Contributions in Merlin

In this paper, we propose a new method to estimate non-fundamental demand shocks for green financial assets based on the arbitrage activity of exchange-traded funds (ETFs). By estimating the monthly abnormal flows into environment-friendly ETFs, we construct a Green Sentiment Index that captures shifts in investors' appetite for environmental responsibility that are not yet priced in the value of the underlying assets. Our measure of green sentiment differs significantly from the news-based climate indexes proposed by the extant literature, and it has additional explanatory power on both stock returns and corporate decisions. Over the period 2010-2020, shifts in green sentiment anticipate a persistent stock-price out-performance of more environmentally responsible firms, (of approximately 53 basis points over six months for a one-standard-deviation higher green sentiment) as well as an increase in their capital investments and cash holdings, particularly for more equity-dependent ones.

We study the impact of the public disclosure of environmental, social, and governance (ESG) ratings on information asymmetry. Our empirical results show, that consistent with our theoretical predictions derived from the strategic information asymmetry model, ESG ratings contain private information. Stocks for which an ESG rating becomes publicly available exhibit a significant reduction in their levels of information asymmetry. This finding is likely to be causal. The results persist in robustness tests in a difference-in-difference setting using a propensity-score-matched control group.

In light of the growing use, acceptance of, and demand for, machine learning in many fields, notably data science, but also other fields such as finance- and this in both industry and academics, some university departments might wish, or find themselves forced to, accord to the winds of change and address this pressing issue. The goal of this document is to assist in designing relevant courses using material at the appropriate mathematical level. It protocols, sorts, evaluates, and contrasts, numerous viable books for a variety of possible courses. The subjects span several levels of, and different avenues in, linear algebra and real analysis, with briefer discussions of material in probability theory and mathematical finance.

This paper investigates how mortgage credit conditions affect housing markets and the demand for homeownership. Using unique data on homeowners' listings and transactions and exploiting policy-driven changes in mortgage credit conditions in China, I provide empirical evidence that tightened mortgage credit conditions have a negative effect on housing demand and prices. Estimating a structural model of households' demand and supply of residential properties, I obtain measures for market liquidity and bargaining power of home buyers and sellers and find that mortgage interest rates and down payment requirements negatively affect the value of owning residential properties. With counterfactual experiments, I quantify the impact of mortgage interest rates, down payment requirements, property tax rates, and transaction tax rates on housing demand, supply, and prices. At the cost of a welfare loss for home buyers (owners), 1 percentage point higher mortgage interest rates (property tax rates) reduce demand-supply imbalance and decrease the average transaction price by 2.3% (1.9%), which cannot be achieved with higher transaction tax rates.

Focusing on the QE program of the Eurosystem in the covered bond market, this study shows that yields of eligible and closely matched ineligible bonds diverge in response to central bank buying activity. The magnitude of this price pressure depends on the country of issuance and is not there in all euro area countries. In countries with high price pressure, banks issue abnormal amounts of bonds in the first two years after the start of the QE program, not before and not after. The source of price pressure is central bank buying activity as it is larger in countries with on average smaller bond sizes and its effect declines after persisting for 12 months as buying activity abates. These findings suggest mispricing between similar bonds is an important factor in explaining abnormal bond issuance during a QE program.

We study the post-Great Recession market for retail investment products. With an experiment, we show that low interest rates drive investment demand but not product differentiation. Elicited margins go hand in hand with investors' underestimation of complex risk exposures. We empirically document that (i) rising complexity follows market growth, (ii) issuer margins increase in complexity, and (iii) simpler products first-order dominate more-complex products. Furthermore, biased dependency perceptions predict margins in the cross-section. Consistent with limited buy-side learning and growing sell-side competition, banks employ strategic price complexity to mitigate competitive pressure. Our findings showcase how low interest rates fuel excessive risk-taking.

We introduce deep equilibrium nets---a deep learning-based method to compute approximate functional rational expectations equilibria of economic models featuring a substantial amount of heterogeneity, significant uncertainty, and occasionally binding constraints. Deep equilibrium nets are neural networks that directly approximate all equilibrium functions and that are trained in an unsupervised fashion to satisfy all equilibrium conditions along simulated paths of the economy. Since the neural network approximates the equilibrium functions directly, simulating the economy is computationally cheap, and training data can be generated at virtually zero cost. We demonstrate that deep equilibrium nets can solve rich and economically relevant models accurately by applying them to solve three different models, all featuring a very high-dimensional state space. Specifically, we solve two overlapping generations models with aggregate and idiosyncratic uncertainty, illiquid capital, a one-period bond, and occasionally binding constraints. Additionally, we solve a Bewley-style model with a continuum of agents, aggregate and idiosyncratic risk, borrowing constraints, and recursive preferences.

This paper provides evidence that corporate boards violate the informativeness principle in their forced CEO turnover decisions by failing to ignore uninformative performance outcome signals. I show that CEOs of firms with barely positive shareholder returns in the previous year are less likely to be dismissed than CEOs of firms with barely negative returns, even though this return outcome is conditionally uninformative. I observe a similar pattern for stock returns relative to the S&P 500 index return: a firm's board is less likely to dismiss its CEO if the firm barely outperformed the S&P 500 index than if the firm barely underperformed the S&P 500 index. Moreover, I demonstrate that the tendency of boards to consider uninformative absolute return outcomes has decreased over time, while their tendency to consider uninformative relative return outcomes has increased over time. This suggests that boards have shifted their focus toward relative returns while continuing to violate the informativeness principle.

Serious games have emerged as a promising new form of education and training. Even though the benefits of serious games for education are undisputed, there is still a further need for research on the efficacy of such games. The main goal of our research is to examine the effectiveness of a serious moral game—uFin: The Challenge—that was designed to promote moral sensitivity in business, a precondition of ethical decision-making and behavior and a core moral competence of moral intelligence. A second goal is to examine the role of metacognitive prompting and prosocial nudging in influencing learning effectiveness. Participants (N = 345) took part in an experimental game-based intervention study and completed a pre- and post-test questionnaire assessing moral sensitivity. The analyses of both questionnaire and game data suggest that the game is effective in promoting moral sensitivity. Neither self-reflection nor exposure to prosocial nudges, however, were determined to be factors that improve learning effectiveness. In contrast, those interventions even decreased the learning outcome in some cases. Theoretical and practical implications, limitations and avenues of further research are discussed.

Given the competitiveness of a market-making environment, the ability to speedily quote option prices consistent with an ever-changing market environment is essential. Thus, the smallest acceleration or improvement over traditional pricing methods is crucial to avoid arbitrage. We propose a novel method for accelerating the pricing of American options to near-instantaneous using a feed-forward neural network. This neural network is trained over the chosen (e.g., Heston) stochastic volatility specification. Such an approach facilitates parameter interpretability, as generally required by the regulators, and establishes our method in the area of eXplainable Artificial Intelligence (XAI) for finance. We show that the proposed deep explainable pricer induces a speed accuracy trade-off compared to the typical Monte Carlo or Partial Differential Equation-based pricing methods. Moreover, the proposed approach allows for pricing derivatives with path dependent and more complex payoffs and is, given the sufficient accuracy of computation and its tractable nature, applicable in a market-making environment.