table 7) in unemployment from the early 1970s to the late 1980s look larger in the data. 36The calibration minimizes unemployment while targeting the rate of labor force participation. A higher value of the matching efficiency parameter (M) lowers unemployment but it increases the incentives to participate in the labor force. The standard deviation of productivity shocks (σ) helps strike a balance to obtain lower unemployment rates among older workers relative to unemployment among prime-age workers.
Quantitative Economics 9 (2018) Turbulence and older workers 769 to both labor market institutions and some technology parameters. To measure the effects of the interaction between economic turbulence and institutions, we would need to remove the difference in technology parameters. We did so in experiments not reported here: we studied a laissez-faire economy with the parameters Mand σof the baseline WS economy. Subtracting technological differences closed almost half of the gap in labor market outcomes between the LF and WS economies. For instance, labor force participation among older workers decreases by 162percentfromtranquiltoturbulent times in the reparametrized LF economy (versus 116percent in the baseline). We conclude that per se the interaction between turbulence and institutions explains at least 50 percent of the differences between the two economies. The other comment relates to the timing of employment changes analyzed in this section. On the one hand, the increase in turbulence as measured by the transitory component of earnings ends during the late 1980s (footnote 29), which is also the period when labor force participation among older workers stabilizes in the U.S. On the other hand, in Europe, the downward trend in labor force participation continues after this period (see Figure 1). It is conceivable that the adoption of new information technologies and the induced changes in organizations and work practices occurred later in Europe than in the U.S. Yet a perhaps more promising explanation is that there were also changes in labor market institutions that impacted labor force participation among older workers in Europe. The next subsection presents results that substantiate this explanation. 6.2 Changing labor market institutions In this subsection, we begin by briefly describing the relevant changes in a specific labor market institution—early retirement schemes—during the period from the 1970s to the 1990s. Then we use our model to analyze the nature and magnitude of the employment effects of those changes.37 Summary of the evidence The chapters collected in Gruber and Wise (2010) document a trend toward policies that incentivized older workers to withdraw from the labor market, followed by a reversal starting in the 1990s. Here, we summarize the salient facts for France,Germany,andItaly. 38 France developed several early retirement schemes targeted at workers aged 60 to 65 in the 1970s. The most important of these was the so-called Guarantie de ressources, 37Existing evaluations of the employment effects of early retirement policies are mostly based on reduced-form analyses. For instance, one regresses the unemployment rate for the younger on the labor force participation rate of older workers, while exploiting some policy discontinuities or controlling for variables that may lead to spurious correlation. In this section, we study the equilibrium response of the labor market following a change in the parameters of early retirement policies. We use the variations of labor force participation among older workers prompted by the policy change to calculate the employment and unemployment elasticities reported in Table 9. 38Our summary for France is based on the chapter written by Ben Salem et al. (2010); for Germany on the chapter by Börsch-Supan and Schnabel (2010); and for Italy on the chapter by Brugiavini (2010). The book by Gruber and Wise (2010) contains specific chapters for five other European countries, namely Belgium, the Netherlands, Spain, Sweden, and the U.K.
