Re: [OPE-L] [Jurriaan on] Estimated percentage share of public sector employment

From: glevy@PRATT.EDU
Date: Tue Oct 11 2005 - 13:58:41 EDT

----- Original Message -----
From: "Jurriaan Bendien" <>
Sent: Tuesday, October 11, 2005 9:23 AM
Subject: Re: [OPE-L] [Jurriaan on] Estimated percentage share of public
sector employment

In a sense, you got to be correct - the relevant measure would be
GDP/employed labour force.

As you may know (and as I indicated in various wiki articles), there are
quite a few criticisms of the notion of GDP; even so, it does measure, even
if in a crude and inexact way, the new incomes generated by production,
affording some comparisons of proportions internationally, and over an
interval of let us say 5-10 years.

As regards productivity, my argument has been that there is no "neutral"
measure of productivity, a measure of productivity is a measure taken "
from a certain point of view", and one ought also to distinguish sharply
between physical productivity and productivity in value or price terms.
Oddly, neoclassical economists who hold a subjective theory of value,
will  nevertheless freely use GDP data based on an objective theory of
value, to  devise output measures. More consistent in this context is
so-called  Austrian economics of the Hayek type, which simply rejects the
validity of  price-aggregation (except where it involves their own bank
accounts, their  own transactions or their own well-paid  jobs). Austrian
economics, stripped of bullshit, is really stunningly simple: it just
says that "the market is  good".

My claim really is that, even using a variety of widely used official
measures, the typical "neoliberal" arguments fail to make much sense of
the factual evidence. You can e.g. have a high GDP and a large public
sector, a  low GDP and a large public sector, a high GDP and a small
public sector, and a low GDP with a small public sector, in a wide range
of countries of varying size and degrees of development. You can also have
a high GDP with a  high exports/GDP ratio, a low GDP with a high
exports/GDP ratio, a high GDP  with a low exports/GDP ratio, and a low GDP
with a low exports/GDP ratio. Sometimes privatisation correlates with
declining GDP, sometimes it correlates with a rising GDP. There is no
pattern here. To see the real pattern takes much more work.

There is no statistical correlation which would corroborate the idea that
a  proportionally large public sector in employment terms results
generally in  a low GDP or a low exports/GDP ratio, or is somehow a
general "drain on
wealth". Much more evidence would be required, to make that argument

Of course, the whole argument is a bit tricky anyway - correlation does not
necessarily mean causation; GDP itself includes public sector outputs, etc.

One of the conclusions that accidentally came out of my own Phd research on
social accounting (which I didn't complete, being priced out of the market
there) is that the measures derived from national accounts to describe the
economy are distorted by the specific way in which "income from production"
is conceptually distinguished from "property income".

That distortion becomes very important, when property income increases,
while income from production stagnates or grows only sluggishly. Gerald
Epstein & Arjun Jayadev have suggested that in the USA today, around 1/5 of
profits are rentier profits, yet to the extent that they base themselves on
gross product data, they actually underestimate the true proportion. Quite
possibly 1/4 is nearer to the truth, if we include property income excluded
from the product account.

Originally, GDP was designed to indicate the value of new net output, the
value-added when intermediate goods were deducted from gross output, but
over the years people started to use GDP as a total  income measure or a
total expenditure measure. That new approach is flawed, simply because
property income and transfer income is excluded from consideration (if
anywhere, it shows up in the income & outlay account, the capital account,
transfers account, or BOP data).

Marxists often fall into the same trap - Jacques Gouverneur writes for
example "The new monetary income that is created (or the sum of the net
prices of commodities) is calculated in the following way, setting out from
national accounts data: created income = net domestic product at market
prices, less income of earners outside the commodity sector. The net
domestic product represents the sum of incomes (gross) for the whole of the
economy" (Kapitalisme Vandaag [Contemporary Capitalism], p. 280). That's
just not true,  he's been hoodwinked by the value-added concept, and
double-entry bookkeeping.

That is to say, GDP does not measure total national income, at best it
measures only that income thought to be generated from production, but even
there, this is not quite accurate. The problems are, among others,

- a significant portion of interest, profit and rents paid from gross
enterprise income are excluded from value-added on the ground that these
are  defined as property income (a distinction is drawn for instance
between  property rents and the rental on leased productive assets),
- gross profit measures derived from tax data are adjusted in many ways -
some more credible than others - to arrive at a profit figure consistent
with the notion of value-added (I still have to write up how this really
- actual depreciation charged, is adjusted to arrive at "economic
depreciation", to which certain expenses directly related to fixed assets
are then added; in the last instance, "economic" depreciation is usually
based on schedules of observed average prices for assets traded at a
various ages.
- compensation of employees includes deferred income (principally social
insurance contributions), certain types of "expensing" by employers (which
can include imputations which are not made explicit), certain self-employed
income, and items such as profit-sharing and stock-options.
- tax considered to be unrelated to production is excluded.
- a valuation adjustment is made for inventories held.
- the imputed rental value of owner-occupied housing is included, even
although largely fictitious.
- there are big differences between countries in terms of the way
government  and household expenditure is accounted for.

A GDP measure thus results, which is a mixture of fact and theory, but
which  in my view neither constitutes an adequate measure of net output,
nor of  actual gross income distributed. Additionally, historical series
of GDP  obscure the changes in the division of labour, e.g. with respect
the growth of the tertiary sector (services).

The main reasons why GDP is still used are:

(1) because there is no other or better measure of total net new output
readily available,
(2) because it hides the true income distribution, i.e. the sources and
recipients of income generated by economic activity.

The second reason has become more and more important, as income inequality
has grown. However, the more foreign transactions expand, the more GDP is
undermined as as a measure which is in some sense a valid indicator of the
social product and income.

One of my favourite examples of confusion in this area concerns GDP versus
GNI (which is really just GNP). The World Bank and similar agencies state
vaguely that GNI is really more an "income" measure, than a "product"
measure. The real point however is, that GNI includes net factor income
from  abroad, and by valuing GNI using an "Atlas method" while valuing GDP
at  purchasing power parity, the amount of factor income transferred
internationally is conveniently obscured. Conceptually, of course, world
GDP  and world GNI are quantitatively identical, but the "Atlas method"
applied to world GNI will reduce world GDP by an increasing margin, which
nowadays  amounts to about US$1.1 trillion, i.e. an amount which comes
close to the reported annual real increase in the variable!

The ActionAid report "State of Disaster: Causes, Consequences & Policy
Lessons From Malawi",  dated June 13, 2002 queried "what the relevance of
development of macroeconomic targets, private sector led GDP growth,
fiscal discipline and export growth targets are, if such variables
seemingly have little impact on the ability of the economy to support food
security?" . Well, quite. While we
gawking at financial statements, an unknown number could  starve to death.
 The categories do not enable us to see the wood for the trees, and in
truth it takes quite a bit of work to uncover the true situation.


This archive was generated by hypermail 2.1.5 : Wed Oct 12 2005 - 00:00:01 EDT