‘Should financial regulation intervene in the portfolio choice of investors?’Why is the gender of the professor chosen to represent the London Business School female? Is it coincidence, or has it been selected to make some ideological point? If the latter, what point? That the School has its share of female professors? Or are readers supposed to regard it as more natural to have an interventionist position expressed (or at least considered) by a woman? Herbert Spencer thought that women were more likely than men to adopt interventionist positions, and one can certainly posit evolutionary models which would fit with this.
The LBS could have asked me to comment on whether the position invoked by the advert is philosophically defensible, or logically flawed, but they did not, perhaps in part because they guessed which answer I would give.
Modern economics is rather keen on the concept of ‘bad choices’, and much has been made in recent years of one version of this, the so-called phenomenon of ‘cognitive bias’. It is supposed to be possible to demonstrate, experimentally, that people make choices that are not optimal. In fact, no such demonstration is possible.
This is not to say that the concept is nonsensical to begin with. I dare say the average human being has a psychology that is full of unresolved conflicts, which trip him up in a way that prevents him getting what he really wants. But demonstrating that this is the case, by comparing his actual choices to supposedly superior ones, is another matter altogether. From a strict point of view, it simply cannot be done.
Consider person X making a choice between option A and option B. The choice is made in favour of A, through some action on his part, and consequences start to follow. How are you, an outside observer, going to assess that it was B which was the optimal choice from his own point of view? Because A causes him harm of a kind that B does not? But how can you assess what weights X places on different benefits and detriments, or whether indeed a detriment really is a detriment, from his point of view? He may choose to smoke, for example, precisely because he does not wish to live to the maximum possible age.
What if things become extreme enough, you may argue. If A causes him to lose his wife, and his job, surely he cannot have wanted to choose A? Of course, we have to distinguish between consequences he foresaw and those he did not. Your assessment of the probabilities of different possible consequences may be different from his. ‘Oh, but the way things turned out proves that my probability assessments were correct.’ Well, no. Also, even if he himself thought that the hoped-for outcome of his decision had a low probability, the value he assigned to that outcome may have been so high that the choice was still rational from his point of view.
What if X now seems miserable in a way that is hard to envisage he would have felt if he had chosen B? Even if it were possible to make comparisons between his current emotional state and the one he might have derived from the other option, and possible to rank one above the other on his behalf (it is not), it still would not prove that he should not have taken the gamble.
What if X claims he would not have made the decision if he had known something which he did not know at the time, but which he now does, i.e. if he had had access to information P? And what if a lot of other people say the same thing? (For example: ‘now we have read the leaflet about lung cancer, we think we would never have become smokers if we had been able to read it when we started smoking’.) Would that mean that if awareness of P were increased, decisions would necessarily be made that were ‘better’ for those kinds of individuals? No.
What if additional knowledge does not come into it, and the individuals just say ‘I was foolish’, or ‘I did not think it through’, or ‘I was under the influence of alcohol’, and claim that in a more normal, sensible state they would have chosen B, not A. Can we not at least then conclude they made a ‘bad’ decision? Not really, for the same sorts of reason.
One cannot reach any strong conclusions from what someone says, especially after the event, because of the principle of ‘cheap talk’. If what you say has little or no effect on what happens to you, then arguably nothing can be read into your statements except (possibly) your intention to influence the listener. Only actual choices can reveal something about a person’s desires or interests, and even those only in an imperfect way.
You (a) do not know whether a person’s claims about what they would have done are meaningful, and (b) do not know what their optimal choices, given all available information, would have been. What is more (allowing for the sake of argument the possibility that there is a better choice, and that others know what it is), you can certainly not assume that intervening to produce the outcome of the optimal choice has the same ranking, from the person’s point of view, as the case in which he chooses the outcome for himself.
Such considerations, showing that (strictly) you can never judge that another’s choices are imperfect, can never know what the optimal choice would have been, and can never recreate it artificially even if you knew it, are routinely waved aside.
We can guess the reason why the possible objections are ignored. The assumptions that we can know what is good for people, and that we can bring it about, may seem to provide legitimacy for intervention. It can be claimed that the intervention is done, not for the intervenor’s benefit, but for the sake of the victim.
Conversely, consideration of the philosophical flaws underlying theories of irrationality tends to undermine the arguments in favour of intervening, whether in portfolio choice or any other area of decision-making.
Theories which provide justification for the exercise of power should always be regarded with greater scepticism than those which are neutral with respect to power, and certainly not with less.