Professional Documents
Culture Documents
Bernard C. Beaudreau - Science and The Wealth of Nations-Cambridge Scholars Publishing (2021)
Bernard C. Beaudreau - Science and The Wealth of Nations-Cambridge Scholars Publishing (2021)
Bernard C. Beaudreau - Science and The Wealth of Nations-Cambridge Scholars Publishing (2021)
of Nations
Science and the Wealth
of Nations:
By
Bernard C. Beaudreau
Science and the Wealth of Nations: The Physics of Economic Growth
By Bernard C. Beaudreau
All rights for this book reserved. No part of this book may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording or otherwise, without
the prior permission of the copyright owner.
Preface ...................................................................................................... xv
Tables
Table 1.1 Material Processes, Energy and Organization ............................ 2
Table 1.2 Neoclassical and Other Enigmas ................................................ 5
Table 1.3 Output and Input Growth Rates: U.S., German and Japanese
Manufacturing ..................................................................................... 12
Table 1.4 Incursions into the Material Sciences: Missed Opportunities ... 13
Table 2.2 Key Developments in Thinking About Production .................. 36
Table 3.1 Manufacturing Output and Input Data 1950-84a ...................... 42
Table 3.2 Factor Income Shares, U.S. Manufacturing Value-Added ....... 44
Table 3.3 KLEP Regression Results, 1950-84' ......................................... 44
Table 3.4 Simple Growth Rates for Output and Inputs in US
Manufacturing, Selected Periods ........................................................ 45
Table 3.5 Changes in Labor Productivity, Total Factor Productivity, and
the Labor Intensity Ratios for the Pre-1973 and post-1973 Periods ... 46
Table 4.1 A Tool Taxonomy .................................................................... 57
Table 4.2 NPA Energy Productivity Measures ......................................... 64
Table 4.3 NPA Organization Productivity Measures ............................... 65
Table 4.4 KLEP Regression Results U.S., Canadian and Japanese
Manufacturing ..................................................................................... 68
Table 4.5 Kummel, Henn and Lindenberger’s Output Elasticities: U.S.,
Germany and Japanese Manufacturing ............................................... 68
Table 4.6 Output and Input Growth Rates: U.S., German and Japanese
Manufacturing ..................................................................................... 69
Table 4.7 Productivity Growth: U.S., German and Japanese
Manufacturing ..................................................................................... 70
Table 5.1 KLEP Regression Results U.S., Canadian and Japanese
Manufacturing ..................................................................................... 84
Table 5.2 Output and Input Growth Rates: U.S., German and Japanese
Manufacturing ..................................................................................... 85
Table 6.1 Contributions to Aggregate Productivity Growth of Industry ICT
Capital Deepening, EU-4 and USA .................................................... 97
Table 6.2 KLEP Regression Results U.S., Canadian and Japanese
Manufacturing ..................................................................................... 99
Table 7.1 Global R&D spending in 2011-billion U.S.$ ......................... 106
xii List of Tables and Figures
Table 7.2 World Energy Consumption by End Use Sector, 1990 .......... 112
Table 7.3 World Electricity Generation: Primary Energy Intensity
Relative to 1990 ................................................................................ 115
Table 7.4 World Transportation: Primary Energy Intensity Relative
to 1990 .............................................................................................. 116
Table 7.5 World Residential Primary Energy Intensity Relative to 1990 ... 119
Table 7.6 World Industrial: Primary Energy Intensity Relative to 1990 .. 121
Table 7.7 World Commercial: Primary Energy Intensity Relative
to 1990 .............................................................................................. 122
Table 7.8 World Primary Energy Intensity Relative to 1990. ................ 123
Table 8.1 Estimates of the Electricity-Use Output Elasticity-
Manufacturing ................................................................................... 139
Table 8.2 OLS-AR(1) Estimates-Log Linear Specification.................... 139
Table 8.3 OLS-AR(1) Estimates-Linear Specification ........................... 140
Table 8.4 OLS-AR(1) Estimates-Log Differences Specification ........... 140
Table 8.5 2-Digit SIC Industry Electric Power Output Elasticities
1947-1984 ......................................................................................... 142
Table 8.6 Multifactor Productivity Growth in Major U.S. Industries,
1948-1985 ......................................................................................... 145
Table 8.7 U.S. Annual Productivity Growth 1947-1980 ........................ 147
Table 8.8 Chronology of Maximum Fourdrinier Newsprint Paper
Machine Speeds ................................................................................ 155
Table 8.10 Output and Input Growth Rates: U.S., German and Japanese
Manufacturing ................................................................................... 160
Table 9.1 Manufacturing Processes and Corresponding Kinetics .......... 177
Table 9.2 Production per Factor Indexes-PFI ......................................... 182
Table 9.3 Estimates of the Electricity-Use Output Elasticity-
Manufacturing .................................................................................. 184
Table 9.4 OLS-AR Estimates-Log Linear Specification ........................ 186
Table 9.5 OLS-AR (1) Estimates-Linear Specification .......................... 186
Table 9.6 OLS-AR (1) Estimates-Log Differences Specification .......... 187
Table 9.7 2-Digit SIC Industry Supervision Input Elasticities
1947-1984 ......................................................................................... 188
Table 10.1 Second-Law Efficiency Increasing (increasing work per unit
energy) .............................................................................................. 196
Table 10.2 Innovations That Increased Energy Use Per Unit Capital Per
Unit Time .......................................................................................... 197
Table 10.3 Innovations/Ideas That Increased Energy Availability ......... 198
Table 10.4 Processes That Have Reached Their Kinetic Limit .............. 203
Science and the Wealth of Nations: The Physics of Economic Growth xiii
Figures
Figure 1.1 Actual and Predicted Growth of U.S. GDP 1960-1978 ........... 11
Figure 5.1 Structure of Production Processes ........................................... 79
Figure 6.1 Manufacturing Production Processes ...................................... 92
Figure 6.2 U.S. Net Stocks of Fixed Assets.............................................. 98
Figure 6.3 IT and non-IT Capital Services ............................................. 100
Figure 8.1 Productivity Growth Slowdown and Past Productivity
Growth, by Manufacturing Industry ................................................. 144
Figure 8.2 Productivity Growth by Manufacturing Industry
1948-1985 ........................................................................................ 147
Figure 8.3 Value-Added per kwh 1948-1980 by 2-Digit SIC Industry ... 149
Figure 8.4 Electric Power to Capital Ratios-Manufacturing Pre-1973 ... 151
Figure 8.5 History of Clock Rate and Power for Intel x86
Microprocessors ................................................................................ 156
Figure 10.1 U.S. R&D by Performing Sectors and Source of Funding
1953-2015 ......................................................................................... 204
Figure 10.2 Ratio of U.S. R&D to Gross Domestic Product by Source
of Funding 1953-2015....................................................................... 205
PREFACE
Chapters-Brief Summary
This section briefly describes each of the ten chapters.
went on record raising what he felt was a paradox, namely the ubiquitous
presence of computers everywhere from manufacturing plants to homes, but
their apparent absence in productivity statistics. Put differently, the much-
anticipated growth from computing had failed to materialize.
Using the energy-organization approach, he argues that because
information is not a source of force/energy, it cannot increase productivity,
thus rationalizing the information paradox as a prediction of the underlying
principles of basic physics.
The end result was a framework that shed considerable light on events
such as the Great Depression and the Productivity Slowdown. In short, the
latter was attributed to the fact that maximum machine speeds had been
reached in most industries/sectors in the late 1960s/early 1970s. As such,
the lower rates of growth of energy consumption pointed out by Reiner
Kummel and others owed not to higher prices, but to kinetics, namely that
machine speeds could not be increased further. In short, the productivity
slowdown concorded with what is generally regarded as the “end of the age
of speed.”
Abstract
In virtually every material science, the process of growth is not only well
understood, it has been systematically reduced to its thermodynamic and/or
kinetic equivalent. That is, growth is a function of resource and energy
availability, whether it be within a stationary or non-stationary environment.
This begs the question, why is economics the outlier, the exception? Why
are models of economic growth decoupled from the basic science of material
processes? This chapter attempts to answer this question by focusing on the
formalization of growth. It will be argued that for a number of reasons, the
economics profession has enigmatized material processes, introducing
concepts that were orthogonal to the laws found in the material sciences,
leading to the current situation where a whole new generation of enigmatic
approaches (quality ladders, institutions, etc.) has emerged to understand
previous engimas.
1.1 Introduction
In virtually every material process-based field/discipline, the process of
growth is not only well understood, it has been and is systematically reduced
to its thermodynamic and/or kinetic equivalent (biology, ecology,
demography). In short, growth is either a function of growth in energy
availability/use–whether it be within a stationary or non-stationary
environment–or an increase in second-law efficiency. The quintessential
example is photosynthesis where the growth of biomass is a function of
solar radiation, the latter being the force that acts on carbon dioxide and
water to produce carbohydrates/sugars. This begs the question, why is
economics the outlier, the exception? This chapter attempts to answer this
question by focusing on the very way in which the profession has formalized
material processes. It will be argued that for a number of reasons, the
economics profession, by enigmatizing a simple energy-based material
2 1. The Enigmatization of Economic Growth
process, has generated findings (i.e. the Solow residual) that have
prompted/led to the increased enigmatization of the growth process.
The chapter is organized as follows. To begin with, we present a
consilient approach to modeling material processes in general, namely the
energy-organization approach according to which output is increasing in
terms of two universal factor inputs, namely broadly-defined energy and
broadly-defined organization. This will then provide the basis for a
comprehensive review of the literature organized around two themes,
namely steady-state growth and non-steady-state growth (technology
shocks). This will be followed by a discussion of consilient approaches to
growth–that is, approaches that incorporate elements or aspects of physics.
We end with a set of external–read: scientific–guidelines for future work on
growth.
The latter, in turn, is a function of S(t) the supervisory input, T(t), tools,
and I(t) information. In keeping with basic physics, the latter three factor
inputs are not physically productive, but rather are organizational in nature,
affecting second-law efficiency. Better tools (e.g. Watt’s external condenser,
the Boulton-Watt dual-action steam engine, electric unit drive) increase
energy efficiency by minimizing losses. As Ș is bounded from above, it
stands to reason that organizational innovations will have limited effect on
output and output growth (Beaudreau and Lightfoot 2015). Equation 1.1
provides a simple description of the EO approach to material processes, with
E(t) being the energy input and Ș being the thermodynamic concept of
second-law efficiency. This can be seen as a measure of energy
productivity, which in this case, is a function of the relevant organizational
variables, including tools T(t), supervision S(t), and information I(t).
Beaudreau (1998) maintained that this simple model was universal in scope,
being applicable to all material processes.
The EO approach to growth is straightforward, namely that growth of
the output is an increasing function of growth of the energy input as well as
growth/innovations in Ș, second-law efficiency. The key as far as we are
concerned is the universality of Equation 1.1. Any and all growth processes
in the material sciences is/are predicated on growth in the energy input, and,
the case that concerns us here, growth in the organizational context. For
example, in the case of economic material processes, growth requires an
increase in energy as well as an equivalent increase in tools and supervision
–conventional capital and labor.
This raises the question of productivity or, put differently, the
contribution of factor inputs to output and growth. In keeping with basic
mechanics and thermodynamics, the only physically productive factor input
is energy/force. All others are organizational inputs, which together define
the material process, but are not productive in the traditional sense. Put
differently, they increase with output but are not the ultimate cause.
4 1. The Enigmatization of Economic Growth
“At the risk of being redundant, let me illustrate what we mean by the
question ‘How do men live?’ by asking what makes a railroad train go. In
one sense or another, credit for the achievement may be claimed by the so-
called ‘engine-driver, ’ the guard, the signalman, the manager, the capitalist,
the share-holder, -or again, by the scientific pioneers who discover the
nature of fire, by the inventors who harnessed it, by labour which built the
railroad and the train. The fact remains that all of them, by their collective
effort could not drive the train. The real engine-driver is the coal. So, in the
present state of science, the answer to the question how men live or how
anything lives, or how inanimate nature lives, in the sense in which we speak
of the life of a waterfall or of another manifestation of continued liveliness,
is, with few and unimportant exceptions, ‘By sunshine.’ Switch off the sun
and a world would result lifeless, not only in the sense of animate life, but
also in respect of by far the greater part of the life of inanimate nature”
(Soddy, 1924, 4).
material processes, which in turn can be traced back to the earliest attempts
on the part of moral philosophers and political economists to understand
wealth and its creation. For example, in Chapter 1 of Adam Smith’s An
Inquiry into the Nature and Causes of the Wealth of Nations, industrial
material processes are modeled as being labor-based, with labor
productivity (a scaler) being a function of (i) worker learning (ii) reduced
downtime, and (iii) the introduction of machinery. Ironically, while labor
had been reduced to a marginal factor input, overseeing steam-powered
machines, Smith put it at the center of his analysis, a decision that would be
heavy in consequences. For roughly a century, labor was front and center,
while the steam engine was couched in a parameter.
Enigma Violation
Capital is physically productive Principles of basic mechanics; tools
(simple and complex) are not
productive.
Labor is physically productive Machine operatives; production
requires both, yet worker-less
factories exist
Physically productive labor and Neither is physically productive
capital can be substituted
Energy/force is absent
Production functions exist Neither labor nor capital is physically
productive.
Solow Residual Roughly half of growth is attributed to
unknown factors. Metaphysical in
nature.
As it turned out, this became the central theme of Karl Marx’s labor
theory of value, namely that labor was the only productive factor input and
as such was entitled to the entirety of the product. Ironically, throughout
Marx’s life, labor was little more than an organizational factor input,
overseeing the workings of machines. Its brawn no longer powered the
material processes of the industrial revolution, yet it remained at the center
of the discourse.
The resulting crisis in classical economics (after all, Marx should be
considered to be a de facto classical economist), led eventually to the
neoclassical rejoinder, one that evacuated the problem of (unearned) profits
6 1. The Enigmatization of Economic Growth
in science. The proof is that some four decades later, not one of these
hypotheses has been confirmed empirically. In time, however, the
profession responded with a veritable flurry of activity aimed at modeling
the residual, understanding the productivity slowdown and ultimately
affecting the policy debate (i.e. increasing growth). The result has since
become known as New Growth Theory, which we now examine.
scaler. The underlying mechanics are, for the most part, not specified. As
such, subsequent developments such as the development in the 1840s of
high-pressure steam engines, and the development in the 1880s of the steam
turbine, both of which paved the way for greater machine speeds and
productivity, are ignored.
While these models or theories of industrial revolutions are enigmatic
with regard to the basic neoclassical production function (i.e. they affect A,
the technology scaler), they are, in essence, consistent with the underlying
principles of the material sciences. This owes in large measure to the
emphasis placed on the role of energy in the various aspects/dimensions of
first industrial revolution. For example, the Watt atmospheric steam engine
(with external condenser) in coal mines, the Boulton-Watt dual-action,
reciprocating engine in spinning, carding and weaving material processes.
In short, the first industrial revolution witnessed a massive increase in
energy use/consumption, resulting in an equally massive increase in
wealth. In other words, they are consistent with the laws of
physics/mechanics/thermodynamics. What could be regarding as an
important advance in the understanding of economic growth was, however,
lost in the ensuing analysis of its effect on productivity and output.
Specifically, it was seen as increasing both labor and capital productivity,
two inert factor inputs. In short, the steam engine was modeled as a scaler
affecting, in a one-shot manner, capital and labor productivity.
Dissatisfaction with material approaches to industrial revolutions led, in
the 1980s and 1990s, to a new approach, namely non-material where the
emphasis was on the institutional and cultural underpinnings of the
technological and institutional change that characterized the industrial
revolution eras. Drawing largely from institutional economics, it sought to
identify the exact set of circumstances or causes. Among the leading non-
material approaches is Joel Moykr’s Republic of Letters according to which
the enlightenment in England–what he refers to as the Baconian program–
combined with a growing interest in practical knowledge, led to the
industrial revolution (Moykr 2016). A good example of this, he argued, is
the Birmingham Lunar Society where men of science rubbed shoulders with
entrepreneurs/practical men. Another is Deirdre McCloskey’s notion of
Bourgeois Dignity, according to which societal values towards business in
general and endeavoring to make a return on one’s investment, laid the
ground for the industrial revolution (McCloskey 2010). In both cases, non-
economic factors were responsible for the cataclysmic change that was the
first industrial revolution.
From a scientific point of view, these approaches only serve to further
muddy the waters for the simple reason that they are virtually not testable.
10 1. The Enigmatization of Economic Growth
In short, the data set is limited to a single observation, thus eliminating any
possibility of testing for regularity. Both Moykr and McCloskey are aware
of this, and have responded with a barrage of anecdotes filling volumes. For
example, McCloskey develops her thesis in a trilogy of works, including
Bourgeois Dignity, Bourgeois Values and Bourgeois Equity. The sheer
volume, however, does not take away from the fact that her theory is not
testable.
Both approaches maintain that the record growth in labor and multifactor
productivity owed to an increase in energy-use intensity–that is, the increase
in energy use/consumption per unit of labor/capital. Implicitly, the enigma
that is the Solow residual is resolved, with productivity growth being
attributed to greater energy use per unit of labor/capital.
This brings us to the question of the productivity slowdown. In keeping
with the laws of classical mechanics, both attribute it to the fall in the rate
of growth of energy use/consumption, itself the result of the OPEC-induced
price increases. In other words, higher energy prices and the specter of even
higher future prices, reduced the rate of growth of energy use/consumption.
Once again, the enigma of the productivity slowdown is resolved, being
attributed to the decrease in the rate of growth of energy use/consumption.
In more recent work, Beaudreau (2017) re-examined the underlying
hypothesis, namely higher energy prices leading to a lower rate of growth
of energy use/consumption. Specifically, he showed that while fossil-fuel
prices increased in the mid-1970s, the price of electricity (primary source
of energy in manufacturing) remained relatively constant and moreover, the
real price of energy (and electricity) had, by the 1980s, returned to its pre-
OPEC crisis level, yet the rate of energy use/consumption did not rebound.
Drawing from his work on the economies of speed, he invoked the laws of
kinetics to attribute the decrease in energy use/consumption to the problems
inherent in speeding up material processes. More to the point, he argued that
maximum machine speed had, by the late 1960s/early 1970s, been reached
in most industries, making for a slowdown in the rate of increase of machine
speeds and consequently in productivity.
Table 1.3: Output and Input Growth Rates: U.S., German and
Japanese Manufacturing
processes in the 19th century led to more enigmas and even more in an
attempt to understand them.
1.6 References
Alting, Leo. 1994. Manufacturing Engineering Processes. New York, NY:
Marcel Decker Inc.
Beaudreau, Bernard C. 1998. Energy and Organization: Growth and
Distribution Reexamined. Westport, CT: Greenwood Press.
Beaudreau, Bernard C. and Douglas Lightfoot. 2015. “The Physical Limits
to Economic Growth by R&D-Funded Innovation,” Energy, 84: 45-52.
