This week I heard economic journalist Tim Harford talk at the London School of Economics (LSE), on a whirlwind tour (7 talks, I think he told us, this week) to promote his new book. Each talk is on one topic covered in the book, and at LSE he talked about the GFC and his suggestions for preventing its recurrence.
Harford’s talk itself was chatty, anecdotal, and witty. Economics is still in deep thrall to its 19th century fascination with physical machines, and this talk was no exception. The anecdotes mostly concerned Great Engineering Disasters of our time, with Harford emphasizing the risks that arise from tightly-coupling of components in systems and, ironically, frequent misguided attempts to improve their safety which only worsen it.
Anecdotal descriptions of failed engineering artefacts may have relevance to the preventing a repeat of the GFC, but Harford did not make any case that they do. He just gave examples from engineering and from financial markets, and asserted that these were examples of the same conceptual phenomena. However, as metaphors for economies machines and mechanical systems are worse than useless, since they emphasize in people’s minds, especially in the minds of regulators and participants, mechanical and stand-alone aspects of systems which are completely inappropriate here.
Economies and marketplaces are NOT like machines, with inanimate parts whose relationships are static and that move when levers are pulled, or effects which can be known or predicted when causes are instantiated, or components designed centrally to achieve some global objectives. Autonomous, intelligent components having dynamic relationships describes few machines or mechanical systems, and certainly none from the 19th century.
A better category of failure metaphors would be ecological and biological. We introduce cane toads to North Queensland to prey upon a sugar cane pest, and the cane toads, having no predators themselves, take over the country. Unintended and unforeseen consequences of actions, not arising merely because the system is complex or its parts tightly-coupled, but arise because the system comprises multiple autonomous and goal-directed actors with different beliefs, histories and motivations, and whose relationships with one another change as a result of their interactions.
Where, I wanted to shout to Harford, were the ecological metaphors? Why, I wanted to ask, does this 19th-century fascination with deterministic, centralized machines and mechanisms persist in economics, despite its obvious irrelevance and failings? Who, if not rich FT journalists with time to write books, I wanted to know, will think differently about these problems?
Finally, only economists strongly in favour of allowing market forces to operate unfettered would have used the dirigismic methods that the LSE did to allocate people to seats for this lecture. We were forced to sit in rows in our order of arrival in the auditorium. Why was this? When I asked an usher for the reason, the answer I was given made no sense: Because we expect a full hall. Why were the organizers so afraid of allowing people to exercise their own preferences as to where to sit? We don’t all have the same hearing and sight capabilities, we don’t all have the same preferences as to side of the hall, or side of the aisle, etc. We don’t all arrive in parties of the same size. We don’t all want to sit behind a tall person or near a noisy group.
The hall was not full, as it happened, so we were crammed into place in part of the hall like passive objects in a consumer choice model of voting, instead of as free, active citizens in a democracy occupying whatever position we most preferred of those still available. But even if the hall had been full, there are less-centralized and less-unfriendly methods of matching people to seats. The 20 or so LSE student ushers on hand, for instance, could been scattered about the hall to direct latecomers to empty seats, rather than lining the aisles like red-shirted troops to prevent people sitting where they wanted to.
What hope is there that our economic problems will be solved when the London School of Economics, of all places, uses central planning to sit people in public lectures?
Update: There is an interesting critical review of Harford’s latest book, here.