Current Projects
The overarching objective of this project is to better understand how consumer inertia affects firms’ strategies, market outcomes, and the suitability of regulatory interventions. For this purpose, this project features three building blocks. In the first block, we aim to develop dynamic market models with behavioral consumers. These models are targeted to explain why consumer inertia is so wide-spread in subscription markets, why policy interventions often fail to activate consumers despite large potential switching gains, and why firms that earn high profits per consumer often refrain from offering even better initial deals to consumers. Importantly, we want to understand the role of technological trends (such as declining costs over time), contract design, consumer learning, and regulation restricting what contracts firms can offer in determining the various dynamic price patterns that lead to “loyalty penalties” (i.e. high prices for inactive consumers). In the second block, we plan to empirically analyze how online firms can use their website design (or choice architecture) to influence consumers’ purchase behavior, and combine our empirical insights with theory to determine what the resulting welfare consequences are. As a first step, we combine experimental measures for consumer mistakes and preferences with cookie data to study how consumers’ journeys are affected by the choice architecture of firms. Based on the empirical regularities, we develop behavioral online-market models that allow the firms’ online choice architecture to influence consumer choices. Thereafter, we will reconsider the data in light of our theories in order to quantify the welfare effects of A/B testing. We anticipate that the online data will help us to understand some issues – such as consumer learning or the lack thereof – that will help formulate dynamic market models in building block one. Based on the findings in the first two blocks and in other projects of this research group, the third block considers important policy questions of the digital economy explicitly allowing for behavioral biases, framing effects, and heterogenous consumer responses to inform the ongoing policy debate about online consumer protection as well as the regulation of large online platforms.
Building on experimental as well as observational field data this project seeks to determine the implications of retail investors’ limited attention for investment choices and firms’ design of financial products. In this project we are interested in those salience effects that financial firms can (easily) exploit, namely effects related to the labelling of financial products, the relative positioning of financial products and the marketed investment strategies for financial products. There is much suggestive evidence that such exploitation is going on. For instance, the number and variety of stock market indices has increased rapidly. For the last five years, there are more investable indices in the US than listed firms, while on average only five funds use a given index and 75 percent of indices are adopted by a single fund. Such niche indices charge higher fees than broadly adopted benchmarks, and just three firms capture 80 percent of the market for benchmark index provision in 2018. While classical economic models have a hard time explaining these patterns, insights from the literature on limited attention can. Learning about the precise channel that drives such patterns is particularly important to draw suitable policy implications. Precisely, we would like to understand experimentally (1) how benchmarks can be chosen to make products look more favorable and thereby increase firms’ profit margins, and (2) which investment strategies firms can prominently advertise in order to increase investments. By means of experiments and structural models, we want to test (3) how attention-grabbing labels can be used to direct demand toward financial products that are dominated in terms of their cost structure. Finally, structural models should provide insights on (4) how policies that restrict the availability of niche benchmarks and extend the consideration set of retail investors affect investor welfare and the profits of index providers and funds.
When facing economic choices, information on the available options is typically vast while consumers’ capacity to process this information is limited. Consumers’ attention, at least in parts, is non-directed and therefore automatically guided toward eye-catching features of the choice environment. Little is known about firms’ incentives to exploit such attentional imperfections of consumers. We investigate non-directed attention allocation for the example of a particularly quickly growing and important market that is currently in the focus of competition authorities around the world, the gaming market. Nowadays, many successful video games feature “loot boxes” that, just like gambling, offer random rewards to be used in-game (with respect to the “FIFA” game such a reward would be, for instance, a virtual Lionel Messi). In 2020 alone, such loot boxes generated $15 billion of worldwide revenue. There is an increasing public concern, however, that video game developers design loot boxes in a way to make gamers “overpay” for the typically small chance of getting the reward. We single out three features of loot boxes that might result in gamers overestimating the probability of winning the reward and, consequently, paying too much. First, developers typically do not disclose the odds, but provide gamers only with an interval-censored distribution of rewards. Second, in many games there is a public announcement whenever someone wins a reward, resulting in gamers observing a biased sample of the reward distribution. Third, high rewards are additionally disclosed in a spectacular visual presentation. In a series of papers, this project seeks to investigate three issues. (i) In laboratory as well as field experiments, we identify the features that drive the success of loot boxes and analyze in how far these features are exploitative in nature. (ii) We ask to what extent firms use these features strategically in order to raise users’ willingness-to-pay. And (iii) with the help of a theoretical model and with market experiments, we investigate how transparency regulations or improved competition in the market can raise welfare.
This project seeks to improve our understanding of how consumers direct attention to different tasks, and how consumer policies that aim to limit the exploitation of inattentive consumers affect the attention allocation of consumers. Competition in the marketplace together with the consumer protection regime determines what consumers need and want to pay attention to, and what consumer pay attention to feeds back into how firms compete for consumers in the first place. We study this nexus of interactions with a focus on consumer attention allocation. To achieve this objective, we will run carefully designed experiments in order to learn more about how well consumer attention allocation decisions correspond to those predicted by existing theories. This will provide a better understanding of how individuals divide cognitive resources across tasks. In market settings, the optimal design of products and contracts by firms affects how consumers allocate their attention across tasks and markets, which in turn feeds back into firms’ decisions how to design products and contracts. Furthermore, consumer policy, which in OECD countries heavily regulates which products and contracts firms may offer in the market, determines how much attention consumers need to spend to avoid being exploited, and thus interacts with consumer attention allocation decisions as well as firms’ strategic decisions. We thus plan to study the impact of various regulatory choices on consumer attention decisions and market outcomes theoretically. Finally, complementing the laboratory evidence as well as the theory developed, we plan to use observational evidence to test how firms respond to limited attention of consumers in the field. We plan to empirically study whether firms discriminate against consumers with more limited attention or lower cognitive ability. We anticipate that our research generates important new insights on the pros and cons of different consumer-protection-regulation regimes.
