Endogenous Initial Conditions Research Articles

Incomplete ordinal panel responses with initial condition and skew-normal random effect

This paper presents a model for the ordinal categorical panel data with skew-normal random effects, endogenous initial condition, and non ignorable (NI) non monotone missing data. We assumed three separate models for the initial condition, subsequent responses, and missing procedure. The baseline or initial response variable and subsequent response variables are modeled as ordinal logistic, whereas the logistic binary model is considered for the missing process where the maximum likelihood estimates are obtained using Monte Carlo methods. The results show that in the presence of missing data, the skew-normal random effects produce better results compared to the normal one, and considering the endogenous initial conditions is effective in modeling the results of the observed and missing responses such that the patients whose severity of schizophrenia is higher at the beginning of the process have a higher chance of missing on the subsequence times. In this research, schizophrenia data and then simulation studies have been used to check the performance of the proposed model. The results show that the skew-normal random effect using the NI non monotone missing mechanism by increasing the skewness value produces better results in the analysis of simulation studies.

Do supplementary jobs for welfare recipients increase the chance of welfare exit? Evidence from Germany

AbstractWelfare recipients in Germany are allowed to take up supplementary jobs while receiving welfare. In the present study, we use the German Panel Study “Labour Market and Social Security” (PASS) for the years 2006–2014 to analyze the impact of these supplementary jobs on the chances of welfare exit. Dynamic multinomial logit models controlling for unobserved heterogeneity and endogenous initial conditions reveal that full‐time employed men and women are more likely to exit welfare into employment than their non‐employed counterparts. For supplementary part‐time jobs, however, we find no or (only in some specifications for men) much smaller stepping stone effects.

Child Poverty in Croatia, Serbia, and Slovenia: The Role of Poverty Persistence and Social Transfers

ABSTRACT We use a dynamic correlated random effects probit model with endogenous initial conditions for three poverty measures to assess child poverty persistence after controlling for structural household characteristics and economic variables. The empirical analysis is based on the 2015–2018 longitudinal sample of the Survey of Income and Living Conditions for Slovenia, Croatia, and Serbia. Results show strong poverty persistence regardless of poverty measure, with much stronger effects observed for Croatia and Serbia than for Slovenia. Social transfers have more than twice the capacity to lift children out of poverty in Slovenia compared to the other two countries.

The dynamics of poverty in Europe: what has changed after the great recession?

This paper provides novel evidence on the importance of the phenomenon of poverty and its heterogeneity across European countries. We analyze the determinants of poverty in Europe and their evolution over time by disentangling the role of genuine state dependence and heterogeneity.We apply alternative dynamic probit models accounting for endogenous initial conditions and correlated random effects to the pre-Great Recession period of 2005–2008 and the post- Great Recession period of 2015–2018 using EU-SILC longitudinal datasets for a sample of European countries in order to estimate genuine state dependence and uncover the role of observable and unobservable factors in determining the risk of poverty. Our findings suggest that the degree of genuine state dependence is relevant in Europe and that it increased slightly from pre- to post-Great Recession. This suggests that measures aimed at lifting individuals out of poverty, including cash transfers, have become even more important during the Europe 2020 decade. Our analysis also reveals that Europe is characterized by an increasing scarring effect of poverty, the trend of which has been exacerbated in the post-recession period. The analysis at the country level clarifies why the evolution of genuine state dependence was heterogeneous. While a clear pattern within macro-regions does not emerge, we find an association between country-level variation in genuine state dependence and some macroeconomic indicators. Finally, our results suggest that the protective role of higher education has diminished over time, while the role of employment stability and of childcare provision during early childhood has become even more important in the post-recession period.

Multilevel Hidden Markov Models for Behavioral Data: A Hawk-and-Dove Experiment

Motivated by the analysis of behavioral data taken from an economic experiment based on the Hawk-and-Dove game, this article describes a multilevel hidden Markov model, that includes covariates, autoregression, and endogenous initial conditions under a unified framework. The data at hand are affected by multiple sources of latent heterogeneity, due to multilevel unobserved factors that operate in conjunction with observed covariates at all the levels of the data hierarchy. We fit a multilevel logistic regression model for repeated measurements of player behaviors, nested within groups of interacting players. The model integrates discrete random effects at the group level and Markovian sequences of discrete random effects at the player level. Parameters are estimated by a computationally feasible expectation-maximization algorithm. We model the probability of playing the Hawk strategy, which implies fighting aggressively for controlling an asset, and test the role played by initial possession, property, and other player-specific characteristics in driving hawkish behaviors. The results from our study suggest that crucial factors in determining hawkish behavior are both the way possession is achieved – which depends on our treatment manipulation- and possession itself. Furthermore, a clear time-dependence is observed in the data at the player level as accounted for by the Markovian random effects.

