Unobserved Fixed Effects Research Articles

A Novel Approach for Testing Fractional Cointegration in Panel Data Models with Fixed Effects

Fractional cointegration in time series data has been explored by several authors, but panel data applications have been largely neglected. A previous study of ours discovered that the Chen and Hurvich fractional cointegration test for time series was fairly robust to a moderate degree of heterogeneity across sections of the six tests considered. Therefore, this paper advances a customized version of the Chen and Hurvich methodology to detect cointegrating connections in panels with unobserved fixed effects. Specifically, we develop a test statistic that accommodates variation in the long-term cointegrating vectors and fractional cointegration parameters across observational units. The behavior of our proposed test is examined through extensive Monte Carlo experiments under various data-generating processes and circumstances. The findings reveal that our modified test performs quite well comparatively and can successfully identify fractional cointegrating relationships in panels, even in the presence of idiosyncratic disturbances unique to each cross-sectional unit. Furthermore, the proposed modified test procedure established the presence of long-run equilibrium between the exchange rate and labor wage of 36 countries’ agricultural markets.

Linear panel regressions with two-way unobserved heterogeneity

We study linear panel regression models in which the unobserved error term is an unknown smooth function of two-way unobserved fixed effects. In standard additive or interactive fixed effect models the individual specific and time specific effects are assumed to enter with a known functional form (additive or multiplicative). In this paper, we allow for this functional form to be more general and unknown. We discuss two different estimation approaches that allow consistent estimation of the regression parameters in this setting as the number of individuals and the number of time periods grow to infinity. The first approach uses the interactive fixed effect estimator in Bai (2009), which is still applicable here, as long as the number of factors in the estimation grows asymptotically. The second approach first discretizes the two-way unobserved heterogeneity (similar to what Bonhomme et al., 2021 are doing for one-way heterogeneity) and then estimates a simple linear fixed effect model with additive two-way grouped fixed effects. For both estimation methods we obtain asymptotic convergence results, perform Monte Carlo simulations, and employ the estimators in an empirical application to UK house price data.

Examining the heterogeneity by employment status in dynamics between BMI and cognition: a longitudinal cohort study of the Korean aging population

PurposeThis study investigated the potential heterogeneity by employment status in the relationship between body mass index (BMI) trajectory and cognitive function among an aging population. MethodsWe analyzed 2010–2018 cohort data from the Korean Longitudinal Study of Aging involving 4889 individuals aged over 45. We used statistical techniques of quadratic terms, interaction terms, time-invariant unobserved fixed effects, and time-lag effects to estimate the dynamic and interactive relationships among study variables. ResultsThe effect of BMI on cognitive function was heterogeneous based on employment status. For the nonemployed group, the impact of BMI on cognitive function showed an inverted U-shape with a turning point (BMI 25); a higher BMI (up to 25) was associated with higher cognitive function, but further increases beyond this threshold led to decreased cognitive function. For the employed group, however, the impact of BMI on cognitive function was nonsignificant. The nonlinear effect of BMI on cognitive function for the nonemployed group was robust across various subgroups and specifications. ConclusionsThe findings highlight the risk of obesity (BMI ≥ 25) on cognitive decline, particularly among nonemployed individuals. This illuminates the critical role of labor activity in regulating the impact of BMI on cognitive function among an aging population.

Distant lending for regional small businesses using public credit guarantee schemes: Evidence from Japan

In this paper, we investigate the extent to which banks use public credit guaranteed loans for distant small business borrowers. Existing studies argue that when banks offer loans for these borrowers, the information asymmetry between them is often severe. These studies then empirically demonstrate how banks can mitigate this problem. We focus instead on the role of Japan’s public credit guarantee scheme in mitigating these same information problems, an approach unaddressed in the literature. If banks offer credit guaranteed loans, they suffer few losses from borrower default because the public credit guarantee corporations (not the small business borrowers) make payments to the banks. Therefore, banks can offer loans to informationally opaque and distant borrowers. To conduct the econometric analysis, we use semiannual bank-region-level data from Japan, which allows us to control for several unobserved fixed effects. The results reveal that credit guarantee loan size is larger and the default rate higher when banks offer credit guaranteed loans to distant borrowers. These findings suggest that banks successfully mitigate the losses of distant lending using the public credit guarantee scheme, thereby providing a valuable contribution to the small business finance literature.