770 Etienne Lalé Quantitative Economics 9 (2018) which was introduced in 1972 for laid-off workers and extended in 1977 to those who voluntarily quit their jobs. An additional phase of early retirement schemes targeted at workers aged over 55 years was implemented in the 1980s (in addition to lowering the normal retirement age from 65 to 60). These schemes worked through an unemployment insurance route: they exempted older workers from searching for a job, and provided them with benefits until they become entitled to a full-rate pension. The 1993 Balladur reform of the pension system marks the end of the trend toward promoting early retirement. In (West) Germany, initially the only option for men to retire before the age of 65 was to prove a disability. A first reform was passed in 1972 with the stated goal of “providing more leisure to the workers” (Börsch-Supan and Schnabel (2010,p.152)).Themostimportant changes took place in the 1980s, when more generous unemployment benefits for workers aged 55 to 59 were introduced in order to create a “bridge to retirement.” These benefits were not means tested, and in addition workers were exempted from the need to meet job-search requirements. The reversal of trend was initiated by the 1992 reform leading to (quasi-)actuarial adjustments to the benefit system. The new phase of policy changes includes the 2001 Riester reform and some elements of the Hartz reforms. In what regards early retirement benefits in Italy, “after World War II, acts of Parliament enacted piecemeal changes that went almost invariably in the direction of increasing generosity, with no concern about the long-term effects of these amendments” (Brugiavini (2010, p. 195)). The first attempts to cut the incentives for workers to withdraw from the labor force long before retirement age were made in 1984. The government introduced a minimum eligibility level for yearly earnings that counted as full for social security tax payments. It also tightened the eligibility criteria for, and limited their duration of, disability insurance benefits. But the trend really came to an end with the 1992 Amato reform and the 1995 Dini reform of the social security system. The WS economy can speak to the effects of changing policies toward early retirement. True, it does not contain an explicit model of the policies, but it can at least capture their effects on the incentives (in terms of flow value) of remaining in the labor force instead of dropping out into early retirement programs. For workers aged 55 to 64,this additional incentive is given by (1−γa)b. Our approach is to explore the range of values between two extreme benchmarks: we do this by varying the parameter γafrom zero to 100 percent to measure the effects of eliminating the flow value of participating in the labor force. Noting that early retirement policies were implemented as early as the 1970s and were seldom repealed before the 1990s, we also explore two extreme cases in terms of economic context: namely, we study their effects in both tranquil times and turbulent times. In what follows, it is important to note that the WS economy is slightly different from that studied in Section 6.1 in that we recalibrate the WS economy in line with changes made to the parameter γa. The calibration procedure and parameter values are provided in Appendix B.2.
Quantitative Economics 9 (2018) Turbulence and older workers 771 Employment effects Table 8reports the effects of changing early retirement incentives on unemployment and labor force participation among older workers. To do so, our main instrument is the parameter γ60−64, which crucially affects the incentives offered to workers aged 60 to 64. In keeping with out approach of exploring extreme benchmarks, we consider two alternatives for the other parameter, γ55−59: we keep it either fixed to zero or we set it equal to γ60−64. The first remark is that early retirement incentives are very effective in reducing labor force participation among older workers. In the various environments considered, we find that changing the parameters governing the generosity of those schemes can decrease participation by about 40 to 45 percent. There is substitution with unemployment: the unemployment rate of older workers decreases by roughly one pp. in tranquil times and two pp. in turbulent times. Again, this effect revolves around reducing the relative losses of nonparticipation shown in equation (19). Note that the effect on unemployment is not always linear. An increase in γacan increase the value of employment (the worker will bargain for a higher wage at older ages) while having little impact on the probability of match formation. This channel implies a higher opportunity cost of nonparticipation, and thus explains the slight increase in unemployment at lower values of γa. Table B.1 in the Appendix is the analogue for prime-age workers of Table 8.Inthat table, we report that labor force participation among workers aged 25 to 54 is very insensitive to early retirement policies, but that their unemployment rates are somewhat more responsive. There are two macro-channels that could be driving this result, and which may interact with one another. First, early retirement schemes improve the bargaining position of workers and thereby lead to lower returns to job creation. Second, increasing the generosity of these schemes leads to an increase in the tax rate τto meet the government budget constraint. We find that the unemployment effects are mostly driven by the negative impact on job creation. Table 8. Effects of early retirement incentives on employment among older workers. Generosity of incentives γ60−64 000 025 050 075 100 A. Tranquil times Unemployment rate γ55−59 =0397 408 407 381 324 γ55−59 =γ60−64 371 384 387 351 264 Participation rate γ55−59 =0772715649583527 γ55−59 =γ60−64 781716634544457 B. Turbulent times Unemployment rate γ55−59 =0581 596 576 494 394 γ55−59 =γ60−64 556 573 557 479 383 Participation rate γ55−59 =0598533463393330 γ55−59 =γ60−64 605540469399335 Note: Results for older workers: calculations are based on the parameter values reported in Table 3and the recalibrated parameter values reported in Appendix B.2. All entries are expressed in percentage points.