Assessing poverty persistence in households with children

The analysis of poverty persistence received considerable attention in recent years. In this paper we explore the role of the adopted poverty measure in the analysis of its persistence. Specifically, we consider three measures: the risk of poverty, the severe material deprivation and subjective poverty, motivated by the understanding of poverty as a complex phenomenon and for which no single measure can effectively capture its several dimensions. The empirical analysis is based on the 2013-2016 longitudinal sample of the EU-SILC survey. We focus on Italian households with dependent children. We apply a correlated random effects probit models with endogenous initial conditions to assess genuine state dependence after controlling for structural household characteristics and variables related to participation in the labour market. A strong state dependence emerges, regardless of the considered poverty measure thus providing evidence of poverty and social exclusion persistence. We also find evidence of relevance of initial conditions for all measures in focus. Nonetheless, structural household characteristics and household level economic variables play roles that are often different in the three parallel models; these differences are consistent with the aims and nature of the alternative measures.

Local Knowledge Spillovers and Innovation Persistence of Firms

Recent empirical evidence has shown that firm’s innovation behavior exhibits high persistency but not much is known about potential contingencies affecting the degree of persistence. This paper focuses on the role of the local knowledge environment and asks how local knowledge spillovers affect firms' innovation persistence. The empirical analysis draws upon a representative panel data set of firms in Germany from 2002-2016, complemented by detailed geographic information of patent activity over discrete distances to proxy local knowledge spillovers. Based on correlated random effects probit models that control for state dependence, unobserved individual heterogeneity and endogenous initial conditions, our results corroborate former evidence that persistency in innovation is driven by true state dependence. More importantly, we find that the local patenting activity positively moderates firms' degree of persistency in innovation behavior. This is a novel firm-level mechanism that can explain the widening of spatial disparities in innovation performance. Estimations with different distance bands show that the strength of knowledge spillovers that contribute to innovation persistence via true state dependence declines rather rapidly with increasing distance.

State Dependence in Welfare Benefits in a Non‐Welfare Context

This study investigates state dependence in social assistance benefits in Turkey, where benefit receipt and persistence rates have significantly increased over the past decade. We estimate state dependence through dynamic random‐effects probit models, controlling for observed and unobserved heterogeneity, and endogenous initial conditions. In particular, we employ Wooldridge’s estimator to achieve consistent and correct estimates of state dependence and compare the results with estimates from Heckman’s reduced‐form approach as a sensitivity check. Both estimators enable us to disentangle true state dependence from its spurious components and address the potential bias due to the short panel length. Our results suggest that the receipt of benefits in the last year increases the likelihood of benefit receipt in the current year, namely the structural state dependence, by 17.2–19.5 percentage points.

Does the Presence of a Disabled Person in the Household Affect the Employment Probabilities of Cohabiting Women? Evidence from Italy, France and the UK

This paper investigates how the presence of a disabled person in the household affects the employment probabilities of cohabiting women. Using a unique data source and a dynamic probit model accounting for unobserved heterogeneity and endogenous initial conditions, we analyze Italy, France, and the UK, three countries that diverge substantially in terms of welfare system regimes, family and employment policies, and social norms. In line with care theory suggestions, we find that in Italy, where formal caring services are limited, and the male breadwinner model persists, women see reduced employment possibilities when cohabiting with disabled persons. In France and the UK, where family and employment policies, such as low-cost formal care and part-time jobs, provide some support for women in reconciling unpaid and paid work, the presence of a disabled person increases employment probabilities. In disentangling the contributions of disability benefits, it appears that they might provide financial resources to support formal caregiving and, therefore, mitigate the caregiving responsibilities of women.

Disability and Social Exclusion in Italian Households

This paper investigates how the presence of disabled member(s) affects the household risk of being socially excluded in Italy. By using a dynamic probit model with correlated random effects and accounting for endogenous initial conditions, we find that the presence of severely disabled member(s) increases the probability of being socially excluded by 2.5%. Importantly, we find that genuine state dependence is greater, by about 20% in relative terms, for households with disabled member(s) than their counterparts, indicating that the former experience greater difficulties in escaping from social exclusion. This suggests that policies aimed at preventing social exclusion would be particularly effective for households with disabled member(s). All these findings hold when accounting for endogeneity of disability and social exclusion because of possible feedback effects. Finally, our estimates confirm that in Italy low education, living in the South and the presence of children increase the probability of a household being socially excluded.

