Bias corrections for two-step fixed effects panel data estimators

This paper introduces large-T bias-corrected estimators for nonlinear panel data models with both time invariant and time varying heterogeneity. These models include systems of equations with limited dependent variables and unobserved individual effects, and sample selection models with unobserved individual effects. Our two-step approach first estimates the reduced form by fixed effects procedures to obtain estimates of the time varying heterogeneity underlying the endogeneity/selection bias. We then estimate the primary equation by fixed effects including an appropriately constructed control variable from the reduced form estimates as an additional explanatory variable. The fixed effects approach in this second step captures the time invariant heterogeneity while the control variable accounts for the time varying heterogeneity. Since either or both steps might employ nonlinear fixed effects procedures it is necessary to bias adjust the estimates due to the incidental parameters problem. This problem is exacerbated by the two-step nature of the procedure. As these two-step approaches are not covered in the existing literature we derive the appropriate correction thereby extending the use of large-T bias adjustments to an important class of models. Simulation evidence indicates our approach works well in finite samples and an empirical example illustrates the applicability of our estimator.     

State minimum wages and business location: Evidence from a refined border approach

This study examines the effect of state minimum wage changes on new and existing business establishments. It employs a refined border approach in conjunction with other differencing methods to control for unobserved heterogeneous area characteristics. The findings suggest that state minimum wage increases deter new establishments from locating in an area, particularly in industries that rely on low-education workforces, such as the retail and manufacturing industries. However, existing establishments, regardless of industry type, are not found to be adversely affected by minimum wage policy.     

Noncognitive skills, occupational attainment, and relative wages

This paper examines whether men's and women's noncognitive skills influence their occupational attainment and, if so, whether this contributes to the disparity in their relative wages. We find that noncognitive skills have a substantial effect on the probability of employment in many, though not all, occupations in ways that differ by gender. Consequently, men and women with similar noncognitive skills enter occupations at very different rates. Women, however, have lower wages on average not because they work in different occupations than men do, but rather because they earn less than their male colleagues employed in the same occupation. On balance, women's noncognitive skills give them a slight wage advantage. Finally, we find that accounting for the endogeneity of occupational attainment more than halves the proportion of the overall gender wage gap that is unexplained.     

The Impact of Training on Productivity and Wages: Evidence from British Panel Data*

It is standard in the literature on training to use wages as a sufficient statistic for productivity. This paper examines the effects of work‐related training on direct measures of productivity. Using a new panel of British industries 1983–96 and a variety of estimation techniques we find that work‐related training is associated with significantly higher productivity. A 1% point increase in training is associated with an increase in value added per hour of about 0.6% and an increase in hourly wages of about 0.3%. We also show evidence using individual‐level data sets that is suggestive of training externalities.     

Seniority wages and establishment characteristics

A representative linked employer-employee panel and an innovative two-step estimation strategy are used to show that large and profitable establishments as well as establishments with a highly qualified workforce pay high seniority wages. Also collective bargaining coverage, works councils and reduced working time for older employees are positively correlated with seniority wages, the share of foreigners, females as well as initial wage levels for job entrants are negatively correlated. These results support an agency based motivation for seniority wages with older employees' wages set higher than their productivity.     

The family gap in wages: What wombmates reveal

We shed new light on the effects of having children on hourly wages by exploiting access to data on the entire population of employed same-sex twins in Denmark. Our second contribution is the use of administrative data on absenteeism; the amount of hours off due to holidays and sickness. Our results suggest that childbearing reduces female hourly wages but the principal explanation is in fact mothers’ higher levels of absence. We find a positive wage premium for fathers.     

City size and urban labor productivity in China: New evidence from spatial city-level panel data analysis

The previous literature has generally found that most Chinese cities were undersized in the 1990s. However, little is known about how urban agglomeration in China has evolved since the country began to experience much faster urbanization and deep marketization after 2000. Based on panel data of 281 cities between 2000 and 2013, our spatial panel data regression results show that the scale impact of city size on urban productivity appears to have an inverted U shape, and its effect will grow when the cities’ industrial structure becomes more dominated by the service sector. There are also significant spatial interactions and spatial heterogeneity of urban agglomeration among Chinese cities. Furthermore, we compute the optimal size for each city and find that most Chinese cities are still undersized in recent years. Based on our findings, we argue that the Chinese government should focus on promoting sufficient growth of medium sized cities in its new urbanization strategy.     

