Trapped in Bad Choices
Why do companies, managers, and entire industries often end up on paths that, in hindsight, clearly turn out to be wrong? Why do we have to stick to inefficient technologies, inferior standards, or strategies that look sensible only until it is too late to change?
The classic economic answer lies in two words: path dependence. Once you've taken a course, turning back becomes expensive, complicated, and sometimes impossible. Early decisions become self-reinforcing, alternatives disappear, and the system remains “stuck”, even when better solutions emerge.
Saeid Kazemi (Bocconi PhD and fellow at ION Management Science Lab – SDA Bocconi), Arnaldo Camuffo (Department of Management and Technology, Bocconi and ION Management Science Lab – SDA Bocconi School of Management), Alfonso Gambardella (Department of Management and Technology and ION Management Science Lab — SDA Bocconi) believe that this explanation is incomplete, as they explain in a recent article in Industrial and Corporate Change. It is not only external events or “historical coincidences” that determine long-term trajectories. What happens before the decision is made is also very important: how decision-makers interpret the situation, what theories they use to read it, and above all, what idea they have about the future.
Knowing that you risk lock-in is not enough
Making decision-makers aware of the mechanisms of path dependence (i.e., the fact that certain choices can become irreversible) is a decisive step toward improving the quality of decisions. But under what conditions does this greater awareness really work?
The authors answer this question with a controlled experiment that draws on a famous case: the birth of the QWERTY keyboard. Participants, all managers, were asked to put themselves in the shoes of inventor Christopher Latham Sholes and choose between three alternatives. The first solves the immediate problem: almost no jamming, but very slow typing speed. The second is a compromise solution: it works reasonably well, without excelling. The third is risky: today it jams easily, but it allows you to type much faster and could become superior if the technology improves. Before choosing, however, participants receive different information. Some receive no additional warning, while others are explicitly warned that once a keyboard becomes widespread, changing standards becomes difficult and expensive. Still others receive, in addition to this warning, a message about the future: in some cases pessimistic (“things could get worse “), in others optimistic (“there are good reasons to believe that technical problems will be solved”).
Well, most participants still plumped for the safest solution in the short term. Only when awareness of the lock-in risk was accompanied by a positive view of the future did a significant proportion of decision-makers agree to take the initial risk by choosing the best option in the long term.
Optimism makes the difference
The situation therefore only changes when a second element is added to awareness: a positive vision of the future. Decision-makers begin to choose the best option in the long term only when they know they are facing a possible lock-in and, at the same time, believe that that path can be improved thanks to favorable future developments.
Awareness of path dependence makes the risks very visible: if you make a mistake now, you may not be able to correct it. Without a positive outlook, this awareness pushes you to take refuge in the safest solution today, even if it may prove mediocre tomorrow. Optimism, on the other hand, rebalances the decision: it makes the initial sacrifices more bearable because it gives meaning to the direction taken over time.
Fear of making mistakes today outweighs benefits of tomorrow
To explain these results, the study refers to Prospect Theory, one of the most established theories in economic psychology. People tend to attach much more importance to immediate losses than to future gains. In the case of the keyboard, the risk of jamming today counts more than the promise of greater efficiency tomorrow.
Awareness of path dependence alone does not mitigate this mechanism. On the contrary, it can reinforce it: it makes the cost of a wrong choice even more apparent. As the authors observe, “only path dependence awareness combined with a concrete, positive expectation can partially offset loss aversion.”
Simulations with agents based on artificial intelligence models confirm this dynamic. When agents “read” the problem carefully, they perfectly understand the meaning of the instructions. But when the time horizon shortens or immediate losses become more apparent, they too end up favoring short-term security.
Therefore, it is not enough to explain to managers that path dependence exists, nor to warn them about the risks of lock-in. Without a credible and positive vision of what can happen along an alternative path, awareness remains sterile or even counterproductive.
To avoid getting trapped in suboptimal choices, two things are needed: a sharp mind to understand that some decisions are hard to reverse, and the ability to imagine a future in which it is worth taking a risk today. Without this second ingredient, rationality remains stuck in the present.