Pension return assumptions and shareholder-employee risk-shifting

Firm managers of defined-benefit (DB) pension plan sponsors reveal their primary motives — risk-shifting or risk-management — through their assumed expected rates of return (ERRs) on the plan assets. Managers with risk-shifting motives choose high ERRs to exploit flexible internal financing from employees via pension underfunding. Those with risk-management motives choose low ERRs to reduce future cash-flow uncertainty by improving the pension funding status. We examine if ERRs predict the firms’ future cash-flow allocation between pension funding and corporate investments, in a Japanese sample that mitigates the selection bias concern for US DB plan sponsors. Using dynamic panel regressions that control for lagged dependent variables, firms’ business prospects, and unobserved fixed effects, we show that higher ERRs precede higher capital investments, R&D expenses, and net pension obligations while revealing managerial aggression, especially among firms with high external financing costs. Higher ERRs predict higher market-to-book ratios for the firms with larger R&Ds and/or underfunding, suggesting that the risk-shifting channel of internal financing with high ERRs can help alleviate underinvestment problems.

Working Still Harder

The authors use data from the British Skills and Employment Surveys to document and to try to account for sustained work intensification between 2001 and 2017. They estimate the determinants of work intensity, first using four waves of the pooled cross-section data, then using a constructed pseudo-panel of occupation–industry cells. The latter approach suggests biases in cross-section models of work intensity, associated with unobserved fixed effects in specific occupations and industries. The pseudo-panel analysis can account for slightly more than half (51%) of work intensification using variables that measure effort-biased technological change, effort-biased organizational change, the growing requirement for learning new things, and the rise of self-employment. The authors interpret the work intensification and these effects within a power-resources framework.

Determinants of regional distribution of AKP votes: Analysis of post‐2002 parliamentary elections

AbstractElectoral politics and regional variation in socio‐economic conditions of individuals receive huge interest in countries with sizable political tension. Even though Turkey has undergone a combination of periods of economic success and a period of economic and political turmoil, it has been governed by a single political party during the last two decades—Adalet ve Kalkinma Partisi (AKP)— Justice and Development Party. This paper examines the determinants of the variation in regional vote shares of AKP between the 2002 and 2018 parliamentary elections using panel data estimation method—effects—which allows us to account for time invariant region‐specific unobserved fixed effects. Our findings demonstrate that inflation, unemployment, per capita GDP growth, provision of healthcare, industrial sector growth rates, change in the ageing of the population and rate of absenteeism (voter turnout) are essential factors in voters' decisions. Our additional results reveal that the effects of socio‐economic factors on the AKP vote shares were distinctly different for the AKP stronghold provinces and the eastern regions that are mostly populated with Kurdish, suggesting that ideological position of the voters also play an essential role in voters' response to the changes in socio‐economic conditions.

Retrospective causal inference via matrix completion, with an evaluation of the effect of European integration on cross-border employment

We propose a method of retrospective counterfactual imputation in panel data settings with later-treated and always-treated units, but no never-treated units. We use the observed outcomes to impute the counterfactual outcomes of the later-treated using a matrix completion estimator. We propose a novel propensity-score and elapsed-time weighting of the estimator's objective function to correct for differences in the observed covariate and unobserved fixed effects distributions, and elapsed time since treatment between groups. Our methodology is motivated by studying the effect of two milestones of European integration -- the Free Movement of persons and the Schengen Agreement -- on the share of cross-border workers in sending border regions. We apply the proposed method to the European Labour Force Survey (ELFS) data and provide evidence that opening the border almost doubled the probability of working beyond the border in Eastern European regions.

Retrospective Causal Inference Via Matrix Completion, with an Evaluation of the Effect of European Integration on Cross-Border Employment

We propose a method of retrospective counterfactual imputation in panel data settings with later-treated and always-treated units, but no never-treated units. We use the observed outcomes to impute the counterfactual outcomes of the later-treated using a matrix completion estimator. We propose a novel propensity-score and elapsed-time weighting of the estimator's objective function to correct for differences in the observed covariate and unobserved fixed effects distributions, and elapsed time since treatment between groups. Our methodology is motivated by studying the effect of two milestones of European integration -- the Free Movement of persons and the Schengen Agreement -- on the share of cross-border workers in sending border regions. We apply the proposed method to the European Labour Force Survey (ELFS) data and provide evidence that opening the border almost doubled the probability of working beyond the border in Eastern European regions.