772 Etienne Lalé Quantitative Economics 9 (2018) Quantitative appraisal Next, we synthesize the impact of early retirement programs by evaluating several employment and unemployment elasticities. As already mentioned, the goal is to get a sense of the nature and magnitude of the main effects, using the variations of labor force participation among older workers triggered by policy changes. We let ωadenote the population share of age group a(a=25–54 for prime-age workers and a=55–64 for older workers), and denote by ea,ua,pathe employment, unemployment, and labor force participation rates, respectively. Also, we use kto denote the elasticity of k∈{eauapa}with respect to participation among older workers. The main accounting equation is: e=ω25−54 e25−54 ee25−54 +ω55−64 e55−64 ee55−64(23) This equation decomposes the effects of older worker participation rates on the aggregate employment rate through two channels: directly through employment in this age group (e55−64 ee55−64) and indirectly through employment among younger workers (e25−54 ee25−54). Table 9reports the employment elasticities that enter equation (23) and two unemployment elasticities, u25−54 and u55−64. Consider first the employment effects among older workers. The relevant elasticities are linked by: e55−64 =1−u55−64 1−u55−64 u55−64.Thus, the calculations verify that nonparticipation among older workers is essentially a substitute for unemployment (u55−64 is positive), which leads to an employment elasticity of around 1minus the unemployment rate in all instances. Turning to the effects on prime-age workers, we see that u25−54 is negative, showing that older worker nonparticipation is complemented by unemployment among younger workers. The effects on the employment rate of prime-age workers is inherently more modest. Here, the relevant accounting equation is: e25−54 =p25−54 −u25−54 1−u25−54 u25−54, which yields smaller absolute values since p25−54 ≈0. The last column of Table 9displays the results based on equation (23). We draw two main conclusions. First, the complementarity between older worker nonparticipation and unemployment at younger ages has a nonnegligible impact on aggregate employTable 9. Adding up the employment effects of early retirement incentives. Elasticities by Age Groups u25−54 e25−54 u55−64 e55−64 Total e A. Tranquil times γ55−59 =0−0236 0022 0396 0983 0201 γ55−59 =γ60−64 −0242 0022 0374 0985 0197 B. Turbulent times γ55−59 =0−0144 0024 0587 0965 0158 γ55−59 =γ60−64 −0164 0025 0574 0967 0159 Note: Calculations are based on the parameter values reported in Table 3and the recalibrated parameter values reported in Appendix B.2.
Quantitative Economics 9 (2018) Turbulence and older workers 773 ment. A simple calculation illustrates this. e55−64 is close to 1, the population share of older workers (among workers aged 25 to 64)ω55−64 is 025, and their relative employment rate, e55−64/e,isabout70 percent in tranquil times and 55 percent in turbulent times. As a result, ω55−64 e55−64 ee55−64 is roughly one-quarter of 70 percent (0175)in tranquil times, and one-quarter of 55 percent (0138) in turbulent times. But due to the negative value of u25−54, the elasticity of aggregate employment is 15 percent higher than this number. Second, and somewhat conversely, the magnitude of the elasticity u25−54 is too small to trigger large unemployment responses. Thus, although in Section 6.1 we ignored potential changes in retirement policies over time, these cannot explain the discrepancy between the WS economy and the outbreak of high European unemployment. 7. Conclusion We provide a novel assessment of the effects of rising economic turbulence and its interaction with labor market institutions. To this end, we develop a rich life-cycle model featuring two sources of nonemployment: there are frictions in the labor market and agents face a nondegenerate labor supply problem. Our first result is that rising economic turbulence consistently explains the lower labor force participation of older workers, and how it has contributed to the decline in aggregate male employment in the U.S. Thus, turbulence is not just an account of the steady U.S. unemployment rate. Second, economic turbulence and institutions explain the much larger decrease in labor force participation among older workers in Europe. However, neither of these factors offer much in terms of explaining the increase in unemployment, which is somewhat in contrast with the standard interpretation of the effects of those shocks. Finally, we find that the early retirement policies of the 1970s–1990s period, although detrimental to employment, cannot bring the model closer to explaining the era of high unemployment in Europe. Our model generates worker transition probabilities across employment, unemployment, and nonparticipation, with some success in explaining how these probabilities change over the life cycle. It would be interesting for future work to develop a version of the model that fits these transition probabilities from labor market entry to labor market exit—in all likelihood, this would be achieved by removing some layers of worker heterogeneity that were relevant for this paper. This model could shed light on the structural determinants (e.g., preferences, technology) of the large life-cycle variations of worker flows observed in the data. It would also be useful to develop cross-country empirical evidence on the life-cycle profile of transition probabilities between the three labor market states. The model, or a modified version of it, could serve as a structural tool to analyze the discrepancies and relate them to cross-country differences in labor market institutions. On a related note, the model could help understand why the effects of certain labor market policies (e.g., minimum wage, employment protection) are so heterogeneous over the life cycle.
