Search results

Researchers in economics and other disciplines are often interested in the causal effect of a binary treatment on outcomes. Econometric methods used to estimate such effects are divided into one of two strands depending on whether they require unconfoundedness (i.e., independence of potential outcomes and treatment assignment conditional on a set of observable covariates). When this assumption holds, researchers now have a wide array of estimation techniques from which to choose. However, very little is known about their performance – both in absolute and relative terms – when measurement error is present. In this study, the performance of several estimators that require unconfoundedness, as well as some that do not, are evaluated in a Monte Carlo study. In all cases, the data-generating process is such that unconfoundedness holds with the ‘real’ data. However, measurement error is then introduced. Specifically, three types of measurement error are considered: (i) errors in treatment assignment, (ii) errors in the outcome, and (iii) errors in the vector of covariates. Recommendations for researchers are provided.

This paper aims to identify the effects of social learning and network externalities by conditioning on product quality and early sales momentum. This approach is demonstrated using film sales data.

This study used econometric modeling approach.

It was found that both social learning and network externalities have significant and comparable impacts on film choice. We show that the relative effects of network externalities and social learning in the film market are robust to different momentum and quality definitions and to alternative estimation methods.

Scholars have long argued that social learning plays a key role in new product diffusion. In some product categories, consumer choice may also be influenced by network externalities, meaning that purchasing popular products may provide the consumer utility above and beyond that derived from product usage directly. We propose a novel identification approach to help quantify the relative magnitude of these two effects on new product sales.

The purpose of this paper is to study the impact of unemployment on coronary heart diseases (CHD) in Italy on a sample of male manual workers in the private sector.

The authors investigate the association between CHD and different unemployment experiences (ever unemployed; short, mid and long cumulative unemployment), exploiting a large Italian administrative database on careers and health. The study design is based on the balancing of individuals' characteristics during a 12-year pre-treatment period; the measurement of unemployment occurrence during a seven-year treatment period; the observation of CHD occurrence during a five-year follow up. The workers characteristics and the probability of receiving the treatment are balanced by means of propensity score matching. Standard diagnostics on the balancing assumption are discussed and satisfied, while the robustness to violations of the unconfoundedness assumption is evaluated by a simulation-based sensitivity analysis.

The authors find a significant increase of CHD probability was found among workers who experience more than three years of unemployment (relative risks (RR)=1.91, p<0.1), and among those who exit unemployment starting a self-employment activity (RR=1.70, p<0.1). Using different selections of the study population, a clear pattern emerges: the healthier and more labour market attached are workers during pre-treatment, the greater is the negative impact of long-term unemployment on health (RR=2.79, p<0.01).

The very large representative sample (n=69,937) and the deep longitudinal dimension of the data (1985-2008) allowed the authors to minimize the risks of health selection and unemployment misclassification. Moreover, the adopted definition of unemployment corrected some undercoverage and misclassification issues that affect studies based on a purely administrative definition and that treat unemployment as a unique career event disregarding the duration of the experience.

This chapter reviews the recent developments at the intersection of personalization and AI in marketing and related fields. We provide a formal definition of personalized policy and review the methodological approaches available for personalization. We discuss scalability, generalizability, and counterfactual validity issues and briefly touch upon advanced methods for online/interactive/dynamic settings. We then summarize the three evaluation approaches for static policies – the Direct method, the Inverse Propensity Score (IPS) estimator, and the Doubly Robust (DR) method. Next, we present a summary of the evaluation approaches for special cases such as continuous actions and dynamic settings. We then summarize the findings on the returns to personalization across various domains, including content recommendation, advertising, and promotions. Next, we discuss the work on the intersection between personalization and welfare. We focus on four of these welfare notions that have been studied in the literature: (1) search costs, (2) privacy, (3) fairness, and (4) polarization. We conclude with a discussion of the remaining challenges and some directions for future research.

We provide straightforward new nonparametric methods for testing conditional independence using local polynomial quantile regression, allowing weakly dependent data. Inspired by Hausman's (1978) specification testing ideas, our methods essentially compare two collections of estimators that converge to the same limits under correct specification (conditional independence) and that diverge under the alternative. To establish the properties of our estimators, we generalize the existing nonparametric quantile literature not only by allowing for dependent heterogeneous data but also by establishing a weak consistency rate for the local Bahadur representation that is uniform in both the conditioning variables and the quantile index. We also show that, despite our nonparametric approach, our tests can detect local alternatives to conditional independence that decay to zero at the parametric rate. Our approach gives the first nonparametric tests for time-series conditional independence that can detect local alternatives at the parametric rate. Monte Carlo simulations suggest that our tests perform well in finite samples. We apply our test to test for a key identifying assumption in the literature on nonparametric, nonseparable models by studying the returns to schooling.

Various types of short training programmes exist in Germany. This article aims to evaluate short training courses for welfare recipients and to detect which programme type works best with respect to different outcome indicators.

The author measures the impact of six short training programmes on the participants. She uses propensity score matching and large‐scale administrative data to find suitable comparison groups. She compares treatment to non‐participation as well as participation in different training types directly.

The results demonstrate that in‐firm training has large positive effects on individual employment prospects and stability. Furthermore, classroom skill training is more effective than other classroom types while application training is rather ineffective. Treating some of the application training participants with an alternative sub‐programme would improve the effectiveness.

