Friday, May 28, 2010

Prosperity based on commodities

An earlier post looked at economic prosperity and standard of living from the point of view of a grain-based agricultural economy. There I singled out intensive, extensive, and technology-based growth, and the effects these scenarios had on the standard of living for a farming population. This is a particularly simple case, since it equates standard of living with food availability per capita. (This is enough, however, to arrive at credible estimates of the standard of living over long stretches of Chinese history, as Bozhong Li has demonstrated in Agricultural Development in Jiangnan, 1620-1850.)  This simplification leaves out markets, prices, and trade; so it doesn't shed much light on economies based substantially on the production of commodities (including farm products, but also including manufactured goods). So how does the situation change when we postulate production for exchange and consumption based on cash income?

Let's once again consider an isolated region, where all products consumed are produced in the region. So there is no interregional trade. And let's suppose there are three goods: grain, shirts, and beer. Every household needs some of each, and households acquire income through ownership of resources: land, capital, and labor power. The income available to a household is the net return it achieves through use of its resources. Goods are produced by "firms" and are bought and sold through competitive markets.

Now to estimate a household's standard of living we need to do a more complex estimation: we need to estimate the household's income and we need to estimate the "purchasing power" of this income in terms of the baskets of goods this income can purchase at prevailing prices. So we need an income model and a price model for the three goods.  (See Robert Allen's detailed efforts at answering these questions across Eurasia (linklink).)

Production requires access to resources.  Each resource can be used in two basic ways: it can be used directly by its owner in production, or it can be "rented" to a firm for use by the firm in production. So there is also a competitive market for resources: rent for land, interest for capital, and wages for labor. And at any given time there is a specific distribution of resources across population; some households have dramatically more of each resource than others.

We can begin our thought experiment by taking as fixed the techniques of production that exist for the three basic commodities. In order to produce at a given level of output, the firm needs access to a known quantity of resources, in a specific proportion. Firms amd households with lots of resources can begin producing shirts, beer, and grain immediately. Poor firms and households will either rent access to more resources through promise of future rents; or they will rent out the resources they currently possess, including labor time. So landless, propertyless households have no choice but to sell their labor time; they become workers. So now let's picture our region as populated by firms and households producing commodities, and all persons functioning as consumers purchasing a bundle of commodities for life needs.

So far we've provided a scene very familiar from the classical political economists and Marx. Much of subsequent economic thought went into solving various parts of this story: what determines prices, what does the distribution of income look like, and how do innovation and organizational and technological change fit into this story?  What does an equilibrium of production, consumption, and price look like with static technology?  What are the dynamic processes of adjustment that occur when there is a substantial change in the process of production?

My question here is a limited one: what needs to occur in this scenario in order for there to be a rising trend in the average and median standard living for this society?

Let's define the standard of living as the size of the wage basket available to the median consumer: the sets of baskets of grain, shirts, and beer that the median income earner is able to purchase. In order for the standard of living to rise in this isolated region, there needs to be an overall increase in the efficiency and productivity of the production process for the three goods. And the money wage of the median consumer needs to rise.  (Amartya Sen provides quite a bit of analysis of the meaning of the standard of living in The Standard of Living.)

Let's refer to the concrete production process at a given time as the current practice; this is the specific way that inputs are organized in order to create the output. As we saw in the graph of output against time borrowed from Mark Elvin (link), we can think of progress here in two ways. First, there is refinement of practice, as producers gradually recognize small modifications that permit removal of costs from the process. And, as Marx and Smith agree, firms and households producing goods for a market have a powerful incentive to seek out these improvements: they can continue to sell their products at the old price until the rest of the producers catch up.

Second, producers can introduce substantial, revolutionary changes in technology. They may replace skilled sewing-machine operators with sewing robots that reduce each of the inputs into the good. Productivity takes a big stride forward.

There is a third mechanism of cost reduction available: the firm/household may speed up the labor process, lengthen the working day, or lower the wage. Volume I of Capital goes into detail on each of these mechanisms within a market-governed firm.  And each of these approaches is negative for the quality of life of the working class.

Now let's get back to the question of the standard of living. Does the process of competition, rising productivity, and falling prices imply an improvement in the standard of living? Or, conceivably, does it lead to a paradoxical immiseration of the bulk of the population? Both outcomes are possible. Rising efficiency and productivity have permitted our little society to produce a rising quantity of beer, grain, and shirts. And this on the basis of a fixed level of basic resources. So in principle everyone may be better off. But it is possible as well that the benefits of rising productivity have been disproportionately captured by a small advantaged group. So income may have become increasingly concentrated at the top. The average wage basket will have increased. But the median consumer may have declined through that process of concentration.

What this story tells us is something fairly simple: the effects of productivity improvement within a commodity economy depend critically on the prior distribution of assets and the institutions through which income and the gains of efficiency are distributed. And this in turn suggests a point much like that of Robert Brenner: the social-property relations embedded within an economy are critical in determining the fate of the median person, and they are subject to profound political struggle (post).

It would be very interesting to use agent-based modeling software to represent a series of scenarios based on this description of a commodity-based economy undergoing growth.  What do distributive outcomes look like when the prior distribution is relatively equal?  How about when they are substantially unequal?  How much difference does the timing of growth make on the eventual distributive and welfare characteristics of the scenario?

(Piero Sraffa's Production of Commodities by Means of Commodities : Prelude to a Critique of Economic Theory picks up some parts of this story in a neo-Ricardian way; Marxian economists have looked at Sraffa's work as also providing a novel basis for the labor theory of value.  The framework provided here also leads into an argument for a new definition of exploitation by John Roemer in A General Theory of Exploitation and Class.)

1 comment:

russell1200 said...

A reduction in population (rather than an increase in productivity) would could to an increase in living standards. Particularly if the society was at a population level were diminishing marginal returns to labor had set in some time ago.

The English at the time of the industrial revolution already had high wages, and that lead to their developing steam pumps and then steam engines:

Or at least that seems to be what
Robert C. Allen "The British Industrial Revolution in Global Perspective" is saying. [I saw him give a talk, but I have not read the book yet].

What I don't know is why they had high wages. The time period of the industrial revolution, I believe was a time period of expanding populations. But it is possible that just enough of the skilled labor went off to the new worlds, or possibly there was competition for skilled labor from world trade.

Of course in your closed economy, there is no trade. But someone could open a pool hall if you had adequate surplus resources. At which point the price for your basic commodities would drop as people choose to play pool rather than over indulge in commodities. And of course gambling (as pointed out by Aaron Brown in "The Poker Face of Wall Street") is one of the few methods for normal people to accumulate capital in pre-modern finance days.

Somewhere within this muddled mess of a comment, I guess I am trying to get at/ and or figure out why, there seems to be at least a correlation between the wages of the common worker, and advances in labor productivity on one hand; but that the high wages do not appear to be driven (at least initially) by the higher productivity.