[OPE] webpage computing dynamic rate of profit

From: Jurriaan Bendien <adsl675281@telfort.nl>
Date: Fri May 22 2009 - 06:53:47 EDT

In 1986, Geoff R. Pearce adopted almost exactly the approach that Peter
Fleissner recommends, for the analysis of New Zealand manufacturing data
(see G.R. Pearce, Where is New Zealand Going? Phd Thesis in sociology,
University of Canterbury, New Zealand, 1986, I worked on this as research

Pearce assumed a turnover of 5 or 6 weeks for manufacturing variable
capital, and from memory used the average of four quarterly inventory levels
as constant circulating capital. However, since fixed capital was about 70%
of the total physical capital outlay, the effect which including cc in the
total capital stock on the trend had was fairly slight.

Subsequently I experimented with "opening stocks" and "closing stocks" from
enterprise surveys, and also estimated the turnover as:

 T = i/x

where i = annual intermediate expenditure and x is the known average stock
level (or an opening/closing stock) held during the year.

The theory is that the inventory stock must rotate a certain number of
times, and that the total annual intermediate expenditure willl reflect
this. In principle, the UNSNA system recommends a distinction between
opening stocks, work-in-progress and closing stocks. But often these are not
measured quarterly, except for manufacturing, which serves as the economic
indicator, and often enterprises themselves are asked to provide the
"increase in stocks" item itself in questionnaires rather than actual stock
levels. I obtained unofficial stock levels from SNZ to create a series of
benchmark stocks for each industry, but the results of that exercise were
not credible. Lacking good data, you had to make just too many assumptions.

I found that all the results I obtained were not credible. Why? The main
reason appears to be, that the inventory level refers only to stockpiled
goods, but intermediate expenditure includes the cost of services and other
operating expenditures. Therefore, intermediate expenditure had to be split
into purchase of goods and purchase of services for the estimation of the
turnover, but this proved not to be feasible, for lack of data.

As an indication, from e.g. the NIPA I/O tables you can calculate that USA
intermediate consumption equals about 45% of the total gross output, and
that 40% of this intermediate consumption is the value of services. But the
proportions differ greately between industries, and in UNSNA the distinction
between goods and services is not made as rigorously as in the NIPAs.

Typically fluctuations in inventory will affect the shortterm trend, but the
longterm trend will be the same irrespective of whether inventory is
included or excluded in total capital, since the proportion of the inventory
in total capital does not change very significantly in the longrun.

Any model or formula used to estimate the turnovers must be based on some
evidential basis, but often the relevant empirical evidence available isn't
very adequate because the data is not collected in that way. Another method
I tried was that I looked at the bookkeeping in annual and quarterly company
reports to find more exact information about intermediate inputs relevant
for estimating the turnovers. However, I could not reconcile the proportions
in company accounts with the national accounting aggregates, I assume e.g.
because the national accounting aggregates include items not included in
company accounts, and apply rules to avoid doublecounting.

The next step for the analysis is to see what difference the inventory
stocklevels make as regards total capital stocks. In reality, the
significant quantities only concern particular industry groups, such as
manufacturing, mining, construction e.d. In almost all other sectors the
value of inventory held is too small to make much of a difference. And thus
if you can get industry level data on stocks of inventories, just including
a few strategic stocks in total capital may be sufficient to capture the


E-mail message checked by Spyware Doctor (
Database version: 5.10260
ope mailing list
Received on Fri May 22 06:56:04 2009

This archive was generated by hypermail 2.1.8 : Sun May 31 2009 - 00:00:03 EDT