For More Robust Decisions, Let Ambiguity In

The future is uncertain, its possible paths coming in and out of focus with time. Nevertheless, individuals and policymakers alike must make choices about which of those paths to follow. Understanding how planning in the face of ambiguity can be made more robust is the research focus of economist Dr. Philipp Eisenhauer. Traditional economic models estimate future outcomes by assigning known probabilities to risk-related components. Risk and uncertainty are not the same thing, though, and unique probabilities cannot be established for all possible random outcomes. In reality, people make decisions under ambiguity all the time and there are insights to be gained from adding it to economic analysis, like better explanations of observed economic phenomena and optimization of government policies, Dr. Eisenhauer explains.
Examining the case of educational choices and their impact on people’s future in the job market, he is developing microeconomic models that incorporate this missing notion of ambiguity. All existing research on the drivers of education decisions has considered risk, but not uncertainty. However, individuals’ school and career paths contain a random component, equivalent to luck on the job market. By leaving ambiguity out of the picture, traditional models assume that people can know the probability of each potential outcome of this uncertain element, luck. For this to be true, the economic environment would have to remain stable, which is unlikely to be the case, Dr. Eisenhauer points out; the advance of technology, for one thing, will alter it and one’s place on the market.
When applied to real datasets tracking people’s education and job experiences, the standard models also have trouble explaining the large numbers of high school and college dropouts. They resort to attributing this discrepancy to something called psychic costs, related to the effort of making a decision. “According to these models, psychic costs appear to dominate people’s decision-making, more than they should,” Dr. Eisenhauer says. “It’s really just an unexplained residual of the model, which is an unsatisfactory conclusion, because it provides no clear interpretation of how we can affect this problem with policies.” With his research, he aims to provide something more plausible and meaningful, with real-world applications.
He has developed a model for the analysis of economic phenomena that allows decision-making agents to confront ambiguous situations. Faced with binary choices like to go to college or not, to graduate from college or drop out, the agents in this likelihood of a range of possible, uncertain outcomes, they attempt to maximize the worst-case scenario outcome. When he fits the data on education decisions to this new model, instead, he gets a better fit: it successfully reduces the discrepancy that has, until now, been explained only (and incompletely) by psychic costs.

In his future research, the economist will elaborate on his ambiguity-inclusive model, allowing it to analyze more complex behavior in the agents it describes. His goal is to build better descriptions of human decision-making that can contribute to rethinking old policy and designing new interventions that could be more effective from the start. The ambiguity around the potential job market returns of investing in one’s education, for instance, could play an important role in people’s decisions. “If this turns out to be an important driver of educational choices,” Dr. Eisenhauer says, “it warrants a re-evaluation of proposed tuition policies and a re-design of optimal policies.” It’s an uncertain world, but his novel approaches could help us confront and cope with this ambiguity and derive the best possible decisions from the information we have.

Scientific title: Robust Dynamic Discrete Choices