Tied Migration and Subsequent Employment: Evidence from Couples in Britain*

We use unique information on migration behaviour and reasons for migration to study the impact of tied migration on labour market outcomes among husbands and wives. Fewer than 2% of couples migrate for job‐related reasons and the majority of these move for reasons associated with the husband's job. Estimates from dynamic random‐effects models indicate that husbands and wives in couples that migrated for job‐related reasons suffer lower job retention rates than non‐migrants. Tied migration reduces the probability of subsequent employment for both husbands and wives and in particular has a large negative impact on job retention rates among wives.     

Negative equity and wages

We examine the relationship between housing equity and wage earnings using nine waves of the national American Housing Survey from 1985 to 2003. Employing a rich set of time and place controls, a synthetic mortgage instrumental variable strategy, and a first difference estimator we find that people underwater on their mortgage command a significantly lower wage than other homeowners. The finding survives a number of robustness checks for reverse causality and unobserved heterogeneity. We also explore other determinants of “house lock” including loss aversion, a low existing mortgage interest rate and property tax assessment caps, but do not find these factors mitigate the effect of negative equity on wages.     

The cyclicality of effective wages within employer-employee matches in a rigid labor market

This study analyzes real wage cyclicality for male full-time workers within employer-employee matches in Germany over the period 1984-2004. Five different wage measures are compared: the standard hourly wage rate; hourly wage earnings including overtime and bonus pay; the effective wage, which takes into account unpaid overtime; and monthly earnings, with and without additional pay. None of the hourly wage measures exhibits cyclicality except for the group of salaried workers with unpaid overtime. Their effective wages show a strongly procyclical reaction to changes in unemployment. Despite acyclical wage rates, salaried workers without unpaid overtime experienced procyclical earnings movements if they had income from extra pay. Monthly earnings were also procyclical for hourly paid workers with overtime pay. These findings suggest that cyclical earnings movements are generated by variable pay components, such as bonuses and overtime pay, and by flexible working hours. The degree of earnings procyclicality revealed for the German labor market is comparable to the United States.     

Estimation for spatial dynamic panel data with fixed effects: The case of spatial cointegration

Yu et al. (2008) establish asymptotic properties of quasi-maximum likelihood estimators for a stable spatial dynamic panel model with fixed effects when both the number of individuals n and the number of time periods T are large. This paper investigates unstable cases where there are unit roots generated by temporal and spatial correlations. We focus on the spatial cointegration model where some eigenvalues of the data generating process are equal to 1 and the outcomes of spatial units are cointegrated as in a vector autoregressive system. The asymptotics of the QML estimators are developed by reparameterization, and bias correction for the estimators is proposed. We also consider the 2SLS and GMM estimations when T could be small.     

Productivity,seniority and wages: new evidence from personnel data

Wages may be observed to increase with seniority because of firm-specific human capital accumulation or because of self-selection of better workers in longer jobs. In both these cases, the upward sloping wage profile in cross-sectional regressions would reflect higher productivity of more senior workers. If this were true, the observation of an effect of seniority on wages would depend on the presence of controls for individual productivity. In this paper we replicate, using personnel data from a large Italian firm, the results of the pioneering work of Medoff and Abraham [Quarterly Journal of Economics (1980); The Journal of Human Resources, 15(2) (1981)] in which supervisors' evaluations were used as productivity indicators. Since the validity of supervisors' evaluations as measures of productivity has been widely criticised, we extend the work of Medoff and Abraham using different direct measures of productivity based on recorded absenteeism and misconduct episodes. Both these indicators and supervisors' evaluation suggest that the observed effect of seniority on wages does not reflect a higher productivity of more senior workers. Only at the lowest levels of the firm's hierarchy, the human capital theory contributes to explain the effect of seniority on wages. At least at all other levels, the explanation of the observed upward sloping profile has to be based on theories in which wages are deferred for incentive or insurance reasons.     

Effect of Immigrant Nurses on Labor Market Outcomes of US Nurses

We study the effect of immigration of foreign-trained, registered nurses (RNs) on the employment and wages of US-trained RNs. We use the “area” approach and study effects of immigration in labor markets defined by the state. We find substantial evidence that immigration by foreign-trained nurses increases the supply of nurses and that this increase in supply is associated with a decrease in annual earnings. Estimates suggest that a 10% increase in supply due to immigration is associated with a 1-4% decrease in annual earnings, although most estimates were not statistically significant and we did not find a similar association between an increase in supply and wages.     

