True State Dependence Research Articles

Self-Employment Dynamics, State Dependence and Cross-Mobility Patterns

This paper analyzes the mobility between self-employment, wage employment and non-employment. Using data for men in West Germany, we find strong true state dependence in all three states. Moreover, compared to wage employment, non-employment increases the probability of self-employment significantly, and self-employment goes along with a higher risk of future non-employment.

Persistence of innovation: stylised facts and panel data evidence

This paper investigates whether firms innovate persistently or discontinuously over time using an innovation panel data set on German manufacturing and service firms for the period 1994–2002. It turns out that innovation behaviour is permanent at the firm level to a very large extent. Using a dynamic random effects discrete choice model and a new estimator recently proposed by Wooldridge (2005), I further shed some light on the driving forces for this phenomenon. The econometric results show that past innovation experience is an important determinant for manufacturing as well as for service sector firms, and hence confirm the hypothesis of true state dependence. In addition, the results highlight the important role of knowledge provided by skilled employees and unobserved individual heterogeneity in explaining the persistence of innovation.

True versus spurious state dependence in firm performance

This paper analyzes the persistence of firms’ exporting behavior in a panel of German manufacturing firms using dynamic binary choice models. We distinguish between true and spurious state dependence in exports and apply fixed effects methods that allow us to verify the robustness of our results to critical assumptions on firms’ initial export status. We find robust evidence of state dependence in the current export status of firms. We also document an important role of unobserved permanent firm heterogeneity (spurious state dependence) and quantify the relative importance of different export determinants. Our results, which are consistent with the findings of previous studies on firms in developing countries and in the United States, show the presence of important sunk costs in export market entry and a depreciation of knowledge and experience in export markets.

Correcting for bias in retrospective data

AbstractWhen panel data are not available, retrospective data are used in the estimation of dynamic choice models. However, retrospective data are not reliable. Previous studies of voting choices, for example, have shown that respondents misreport their past choices in order to appear more consistent with their current choice. Such retrospective bias leads to inconsistent estimates, especially when there is state dependence in choices. Specifically, observed persistence in retrospective data may be due to (a) true state dependence, (b) unobserved heterogeneity, and (c) retrospective bias in reporting previous choices. Whereas Heckman in his 1981 study deals with (a) and (b), we introduce a method to estimate true state dependence while accounting for both unobserved heterogeneity and retrospective reporting bias. Our method is based on modeling the reporting behavior and integrating it into the estimation. The identification strategy is based on the correlation between the reported previous choices and current exogenous variables. Using data on Israeli voters, we find that the probability that a respondent whose vote intention in 1991 differed from his or her past voting choices would lie about their past choices is 0.23. Copyright © 2007 John Wiley & Sons, Ltd.

Utilities Deprivation Dynamics and Energy Sector Reforms in Europe

In the late 1990s many European countries started comprehensive restructuring of their energy industries, the typical ingredients of the reforms are full or partial privatization, vertical disintegration, liberalization. In this paper we focus on the way in which energy sector reforms affect social affordability. The aim of this paper is to analyze the effects of energy reforms on the household probability of experiencing utilities deprivation (that is, to be unable to pay scheduled utility bills) in seven European Countries: Denmark, Belgium, France, Ireland, Italy, Netherlands and Spain. The period of analysis is 1994-2001. We also explore the dynamics of utilities deprivations focusing on the causes behind deprivation persistence. We differentiate between household heterogeneity and true state dependence. Then, controlling for observed and unobserved heterogeneity, we use the magnitude of average partial effects to investigate the relevance of any state dependence and the impact of energy sector reforms on the probability of experiencing utilities deprivations and on state dependence. We find evidence that vertical disintegration in the energy sector and privatization increase the household probability of experiencing utilities deprivation. Moreover, vertical disintegration also increases the household persistence in the status of deprivation.

Does persistence of social exclusion exist in Spain?

The aim of this paper is to analyse the causes leading to social exclusion dynamics. In particular, we wish to understand why an individual experiencing social exclusion today is much more likely to experience it again. In fact, there are two distinct processes that may generate a persistence of social exclusion: Heterogeneity (individuals are heterogeneous with respect to some observed and/or unobserved adverse characteristics that are relevant to experiencing social exclusion and persist over time) and true state dependence (experiencing social exclusion in a specific time period, in itself, increases the probability of undergoing social exclusion in subsequent periods). Distinguishing between the two processes is crucial since the policy implications are very different. The results suggest that both individual heterogeneity and true state dependence are related to the probability of experiencing social exclusion.

