Fixed link

Why Most Things Fail

Paul Ormerod

posted on 03 May 2006

reviewed by Joseph L. McCauley

Paul Ormerod’s third book rightly emphasizes that failure is more common than success. Species disappear, firms fail, nothing persists forever. This is a nice and useful contrast with textbook economics, where nothing is allowed to change. In an economics classroom, nothing can fail but the student.

Of course, many enterprises fail for lack of financing. Financial markets dominate all other markets worldwide so it’s quite difficult to discuss markets without an adequate treatment of finance as basis. E.g., on pg. 173 fat tails are named as the cause of the collapse of LTCM but this explanation is completely false: LTCM collapsed because finance market liquidity dried up. Fat tails describe large returns in normal liquid markets. Details do matter if we’re to understand why some things fail and others don’t. With adequate finance market liquidity LTCM would still be in business. They begged for a quick fix, a loan, but Greenspan wouldn’t ok it. Neither would Warren Buffett (see Dunbar’s “Inventing Money”).

Having said that, let me proceed to my own naive observations why some things fail and others don’t. Financing being equal or at least adequate, technological competition is the most important factor. Consider the 1954 Cadillac and Packard models (my parents owned the latter, leather upholstery and all).

The Packard had a flathead straight eight motor, old technology. The Cadillac had the most modern technology, an OHV V-8. Far more hp/liter. Packard was bought a few years later by Studebaker which suffered from an inadequate dealer-service network and both collapsed together, but not before the Packard got the Studebaker OHV V-8. In 1954 when the Dollar was pegged to gold at $35/oz. the Packard cost about $3500. In 1974, two years after Nixon deregulated the Dollar, a VW 411 cost $3000 (the price of the Beetle had doubled in the US from $800 to $1600) and the price of gasoline doubled soon thereafter. We live in a credit economy, the enormous consumption that drives worldwide production would be impossible without credit, but economists tend to regard financial markets as ‘just another market’, not the main market.

Then there’s the outboard motor market. Fifty-five years ago there were over ten outboard motor manufacturers in the US, now there are (thank goodness, still) two. Consider the Cross Radial, ca. 1930.

The air cooled radial was a good idea for airplanes but was too heavy and bulky for boats, where it was easy to suck water out of the lake to cool the motor and then spit it back in. No radiator needed, all the cool water a motor needs is already there. The water cooled radial was too bulky and didn’t compete well with the opposed two cylinder OMC outboards that dominated racing until the fifties.

Then there was Scott-McCulloch, still manufactured in the sixties.

They produced a beautiful motor with a good engineering design before 1960, three cylinders, but couldn’t keep up with the ‘big two’ in the horsepower race. They were quite active in racing (the testing ground for new technology, as well as an advertising benefit) but lacked the engineers to build a motor fast enough to compete with the OMC V-4

or the Mercury straight 6.

Fifteen years later OMC (Evinrude-Johnson) developed the loop-charged in-line three into a fine racing motor that still competes. I drove one from 1977-1985 and held the national speed record until the class became defunct.

My propeller technology, due to having had a free hand in the UH Physics Department machine shop and also in a local propeller repair shop, was far better than that of my competition. I even sold propellers to competitors, but never my best ones. It was seat of the pants engineering guided by the general theory of airfoils/hydrofoils. I was later bared from both shops, but by then I’d taken up long distance hiking.

In the sixties Saab produced good cars with the same two stroke three cylinder design. They were popular in the northern part of the US. So details matter, there are no general laws, there is no ‘iron law of failure’. There are largely details of competition without any general rules for guidance except that: (a) you have to have adequate financing to ante up and play, and (b) hire the best engineers you can get. If you don’t stay ahead of the competition, you’ll lose. And you can never quit because they’ll copy your results as fast as they can when you beat them. I ask the reader to forgive me for choosing markets that I understand (finance, motors, boats): I was a mechanic even as teenager long before I became a physicist. It was, in fact, mechanical explanations that interested me in physics in the first place. Merton was a stock car mechanic, and he did good theoretical work. He was just too uncritical of the neo-classical idea of stable equilibrium, an far less trustworthy tool than a good drop forged stainless steel wrench.

Why have I emphasized interesting examples from the real world instead of trying to present an overall theory? First, I don’t have such a theory, nor does anyone else. Second, global explanations are suspect. We know from nonlinear dynamics that we have to learn to live with local explanations. Local explanations of biological evolution are clear: evolution takes place mutation by mutation at the DNA level, but we don’t have and will never have a geometric picture or falsifiable mathematical model of how a fish evolved into a bird.

I still like Ormerod’s first book the best of the three. I don’t like the advice given lightly in his last two books that government should keep hands off because we can’t predict the future. We can apply the same reasoning to global business and say that we shouldn’t let international organizations (World Bank, IMU, WTO) and corporations set free market rules that generate surprises that upset local societies. In particular, how should western countries continue to buy products when unemployment eventually reaches high levels due to loss of industry?

Soros has it right, markets are unstable: the only way to govern effectively is to remain sceptical and be prepared to change rules/behavior/expectations when you find out that you’re wrong. Rigidity caused the USSR to collapse, lack of rigidity allows China, although still a dictatorship, to compete successfully in world markets. Rigidity in the West in the form a belief in the neo-classical prediction that deregulation provides the best of all possible worlds, the Milton Friedman lesson that ‘if they can make it cheaper then let them produce it”, is a liquidity illusion: no one is inventive enough to generate new, womanpower-intensive industry to replace what’s been lost. This is the problem that economists had better face, the one written about by Jane Jacobs. Asia is implicitly taking her advice, western countries are ignoring it in favor of ill-advised reliance on the Pareto Optimum, an assumption of nonexistent liquidity based on pure barter reasoning! On the subject of industrial competition between east and west, one would do better to read Spengler and Jacobs rather than the economics texts. Jacobs even recommends local currency, the opposite of the reasoning that lead to the Euro. We know for sure to one can predict the future reliably, but Soros made money by guessing the future most probable value of financial instruments, and Spengler correctly anticipated and even explained the reason for the economic disadvantage of the west against the east today.