The results show which training types work best. However, the results do not imply macro or cost‐benefit effects.

Several of the short training courses, mostly occupation‐specific sub‐programmes, are short and relatively inexpensive options to activate welfare recipients.

The paper analyses the effects of six short training programmes for welfare recipients that have not been analysed before. The sensitivity of the results is examined using a wide range of estimators and different outcomes.

Job Corps is the United State’s largest and most comprehensive training program for disadvantaged youth aged 16–24 years old. A randomized social experiment concluded that, on average, individuals benefited from the program in the form of higher weekly earnings and employment prospects. At the same time, “young adults” (ages 20–24) realized much higher impacts relative to “adolescents” (ages 16–19). Employing recent nonparametric bounds for causal mediation, we investigate whether these two groups’ disparate effects correspond to them benefiting differentially from distinct aspects of Job Corps, with a particular focus on the attainment of a degree (GED, high school, or vocational). We find that, for young adults, the part of the total effect of Job Corps on earnings (employment) that is due to attaining a degree within the program is at most 41% (32%) of the total effect, whereas for adolescents that part can account for up to 87% (100%) of the total effect. We also find evidence that the magnitude of the part of the effect of Job Corps on the outcomes that works through components of Job Corps other than degree attainment (e.g., social skills, job placement, residential services) is likely higher for young adults than for adolescents. That those other components likely play a more important role for young adults has policy implications for more effectively servicing participants. More generally, our results illustrate how researchers can learn about particular mechanisms of an intervention.

The purpose of the paper is to estimate the effectiveness of two start‐up programs (bridging allowance and start‐up subsidy) in East Germany.

Using a combination of administrative data from the Federal Employment Agency and a follow‐up survey (for roughly 1,300/1,000 participants/non‐participants), the analysis is based on the conditional independence assumption. Kernel matching estimators are applied to estimate the treatment effects and test the sensitivity of the results using a conditional difference‐in‐differences approach.

At the end of the observation period both programs are effective: unemployment rates of participants are lower, and employment rates and personal income are higher when compared to non‐participants. Additionally, first descriptive evidence of the additional employment effects through direct job creation is presented, which is quite significant for the bridging allowance and negligible for participants in the start‐up subsidy.

Participants in the start‐up subsidy are in their third year of participation at the end of the observation period and mostly still receive further support (although at a reduced rate). Therefore, the results for this program have to be treated as preliminary.

In contrast to other active labor market programs that have been evaluated recently (including job‐creation schemes and vocational training programs) this paper finds considerable positive effects for start‐up subsidies. Hence, programs aimed at turning the unemployed into entrepreneurs may be a promising strategy in East Germany.

This is one of the first studies that allows inferences to be drawn about the effectiveness of start‐up programs in East Germany. Most previous studies on the effectiveness of active labour market policies in the Eastern part of Germany neglected these programs.

Works councils have the legal right to participate in a firm’s training process and, where necessary, call for a replacement of training instructors. The purpose of this paper is to empirically test whether works councils are associated with a higher quality of apprenticeship training – or its inputs or outputs – in Germany.

The authors use two representative cross-sectional surveys of German workplaces in 2007 and 2012/2013 that were conducted by the Federal Institute for Vocational Education and Training in Germany. To account for selection on observables, the authors apply nearest neighbor matching models to estimate the extent to which works councils are associated with training quality.

The results shed light on the influence of works councils on the quality of apprenticeship training in Germany. Based on a quality model, the authors show that works councils are associated with a (moderately) higher output quality of apprenticeships, particularly with respect to the share of retained apprentices. However, the authors do not find empirical evidence for a positive association between works councils and input- and process-quality indicators.

Although the identification of causal effects due to the existence of works councils is difficult and cannot be fully addressed in the analysis, the authors can use a number of important control variables at the workplace level. The results suggest that a works council only plays a moderate role in enhancing the quality of the German apprenticeship system.

The authors provide the first direct empirical evidence of how the existence of a works council is associated with the input-, process- and outcome-quality measures of the German apprenticeship system.

This study aims to provide a better understanding of how gift card receivers react to the types of gift cards. This study examined the effect of gift card types (intangible experiences vs less intangible experience vs tangible goods) on a recipient’s willingness to spend more through emotions and perceived effort (Study 1) and on feeling of appreciation (Study 2).

Study 1 adopted a scenario-based 2 (tangible vs intangible) × 3 ($100 vs $200 vs $300) between-subjects design. Study 2 narrowed the scope of gift card type (intangible vs less intangible).

Receivers tended to perceive less effort in gift card selection and feel less emotion when receiving gift cards for intangible experiences than when receiving gift cards for both tangible and less intangible products. However, as face value increased, gift card receivers for intangible experiences felt more pleasure and, in turn, rated higher willingness to spend more money than face value than those with gift cards for tangible products.

Future studies can rule out alternative explanations related to brand-related effects, previous experiences and personal preferences.

Service providers should put more effort into tangibilizing the intangibles to reduce receivers’ uncertainty. Also, they can increase their profitability by stimulating gift card receivers’ willingness to spend more money through pleasure.

Answering research calls for examining consumers’ perceptions of different gift card types, this study might be the first to unveil the differential effect of gift card types associated with the tangibility of products on purchase behavior and the underlying emotional mechanism.