774 Etienne Lalé Quantitative Economics 9 (2018) Appendix A: Data Appendix A.1 Cross-country time series Our analysis of cross-country time series is mostly based on data from the Organisation for Economic Co-operation and Development (OECD) labor force database (http: //stats.oecd.org/). The OECD provides employment and labor force participation statistics harmonized for the purpose of developing cross-country comparisons. These data are available at the country level but also at a finer level, namely gender and different age groups. The disaggregated data are not census-based, however: they are taken from labor force surveys, which usually span a shorter period of time. Therefore, we complement our analysis of OECD data as follows: •For France, the OECD data coverage begins in 1983. We compute the time series prior to 1983 directly from the French Labor Force Survey. The 1968–1982 waves of the survey are obtained from the repository of the Réseau Quetelet (http://www. reseau-quetelet.cnrs.fr/). •For Germany, the OECD database before 1991 covers West Germany only. There is no ideal method to address this data issue. Meanwhile, the OECD West German data is available up until 1998, meaning we can conduct sensitivity checks using these data. In results available upon request, we find that the stylized facts of Section 2are also borne out by the West German data. •For Italy, several time series from the OECD database exhibit large discontinuities in 1982 and 1993. We remove the breaks in the OECD data by aligning those time series to their respective counterparts provided by the Italian National Institute of Statistics (http://dati.istat.it/). •For Spain and Norway, the OECD data coverage begins in 1972. For Portugal, the data coverage begins in 1974. We make no attempt to expand these data before the first period of observation. No adjustment is required for Sweden, as the OECD Swedish data begin as early as 1963. A.2 Life-cycle profile of transition probabilities In order to study transition probabilities, we use microdata that allow us to link respondents longitudinally over time. The data come from the French Labor Force Survey (LFS), the German Socio-Economic Panel (GSOEP), the Italian sample of the European Union Statistics on Income and Living Conditions (EU-SILC), and the monthly Current Population Survey (CPS). The latter data are available as of 1976, so that we can construct transition probabilities for the U.S. even before the 1980s (see Figure 3). In Figure 2,we use CPS data from the years 2005–2015 to match the time period spanned by the other datasets. Using the linked data, we compute the transition probability of moving across labor market states (employment, unemployment, nonparticipation) for each group of individuals of age aobserved during period t.Letqij denote the transition probability of
Quantitative Economics 9 (2018) Turbulence and older workers 781 Bertola, G. and A. Ichino (1995), “Wage inequality and unemployment: United States vs. Europe.” NBER Macroeconomics Annual, 10, 13–54. [739,758] Blanchard, O. (2006), “European unemployment: The evolution of facts and ideas.” Economic Policy, 21 (45), 5–59. [735,740] Blanchard, O. and J. Wolfers (2000), “The role of shocks and institutions in the rise of European unemployment: The aggregate evidence.” The Economic Journal, 110 (462), 1–33. [739] Boeri, T., P. Garibaldi, and E. R. Moen (2017), “Inside severance pay.” Journal of Public Economics, 145, 211–225. [757] Börsch-Supan, A. and R. Schnabel (2010), “Early retirement and employment of the young in Germany.” In Social Security Programs and Retirement Around the World: The Relationship to Youth Employment, 147–166, University of Chicago Press. [769,770] Brugiavini, A. (2010), “Social security and retirement in Italy.” In Social Security Programs and Retirement Around the World: The Relationship to Youth Employment (J. Gruber and D. A. Wise, eds.), 181–237, University of Chicago Press. [769,770] Chang, Y. and S.