Beaudreau, Bernard C. 2017. “The Economies of Speed, KE=1/2mv2, and
the Productivity Slowdown,” Energy. 124: 100–113.
Beaudreau, Bernard C. 2018. “A Pull-Push Theory of Industrial
Revolutions,” Université Laval, manuscript.
Beiser, Arthur. 1983. Modern Technical Physics. Menlo Park, CA: The
Benjamin/Cummings Publishing Company.
Berndt, Ernst and David Wood. 1975. “Technology, Prices and the Derived
Demand for Energy.” Review of Economics and Statistics, 57, no.
3:259–268.
Denison, Edward F. 1962. The Sources of Economic Growth in the United
States and the Alternatives Before Us. New York, NY: Committee for
Economic Development.
Denison, Edward F. 1985. Trends in American Economic Growth 1929–
1982. Washington, DC: The Brookings Institution.
Jevons, W.S. 1865. The Coal Question. London: MacMillan and Co.
Jevons, W.S. 1965[1874]. The Theory of Political Economy. New York,
NY: Augustus M. Kelley, Bookseller.
Krugman, Paul. 2013. “The New Growth Fizzle.” The New York Times,
August 18, 2013.
Kummel, Reiner. 1982. “The Impact of Energy on Industrial Growth.”
Energy. 7, no. 2: 189- 201.
Kummel, Reiner, Dietmar Lindenberger, and Wolfgang Eichhorn. 1998.
“The Productive Power of Energy and Economic Evolution,” Indian
Journal of Applied Economics, Special Issue on Macro and Micro
Economics.
Lindenberger, Dietmar and Reiner Kummel. 2002. “Thermodynamics and
Economics.” Post-Autistic Economics Review, 14.
Lloyd-George David. 1924. Coal and Power. London: Hodder and
Stoughton.
Marshall, Alfred. 1890. Principles of Economics. London: MacMilllan.
Science and the Wealth of Nations: The Physics of Economic Growth 17
Abstract
It can therefore be surmised that the Wealth of Nations was first and
foremost about the new power drive technology that was the steam engine.
Processes that had been the realm of artisans and craftsmen were now
mechanized, increasing output manifold. Yet, despite this, the resulting
view of production was long on labor and short on force/energy. More
specifically, classical production theory posited a monotone increasing
relationship between labor and output. Put differently, Smith who
admittedly was a moral philosopher had failed to incorporate the basic
principles of natural philosophy (i.e. classical mechanics) into his analysis.
The great increasing of the quantity of work which, in consequence of the
division of labor, the same number of people are capable of performing, is
owing to three different circumstances; ¿rst, to the increase in dexterity in
every particular workman; secondly, to the saving of the time that is
commonly lost in passing from one species of work to another; and lastly,
to the invention of a great number of machines which facilitate and abridge
labor, and enable one man to do the work of many. (Smith 1776, 10)
Value, then, essentially differs from riches, for value depends not on
abundance, but on the difficulty or facility of production. The labour of a
million of men in manufactures will always produce the same value, but will
not always produce the same riches. By the invention of machinery, by
improvements in skill, by a better division of labour, or by the discovery of
new markets, where more advantageous exchanges may be made, a million
of men may produce double, or treble the amount of riches, of “necessaries,
22 2. Thinking About Production
This raises the obvious question, why were early moral philosophers
unable to provide a more thorough description of the steam engine and its
role in generation wealth or riches? In the grand scheme of things, the focus
remains on labor, with machinery as an aid of sorts, increasing conventionally-
measured labor productivity.
The first writer/political economist to refer to science, or the science of
the steam engine, was Robert Owen who in his Report to the County of
Lanark, pointed to “scientific improvements and arrangements” in reference
to advances in steam-driven power drive technology.
It is well known that, during the last half-century in particular, Great Britain,
beyond any other nation, has progressively increased its powers of
production, by a rapid advancement in scientific improvements and
arrangements, introduced, more or less, into all the departments of
productive industry throughout the empire. The amount of this new
productive power cannot, for want of proper data, be very accurately
estimated; but your Reporter has ascertained from facts which none will
dispute, that its increase has been enormousʊthat, compared with the
manual labour of the whole population of Great Britain and Ireland, it is, at
least, as forty to one, and may be easily made as 100 to one; and that this
increase may be extended to other countries; that it is already sufficient to
saturate the world with wealth and that the power of creating wealth may be
made to advance perpetually in an accelerating ratio. (Owen 1820, 19)
In turning from the smaller instruments in frequent use to the larger and
more important machines, the economy arising from the increase in velocity
becomes more striking. In converting cast into wrought iron, a mass of
metal, of about a hundredweight, is heated almost to white heat and placed
under a heavy hammer moved by water or steam power. This is raised by a
projection on a revolving axis; and if the hammer derived its momentum
only from the space through which it fell, it would require a considerably
greater time to give a blow. But it is important that the softened mass of red-
hot iron should receive as many blows as possible before it cools, the form
of the cam or projection on the axis is such, that the hammer, instead of
being lifted to a small height, is thrown up with a jerk, and almost the instant
after its strikes a large beam, which acts as a powerful spring, and drives it
down on the iron with such velocity that by these means about the double
the number of strokes can be made in a given time. (Babbage 1832, 26)
Day by day it becomes more evident that the Coal we happily possess in
excellent quality and abundance is the mainspring of modern material
civilization. As the source of ¿re, it is the source at once of mechanical
motion and of chemical change. Accordingly, it is the chief agent in almost
every improvement or discovery in the arts which the present age brings
forth. It is to us indispensable for domestic purposes, and it has of late years
Science and the Wealth of Nations: The Physics of Economic Growth 25
Undoubtedly, this citation could easily find itself into the first chapter of
any thermodynamics textbook or manual, given its perspicacity. Yet, less
than a decade later, all of this was forgotten in what is considered to be
Jevon’s magnum opus, namely The Theory of Political Economy published
in 1874 where wealth is defined as an increasing function of labor and
capital. In short, coal or energy was conspicuous by its absence.
One could argue that Jevons was of two minds, one positive, and the
other normative. His early work predated the publication of Das Kapital by
Karl Marx by a few years. Clearly, the thinking about production had been,
in many ways, subverted by the larger issue of legitimacy, the legitimacy of
profits.
Surprisingly, this duality is also present in Marx’s writings, specifically
in Chapter 15 of Das Kapital where he noted:
and ¿nally the tool or working machine. The motor mechanism is that which
puts the whole in motion. It either generates its own motive power, like the
steam-engine, the caloric engine, the electromagnetic machine, etc., or it
receives its impulse from some already existing natural force, like the water-
wheel from a head of water, the wind-mill from wind, etc. The transmitting
mechanism, composed of Ày-wheels, shafting, toothed wheels, pullies,
straps, ropes, bands, pinions, and gearing of the most varied kinds, regulates
the motion, changes its form where necessary, as for instance, from linear to
circular, and divides and distributes it among the working machines. These
two ¿rst parts of the whole mechanism are there, solely for putting the
working machines in motion, by means of which motion the subject of
labour is seized upon and modi¿ed as desired. The tool or working machine
is that part of the machinery with which the industrial revolution of the 18th
century started. And to this day it constantly serves as such a starting-point,
whenever a handicraft, or a manufacture, is turned into an industry carried
on by machinery. (Marx 1867, 357)
As it turned out, the early 20th century witnessed a second energy surge,
as it were, with the widespread use of a new power drive technology, namely
electric motors/dynamos. Like other energy-related innovations, it would
open the way to even greater energy use per unit of capital/labor, again in
the form of greater machine speeds, often referred to as machine speed-ups.
In short, shafting and belting would be replaced by electric unit drive,
untethering material processes and subprocesses from the constraints of
shafts and belts, resulting in autonomy and, more importantly, faster
machine speeds, the latter being achieved by increasing kinetic energy.
Like the Boulton-Watt reciprocating steam engine, this development/
innovation would go on to define an era, known commonly as the second
industrial revolution. Not only were belting and shafting-driven material
processes speeded up, processes that had not been mechanized, were. The
overall result was a massive increase in productivity and output. As the
General Electric Company was to describe it, ”Today is the Age of Speed.”
As Alfred D. Chandler pointed out perspicaciously, speed not physical scale
was of the essence.
Unfortunately, this massive, even paradigmatic shock failed to appear
on the economics radar. While productivity records were being shattered,
the profession remained impervious. After all, nothing had changed. In fact,
if anything, employment and capital spending were down. As it turned out,
electric unit drive was less cumbersome and less costly than belting and
shafting, resulting in a decrease in capital intensity, the latter being
measured in terms of costs. It is our view that this was perhaps the most
egregious failing of neoclassical production theory as it de facto preempted
a deeper understanding of the changes that were transforming not only the
U.S., but the World in general.
Not surprisingly, discordant voices began to appear. Clearly, a revolution
was underway. Productivity records were being shattered. New, electric-
powered goods were changing life as we knew it. These included household
appliances, electric-powered tools, electric-powered transportation, etc.
As it turned out, this prompted a response on the part of a handful of
economists, one that like Marx and Jevons’ forays in the 19th century,
focused on the role of energy/force/power in production. For example,
consider the cases of Columbia University professor Rexford G. Tugwell
and Yale University professor Irving Fisher. In a book entitled Industry's
Coming of Age, published in 1927, Tugwell described various changes to
industry, including “the bringing into use of new and better power resources
more suited to our technique, more flexible and less wasteful.”
flexible means of moving things cannot have taken place without numerous
secondary results in lowering costs, improvements in quality, and a
heightened moral among workers. For the new power is not only cheaper to
use; it is also cleaner, more silent and handier. On the whole, the
electrification of industry must be set down as the greatest, single cause of
the industrial revolution. (Tugwell 1927, 182)
In The Stock Market Crash and After, Irving Fisher ranked the
electrification of U.S. industry as one of the contributing factors behind the
stock market's rise in the late 1920s. Changing economic fundamentals, he
maintained, were the underlying cause.
But after 1919, something happened. The implications of which are not yet
sufficiently gauged. It was of enough significance to cause President
Hoover's Committee on Recent Economic Changes to remark that
“acceleration rather than structural change is the key to an understanding of
our recent economic developments.” The committee added: “But the breath
of the tempo of recent developments gives them new importance.” What
happened was indicated by the fact that in the United States, eight million,
three hundred thousand workers produced in 1925 one quarter more than
nine million wage workers turned out in 1919. The new indexes of the
Federal Reserve Board measuring production record this gratifying advance
which reflects an increase in the American standard of living. The indexes
cover, directly and indirectly, four-fifths of the industrial productivity of the
nation directly in about thirty-five industries, and collaterally, in many more.
The general volume of production had increased between 1919 and 1927 by
46.5 percent, primary power by 22 percent, and primary power by wage
earner by 30.9 percent (between 1919 and 1925) and productivity per wage
worker by 53.5 percent between 1919 and 1927. (Fisher 1930, 120)
be gauged by its use of power. The great advance in material standard of life
in the last century was made possible by an enormous increase in the
consumption of energy, and the prospect of repeating the achievement in the
next century turns perhaps more than anything else on making energy
cheaper and more abundant. A theory of production that will really explain
how wealth is produced must analyze the contribution of this element of
energy……These considerations have prompted the Institute of Economics
to undertake a reconnaissance in the field of power as a factor of production.
One of the first problems uncovered has been the need of a long-time index
of power, comparable with the indices of employment, of the volume of
production and trade, of monetary phenomena, that will trade the growth of
the factor of power in our national development. (Tryon 1927, 281)
Suppose that the thirteen million people living in the United States in 1830
had awakened on the morning of January 1, 1831, with forty times the
physical energy they had gone to bed with the night before. An active picture
meets the mind's eye; a very active picture. A lumberman can fell forty times
as many trees in a week, a housewife sweeps forty times as many square feet
of floor; forty barns can be built in the time hitherto required for oneand
forty chests and forty chairs. Porters can transport forty times their
accustomed load in a day; weavers ply their shuttles forty times as fastif
the shuttles can brook the strain; and children raise forty times their normal
rumpus.
Assuming no increase in the invention of labor-saving devicesand
where would be the point with such an exuberance of labor availablewhat
might we logically expect in the way of economic changes in a culture
essentially handicraft? From an economy of scarcity, with barely enough to
go around, the young republic would almost immediately enter an economy
of abundance. (Chase 1933, 1)
Clearly, frustration levels were on the rise as the thinking about production
had not kept pace with technology, making for dissidence. As it turned out,
this spilled over in the related field of engineering where the failure of
economics to adequately describe the paradigm changes underway at the
time, led to the emergence of a new approach not only to wealth, but to the
very workings of an industrial economy. This was Technocracy, which
offered up a new approach to understanding wealth, and a blueprint for
guaranteeing the transition to the new higher equilibrium growth path.
Consider, for example, the following excerpt from Howard Scott's Introduction
to Technocracy, published in 1933:
2.7 References
Beaudreau, Bernard C. 1995. The Impact of Electric Power on Productivity:
The Case of U.S. Manufacturing 1958–1984. Energy Economics. Vol.
17:231–236.
Science and the Wealth of Nations: The Physics of Economic Growth 37
Abstract
3.1 Introduction
It is generally agreed that the electrification of U.S. industry contributed
greatly to raising productivity and output (National Bureau of Economic
Research 1929, Schurr 1983, Rosenberg 1983, Jorgenson 1983, 1984). For
example, as early as 1929, the Committee on Recent Economic Changes,
chaired by Herbert Hoover, identified the increased use of electric power as
the most significant change in U.S. industry (National Bureau of Economic
Research 1929). More recently, Nathan Rosenberg observed that “the
spreading use of electric power in the 20th century has been associated with
the introduction of new techniques and new arrangements which reduce
total costs through their saving of labor and capital” (Rosenberg 1983, 24).
He concludes that “there has been a very wide range of labor-saving
innovations throughout industry which have taken an electricity-using form.
As a consequence, greater use of electricity is, from a historical point of
40 3. The Impact of Electric Power on Productivity
view, the other side of the coin of a labor-saving bias in the innovation
process” (Rosenberg 1983, 36).
In a study of 35 industrial sectors in the United States, Dale Jorgenson
found that technical change was electricity using in 23 of the 35. The decline
in real electricity prices prior to 1973, he argues, prompted increased
electrification via the substitution of electricity for other forms of energy
and through the substitution of energy for other inputsespecially labor
(Jorgenson 1983, 21). Electrification, he concludes, plays a fundamental
role in productivity growth.
Studies of the various sources of output growth, however, paint an
altogether different picture of the role of energy in general and electric
power in particular in the growth process (Denison 1985, Gollop and
Jorgenson 1980, Gullickson and Harper 1987). Nowhere is this more
apparent than in the recent debate over the productivity slowdown. Time-
series data show that output growth, however, measured, has fallen
dramatically from the mid-1970s on. For example, Gullickson and Harper
(1987) report that manufacturing output increased at an average annual rate
of 4.2% from 1949-73, but decreased to 0.6% from 1973 to 1984. A number
of writers pointed to the energy crisis as a cause. After all, everything
seemed to fit: energy prices, including electricity, began their upward climb
in 1973 which corresponded to the time at which productivity growth began
its downward slide.
Growth accountants, however, dismissed such hypotheses (Denison
1985). While they concurred that the rate of growth of energy consumption
had fallen dramatically, it alone could account for no more than 0.10 of the
decrease in output growth (i.e. 4.2%) (Gullickson and Harper 1987). Central
to their argument is the weight attributed to energy in deriving measures of
total factor productivity, notably 0.02-0.04. Put differently, the share of
energy in total factor income is largely insignificant. The energy crisis, they
surmise, could not explain the productivity slowdown.
This leaves us in a conundrum. The bulk of the historical evidence points
to energy in general, and electric power in particular, as an important cause
of productivity and output growth, yet, on the other hand, most, if not all,
studies of the sources of growth find energy to be marginally important.
Thus, either the historical record is incorrect, or previous studies have
underestimated the role of energy in general and electric power in particular
in the process of economic growth. In a recent study, Hisnanick and Kymm
(1992) provide new evidence which suggests that part of the productivity
slowdown can be attributed to the energy crisis. Specifically, they show that
by disaggregating broadly defined energy into a petroleum component and
Science and the Wealth of Nations: The Physics of Economic Growth 41
1 The relevant Divisia input indices are defined as follows: σୀଵ ߚ ݔ where Ei is the
relevant factor output elasticity and xi, is the relevant rate of growth of factor i.
2 Previous studies (e.g. Berndt and Wood 1975) estimate the relevant elasticities
electric component.
42 3. The Impact of Electric Power on Productivity
Year Q EP L K
1950 79.06 56.79 93.80 74.00
1951 84.25 62.74 99.26 77.28
1952 88.14 65.62 102.01 80.63
1953 97.34 74.69 108.23 83.78
1954 92.28 77.57 101.44 86.64
1955 104.95 93.73 106.05 89.48
1956 108.92 101.28 108.24 93.82
1957 107.13 101.94 107.77 97.89
1958 100.00 100.00 100.00 100.00
1959 112.33 109.81 104.15 101.67
1960 112.17 113.38 104.71 104.29
1961 110.96 114.91 101.99 106.69
1962 119.58 121.60 104.74 109, 27
1963 126.59 127.31 105.24 112.09
1964 133.90 136.89 106.89 116.01
1965 144.58 142.17 111.85 122, 09
1966 155.62 150.31 118.01 129.95
1967 157.39 158.44 119.90 137.24
1968 164.68 169.66 121.12 143.84
1969 167.78 181.79 124.20 150.74
1970 156.89 182.95 118.59 156.95
1971 157.40 187.13 112.99 161.97
1972 171.69 201.92 116.92 166.97
1973 185.22 218.46 122.36 173.05
1974 186.18 218.51 121.33 181.22
1975 166.83 206.75 111.35 187.56
1976 182.32 220.51 114.65 193.90
1977 195.95 227.51 120.05 200.81
1978 204.49 230.96 124.74 209.20
1979 208.94 234.58 128.10 217.79
1980 190.57 224.55 125.21 226.49
1981 186.87 225.72 122.67 234.84
1982 173.26 209.76 115.53 241.05
1983 179.66 214.18 113.16 244.95
1984 192.09 233.15 115.77 250.94
aThe output and input indices used in this study were derived as follows. The
4 Implicitly, we assume that total manufacturing output and materials are weakly
separable, which is also known as the Leontief aggregation condition. This can be
justified on a number of grounds. For example, by contrast with Berndt and Wood
(1975), our purpose here is not to predict the demand for energy, specifically,
electric power. Second, manufacturing output and value-added were found to be
highly correlated in the chosen sample period (U= 0.9652).
5 Among the others is a linear homogeneous production function (i.e. constant
returns to scale).
6 In 1967, 1974 and 1981 total energy costs in US manufacturing stood at USD$7,
691.7 million, USD$19, 468.3 million and USD$55, 255.1 million respectively. The
corresponding costs of electric power (purchased and generated) are USD$4, 398.11
million, USD$9576.26 million and USD$27, 609.85 million. Thus, the ratio of
electric power to total energy in US manufacturing stood at 0.5717 in 1967, 0.49188
in 1974 and 0.4996 in 1981.
