Rewards, punishments, and rationality in the stock market

When public companies announce financial results that fail to meet certain benchmarks (such as earnings, income, or their growth rate), their stock prices decline. In fact, they decline significantly more than they would have increased had the company managed to exceed the benchmarks by the same amount. In his Master’s thesis, MTEC MSc student Jonas Reinecke sought to find out whether there is a rational basis for this phenomenon.

Jonas wrote his Master’s thesis, “Meet-or-Beat: A Savings Model of Discretionary Accruals”, at Stanford University under the co-supervision of D-MTEC Professor Didier Sornette and Professor Anne Beyer of the Stanford Graduate School of Business. He hypothesised that managers can generally manipulate their earnings figures to a certain extent. When a company admits that it has failed to meet a benchmark, then, it must no longer have any leeway for such manipulation. This in turn implies that it has already been overstating its earnings in the past. According to this hypothesis, then, investors’ behaviour is indeed rational.

The mathematics

Jonas was able to draw on his background in robotics when creating his model. Like moving robots that determine their location based on data feeds from sensors, investors continually update their behaviour based on the earnings reported by companies. The agents in Jonas’ model follow formal strategies that result in asymmetrical behaviour with respect to companies’ benchmarking, but in doing so they remain rational actors.

“Solving the model turned out to be significantly more challenging than I expected,” Jonas notes, “but I had an absolute blast working on all the mathematical challenges I faced.” Not only was the mathematics fun, however. Jonas adds that “My supervisor at Stanford brought me along to invite-only conferences where I got to talk to Nobel Prize Winners – once without even knowing it. I am very grateful for this exciting experience.”

Democratising automation

Besides his time at Stanford, what Jonas most fondly remembers from his ETHZ days are the robotics courses he took from Professor Raffaello d’Andrea and the skiing trip to Pizol with his MSc MTEC peers. After earning his MSc with distinction and receiving the Ernst Blickle Graduate Award, Jonas joined Daedalus, a Karlsruhe- and San Francisco-based company that builds factories using AI-enabled robotics. He describes the company’s mission as to “democratise automation in manufacturing by building autonomous and instantly reconfigurable factories.” The time at ETH and Stanford, Jonas thinks, “gave me a solid foundation in business and management, a lot of which I am now using to manage strategy and growth at Daedalus.”