To send content items to your account,
please confirm that you agree to abide by our usage policies.
If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your account.
Find out more about sending content to .
To send content items to your Kindle, first ensure email@example.com
is added to your Approved Personal Document E-mail List under your Personal Document Settings
on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part
of your Kindle email address below.
Find out more about sending to your Kindle.
Note you can select to send to either the @free.kindle.com or @kindle.com variations.
‘@free.kindle.com’ emails are free but can only be sent to your device when it is connected to wi-fi.
‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.
Difference-in-differences is a widely used evaluation strategy that draws causal inference from observational panel data. Its causal identification relies on the assumption of parallel trends, which is scale-dependent and may be questionable in some applications. A common alternative is a regression model that adjusts for the lagged dependent variable, which rests on the assumption of ignorability conditional on past outcomes. In the context of linear models, Angrist and Pischke (2009) show that the difference-in-differences and lagged-dependent-variable regression estimates have a bracketing relationship. Namely, for a true positive effect, if ignorability is correct, then mistakenly assuming parallel trends will overestimate the effect; in contrast, if the parallel trends assumption is correct, then mistakenly assuming ignorability will underestimate the effect. We show that the same bracketing relationship holds in general nonparametric (model-free) settings. We also extend the result to semiparametric estimation based on inverse probability weighting. We provide three examples to illustrate the theoretical results with replication files in Ding and Li (2019).
Can crime victimization increase support for iron-fist crime-reduction policies? It is difficult to assess the political effects of crime, mainly because of the presence of unmeasured confounders. This study uses panel data from Brazil and strategies for reducing sensitivity to hidden biases to study how crime victims update their policy preferences. It also examines survey data from eighteen Latin American countries to improve the external validity of the findings. The results show that crime victims are more likely to support iron-fist or strong-arm measures to reduce crime, such as allowing state repression. Affected citizens are also found to value democracy less, which might explain their willingness to accept the erosion of basic rights in favor of radical measures to combat delinquency. These findings reveal that exposure to crime can change what people think the state should be allowed to do, which can have important political implications.
Does exposure to the refugee crisis fuel support for extreme-right parties? Despite heated debates about the political repercussions of the refugee crisis in Europe, there exists very little—and sometimes conflicting—evidence with which to assess the impact of a large influx of refugees on natives’ political attitudes and behavior. We provide causal evidence from a natural experiment in Greece, where some Aegean islands close to the Turkish border experienced sudden and drastic increases in the number of Syrian refugees while other islands slightly farther away—but with otherwise similar institutional and socioeconomic characteristics—did not. Placebo tests suggest that precrisis trends in vote shares for exposed and nonexposed islands were virtually identical. This allows us to obtain unbiased estimates of the electoral consequences of the refugee crisis. Our study shows that among islands that faced a massive but transient inflow of refugees passing through just before the September 2015 election, vote shares for Golden Dawn, the most extreme-right party in Europe, moderately increased by 2 percentage points (a 44 percent increase at the average). The finding that mere exposure to the refugee crisis is sufficient to fuel support for extreme-right parties has important implications for our theoretical understanding of the drivers of antirefugee backlash.
The fixed-effects estimator is biased in the presence of dynamic misspecification and omitted within variation correlated with one of the regressors. We argue and demonstrate that fixed-effects estimates can amplify the bias from dynamic misspecification and that with omitted time-invariant variables and dynamic misspecifications, the fixed-effects estimator can be more biased than the ‘naïve’ OLS model. We also demonstrate that the Hausman test does not reliably identify the least biased estimator when time-invariant and time-varying omitted variables or dynamic misspecifications exist. Accordingly, empirical researchers are ill-advised to rely on the Hausman test for model selection or use the fixed-effects model as default unless they can convincingly justify the assumption of correctly specified dynamics. Our findings caution applied researchers to not overlook the potential drawbacks of relying on the fixed-effects estimator as a default. The results presented here also call upon methodologists to study the properties of estimators in the presence of multiple model misspecifications. Our results suggest that scholars ought to devote much more attention to modeling dynamics appropriately instead of relying on a default solution before they control for potentially omitted variables with constant effects using a fixed-effects specification.
A better understanding of the relative importance of factors related to climate change and to changes associated with economic growth would serve to inform water policy and to focus scarce public resources on anticipated problems arising from distinct sources of changes in water demand. This article investigates the determinants of residential water consumption in Chile, a developing country that has seen noteworthy changes in incomes, household size, poverty rates and levels of urbanization, and which is projected to experience significant climatic but varied changes, depending on the region of the country. Panel data for 1998-2010 at the municipal level is used to analyze the sensitivity of residential water demand to climate and development-related factors. In the case of Chile, the effect on water consumption of these development-related changes is estimated to be several times that of the changes associated with climate projections for 50 to 80 years in the future.
Party identification has been thought to provide the central organizing element for political belief systems. This article makes the contrasting case that core values concerning equality and government intervention versus individualism and free enterprise are fundamental orientations that can themselves shape partisanship. The authors evaluate these arguments in the British case using a validated multiple-item measure of core values, using ordered latent class models to estimate reciprocal effects with partisanship on panel data from the British Household Panel Study, 1991–2007. The findings demonstrate that core values are more stable than partisanship and have far stronger cross-lagged effects on partisanship than vice versa in both polarized and depolarized political contexts, for younger and older respondents, and for those with differing levels of educational attainment and income, thus demonstrating their general utility as decision-making heuristics.
