Herbert Simon on bounded rationality and decision-making

Bounded rationality — Herbert Simon on decision-making under uncertainty and complexity

Rationality, search, and satisfaction

You’re likely to find some tension in any organization you walk into. Even when everybody agrees on the broad direction, there’s going to be some disagreement about the best course of action and allocation of capital. I’m willing to bet that almost everybody, from managers to policymakers, can attest to this experience.

But as Herbert Simon asks tongue in cheek in his lecture Rational Decision-Making in Business Organizations: “why [do] they not, as [our] economics books suggest, simply balance off the marginal return of one activity against that of the other?”

The answer, Simon notes, is obvious (at least to the lay person): they can’t! The cost-benefit calculus necessary for such a solution is inherently difficult. And disagreement arises in part because people have different values, perceptions, and priorities. Put another way, “there is no unique determination of [their] subordinate goals”.

Sure, groups can and do reach resolutions. Sometimes, their choices are even optimal. But it depends a lot on the knowledge, structures, experience, and conditions (e.g., time constraints) of the decision-making body. More crucially, the mechanism that gets them there differs mightily from what classical economic theory puts forward.

Skip ahead

Bounded rationality, uncertainty and complexity
Satisficing and optimizing
Generators, evaluators, and stop rules
Plumbing tools and Bayesian dials

Bounded rationality, uncertainty and complexity

The rational firm, as classical theory assumes, has perfect knowledge, stable preferences, and the ability to undertake massive number crunching — in essence, they possess what Simon calls global rationality. We know in practice, however, that human rationality is limited (or bounded); and the product of two chief challenges: uncertainty and complexity.

Uncertainty

Firms, governments, and households have incomplete information. They see through the glass darkly. Customers do not know whether they like some new flavor of ice cream until they taste it. The startup, likewise, does not know much about their demand or production curves until it goes to market. Then, like a wild insect mating ritual, an intricate dance of price and product discovery takes place. All of us are learning and adapting in a sea of uncertainty.

Complexity

There is also the issue of complexity. In games of chess, business, or life, there are just far too many permutations and possibilities to consider. Decision-making in turn isn’t some instantaneous process. There is a real cost to search and calculation. Even chess-playing supercomputers have a finite search depth. Humans, likewise, have to rely on conscious and intuitive selection devices to cull their complex problem and decision tree into something manageable.

Satisficing and optimizing

Simon outlines three ways in which people grapple with uncertainty and complexity. That is, to “transform intractable decision problems into tractable ones”. The first, he says, “is to look for satisfactory choices instead of optimal ones”.

Second is to “replace abstract, global goals with tangible sub-goals”. The third then is to “divide up the decision-making task among many specialists, coordinating their work by means of a structure of communications and authority relations”.

Indeed, we have to recognise that much of our decision making in business and governments consist of approximations, heuristics, and satisficing. While these solutions are not necessarily the best, they are ‘good-enough’ given our constraints.

Even consultants who peddle whizzbang models and ‘optimal’ solutions to Fortune 500 clients are satisficing. Their work is premised on a simplification of reality, consisting of assumptions that they believe are good enough. As Simon writes:

“Decision makers can satisfice either by finding optimum solutions for a simplified world, or by finding satisfactory solutions for a more realistic world. Neither approach, in general, dominates the other, and both have continued to co-exist in the world of management science.”
Herbert Simon. (1978). Rational Decision-Making in Business Organizations.

If we are honest about our abilities, the “survival condition” in business is not “maximum profits” but “positive profits”. It frustrates me to no end when consultants, investors, and economists boast about their ability to find optimum solutions to complex, real-life problems. What does it mean to maximize in a world that we cannot fully appreciate?

Generators, evaluators, and stop rules

It follows that to make good decisions, we need cost-effective: (1) move generators, (2) evaluators, and (3) stop rules. Unlike Homo economicus, Homo sapiens are not endowed with a complete set of information and alternatives. When we don’t know what our options are, we have to hunt for them. Again, this is not a cost-free process. A good stop rule is necessary to balance what we know or have with our opportunity cost. Bounded rationality, Simon writes, “must incorporate a theory of search”.

Aspiration levels

Underpinning all this is what Simon refers to as an aspiration level. As we sample and evaluate more and more alternatives, our aspiration or standard for ‘good enough’ rises and falls with our observations. Right or wrong, it works like an anchor or reference point that adjusts to the individual’s circumstances.

“In marriage, you shouldn’t look for someone with good looks and character. You look for someone with low expectations.”
Charlie Munger in Clark, David. (2017). Tao of Charlie Munger.

Heroic assumptions

To be fair, mainstream economics has incorporated some of these ideas into its theories and models. We can expand our cost functions, for example, to account for the cost of search and information transfer. Game theory, likewise, can incorporate fallibility and incomplete information into its modelling.

The problem with these inroads, Simon says, is that it assumes perfect maximization, greater computation, and other “heroic assumptions” on the decision maker’s behalf — “they do not even remotely describe the processes that human beings use for making decisions in complex situations”.

“Human behavior, even rational human behavior, is not to be accounted for by a handful of invariants. It is certainly not to be accounted for by assuming perfect adaptation to the environment. Its basic mechanisms may be relatively simple, and I believe they are, but that simplicity operates in interaction with extremely complex boundary conditions imposed by the environment and by the very facts of human long-term memory and of the capacity of human beings, individually and collectively, to learn.”
Herbert Simon. (1978). Rational Decision-Making in Business Organizations.

Plumbing tools and Bayesian dials

So, what should we make of bounded rationality and the economic creed? As Mervyn King and John Kay discuss in Radical Uncertainty, we should treat the methods in economics like tools in a plumber’s toolbox. Models and their small world descriptions are useful in some contexts and useless in others. Intellectual honesty about their limitations is paramount.

King and Kay also recommend that we use our ‘Bayesian dials’. That is, to recognize our incomplete knowledge and the unquantifiable risks that characterize our world; and to update our viewpoint as useful evidence comes along. We are, after all, limited creatures, not omniscient beings, trying to make do in a complex, uncertain world.

Indeed, in Principles more than a century ago, Alfred Marshall referred to economics as a psychological science. While major strides in behavioral, complexity, and evolutionary economics have been made since then, much of the discipline remains tethered to rational reasoning and calculation. With a touch of irony, Herbert Simon tells us that:

Sources

Simon, Herbert. (1955). A Behavioral Model of Rational Choice. The Quarterly Journal of Economics.
Simon, Herbert. (1972). Decision and Organization. Chapter 8, Theories of Bounded Rationality.
Simon, Herbert. (1978). Rational Decision-Making in Business Organizations. The American Economic Review. Vol. 69, No. 4 (Sep., 1979), pp. 493-513
Clark, David. (2017). Tao of Charlie Munger: A Compilation of Quotes from Berkshire Hathaway’s Vice Chairman.
Kay, John., and King, Mervyn. (2020). Radical Uncertainty: Decision-making for an Unknowable Future.

Bounded Rationality — Herbert Simon on Decision-Making under Uncertainty and Complexity