By Daniel M. Ryan
There's an old economic fallacy pegged by Henry Hazlitt in Economics In One Lesson as "the blessings of destruction." It claims that destroying older capital goods conveys a net benefit to the economy, because the older ones must be replaced by new ones that are more efficient. This extra efficiency, so the claim goes, would more than make up for the loss of the old capital goods.
If there's any word that pegs this fallacy correctly, it would be "unctuous." It's unveiled when natural disasters strike; in this capacity, it can serve as a salve to the victims by assuring them that their lives will get better despite the disaster's damage. It could be pegged as bonhomie used by an economist who's trying to put on the house-call hat and deploy a good bedside manner. Few people (outside of those economists who aren't taken in by it) would challenge it right after a disaster such as Hurricane Ike. Most people would take it in stride, as an exaggerated way of saying "things aren't all that bad; you can salvage something from the mess."
A fallacy it is, though, and here's why: it confuses technological efficiency with economic efficiency. To use a homely example, an old stove is technologically less efficient than a new stove. Almost certainly, it uses more energy than a new stove of comparable power would. Unfortunately, the new stove costs a lot more than the old stove would yield from a sale: this makes for a loss that has to be recouped through future energy saving. If the net present value of the energy savings isn't as much as the net cost of the new stove, net of the proceeds from the sale of the old one, then the new stove isn't economically more efficient. Technologically, it is. In terms of energy use, it is. In terms of consumer satisfaction, it may be. In monetary-value terms, it isn't. The "Blessings of Destruction" fallacy, a more complex derivative of the old broken-window fallacy, confuses different kinds of efficiency. It rests on a subtle equivocation.
It can, of course, be used as a palliative to victims – but it becomes less defensible when used to justify deliberate remaindering. In particular, there's one kind of "remaindering" that makes it somewhat less than benign when used as a deliberate policy tool. Employees, or former employees, in troubled financial-sector firms could identify the human costs of such a policy.
Capital goods don't run themselves. It takes workers, most notably skilled workers, to use them to produce either a finished or an intermediate product. Unlike consumer goods, where ease of use is part of the package, capital goods are not designed to be user-friendly. That's why it takes learned skills to use so many of them.
These skills tend to be specific to the capital good used. Anyone who's ever worked in a factory, or observed work in one, has seen so. A worker gets used to his/her tools as well as the work. If another set of tools are substituted for the ones that (s)he's used to, the performance level will go down unless the new ones are almost a perfect match for the old. Brand loyalty amongst skilled workers is almost legendary.
As consumers, we're used to new and improved consumer products that are used in the same way as the old unless they're more convenient. The few exceptions are so glaring, we joke about them. (The VCR was a notorious example of a new but relatively user-unfriendly consumer good; so, I suggest, is the home theatre as of now.) With capital goods, though, ease of use is usually a secondary feature. The purpose of capital goods is to get products made and on the shelves, not to make life more enjoyable through their use.
Thus, there's little incentive for newer and more productive capital goods to be designed for greater ease of use. In fact, they're typically not: another popular fallacy, the Luddite fallacy, is indirect evidence for this point. If switching from old and relatively inefficient tools to new and relatively efficient ones wasn't that big a deal, then why all those protests against automation and new industries? The typical worker may not know that much economics, but (s)he does know how to induce from experience.
Presumptively, there is a hidden cost to ripping out old economically viable machines and replacing them with new ones. That cost is added skill obsolescence – meaning, "pushing more people out of jobs."
The free market, dynamic creature that it is, causes enough skill obsolescence to make it controversial. Why would forcing technological innovation beyond the free market's natural rate make said "remaindering" anything but worse?
Of course, the governmentally-inclined person has an answer to this question: a government-encouraged retraining program will make things better. Programs such as these can be seen as mitigation for forced skill obsolescence, and as worthy social programs in themselves. (I note that the first attribute isn't mentioned all that much.) As a program to help those marooned by free-market obsolescence, it fits in with the modern conception of the welfare state. As a mitigation program, though, it does beg the question: instead of trying to fix what was already broken, wouldn't it be better not to have broken it at all?
By a happy coincidence, Professor Paul David found out in 1989 that free market making replacements by economic-obsolescence measures has the effect of smoothing out the boom-town effect of technological innovations. (This study is cited in Alan Greenspan's The Age of Turbulence, pp. 474-6 hc.) Letting individual actors in the marketplace work innovations in to the economy has a relatively countercyclical effect when compared with governmental forcings. When you consider the amount of government effort and tax treasure put into smoothing out the business cycle, then where's the sense in promoting a fallacious measure that exacerbates it?