We propose a new parametric model for the modelling and estimation of event distributions for individuals in different firms. The analysis uses panel data and takes into account individual and firm effects in a non-linear model. Non-observable factors are treated as random effects. In our application, the distribution of accidents is affected by observable and non-observable factors from vehicles, drivers and fleets of vehicles. Observable and unobservable factors are significant to explain road accidents, which mean that insurance pricing should take into account all these factors. A fixed effects model is also estimated to test the consistency of the random effects model.
The EU infrastructure policy has relied on Public-Private Partnerships (PPPs) as a means to successfully deliver infrastructure of benefit for the EU. To reach its infrastructure policy objectives, the EU has implemented support mechanisms aimed at facilitating the delivery of PPPs. This article is aimed at evaluating to what extent these mechanisms have actually contributed to improving the economic performance of PPPs. To that end, we have selected the case of Spanish road PPPs for empirical analysis. The main result shows that EU support positively influences the economic performance of PPP projects. This is caused by the fact that the EU conditions its financial support on a project’s meeting a set of requirements that help assure the success of the project. From this result, we obtain a set of conclusions that may be generalised to other cases, and provide a contribution to the body of knowledge on PPPs.
Ce papier analyse la dynamique de la spécialisation des activités technologiques mesurées par les brevets de 221 régions européennes sur la période 1989-2000. L'approche adoptée repose sur la spécification d'un modèle à erreurs composées et à coefficients variables. La modélisation prend en compte l'impact des spécificités régionales sur la vitesse de convergence technologique et étudie comment la croissance économique affecte la spécialisation des régions. Les résultats d'estimation montrent (i) une convergence (dé-spécialisation) des activités technologiques régionales. Les vitesses de déspécialisation varient selon les régions de 0% à 8% par an. (ii) Les capacités inventives initiales et les effets nationaux apparaissent comme des facteurs de décélération de la vitesse de convergence technologique, (iii) l'effet de la croissance économique sur la spécialisation est positif pour les régions à forte propension à breveter mais négatif pour les régions à faibles capacités inventives, toutes choses égales par ailleurs.
This paper proposes a Bayesian approach to estimating a factor-augmented GDP per capita equation. We exploit the panel dimension of our data and distinguish between individual-specific and time-specific factors. On the basis of 21 technology, infrastructure, and institutional indicators from 82 countries over a 19-year period (1990 to 2008), we construct summary indicators of each of these three components in the cross-sectional dimension and an overall indicator of all 21 indicators in the time-series dimension and estimate their effects on growth and international differences in GDP per capita. For most countries, more than 50% of GDP per capita is explained by the four common factors we have introduced. Infrastructure is the greatest contributor to total factor productivity, followed by technology and institutions.
This study examines in an empirical comparison how different econometric specifications of stochastic frontier models affect the decomposition of total factor productivity growth. We estimate nine stochastic frontier models, which have been widely used in empirical investigations of sources of productivity growth. Our results show that the relative contribution of components to total factor productivity growth is quite sensitive to the choice of econometric model, which points to the need to select the “right” model. We apply various statistical tests to narrow the range of applicable models and identify additional criteria upon which to base the choice of non-nested models.
Ratemaking is one of the most important tasks of non-life actuaries. Usually, the ratemaking process is done in two steps. In the first step, a priori ratemaking, an a priori premium is computed based on the characteristics of the insureds. In the second step, called the a posteriori ratemaking, the past claims experience of each insured is considered to the a priori premium and set the final net premium. In practice, for automobile insurance, this correction is usually done with bonus-malus systems, or variations on them, which offer many advantages. In recent years, insurers have accumulated longitudinal information on their policyholders, and actuaries can now use many years of informations for a single insured. For this kind of data, called panel or longitudinal data, we propose an alternative to the two-step ratemaking approach and argue this old approach should no longer be used. As opposed to a posteriori models of cross-section data, the models proposed in this paper generate premiums based on empirical results rather than inductive probability. We propose a new way to deal with bonus-malus systems when panel data are available. Using car insurance data, a numerical illustration using at-fault and non-at-fault claims of a Canadian insurance company is included to support this discussion. Even if we apply the model for car insurance, as long as another line of business uses past claim experience to set the premiums, we maintain that a similar approach to the model proposed should be used.
Long-run income convergence is investigated in the U.S. context. We employ a novel pairwise econometric procedure based on a probabilistic definition of convergence. The time-series properties of all the possible regional income pairs are examined by means of unit root and non-cointegration tests, where inference is based on the fraction of rejections. We distinguish between the cases of strong convergence, where the implied cointegrating vector is [1, −1], and weak convergence, where long-run homogeneity is relaxed. To address cross-sectional dependence, we employ a bootstrap methodology to derive the empirical distribution of the fraction of rejections. We find supporting evidence of U.S. states sharing a common stochastic trend consistent with a definition of convergence based on long-run forecasts of state incomes being proportional rather than equal. We find that the strength of convergence between states decreases with distance and initial income disparity. Using Metropolitan Statistical Area data, evidence for convergence is stronger.