7 Formally, V
EP = O.5VG.
44 3. The Impact of Electric Power on Productivity
Period vn vk ve
Berndt and Wood (1975)a 1947-71 0.6736 0.2902 0.0360
Gullickson and Harper (1987) 1949-83 0.7162 0.1590 0.1246
Denison (1962)b 1967-82 0.8263 0.1021
Kendrick and Grossman (1980) 1948-73 0.6552 0.3448
US Dept of Commerce (1985) 1967-81 0.4282 0.5233' 0.0459
aCost-based estimates of the relevant output elasticities. The values reported are net
8 lmplicitly, we assume that E , the scalar is equal to unity ((E =1). When E is
0 0 0
included, restrictions on E1, E1, and E1 are needed. Specifically, by imposing linear
homogeneity, we obtained the following E0, E1, E1, and E1 estimates (t-statistic): E0=
Science and the Wealth of Nations: The Physics of Economic Growth 45
1.073 (11.639); E1= 0.5372 (12.883); E2= 0.3997 (23.377); and, E3 =0.0630 (1.313).
The corresponding R2 is 0.984.
9 This result was found to be robust with regard to the sample period. Estimates for
the period 1964-84 were not statistically different from those reported in Table 3.3.
46 3. The Impact of Electric Power on Productivity
Unlike previous studies which found a sizable gap between the actual
rates of growth of output and aggregate input, our results show that growth
in U.S. manufacturing value-added is fully explained by growth in the
relevant Divisia index of factor input growth. For the complete period of
1950-84, manufacturing value-added increased at an average annual rate of
2.99%, while the aggregate input increased at 2.939%. For the first
subperiod 1950-73, manufacturing value-added increased at 4.217%, while
the aggregate input increased at 4.129%. Lastly, in the second subperiod,
1974-84, manufacturing value-added increased at an average annual rate of
0.3309, while the aggregate input increased at 0.3215.10
10 The corresponding values for tfp, defined earlier, are 0.063 (1950-84), 0.088
(1950-73) and 0.0093 (1974-84).
Science and the Wealth of Nations: The Physics of Economic Growth 47
For the entire sample period (i.e. 1950-84), labor productivity increased
at an average annual rate of 2.211%. In this period, the electric powerlabor
ratio increased at an average annual rate of 3.6704%, which when multiplied
by the relevant elasticity (i.e. 0.53304) yields a value of 1.956 percent which
measures the effects on labor productivity of the substitution of labor for
electric power referred to by Dale Jorgenson (Jorgenson 1984). In this
period, the capitallabor ratio increased at an average annual rate of 2.7801,
which when multiplied by the relevant elasticity yields a value of 0.178,
which measures the effects on labor productivity of the substitution of labor
for capital. The sum of these two effects accounts for 97% of the growth in
labor productivity in US manufacturing.
Prior to the energy crisis, labor productivity increased at an average
annual rate of 2.842% of which 2.585% can be attributed to laborelectric
power substitution and 0.146% can be attributed to laborcapital
substitution. Together, these two effects account for 96% of the overall
increase in labor productivity. In the ensuing decade, labor productivity
increased at an average annual rate of 0.834% of which 58.3% can be
attributed to laborelectric power substitution and 24.9% can be attributed
to laborcapital substitution.
3.5 Conclusions
For years, writers have pointed to the increased use of electric power as
a key factor in the growth of productivity. As such, to most observers, it
seemed perfectly reasonable to assume that the productivity slowdown
which began in 1973-74 was the direct result of the energy crisis. Oil prices
had increased dramatically, driving up all energy prices, including the prices
of electricity. Growth accountants, however, dismissed such claims, arguing
that while energy consumption in general and electric power in particular
had indeed slowed down markedly, neither could explain more than 1-2%
of the productivity slowdown.
Our findings, however, reconcile the historical record with the empirical
facts. By using direct as opposed to indirect estimates of the relevant output
factor input elasticities, we were able to reconcile the historical record as
described by Schurr (1983), Jorgenson (1983, 1984), and Rosenberg (1983)
48 3. The Impact of Electric Power on Productivity
3.6 References
Berndt, Ernst and Wood, David O. 1975. “Technology, Prices and the
Derived Demand for Energy.” The Review of Economics and Statistics.
August: 259-268
Christensen, L. R. and Dale Jorgenson. 1970. “US Real Product and Real
Factor Input” The Review of Income and Wealth, March.
Denison, Edward F. 1985. Trends in American Economic Growth 1929-
1982. The Brookings Institution, Washington, DC (1985).
Gollop, F. M. and Dale Jorgenson. 1980. “US productivity Growth by
Industry, 1948-1973” in Kendrick, J. W. and Vaccara B. (eds.) 1980
New Developments in Productivity Measurement and Analysis.
Chicago: National Bureau of Economic Research.
Gullickson, W. and M. J. Harper, 1987. “Multifactor Productivity in US
Manufacturing, 1949-1983.” Monthly Labor Review. October:18-28
Hisnanick, J. J. and Kymm, K. O. 1992. “The Impact of Disaggregated
Energy on Productivity.” Energy Economics. October: 274-278.
Hounshell, David A. 1984. From the American System to Mass Production
1800-1932: The Development of Manufacturing Technology in the
United States. Baltimore, MD: The Johns Hopkins University Press.
Jorgenson, Dale W. 1983. “Energy Prices and Productivity Growth,” in
Schurr, Sam. et al (eds.) Energy, Productivity, and Economic Growth.
Cambridge, MA: Oelgeschlager, Gunn and Hain.
Jorgenson, Dale W. 1984. “The Role of Energy in Productivity Growth,” in
Kendrick, J. W. (ed) International Comparisons of Productivity and
Causes of the Slowdown. Ballinger, Cambridge, MA: Ballinger.
Kendrick, J. W. and Grossman, E. S. 1980. Productivity in the United States,
Trends and Cycles. Baltimore, MD: The John Hopkins University Press.
National Bureau of Economic Research. 1929. Recent Economic Changes
in the United States. New York, NY: McGraw-Hill.
Science and the Wealth of Nations: The Physics of Economic Growth 49
4.1 Introduction
As pointed out in the Introduction, 19th and 20th-century political
economists failed to provide a theory or model of production that was consilient
with classical mechanics and/or the emerging field of thermodynamics. The
fallout was to be fatal, to say the least. Labor, by then a supervisory input,
was seen as physically productive; capital (tools) was simply decreed to be
physically productive, in abject contradiction with classical mechanics. To
begin with, we present a consilient model of material processes as they
apply to political economy, namely the energy-organization model
developed in Beaudreau (1998). Drawing from the pure and applied
sciences, the energy-organization approach focuses on two universal inputs,
namely broadly-defined energy and organization. Broadly-defined energy
includes both animate and inanimate forms of energy. By animate energy,
it should be understood muscular (human and animal) power; by inanimate
energy, it should be understood wind, fossil fuel-based, hydraulic and
nuclear power. All production processes, Beaudreau (1998) argued, involve
the consumption of energy of one type or another.1 Organization will be
defined as the conception/design of, and the overseeing (i.e. supervision) of
energy-consuming (i.e. entropic) production processes. The development of
the steam engine by Papin, Savary, Newcommen and Watt in the 17th and
18th centuries is an example of the former, while its day-to-day operation is
available work of a particular energy source, keeping in mind that energy cannot be
created or destroyed (First Law of Thermodynamics). As such, energy consumption
is equivalent to the concept of entropy.
52 4. The Energy-Organization Framework
an example of the latter. In the natural world, the design of and supervision
of energy-consuming processes is governed by forces that are not fully
understood. We shall refer to such processes as naturally-occurring/spontaneous
entropic processes. Where these differ from man-made (i.e. anthropomorphic)
entropic processes is in their organization, broadly defined. From the
Paleolithic era to the present, Homo sapiensneanderthalensis and
sapienshave designed and redesigned man-made entropic processes (i.e.
anthropic entropic processes). For example, the development of stone tools
in the Paleolithic era altered the very nature of work. By reducing waste (i.e.
increasing efficiency), primitive tools such as hammers and knives
increased the amount of work that could be carried out by a given quantity
of muscular energy. In fact, the development and improvement of
anthropomorphic entropic processes are what define various pre-historical
and historical eras (e.g. the stone age, the bronze age, the machine age).2
As we shall see, the resulting model of production has important
implications for distribution theory. Specifically, given that energy and
energy alone is productive in the physical sense, existing theories of
distribution in factory production (i.e. classical, radical-Marxian,
neoclassical) are incomplete, not to mention misspecified. For example, the
notion of marginal product of labor and capital is theoretically irrelevant,
labor and capital not being physically productive. What’s more, they violate
basic mechanics. In their place, we offer a bargaining approach to
distribution in which the owners of energy and organization bargain over
the final product, in this case, value-added, one in which only energy is
physically productive. The notion of energy rents is introduced. Energy
rents are, by definition, the difference between the value product of energy
and its price (cost). The owners of the energy and organization inputs, it is
argued, bargain over these rents, the result (solution) of which is the
observed functional distribution of income.
2 Those interested in the evolution of production processes over time are referred to
force, and d = distance: work equals force times distance. Thus, the greater
is f, ceteris paribus, the greater is W. The greater is d, ceteris paribus, the
greater is W. In short, the more force exerted and the longer the distance (or
time period) over which the force in question is exerted, the more work
performed. Redefining work as output permits us to write the basic axiom
of the energy-organization approach, namely that for any given, well-
defined man-made entropic process, output is an increasing function of
energy consumption.
Contrast this with the standard, seemingly time-invariant definition of
work found in political economy, namely that work is an increasing function
of capital and labor (i.e. W=f(K, L)). Capital and labor, it is argued, produce
output (value-added). Both are assumed to be productive. Just what it is that
capital and labor do in production processes is unspecified. Terms such as
capital productivity and labor productivity, however, connote the idea that
both somehow work. Inanimate forms of energy such as oil, gas and
electricity consumption are assumed to be intermediate inputs, and, hence,
are not productive in the conventional sense. Put differently, they are not
factors of production (value-adding). In short, capital and labor are assumed
to add value to energy and other raw materials. Broadly-defined organization
is also ignored. Production processes are assumed to exist. Management
issues are, in general, ignored.
Clearly, the physical and economic definitions of work are worlds apart.
For three centuries, physicists have focused on force and energy as the basis
of work; economists, on the other hand, have focused on capital and labor.
In physics, tools and machines (i.e. capital) modify and transmit force/
energy, but are not, as such, sources of energy. In political economy energy
(inanimate energy forms) is viewed as an intermediate good, and, thus, is
not productive in the traditional sense (value-adding).
where: W(t) = work in period t; and E(t) = energy (i.e. force) in period t.
54 4. The Energy-Organization Framework
At the risk of being redundant, let me illustrate what I mean by the question
How do men live? by asking what makes a railroad train go. In one sense or
another, credit for the achievement may be claimed by the so-called engine-
driver, the guard, the signalman, the manager, the capitalist, the share-
holderor, again, by the scientific pioneers who discovered the nature of
fire, by the inventors who harnessed it, by labour which built the railroad
3The view that conventionally-defined labor can be broken down into a force (i.e.
energy) and supervisory component is as old as thermodynamics itself. German
physicist and physiologist Hermann von Helmholtz argued that the forces of nature
(mechanical, electrical, chemical, etcetera) are forms of a single, universal energy
or Kraft, that cannot be either added to or destroyed. According to Ansom
Rabinbach, As Helmholtz was aware, the breakthrough in thermodynamics had
enormous social implications. In his popular lectures and writings, he strikingly
portrayed the movements of the planets, the forces of nature, the productive forces
of machines, and, of course, human labor power as examples of the principle of
conservation of energy. The cosmos was essentially a system of production whose
product was the universal Kraft, necessary to power the engines of nature and
society, a vast and protean reservoir of labor power awaiting its conversion to work
(Rabinbach 1990, 3).
Science and the Wealth of Nations: The Physics of Economic Growth 55
and the train. The fact remains that all of them by their collective effort could
not drive the train. The real engine-driver is the coal. So, in the present state
of science, the answer to the question how men live, or how anything lives,
or how inanimate nature lives, in the sense in which we speak of the life of
a waterfall or of any other manifestation of continued liveliness, is, with few
and unimportant exceptions, By sunshine. Switch off the sun and a world
would result lifeless, not only in the sense of animate life, but also in respect
of by far the greater part of the life of inanimate nature. (Soddy 1924, 4)
Accordingly, because tools are not a source of energy, they are not
productive in the physical sense. That is, they do not work. Nowhere is this
better seen than in the physicist’s definition of a machine which consists of
an instrument used to transmit or modify force/energy.
By using primitive hammers and knives, early man (i.e. Paleolithic and
Neolithic man) was able to better direct and apply his force. Analytically
speaking, however, while tools allowed him to minimize energy loss (i.e.
wasted energy), they did not allow him to increase the total amount of work
beyond the initial level of force. That is, they were not a source of additional
energy.4 Implicit here is the basic notion of thermodynamic efficiency,
defined as work out versus work in. By better transmitting muscular force,
tools improved early man’s thermodynamic efficiency. With the same
amount of energy, more work could be done (e.g. skinning animals, cutting
firewood).
This raises the question of the relevant measure of tool/capital
productivity. Specifically, how do we measure and define the productivity
of tools/machines (i.e. capital)? The answer, we believe is straightforward.
Since tools/machines/capital are not a source of energy, they cannot be
regarded as productive in the classical sense (i.e. performing work/working).
4 This has important consequences for distribution. Clearly, the owner(s) of the
tools/machines cannot lay claim to a portion of the output on the basis of work.
Instead, their claim has to be based on tools/machines contribution to second-law
efficiency. Tools/machines improve second-law efficiency, thus increasing output.
56 4. The Energy-Organization Framework
that while tools/machines can better distribute the overall amount of work to be
done, they do not, in any way, reduce it. In other words, because they are not a source
of energy, they cannot increase the overall amount of work being performed.
Science and the Wealth of Nations: The Physics of Economic Growth 57
New machinery, when just invented, generally requires a great deal of care
and attention. But the work of its attendant is always being sifted; that which
is uniform and monotonous is gradually taken over by the machine, which
thus becomes steadily more and more automatic and self-acting; till at last,
there is nothing for the hand to do, but to supply the material at certain
intervals and to take away the work when it is finished. There still remains
the responsibility for seeing that the machinery is in good order and working
smoothly; but even this task is often made light by the introduction of
8One could argue that natural entopic processes are also subject to breakdown. For
example, take the human body and the numerous diseases that prevent energy from
being transformed into work.
Science and the Wealth of Nations: The Physics of Economic Growth 59
where S(t) and T(t) correspond to the level of supervision at time t and tools
at time t, respectively. For the time being, we simply assume that second-
law efficiency is an increasing function of tools and supervision. Thus, for
a given quantum of energy, the more/better tools and supervision, the
greater is second-law efficiency (i.e. Ș(t)), and hence, the greater is output
(i.e. W(t)). Capital and supervision, it therefore follows, are not directly
productive; rather, they contribute to work via their effect on second-law
efficiency.
Here, we stop short of explicitly modeling supervisory activity, except
to point out the obvious, namely that, historically, it has been carried out by
individuals (animate supervision), and secondly, has been organized
hierarchically, with conventionally-defined workers (lower-level supervisors)
at the bottom, line supervisors above, and senior managers (e.g. CEOs,
CFOs and the Board of Directors) at the top. It is clear that while not a
source of energy, the supervisory input is nonetheless a sine quo non of
virtually all production processes. Without supervisors, output becomes
probabilistic (including the null set).9 Tools and machines break down,
resulting in a loss of energy and output.
We may now pass to the effects which machinery has in relieving that
excessive muscular strain which a few generations ago was the common lot
of more than half the working men even in such a country as England…. in
other trades, machinery has lightened man’s labours. The house carpenters,
for instance, make things of the same kind as those used by our forefathers,
with much less toil for themselves….. Nothing could be more narrow or
monotonous than the occupation of a weaver of plain stuffs in the old time.
But now, one woman will manage four or more looms, each of which does
many times as much work in the course of a day as the old handloom did;
and her work is much less monotonous and calls for much more judgment
than his did. (Marshall 1890, 218)
We may readily apply the principle here illustrated to the actual historical
facts. The introduction of machinery during the last hundred years has to a
certain extent changed the directions of man’s occupations. Instead of
making things with their own hands, as they formally had to do, they are
now managing machines or assisting in various ways in working them. The
pin-makers are no longer at work; a few of them are feeding pin-making
machines, but the majority of them have learned other employments. A large
class of carpenters no longer push the plane; a portion of them feed the
planning machines, and the remainder are fully occupied in executing work
that increased refinement which demand has encouraged. The same thing
may be traced all through the channels of industry. (Newcomb 1886, 390)
10This explains the increasing use of child labor (i.e. child supervision) in factories
11 This is especially true at the aggregate level. Production processes differ markedly
within and across sectors.
12 Unfortunately, qualitative measures or indexes (information) of capital and labor
in manufacturing (and for other sectors) are unavailable at the aggregate level.
Ideally, such measures could be used to proxy innovations in second-law efficiency.
For more on the role of information in production, see Jorgenson (2001), where
investment in information technology (IT)-based capital is used as a proxy for
information. Unfortunately, such a measure would be of limited value in our work,
given the complementary nature of IT-capital and supervision (automation), and the
lack of a reliable measure of information quality.
Science and the Wealth of Nations: The Physics of Economic Growth 67
post-WWII input and output data for U.S., German and Japan manufacturing,
were then used in standard growth accounting.
The relevant output elasticities were estimated directly (i.e. as opposed
to indirectly using cost data). Chief among the reasons for proceeding this
way was the absence of competition in both energy and labor markets
(Blanchflower, Oswald and Sanfey 1996, Van Reenen 1996). Specifically,
electric power markets are, in general, regulated, via public utility
commissions, while labor markets, according to Blanchflower et al and Van
Reenen, are, in general, non-competitive. This violates one of the key
assumptions of cost-based growth accounting, namely that factor markets
are assumed to be competitive. In the presence of non-competitive factor
markets, factor shares do not mirror physical productivity. To get around
this problem, output elasticities were estimated directly. Data on value-
added, electric power consumption, total employment and capital for U.S.,
German and Japanese manufacturing were used to estimate the Cobb-
Douglas KLEP output elasticities.
The estimated output elasticities (ȕ’s) for all three countries are presented
in Table 4.4. What is striking are the similarities across countries. In all three
cases, electric power consumption is, by far, the most important factor input,
as evidenced by output elasticities for U.S. manufacturing, German
manufacturing and Japanese manufacturing of 0.537244, 0.747482 and
0.605599, respectively. Capital and labor output elasticities are lower than
in previous studies (i.e. Berndt and Wood 1975).