Job security and employee well-being: Evidence from matched survey and register data

We examine the effects of establishment- and industry-level labor market turnover on employees' well-being. The linked employer-employee panel data contain both survey information on employees' subjective well-being and comprehensive register-based information on job and worker flows. We test for the existence of compensating wage differentials by explaining wages and job satisfaction with average uncertainties, measured by an indicator for a high excessive turnover (churning) rate. The results are consistent with the theory of compensating wage differentials, since high uncertainty increases real wages, but high uncertainty has no effect on job satisfaction while not controlling for wages.     

Maternal employment,migration, and child development

We analyze the roles of and interrelationships among school inputs and parental inputs in affecting child development through the specification and estimation of a behavioral model of household migration and maternal employment decisions. We integrate information on these decisions with observations on child outcomes over a 13-year period from the National Longitudinal Study of Youth (NLSY). We find that the impact of our school quality measures diminishes by factors of 2 to 4 after accounting for the fact that families may choose where to live in part based on school characteristics and labor market opportunities. The positive statistical relationship between child outcomes and maternal employment reverses sign and remains statistically significant after controlling for its possible endogeneity. Our estimates imply that when parental responses are taken into account, policy changes in school quality end up having only minor impacts on child test scores.     

Twin Data vs. Longitudinal Data to Control for Unobserved Variables in Earnings Functions – Which Are the Differences?*

This paper compares two different approaches empirically to control for unobserved characteristics when estimating the effect of marriage on male and female earnings: the longitudinal and the twins approach. The estimates were obtained by exploiting the longitudinal dimension of a large sample of Swedish twins, so that longitudinal and twin‐based estimates could be obtained in the same sample. The two approaches lead to different conclusions both regarding the role of unobserved characteristics in the cross‐sectional earnings–marriage relationship and the effect of marriage on earnings. The paper investigates three potential explanations of this difference.     

Wage inequality, technology and trade: 21st century evidence

This paper describes and explains some of the principal trends in the wage and skill distribution in recent decades. Increases in wage inequality started in the US and UK at the end of the 1970s, but are now widespread. A good fraction of this inequality trend is due to technology-related increases in the demand for skilled workers outstripping the growth of their supply. Since the early 1990s, labor markets have become more polarized with jobs in the middle third of the wage distribution shrinking and those in the bottom and top third rising. I argue that this is because computerization complements the most skilled tasks, but substitutes for routine tasks performed by middle wage occupations such as clerks, leaving the demand for the lowest skilled service tasks largely unaffected. Finally, I argue that technology is partly endogenous, for example it has been spurred by trade with China. Thus, trade does matter for changes in the labor market, but through a different mechanism than conventionally thought.     

Local labor market impacts of energy boom-bust-boom in Western Canada

The impacts of energy price boom and bust are analyzed through the differential growth in employment and earnings between local labor markets with and without energy resources in Western Canada. The estimated differentials attributed to the boom-induced labor demand shocks show significant direct and indirect impacts on the earnings and employment within the energy extraction and other non-energy local sectors respectively. The local job multipliers indicate that job creation within the energy extraction sector leads to modest job creation within the non-energy local sectors during boom periods. For every ten energy extraction jobs created during a boom period, approximately three construction jobs, two retail jobs, and four and a half service jobs are created.     

Assessing inherent model bias: An application to native displacement in response to immigration

There is a long-standing debate among academics about the effect of immigration on native internal migration decisions. If immigrants displace natives this may indicate a direct cost of immigration in the form of decreased employment opportunity for native workers. Moreover, displacement would also imply that cross-region analyses of wage effects systematically underestimate the consequences of immigration. The widespread use of such area studies for the US and other countries makes it especially important to know whether a native internal response to immigration truly occurs. This paper introduces a microsimulation methodology to test for inherent bias in regression models that have been used in the literature. We show that some specifications have built biases into their models, thereby casting doubt on the validity of their results. We then provide a brief empirical analysis with a panel of observed US state-by-skill data. Together, our evidence argues against the existence of native displacement. This implies that cross-region analyses of immigration’s effect on wages are still informative.     

Differences in Opportunities? Wage,Employment and House‐Price Effects on Migration*

Most empirical studies of individual migration choice analyse factors associated with out‐migration from an origin location. In contrast, we model the migration decision within the context of potential destinations, combining British panel data over the period 1992–2008 with other data sources. Contrary to earlier micro studies, we show that differences in house prices levels (but not growth) are important determinants of household migration for homeowners. Unemployed individuals respond to regional differences in expected individual wages, whereas the employed are more sensitive to employment opportunities. Our evidence is consistent with partners of heads of households being tied migrants.