-B. Kim (2006), “From individual to aggregate labor supply: A quantitative analysis based on a heterogeneous agent macroeconomy.” International Economic Review, 47 (1), 1–27. [755] Chéron, A., J.-O. Hairault, and F. Langot (2009), “The role of institutions in transatlantic employment differences: A life-cycle view.” Annals of Economics and Statistics, 95–96, 121–138. [736,739,740,766] Chéron, A., J.-O. Hairault, and F. Langot (2013), “Life-cycle equilibrium unemployment.” Journal of Labor Economics, 31 (4), 843–882. [739,766] Choi, S., A. Janiak, and B. Villena-Roldán (2015), “Unemployment, participation and worker flows over the life-cycle.” The Economic Journal, 125 (589), 1705–1733. [740] den Haan, W. J., C. Haefke, and G. Ramey (2005), “Turbulence and unemployment in a job matching model.” Journal of the European Economic Association, 3 (6), 1360–1385. [739,747] Elsby, M. W. L., B. Hobijn, and A. ¸Sahin (2013), “Unemployment dynamics in the OECD.” Review of Economics and Statistics, 95 (2), 530–548. [740] Flinn, C. J. and J. J. Heckman (1983), “Are unemployment and out of the labor force behaviorally distinct labor force states?” Journal of Labor Economics, 1 (1), 28–42. [743] Garibaldi, P. and G. L. Violante (2005), “The employment effects of severance payments with wage rigidities.” The Economic Journal, 115 (506), 799–832. [757] Garibaldi, P. and E. Wasmer (2005), “Equilibrium search unemployment, endogenous participation, and labor market flows.” Journal of the European Economic Association,3 (4), 851–882. [738,740,746]
782 Etienne Lalé Quantitative Economics 9 (2018) Gottschalk, P. and R. A. Moffitt (1994), “The growth of earnings instability in the US labor market.” Brookings Papers on Economic Activity, 25 (2), 217–272. [738,759,776] Gottschalk, P. and R. A. Moffitt (2009), “The rising instability of US earnings.” Journal of Economic Perspectives, 23 (4), 3–24. [738,759,776] Gruber, J. and D. A. Wise (2010), Social Security Programs and Retirement Around the World: The Relationship to Youth Employment. University of Chicago Press. [769] Guvenen, F. and B. Kuruscu (2010), “A quantitative analysis of the evolution of the US wage distribution, 1970–2000.” NBER Macroeconomics Annual, 24 (1), 227–276. [762] Hairault, J.-O., F. Langot, and A. Zylberberg (2015), “Equilibrium unemployment and retirement.” European Economic Review, 79, 37–58. [766] Hairault, J.-O., T. Sopraseuth, and F. Langot (2010), “Distance to retirement and older workers’ employment: The case for delaying the retirement age.” Journal of the European Economic Association, 8 (5), 1034–1076. [739] Hornstein, A., P. Krusell, and G. L. Violante (2007), “Technology-policy interaction in frictional labour-markets.” Review of Economic Studies, 74 (4), 1089–1124. [739] Jacobson, L. S., R. J. LaLonde, and D. G. Sullivan (1993), “Earnings losses of displaced workers.” American Economic Review, 83 (4), 685–709. [763,764] Jones, S. R. G. and W. C. Riddell (1999), “The measurement of unemployment: An empirical approach.” Econometrica, 67 (1), 147–162. [744,746] Jones, S. R. G. and W. C. Riddell (2006), “Unemployment and nonemployment: Heterogeneities in labor market states.” Review of Economics and Statistics, 88 (2), 314–323. [744] Kambourov, G. and I. Manovskii (2009a), “Accounting for the changing life-cycle profile of earnings.” Report, University of Pennsylvania. [758,759,762,776] Kambourov, G. and I. Manovskii (2009b), “Occupational specificity of human capital.” International Economic Review, 50 (1), 