Fred Moseley (fmoseley@laneta.apc.org)
Tue, 1 Oct 1996 13:03:21 -0700 (PDT)

[ show plain text ]

Thanks very much to Andrew for sending me his earlier post (3063) to which
Duncan was responding in (3074), which started the recent round of
Jerry for sending me an extra copy.)

After studying Andrew's example (which he recently reposted in (3176)), it
seems to me that Andrew is correct and the inference I drew from Duncan's
critique of Andrew is wrong. I also think that Duncan's critique is wrong -
more on that below.

The inference I drew from Duncan's critique is that the TSS interpretation
implies that, as one moves from period to period, value added changes as a
result of a change in the price of material inputs without a change in
living labor. I now see that value added in Andrew's example is assumed to
be = 100 in every period, determined by an unchanging 100 units of labor.
This amount of value added is then taken as given in the determination in
the price of the output of each period.

Alan is also correct (in (3189)) that a possible source of confusion is the
distinction between the TOTAL value added for the total output and the PER
UNIT value added. Andrew's example does not imply that the TOTAL value
added changes without a change in the total living labor. The per unit
value added does indeed decline from period to period, but this is not due
to the effect of a change in the price of raw materials without a change in
living labor, but is instead due to the 25 0ncrease in productivity that
Andrew assumes from one period to the next. The per unit value added
declines instead because a given total value added, as determined by the
amount of living labor, is spread out over a greater quantity of output. In
Andrew's example, output increases from 5 in period 0 to 6.25 in period 1,
so the given amount of value added (=100) is spread out over 6.25 rather 5
units of output, so that the per unit value added is 16 rather than 20.

Duncan presents the following equation for the per unit value added:
va = p(t) - ap(t-1)
Duncan's interpretation of this equation makes it appear as if this is the
way that value added is determined, as a residual, with both p(t) and p(t-1)
determined prior to value added. Assuming this, Duncan then subdivides
value added into two components: [p(t) - ap(t)] and a[p(t) - p(t-1)], the
first of which is the standard definition of value added in terms of current
costs and the second of which is the inventory valuation adjustment (IVA).
This equation and decomposition makes it appear as if the IVA affects the
magnitude of the value added. This is the way I interpreted Duncan's post
and hence interpreted Andrew's example before I saw Andrew's actual post.

But this interpretation is mistaken. In Andrew's example, the total value
added is determined prior to the determination of p(t) and then used to
determine p(t). The total value added is assumed to be equal to 100, which
is determined by the living labor and is independent of the IVA. The IVA
can be determined only after p(t) is determined.

Duncan's equation for value added (above) and its division into two
components in effect: (1) first determines a different value added (the
standard value added in terms of current costs) which is larger than
Andrew's value added, and (2) then subtracts the IVA from the standard value
added. The result is then Andrew's value added. But this does not mean
that Andrew's value added is determined by or affected by the IVA. In
Andrew's example, the total standard value added would be 120 (thus
contradicting the shared assumption that 100 units of labor produce a value
added of 100) and the IVA would be -20, thus arriving at Andrew's 100. But
this is not the way Andrew's value added = 100 is determined.

So I think that Andrew is correct that in his example (and I imagine in his
TSS interpretation in general) that value added is not affected by the IVA.
I realize that Duncan may not be able to respond for a while, but we can
wait, and hopefully others will respond as well.

This does not mean that I think that the TSS interpretation is a correct
interpretation of Marx - more on this to come - but only that it is not
subject to the particular criticism that value added is affected by the IVA,
independent of the amount of labor. I look forward to continued discussion.