Durations in Panel Data Subject to Attrition: A Note on Estimation in the Case of a Stock Sample1

In the empirical analysis of unemployment durations or job durations, it is generally assumed that the stochastic processes underlying labour market behaviour and the behaviour concerning participation in a panel survey are independent. As a consequence, spells that are incomplete due to attrition can be treated as spells that are subjected to independent right censoring. However, if the assumption of independence is violated, i.e. if for example the probability of dropping out of the panel is related to the rate at which a job is found, then attrition may have to be modelled and estimated jointly with the unemployment duration distribution to avoid biased estimates of the rate at which individuals become employed. A way to model the joint dependence is by means of stochastically related unobserved determinants. We discuss some properties of these kinds of models and state conditions needed to estimate such models in the case of stock sampled duration data.     

Estimation of sons' intergenerational earnings mobility in the presence of censoring

Based on intergenerational data from the Panel Study of Income Dynamics, I provide estimates of income mobility across generations for men. Special attention is paid to issues of censoring caused by a son's unemployment. Employing non‐parametric bounds estimates, I illustrate that previous income mobility estimates rely heavily on (unjustified) assumptions of exogenous selection. Assuming a son's potential income is instead a function of his reason for unemployment and work history, I re‐estimate mobility. Allowing for sampling variability, the range of feasible slopes consistent with these modified bounds restrictions is 0.27 to 0.55. Copyright © 2002 John Wiley & Sons, Ltd.     

Socio‐economic distance and spatial patterns in unemployment

This paper examines the spatial patterns of unemployment in Chicago between 1980 and 1990. We study unemployment clustering with respect to different social and economic distance metrics that reflect the structure of agents' social networks. Specifically, we use physical distance, travel time, and differences in ethnic and occupational distribution between locations. Our goal is to determine whether our estimates of spatial dependence are consistent with models in which agents' employment status is affected by information exchanged locally within their social networks. We present non‐parametric estimates of correlation across Census tracts as a function of each distance metric as well as pairs of metrics, both for unemployment rate itself and after conditioning on a set of tract characteristics. Our results indicate that there is a strong positive and statistically significant degree of spatial dependence in the distribution of raw unemployment rates, for all our metrics. However, once we condition on a set of covariates, most of the spatial autocorrelation is eliminated, with the exception of physical and occupational distance. Racial and ethnic composition variables are the single most important factor in explaining the observed correlation patterns. Copyright © 2002 John Wiley & Sons, Ltd.     

Fractional Integration and the Dynamics of UK Unemployment*

This article is concerned with estimating the dynamic behaviour of UK unemployment using fractional integration methods. The question it considers is whether an unemployment model using a relatively small set of determinants of unemployment is consistent with the persistence which estimates an integrated model yield, or the much long‐lasting estimates obtained from fractional integration. Our empirical tests favour the latter version. The results show that when accounting for UK unemployment in terms of lagged values of the real oil price and the real interest rate, unemployment appears fractionally integrated. This finding means that although unemployment is mean reverting, once it is shocked it may take a very long time to recover, and our estimates of the response times are considerably in excess of those previously reported in the literature.     

Analysis of Hybrid Life-tests in Presence of Competing Risks

The mixture of type-I and type-II censoring schemes, called the hybrid censoring scheme is quite common in life-testing or reliability experiments. In this paper, we consider the competing risks model in presence of hybrid censored data. Under this set up, it is assumed that the item may fail due to various causes and the corresponding lifetime distributions are independent and exponentially distributed with different scale parameters. We obtain the maximum likelihood estimators of the mean life of the different causes and derive their exact distributions. Using the exact distributions, all the moments can be obtained. Asymptotic confidence intervals and two bootstrap confidence intervals are also proposed. Bayes estimates and credible intervals of the unknown parameters are obtained under the assumptions of independent inverted gamma priors of the mean life of the different causes. Different methods have been compared using Monte Carlo simulations. Onereal data set has been analyzed for illustrative purposes. Part of the work was supported by a grant from the Natural Sciences and Engineering Research Council.     