Dans cet article, nous analysons la structure des coûts d'alimentation en eau potable des villes de Côte d'Ivoire à partir d'un panel des centres de production. Considérant les services d'eau comme des monopoles multi-produits fournissant conjointement deux biens (volumes d'eau facturés et volumes d'eau perdus), nous estimons une fonction de coût translog multi-produits sur la base de la dualité entre fonctions de production et de coût. Les différentes mesures de rendements calculées révèlent que le service d'eau moyen ivoirien produit dans la zone des rendements constants. Cependant, en classant les services en petits, moyens et grands selon différents critères, il apparaît clairement que l'opérateur a un avantage économique à accroître sa production et les connections dans la plupart des petits et moyens services. Aussi, la présence d'économies de gamme révèle que la production conjointe des deux biens considérés dans certaines proportions est plus bénéfique que l'amélioration du rendement du réseau. Enfin, les évaluations des coûts estimés font apparaître un coût marginal supérieur en moyenne aux prix marginaux des premières tranches de la grille tarifaire.
The aim of this paper is to measure the influence of the railroad in the urbanization of Spain between 1860 and 1910. Our sources are from quantitative information –censuses of population– and qualitative one –coastal condition, existence of mining industry or industry, administrative capital and date of the railway connection–. We have estimated a first model of data panel in differences. Based on this model we have employed different estimation techniques in order to address omitted variables and/or endogeneity of the train variable. Results from all estimations give us clear evidence of the positive influence of the railroad on the urban growth. In addition a quasi-experiment design reinforces this conclusion. In short, although moderate, our paper shows strong evidence of the influence of the railroad on Spanish urbanization. This conclusion is coherent with other research.
To assess the short-term impact of a nutritional intervention aimed at reducing childhood overweight in German pre-school children.
Using a cluster-randomized study design with waiting-list controls, we tested a 6-month intervention administered once weekly by a nutrition expert consisting of joint meal preparation and activities for children and parents such as tasting and preparing nutritious, fresh foods. At baseline, 6 and 12 months, a parent-completed questionnaire assessed fruit and vegetable intakes (primary outcomes) and water and sugared drinks consumption (secondary outcomes). Direct measurement assessed BMI, skinfold thickness and waist-to-height-ratio. An intention-to-treat analysis used random-effects panel regression models to assess the intervention effect, adjusted for each child's age, gender, immigrant background and maternal education.
Eighteen pre-schools from three south German regions.
Healthy children aged 3–6 years.
Three hundred and seventy-seven (80 %) eligible pre-school children participated in the study. Of these, 348 provided sufficient data for analysis. The sample mean age was 4·26 (sd 0·78) years; the majority (53·2 %) were boys. Children's fruit and vegetable intakes increased significantly (P < 0·001 and P < 0·05, respectively); no significant changes in the consumption of water, sugared drinks or anthropometric measurements were noted.
Nutritional interventions in pre-schools have the potential to change eating behaviours in young children, which in the long term might reduce risk for developing overweight.
Panel data are used in almost all subfields of the agricultural economics profession. Furthermore, many research areas have an important spatial dimension. This article discusses some of the recent contributions made in the evolving theoretical and empirical literature on spatial econometric methods for panel data. We then illustrate some of these tools within a climate change application using a hedonic model of farmland values and panel data. Estimates for the model are provided across a range of nonspatial and spatial estimators, including spatial error and spatial lag models with fixed and random effects extensions. Given the importance of location and extensive use of panel data in many subfields of agricultural economics, these recently developed spatial panel methods hold great potential for applied researchers.
These articles provide a discussion of studies presented in a session on spatial econometrics, focusing on the ability of spatial regression models to quantify the magnitude of spatial spillover impacts. Both articles presented argue that a proper modeling of spatial spillovers is required to truly understand the phenomena under study, in one case the impact of climate change on land values (or crop yields) and in the second the role of regional industry composition on regional business establishment growth.
In this paper, the impact of lifecourse family and labour market experiences on household incomes of older people in Belgium and the United Kingdom (UK) is analysed. To this end, panel data and life-history information from the Panel Study of Belgian Households and the British Household Panel Survey are combined. The results show that old-age income is indeed influenced by previous lifecourse experiences, and that differences between Belgium and the UK can be explained in terms of (the development over time of) welfare regime arrangements. Family experiences have a larger impact on old-age incomes in ‘male-breadwinner’ Belgium, while in Britain labour market events are more important. As social transfers in Britain are more aimed at poverty prevention and less at income replacement, a ‘scarring effect’ of unemployment persists even into old age. Also, the more of one's career is spent in blue-collar work or self-employment/farming, the lower the income in old age. A new finding is that, notwithstanding the high level of ‘de-commodification’ achieved by the Belgian welfare state, this effect turns out to be significantly stronger in Belgium than in the UK. Compared to the market, the welfare state is hence a more efficient ‘mechanism’ of stratification for incomes in old age.