These results, we argue, provide some support, albeit limited, for the
energy-organization model of production presented above. ȕ1, the electric
power output elasticity, is greaterby a magnitude of tenthan estimates
reported elsewhere in the literature. The labor output elasticity is smaller
than typically reported, while the capital output elasticity is somewhat
comparable. Does the fact that ȕ2 and ȕ3 take on positive and statistically
significant values weaken the case for the energy-organization approach?
The answer, we believe, is no, as such elasticities only highlight the
weaknesses of conventional regression analysis in this context. Correlation
does not imply causality. It is important to recall that organization is as
important a part of the energy-organization approach as energy, but that
energy and only energy is physically productive. Organization is a
necessary input, one that is organizationally productive, but not physically
productive.
68 4. The Energy-Organization Framework
Using these estimates, Beaudreau (1995, 1998) and Kummel, Henn and
Lindenberger (2002) reexamined growth of manufacturing output in the
post-WWII period, with particular emphasis on the pre- and post-energy
crisis subperiods. Specifically, they set out to demonstrate that contrary to
the accepted wisdom, the two energy crises were a leading cause of the
productivity slowdown. In short, the energy crises (1973 and 1979) ushered
in a period of zero energy-consumption growth, which, given the prominent
Science and the Wealth of Nations: The Physics of Economic Growth 69
Table 4.6: Output and Input Growth Rates: U.S., German and
Japanese Manufacturing
Similar results were reported by Kummel, Henn and Lindenberger using the
Linex approach.
13As pointed out earlier, this view of productivity is archaic. Labor is not productive
in the physical sense; instead, it is productive in the organizational sense. As such,
labor productivity is analogous to management productivity.
Science and the Wealth of Nations: The Physics of Economic Growth 71
At the risk of being redundant, let me illustrate what I mean by the question
“How do men live?” by asking what makes a railroad train go. In one sense
or another, credit for the achievement may be claimed by the so-called
“engine-driver,”, the guard, the signalman, the manager, the capitalist, the
share-holder, -or, again, by the scientific pioneers who discovered the nature
of fire, by the inventors who harnessed it, by labour which built the railroad
and the train. The fact remains that all of them by their collective effort could
not drive the train. The real engine-driver is the coal. So, in the present state
of science, the answer to the question how men live, or how anything lives,
or how inanimate nature lives, in the sense in which we speak of the life of
a waterfall or of any other manifestation of continued liveliness, is, with few
and unimportant exceptions,”By sunshine.” Switch off the sun and a world
would result lifeless, not only in the sense of animate life, but also in respect
of by far the greater part of the life of inanimate nature. (Soddy 1924, 4)
This approach, which is consilient with Soddy’s view of how men live,
and a response to Tryon’s critique of production theory, provides the
analytical framework that will be used throughout the next eight chapters of
the book. Soddy’s coal is the energy and his so-called “engine-driver,”
guard, signalman, manager, capitalist, share-holder, scientific pioneers who
discovered the nature of fire, inventors who harnessed it, labor which built
the railroad and the train, are the organization.
4.9 Bibliography
Abraham, Katherine G., James R. Spletzer, and Jay C. Stewart. 1999. "Why
Do Different Wage Series Tell Different Stories?" American Economic
Review. 89, no. 2: 34-39.
Adams, Gregory & Rausser, Gordon. & Simon, Leo. 1996. "Modelling
Multilateral Negotiations: An Application to California Water Policy,”
Journal of Economic Behavior & Organization. 30, no. 1: 97-111.
Aghion, Philippe and Peter Howitt. 1998. Endogenous Growth Theory.
Cambridge, MA: MIT Press.
Alting, Leo. 1994. Manufacturing Engineering Processes. New York:
Marcel Decker Inc. 1994.
Beaudreau, Bernard C. 1995. The Impact of Electric Power on Productivity:
The Case of U.S. Manufacturing 1958–1984. Energy Economics.
17:231–236.
Beaudreau, Bernard C. 1996. Mass Production, The Stock Market Crash,
and The Great Depression: The Macroeconomics of Electrification.
Westport, CT: Greenwood Press.
Beaudreau, Bernard C. 1998. Energy and Organization: Growth and
Distribution Reexamined. Westport, CT: Greenwood Press.
Beaudreau, Bernard C. 1999. Energy and the Rise and Fall of Political
Economy. Westport, CT: Greenwood Press.
Beaudreau, Bernard C. 2004. Making Markets and Making Money, Strategy
and Monetary Exchange. New York: iUniverse.
Beiser, Arthur. 1983. Modern Technical Physics. Menlo Park, California:
The Benjamin/Cummings Publishing Company.
Bernard, Jean-Thomas and Benoît Coté. 2002. “The Measurement of the
Energy Intensity of Manufacturing Industries: A Principal Components
Analysis.” Resources for the Future Discussion Paper 02-31.
Berndt, Ernst and David O. Wood. 1975. “Technology, Prices and the
Derived Demand for Energy.” The Review of Economics and Statistics:
259–268.
Science and the Wealth of Nations: The Physics of Economic Growth 73
Dale W. Jorgenson and Kevin J. Stiroh. 2000. “Raising the Speed Limit:
U.S. Economic Growth in the Information Age.” Brookings Papers on
Economic Activity: 1, Brookings Institution.
Kahn, Alfred E. 1988. The Economics of Regulation: Principles and
Institutions. Cambridge, MA: MIT Press.
Krueger, Alan B. 1999. Measuring Labor’s Share. Cambridge, MA,
National Bureau of Economic Research, Inc. (Working Paper 7006.)
Kummel, Reiner, Dietmar Lindenberger and Wolfgang Eichorn. 1998. The
Productive Power of Energy and Economic Evolution. University of
Wurzburg Working Paper.
Kummel, Reiner, Julian Henn, Dietmar Lindenberger. 2002. “Capital,
Labor, Energy and Creativity: Modeling Innovation Diffusion.”
Economic Dynamics and Structural Change. 13: 415-433.
Landes, David S. 1980. “The great drain and industrialization: Commodity
flows from periphery to center in historical perspective,” in R.C.O.
Matthews (Ed.) Economic Growth and Resources, Trends and Factors.
London: MacMillan.
Leibenstein, Harvey. 1968. “Entrepreneurship and Development.” American
Economic Review. 58: 72-83.
Lloyd George, David. 1924. Coal and Power. London: Hodder and
Stoughton.
Maddison, Angus. 1987. “Growth and Slowdown in Advanced Capitalist
Economies: Techniques of Quantitative Assessment.” Journal of
Economic Literature. 25: 649–698.
Malthus, Thomas R. 1827. Principles of Political Economy Considered with
a View to Their Practical Application. New York: Augustus M. Kelley.
Marshall, Alfred. 1890. Principles of Economics. London: MacMillan.
Marx, Karl. 1867. Das Kapital. Chicago: Encylcopaedia Britannica (1992).
Newcomb, Simon. 1886. Principles of Economics. New York: Augustus
Kelley.
Nye, David E. 1990. Electrifying America: Social Meaning of a New
Technology. Cambridge, MA: MIT Press.
Odum, H.T. and E.C. Odum. 1976. Energy Basis for Man and Nature. New
York: McGraw-Hill.
Owen, Robert. 1817. The Life of Robert Owen. London: Cass, 1967.
Owen, Robert. 1820. A New View of Society and Other Writings, London:
J.M. Dent and Sons, Ltd. (1927).
Rabinbach, Anson. 1990. The Human Motor: Energy, Fatigue and the
Origns of Modernity. New York, NY: Basic Books.
Ricardo, David. 1817. The Principles of Political Economy and Taxation.
New York: Everyman's Library.
76 4. The Energy-Organization Framework
Rifkin, Jeremy. 1995. The End of Work. New York: G.P.Putnam’s Sons.
Rosenberg, Nathan. 1972. Technology and American Economic Growth.
Armonk, NY: M.E. Sharpe.
Rubinstein, Ariel and Delip Abreu. 1988. “The Structure of Nash
Equilibrium in Repeated Games with Finite Automata.” Econometrica.
56: 1259-1282
Scott, Howard et al. 1933. Introduction to Technocracy. New York: The
John Day Company.
Smith, Adam. 1776. An Inquiry into the Nature and Causes of the Wealth
of Nations. Chicago: Encyclopaedia Britannica, 1990.
Sobel, Robert M. 1972. The Age of Giant Corporations, A Microeconomic
History of American Business 1914-1970. Westport, CT: Greenwood
Press.
Soddy, Frederick. 1924. Cartesian Economics, The Bearing of Physical
Sciences upon State Stewardship. London: Hendersons.
Solow, Robert M. 1974. “The Economics of Resources or the Resources of
Economics.” American Economic Review. 64, no. 2: 1-14.
Stahl, I. 1972. Bargaining Theory. Stockholm Research Institute, Stockholm.
Stiroh, Kevin J. 2001. “Information Technology and the U.S. Productivity
Revival: What Do the Industry Data Say?” Federal Reserve Bank of New
York Staff Reports. no. 115.
Tylecote, Mabel. 1957. The Mechanics Institutes of Lancashire and
Yorkshire before 1851. Manchester: Manchester University Press.
Tryon, F.G. 1927. “An Index of Consumption of Fuels and Water Power,”
Journal of the American Statistical Association. 22: 271-282.
United Nations. 1984. Industrial Statistics Yearbook 1984. New York,
United Nations.
U.S. Department of Commerce. 1975. Historical Statistics of the U.S.:
Colonial Times to 1970. Bicentennial Edition. Washington, DC: Bureau
of the Census.
U.S. Department of Commerce. 1986. Survey of Current Business.
Washington, DC: Bureau of Economic Analysis.
Van Reenen, John. 1996. “The Creation and Capture of Rents: Wages and
Innovation in a Panel of U.K. Companies,” Quarterly Journal of
Economics. 61, no.1:195–226.
5
Abstract
5.1. Introduction
The economist’s purview is, to a large extent, circumscribed and
consists of studying consumer and producer behavior, tastes and technology
being provided by the psychologist (social psychologist) and the production
engineer, respectively. The firm’s problem, it therefore follows, consists of
combining factor inputs to either minimize costs, or maximize profits.
Technology, as such, is assumed to be parametric. For over three decades
(1945-1980), this view pervaded economics. The growth and productivity
slowdowns of the mid-1970s, however, changed this (Maddison 1987). By
the early 1980s, it was clear that the trend rate in growth had shifted
downward, prompting economists in general, and “growth economists” in
particular, to take a closer look at the A (technology) scaler in pro-duction
functions. It would be reasonable to expect that at this point in time, the
profession would have turned to the experts, namely, production engineers.
However, instead of calling upon production engineers for insights into the
“growth and productivity slowdowns,” the profession took it upon itself to
78 5. Engineering and Economic Growth
1 The formalization notwithstanding, new growth theory offers little by way of new
ideas about growth per se. Early work by Paul Romer (Romer 1986, 1994) formalized
the notion of increasing returns to scale found in Edward Denison’s work on the
sources of economic growth. Recent work by Philippe Aghion and Peter Howitt
(Aghion and Howitt 1998) formalizes Joseph Schumpeter’s notion of creative
destruction.
Science and the Wealth of Nations: The Physics of Economic Growth 79
point of application of required forces in order to make tasks easier. The output of
useful work from any machine, however, can never exceed the total input of work
and energy (Betts 1989).” See also Beiser (1983).
Science and the Wealth of Nations: The Physics of Economic Growth 81
manufacturing data for the United States, Germany and Japan. Unlike
previous attempts at integrating energy into production theory, the energy-
organization framework provides a theoretically-consistent model of
production that does not violate the basic principles of classical mechanics
and basic production engineering.
This theory predicts that output growth is increasing in the rate of growth
of energy consumption growth, and variations in second-law efficiency.4
Energy and organization are, in general, complementary, and, as such,
collinear factor inputs. Increases in energy consumption will, in general, be
accompanied by increases in tools and supervision (organization). Variations
(growth) in second-law efficiency are the result of technological change.
That is, as a result of technological change, tools and supervision (organization)
are able to extract more work (value-added) out of a given amount of
energy.5 In general, output growth will be driven by energy consumption
growth, variations in second-law efficiency, and accompanied by a
corresponding increase in tools and supervision.
In the next section, this theory of production is tested against data for
U.S., German and Japanese manufacturing. As energy is assumed to be the
only physically productive factor input (as opposed to organizationally
productive), the theory predicts that the output elasticity for energy should
be greater than the output elasticity for either conventionally-defined capital
(tools) and/or conventionally-defined labor (supervision).
4 For the case in which second-law efficiency and energy consumption enter
multiplicatively, the rate of growth of output is simply the sum of the rate of growth
of energy consumption, and the rate of growth of second-law efficiency.
5 A good example of this is James Watt’s external condenser, which raised the
note that collinearity does not necessarily imply causality nor physical
productivity.
Despite these shortcomings, conventional testing of production
relationships can nonetheless yield interesting insights, notably with regard
to the relative contribution of energy, capital and labor. Standard production
theory (e.g. the KLEMS approach) views energy as a minor factor input,
relative to capital and labor. The energy-organization approach sees the
converse, namely that capital and labor are minor factor inputs (productivity-
wise), and energy as the prime factor input.
In this section, the energy-organization model of production is tested
indirectly, and the resulting output elasticities are used to reexamine the
productivity slowdown. Drawing from earlier work (Beaudreau 1995,
1998), it was shown that by widening the set of productive factor inputs to
include energy, specifically electric power consumption, and estimating the
output coefficients directlyas opposed to via cost datafundamentally-
different output elasticities are obtained, which, when used in standard
growth accounting exercises reduces the unexplained variation (i.e. the
Solow residual), to the point of explaining almost entirely the productivity
slowdown.
In earlier work (Beaudreau 1995, 1998), we estimated the energy-
organization production function using a modified KLEMS production
function, the KLEP (capital, labor and electric power) production function.
Electric power was used as a proxy for energy consumption given the
prevalence of electricity as the prime mover in modern manufacturing (see
Sonenblum 1990). Capital and labor were used as proxies for tools and
supervision.7 The resulting output elasticities, estimated directly using post
World War II input and output data for U.S., German and Japan
manufacturing, were then used in standard growth accounting exercises.
The relevant output elasticities were estimated directly (i.e. as opposed
to indirectly using cost data). Among the reasons for this is the absence of
competitive factor markets for both energy and labor (Blanchflower et al.
1996, Van Reenen 1996). Specifically, electric power markets are, in
general, regulated, via public utility commissions, while labor markets,
These estimates, one could argue, provide some support for the energy-
organization view of production presented above. Each of these output
elasticities, in conjunction with the laws of basic mechanics (i.e. regarding
tools), and the supervisory nature of labor, is consistent with the predictions
of the energy-organization model. As pointed out earlier, positive capital
and output elasticities have little theoretical relevance, beyond the collinear
nature of organization-related factor inputs.
These elasticities were then used to construct a fixed-weight aggregate
input index for each country, both before and after the energy crisis (circa
1974). Table 5.2 reports the relevant growth rates for manufacturing value
added (Q), electric power (EP), labor (L) and capital (K), as well as the
relevant fixed-weight aggregate input index (AI) (Hisnanick and Kymm
1992; Gullickson and Harper 1988) for three time intervals: 1950-1984,
1950-1973 and 1974-1984, for all three countries.
Table 5.2: Output and Input Growth Rates: U.S., German and
Japanese Manufacturing
average annual rate of 4.052 percent over the period 1950–1984. Per worker
consumption of electric power in U.S. manufacturing in this period goes
from 12, 534kwh in 1950 to 41, 688kwh in 1984, a total increase of 232
percent (U.S. Department of Commerce various years). When multiplied by
the relevant output elasticity (i.e. 0.537244), the growth in electric power
consumption in U.S. manufacturing accounts for 82% of overall output
growth (i.e. 2.17/2.68), which corroborates the predictions of energy
organization-based production analysis. Prior to the energy crisis (i.e.
1973), electric power consumption in manufacturing increased at an average
annual rate of 5.371%. Output increased at an average annual rate of
3.466%. Electric power consumption growth, it then follows, accounts for
83 percent of output growth (i.e. 2.88/3.46). In the post-energy-crisis period,
electric power consumption increased at an average annual rate of 0.246,
while output increased at an average annual rate of 0.121.
5.5 Conclusions
While this chapter has been somewhat critical of growth theory in its
current state, especially the lack of diversity in research programs, it
findings and conclusions should nonetheless be seen as prospective in
nature, providing an additional—we hesitate to use the word novel—
approach to thinking about and modeling growth. In other words, there is
room for competing and complementary approaches. For example, the
energy-organization approach to growth, used here, can be viewed as
complementary to general purpose technology (GPT) analysis (Bresnahan
Science and the Wealth of Nations: The Physics of Economic Growth 87
5.6 References
Aghion, P. and P. Howitt. 1998. Endogenous Growth Theory. Cambridge,
MA: MIT Press.
Alting, L., 1994. Manufacturing Engineering Processes. New York, NY:
Marcel Decker.
Beaudreau, B.C. 1995. The Impact of Electric Power on Productivity: The
Case of U.S. Manufacturing 1958-1984. Energy Economics. 17: 231-
236.
Beaudreau, B.C., 1996. “R&D: to Compete or to Cooperate.” Economics of
Innovation and New Technology. 4, 173–186.
Beaudreau, B.C. 1998. Energy and Organization: Growth and Distribution
Reexamined. Westport, CT: Greenwood Press.
Beaudreau, B.C. 1999. Energy and the Rise and Fall of Political Economy.
Westport, CT: Greenwood Press.
Beiser, A. 1983. Modern Technical Physics. Menlo Park, CA: The
Benjamin-Cummings Publishing Co.
Berndt, E., Wood, D.O. 1975. “Technology, Prices and the Derived Demand
for Energy.” The Review of Economics and Statistics: 259–268.
Betts, J.E. 1989. Essentials of Applied Physics. Englewood Cliffs, NJ.
Blanchflower, D.G., Oswald, A.J. and S. Peter. 1996. “Wages, Profits, and
Rent Sharing.” Quarterly Journal of Economics. 60: 227–251.
Bresnahan, T. and Trajtenberg M. 1992. General Purpose Technologies:
Engines of Growth? National Bureau of Economic Research Working
Paper No. 4148. Cambridge, MA.
Denison, E.F. 1962. The Sources of Economic Growth in the United States
and the Alternatives Before Us. Committee for Economic Development,
New York, NY.
88 5. Engineering and Economic Growth
Abstract
6.1 Introduction
The information and communications technology (ICT) revolution
remains an enigma. Three decades after the introduction of the microchip,
and 15 years after the birth of the internet, productivity, as measured by
output per worker, has increased, especially in the USA, and in the period
90 6. The Dynamo and the Computer
1995-2001, but by less than expected (Gordon 2000, 2006, Oliner and
Sichel 2000, Gust and Marquez 2000). In the 1980s and 1990s, it was
generally believed that ICTs would usher in a third industrial revolution,
and more importantly, a period of uninterrupted productivity growth, not
unlike the first two industrial revolutions, the first resulting from the steam
engine, and the second, from the dynamo. Both were associated with
extended periods of above-average growth, lasting decades:1
The advent of microcomputers since the beginning of the 1970s has brought
a new dimension to power electronics and drive technology. The impact of
this evolution is as significant as the advent of power semiconductor devices
in the 1950s. In the forthcoming industrial revolution, that is the
“computerized automation of factories,” microcomputers will not only
provide intelligence to higher-level factory automation, but will play a vital
role in the control of lower-level power electronics and motion control
systems. Microcomputers have now been accepted universally for the
control of power electronic and drive systems. It is interesting to note that
both the ends of the power-electronics spectrum are digital: one provides the
brain, and the other provides the muscle (Bose 1987).