State Dependence at Internet Portals

This study offers evidence of the existence of switching costs on the Internet. It uses more flexible methods than previously possible to separate switching costs from serially correlated unobservables at Internet portals. The data contain nearly 1,000 observations per household, allowing for household‐specific regressions that control for all household‐specific heterogeneity. The results show that households exhibit switching costs. The loyalty generated by these costs drives a large fraction of portal visits and generates considerable revenues; however, these revenues are not large enough to justify the losses incurred by Internet portals in the 1990s while building market share. The results also suggest that random coefficients models overestimate true state dependence.

The Decision to Adopt Alternative Payment Methods: Testing for the Role of State Dependence

Many have predicted that through technological progress, the use of paper and coin currency would become almost obsolete. The obvious advantage of these payments technologies, i.e. interest-bearing bank accounts and ATM/debit cards, is that they allow households to economize on their money holdings while still providing liquidity services. However, studies by Humphrey, Pulley, and Vesala (1996), Evans and Schmalansee (1999), Mulligan and Sala-i-Martin (2000), and Stavins (2001) have shown that large segments of the population in Europe, North America, and Japan have yet to adopt financial innovation. One major reason used to explain the low adoption rates are the role of fixed costs involved with adopting financial innovation. This paper investigates the role of fixed costs in the household decision to adopt alternative payment technologies (i.e. interest-bearing bank accounts and ATM/debit cards). These issues are investigated by applying discrete-choice dynamic panel data techniques to the Bank of Italy Survey of Household Income and Wealth dataset in order to test for the presence of fixed costs or state dependence. One of the major obstacles is distinguishing between true state dependence and unobserved heterogeneity. The empirical evidence indicates that state dependence/fixed costs play a role in the adoption decision.

Persistence of Innovation: Stylised Facts and Panel Data Evidence

This paper investigates whether firms innovate persistently or discontinuously over time using an innovation panel data set on German manufacturing and service firms for the period 1994-2002. We find that innovation behaviour is permanent at the firm-level to a very large extent. Using a dynamic random effects discrete choice model and a new estimator recently proposed by Wooldrigde (2005), we further shed some light on the driving forces for this phenomenon. The econometric results confirm the hypothesis of true state dependence for manufacturing as well as for service sector firms. In addition to past innovation experience, the results further highlight the important role of knowledge provided by skilled employees and unobserved individual heterogeneity in explaining the persistence of innovation.

State dependence and heterogeneity in fishing location choice

To explore the distinction between state dependence and heterogeneity in repeated decisions, this paper combines a Mixed Logit model with a state dependence parameterization from the marketing literature to study fishing location choices of commercial sea urchin divers in California. It examines implications of ignoring either effect and finds in all cases that true state dependence is an important determinant of location choice. Consequently, spatial policies like marine reserves can lead to differences in the short- and long-run behavioral responses of the fishing fleet. Under some specifications, random preference parameters are statistically significant when state dependence is excluded from the model, but when it is included, random preference parameters are not significant. In other specifications, including state dependence only dampens the variability in preference parameters. These results highlight the importance of gathering and analyzing diary-type data for commercial fisheries as well as for similar choice problems in recreation demand.

State Dependence in a Multi-State Model of Employment Dynamics

A multinomial choice framework is used to investigate the nature of women's transitions between full-time employment, part-time employment and non-employment. The stochastic framework allows time varying and time invariant unobserved preferences, and also controls for the possible endogeneity of education, fertility and non-labor income. Significant positive true state dependence is found in both full-time and part-time employment. This finding is robust to the specification of unobserved preferences. The results are used to assess the dynamic effects of three temporary wage subsidies. All three policies have substantial effects on employment behavior for up to 6 years. However, obtaining a permanent increase in employment requires sustained or repeated interventions.

True Versus Spurious State Dependence in Firm Performance: The Case of West German Exports

This paper analyzes the persistence of firms' exporting behavior in a panel of West German manufacturing firms. Dynamic binary choice models allow us to distinguish between true and spurious state dependence in firm performance. Using random effects models as well as a recent fixed effect approach which imposes few restrictions on unobservables, we find robust evidence of state dependence in the current export status of firms. Unobserved permanent firm heterogeneity (spurious state dependence) is found to be less important than suggested by earlier studies. The existence of true state dependence in exports has direct economic policy implications: if policy successfully turns non-exporters into exporters, the effect is likely to be lasting.

