Picking nits with smart people.

Paul Krugman is understandably exasperated: Steve Williamson is saying some things about money that ... don't quite add up. Krugman does a nice job of explaining the problems with Williamson's position (and Noah Smith does as well, from a slightly different angle), so there's no point in me rehashing the whole argument. (Though if you find yourself thinking that you're just not spending enough hours in the day reading about money, you could wander through my more folksy take on the subject.)

Found here

But as the title of the post suggests, I do have one nit to pick with Krugman, based on his stemwinder of a finish:

A final thought: the notion that there must be a “fundamental” source for money’s value, although it’s a right-wing trope, bears a strong family resemblance to the Marxist labor theory of value. In each case what people are missing is that value is an emergent property, not an essence: money, and actually everything, has a market value based on the role it plays in our economy — full stop.

I'll agree that there are big holes in the labor theory of value, just as there are similar holes in the energy theory of value (something I flirted with in graduate school). But I think Krugman's position would be improved with an explicit biophysical grounding.

The labor theory and the energy theory both leave out the aspect of marginal utility: people's willingness to give up one thing to get a little bit more of something else. Say for some arbitrary reason, people's  tastes changed, and everyone went from wanting Hummers (11 mpg) to Priuses (45 mpg). The price of Hummers would collapse (no demand), the price of Priuses would rise, labor and capital (and energy) would shift from making Hummers to making Priuses, and the price of the Prius would come back down. People would be getting the same amount of value from driving as before (remember, their tastes changed, so they're just as happy in their Priuses now as they were in their Hummers before), but driving only takes 1/4 as much gasoline as before. Or looked at from the other end, we started getting four times as much value from each gallon of gas (unit of energy) as before. There was no change in technology, or in resource availability, just a change in tastes.

In other words, human preferences matter. Any theory of value that leaves that out, whether it's based on labor or energy, has a problem.

At the same time, the difficulty of procuring something matters as well. (Maybe that's incorporated in Krugman's description of a thing's market value being "based on the role it plays in our economy," but it's not clear that it is.) Look at another thought experiment, going the other direction from the Hummer-vs.-Prius one above. Imagine the economy produces only two goods: food and clothes. And the only input to production is labor, and all labor is equally skilled. We don't have enough information to say what the relative prices of food and clothes would be, because we don't know people's preferences, but we can say that, whatever those prices are, they must end up such that the wages earned by farmers are the same as the wages earned by tailors. Why? Because if farmers were making more, some tailors would shift jobs and become farmers; the supply of clothes would fall while the supply of food rose, raising the price of clothes relative to food. This would go on until farmers and tailors earned the same amount. Which would mean in turn that the value of each item was in fact directly related to the labor needed to produce it.

Now imagine that one of the tailors hits upon a new way of making clothes, and this new technique spreads among all the tailors, so now an hour of a tailor's time makes twice as many clothes as before. Have clothes become half as valuable as before? Not at all. The clothes supply will rise, the price will fall, wages will fall, labor will shift to farming, and the prices will restabilize where once again wages are the same in both sectors. The price of clothes relative to food will end up lower than it started, but we'll produce more clothes than we used to (because the remaining tailors are more productive than they used to be), and we'll also have more food than before (because we have more farmers).

It would make sense to say that the economy is creating more value--there's more clothes and more food, for the same amount of work--so the value of labor went up as well, because of the new way of making clothes.

Something very much like this has been happening since the Industrial Revolution. In the goods sector, making physical things--everything from textiles to machines and ultimately to food--workers' productivity went up. You could blandly describe this as "technological progress," but it was technology with a strong tendency in one direction: each worker was able to direct the flow of more energy. It was true in manufacturing, where handcrafts were replaced by machines where a few workers tending the machines produced far more output than their grandfathers could have; but it wouldn't have been possible without the coal to power the machines. It was true in agriculture, where mechanization and new crops meant that one farmer could produce enough food for 50 neighbors; but that too would not have been possible without the energy for the tractors, and the energy to make the fertilizer and pump the irrigation water that made the new crops so spectacularly productive.

And so the price of goods--relative to services--fell. Labor left farming and manufacturing and went into services. In my toy economy above with nothing but food and clothing and all labor equally skilled, the wage equalization was perfect; in the real world, the equalization was far from perfect, but some of it did happen.

So: labor got more productive at making goods, which caused labor to move into services and (relatively speaking) the wages of service labor to rise, independently of labor getting any more productive in the service sector.

This introduces a systematic downward bias in our estimate of how important energy is. We look at all the value being produced in the service sector, using relatively little energy. But the only reason it's able to produce so much value is that workers in the goods sector are so proficient at turning stuff out. And a fundamental part of why they're so proficient is that they can make use of abundant, inexpensive energy. So the current rate of producing value from energy in the service sector is a very poor indicator (a large overestimate) of how good we'll be at turning energy into value if energy stops being cheap. (That's why I would differ with Jim Hamilton's take--while I'm picking nits with people far more experienced than I.)