Local Labour Markets and Theft: New Evidence from Canada

This paper provides causal evidence that labour market opportunities affect theft‐related crime rates in Canada. Synthetic panel data from 2007–2011 combine the Labour Force Survey and Uniform Crime Reports microdata. Low‐skill unemployment rates and corresponding crime rates are measured for age‐city‐specific groups of young males. IV estimates exploit the exposure of low‐skill employment to exogenous demand for exports to the US. Causal estimates of the elasticity of theft‐related crimes with respect to low‐skill unemployment range from 0.357 to 0.654. The use of aggregated unemployment rates appears to bias OLS estimates downward. IV estimates are found to mitigate this aggregation bias.     

Wage Formation,Regional Migration and Local Labour Market Tightness*

Empirical studies of regional wage formation and interregional migration routinely include the regional unemployment rate as indicator of local labour market tightness. However, these studies are usually motivated by economic theories that emphasize transition probabilities between unemployment and employment, and the unemployment rate is an imperfect proxy for these probabilities. We use a large micro data set to compute estimates of the rate of outflow from unemployment for 90 Norwegian travel‐to‐work areas. The outflow rates perform better than traditional measures of regional labour market tightness in panel data analyses of regional wages and interregional migration.     

Unemployment or credit: Which one holds the potential? Results for a small open economy with a low degree of financialization

This paper investigates the relative importance of unemployment versus credit in determining the potential level of real activity for a small open economy with a low degree of financialization. We use a multivariate unobserved component model (MUC) to derive the potential output and the associated output gap for the Lithuanian economy. The model is estimated via Bayesian methods and the time paths of unobserved variables are extracted via the Kalman filter. The inclusion of unemployment in the MUC model substantially improves the estimates of the output gap in real time. Adding information about credit further emphasizes the overheating of the economy in the pre-crisis period, both in real time and ex post. Including credit preserves the conclusions regarding turning points. We uncover a strong negative correlation between the model-implied unemployment gap (without accounting for credit) and real credit growth. Data revisions do not appear to be the primary source of revisions of output gap estimates.     

Inference on endogenously censored regression models using conditional moment inequalities

Under a quantile restriction, randomly censored regression models can be written in terms of conditional moment inequalities. We study the identified features of these moment inequalities with respect to the regression parameters where we allow for covariate dependent censoring, endogenous censoring and endogenous regressors. These inequalities restrict the parameters to a set. We show regular point identification can be achieved under a set of interpretable sufficient conditions. We then provide a simple way to convert conditional moment inequalities into unconditional ones while preserving the informational content. Our method obviates the need for nonparametric estimation, which would require the selection of smoothing parameters and trimming procedures. Without the point identification conditions, our objective function can be used to do inference on the partially identified parameter. Maintaining the point identification conditions, we propose a quantile minimum distance estimator which converges at the parametric rate to the parameter vector of interest, and has an asymptotically normal distribution. A small scale simulation study and an application using drug relapse data demonstrate satisfactory finite sample performance.     

Using a Structural‐Form Model to Analyze the Impact of Home Ownership on Unemployment Duration

It is often found that the impact of home ownership on the hazard rate for leaving unemployment is positive, indicating that home ownership helps workers to leave unemployment for a paid job. However, little emphasis has been given to how such a relationship can be explained. This paper estimates a structural‐form model that allows for self‐selection into home ownership and the risk of home owners losing their property during a spell of unemployment. We find a substantial amount of self‐selection using indirect inference based on a mixed proportional hazards‐rate model and find virtually no impact of home ownership on individual labor market performance. Copyright © 2016 John Wiley & Sons, Ltd.     

