money doesn't measure

From: Allin Cottrell (
Date: Mon Jul 19 2004 - 20:24:00 EDT

I'm going to have another try at this, from first principles.  This is
partly a reply to Rakesh, but it's also a more general reflection on
the long-running "measurement of abstract labor" thread.

My thesis is that money as such does not "measure" anything.

Money (e.g. the dollar) is a unit of account, a "standard of value" if
you like.  In that respect it bears some analogy to a "standard of
weight", such as the standard kilogram.

Note that the standard kilogram is not, of itself, capable of
measuring the weight of anything: for that you need a balance or some
such apparatus.  Suppose I give you some item x, and the standard
kilogram, and ask you to use the kilo to measure the weight of x --
but you're not allowed a balance or anything of the sort.  It can't be

If you're allowed a balance, you're in business.  Suppose x is some
standardized item, weighing substantially less than a kilo.  You then
see how many x's you can put in the scales against the standard kilo
before tipping it, and you can bracket the weight of x quite

Now my point is that in the case of money the "balance" (that actually
does the "measuring") is a much more complex apparatus, involving
production, supply-and-demand, and profit rates.  Unlike the physical
balance it is not a simple bilateral mechanism (the standard kilo
versus x, the monetary standard versus an individual commodity -- no,
it doesn't work, the analogy breaks down).

And with the social "valuation apparatus", it turns out that it
doesn't matter very much whether or not the standard (e.g. the dollar)
remains constant over time.  The system is in trouble if the standard
fluctuates wildly over a short time horizon, but it is remarkably
robust in face of substantial ongoing change, at a non-constant but
boundedly-varying rate.

The condition for this is that economic agents are primarily
interested in relative prices, and they're "smart enough" to perceive
relative prices even when the absolute price level is nowhere near
constant.  Of course, the "smart enough" here is not really an
individualistic thing: it's a matter of the evolution of social
institutions such as the calculation and timely publication of
economy-wide price and wage indices.


This archive was generated by hypermail 2.1.5 : Wed Jul 28 2004 - 00:00:01 EDT