Commodity Relations, part 2

Commodity relations describe the economic activity of individuals and small* groups:

1. Making stuff
2. Exchanging that stuff for other stuff
3. Using the stuff we've exchanged

It therefore stands to reason that we're interested in three properties of the stuff we make, exchange and use: its cost, price (or exchange value) and use-value.

*Even a huge corporation is relatively small compared to the overall population.

The cost of a commodity is how much it takes to create the commodity. (How much of what? Good question.) The use-value is how much subjective enjoyment a person receives from consuming the commodity, or from consuming what the commodity produces.

(Note too that we use "commodity" to represent both an individual thing, such as a particular pair of shoes, as well as a type of thing, such as pairs of shoes in general.)

Since an essential feature of commodity relations is about the exchange of commodities, we want to focus on how commodities are exchanged, i.e. how we socially determine the price of a commodity.

When we exchange commodities directly for other commodities, the price is determined by what is exchanged. If I give you one hat and you give me two pairs of shoes, the price of two pairs of shoes for me right now is one hat; the price of one hat for you right now is two pairs of shoes. Money (especially fiat money like 2009 dollars instead of commodity money like gold) just adds a level of indirection: If I give you one computer for $100, and go to the grocery store and spend $100 on a bag of groceries, then the price of a bag of groceries for me right now is one computer.

There are a lot of people in a society, all exchanging stuff at various prices. We therefore want to talk about various statistical properties of all of these exchanges. Also, we want to talk about the dynamics of these properties, how the individual and statistical properties change over time.

When we look at theoretical models of how we expect human beings to conduct commodity relations, we draw a possibly surprising conclusion:

Statistically, the price of a commodity is a function of its cost, not its value

It's extremely important to understand that we're talking about price, not use-value. Economists' habit of using exchange-value to talk about price (and using price to talk about the exchange value in terms of money) has led to many misconceptions, from Santayana to Heinlein. (Read Heinlein's Starship Troopers for a hilarious, but egregious and indefensible misunderstanding of economic theories of price.) If it costs the same to make a six-foot high pile of toxic waste and an apple pie, that does not mean that a pile of toxic waste is worth an apple pie. It just means that if people were to exchange piles of toxic waste for apple pies, they would exchange one apple pie for one six-foot high pile of toxic waste. Marx did not invent this relationship, and in Das Kapital he explicitly disambiguates exchange-value and use-value, but he still bears the brunt of uninformed criticism of this relationship.

This law of commodity relations is statistical, not individual. Just because it costs Alice twice as much as Bob to each make an identical pair of shoes does not mean that Alice's shoes "deserve" or will command twice the price of Bob's. Also, the this law of commodity relations is dynamic: it could be phrased more precisely by saying that over time a statistical property of the price of a commodity approaches a statistical property of its cost.

All microeconomic models predict a strong correlation between price and cost, even the most conservative "freshwater" capitalist economic models, i.e. Marginalism.

If we use Marginalism to compare patterns of subjective use-value, we find a society will produce two different commodities with different patterns of use-value in the proportion necessary to exchange those commodities on the basis of equivalent cost. If we use Marginalism to compare patterns of production cost vs. use-value, we determine the statistical property of the cost that corresponds to the price.

We can empirically measure price and cost independently, and verify or falsify the theoretical models by looking at the correlation between price and cost in a number of commodities. There are other factors than cost affecting price, so the correlation won't be perfect, but if the law has any value at all under present circumstances, we should see at least a strong correlation.

There are many other arguments in favor of the relationship between price and cost. For example, a computer has (at least for me) more use-value than a car: I have eight computers (some not working) and one car; I spend more time using my computer than I do driving my car; and if I had to choose between a car and a computer I wouldn't hesitate to keep the computer and ride the bus. Yet a car costs two orders of magnitude more than a computer, because it costs more to build a car than a computer.

I could go on but I hope I have you convinced:

Under commodity relations, people exchange commodities on the basis of equivalent cost.