There are three things we need to make wise and rational decisions with confidence: time, information and smarts. Unfortunately, as mere humans, we know that these resources are in short supply. Indeed, the vast range of choices we have to make daily and the overwhelming influx of information and stimuli can limit or confuse all three, leading to decision fatigue.
Which career path to take, which asset-allocation to follow or (in my case) what to order from the restaurant menu are questions whose answers we can never be entirely confident of because of limits on our time, information and smarts. These practical limits put “bounds” on our rationality. Indeed, there may not even be a single best choice for any hard decision. Consequently, people make decisions using rules of thumb that are ‘good enough’ (or “satisficing” in the jargon.) Many philosophers would consider even satisficing as rational insofar as we “adopt suitable means to our ends”.
In contrast, classical economics treats people as entirely rational entities. This means they always make the decision that maximizes their welfare based on complete information, sufficient time as well as super-human cognitive skills and computing power. I’ve been thinking a lot about whether and how we can improve our decision-making by pushing these bounds and approach this ideal of rationality. Satisficing will usually get us outcomes that are less than the best possible . Can we do better?
Herbert Simon (who happens to have taught at my alma mater), wrote extensively on the concept of bounded rationality. He proposed it as a more realistic way to understand and model how real people actually make economic decisions. As early as 1957 he described decision-making as a process of searching for a way to get the value of a “goal variable” to exceed a minimum “aspiration level”. The goal variable could be one’s happiness or income, for example. The decision-maker adjusts the aspiration level down if finding a solution were too difficult; up if it were too easy. Real humans emphatically do not try to optimize, i.e., obtain mathematically the best possible outcome. This was a pretty radical thought at the time because it contradicted the foundation of prevailing economic theory.
In my career as an investment banker to municipalities and nonprofits, I have tried to expand the bounds on rationality by designing decision tools for finance staff. I theorized that many borrowers choose to borrow with traditional fixed rate bonds rather than lower cost alternatives because it is the default choice, using the “default heuristic”. Even if it’s a lot more expensive than alternatives, it was good enough for their purposes and didn’t require a whole lot of comparative analysis, which their limited staff may not have had the time or capacity to perform.
To address limits in finance staff time, information and computing power, I developed a decision-making framework and Excel optimization tool that addressed the key risks and goals of financing decisions. Staff input the most important risks, the various financing alternatives and the model selects the alternative with 1) the best expected result and 2) which does not present too much risk. Normally, investing a few hours in this process saves millions annually in financing costs.
Bounded rationality is a fact of human life. So is the use of tools. We can design tools to aid decision-making in environments like Excel to help push the bounds of rationality and achieve better decisions.