Internal Lunch Seminars

Gender-based wage discrimination and the backlash effect in recruitment and dismissal processes: Experimental evidence from Slovakia

Abstract: A sample of 155 HR managers participated in an experimental vignette study. In Task 1, they evaluated three CV resumes in terms of the candidates’ competence, hireability, likeability and wage proposal of three candidates applying for a regional manager post. Half of the sample received CVs presented as females’ and half as males’, otherwise the CVs were identical. Generally, male and female candidates were evaluated similarly in terms of competence and hireability. Average and worst male candidates were evaluated as less likeable than identical females. However, wages offered to female candidates were significantly lower than those offered to male candidates. We were unable to identify moderators of the phenomenon other than female HR managers driving the effect. In Task 2, participants were showed a set of vignettes presenting six employees (3 men and 3 women) preselected to be dismissed due to the economic crisis. Apart from basic demographics, the employees were described in terms of age, years in the company and frequency of absences. Again, we switched employees’ gender for half of the sample. We have found that HR managers are more likely to dismiss male employees and that they are particularly unforgiving to male workers with frequent absences.

MS Teams link: http://tiny.cc/mues2020

Big data analytics and innovation performance: the role of dynamic capabilities view

Big Data Analytics (BDA) has become a crucial source of competition and over the last several years been ranked among the top agenda items of business and IT executives. While some research has been conducted to explore the link between BDA and its potential business value, current knowledge on the link between BDA and innovation performance remains unclear. To fill this gap, based on the dynamic capabilities view, this research proposes a conceptual model to explore how and under what mechanisms, using BDA can influence innovation performance. This research conceptualizes firm agility, in terms of dynamic capability, and introduces three related constructs (i.e., sensing agility, decision-making agility, and acting agility). We also consider the moderating roles of data-driven culture and BDA team sophistication. Using the survey data, we uncover that dynamic capabilities mediate the link between BDA use and innovation performance. Besides, we find out that data-driven culture moderates the link between sensing agility and decision-making agility; however, such a moderating effect is not observed on the relationship between BDA use and sensing agility. This research also supports the moderating role of BDA team sophistication on the link between BDA use and sensing agility. We conclude the paper by providing contributions and future research directions.

Adding fuel to the fire: Experiment on a role of provocation in conflict

Agents with extreme positions may take actions in order to trigger a conflict. In the degree of violence, these actions may range from violent riots to symbolic political acts and the creation of fake news. We develop experimental design that investigates whether a provocateur can trigger a conflict by making cheap-talk statements. The experiment consist of three treatments: control without provocateur, treatment with provocateur and treatment where provocateur's preferences are known.

Choosing the Mode of Transport - Case Study of Bratislava

We analyse commuting patterns in Bratislava's fast growing sub-urban region with sub-optimal developed infrastructure. Standardized discrete choice model is used to estimate demand for individual car transport as well as for public buses and trains and to obtain corresponding elasticities with respect to travel costs, times and income. We find low rate of substitution between available modes. Direct price elasticity for public modes is in accordance with often cited rule of thumb -0.3. Negative income elasticities of demand for buses and trains, together with low direct price elasticity for car transport can be hard to overcome when looking for solution of current traffic problems in the region. We use modelled demand to predict effects of two recently proposed policies - new parking system in Bratislava city and construction of highway D4R7. In case of first policy, we expect massive reduction in car usage due to increased costs for car commuters. On the other hand, new highway would have only limited impact on mode choice and could reduce number of train commuters.

Secessionism and Foreign Trade: The Case of Catalonia

The paper aims to determine whether the political conflict between Catalonia and Spain leads to higher foreign exports in the region. Empirical analysis, based on the gravity model approach, indicates that the trade and intrastate conflict in Catalonia are related. Both the Poisson model with country and time fixed effects, and the fixed-effects panel data model show that Catalan foreign exports are positively associated with the past values of conflict in the region. Testing for endogeneity fails to support the view of a reciprocal relationship between domestic conflict and foreign trade.

