(1) Cultural values and personality traits and their relation with financial decision making (Bannier/Walter/Winker)
Understanding individuals’ financial decision making has become ever more important. From an individual perspective, efficient savings and investment decisions are relevant to reach the desired level of prosperity and financial independence. From the viewpoint of society, changes to social security and pension systems make households’ private financial provisions indispensable for upholding a certain standard of living in times of crises or old age. Finally, against the background of an ever increasing complexity of financial instruments, financial institutions strive towards a better understanding of the needs for financial advice.
Research studies in the last years have considered different issues related to households’ financial decision making, among them the effect of socio-demographic characteristics such as gender, age, education or income. Personality traits, cultural values or ethical views have so far received scant attention, however. This is where the research project tries to contribute, using survey datasets that have not yet been employed for studying questions on financial decision making, but that nevertheless deliver some information on individual households’ savings behavior and wealth accumulation.
(2) Analyses of sentiment in German business texts (Bannier/Walter)
This project aims at applying (and broadening) the BPW wordlist by Bannier et al. (2017) on different types of business texts to examine the sentiment in German business texts and its effects on financial markets. This context-specific German word list allows to compare the results from earlier analyses derived from English texts and find out whether the German background delivers a different outcome that can be related to social or cultural differences between the Anglo-Saxon and the European business world. The BPW wordlist moreover enables us to examine texts that are available only or predominantly in German, such as the speeches of CEOs at companies’ annual general meeting or interviews of corporate representatives in German newspapers.
(3) Exchange Traded Funds (ETFs) and market efficiency (Bannier/Tillmann)
Passive investment strategies have seen an unprecedented upsurge over the last years. Ever more investors shy away from an active – and hence expensive - management of their money and rather invest in vehicles that simply mirror the movement of a broad market index and that are typically traded on an exchange. As this simple investment strategy does not require any research on the fund manager’s side, these exchange-trade funds are much cheaper than actively managed funds. This tremendous movement out of active and into passive investment has shaken the fund industry to its core.
While the repercussions on the industry are currently hotly debated, a question that has received much less attention is what the consequences of this development on financial markets will be. A concern recently aired is that the lack of any research and the “blind” investment in the current market, i.e., index, structure via ETFs will lead to less efficient capital markets. While this concern is easily explained, it is much more difficult to test. This research project aspires nevertheless to undertake such a test by use of daily data on emerging market fund in- and outflows.
(4) Debiasing Techniques (Ewelt-Knauer/Walter/Wöhrmann)
Behavioral research in accounting and finance has identified various decision-making and judgment processes where biases lead to poor results. This encompasses but is not limited to management decisions, nonprofessional investor decisions, auditor judgements or supervisor evaluations. For example, managers are susceptible to escalate commitment, investors are overconfident and auditors are affected by anchoring heuristics.
To overcome such biases “debiasing techiques” are required. Prior research has undertaken first steps in this field and identified techniques on a case-by-case basis. A coherent approach, however, that also considers how personal characteristics interact with specific debiasing techniques is still absent. This is where this research project intends to contribute.
(5) Information-overload (Ewelt-Knauer/Walter/Wöhrmann)
While econ theory predicts that more information is always beneficial, behavioral research draws a more differentiated picture. Specifically, increasing the amount of information beyond a certain limit reduces decision-making quality. Today, firms have elaborated information systems that provide management with a huge amount of data. Management accountants are in charge of designing reports that trade-off information needs and the quantity of information. From an external perspective, the length of financial reports is also constantly growing. Investors struggle with the volume of these reports and other sources of information such as information documents for packaged retail investment and insurance products.
Eye-tracking and other techniques are needed to gain a deeper understanding of how information is acquired and processed. Research in this field is required to better understand whether the presentation of information can be altered in a way that decision-makers are better able to cope with information overload.
(6) Identifying policy shocks through text analysis (Tillmann/Winker)
Monetary policy shocks are unexpected changes of current and expected policy. We aim at identifying shocks based on textual analysis of central bank communication. In particular, measures of text similarity are used to locate and quantify significant changes in the minutes of policy meetings, central bank speeches and publications. We also interact the similarity measure with other properties of texts such as the degree of hawkish/dovishness or the sentiment conveyed. The series of policy shocks is then used to quantify the effects of monetary policy. Results are compared to those obtained by conventional multivariate time series modelling.
(7) Measuring trust from text data (Tillmann/Winker)
We aim to measure the degree of public trust in selected decision makers with regard to central goals, e.g., the fulfillment of the ECB’s mandate. For that purpose we use different methods of text analysis applied to large sets of publications addressing the selected decision makers, e.g., newspaper articles. The resulting measures of trust are compared to survey-based indicators and is used in further empirical analysis of ECB policy. These methods are also applicable to measuring trust based on financial reporting and business communication of firms.