That the ICT revolution has not, at least yet, produced anything
comparable to the growth rates in the 1920s (the roaring 1920s), has raised
a number of questions (Maddison 1987, David 1990, Gordon 2000, 2006,
Gust and Marquez 2000). Foremost among these has, without a doubt, been
why? Why have ICTs not, as was generally expected, increased overall
productivity, and, consequently, wealth? Related questions include, how
does the computer stack up against the steam engine, against the dynamo?
David (1990), in an article entitled The Dynamo and the Computer: An
Historical Perspective on the Modern Productivity Paradox, raised the
possibility of lengthy diffusion lags, not unlike those experienced by U.S.
manufacturers with the dynamo (electrical motor) in the late 19th/early 20th
century:
1 See Jorgenson (2001) for more on the ICT revolution and productivity in the USA.
Science and the Wealth of Nations: The Physics of Economic Growth 91
Figure 6.1, taken from Alting (1994), is the corresponding flow diagram.
In short, production processes are energy and information based. Capital
and labor, the workhorses of production theory in economics, are, at least
on the surface, not mentioned, at least not explicitly. However, as it turns
out, they play a crucial role in organizing energy and information. We will
return to this later.
Unlike standard production theory where energy and information enter
parametrically via the technology scaler (A in the Cobb-Douglas specification),
in engineering they constitute the two key factor inputs (Alting 1994, Beiser
1983). According to Figure 6.1, energy and information are what transform
materials. This raises a number of fundamental questions, notably, what is
the exact role of each? Are they both physically productive? Are they
comparable as Paul David and Dale Jorgenson maintain? Consider classical
mechanics and thermodynamics. According to classical mechanics, energy
being the source of force, is the basis of work. For example, work is defined
as the product of force and distance (W=fd). This principle carries over to
thermodynamics where energy, work and heat behave according to the
equation ܷ݀ = ߜܳ െ ߜܹ, where dU is the net increase in the internal
energy of a thermodynamic system, GQ is the heat entering the system and
GW is the work done by the system.3 The amount of useful work that can be
extracted from a thermodynamic system is determined by the second law of
thermodynamics. In essence, it states that not all energy can be transferred
into useful work. The better is the system, the more work can be extracted
4Useful energy (i.e. that which can be extracted from a given supply of energy) is
sometimes referred to as exergy or energy availability.
94 6. The Dynamo and the Computer
We may now pass to the effects which machinery has in relieving that
excessive muscular strain which a few generations ago was the common lot
of more than half the working men even in such a country as England in
other trades, machinery has lightened man’s labours. The house carpenters,
for instance, make things of the same kind as those used by our forefathers,
with much less toil for themselves. Nothing could be more narrow or
monotonous than the occupation of a weaver of plain stuffs in the old time.
But now, one woman will manage four or more looms, each of which does
many times as much work in the course of a day as the old hand loom did;
and her work is much less monotonous and calls for much more judgment
than his did (Marshall 1890, 218).
The next step is to relieve human of the task of guiding or controlling the
tool or machine. Tools or machines that control their own motions within
prescribed limits are said to be automatic; or by virtue of having control
devices connected to them, tools or machines are automated [.. .]. In many
instances, a tool or machine is automated to maintain product quality, A
hamburger grill may become too hot to produce quality hamburgers, unless
controlled. Likewise, the chemistry of an uncontrolled electroplating bath
may vary from the acceptable range, necessitation the stopping of production
while adjustments are made[. . .]. Automatic machines can often produce
Science and the Wealth of Nations: The Physics of Economic Growth 95
parts with greater accuracy, uniformity, and speed than can manually-
operated machines; they also relieve humans of some tedious, dangerous,
and hard jobs.
5Perhaps this explains why workers who use computers have higher earnings than
workers who do not (Kreuger 1993, Black and Lynch 1996).
96 6. The Dynamo and the Computer
1979-1995 1996-2000
EU-4 USA USA-EU EU-4 USA USA-EU
Total economy 0.33 0.46 0.12 0.53 0.86 0.33
ICT-producing 0.04 0.06 0.02 0.07 0.11 0.04
industries
Electrical and electronic 0.01 0.04 0.02 0.02 0.05 0.04
Equipment and 0.03 0.02 0.00 0.05 0.05 0.00
instruments
communications
ICT-using industries 0.21 0.28 0.07 0.35 0.57 0.22
ICT-using 0.02 0.02 0.01 0.03 0.03 0.01
manufacturing
Wholesale Trade 0.03 0.08 0.05 0.07 0.13 0.06
Retail trade 0.01 0.04 0.03 0.03 0.05 0.02
Financial intermediation 0.08 0.11 0.03 0.10 0.27 0.17
Business services 0.07 0.04 20.03 0.12 0.09 20.04
Non-ICT industries 0.08 0.11 0.03 0.11 0.18 0.07
Agriculture, forestry 0.00 0.00 0.00 0.00 0.00 0.00
and fishing
Mining and quarrying 0.00 0.01 0.01 0.00 0.00 0.00
Non-ICT manufacturing 0.04 0.04 0.00 0.04 0.05 0.01
Transport and storage 0.00 0.01 0.00 0.01 0.02 0.01
Social and personal 0.01 0.01 0.01 0.01 0.03 0.02
services
Non-market services 0.01 0.03 0.01 0.02 0.04 0.03
Other non-ICT 0.02 0.02 0.00 0.03 0.03 0.00
Source: Inklaar et al. (2005, 519)
control devices, firms reduce their costs, and increase their profits. The
automation and robotization literature contains numerous references to such
windfalls. It therefore stands to reason that the only place computers are
likely to show up is in labor (production and non-production worker) and
capital statistics, the result of firms substituting inanimate forms of
supervision (computers and control devices) in the place of animate (read:
human) forms. Labor input per unit output will decrease, while capital input
per unit output will increase, which is precisely what we observe in the data.
Output will not increase as a result; however, output per remaining labor–
predominantly, non-production labor–will increase.7
Evidence is provided by the output and input data for the U.S., German,
and Japanese manufacturing presented in Table 6.2 taken from Beaudreau
(1999). Output is measured in terms of real VA, while inputs include electric
power consumption, broadly defined labor (L), and capital (K). We see that
in all three cases, broadly defined labor has actually decreased from 1974 to
1988. Broadly defined capital, on the other hand, has increased at rates equal
to or greater than the rate of growth of VA. For example, in US
manufacturing, VA increased at an average annual rate of 0.121 percent
6.4 Conclusions
The purpose of this chapter was exploratory: draw from the laws of
classical mechanics and thermodynamics in search of new leads, new
perspectives in the debate over the modern productivity paradox. As pointed
out, historical comparisons such as Paul David’s and indeed those found in
the current growth literature, while extremely useful, have to rest on solid
theoretical grounds. Alikes have to be compared to alikes; apples cannot
and should not be compared to oranges. As we have shown in this note,
according to classical mechanics and thermodynamics, the dynamo and the
computer are fundamentally different, adding credence to Joel Mokyr’s
warning that: “Historical analogies often mislead as much as they instruct
and in technological progress, where change is unpredictable, cumulative,
and irreversible, the analogies [are] more dangerous than anywhere”
(Mokyr 1997, 35). One is, according to classical mechanics, the source of
all transformations in the universe, including value-added (wealth), while
the other is an organizational input.
Information, like tools, has never been, is not, and will never be
physically productive per se. This, we argue, provides an interesting
perspective on the modern productivity paradox. Computers have not shown
up in the productivity data for the simple reason that information is not
physically productive. Further, it corroborates Gordon’s (2000, 2006) claim
Science and the Wealth of Nations: The Physics of Economic Growth 101
to the effect that the ICT revolution and the second industrial revolution
appear to be, qualitatively and quantitatively, incomparable. Clearly, more
work on the role of information and indeed, on all other non-traditional
factor inputs (e.g. energy) in wealth-generating material processes is needed
if only to reassess and reevaluate growth-related public policy. If
information is not productive as originally thought, then investments in ICT
has to be reassessed as does growth policy in general. For the past three
decades, growth policy has been based on massive investment in ICT via
direct investment and various policy incentives (tax credits, etc.). Lastly,
our findings have important implications for the question of productivity at
the firm level. For years, ICT has been front and center in the productivity
debate. Firms have invested billions in ICT in the hope of increasing profits
and dividends. As we have argued, profits and dividends have increased but
for entirely different reasons, namely by reducing costs.
It is our view that more work along the lines suggested by our model and
findings is needed. Specifically, micro-models of the role of information in
production and management technology are needed. Such models are
needed to better evaluate the various ICT innovations at the firm, industry
and national level. In sum, the time has come to integrate ICT in a
meaningful way into production and management technology.
6.4 References
Alting, L. 1994. Manufacturing Engineering Processes. New York, NY:
Marcel Decker.
Babbage, C. 1832. The Economy of Machinery and Manufacturing.
London: C. Knight.
Beaudreau, B.C. 1999. Energy and the Rise and Fall of Political Economy.
Westport, CT: Greenwood Press.
Beiser, A. 1983. Modern Technical Physics. Menlo Park, CA: Cummings
Publishing Company.
Berndt, E. and Wood, D.O. 1975. “Technology, Prices and the Derived
Demand for Energy.” The Review of Economics and Statistics. Vol. 57:
259-68.
Betts, J.E. 1989. Essentials of Applied Physics. Englewood Cliffs, NJ:
Prentice-Hall.
Black, S.E. and Lynch, L.M. 1996. “Human Capital Investments and
Productivity.” American Economic Review: Papers and Proceedings.
May.
102 6. The Dynamo and the Computer
Oliner, S.D. and D.E. Sichel. 1994. “Computers and Output Growth
Revisited: How Big is the Puzzle?” Brookings Papers on Economic
Activity. 2: 273-317.
Oliner, S.D. and D.E. Sichel. 2000. “The Resurgence of Growth in the Late
1990s: Is Information Technology the Story.” Journal of Economic
Perspectives.14, no. 4: 3-22.
Triplett, J. and B. Bosworth. 2008. “The State of Data for Services
Productivity Measurement in the United States.” International
Productivity Monitor. 16: 53-71.
U.S. Department of Commerce. 2004. Bureau of Economic Analysis U.S.
Net Stock of Private Fixed Assets. Washington, DC.
7
Abstract
For over three decades, worldwide R&D expenditure has risen steadily,
reaching $1.3 trillion in 2011. Underlying this unprecedented growth is a
deeply-held belief that R&D is a prime mover of economic growth.
Ironically, despite three decades of massive R&D expenditure, growth levels
have remained substantially lower than that of the immediate post-World
War II period. This raises important theoretical questions regarding R&D
and its impact on growth per se. For example, R&D-growth has been
modeled and continues to be modeled as an unbounded set. This has not
been inconsequential because it has introduced an upward bias in growth
projections as evidenced in the literature. More importantly, are there
physically-determined upper limits to R&D-based growth and, if so, what
are they? This chapter uses the physical sciences to map the physical limits
to R&D-based innovation. A consilient model of economic growth is
presented and upper bounds for energy efficiency-based growth rates are
provided, both for individual energy sectors and globally. We find that with
economic growth by innovation being limited by physical conditions,
increasing the rate of economic growth can only come through increasing
the rate of energy consumption
7.1 Introduction
Virtually ignored in economics for most of the 20th century, research and
development (R&D) literally took off in the 1980s. R&D and innovation
took center stage in the debate over economic growth; theoretically,
empirically and policy-wise. Building on the pioneering work of Denison
(1962) and Griliches (1995) and the writings of Moses Abramovitz1 the
growth literature virtually exploded with articles about the myriad aspects
of innovation. Governments heeded the call and re-examined incentives,
industry structure and corporate taxation.
Now, some three decades later, massive spending on R&D (Table 7.1)
growth has not returned GDP2 growth to the levels in the 1950-1970s. Why
have record levels of R&D failed to jump-start growth? This chapter is an
attempt to provide answers to this and other R&D-growth related questions.
Our starting point is the physical R&D-growth boundary, defined as the
upper limits of R&D-based economic growth, which is defined by the
underlying physical laws. By the latter, it should be understood as the
fundamental laws of classical mechanics and thermodynamics as applied to
material processes. Among the problems associated with the dominant R&D
growth model is its unbounded nature. For example, the technology scalar
in the Solow-Swan growth model (Solow 1956, Guerrini 2005) is
theoretically unbounded from above, which is in clear violation of the first
and second laws of thermodynamics and of matter. In other words, R&D
cannot create matter or energy and cannot violate the laws of physics
(Beaudreau 1998, 1999).
Americas 491.8
US 427.2
Other 64.6
Asia 473.5
Japan 152.1
China 174.9
India 38
Europe 326.7
Rest of world 41.4
Total 1, 333.0
2GDP (gross domestic product) is the sum of all of the goods and services produced
and delivered. GDP is a measure of wealth and can be expressed as GDP/capita
when comparing the wealth of populations.
Science and the Wealth of Nations: The Physics of Economic Growth 107
there were two consecutive 50% increases in energy efficiency for the fuel
consumption of an automobile that was initially 10 l/km. The first 50% increase
allows the automobile to travel 150 km on 10 l of fuel and energy intensity would
be (10x(100/150))1/4 6.67 l/km. The second 50% increase in energy efficiency
allows the automobile to travel 150 km on 6.67 l of fuel. The energy intensity is now
(6.67x(100/150)) 1/4 4.44 l/km. The first 50% increase in energy efficiency reduced
fuel consumption by 3.3 l/km and the second by 2.2 l/km. Thus, the second 50%
increase in energy efficiency gives less actual reduction in energy intensity.
108 7. The Physical Limits to Economic Growth by R&D Funded Innovation
߮ ట
ݐݔ ݐܶ ݐݏ = ݐݍ 7.3
terms of Equation 7.4 where Wt, Et, Lt, and St refer to wealth, energy, tools
and supervision at time t, respectively and K refers to second-law efficiency,
which, as shown, is a function of Lt and St.
to that of Tables 6.4-6.7. Columns A and B are 1990 numbers for each sector.
Column C is the potential increase in energy efficiency as previously described in
116 7. The Physical Limits to Economic Growth by R&D Funded Innovation
the text. Column D is the decrease in energy intensity that corresponds to the
increase in energy efficiency. Column E is the percentage of each energy source in
the mix in 2100. Multiplying the values in Column D with those in Column E results
in Column F, which is the decrease in energy intensity for each source in the mix.
The individual decreases in energy intensity are added to obtain the total of 31.4%.
The residual energy intensity is 68.6% of that in 1990 and the equivalent average
increase in energy efficiency is 45.9%
A B C D E
U.S. trans. Potential Potential World trans. Energy
Energy increase in decrease in energy intensity
consumed in energy energy consumed in decrease
2012 (%) efficiency to intensity 2100 (%) 1990-
2100 (%) 1990-2100 2100C x D
(%) (%)
Cars and light 58.6 270 73.0 58.6 42.8
trucks
Trucks, ships, 29.6 50 33.3 29.6 9.9
trains
Air 8.0 100 50.0 8.0 4.0
Other 3.8 50 33.3 3.8 1.3
Totals 100.0 - - 100 57.9
Energy 57.9
intensity
decline from
1990 to 2100
Residual 42.1
energy
intensity in
2100
Corresponding 137.8
overall increase
in energy
efficiency
7.4.6 Co-Generation
Co-generation is defined as the recovery of thermal energy that is
normally lost or wasted. It is often difficult to obtain large and/or consistent
benefits from co-generation because normally waste heat cannot be stored
until needed. Thus, it is necessary to balance the amount and timing of the
loads between electricity generation and heat utilization.
Science and the Wealth of Nations: The Physics of Economic Growth 117
7.4.7 Transportation
Transportation in Table 7.2 is estimated to consume 18.6% of world
energy in 1990. Currently, 95% of U.S. transportation energy is supplied by
oil. The figures in Table 7.4 are for the U.S. in 2012 (Center for Climate and
Energy Solutions 2012, U.S. Department of Energy 2014) because
comparable figures for the world are not available. As such, they overstate
the contribution of cars and light trucks. However, it could be argued that
the rest of the world will tend towards a similar breakdown over the course
of this century.
Highway weight and size restrictions limit the size and weight of heavy
trucks, and relaxation of these restrictions is not likely. Improvements in
energy efficiency must come, therefore, from propulsion systems and
reductions in air and rolling resistance. For purposes of our analysis, the
maximum increase in energy efficiency for large trucks was estimated at
50%, which raises current efficiency from 45% to 67%. Trains have the
advantage of the low rolling resistance of steel wheels on steel rails. The
maximum increase in efficiency here is set at 50% because all of the
components of the propulsion system are well developed. Doubling the size
of a ship increases the carrying capacity by eight times, but the power
required increases by only four times. Thus, there is an energy advantage
for large ships as evidenced by the emergence of large supertankers, which
require special port facilities. The size of port and canal facilities, as it turns
out, limits the size of most ships. It is unlikely that this will change
significantly and, therefore, the average size of ships cannot increase
significantly. The efficiency of large ship propulsion systems from fuel to
propeller is currently at 42% (U.S. Department of Energy 2001). Large
increases in efficiency do not appear likely from improved hull shapes and
anti-fouling methods. Therefore, the maximum increase in energy
efficiency is set at 50% for purposes of this analysis.
7.4.7.3 Aircraft
Similarly, large aircraft are more efficient than smaller ones, which
explains the trend towards larger aircraft. However, limiting the size of
aircraft are route traffic patterns, runway construction and airport facilities.
With these constraints, it is unlikely that the average size of aircraft will
A B C D E
U.S. Potential Potential World Energy
residential increase in decrease in residential intensity
energy energy energy energy decrease
consumed in efficiency to intensity consumed in 1990-
2010 (%) 2100 (%) 1990-2100 2100 (%) 2100C x D
(%) (%)
Space heating 44.7 300 75 45 33.8
Water heating 16.4 50 33 17 5.7
Space cooling 9.2 200 12 8.0
Lighting 5.9 400 80 4 3.2
Refrigeration 3.9 300 75 4 3.0
Electronics 4.7 300 75 5 3.8
Wet cleaning, 3.3 100 50 3 1.5
drying
Cooking 3.7 300 75 4 3.0
Computers 1.5 100 50 2 1.0
Other 6.8 100 50 4 2.0
Totals 100.0 - - 100 64.9
Energy 64.9
intensity
decline from
1990 to 2100
Residual 35.1
energy
intensity in
2100
Corresponding 184.9
overall
increase in
energy
efficiency
120 7. The Physical Limits to Economic Growth by R&D Funded Innovation
7.6.5 Residential
Residential energy consumption in the U.S. is given in Table 7.5 (U.S.