Equilibrium Unemployment Dynamics in a Panel of OECD Countries

We focus on the equilibrium unemployment rate as a parameter implied by a dynamic aggregate model of wage and price setting. The equilibrium unemployment rate depends on institutional labour market institutions through mark‐up coefficients. Compared with existing studies, the resulting final equation for unemployment has a richer dynamic structure. The empirical investigation is conducted in a panel data framework and uses OECD data up to 2012. We propose to extend the standard estimation method with time dummies to control and capture the effects of common and national shocks by using impulse indicator saturation (WG‐IIS), which has not been previously used on panel data. WG‐IIS robustifies the estimators of the regression coefficients in the dynamic model, and it affects the estimated equilibrium unemployment rates. We find that wage co‐ordination stands out as the most important institutional variable in our data set, but there is also evidence pointing to the tax wedge and the degree of compensation in the unemployment insurance system as drivers of equilibrium unemployment.     

The German federal employment program for regions with special employment problems: An evaluation

A special labor-market program to support regions with high unemployment was introduced in the Federal Republic of Germany in the summer of 1979. The program is evaluated through analyses of the determinants of the take-up rate and estimates of the impact on the unemployment rate using Box-Jenkins analysis. Evaluation of the program indicates some success with respect to training and public works creation, but also some serious shortcomings in design and implementation.     

A Logit Model with Missing Information Illustrated by Testing for Hidden Unemployment in Transition Economies

In an important paper, Dempster, Laird and Rubin (1977) showed how the expectation maximization (EM) algorithm could be used to obtain maximum likelihood estimates of parameters in a multinomial probability model with missing information. This article extends Dempster, Laird and Rubin's work on the EM algorithm to the estimation of a multinomial logit model with missing information on category membership. We call this new model the latent multinomial logit (LMNL) model. A constrained version of the LMNL model is used to examine the issue of hidden unemployment in transition economies following the approach of Earle and Sakova (2000) . We found an additional 0.5% hidden unemployment among workers describing themselves as self‐employed in the transition economies of Central and Eastern Europe.     

New Insights into Unemployment Duration and Post Unemployment Earnings in Germany*

Building on non‐stationary search theory ( Mortensen, 1977 ; Van den Berg, 1990 ), this article estimates the effects of UB on unemployment durations and future earnings using unique administrative data in Germany. We apply censored Box–Cox quantile regression. Our results imply that the length of entitlement shows only a weak effect on unemployment duration for entitlement lengths up to 12 months and no effect on post unemployment earnings. There are noticeable effects on exits from unemployment for entitlement lengths above 12 months. A high wage replacement rate for low‐wage earners is associated with a longer duration of unemployment and higher post unemployment earnings.     

Reconciling theory and empirics on the role of unemployment in mortgage default

Empirical models of mortgage default typically find that the influence of unemployment is negligible compared to other well known risk factors such as high borrower leverage or low borrower FICO scores. This is at odds with theory, which assigns a critical role to unemployment in the decision to stop payment on a mortgage. We help reconcile this divergence by employing a novel empirical strategy involving simulated unemployment histories to measure the severity of attenuation bias in loan-level estimations of default risk due to a borrower becoming unemployed. Attenuation bias results because individual data on unemployment status is unobserved, requiring that a market-wide unemployment rate be used as a proxy. Attenuation is extreme, with our results suggesting that the use of an aggregate unemployment rate in lieu of actual borrower unemployment status results in default risk from a borrower becoming unemployed being underestimated by a factor more than 100. In addition, our analysis indicates that adding the unemployment rate as a proxy for the missing borrower-specific unemployment indicator does not improve the accuracy of the estimated model over the specification without the proxy variable included. Hence, aggregate portfolio-level risk estimates for mortgage guarantors such as FHA also are not improved.These views represent those of the authors and not necessarily those of the Federal Reserve Bank of New York or the Federal Reserve System. This is a revised version of a paper that previously circulated under the title “Unemployment and Unobserved Credit Risk in the FHA Single Family Mortgage Insurance Fund (NBER Working Paper No. 18880). John Grigsby provided excellent research assistance. We appreciate the helpful comments of Andrew Haughwout, Wilbert van der Klaauw, the editor (Stuart Rosenthal) and referees, but remain responsible for any errors.     

