I was reluctant to buy back in on this issue if there was no general
interest, but I am pleased to see that Polly Cleveland has re-opened the
thread so I will join in again. 

In regard to Bruce Caldwell’s comment that Hayek had written with
approval that “demand for commodities is not demand for labour”, I
agree that he had but I am less certain that he took Mill’s point.
Hayek is like Marshall in approving Mill's words, but then amending the
statement to say something else. 

What Marshall concluded was this: “demand for commodities is in
general demand for labour” (Principles 1920: 828).  The "not" has
completely disappeared. 

And this was Hayek’s conclusion: “More than ever it seems to me to
be true that the complete apprehension of the doctrine that ‘demand
for commodities is not demand for labour’ – and of its limitations
– is ‘the best test of an economist’.” (Pure Theory of
Capital: Appendix III: 439) Did Mill really have some limitations in
mind? 

Marshall for all practical purposes contradicts Mill writing that “in
general” demand for commodities is indeed demand for labour. Hayek
puts in a qualification – “and of its limitations” – that I do
not think belongs. Hayek more or less states that demand for commodities
is not demand for labour except where it is. Not such a fundamental
principle after all but I think Hayek is wrong (I hope I won’t be
struck by lightening for saying this). 

Further, James Ahiakpor writes that “it is puzzling to deny that the
demand for commodities may be a demand (indirectly) for labor, and that
more labor would be hired relative to capital goods depending upon the
wage-rental ratio.” 

I agree. It is puzzling, and has been for over a century to economists
who have been educated since the marginal revolution. I don’t myself
think that the wage-rental ratio is what Mill had in mind. 

Polly Cleveland wrote that “public policy generally biases investment
decisions to greater capital-intensity” and therefore, because of the
growth of more capital-intensive public works programs in the present
day, there has been a fall in the demand for labour as the proportion of
total national expenditure on public works has increased. It is an
interesting idea but it is surely not what Mill had in mind writing as
he did in 1848. He was writing about the effect on employment caused by
diverting productive efforts from the production of capital to the
production of goods and services devoted to current consumption. It is
this in itself that costs jobs. 

Finally, Michael Perelman wrote, “In my book on Keynes, I argued that
his neglect of replacement investment left a gaping hole in his
analysis. In later works, I explained how an insufficient level of
replacement investment was leaving the economy vulnerable.  Finally in
2007 (before the crisis) I wrote about how this vulnerability would lead
to a crisis.” 

I am tempted to agree with this if I have understood him correctly. If
he is saying that allowing existing capital to deteriorate will lower an
economy’s ability to employ, then I agree and that seems to me to be
an extension of Mill’s point. But if he is also saying that this
absence of replacement investment was the cause of the crisis, I would
have to withhold my assent although it is something worth thinking
about. 

I do not myself believe that any of the comments have actually shone
much light on what Mill meant and on what had received virtually
unanimous assent from his contemporaries. And the fact remains, given
the way our economies have unfolded in the past two years, demand for
commodities does not seem to have been demand for labour. Mill’s
conclusion has been demonstrated to be right. The only question is
whether his explanation is also right. I think it is, but the fact
remains that just about no one seems to have understood his point for
more than a hundred years. 


Dr Steven Kates
School of Economics, Finance
    and Marketing
RMIT University
Level 12 / 239 Bourke Street
Melbourne Vic 3000

Phone: (03) 9925 5878
Mobile: 042 7297 529