Department of Energy 2012). The U.S. data is used as a base because it is
readily available and over the next century it is highly probable that world
residential energy consumption patterns will approach U.S. levels. The
values in Column D tend to reflect this.
The projected increases in energy efficiency are listed in Column B and
range from 50 to 300% (Letschert et al. 2013). Only a minor part of the
projected increase in energy efficiency for space heating and cooling derives
from increased efficiency of furnaces, heaters and air conditioners as energy
efficiency of these items is already well developed (Wada et al. 2012). For
example, the efficiency of older heating systems is 56-70% and the current
annual fuel utilization efficiency recommendation is 78-83%. The bulk of
the increase is from better windows, reduced air infiltration, better
insulation, and the replacement of less energy-efficient heating and cooling
systems. Energy efficiency increases of 50% for water heating will come
mainly from better storage tank and piping insulation and, possibly, from
combined space and water heating systems. The energy efficiency of
refrigerators has increased 300% since 1974 and most of the improvement
in energy efficiency is likely to come from replacement of old, less energy
efficient models. Lighting holds the potential for large increases in energy-
efficiency mainly through wider use of more efficient light bulbs.
Overall energy intensity for the residential sector is projected to decrease
by 64.9% to a level of 35.1%, which represents an increase in energy
efficiency of 184.9% from that in 1990.
7.6.6. Industrial
Table 7.6, Column A, is a breakdown of industrial energy consumption
for the U.S. in 2004. Once again, as similar information is not available for
the world, estimates were made of how energy might be used by all world
industry in 2100 (Column D). Based on historical experience and the fact
that the use of robots and automation is expected to increase, the estimates
for percentage of boiler fuel and process heating and cooling are reduced in
2100 and machine drives and facilities are higher.
The energy efficiencies of industrial boilers, process heating and
cooling, and machine drives are generally significantly higher than 50%
today. The projected increase in energy efficiency for these three categories
is 25%. Facilities include buildings, which may have a potential for energy
efficiency increase as high as 200%. “Other” is unknown and has been
Science and the Wealth of Nations: The Physics of Economic Growth 121
7.7 Commercial
Column A in Table 7.7 gives the breakdown for commercial primary
energy consumption for the U.S. in 2010 (U.S. Department of Energy 2010)
which, given the nature of the technology, appears to be reasonable for the
world. Estimated overall energy intensity for commercial can decrease by
43.7% to a residual of 56.3%, which is equivalent to an increase in energy
efficiency of 77.6% over that of 1990.
A B C D E
U.S. Potential efficiency to Potential World
industrial increase in 2100 (%) decrease in industrial
energy energy energy energy
consumed in intensity consumed in
2004 (%) 1990-2100 2100 (%)
(%)
Boiler fuel 35 25 20 20 4
Process heat 39 25 20 20 4
and cool
Machine drive 12 25 20 25 5
Facilities 8 200 66.7 15 10
Electro- 2 25 20 0 0
chemical
Other 4 200 66.7 20 13.3
Totals 100 - - 100 36.3
Energy 36.3
intensity
decline from
1990 to 2100
Residual 63.7
energy
intensity in
2100
122 7. The Physical Limits to Economic Growth by R&D Funded Innovation
Corresponding 57.1
overall
increase in
energy
efficiency
A B C D E
U.S. Potential efficiency to Potential World
commercial increase in 2100 (%) decrease in commercial
energy energy energy energy
consumed intensity consumed in
in 2010 (%) 1990-2100 2100 (%)
(%)
Lighting 13.6 100 50 15 7.5
Space heating 26.6 100 50 25 12.5
Space cooling 10.1 50 33.3 10 3.3
Ventilation 6.1 50 33.3 6 2.0
Refrigeration 4.5 200 66.7 5 3.3
Water heating 6.7 50 33.3 7 2.3
Electronics/co 5.4 200 66.7 6 4.0
mputers
Cooking 2.3 50 33.3 5 1.7
Other 24.6 50 33.3 21 0.7.0
Totals 100 - - 100 43.7
Energy 43.7
intensity
decline from
1990 to 2100
Residual 56.3
energy
intensity in
2100
Corresponding 77.6
overall
increase in
energy
efficiency
Science and the Wealth of Nations: The Physics of Economic Growth 123
A B C
Portion of world Decrease in energy Contribution to
energy consumption in intensity by sector to decrease of each
1990 (%) 2100 (%) energy sector to world
energy intensity A x B
(%)
Electricity 37.5 31.4 11.8
generation
Transportation 18.6 57.9 10.8
Residential 12.1 64.9 7.9
Commercial 21.9 36.3 7.9
Industrial 9.9 43.7 4.3
Totals 100 - 42.7
World primary 57.3
energy intensity
as a percent of
1990
Equivalent 74.5
increase in world
energy efficiency
7.7 Conclusions
In this chapter, we examined the question of the physical R&D-Energy
Efficiency-Growth upper boundary. Until now, there have been no physical
constraints imposed on economic growth based on R&D as R&D-based
growth was, for the most part, assumed to be unbounded. Using basic
physics, we examined the physical limits to non-energy consumption-based
growth, i.e., the physical limits to energy efficiency. Our conclusions are:
These findings are consistent with the laws of physics that underlie the
material processes that constitute the world economy. They clearly indicate
that R&D alone cannot restore the high levels of growth experienced in the
post-WWII era up to the mid-1970s.
7.8 References
Abdih Y, F. Joutz. 2006. Relating the Knowledge Production Function to
Total Factor Productivity. International Monetary Fund Staff Papers.
Armstrong P, D. Agir, I. Thompson, and M. McCulloch. 2014. “Improving
the Energy Storage Capability of Hot Water Tanks Through Wall
Material Specification.” Energy. 78:128-40.
Ayres R, H. H. Turton H, and T. Casten. 2007. “Energy Efficiency,
Sustainability and Economic Growth.” Energy 32:634-48.
Beaudreau B.C. 1998. Energy and Organization: Growth and Distribution
Reexamined. Westport, CT: Greenwood Press; 1998.
Beaudreau B.C. 1999. Energy and the Rise and Fall of Political Economy.
Westport, CT: Greenwood Press; 1999.
Burman E, D. Mumovic, and J. Kimpian. 2014. Towards Measurement and
Verification of Energy Performance under the Framework of the
European Directive for Energy Performance of Buildings.” Energy.
77:153-63.
Caballero R. and A. Jaffe. 1993. “How High are the Giant's Shoulders: An
empirical Assessment of Knowledge Spillovers and Creative Destruction
in a Model of Economic Growth.” NBER Macroeconomics Annual
Cambridge, MA: MIT Press.
Center for Climate and Energy Solutions. 2012. Annual Report.
Chang M.C. 2014. “Energy Intensity, Target Level of Energy Intensity, and
Room for Improvement in Energy Intensity: An Application to the Study
of Regions in the EU.” Energy. 67:648-55.
Choi J.K, D. Morrison, K. Hallinan, and R. Brecha. 2014. “Economic and
Environmental Impacts of Community-Based Residential Building
Energy Efficiency Investment.” Energy.78:877-86.
126 7. The Physical Limits to Economic Growth by R&D Funded Innovation
Denison E. 1962. The Sources of Economic Growth in the United States and
the Alternatives Before Us. Committee for Economic Development.
Washington, DC.
Giacone E. and S. Manco. 2012. “Energy Efficiency Measurement in
Industrial Processes.” Energy. 38:331-45.
Green, M.A., K. Emery, Y. Hisikawa and W. Warta. 2007. Solar Cell
Efficiency Tables (Version 30) Progress in Photovoltaics. 15:425-30.
Griliches Z. 1995. “The Discovery of the Residual: An Historical Note.”
National Bureau of Economic Research Working Paper 5348.
Guerrini L. 2005. The Solow-Swan with Abounded Population Growth
Rate. Journal of Mathematical Economics. 42, no. 1:14-21.
Hall C and K. Klitgaard. 2011. Energy and the Wealth of Nations:
Understanding the Biophysical Economy. New York, NY: Springer
Press; 2011.
International Energy Agency. 2008. Report OECD/IEA.
Johansson B. 2013. “A Broadened Typology on Energy and Security.”
Energy. 53: 199-205.
Jones C. 1995 “R&D-Based Models of Economic Growth.” Journal of
Political Economy.103: 759-84.
Krusell P. 1998 “Investment-Specific R&D and the Decline in the Relative
Price of Capital.” Journal of Economic Growth. 3:131-41.
Ladu M, and M. Meleddu. 2014. “Is There Any Relationship Between
Energy and TFP (Total Factor Productivity)? A Panel Cointegration
Approach for Italian Regions.” Energy 75:560-7.
Letschert V. L.B. Derroches KeJ, M. McNeil. 2013. “Energy Efficiency:
How Far Can We Raise the Bar? Revealing the Potential of Best
Available Technologies.” Energy. 59:72-82.
Lightfoot H.D. 2006. “A Strategy for Adequate Future World Energy
Supply and Carbon Emission Control.” in: Climate Change Technology
Conference: Engineering Challenges and Solutions in the 21st Century.
Engineering Institute of Canada, Ottawa, Canada.
Lightfoot H.D. 2007. “Understand the Three Scales for Measuring Primary
Energy and Avoid Errors.” Energy. 32:1478-83.
Lightfoot H.D. W. Manheimer D.A. Meneley, D. Pendergast and G.
Stanford. 2006. “Nuclear Fission Fuel is Inexhaustible.” in: Climate
Change Technology Conference: Engineering Challenges and Solutions
in the 21st Century. Engineering Institute of Canada, Ottawa, Canada.
Maddison A. 1987. “Growth and Slowdown in Advanced Capitalist
Economies: Techniques of Quantitative Assessment.” Journal of
Economic Literature. 25, no.2:649-98.
Science and the Wealth of Nations: The Physics of Economic Growth 127
Abstract
8.1. Introduction
Energy-based theories/explications of the productivity slowdown
attribute the marked decrease in output growth in the 1970s to the OPEC
crisis and the ensuing energy price increases (Baily 1982, Jorgenson 1981,
1983). There are however a number of problems with these theories. First,
there is the question of timing. Oil prices increased in 1973, but electricity
prices remained stable well into the late 1970s. So, why did electricity use
growth rates fall in 1973? Second, there is the question of expectations.
Were higher energy prices perceived of by firms as temporary or permanent? If
130 8. The Economies of Speed, KE=1/2mv2 and the Productivity Slowdown
industrialized economies in the 1970s may have been demand, not supply,
based. Put differently, physical constraints on machine speeds may have put
an end to energy deepening, resulting in a lower rate of growth of output.
We end with a discussion of the possible implications of our findings.
9398. State why you believe that the labour of those employed has doubled
since the first introduction and use of cotton machinery, or at least since you
first knew it?The reason why I believe so is from some calculations which
I have been obliged to make, and by my own observation during the time I
was manager of a mill in 1830 and 1831, when I had some of the same
operations under my own observation.
9399. Have you any objection to put in those calculations Certainly not. [The
following document was then put in and read.]
9400. It appears by this document that the work done is very greatly
increased between the years 1810 and 1832; has the machinery been so
altered as to produce that amazing difference, or does it result from
accelerating the speed of the machinery?—It is from accelerating the speed
generally; and another cause is, that more and more exertion is required from
the individual working at the machine; these are the two causes ([33]; 430).
Mass production is not merely quantity production for this may be had with
none of the requisites of mass production. Nor is it merely machine
production, which also may exist without a resemblance to mass production.
Mass production is the focusing upon a manufacturing project of the
principles of power, accuracy, economy, system, continuity and speed (Ford
1926)
Industrial processes have been speeded up, new inventions are being added
to manufacturing, new economies of personnel and of management have
been made in industry. The 1929 production can undoubtedly be achieved
with thousands, and probably millions, fewer workers (Laing 1933).
com/watch?v=CUYajEF7X-U.
3 Alfred Chandler's economies of speed are consistent with Dale Jorgenson and
During these years, the focus of managerial attention shifted from enlarging
the scale of operations to increasing operating efficiency by speeding up the
rate of throughput in the plant. High priority was assigned to modifications
of factory design and layout in order to better integrate worker and machine
tasks. Advances in the electrification of machine drive were indispensable
to the realization of these new objectives and may, indeed, have served to
stimulate the new managerial perspectives that emerged (Sonenblum 1990,
291)
4 The nature of control technologies changed in the 1980s and 1990s. Instead of
focusing on minimizing downtime and increasing average machine speed, they
focused on automating supervision, rendering labor expendable. In others words,
they ushered in the era of the intelligent plant/factorythat is, a factory without
workers.
Science and the Wealth of Nations: The Physics of Economic Growth 135
8.3 Model
Surprisingly, little is known of the role of energy in material processes
in general, and in productivity in general. Typically, it has been modeled as
another input in a multi-input Cobb-Douglas, or constant elasticity of
substitution (CES) production function (Beaudreau 1998, 1999, Kummel
1982). Accordingly, firms are free to either increase of decrease the amount
of energy deployed within a given plant (or material process) at will—that
is, without limit. Left unspecified are the underlying mechanics (i.e.
microfoundations) of say a higher energy-capital ratio, other than the
standard reference to the laws of diminishing marginal productivity. This
raises a number of questions regarding energy use in manufacturing
processes. Can it be increased at will, and what the physical laws that govern
its use in? In this section, we attempt to answer these questions by
introducing the laws of physics, specifically, the laws of machine kinetics
into a standard model of production.5
5 In so doing, we are attempting to shed light on the black box of not only
neoclassical economics, but of energy-augmented models of production.
6 The quantity theory of money can be seen as a special case, where V, the velocity
of money, is defined as the speed at which money is exchanged. M, the money stock,
can be seen as the capital in this case. Hence, the greater is V, the more productive
is the money stock.
7 This equation is derived from special relativity, when v, the velocity, is much less
than c, the speed of light. The Lorentz factor allows us to derive KE = 1/2mv2 from
E = mc2.
8 Machine speed can be defined either in time or in number of product (i.e. maximum
speed is 10s per product, maximum speed is six products per minute).
136 8. The Economies of Speed, KE=1/2mv2 and the Productivity Slowdown
Machine downtime in turn can be broken down into (i) maintenance (ii)
idleness due to lack of coordination between sub¬processes (iii) retooling
for a new product and (iv) market conditions. To capture these effects, we
define machine average speed in terms of Equation 8.2 where ߛ ( )ݐis a
downtime scaler. If machines are never idled owing to any one or a
combination of factors (e.g. maintenance, market conditions), then ߛ ( )ݐis
zero. However, the more a machine is shut down as opposed to slowed down
the greater will ߛ ( )ݐbe [0< ߛ ( < )ݐ0.5].9 Equation 8.3 defines maximum
sub-process i machine speed.
ݏ[݊݅݉ = )ݐ(ݕଵ (݇)ݐଵ , ݏଶ (݇)ݐଶ ()ݐ, ݏଷ (݇)ݐଷ ()ݐ, … , ݏ (݇)ݐ ( ])ݐ8.1
ݏ (݁ = )ݐ ()ݐ.ହାఊ (௧) 8.2
Assume that ݏ ( )ݐis bounded from above. In other words, for each sub-
process, there exists a maximum speed beyond which it is impossible to
go.10 It will be defined as a combination of (i) translational/rotational
kinetics (ii) material tolerances and (iii) minimal downtime (owing to
maintenance, retooling, etc).
9 Consider the following example. Suppose that a firm has two machines, both of
which operate according to the law of kinetics. Suppose that demand is halved in
which case it can choose to either slow both down by 50 percent, or shut one of the
two machines down. In the former case, will be zero and the law of kinetics will be
in force. However, in the latter case, as one of the two machines will be shut down,
output will be linear in the energy input (i.e. homogeneous of degree one). When it
restarts the second machine (i.e. increases energy consumption), output will double.
Hence, in this case ߛ ( )ݐwill be non-zero
10 ݏ = )ݐ( ݏҧ ( )ݐcan be viewed as a combination of the asymptote of = )ݐ( ݏ
݁ ()ݐ.ହାఊ (௧) , and the physical upper limits of machine speed.
11 In this case, machine speed defines total factor productivity or capital productivity.
This would continue to hold if labor was included. It is important to keep in mind
Science and the Wealth of Nations: The Physics of Economic Growth 137
tools, providing mechanical advantage, but not being a source of energy per se.
Interestingly, labor as a factor input is ignored altogether. See Alting (1994) and
Beiser (1983).
138 8. The Economies of Speed, KE=1/2mv2 and the Productivity Slowdown
16 No standard errors or t-statistics are provided in Kummel et al. (2002) and Stresing
et al. (2008).
17 For indirect estimates, see Berndt and Woods (1975). Indirect estimates subsume
perfect competition in all factor markets. Direct estimates, on the other hand, make
no such assumption.
18 Beaudreau (1999) conducted a simple growth accounting exercise using these
output elasticities as well as input-output growth rates before and after 1973. He was
able to show that factor input growth (energy, capital and labor) accounted for
almost all of the variation in manufacturing output growth. See the results in the
Appendix.
Science and the Wealth of Nations: The Physics of Economic Growth 139
Country Constant EP K L R2
U.S. 1.1678 0.4902 0.7826 0.1034 0.9990
(1958-1984) (0.5342) (3.169) (0.9005) (5.136) ȡ=0.688
Germany -0.2128 0.6124 0.5587 -0.1086 0.9494
(1963-1988) (0.1961) (3.710) (0.5520) (3.2171) ȡ=0.557
Japan -3.1852 0.6970 0.5175 0.3789 0.9818
(1965-1988) (1.328) (2.514) (2.383) (0.9804) ȡ=0.885
Canada -3.6435 0.2451 0.2655 1.2673 0.9844
(1962-1988) (2.421) (0.8631) (1.216) (3.070) ȡ=0.660
Britain 1.2630 0.9141 -0.0972 -0.0870 0.9232
(1963-1988) (0.936) (5.762) (4.000) (0.7156) ȡ=0.502
Finland -0.4482 0.7081 0.3040 0.0583 0.9856
(1963-1988) (0.6075) (4.772) (1.702) (0.3133) ȡ=0.364
140 8. The Economies of Speed, KE=1/2mv2 and the Productivity Slowdown
Country Constant EP K L R2
U.S. -64.7637 0.4554 0.0963 1.1141 0.9891
(1958-1984) (3.186) (4.009) (1.084) (5.417) ȡ=0.663
Germany -0.2128 0.6124 -0.1086 0.5587 0.9994
(1963-1988) (0.1961) (3.710) (0.5222) (3.217) ȡ=0.557
Japan -164.97 0.5618 0.2374 1.3221 0.9820
(1965-1988) (1.814) (2.983) (3.683) (1.643) ȡ=0.827
Canada -130.442 0.3426 0.1582 1.6709 0.9821
(1962-1988) (2.347) (1.210) (6.2092) (2.836) ȡ=0.651
Britain 17.1535 0.8570 -0.0289 -0.0459 0.911
(1963-1988) (0.3206) (5.605) (0.1008) (0.1720) ȡ=0.583
Finland -17.9512 0.57475 0.3122 0.1854 0.9918
(1963-1988) (0.8936) (6.931) (2.553) (0.9723) ȡ=0.294
Country Constant EP K L R2
U.S. 0.0447 0.3824 0.9839 0.9646 0.9179
(1958-1984) (3.168) (2.498) (2.9561) (6.499)
Germany -0.0016 0.6093 0.0650 0.4675 0.6682
(1963-1988) (0.1469) (3.682) (0.2197) (1.872)
Japan 0.0360 0.7176 0.3052 0.6291 0.6710
(1965-1988) (1.930) (2.942) (1.128) (1.228)
Canada 0.0572 0.0528 -0.9138 1.4123 0.715
(1962-1988) (2.391) (0.2072) (1.933) (4.244)
Britain -0.0118 0.7898 -0.2098 0.0891 0.6432
(1963-1988) (0.9165) (5.122) (0.4099) (0.5597)
Finland 0.0139 0.7066 -0.1171 -0.3518 0.48
(1963-1988) (0.371) (4.364) (0.1352) (0.8735)
19 Output, labor and electricity use data were obtained from United Nations
Industrial Statistics Yearbook (1960-1988) (United Nations 1984); capital data were
obtained from OCED, Flows and Stocks of Fixed Capital 1989 (OECD 1989). These
data are available from the author upon request.
