With respect to the ongoing discussion of "bias" in technical change, I'd like
first to reiterate that, regarding macro data, Frank Thompson's calculations
for the U.S. show a basically trendless relation between GDP and "capital,"
for almost all measures of capital.
With respect to micro relations, William H. Peterson ("Capital-intensive
industry," in _The McGraw-Hill Encyclopedia of Economics_, 2d ed., 1994, pp.
123-25) writes:
"Whatever the industry, almost all technological advances tend to lead to the
substitution of more efficient capital equipment for less efficient capital
equipment. But in a larger sense, these substitutions amount to factor
substitution -- the replacement of labor by capital with a decline in unit
labor costs usually resulting. This phenomenon is especially the case in
decreasing-cost industries. Such industries as well as their customers are the
beneficiaries of economies resulting from large-scale production.
Manufacturing in general is a decreasing-cost industry, depending on the
degree of capital intensity.
"Given high intensity, volume is the key to decreasing costs, as in the
automobile industry. Assembly-line production, automatic multiple boring of
engine blocks, giant body-stamping equipment, and the like all demand
intensive, almost full-time use for the heavy capital investments to pay off.
When the investments do pay off, unit costs tend to fall."
The pattern Peterson notes would be classified, using the au courant
terminology, as capital-saving and (even more) labor-saving .
His discussion of economies of scale leads me to wonder whether those who
relate the falling profit rate to a rise in the "capital/output" ratio may
have the causation reversed. Given economies of scale, if output falls --
due, say, to a slump brought on by a falling rate of profit -- we should
expect, ceteris paribus, that the "capital/output" ratio will rise *as a
consequence*. It seems to me that the appropriate way to control for this is,
when measuring the productivity of "capital," to adjust the "capital" figures
for capacity utilization, to get a measure of employed "capital," as Solow's
famous study does.
Andrew Kliman (AX)