FIRM Research Conference

Sasan Mansouri wins the 2024 research prize

Dr. Sasan Mansouri wins the FIRM Research Prize 2024 with his dissertation on the information behaviour of managers at analyst conferences. He shares the prize money of 30,000 euros with the chair of Prof. Dr. Mark Wahrenburg from Goethe University Frankfurt, who supervised Mansouri’s work.

Every two years, the FIRM research conference focuses on talented young researchers whose dissertations deal with current topics in risk management. FIRM brings the best to Frankfurt with the Research Award competition. 13 researchers from four nations applied for the Research Award this year. Three made it to the final. They were able to present their work in front of a packed audience and take part in discussions with practitioners and academics. In addition to Mansouri, Dr. Alina Steshkova from the Vienna University of Economics and Business and Dr. Urban Ulrych from the University of Zurich made it to the final.

It is no secret that managers often talk a lot in response to specific questions at the quarterly analyst conferences and yet say nothing. They try to maintain the information asymmetry between companies and investors with vague answers. But how do investors react when their questions are not answered or only answered inadequately? Mansouri investigated this in his dissertation with the help of AI-supported language models.

KI-SUPPORTED LANGUAGE MODEL

His dissertation comprises five in-depth studies on the dynamics of information disclosure and perception by the financial market. In order to measure how investors react to inadequate answers from interviewed company managers, a language model was trained to recognize so-called non-answers. The machine learning algorithm identifies phrases that often consist of non-answers or evasive answers, such as “[let me get]back to you” and “[it’s]too early to[tell]”. These identified non-answer markers are context-independent and can therefore be used universally – for example, in interviews with politicians.

REACTION OF MARKET PARTICIPANTS

For his research, Mansouri examined 600,000 questions and answers in telephone conferences between companies and financial analysts. The results show that financial markets react negatively when managers impede the flow of information through non-answers. In particular, non-answers increase investors’ uncertainty and their need to hedge against equity risks. Not surprisingly, non-answers occur more frequently when analysts’ questions are difficult, e.g. forward-looking or asked by less favorable analysts.

LACK OF CLARITY IS PUNISHED

Mansouri’s presentation was most convincing to the conference participants because his questions have a high practical relevance and he also uses innovative methods to process enormous amounts of data. “Mansouri’s study results prove that the capital market punishes a lack of clarity. He thus makes an important contribution to understanding the mechanisms of the capital market,” explains jury chairman Prof. Dr. Dr. h.c. Günter Franke, University of Konstanz. The decision to award the prize was not an easy one, as the two other papers presented were also innovative, relevant to practice and very well structured.

TRANSMISSION OF MONETARY POLICY DECISIONS

With her work on downside risks on the currency and equity markets, Dr. Alina Steshkova was able to show how monetary policy decisions by the Federal Reserve Board are transferred to the financial markets. In particular, she examined the relationship between exchange rates, interest rates and share prices. Traditional economic theory states that a tightening of monetary policy leads to an appreciation of the domestic currency. However, empirical studies show that domestic currencies sometimes also depreciate after interest rate hikes.

ASSET PRICES AND ANNOUNCEMENTS

To analyze the transmission of monetary policy to foreign exchange markets, Steshkova examined the co-movement of interest rates and stock prices in the wake of Federal Reserve meetings. The study shows that the US dollar appreciates following monetary tightening in the US when the co-movement between bond yields and the S&P 500 index is negative. Conversely, the US dollar depreciates when the simultaneous reaction is positive, challenging the conventional wisdom. This inverse reaction of the US dollar could be due to the information effect, whereby central banks also convey new assessments of future economic growth in their decisions.

STATISTICAL  LEARNING  IN  RISK MANAGEMENT

Dr. Urban Ulrych’s dissertation deals with the possibilities of statistical learning in quantitative finance and novel solutions for use in quantitative risk management. Urban shows that traditional asset allocation assumes known probability distributions of security returns and exchange rates. In the case of ambiguity, this assumption is weakened to the extent that the probability of a certain return is no longer fixed, but can itself vary stochastically. Ambiguity measures are therefore used alongside traditional risk measures.

AMBIGUITY AS A SUPPLEMENTARY RISK MEASURE

Ulrych examines the optimal currency allocation for risk- and ambiguity-averse international investors. A theoretical model is developed which shows that in a securities portfolio a higher aversion to ambiguity induces a stronger hedging of currency risks. This correlation is confirmed by way of example. The model provides new insights into the currency-related home bias and improves the stability of the currency allocation. The study also presents an optimization approach that can be used to optimize a portfolio of securities and currency positions simultaneously.

Other important topics at the FIRM research conference included a study on credit standards at banks and a practical report on the use of quantum computing in risk management.

Impressions of the research conference