Estimation of average treatment effects using panel data when treatment effect heterogeneity depends on unobserved fixed effects

SummaryThis paper proposes a new panel data approach to identify and estimate the time‐varying average treatment effect (ATE). The approach allows for treatment effect heterogeneity that depends on unobserved fixed effects. In the presence of this type of heterogeneity, existing panel data approaches identify the ATE for limited subpopulations only. In contrast, the proposed approach identifies and estimates the ATE for the entire population. The approach relies on the linear fixed effects specification of potential outcome equations and uses exogenous variables that are correlated with the fixed effects. I apply the approach to study the impact of a mother's smoking during pregnancy on her child's birth weight.

Revealing preferences for teachers in rural Myanmar

ABSTRACT In November 2010, Myanmar1 began a process of significant democratic reform ending a half century of authoritarian rule by military junta. Rural education, having long suffered from a lack of qualified teachers in poorly equipped, overcrowded schools, has received little attention from empirical scholarship aimed at informing policymakers. This article pilots a direct survey of university students entering into the teaching profession, producing clear priorities to support teachers, especially in rural areas. Survey data were gathered at the Yangon University of Education using an improved conjoint analysis. Survey attributes were completely and independently randomised then analysed by ordinary least squares regression and the Sarafidis and Weber partitional-clustered regression with unobserved fixed effects. Out of many competing priorities, the analysis reveals that secure housing is at the highest priority for respondents, with secondary and tertiary preferences varying based on group clusters. This study aims to inform teacher placement policy by demonstrating methodologically how observed and unobserved characteristics of the respondents and the teaching environment influence teacher retention. Given the present challenges facing teachers in rural areas, better meeting the preferences of teachers can have long-term educational benefits for children in these areas.

Trading firms and the gender wage gap: evidence from South Africa

abstractFirms engaged in international trade account for a large share of South Africa’s output and employment. Therefore, their behaviour has potentially large effects on national outcomes including gender inequality. In this paper, we test whether the gender wage gap (GWG) of trading firms differs from that of domestic firms based on a growing literature that has outlined how the special characteristics of trading firms could lead to such a different GWG. Using a unique employer–employee matched data panel from 2011 to 2016 for South Africa based on novel tax record data and employing various fixed effects regressions, we find that the GWG of trading firms is significantly larger than that of domestic firms. This holds even when controlling for unobserved individual and firm fixed effects that account for factors such as worker education or firm profitability. We also find that firms which both import and export (a crude proxy for foreign-owned firms) behave more equally than other trading firms. This could be driven by foreign owned firms that impose their more equal domestic pay structures on their South African affiliates and would emphasize the role of foreign investment for gender equality.

Estimation of Average Treatment Effects Using Panel Data when Treatment Effect Heterogeneity Depends on Unobserved Fixed Effects

This paper proposes a new panel data approach to identify and estimate the time-varying average treatment effect (ATE). The approach allows for treatment effect heterogeneity that depends on unobserved fixed effects. In the presence of this type of heterogeneity, existing panel data approaches identify the ATE for limited subpopulations only. In contrast, the proposed approach identifies and estimates the ATE for the entire population. The approach relies on the linear fixed effects specification of potential outcome equations and uses exogenous variables that are correlated with the fixed effects. I apply the approach to study the impact of a mother's smoking during pregnancy on her child's birth weight.

Path-Dependent Dynamics and Technological Spillovers in the Brazilian Regions

This article investigates the influence of path dependence, of spatial spillovers and of production specialization on regional technological specialization. We use patent data and characteristics of industrial activity by Brazilian regions in the period of 2000–2011 to estimate a spatial dynamic panel using the Generalised Method of Moment (GMM) estimator, which deals with unobserved fixed effects and with the endogeneity problem. The results show that the regional production specialization influences technological specialization in Brazilian regions. Furthermore, this article finds that regional technological development is highly path-dependent and characterized by spatial spillovers. The former result means that regional technological development is influenced by its own technological specialization trajectory. The latter shows that the technological specialization of the neighborhood has proved to be a determining factor in local technological specialization. These results may help in the understanding of the development of technological clusters, suggesting that the strategies to reinforce the regional innovation processes should consider the specificities of the regional production pattern.

The Determinants of the Value Created by Stock Repurchase Programs

This paper aims to extend the empirical literature on the determinants of the value created by stock repurchase programs by analyzing the characteristics of the repurchase decision-making groups. Based on a sample of 200 US-listed firms from 1998-2004, the paper examines empirically the relationship between value created by stock repurchase programs and board of directors’ characteristics. The analysis depends on panel data analysis to consider unobservable fixed-effects in this relationship. The results provide evidence of a significant relation between board of directors’ characteristics and value created by the stock repurchase programs. The two important determinants are directors’ independence and directors’ outside experience. However, we find that this relationship changed significantly for some measures after the adoption of the 2002 Sarbanes-Oxley Act.