A Bounded Model of Time Variation in Trend Inflation,Nairu and the Phillips Curve

In this paper, we develop a bivariate unobserved components model for inflation and unemployment. The unobserved components are trend inflation and the non‐accelerating inflation rate of unemployment (NAIRU). Our model also incorporates a time‐varying Phillips curve and time‐varying inflation persistence. What sets this paper apart from the existing literature is that we do not use unbounded random walks for the unobserved components, but rather bounded random walks. For instance, NAIRU is assumed to evolve within bounds. Our empirical work shows the importance of bounding. We find that our bounded bivariate model forecasts better than many alternatives, including a version of our model with unbounded unobserved components. Our model also yields sensible estimates of trend inflation, NAIRU, inflation persistence and the slope of the Phillips curve. Copyright © 2015 John Wiley & Sons, Ltd.     

The wage scar from male youth unemployment

We utilise the National Child Development Survey to analyse the impact of youth unemployment upon the wage up to twenty years later. We find a large and significant wage penalty, even after controlling for education, region and a wealth of family and individual characteristics. Our estimates are robust to an instrumental variables technique, indicating that the relationship estimated between youth unemployment and the wage is causal. Our results suggest a scar from early unemployment in the magnitude of 13–21% at age 42. However, this penalty is lower, at 9–11%, if individuals avoid repeat exposure to unemployment.     

Information shocks and the empirical evaluation of training programs during unemployment spells

We study the role of notifications in the evaluation of training programs for unemployed workers. Using a unique administrative data set containing the dates when information is exchanged between job seekers and caseworkers, we address three questions: Do information shocks, such as notification of future training, have an effect on unemployment duration? What is the joint effect of notification and training programs on unemployment? Can ignoring information shocks lead to a large bias in the estimation of the effect of training programs? We discuss these issues through the lens of a job search model and then conduct an empirical analysis following a “random effects” approach to deal with selectivity. We find that notification has a strong positive effect on the training probability but a negative one on the probability of leaving unemployment. This “attraction” effect highlights the importance of accounting for notifications in the evaluation of active labor market policies.     

On generalized progressive hybrid censoring in presence of competing risks

The progressive Type-II hybrid censoring scheme introduced by Kundu and Joarder (Comput Stat Data Anal 50:2509–2528, 2006), has received some attention in the last few years. One major drawback of this censoring scheme is that very few observations (even no observation at all) may be observed at the end of the experiment. To overcome this problem, Cho et al. (Stat Methodol 23:18–34, 2015) recently introduced generalized progressive censoring which ensures to get a pre specified number of failures. In this paper we analyze generalized progressive censored data in presence of competing risks. For brevity we have considered only two competing causes of failures, and it is assumed that the lifetime of the competing causes follow one parameter exponential distributions with different scale parameters. We obtain the maximum likelihood estimators of the unknown parameters and also provide their exact distributions. Based on the exact distributions of the maximum likelihood estimators exact confidence intervals can be obtained. Asymptotic and bootstrap confidence intervals are also provided for comparison purposes. We further consider the Bayesian analysis of the unknown parameters under a very flexible beta–gamma prior. We provide the Bayes estimates and the associated credible intervals of the unknown parameters based on the above priors. We present extensive simulation results to see the effectiveness of the proposed method and finally one real data set is analyzed for illustrative purpose.     

Labor quality,natural unemployment,and US inflation

The recent combination of low inflation and low unemployment has led some to question whether the short-run, Phillips curve trade-off is dead. We argue that the improved trade-off has resulted, in part, from improved labor quality in the form of increased average years of work experience and education, and use these variables to calculate new estimates of the natural unemployment rate. Based on evidence from inflation equations, we find strong support for a time-varying natural unemployment rate, and find that our measure based on labor quality outperforms other leading measures of natural unemployment.