"If economists wished to study the horse, they wouldn't go and look at horses. They'd sit in their studies and say to themselves, 'What would I do if I were a horse?" Two things I've seen recently remind me of that line of Ely Devons.

When I approvingly tweeted Stephen Marglin's account of the rise of the factory system, Pseudoerasmus referred me to Greg Clark's paper (pdf). He argues that factory discipline emerged because when workers had freedom to choose their hours they short-sightedly worked too little and earned too little:

Whatever the workers themselves thought, they effectively hired the capitalists to discipline and coerce them. Even in the factories of the Industrial Revolution they were the ultimate masters of their fate, but weakness of the will meant they delegated that mastery to the capitalists... Though factory discipline was coercive, forcing the worker to do what he or she would otherwise not have done, the worker was in no sense exploited by the introduction of discipline. The workers voluntarily entered into the temporary servitude of the factory...Had they been able to exercise more self-control, factory discipline could have been avoided for most technologies in the 19^th century.

There is, though, a massive hole here. Clark gives us no direct empirical evidence that this actually happens. He points to no real worker who freely chose factory discipline as a commitment device; to no evidence as to how workers actually chose among different capitalists; and to no worker who was with hindsight grateful to capitalists for saving him from himself. The voice of actual workers is wholly absent*. Clark is not studying horses, but asking: what would I do if I were a horse?

He is not atypical. This is an example of something common in economics - "as if" modelling. Clark describes the rise of factories as if workers chose them as a commitment device, just as (say) Murphy and Becker describe smackheads as if they were rational maximizers. But this is insufficient. You can describe a woman's black eye as if she had walked into a door. But if in fact her husband had beaten her, you are missing the truth and overlooking genuine oppression and injustice.

My second example was an exchange on Twitter about Cobb-Douglas production functions. Nobody saw fit to point out that if you want to know how useful they are, you should look at how actual firms produce actual stuff: do Cobb-Douglas functions describe the real world or not? Again, nobody's looking at the horses.

What happens when they do? One nice example is a study (pdf) of a steel mill by Igal Hendel and Yossi Spiegel. They showed that it doubled production over 12 years with the same plant and workforce. Every time the mill seemed to be at "full capacity", its managers found ways of tweaking production methods to eke out more output. "Capacity is not well defined" they conclude. If this is true of an old economy steel mill, how much more true might it be of intangibles-intensive firms which are more scalable?

To the extent that this is the case, it has an important implication. It means that growth in demand and a zero output gap will lead not to capacity constraints and rising inflation, but to productivity improvements. Hendel and Spiegel give us micro-empirical evidence for a claim by Joan Robinson:

When entrepreneurs find themselves in a situation where potential markets are expanding but labour hard to find, they have every motive to increase productivity.

Recent data is consistent with this. The UK and US have both recently seen unemployment fall to multi-year lows. And both saw productivity jump in Q3 after years of stagnation.

If all this is the case, then Unlearning Economics is right: long-run growth can indeed be influenced by aggregate demand. Trend and cycle might not be so cleanly distinguished. A zero output gap is to be welcomed as a spur to productivity growth, not feared as a portent of inflation.

Now, I stress the "ifs" here. I'm not saying the evidence for this is overwhelming. All I'm doing in endorsing Sarah O'Connor's point - that the best economists are those with dirty shoes.

* You might reply that we have few first-hand accounts of the lives of workers: history is written by the winners, or at least the literate. We must not, however, let our thinking be distorted by the availability heuristic.