Attendance of cultural events and involvement with the arts—impact evaluation on health and well-being from a Swiss household panel survey

ObjectivesAlthough there is strong uptake of active or passive engagement with the cultural and creative activities as determinants of individual health, well-being and social participation, few population studies report any causal influence on self-reported and physical health or life satisfaction from voluntary engagement with the arts (playing an instrument or singing, painting, sculpture) or passive cultural participation (attending the cinema, theatre, opera and exhibitions). This study set out to investigate any potential derived benefits to the Swiss population. Study designThe 2010 and 2013 waves of the Swiss Household Panel study were used for analysis. The data are representative for the Swiss population aged 14 years and older with respect to major demographic variables. MethodsUsing longitudinal data, the strengths of the two approaches to evaluating causal inference were simultaneously applied: propensity score matching and difference-in-differences. Propensity score matching attempted to eliminate selection bias by conditioning on confounding variables. Difference-in-differences estimator was applied to remove unobserved fixed effects via intra-individual comparisons over time by comparing the trends in a matched treatment and control group. ResultsThe study showed that voluntary cultural activity—of any type, passive or active—did not seem to have any causative influence on health and well-being. Results showed that long-term health and well-being did not improve significantly as a result of any specific activity in the cultural arena. ConclusionsThe investigation provided little evidence to justify health promotion messages for involvement with the arts. Nevertheless, these findings do not contest that active or passive participation in cultural- and arts-related activities may be beneficial to health and well-being when guided by qualified therapists to treat specific health-related problems.

Estimation of time-varying average treatment effects using panel data when unobserved fixed effects affect potential outcomes differently

This paper proposes a new approach to estimate the time-varying average treatment effect using panel data to control for unobserved fixed effects. The approach allows identifying the average treatment effect on the entire population, even if the fixed effects affect potential outcomes under treatment and no treatment differently, which can cause heterogeneity in treatment effects among unobserved characteristics. Note that a popularly used standard difference-in-differences approach can only identify the average treatment effect on the treated. Moreover, the proposed approach allows time-varying treatment effects. The approach exploits panel data with a specific structure in which the treatment exposure expands to the entire population over time. I apply the proposed approach to estimate the effect of the introduction of electronic voting technology for the reduction of residual votes in Brazilian elections.

Term Structure of Hedging Premia and Basis Arbitrages in Synthetic-Cash Credit Market

This paper investigates the price formation of credit risk premia across European sovereign countries. A metric of such premia is retrieved under the statistical measure using bootstrap techniques on hedging portfolios. This latter is retrieved in the cash-synthetic market by means of comparison to standard credit spreads vis-a-vis Germany. Price discovery is then analyzed in a VECM setting by means of lead-lag analysis. Discrepancies between the two credit metrics induces the CDS-bond basis, here defined as the spread of a benchmark rate on target hedged position. The basis is explained across a panel of four core and four non-core countries using idyosincratic and common regressors. The retrieved unobserved fixed effects are then used as a metric for liquidity premia.

Breaking the Curse of Sisyphus: An Empirical Analysis of Post-Conflict Economic Transitions

This paper provides an empirical analysis of the principal determinants of post-conflict economic transitions during the period 1960–2010 using a dynamic panel estimation approach. In addition to demographic, economic, geographic, and institutional variables, we introduce a novel measure of conflict recurrence risk, estimated with a logistic regression approach controlling for unobserved fixed effects in a non-linear probability model. The empirical results show that the risk of conflict recurrence is a significant determinant of post-conflict economic performance, even after controlling for a broad set of demographic, economic, geographic, and institutional factors.

Ownership Structure and Performance Among Entrepreneurial Firms: The Growth-Control Trade-Off

Roughly one-third of our sample firms show substantial ownership structure changes over time. Growth potential, risk sharing, and demand for external capital seem to be the main reasons behind dilutions of ownership structures, and these firms show strong growth in the following five years. After controlling for the unobserved firm fixed effects or the sample selection bias, we do not find any evidence suggesting a link between ownership structure and other dimensions of firm performance. Therefore, the trade-off between growth aspiration and retention of dominant control of the firms may be critical in ownership structure decisions among entrepreneurial firms.