In a previous post we began discussing the work of 2 researchers from the well-respected Max Planck Institutes, on how financial professionals make decisions in the face of uncertainty rather than probability; and stated that the research could be instructive in understanding how a financial counterparty might behave in addressing the negotiation of a complex transaction.
In this further post, we explore some of the approaches followed by individuals in such circumstances.
Our readers will no doubt be familiar with the concept of heuristics, mental “rules of thumb” which all of us employ, whether we admit it or not, in making decisions, complex or simple. Not surprisingly, the subjects of the research made use of a range of such approaches in addressing two related factors, i) model uncertainty: trying to address the fact that the economic world is so complex that no single participant can hope to identify and assess all the probabilistic factors; and ii), strategic uncertainty: the fact that a participant cannot always identify or understand what the actions of other parties might be, whose own decisions may further influence outcomes.
In this post, we shall focus on model uncertainty.
The authors identified a number of ways in which participants addressed this risk; but 3 in particular seemed to stand out across both sets of interviewees (from a large investment bank, and from a significant central bank):
- A “scientific approach”, in which hypotheses were developed and attempts made to test them with data. Of course, this might simply lead to the conclusion that the problem was one of uncertainty, not probability;
- Institutionalizing group discussion by having a process that involved some element of formal committee structure that was part of the decision-making process, or available to offer advice and criticism (in the proper, neutral sense) to decision-makers. Naturally, that then raised issues of governance and group dynamics, but was a widely-adopted mechanism; and
- Constructing and evaluating narratives to put “flesh” on the bones of statistical and quantitative models and thus enable those involved to have frames of reference on which to base decisions. This, of course, raises the issue, which we have mentioned in previous posts, of “framing”, in which participants manage to limit their analysis and discussion to areas within their level of understanding and “comfort” and thus ignore or miss potential tail risks.
Interestingly, managing risk per se was frequently also considered as being as important as predicting uncertain outcomes. Frankly, this should be considered a statement of the obvious and the essential; but can sometimes be forgotten in the “excitement” of making a decision that may bring significant rewards, but also bears significant risk. After all, uncontrolled, or uncontrollable risks are what tend to blow up seemingly sound institutions.
Each of the above approaches has its own limitations; and it is clear to us at Awbury (as it was to the study’s authors) that what is needed is a multi-disciplinary approach to decision-making, in which those involved are aware of the boundaries of the their own knowledge and understanding of the model(s) they are using; are able to assess the extent to which these factors may or should limit their ability to make an effective decision that is properly risk-aware and risk-adjusted; understand the limitations of the model(s) used; and have the capacity both to recognize and assess “tail risks”, and the discipline to walk away from a risk that is not capable of being “boxed”. In short, process, knowledge, experience, maturity and dispassionate judgement are essential factors in being able to manage the risks faced by all those who operate in our markets. At Awbury, we strive to practice such an approach consistently and, by doing so, provide our clients with timely and effective solutions to their often complex and seemingly intractable problems.
In our third post on this topic, we shall explore decision-making as applied to strategic uncertainty.
-The Awbury Team