20 The variables used were rva, real value added, ep electric power use in kwh, k
capital stock, and l, hours workers. The estimated equations were: in the linear case,
rva(t) = a0 + a1ep(t) + a2k(t) + a3l(t); in the log-linear case, lnrva(t) = a0 + a1lnep(t)
Science and the Wealth of Nations: The Physics of Economic Growth 141
squares with an autoregressive error of order one was used in the case of the
first and second, while simple OLS was used in the third. The results are
presented in Tables 8.2-8.4, where we see estimates of the EP (electric
power use) coefficients in the predicted range of 0.50+Ȗ(t). For example, in
the per unit of capital on electric power per unit capital and labor per unit
capital. The results were however not significantly different.21
We then proceeded to test the model using 2-digit SIC data for U.S.
manufacturing from 1947 to 1984. The data in this case were taken from the
U.S. Annual Surveys of Manufactures (1947-1984) as well as the U.S.
Census of Manufactures (1947-1984). The results are presented in Table
8.5, where two sets of electric power output elasticities are reported.
Column 1 contains the electric power consumption output elasticity when
both electric power consumption (purchased and generated) and production
workers were included as independent variables, while Column 2 contains
the output elasticity when only electric power consumption was used as the
independent variable. As was the case with aggregate data, the output
elasticities are once again centered around the predicted-by-the-law-of-
kinetics value of 0.50. Column 3 reports the R2 for the latter case. What is
noteworthy is the extent to which variations in electric power consumption
explain roughly 80 percent of the variation in value-added by 2-digit SIC
industry.
Our model predicts that output per unit of energy input (i.e. in this case,
electric power), will be decreasing in the use of energy per unit of capital.
Hence, greater electric power use per unit of capital (machinery and
equipment) will result in lower incremental output. Once machinery and
equipment reaches its maximum speed, then output per unit energy input
should stabilize. Figure 8.3 presents the ratio of value-added per kwh for 20
U.S. manufacturing industries from 1947 to 1984. What is immediately
obvious is (i) the similarity of these results and those of Baily and
Chakrabarti (1988) reported earlier, and (ii) the fact that electric power
productivity appears to stabilize in virtually all industries after 1970. This
leads us to conclude that the energy deepening that had characterized the
post-WWII period appeared to have come to an end in the latter half of the
60s, well ahead of the OPEC-induced energy crises.
Science and the Wealth of Nations: The Physics of Economic Growth 145
22 Most of the SR-71 fleet has now been retired, except for two Blackbirds currently
on loan to NASA's Dryden Flight Research Center where the aircraft are being used
as "test beds" for high altitude research.
23 A good illustrative example is provided by supersonic air travel, more specifically,
the joint British/French Concorde that flew at Mach II. Energy costs were
astronomical, which in addition to the problems associated with the associated sonic
boom, did the venture in. Today, commercial jet aircraft travel at speeds which in
many cases are lower than those in the 1970s, owing principally to the associated
energy costs.
Science and the Wealth of Nations: The Physics of Economic Growth 147
*The four data points plotted here refer to the four periods Industry 1948-1965,
1965-1973, 1973-1979, 1979-1985.
24 It was suggested that perhaps the OPEC oil embargo precipitated the decline in
energy use. We remain skeptical as, if nothing else, higher energy prices will,
assuming that machine speed can be increased, result in further profit-increasing
speed ups.
25 These ratios were obtained using the same data.
Science and the Wealth of Nations: The Physics of Economic Growth 149
26 This is known as Arrenius’ Law, according to which the rate of a chemical reaction
rotation of an object and is part of its total kinetic energy. Looking at rotational
energy separately around an object's axis of rotation, one gets the following
dependence on the object's moment of inertia: Erotational=1/2iȝ2 , where ȝ is the
angular velocity i is the moment of inertia around the axis of rotation E is the kinetic
energy. The mechanical work required for/applied during rotation is the torque times
the rotation angle. The instantaneous power of an angularly accelerating body is the
torque times the angular velocity. For free-floating (unattached) objects, the axis of
rotation is commonly around its center of mass. Note the close relationship between
the result for rotational energy and the energy held by linear (or translational)
motion: Etranslational=1/2mv2. In the rotating system, the moment of inertia, i, takes
the role of the mass, m, and the angular velocity, ȝ, takes the role of the linear
velocity, v. The rotational energy of a rolling cylinder varies from one half of the
150 8. The Economies of Speed, KE=1/2mv2 and the Productivity Slowdown
In short, kinetic energy increases at the square of velocity. Hence, the cost
of continuous Àow process speed-ups will increase exponentially over time.
Other factors limiting speed-ups are factors like material tolerances, bearing
speeds, bottlenecks, etc.
translational energy (if it is massive) to the same as the translational energy (if it is
hollow). Source: http://en.wikipedia.org/wiki/Rotational-energy.
28 This is surprising given the important role of machine speed in productivity
growth.
Science and the Wealth of Nations: The Physics of Economic Growth 151
29 What eventually did them in was their sizemore specifically, their width.
152 8. The Economies of Speed, KE=1/2mv2 and the Productivity Slowdown
8.7.2 Transportation
The transportation sector provides what we feel is the best example of
the principles at play in this chapter and indeed throughout the post-WWII
period. A motor vehicle provides transportation services. The greater the
speed at which it travels, the greater its productivity. Doubling the speed of
30 No mention is made of speed-ups which in this case consist of increasing the
a truck, for example, will double the number of miles traveled, and hence,
will double its supply of transportation servicesits productivity. Second,
the fewer hours the truck in question is idle (e.g. sitting in the yard), the
greater is its productivity. The post-WWII period, we maintain, witnessed
substantial increases in vehicle productivity, increases that owed primarily
to increased speed and more efficient utilization of rolling stock. By the late
1960s/early 1970s, most of these gains had been achieved. Maximum
speeds had been reached, and downtime (idle time) had been reduced
considerably. A good example of a speed-increasing technological change
in the post-WWII period was the introduction of container shipping or
containerization. Containerization increased productivity markedly. It did
so in two ways, first by increasing the productivity of ports. Less time was
needed to unload large cargo ships (i.e. lower Ȗ(t)). Second, owing to the
latter, ship downtime was reduced considerably so that instead of lying idle
in port, they could be providing services, which in this case consists of
“transporting” cargo. Conceptually, this corresponds to an increase in the
relevant si(t) (see Equation 8.1). It is important to note that this was a one-
shot occurrence. Once minimal port time is reached, no further gains can be
achieved.31
31 Port productivity, like that of the retail sector, is increasing in the speed at which
merchandise moves through the port itself. The greater the turnover, the greater the
productivity. In the retail sector, this is measured in terms of average shelf life and
sales per square foot. The lower is shelf life, the greater is store productivity, and the
greater are sales per square foot, the greater is productivity.
154 8. The Economies of Speed, KE=1/2mv2 and the Productivity Slowdown
8.8 Implications
There are limits in the physical world. And non-linearities. One such
non-linearity is the relationship between speed and energy use. In an ideal
156 8. The Economies of Speed, KE=1/2mv2 and the Productivity Slowdown
world, the relationship between the two would be linear. Doubling energy
would double speed, and in so doing double output. Unfortunately, this is
not the case. As argued, most production processes have presently reached
their maximum operating speed (both physically-viable and economically
viable). Large speed-based productivity gains were achieved in the post-
WWII period, owing to greater machine speeds and to new management
techniques (e.g. reprogrammable control technologies) and exhausted by the
mid-to-late 1960s. In this section, we examine some of the implications of
our findings.
Figure 8.9: History of Clock Rate and Power for Intel x86
Microprocessors
32 A good illustration is provided again by the Concorde which doubled the speed
of air travel to Mach II, but at an exorbitant cost ($5000 versus $800), one that
eventually did the project in.
33 Our results also rationalize Olson's (1988) findings to the effect that post-WWII
Oliner and Sichel (2002) and others have argued that the higher
productivity rates observed in the late 1990s and early 2000s owed in large
measure to productivity gains in the ICT sector. This, we maintain, is
consistent with our findings. As with all new industries/sectors, the
associated sub-processes and processes benefit from important machine
speed-ups and better control technologies, which has the effect of increasing
overall productivity. However, like all other industries, these gains diminish
over time, and eventually disappearthat is, once maximum machine speeds
have been reached.
8.10 Appendix
Table 8.10: Output and Input Growth Rates: U.S., German and
Japanese Manufacturing
8.11 References
Alting Leo. 1994. Manufacturing Engineering Processes. New York, NY:
Marcel Decker Inc.
Baily N, Chakrabarti A.K. 1988. Innovation and the Productivity Crisis.
The Brookings Institution.
Baily Martin Neil. 1982. The productivity slowdown by industry. Brookings
Papers on Economic Activity. 2:423-59.
Beaudreau Bernard C. 1995. “The Impact of Electric Power on Productivity:
the Case of U.S. Manufacturing 1958-1984.” Energy Economics. 17,
no.3:231-6.
Beaudreau Bernard C. 1998. Energy and Organization: Growth and
Distribution Reexamined. Westport, CT: Greenwood Press.
Beiser Arthur. 1983. Modern Technical Physics. Menlo Park, CA: The
Benjamin/ Cummings Publishing Company.
Berndt Ernst, Wood David. 1975. “Technology, Prices and the Derived
Demand for Energy.” Review of Economics and Statistics: 259-69.
Chandler Jr Alfred D. 1977. The Visible Hand, the Managerial Revolution
in American Business. Cambridge, MA: Harvard University Press.
Cohen Avi J. 1984. “Technological Change as Historical Process: the case
of the U.S. Pulp and Paper Industry, 1915-1940.” Journal of Economic
History. 44, no.3:775-99.
Devine Warren D. 1990. “Electricity in Information Management: the
Evolution of Electronic Control.” in: Schurr Sam H. et al., eds.
Electricity in the American Economy. Westport CT: Greenwood Press.
Feyrer J. 2011. “The U.S. Productivity Slowdown, the Baby Boom and the
Management Quality.” Journal of Population Economics. 24, no. 1:267-
84.
Fisher Stanley. 1988. “Symposium on the Slowdown in Productivity
Growth.” Journal of Economic Perspectives. 2, no. 4:37.
Ford, Henry. 1926. Mass Production. Encyclopaedia Britannica.
Giraud Gael and Kahraman Zeynep. 2014. “How Dependent is Growth from
Primary Energy? Output Energy Elasticity in 50 Countries.” Paris
School of Economics. Working Paper.
Gordon Robert J. 2012. “Is Economic Growth Over? Faltering Innovation
Confronts the Six Headwinds.” NBER Working Paper No. 18315.
Greenberg Leon. 1964. Productivity and Technological Change in the
Petroleum Refining Industry. U.S. Senate, Washington: Select
Committee on Small Business.
Griliches Zvi. 1994. “Productivity, R and D, and the Data Constraint.”
American Economic Review. 84, no. 1:1-23.
162 8. The Economies of Speed, KE=1/2mv2 and the Productivity Slowdown
A KINETICS-BASED APPROACH
TO PRODUCTION:
THEORY AND EVIDENCE*
Abstract
Production theory has, over the course of the past two centuries, been
besieged by criticism, ranging from its weak fundamentals to its lack of
coherence with the physical sciences. Yet, no real alternatives have emerged
with the result that neoclassical production theory stands today, much as it
did over a century ago. This chapter presents a consilient theory of
production, one that is grounded in classical mechanics, is empirically
validated, and sheds light on myriad productivity-related phenomena.
Specifically, a two-tiered approach to modeling production is proposed. In
the first tier, a kinetics-based theory is developed where output is an
increasing function of energy consumption, in keeping with basic physics.
In the second tier, the organization of energy-based material processes (the
first tier) is modeled. The resulting model is estimated using U.S.
manufacturing two-digit SIC data from 1947 to 1989.
9.1 Introduction
Economics is, by definition, the science of wealth, a fact that places
wealth at the center of the analysis. Understanding wealth is, as such, the
key to, and basis of economics as an intellectual endeavor. Getting wealth
right, it therefore follows, is a sine qua non condition for success in all other
sub-fields (e.g. labor economics, macroeconomics, industrial organization).
Unfortunately, despite over two centuries of effort, wealth and its creation
remain a challenge, as evidenced by the many puzzles and paradoxes that
characterize the literature, including the decades-old productivity slowdown
and the current information paradox.
This raises the question of scientific validity, namely when is a hypothetical
model of wealth valid or correct? Clearly, it can be internally valid (i.e. to
the profession) if it is accepted by the majority. Take, for example, classical
166 9. A Kinetics-Based Approach to Production
When in 1735, John Wyatt brought out his spinning machine, and began the
industrial revolution of the 18th century, not a word did he say about an ass
driving it instead of a man, and yet this part fell to the ass. He described it
as a machine to spin without fingers.
All fully developed machinery consists of three essentially different
parts, the motor mechanism, the transmitting mechanism, and finally the tool
or working machine. The motor mechanism is that which puts the whole in
motion. It either generates its own motive power, like the steam-engine, the
caloric engine, the electromagnetic machine, etc., or it receives its impulse
from some already existing natural force, like the water-wheel from a head
of water, the wind-mill from wind, etc. The transmitting mechanism,
composed of fly-wheels, shafting, toothed wheels, pullies, straps, ropes,
bands, pinions, and gearing of the most varied kinds, regulates the motion,
changes its form where necessary, as for instance, from linear to circular,
and divides and distributes it among the working machines. These two first
parts of the whole mechanism are there, solely for putting the working
machines in motion, by means of which motion the subject of labour is
seized upon and modified as desired. The tool or working machine is that
part of the machinery with which the industrial revolution of the 18th
century started. And to this day it constantly serves as such a starting-point,
whenever a handicraft, or a manufacture, is turned into an industry carried
on by machinery. (Marx 1867, Chapter 15).
Clearly, there was more to Marx’s thought than the simple labor theory
of value. In fact, one could argue that he was well aware of classical
mechanics and the role of force in material processes, not to mention the
role of tools in material processes.
Perhaps the most influential of 19th-century iconoclastsin large part,
much in spite of himselfwas William Stanley Jevons, the father of
neoclassical production theory. In the The Theory of Political Economy
published in 1874, he outlined what was to become neoclassical production
theory, namely that wealth is an increasing, continuous, twice-differentiable
function of homogenous labor and capital. A lesser known, but equally
important contribution, of his was The Coal Question: An Inquiry Concerning
the Progress of the Nation, and the Probable Exhaustion of Our Coal-
Mines, published in 1865 in which he addressed the question of Great
Britain’s dwindling coal reserves. In the opening salvo, he declared:
Day by day it becomes more evident that the Coal we happily possess in
excellent quality and abundance is the mainspring of modern material
civilization. As the source of fire, it is the source at once of mechanical
motion and of chemical change. Accordingly, it is the chief agent in almost
every improvement or discovery in the arts which the present age brings
forth. It is to us indispensable for domestic purposes, and it has of late years
Science and the Wealth of Nations: The Physics of Economic Growth 169
Paradoxically, some nine years later (i.e. in 1874), coal or the energy
input had disappeared completely from what is largely considered to be his
magnum opus, namely The Theory of Political Economy, where capital is
included in the production function and, more importantly, is assumed to be
physically productive. In short, both labor and capital were assumed to by
physically production and more importantly, were substitutable. One could
argue that internal validity (i.e. vis-à-vis the debate over the role of capital
in wealth) is what prevented Jevons from incorporating energy into the
corpus of neoclassical analysis.
Perhaps the greatest of British iconoclasts was Nobel-prize laureate
chemist Frederick Soddy, who after his pioneering work with Ernest
Rutherford on atomic transmutation turned his attention to economics,
largely in response to the alleged “misspecification” of production theory,
more to the point, to the absence of energy from the analysis. The gist of his
critique can be found in the following allegory:
At the risk of being redundant, let me illustrate what I mean by the question,
How do men live? by asking what makes a railway train go. In one sense or
another the credit for the achievement may be claimed by the so-called
engine-driver, the guard, the signalman, the manager, the capitalist, or share-
holder, or, again, by the scientific pioneers who discovered the nature of fire,
by the inventors who harnessed it, by labour which built the railway and the
train. The fact remains than all of them by their united efforts could not drive
the train. The real engine-driver is the coal. So, in the present state of
science, the answer to the question how men live, or how anything lives, or
how inanimate nature lives, in the sense in which we speak of the life of a
waterfall or of any other manifestation of continued liveliness, is, with few
and unimportant exceptions, By sunshine. Switch off the sun and a world
170 9. A Kinetics-Based Approach to Production
would result lifeless, not only in the sense of animate life, but also in respect
of by far the greater part of the life of inanimate nature. The volcanoes, as
now, might occasionally erupt, the tides would ebb and flow on an otherwise
stagnant ocean, and the newly discovered phenomena of radioactivity would
persist. But it is sunshine which provides the power not only of the winds
and waters but also of every form of life yet known. The starting point of
Cartesian economics is thus the well-known laws of the conservation and
transformation of energy, usually referred to as the first and second laws of
thermodynamics. (Soddy 1924, xi)
One of the first problems uncovered has been the need of a long-time index
of power, comparable with the indices of employment, of the volume of
production and trade, of monetary phenomena, that will trace the growth of
the factor of power in our national development (Tryon 1927, 281).
