I remember being told by a mentor that the problem with my skill set was that I was a ‘generalist’. Being early in my career, I didn’t think too much about the feedback; as I really hadn’t worked long enough to build a specialty even if I knew what I wanted that to be.
But, I did find it confusing. After all, didn’t the field of economics actively poke its nose into nearly every area of human existence – whether it be the links between abortion and crime or how to start a city? And weren’t the intellectual giants of the field extremely multi-faceted? In the words of John Stuart Mill:
“There is very little chance of being a good economist if you are nothing else.”
Of course, this wasn’t the last time I’d hear the the term ‘generalist’ used in the derogatory sense: A real economist should pick a specialty, stick to it, and work hard at completing a phd on as obscure a topic as possible.
A recipe for a life well lived (albeit in a cubical).
Wicked and Kind Problems
What bothered me most about the conventional wisdom of what an economics career should look like is not that it sounded like a nightmare to me personally, but that it was in direct contradiction to what I’d seen when studying leaders in the field as part of a postgraduate course on the history of economic thought.
After all, given that the field of economics attempts to solve such a diverse range of human problems shouldn’t an economist’s career path also exhibit this diversity? Wouldn’t it make sense for economists to see how different the dreams, aspirations and day-to-day lives of people are before developing solutions that affect their lives?
As controversial as it might be for me to suggest that economists (and economics) should be exposed to a diverse range of views about the world before attempting to tinker with it, David Epstein’s book ‘Range’ suggests a more practical reason for generalists making more sense in the world of economic policy.
In essence the idea Epstein presents is that specialist training works well for kind domains – where the rules are clear, the possible outcomes are limited and you’ll receive quick feedback on whether your approach has worked. Whereas generalist experience works better for wicked domains: where the rules are unclear and may alter drastically in different contexts, outcomes are unpredictable and any feedback you receive (if at all) is incomplete, difficult to interpret and occurs with a lag.
| Kind Domains: | Wicked Domains: |
| “Patterns repeat over and over, and feedback is extremely accurate and usually very rapid. In golf or chess, a ball or piece is moved according to rules and within defined boundaries, a consequence is quickly apparent, and similar challenges occur repeatedly. Drive a golf ball, and it either goes too far or not far enough; it slices, hooks, or flies straight. The player observes what happened, attempts to correct the error, tries again, and repeats for years.“ | “In wicked domains, the rules of the game are often unclear or incomplete, there may or may not be repetitive patterns and they may not be obvious, and feedback is often delayed, inaccurate, or both. In the most devilishly wicked learning environments, experience will reinforce the exact wrong lessons.“ |
Economic Policy is Not Golf
Apologies for the spoiler, but the world is not golf. Meaning that stacking the field of economic policy (and public policy more generally) with specialists golfers is a bad idea. As by encouraging specialization in a field that routinely operates in wicked domains we are creating a scenario where poorly designed public policies are more likely to be designed and implemented – impacting all of us.
Taleb’s book the Black Swan (2007) provides another reason why having golfers manage the world might be bad idea which he calls the ‘Ludic Fallacy‘:
“Mistaking the well-posed problems of mathematics and laboratory experiments for the ecologically complex real world. Includes mistaking the randomness in casinos for that in real life.“
Nassim Nicholas Taleb – Antifragile: Things That Gain from Disorder.
The reason why I think the ‘Ludic Fallacy’ dovetails well with Epstein’s point is that if incorrect statistical models of the world are being used in wicked environments, we’ll be less likely to know until disaster strikes – particularly if everybody using the model agrees with its conception of reality. This is both because: with a less diverse field dissenting views are less likely; and as feedback available on the model’s performance is more likely to be ambiguous for wicked environments. Bannarjee provided a nice description of this issue when accepting the Nobel prize:
“Economic theorists build models, which are toy universes where they deliberately assume away much of the complexity that we experience in everyday life, in order to be able to highlight specific mechanisms that might operate in the world...
…We also make assumptions about the exact shape of the cost of effort, but equally importantly, very specific assumptions about how the cost depends for example on the nature of the work (collecting trash or sitting in an office?), the environment, physical and social, in the workplace (Is it hot? Is it friendly?), and the home environment of the workers (say whether or not they live near their parents and can therefore rely on them for childcare). One might imagine that accounting for these features can be very important before confidently generalizing evidence from the place where, say, we had the data, to other places with somewhat different circumstances. Unfortunately, more often than not, researchers estimating models choose to ignore most of these complexities (or to think of them as un-modeled sources of variation in behavior that are, rather implausibly, unrelated to everything else that is going on in the model).”
Abhijit Vinayak Banerjee (8/12/19) – Field Experiments and the Practice of Economics (Nobel Prize Lecture)
A general problem
And while whether a specific characteristic is relevant when implementing a policy might be hard to predict, there is a lot of evidence that they matter in practice. With research by Eva Vivalt finding that similar policy interventions often haven’t resulted in the same outcomes when implemented elsewhere. Meaning that even when we know what works, as soon as we try to apply it in a different context it often doesn’t.
Of course, the point of this blog post isn’t to provide a nuanced overview of the challenges of replicating policy interventions. Rather, it’s to suggest that given the problems economists are meant to solve a preference for specialization is not only misguided, but more than likely costly to society. After all, just as we shouldn’t expect a soccer team made up of golfers to be competitive; we shouldn’t think that having a sector of specialists operating in wicked domains is a good idea.
“The master-economist must possess a rare combination of gifts …. He must be mathematician, historian, statesman, philosopher—in some degree. He must understand symbols and speak in words…. He must be purposeful and disinterested in a simultaneous mood, as aloof and incorruptible as an artist, yet sometimes as near to earth as a politician.”
John Maynard Keynes (1924 essay on Alfred Marshall)
Leave a Reply