Rejection stings! How does being deliberately chosen to be (less or more) relevant affect performance?

with Ondřej Krčál

Previous literature has explored how agents respond to a signal of perceived distrust and shows that it lowers the agent’s motivation to act in the trustor’s (or principal’s) best interests. Distrust, in these studies is conveyed through either limiting the agent’s choice set or increasing the extent to which the agent is monitored. The proposed study is based on the idea that distrust may also be conveyed when a principal deliberately reduces the power of an agent to influence a given outcome based on the agent’s past track-record of performance. This may occur in situations in which a superior should decide, based on the past performance of the workers reporting to her, how to allocate responsibility among them. In such situations, the level of responsibility allocated to a worker may reflect the superior’s expectations of that worker’s performance relative to others. In other words, it may convey a signal of the extent to which a superior (dis)trusts how well one worker would perform relative to others. While it may seem natural for superiors to place the most responsibility possible with the worker that performed the best in the past, this could backfire if there is a large drop in the performance of the employee who feels he has been made needlessly irrelevant. This is because this worker might perceive the allocation decision as a signal that he has been distrusted as a result of a superior having an (unreasonably) low expectation of his performance. In the real world, it could be that this feeling that one has been made (disproportionately) less relevant is actually independent of the level of responsibility one has. In order to measure the pure effect of being made to feel less relevant (or feeling distrusted), we propose a lab experiment that varies the deliberate nature of the allocation decision while keeping the level of responsibility constant. Based on the previous literature, our hypothesis is that workers who are deliberately made less relevant will perform worse than those for whom the decision was not deliberate, keeping the level of responsibility constant. Additionally, we predict that this effect will be stronger for males than females and so we plan to gender balance the experimental treatments.

Why gender matters and how to approach it in economic research

The presentation introduces the issue of gender in economics. I will show why gender matters in economic research and how should it be investigated. I will discuss three working papers: one on the approaches to measure gender-related self and gender differences and two showing possible applications of the approach in the research practice. I will start with explaining what a good “gender measure” is and which conditions it should satisfy. Then, I will proceed with discussing pitfalls of the dominating simplistic “gender as a variable” paradigm contrasting it with a “gender as an influence” approach. The main aim of the presentation is to point out possible gains from abandoning the traditional perspective treating biological sex as an explanation of observed differences in favour of a more sensitive, gender-oriented approach.

Social Capital and Mobility: An Experimental Study

(with Ondřej Krčál and Štěpán Mikula)

Theoretical models of social capital (David et al., 2010; Bräuninger and Tolciu, 2011) provide an explanation for the different outcomes of integration efforts documented by the recent empirical literature. The models show that communities may find themselves in two different equilibria, one with a high level of social capital and low outmigration or one with a low level of social capital and high outmigration. However, empirical tests of this explanation based on quasi-experiments typically suffer from the selection of immigrants which makes the identification challenging. In order to address this issue, we take the setup used in the theoretical models into the laboratory. We implement a treatment in which we change the initial level of social capital without affecting the equilibrium outcomes. In a finitely repeated game, most of the experimental communities end up in one of the two equilibria predicted by the theoretical models, with a higher proportion of the treatment communities attaining the equilibrium with a high level of social capital and low migration. This suggests that the initial levels of social capital might facilitate the effect of immigration on community levels of social capital.

Do People Prefer Inefficient Rules Over Discretion?

(with Rostislav Staněk and Katarína Čellárová)

When lawmakers create laws, they must decide not only on the substance but also the form of the law. The choice of the legal form may be described as a choice between rules and discretion. Although the discretion may more efficient, recent studies suggest that individuals value institutions and procedures for their intrinsic value. i.e. beyond the expected utility associated with the achieved outcome. The aim of our experiment is to identify discretion aversion. Our design combines an effort provision experiment and a voting experiment. Despite the fact that discretion regime is more efficient than the strict rule regime in our experiment, subjects vote significantly more often for the strict rule regime. In order to distinguish whether the unpopularity of the discretionary regime is driven by the presence of the human factor, we ran additional treatment where the human officer is replaced by the computer. The discretionary regime remains unpopular even in the treatment with the computer officer.  

Retirement Savings of the Self Employed - Experimental Desing

Calculation of retirement savings shows that vulnerable subgroup of the self-employed does not save enough to cover their pensions at an acceptable level. Their status in the Czech tax and savings system allows them, within limits, to choose their own level of savings. It might, therefore, be possible to provide a nudge to influence the decision making of the self-employed in order to increase their savings rate. An experiment to test the most promising route - to increase the information level - will utilize the life cycle experiment in ztree. Details of the experiment will be discussed in the presentation.