At this point we are still in just the third decade...

...of the age of the microprocessor. Productivity will rebound. In a few years it will "suddenly" show up in elevated percentages. But contrary to Krugman's assertion, in the long run productivity is almost nothing. Not because productivity increases won't happen; they will. But because, like the universal learning curve that brings costs plunging down, increased productivity is a rote process.

The learning curve of inverted prices was first observed by T. P. Wright, a legendary engineer who built airplanes after the First World War. Wright kept records of the numbers of hours it took to assemble each plane and calculated that the time dropped as the total number of units completed increased. The more experience assemblers had, the greater their productivity. At first this was thought to be relevant only to airplanes, but in the 1970s engineers at Texas Instruments began applying the rule to semiconductors. Since then the increase of productivity with experience is seen everywhere. According to Michael Rothschild, author of Bionomics, "Data proving learning-curve cost declines have been published for steel, soft contact lenses, life insurance policies, automobiles, jet engine maintenance, bottle caps, refrigerators, gasoline refining, room air conditioners, TV picture tubes, aluminum, optical fibers, vacuum cleaners, motorcycles, steam turbine generators, ethyl alcohol, beer, facial tissues, transistors, disposable diapers, gas ranges, float glass, long distance telephone calls, knit fabric lawn mowers, air travel, crude oil production, typesetting, factory maintenance, and hydroelectric power."

As the law of increasing productivity per experience was seen to be universal, another key observation was made: The learning didn't have to take place within one company. The experience curve could be seen across whole industries. Easy, constant communication spreads experience throughout a network, enabling everyone's production to contribute to the learning. Rather than have five companies each producing 10,000 units, network technologies allow the five to be virtually grouped so that in effect there is one company producing 50,000 units, and everyone shares the benefits of experience. Since there is a 20% drop in cost for every doubling of experience, this network effect adds up. Advances in network communications, standard protocols for the transmission of technical data, and the informal, ad hoc communities of technicians all spread this whirlwind of experience, and ensures the routine rise of productivity.

Analyst Andrew Kessler of Velocity Capital Management compares the plummeting of prices due to the universal learning curve to a low pressure front in the economy. Just as a meterological low pressure system sucks in weather from the rest of the country, the low pressure point generated by sinking prices sucks in investments and entrepreneurial zeal to create opportunities.



Archives - This site operates under a Creative Commons License.

This is a blog version of a book of mine first published in 1998. I am re-issuing it (two posts per week) unaltered on its 10th anniversary. Comments welcomed. More details here.
-- KK