Recurrent unemployment, welfare benefits and heterogeneity

The objective of this paper is to investigate the causes of the recurrences of individuals in unemployment during benefit periods. So as to attain this objective, we use administrative data from the Spanish Employment Agency to estimate a duration model with multiple spells that allows for unemployment state dependence through lagged unemployment duration in order to distinguish the heterogeneity and scarring effects. We find that an increase in the duration of previous unemployment benefit periods lengthens the expected duration of future unemployment benefit periods. True state dependence and heterogeneity, intensity of job search and local labour market conditions are among the elements that explain this unemployment state dependence.

THE NET BENEFIT TO GOVERNMENT OF HIGHER EDUCATION: A “BALANCE SHEET” APPROACH

Previous studies find a positive relationship between juvenile and adult criminal involvement. Using data on males from the Delinquency in a Birth Cohort II study, we investigate whether this correlation is due to unobserved characteristics that increase the probability of both juvenile and adult crime, or whether it is due to true state dependence in crime. Distinguishing between state dependence and heterogeneity is important from a policy perspective. For example, if youthful crime causes adult crime, then policies that reduce a juvenile's criminal behavior will also reduce criminal behavior as an adult. Using a treatment effects model, we find evidence of both state dependence and heterogeneity in the relationship between juvenile and adult crime. The causal influence of delinquency on adult crime is largest for white males and males with fewer years of schooling. The findings suggest that preventive policies that divert juveniles from crime are a viable policy tool for reducing the overall rate of crime.(This abstract was borrowed from another version of this item.)

Wage differentials and mobility in the urban labour market: a panel data analysis for Mexico

We analyze wage differentials and transitions between the formal and informal sectors in urban Mexico. We use panel data on five quarterly waves from Mexico's Urban Employment Survey. We develop a dynamic random effects panel data model consisting of separate wage equations for the two sectors and a logit part explaining the labour market state with wages included as explanatory variables. The model is estimated using simulated maximum likelihood. The estimates show that wage differentials increase with education level. The probability of formal sector employment strongly increases with the wage differential. For male workers, the choice between formal and informal sector is driven by wage differentials and unobserved heterogeneity, and true state dependence is not important. For women, non-participation is the most common labour market state, and true state dependence plays a much larger role.

Wage Differentials and Mobility in the Urban Labor Market: A Panel Data Analysis for Mexico

We analyze wage differentials mobility between the formal and informal sector in urban Mexico, using panel data on five quarters drawn from Mexico's Urban Employment Survey. We develop a dynamic random effects panel data model. It consists of two separate wage equations for the two sectors and a multinomial logit part explaining the labor market state, in which wages are included as explanatory variables. The model is estimated using simulated maximum likelihood. The estimates show that wage differentials increase with education level. The probability of formal sector employment strongly increases with the wage differential. Simulated transition probabilities show that for male workers, the choice between formal and informal sector is driven by wage differentials and unobserved heterogeneity, while true state dependence is much less important. For women, nonparticipation is the most common labor market state, and true state dependence plays a much larger role.

True state dependence and heterogeneity in off-farm labour participation

This paper discusses models of off-farm labour participation estimated with panel data. With this kind of model, unexplained persistence in a labour state may be due to true state dependence (when a previous state modifies attitudes, constraints or parameters) and/or to heterogeneity (unobservable permanent individual characteristics). Probit models assuming either heterogeneity or true state dependence are estimated from a two-period panel survey of Pennsylvania farms, and compared with cross-section or pooled sample models. Statistical tests to infer the source of persistence are proposed. The results suggest that for Pennsylvania farmers and spouses, in addition to other personal, household and farm characteristics found to be important in previous studies, previous off-farm labour state is relevant in off-farm labour participation decisions.