63–115. [756,775] Kitao, S., L. Ljungqvist, and T. Sargent (2017), “A life cycle model of trans-atlantic employment experiences.” Review of Economic Dynamics, 25, 320–349. [736,740,757] Krusell, P., T. Mukoyama, R. Rogerson, and A. ¸Sahin (2011), “A three state model of worker flows in general equilibrium.” Journal of Economic Theory, 146 (3), 1107–1133. [740,756] Krusell, P., T. Mukoyama, R. Rogerson, and A. ¸Sahin (2017), “Gross worker flows over the business cycle.” American Economic Review. (Forthcoming). [740] Lalé, E. (2016), “Turbulence and the employment experience of older workers.” IZA Discussion Paper 10061. [745,754] Layard, R., S. Nickell, and R. Jackman (2005), Unemployment: Macroeconomic Performance and the Labour Market. Oxford University Press, USA. [740]
Quantitative Economics 9 (2018) Turbulence and older workers 783 Lazear, E. P. (1988), “Employment-at-will, job security, and work incentives.” In Employment, Unemployment and Labor Utilization (R. A. Hart, ed.). Unwin Hyman, Boston, USA. [748] Ljungqvist, L. and T. J. Sargent (1998), “The European unemployment dilemma.” Journal of Political Economy, 106 (3), 514–550. [737,738,748,756,758] Ljungqvist, L. and T. J. Sargent (2007), “Understanding European unemployment with matching and search-island models.” Journal of Monetary Economics, 54 (8), 2139–2179. [747] Ljungqvist, L. and T. J. Sargent (2008), “Two questions about European unemployment.” Econometrica, 76 (1), 1–29. [735,736,737,738,740,745,747,748,756,763,764] Machin, S. and A. Manning (1999), “The causes and consequences of long-term unemployment in Europe.” Handbook of Labor Economics, 3, 3085–3139. [740] Mankart, J. and R. Oikonomou (2017), “Household search and the aggregate labour market.” Review of Economic Studies, 84 (1), 1735–1788. [740] Marimon, R. and F. Zilibotti (1999), “Unemployment vs. mismatch of talents: Reconsidering unemployment benefits.” The Economic Journal, 109 (455), 266–291. [739] McCall, J. J. (1970), “Economics of information and job search.” Quarterly Journal of Economics, 84 (1), 113–126. [745] Mortensen, D. T. and C. A. Pissarides (1994), “Job creation and job destruction in the theory of unemployment.” Review of Economic Studies, 61 (3), 397–415. [737,745] Mortensen, D. T. and C. A. Pissarides (1999), “Unemployment responses to ‘skill-biased’ technology shocks: The role of labour market policy.” The Economic Journal, 109 (455), 242–265. [739] Nickell, S., L. Nunziata, and W. Ochel (2005), “Unemployment in the OECD since the 1960s: What do we know?” The Economic Journal, 115 (500), 1–27. [739] Petrongolo, B. and C. A. Pissarides (2001), “Looking into the black box: A survey of the matching function.” Journal of Economic Literature, 39 (2), 390–431. [756] Prescott, E. C., R. Rogerson, and J. Wallenius (2009), “Lifetime aggregate labor supply with endogenous workweek length.” Review of Economic Dynamics, 12 (1), 23–36. [736] Pries, M. and R. Rogerson (2009), “Search frictions and labor market participation.” European Economic Review, 53 (5), 568–587. [740] Ravikumar, B. and G. Vandenbroucke (2017), “Why are life-cycle earnings profiles getting flatter?” Federal Reserve Bank of St. Louis Review, 99 (3), 245–257. [762] Rogerson, R. and R. Shimer (2011), “Search in macroeconomic models of the labor market.” Handbook of Labor Economics, 4, 619–700. [740]
784 Etienne Lalé Quantitative Economics 9 (2018) Shimer, R. (2013), “Job search, labor force participation, and wage rigidities.” In Advances in Economics and Econometrics: Volume 2, Applied Economics: Tenth World Congress, Vol. 50, 197–234, Cambridge University Press. [740] Co-editor Karl Schmedders handled this manuscript. Manuscript received 26 March, 2015; final version accepted 24 October, 2017; available online 5 December, 2017.