Consumers often make decisions without being fully informed. We aim to structurally estimate the extent to which different consumers take into account the available choices in differentiated products markets. To obtain credible estimates of limited information of consumers, our modelling approach is going to allow consumers to differ in the extent to which available options are considered and, in their willingness, to pay for product characteristics. Joint identification of preference and consideration set heterogeneity will be based on observed asymmetries in the substitution between different products. Structural demand models with a full information assumption imply symmetric diversion of demand between different products. Deviations from this symmetry indicate that not all consumers are fully informed. Such observed asymmetries therefore allow us to identify parameters that determine the probability that specific products are considered. The approach avoids imposing an exclusion restriction as for example assuming that marketing expenditure only shifts consumer attention but not consumer utility. This also allows price and other product characteristics to determine consumer attention and thereby yields a substantially more flexible demand model. Our aim is to apply this modelling approach to a large number of categories from the U.S. consumer packaged goods industry, as covered in the Kilts-Nielsen data at Chicago Booth. The data offers both scanner, consumer panel and marketing exposure data. Obtaining estimates of consumers’ consideration technology will allow us to quantify if firms have been able to raise their profitability over the last decade by steering consumers to their products through shifting consumer attention. This can shed some light on the potential drivers for systematic markup changes in the U.S. economy. It can also provide important insights on the effect of horizontal mergers in differentiated product markets, where typically full information is assumed. If and how merged firms adjust their efforts of gaining consumer attention post-merger is to the best of our knowledge an under investigated question with potentially important implications for competition policy.
In classical discrete choice models, consumers are assumed to make a comprehensive and careful choice between the available alternatives in terms of their utility consequences. It is well known that this is not an accurate description of the real-world consumer. Brand loyalty and inertia are often found to have a substantial on consumer choice. If inertia and other deviations from strict rationality are strong determinants of daily shopping decisions, this has important consequences for (i) understanding consumers themselves, (ii) the validity of standard choice analyses, and (iii) the strategic behavior of producers and retailers. Our goal in this project is to shed light on all of these aspects. While previous literature mostly focusses on single product categories, the fact that during a shopping trip consumers make dozens of choices among different categories suggests great opportunities to identify a more wholistic picture of behavior from a wealth of repeated and interrelated decisions and large amounts of data. By building on techniques from the machine learning literature, demand can be modelled simultaneously for many product categories. The approach is based on splitting behavioral parameters across individuals and categories. This allows the individual parameters to be identified from choice behavior in all categories at the same time: If we observe many yoghurt purchases of a household, this helps to identify the individual behavior for toothpaste. We will study various aspects of joint decision making using a large consumer panel data set by Nielsen. How often do people automatically pick the same kind of cereal or tooth paste without a deliberate choice? How loyal to a brand are they? How much do these kinds of effects differ across individuals? Is this an individual trait which is correlated over product categories? How much are future choices affected by current shocks such as advertising or product availability? How does a consumer react to bundling and similar store strategies? Finally, we will use the developed methods and behavioral insights to study the strategic behavior of producers and retailers in highly concentrated markets and discuss whether there is any role for legislation or regulation.
This project contributes expertise in foundational questions of behavioral economics and decision-making to the unit. We develop novel methodological tools for the measurement of time preferences, the sophistication about them, and their stability over time. One important aspect of time preferences is time-(in)consistency. Actual behavior may deviate from planned and rational behavior even though the decision environment has not changed. We aim at advancing our understanding of time-inconsistency, theoretically with regard to its causes, empirically through its measurement, and in an applied way, by investigating its stability and implications for real-world behavior as well as by providing tractable modelling approaches for other economists. Advancing the latter will involve axiomatic, applied theoretical, and econometric approaches. Moreover, since time-inconsistency is one important way to model self-control failure, we will address an open question concerning the nature of self-control, namely whether it can be understood as a unitary construct or, alternatively, as an umbrella term for a number of potentially related but ultimately different constructs. We will address this issue empirically by providing a comprehensive synopsis of empirical measurement tools of self-control from various disciplines in order to examine whether different self-control measures converge in what they suggest about a given individual (i.e., study their “convergent validity”).
What impact does household heterogeneity have on the transmission of monetary policy? Conventional New Keynesian macroeconomic models, which examine the transmission of monetary policy to aggregate output, employment, and prices, typically assume a representative household and do not account for household heterogeneity. However, households differ in terms of income, wealth, gender composition, and age structure, leading to a different consumption behavior (consumption patterns) and inflation rates. To address this, the so-called HANK (Heterogeneous Agent New Keynesian) models were developed. So far, these models primarily focus on differences in household income and wealth. This research project aims to extend HANK models by incorporating additional dimensions, such as income- and wealth-specific consumption patterns, as well as the fact that households experience different inflation rates.
The informative value and generalizability of panel surveys depends crucially on whether the panel participants represent the population well. Studies show that the response rates for surveys are falling. This is a major problem for panel surveys in particular. The aim of this project is to contribute to the explanation of non-response in panel surveys and to develop methods to determine the nonresponse bias of estimators.
What effects do vertical mergers and acquisitions (M&A) have on the success of companies, their pricing, market power and innovation? While the focus of competition policy has traditionally been on horizontal mergers, recent sector-specific studies suggest that a reassessment of vertical mergers is necessary. This research project aims to analyse vertical mergers for a broader range of industries. Modern econometric techniques and a unique dataset on the vertical relationships of individual companies are used for the analysis. The results can provide valuable information for policy makers and competition regulators.