A few months ago, Duncan and I discussed the effect of
the stratification of fixed capital on the rate of profit.
As I recall, Duncan maintained that with a constant rate
of investment use of the rate of profit as a measure of 
profitability was unproblematic.  Since then I have been 
able to review a bit of the literature on the using the 
rate of profit to measure profitability.
1.  In their "On the Misuse of Accounting Rates of Return to
    Infer Monopoly Profits" (AEA, March 1983), Fisher and 
    McGowan argue that "Economists (and others) who believe
    that analysis of accounting rates of return will tell 
    them much (if they can overcome the various definitional
    problems which separate economists and accountants) are
    deluding themselves.  (p 91)  To be sure their piece 
    focuses on the difference between accounting depreciation 
    and economic depreciation at the level of the firm.  On 
    this basis, they develop their critique of the use of the 
    accounting rate of profit.  It's unclear to me why the
    same criticism would not be applicable to the macro level.
2.  Fisher responded to criticisms of the article in "The Misuse
    of Accounting Rates of Return: Reply" (AEA, June 1984).  In
    that response he also acknowledged his precursors.  They include
    G.C. Harcourt's "The Accountant in a Golden Age" (Oxford Economic
    Papers, March 1965).  In that piece (I have a reprinted version of
    it.), Harcourt maintains that anyone "...who compares rates of 
    profit of different industries, or of the same industry in different
    countries, and draws inferences from their magnitudes as to the 
    relative profitability of investments in different uses or countries,
    does so at his own peril." (Para 6.2 in the article)  
3.  Harcourt makes mention of other works in this area which I have
    not yet had a chance to review.  Among them is Joan Robinson's
    piece, "Depreciation."  I also have not tracked down the earliest
    cited article concerning the matter, "A General Mathematical 
    Theory of Depreciation" by Harold Hotelling in the Journal of 
    the American Statistical Association, September 1925, 20, 
    pp. 340-353.
4.  In their "Post Depression Trends in the Economic Rate of 
    Return" (The Review of Economics and Statistics, LXXII,
    406-413),  Dumenil and Levy use data to show that while
    there are differences between various accounting rates
    of return and the economic rate of return, the manner in
    which these measures move in U.S. post depression period 
    are similar. Here, as elsewhere, they use the vintage 
    technology to measure the economic rate of return.  In 
    their The Economics of the Profit Rate, they include sectors
    other than those of manufacturing in the U.S. economy and
    note that the economic and accounting rates of return
    are lagged by about 10 years.  They also point out that
    the greatest degree of stratification of fixed capital
    occurs prior to and within the Great Depression.  The 
    how's and why's of these phenomena are unexplored.
5.  Recognition of the problematic nature of the accounting 
    rate of profit at the very least forces those measuring 
    Marx's rate of profit (an accounting rate of profit) 
    for a given country over time to justify its use on 
    theoretical grounds.  What are the hidden assumptions 
    involved?  For example, are we tacitly assuming balanced 
    growth without technical change?  
6.  For those of us seeking to understand Marx's own efforts
    in CAPITAL, problems concerning this matter also arise.
    For example, given that capitalists use the economic rate
    of return to evaluate their investments, what is the 
    significance of the accounting rate of profit?  Marx himself
    notes that, on the one hand, a falling rate of profit blunts
    the stimulus to invest  and, on the other hand, a falling 
    rate of profit can lead to an insufficient mass of profit 
    to meet the demands for growth of capital in the next period.
    If capitalists are using the economic rate of profit for 
    investment decisions, then it is difficult to agree with 
    Marx that a fall in the accounting rate profit blunts the
    stimulus to invest.  However, it would seem that a fall
    in the accounting rate could lead to a shortage in the 
    mass of profit produced even if the economic rate of profit
    is rising.
John