How can political actors identify which putative expert is truly expert, given that any putative expert may be wrong about a given policy question; given that experts may therefore disagree with one another; and given that other members of the polity, being non-expert, can neither reliably adjudicate inter-expert disagreement nor detect when a consensus of experts is misguided? This would not be an important question if the problems dealt with by politics were usually simple ones, in the sense that the answer to them is self-evident. But to the extent that political problems are complex, expertise is required to answer them—although if such expertise exists, we are unlikely to know who has it.
If Friedman is correct, how can we have any confidence in any social realm that outcomes are in any sense good or efficient? If no one can judge whether a counterpart is expert, then how can we, say, invest in shares of corporations or lend to borrowers (buy bonds) with any confidence? How do we know, say, that corporate managers are expert at their jobs. And if no one can judge expertise, how can consumers buy anything with any sense of confidence except by experience – works for toothpaste but not for infrequent purchases such as refridgerators. And if no one can form reasonable judgments about other people’s expertise, how can we have confidence that laissez faire market outcomes are good or efficient? If the answer is competition and trial and error, well, that could work in the public sphere as well given some 60 states and territories, and 1000s of counties, cities and towns (except for nationwide public goods like national defense or the justice system).
how can we have any confidence in any social realm that outcomes are in any sense good or efficient
That is the extremely difficult problem that we might hope to tackle once we’ve solved the preceding extremely difficult problem of coming to a workable agreement of what “good” or “efficient” mean.
This is why many people favor Burkean conservatism; it contains a strong bias toward preserving systems that actually exist and work tolerably well. Change is conceived as a set of small, (relatively) analytically tractable modifications of existing practices, rather than trying to tear everything down and rebuild from first principles (which reliably fails).
One trick is have voters pay the cosr of government services as soon as possible.
Thanks Arnold, very interesting paper (and you have a valuable blog, which I found a year or so ago).
My observation from politics and working in many companies – people in general much prefer confident conclusions to a more accurate delineation of the uncertainties.
The paper describes the policy confusion around one very important branch of economics.
Even more interesting would be to compare sociology (which he clearly has a low opinion of) with behavioral genetics. In government policy (and journalism and education) the perspectives of sociology dominate. Occasionally the sociologists take a swipe at behavioral genetics in papers run by their crowd – e.g., Pulling back the curtain on heritability studies-biosocial criminology in the postgenomic era, Burt & Simons, Criminology (2014). On the other hand Nature – the highest ranking science journal – is happy to publish papers like Meta-analysis of the heritability of human traits based on fifty years of twin studies, Tinca J C Polderman et al (2015) and Top 10 Replicated Findings From Behavioral Genetics, Robert Plomin et al (2016).
If Burt & Simon are correct then behavioral genetics is at best some kind of sham, at worst a stalking horse of Nazism.
If Nature and the behavioral geneticists are correct then most of past and current sociology research & conclusions will be swept into the dustbin of history and much government policy is, at best, pointless and, at worst, completely counter-productive.