Social capital and having a regular family doctor: Evidence from longitudinal data

Access to a regular source of health-care is problematic for some, irrespective of whether the regime is publicly or privately funded. Yet, evidence shows that access to a regular family doctor improves health outcomes. We are the first to examine the impact of social capital (e.g., tangible support, friends and family) on having a regular family doctor taking into account that social capital may be endogenously determined. Using the Canadian National Population Health longitudinal survey (1994–2010: n = 41,022) and a dynamic random effects probit model (with and without endogenous initial conditions) we find robust evidence of a statistically significant and positive causal relationship between social capital and the probability of having a regular family doctor. Since past access to a family doctor is a strong predictor of both current and future access, we show that social capital is much more important in helping individuals find a family doctor than for keeping one.

Importing exporters and exporting importers: A study of the decision of Chinese firms to engage in international trade

AbstractThis paper examines the complex and interdependent relationship between importing and exporting for a panel of Chinese manufacturing firms. We estimate the decision to import and export simultaneously within a dynamic random‐effects bivariate probit framework addressing the endogenous initial conditions problem. Results show that decisions to export and import are simultaneously determined and that sunk‐entry costs play a significant role in a firm's decision to enter international markets. Costs are larger for exporting. We also find a substitution effect between the two decisions. The substitutability between exporting and importing is greater for financially constrained private firms.

Managing Wildlife Faced with Pathogen Risks Involving Multi-Stable Outcomes.

Most models designed to understand how to manage infected wildlife systems with bioeconomic multi-stability take the initial conditions as given, thereby treating pathogen invasion as unanticipated. We examine how ex ante management is an opportunity to influence the ex post conditions, which in turn affect the ex post optimal outcome. To capture these ex ante management choices, we extend the Poisson “collapse” model of Reed and Heras (Bull Math Biol 54:185–207, 1992) to allow for endogenous initial conditions and ex post multi-stability. We account for two uncertain processes: the introduction and establishment of the pathogen. Introduction is conditional on anthropogenic investments in prevention, and both random processes are conditional on how we manage the native population to provide natural prevention of invasion and natural insurance against establishment placing the system in an undesirable basin of attraction. We find that both multi-stability of the invaded system and these uncertainty processes can create economic non-convexities that yield multiple candidate solutions to the ex ante optimization problem. Additionally, we illustrate how the nature of natural protection against introduction and establishment risks can play an important role in the allocation of anthropogenic investments.

Are there differences in persistence across different innovation measures?

In this study we empirically examine the persistence across input and output indicators of innovation using a panel of Spanish manufacturing firms during the period 1990–2013. We use R&D as innovation input and technological innovation, product innovation and process innovation as output measures. We analyse the transition matrices and design an econometric strategy consisting of the estimation of various specifications using random effects dynamic probit models that account for state dependence, unobserved heterogeneity and endogenous initial conditions. Findings indicate the presence of true persistence in all our innovation indicators. They also show the existence of different degrees of persistence depending on the innovation measure used. Persistence in R&D increases around 20% the probability of conducting R&D. The highest persistence is associated with technological innovation. The probability of introducing a technological or process innovation is 28% higher for firms that have innovated in the previous year. When technological innovation is broken down into product and process innovation, the degree of persistence is higher for process innovation than for product innovation. This result is not unexpected considering the Spanish productive structure, which is made up mainly of low and middle-low technological sectors which are based on process innovations rather than on product innovations.

Understanding Export Market Success: Evidence from Manufacturing Firms

In this paper, we analyse empirically the determinants of the decision to export, and the export share of manufacturing firms using a large balanced panel database of firms from Spain, spanning from 1990 to 2012. We are particularly interested in investigating these determinants by controlling for state dependence, unobserved heterogeneity and endogenous initial conditions of firms. Our findings reveal that firms show a high persistence in export behaviour, and that export volume changes are determined primarily by changes in the intensive margin. In addition, our research suggests that having exported before, having achieved a certain productivity level and being large and foreign-owned, greatly increased the probability of exporting. Moreover, these results hold even for the export share of firms, except for productivity and foreign capital, which seem to be less relevant.

State Dependence in Welfare Benefits in a Non-Welfare Context

This study contributes to the ongoing debate about welfare dependency centered on the western societies through an empirical analysis, within the context of a developing country. It examines state dependence in social assistance benefit receipt using longitudinal data from Turkey, where benefit receipt and persistence rates have witnessed a significant increase since the last decade. Identification is achieved by dynamic random effects probit models, controlling for unobserved heterogeneity and endogenous initial conditions. Particularly, Wooldridge's (2005) estimator and its extensions are applied for achieving consistent and correct estimates of state dependence. In order to check for sensitivity, the results are compared with the results from Heckman's (1981) reduced form approach. Both estimators enable us to deal with the potential bias due to the short panel length. It is found that the benefit receipt of the last year increases the likelihood of benefit receipt in the current year by 17 to 22 percentage points. This evidence suggests that state dependence in social assistance might also be a relevant phenomenon for developing countries.