Low‐wage mobility in the Italian labour market

Uses Bank of Italy’s Survey on Household Income and Wealth (SHIW) panel data for 1993 and 1995 to model transition probabilities at the bottom of the Italian wage distribution and to investigate the features and determinants of low‐wage mobility. The analysis is based on a bivariate probit model with endogenous switching which allows tackling the initial conditions problem, i.e. the potential endogeneity of the conditioning starting state. Results show the appropriateness of such a choice, the hypothesis of exogenous initial conditions being always rejected. Shows that while some factors such as education, sex and geographical location have an effect on low‐pay persistence, job‐related variables are more effective in avoiding falls into low pay from higher pay. It is also shown how raw persistence involves a considerable share of true state dependence, i.e. the experience of low pay raises, per se, the probability of future low‐pay episodes, irrespective of personal attributes.

Modeling Heterogeneity and State Dependence in Consumer Choice Behavior

In frequently-purchased-consumer-goods markets, consumer brand choices exhibit substantial persistence across purchase occasions. In this article, I estimate a choice model that admits of both heterogeneity in preferences and true state dependence as sources of this persistence, using Nielsen scanner data on ketchup. I find evidence for true state dependence in the choice process, even after controlling for a rich heterogeneity structure. Simulation of the model indicates that the long-term effect of a promotion-induced purchase on future purchase probabilities is positive but small.

A Framework for Investigating Habits, “The Hand of the Past,” and Heterogeneity in Dynamic Brand Choice

In this paper we develop a general class of dynamic brand choice models, called Lightning Bolt (LB) models, which are consistent with the theory of random utility maximization of consumer choice behavior. The underlying random utility process is Markov, and the inter-temporal evolution of the (utility-maximizing) brand choice process is also Markov. The models permit parsimonious parameterizations of the random utility process in brand choices with the resulting switching probabilities being functionally related to explanatory variables. The model allows for structural state dependence (feedback), habit persistence (inertia), and unobserved heterogeneity. The theoretical development shows that several well known stochastic brand choice models can be deduced from random utility maximization theory. From a managerial perspective, the usefulness of the proposed model stems from its ability to separate out the effects of habits, state dependence and heterogeneity. Strong state dependence effects imply incentives for inducing trial of a brand (e.g., product sampling). In contrast with state dependence, a strong habit persistence effect may be indicative of buyer behavior where inducement of trial of a different brand may not be sufficient to maintain a (sustained) defection of the consumer from the habitually purchased brand to the trial brand. Failure to distinguish between these two effects has important implications. For example, a model that only accounts for state dependence effects would, in the presence of only habit persistence, incorrectly attribute it to state dependence. Based on this the manager could decide to embark on an expensive sampling program that might prove ineffective due to the absence of state dependence. It is also important to distinguish between the effects of unobserved heterogeneity and state dependence. In the absence of true state dependence, failing to account for unobservable variations across households (such as differences in price sensitivities), results in the temporally persistent unobservable elements showing up as state dependence in the model. Hence, a manager may incorrectly opt for sampling as the appropriate marketing action, whereas, a couponing or price promotion strategy should have been preferred. We estimate the model parameters using the AC Nielsen household scanner panel data set on catsup purchases. Further, we investigate empirical techniques for overcoming the “initial conditions” problem that affects many dynamic models of brand choice. Through simulation analysis using the estimated parameters, we show that the calculated profitability of a promotion must take into account the multi-period impact due to state dependence. Further, we demonstrate how ignoring heterogeneity can result in spurious state dependence, thereby making marketing tactics such as product sampling appear far more attractive than they actually are. Specifically, the model that accounts only for habit persistence and state dependence effects predicts a share increase of 2.4% points through sampling for one of the brands in the empirical analysis. Once the effects of unobserved heterogeneity are accounted for, however, this number drops precipitously to 0.3%. While even this low number might fulfill the objectives of the brand manager, it ensures that expectations are not eight times that number. It is important to note that this paper is a first attempt at analyzing the three fundamental dimensions of dynamic brand choice behavior. Our formulation represents a “reduced form” approach as opposed to a “structural” approach. Other limitations of the model include our focus on time-invariant choice sets. Extensions of the model to situations where choice sets vary over time are possible, but are difficult. Another limitation is the dependence on multivariate extreme value distributions. An alternative would be to use multivariate normal distributions, i.e., a probit-like structure which has certain useful properties. A comparison of the two approaches would be useful. Empirical extensions of the model to allow for higher-order processes and nested logit type model structures could also be fruitful. Explicitly incorporating variety-seeking behavior in the model would further enrich the theoretical framework proposed in this paper.