A dynamic hurdle model for zeroinflated panel count data

This article proposes an approximate conditional dynamic finite mixture hurdle model for panel count data with excess of zeros and endogenous initial conditions. We provide parameter estimates by using the Expectation-Maximization (EM) algorithm in a Nonparametric Maximum Likelihood (NPML) framework. An application to a unique data set on traffic violation counts of a subpopulation of Italian drivers is given.

Young people, the disabled, and immigrant workers

PurposeThe purpose of this paper is to introduce the special issue on “Disadvantaged workers in the labour market” and to draw some policy implications that are common to the studies included in the special issue.Design/methodology/approachThe paper introduces several econometric methodologies adopted in the special issue. Such methodologies represent an excellent overview of the typical approaches developed by applied economists to address the issue under discussion. They include: dynamic probit models allowing for unobserved heterogeneity and endogenous initial conditions; propensity score matching; the non parametric Ñopo decomposition to take into account sample selection bias; endogenous switching models to estimate employment choice; and fixed effects models and Arellano and Bond dynamic panel estimator.FindingsThe articles in the special issue highlight several common factors that make the disadvantage at the labor market emerge. The first factor is the recent financial crisis which especially affects young people. Differences in common support – generally due to segregation in low pay jobs – are also important. Several factors favor persistence of the disadvantage, namely the inefficiency of the educational system and the lack of financial support to the weakest groups.Practical implicationsThe paper adopts the EU definition of “disadvantaged workers”, namely workers who find it hard to enter the labor market without assistance. The authors show the relevance of the analysis produced in this special issue to design specific policies fine‐tuned to the needs of disadvantaged workers. In addition, institutional reforms, and/or interventions on structural variables are also in order.Originality/valueThe paper introduces a number of articles proposing innovations in the interpretation and application of a wide range of theoretical approaches and econometric methodologies.

Three Essays on Health and Labor Outcomes

This dissertation is composed of three essays which examine the effects of health on labor market outcomes. Chapter 1 reviews the literature on health and the labor market. It also emphasizes the inherent endogeneity of health when included in models for labor market outcomes. It goes on to highlight the empirical methods most often used to accommodate that endogeneity. In chapter 2, I use 2000 to 2007 data from the Medical Expenditure Panel Survey (MEPS) to examine the role of health status in decisions to transition to self-employment. Much of the past literature has incorporated health status in models for self-employment in a perfunctory fashion. I account for unobserved heterogeneity and endogenous initial conditions using a discrete factor random effects model. Three hypotheses for the direct effect of health on the self-employment decision are put forth. The indirect effect that health may have in determining one’s valuation of health insurance coverage is controlled for in the model. Regression results indicate that individuals who experience any sort of functioning limitation, or who report relatively poorer health, are more likely to transition to self-employment over wage-employment, holding all else constant. Although the magnitude of the impact of health status varies between two sub-groups of the population studied. Chapter 3 examines the extent to which a spouse’s ill-health influences the labor supply decisions of the older men and women. Spouses’ ill-health is likely to affect their partner’s labor supply decision in off-setting ways. I control for the income effect due to the increase in the probability of an ill spouse to leave the labor force. Therefore, my estimates reflect the direct impact of a spouse’s ill-health on the partner’s labor supply decision through its effect on the partner’s reservation wage. However, it is likely that spouses’ earnings are endogenous in these models due to unobserved characteristics common to husbands and wives. I find that the estimated effect of a wife’s ill health on their partner’s labor supply decision is dependent on whether I instrument the spouse’s earnings. I also find that the estimated effect of husbands’ and wives’ ill health on their partners’labor supply decision is dependent on the health measure used in the models.

Immigrant-Native Differences in Earnings Mobility Processes: Evidence from Canadian and Danish Data

This study compares the earnings mobility between immigrants and natives within and between Denmark and Canada. Both countries have different labour market conditions and immigration history which leads to an interesting comparison of earning mobility processes. The paper employs a dynamic multinomial logit model with discrete factor approximation for the specification of unobserved individual heterogeneity. The model takes into account the effect of the endogenous initial conditions problem and unobserved heterogeneity to separate structural and spurious state dependence. The results show that immigrants-native differences in earnings mobility, structural state dependence, and segmentation of earnings distribution are relatively more prominent in Denmark compared to Canada.