The problem with residuals in growth theory, as it turned out, was not
with their existence, but rather with their disappearance. And this is what
happened in the 1970s when the Solow residual suddenly disappeared,
ushering in the productivity and growth slowdown, a slowdown that has not
been since been reversed. Among the alleged causes was the 1973 OPEC
oil embargo. Unfortunately, because energy was absent from production
theory, it was unclear if and how higher oil prices could impact GDP. The
response was not long in coming. In 1975, Ernst Berndt and David Wood
proposed the KLEMS approach to study the effects of oil prices shocksand
energy price shocks in generalon the economy. The upshot was damning
of the energy input: only 4-5 percent of output could be attributed to it.
Hence, the OPEC oil shock could only lead to a downturn in the presence
of factor input complementarities, specifically the capital-energy
complementarityor substitutability.
At roughly the same time, the internal validity of standard neoclassical
production theory came under fire from a Romanian economist, Nicholas
Georgescu-Roegen, who argued that like all other material processes in the
universe, production can and indeed should be seen as entropy increasing,
hence as an irreversible process. His principal target was the standard
neoclassical approach to wealth creation which, in its simple version, was
reversible.
output via the law of kinetics in general, and via machine speed in particular.
In non-displacement-based material processes, more energy per unit of
capital will lead to greater operating temperatures, greater material
breakdown and higher costs (via Arrhenius’ Law). In the remainder of this
section, we examine the literature—however scant—on the role of
speed/kinetics in production and in productivity growth, starting with
William Longston’s testimony on working conditions in early 19th-century
textiles industry before the Committee on the Factories Bill.
9398. State why you believe that the labour of those employed has doubled
since the first introduction and use of cotton machinery, or at least since you
first knew it? The reason why I believe so is from some calculations which
I have been obliged to make, and by my own observation during the time I
was manager of a mill in 1830 and 1831, when I had some of the same
operations under my own observation.
9399. Have you any objection to put in those calculations Certainly not. [The
following document was then put in and read.]
9400. It appears by this document that the work done is very greatly
increased between the years 1810 and 1832; has the machinery been so
altered as to produce that amazing difference, or does it result from
accelerating the speed of the machinery?It is from accelerating the speed
generally; and another cause is, that more and more exertion is required from
the individual working at the machine; these are the two causes. (Committee
on Factories Bill 1832, 430)
better standard of living for the average man. These are the benefits of ever-
increasing speed and accurate control.
Mass production is not merely quantity production for this may be had with
none of the requisites of mass production. Nor is it merely machine
production, which also may exist without a resemblance to mass production.
Mass production is the focusing upon a manufacturing project of the
principles of power, accuracy, economy, system, continuity and speed.
(Ford 1926, 821).
Industrial processes have been speeded up, new inventions are being added
to manufacturing, new economies of personnel and of management have
been made in industry. The 1929 production can undoubtedly be achieved
with thousands, and probably millions, fewer workers. (Laing 133, 23)
Alfred Chandler, in his definitive work on the early 20th century, echoed
this view, generalizing it to the U.S. economy as a whole.
During these years, the focus of managerial attention shifted from enlarging
the scale of operations to increasing operating efficiency by speeding up the
rate of throughput in the plant. High priority was assigned to modifications
of factory design and layout in order to better integrate worker and machine
tasks. Advances in the electrification of machine drive were indispensable
to the realization of these new objectives and may, indeed, have served to
stimulate the new managerial perspectives that emerged. (Sonenblum 1990,
291)
Science and the Wealth of Nations: The Physics of Economic Growth 175
9.4 Model
In keeping with Beaudreau (1998) who classified inputs in terms of two
main categories, namely broadly-defined energy and organization, we
propose a two-tiered approach to understanding production. The first tier is
purely physical and is governed by the laws of physics, specifically, the laws
of machine and chemical kinetics. Neither tools and/or equipment nor
conventionally-defined labor (supervisors) is physically productive, and
hence is parametric to this tier. Tier I is universal in its application and
reach, accessible to industrial engineers, to physicists as well as to
economists and production specialists, thus ensuring both internal and
external validity. The second tier, Tier II, is the organization tier which
Science and the Wealth of Nations: The Physics of Economic Growth 177
We may now pass to the effects which machinery has in relieving that
excessive muscular strain which a few generations ago was the common lot
of more than half the working men even in such a country as England . . . in
other trades, machinery has lightened man’s labours. The house carpenters,
for instance, make things of the same kind as those used by our forefathers,
with much less toil for themselves . . . . Nothing could be more narrow or
monotonous than the occupation of a weaver of plain stuffs in the old time.
But now, one woman will manage four or more looms, each of which does
180 9. A Kinetics-Based Approach to Production
many times as much work in the course of a day as the old hand loom did;
and her work is much less monotonous and calls for much more judgment
than his did. (Marshall 1890, 218)
This change was echoed in official statistics. For example, the U.K.
Board of Trade, in its Censuses of Production, no longer referred to workers
or production workers, but rather to operatives. The concept of labor
productivity, it therefore follows, took on a new meaning, specifically as a
measure of output per machine manager or operative. Implicitly, labor was
not physically responsible for/involved in generating wealth, but rather was
responsible for overseeing/managing the corresponding machines. To
capture this, we model the demand for supervision/machine operatives as a
function of output, specifically desired output. The greater the desired or
targeted level of output on the part of firms, the greater the demand for
supervision.
As machine operatives are involved in all nm and nc subprocesses defined
by Equation 9.1, it stands to reason that the demand for supervision will
depend on a number of factors, from the individual sub-process supervision
technology, to average overall machine speed, to the overall scale of
operation. For example, if the firm automates a given sub-process, then it
would stand to reason that the demand for supervision per unit of output
would fall as a result. The same would hold for an increase in machine
speed. Only with an increase in the overall scale of operations (i.e. all sub-
processes are increased by the same factor) will the demand for supervision
per unit of output stay the same.
9.3.1 Tier I
Y(t) = S(t)K(t) 9.6
S(t) = E(t)0.5+ī(t) 9.7
Y(t) = E(t)0.5+ī(t)K(t) 9.8
182 9. A Kinetics-Based Approach to Production
9.4.2 Tier II
N(t) = A[S(t), ȝ(t)]Y(t) 9.9
K(t) = B[S(t), K(t)]Y(t) 9.10
has been regressed against all factor inputs (e.g. the KLEMS method). This
approach is abandoned on the grounds that it is theoretically unjustifiable
(i.e. physically-productive versus organizational inputs) and serves to
confuse rather than illuminate. Labor is not and has not been physically
productive for over two centuries (starting with the introduction of the steam
engine). Capital has never been, nor will never be physically productive.
Neither can be substituted for each other as each fulfills an entirely different
function. Moreover, neither can be substituted for energy as neither is a
source of energy. Hence, for these and innumerable other reasons, we
proceed by (i) testing the predictions of machine and chemical kinetics in
economics and (ii) testing the derived input demand for organization
specifically conventionally-defined labor.
1 Output, labor and electricity use data were obtained from United Nations Industrial
Statistics Yearbook (1960-1988); capital data were obtained from OCED, Flows and
Stocks of Fixed Capital 1989. These data are available from the author upon request.
2We opted to estimate the output elasticities directly as opposed to indirectly. This
owed to a number of factors, including the nature of our work (i.e. estimating the
production function itself) and the belief that factor markets are not competitive,
especially the electricity market, making indirect estimation techniques inappropriate.
186 9. A Kinetics-Based Approach to Production
Country Constant EP K L R2
U.S. 1.1678 0.4902 0.7826 0.1034 0.9990
(1958-1984) (0.5342) (3.169) (0.9005) (5.136) ȡ=0.688
Germany -0.2128 0.6124 0.5587 -0.1086 0.9494
(1963-1988) (0.1961) (3.710) (0.5520) (3.2171) ȡ=0.557
Japan -3.1852 0.6970 0.5175 0.3789 0.9818
(1965-1988) (1.328) (2.514) (2.383) (0.9804) ȡ=0.885
Canada -3.6435 0.2451 0.2655 1.2673 0.9844
(1962-1988) (2.421) (0.8631) (1.216) (3.070) ȡ=0.660
Britain 1.2630 0.9141 -0.0972 -0.0870 0.9232
(1963-1988) (0.936) (5.762) (4.000) (0.7156) ȡ=0.502
Finland -0.4482 0.7081 0.3040 0.0583 0.9856
(1963-1988) (0.6075) (4.772) (1.702) (0.3133) ȡ=0.364
Country Constant EP K L R2
U.S. -64.7637 0.4554 0.0963 1.1141 0.9891
(1958-1984) (3.186) (4.009) (1.084) (5.417) ȡ=0.663
Germany -0.2128 0.6124 -0.1086 0.5587 0.9994
(1963-1988) (0.1961) (3.710) (0.5222) (3.217) ȡ=0.557
Japan -164.97 0.5618 0.2374 1.3221 0.9820
(1965-1988) (1.814) (2.983) (3.683) (1.643) ȡ=0.827
Canada -130.442 0.3426 0.1582 1.6709 0.9821
(1962-1988) (2.347) (1.210) (6.2092) (2.836) ȡ=0.651
Britain 17.1535 0.8570 -0.0289 -0.0459 0.911
(1963-1988) (0.3206) (5.605) (0.1008) (0.1720) ȡ=0.583
Finland -17.9512 0.57475 0.3122 0.1854 0.9918
(1963-1988) (0.8936) (6.931) (2.553) (0.9723) ȡ=0.294
Science and the Wealth of Nations: The Physics of Economic Growth 187
Country Constant EP K L R2
U.S. 0.0447 0.3824 0.9839 0.9646 0.9179
(1958-1984) (3.168) (2.498) (2.9561) (6.499)
Germany -0.0016 0.6093 0.0650 0.4675 0.6682
(1963-1988) (0.1469) (3.682) (0.2197) (1.872)
Japan 0.0360 0.7176 0.3052 0.6291 0.6710
(1965-1988) (1.930) (2.942) (1.128) (1.228)
Canada 0.0572 0.0528 -0.9138 1.4123 0.715
(1962-1988) (2.391) (0.2072) (1.933) (4.244)
Britain -0.0118 0.7898 -0.2098 0.0891 0.6432
(1963-1988) (0.9165) (5.122) (0.4099) (0.5597)
Finland 0.0139 0.7066 -0.1171 -0.3518 0.48
(1963-1988) (0.371) (4.364) (0.1352) (0.8735)
We then tested the model using 2-digit SIC data for U.S. manufacturing
from 1947 to 1984. The data, in this case, were taken from the Annual
Surveys of Manufactures as well as the Census of Manufactures. The results
are presented in Table 9.7, where three sets of electric power output
elasticities are reported. Column 1 presents the relevant output elasticity
when both electric power consumption (purchased and generated) and
production workers were included as independent variables, while Column
2 presents the output elasticity when only electric power consumption was
used as the independent variable. Column 3 presents the output elasticity
when the dependent and independent variables were measured relative to
the level of production workers, the idea being that this would eliminate
cyclical biases/effects. As was the case with aggregate data, the output
elasticities were centered around the predicted-by-the-law-of-kinetics value
of 0.50. In fact, in the first case, the average output elasticity was 0.493.
would stand to reason that the elasticity with regard to energy would be
systematically less than that with regard to output. The estimates presented
in Table 9.6 confirm this. The demand for supervision per kwh was less than
the demand for supervision per unit of output in virtually all industries.
namely the speed phase and the automation phase. The former refers to the
increase in labor PFI in the immediate post-WWII period owing to greater
machine speeds, which increased the amount of product per labor or
supervisory input. Again, it is imperative to point out that labor was not
responsible for the increase, but rather was simply a witness to greater
machine speeds.
The second phase, which began in the 1980s and continues to this day is
the automation phase which witnessed the increasing use of inanimate
supervision technologies, commonly referred to as factory automation.
Here, product per factor input increased via a decrease in the denominatoras
opposed to an increase in the numerator in the first phase. As the remaining
supervisors (i.e. labor) were not responsible for the increase in the Labor
PFI, it stands to reason that their remuneration would not, in any noticeable
way, be affected. Perhaps this explains the wage-productivity gap that has
been identified in the literature.
at all, given that information is not physically productive, and can only
contribute marginallyif at allto productivity via second-law efficiency.
As the latter is bounded from above and highly stable, it stands to reason
that ICT has not, cannot, and will not increase productivity. This stands in
contrast to the two other GPTs, namely the steam engine and the electric
motor, both of which resulted in greater energy consumption per machine,
and hence, greater productivity and output.
Lastly, they provide a long, overdue bridge between classical mechanics,
basic physics, process engineering and economics. While the economics
profession has paid and continues to pay lip service to the fact that its
formalizations of production are grounded in engineering data, the resulting
models have been and continue to be orthogonal to material processes as
seen in the physical sciences. It was shown that this bridge provides
valuable insights into such things as productivity and product indexes, the
most telling example being the theoretically-correct measure of Labor’s
Product Per Factor Input, which measures output per unit labor, without
connoting of physical productivity. Such insights are immensely important
in moving the debate over output, wages and profits (i.e. the debate
instigated by Thomas Piketty’s Capital in the 21st Century) along.
9.9 References
Alting, Leo. 1994. “A Morphological Process Model.” in Manufacturing
Engineering Processes. New York: Marcel Decker.
Beaudreau, Bernard C. 1995. “The Impact of Electric Power on
Productivity: The Case of U.S. Manufacturing 19581984.” Energy
Economics. 17, no. 3: 231–236.
Beaudreau, Bernard C. 1998 Energy and Organization: Growth and
Distribution Reexamined. Westport, CT: Greenwood Press.
Beiser, Arthur. 1983. Modern Technical Physics. Menlo Park, CA: The
Benjamin/Cummings Publishing Company.
Chandler, Alfred D. Jr. 1977. The Visible Hand, The Managerial Revolution
in American Business. Cambridge, MA: Harvard University Press.
Cobb, Charles and Douglas, Paul. 1928. “A Theory of Production.”
American Economic Review. 18, 1928: 139–165.
Devine, Warren D. 1990. “Electricity in Information Management: The
Evolution of Electronic Control.”, in Schurr, Sam. H. et al. (eds.),
Electricity in the American Economy. Westport CT: Greenwood Press.
Giraud, Gael and Zeynep Kahraman. 2014. “How Dependent is Growth
from Primary Energy-Output Energy Elasticity in 50 Countries.”
Working Paper, Paris School of Economics.
Science and the Wealth of Nations: The Physics of Economic Growth 193
Jevons, W.S. 1865. The Coal Question. London: MacMillan and Co.
Kummel, Reiner, 1982. “The Impact of Energy on Industrial Growth.”
Energy 7, no.2: 189-201
Kummel, Reiner, Dietmar Lindenberger, and Wolfgang Eichhorn. 1998.
“The Productive Power of Energy and Economic Evolution.” Indian
Journal of Applied Economics, Special Issue on Macro and Micro
Economics.
Laing, Graham. 1933. Towards Technocracy. New York, NY: Angelus
Press.
Marshall, Alfred. 1890. Principles of Economics. London, MacMillan.
Marx, Karl. 1867. “Machinery and Modern Industry.” in Marx, Karl, Das
Kapital. New York, NY: The Modern Library.
Rifkin, Jeremy. 1995. The End of Work. New York, NY: G.P. Putnam’s
Sons.
Samuelson, Paul. 1976. “A Comprehensive Restatement of the Theory of
Cost and Production,” in Samuelson, Paul A. Foundations of Economic
Analysis. New York, NY: Atheneum: 57-89.
Soddy, Frederick. 1924. Cartesian Economics, The Bearing of Physical
Sciences upon State Stewardship, London: Henderson
Solow, Robert. 1974. “The Economics of Resources or the Resources of
Economics.” The American Economic Review, 64(2), Papers and
Proceedings of the Eighty-sixth Annual Meeting of the American
Economic Association:1-14.
Thomas, Woodlief. 1928. “The Economic Significance of the Increased
Efficiency of American Industry.” American Economic Review. 18:
122–138.
Tryon, F.G. 1927. “An Index of Consumption of Fuels and Water Power.”
Journal of the American Statistical Association. 22: 271–282.
U.S. Department of Commerce. Annual Survey of Manufactures.
Washington, DC: Bureau of the Census.
U.S. Department of Commerce. 1986. Survey of Current Business.
Washington, DC: Bureau of Economic Analysis.
U.S. Department of Labor. 1966. Technical Trends in Major North
American Industries. Department of Labor Statistics, No. 1474,
Washington, DC.
Wibe, Soren. 1984. “Engineering Production Functions: A Survey,”
Economica: 401-411.
10
Abstract
10.1 Introduction
The gist of this chapter is quite straightforward, namely that all existing
forms of man-made technology, including material process technology,
started with an idea and thus, is idea-based. However, not all goods and
services ideas affect material processes. Ergo, some ideas matter for growth,
while others don’t. In this chapter, we document the former using the
science of material processes as our guide. Specifically, the model of
production developed in Chapters 3, 4, 8, and 9 will be used to separate the
proverbial wheat from the chaff.
Table 10.2: Innovations That Increased Energy Use Per Unit Capital
Per Unit Time
did the demand for iron, which in turn, put a premium on cheap iron and
copper bar. This led to a number of developments including the use of the
atmospheric steam engine as a source of power to pump water out of English
iron mines, the puddling technique, and the development of an indigenous
copper industry.
growth, the vertiginous rise in the use of computers in the 1990s, 2000s, and
2010s had little to no effect on overall productivity growth. While a paradox
to most, this is consistent, indeed even predicted, by the energy-organization
approach outlined in Chapter 5. In short, information is not physically
productive. As such, the information and communications technology
revolution, by increasing the rate of transmission of information and
reducing its cost of storage, had no effect on growth.
There is, however, one exception, namely the capacity of ICT to increase
second-law efficiency. Specifically, improvements in control technology
made possible by ICT could, conceivably, increase growth, but only on a
punctual basis. In other words, these gains are made at a point in time and
not necessarily over time. Hence, while they could potentially affect growth
in the year they are applied, they will have little to no bearing on the overall
rate of growth.
high of $499 billion in 2015. Of that total, the federally sponsored share fell
to a record-low 23 percent while the business sector’s share rose to a record-
high 69 percent. According to new estimates from the National Science
Foundation’s National Center for Science and Engineering, total spending
on R&D reached $499 billion last year, buoyed by record levels of business
spending. If confirmed, this will represent the largest amount the U.S. or
any nation has ever spent on R&D in a single year. It is one of a number of
indicators that the U.S. remains the international leader in science and
technology even as China poses a challenge to U.S. dominance in the field.
Of total U.S. R&D in 2015, businesses funded $355 billion, or 69 percent,
continuing a long-term trend of private enterprise financing an increasingly
large majority of R&D nationwide. The federal government, the second-
largest funder of U.S. R&D, sponsored an estimated $113 billion, or 23
percent of the total.
Business-funded R&D has grown over the decades and now comprises
about two-thirds of U.S. R&D spending, while federally funded R&D has
gradually declined to its lowest point as share of the economy since records
began in 1953 (see Figures 10.1 and 10.2). Public Domain Federal funding
for R&D has declined for four years in a row, reaching its lowest level since
Science and the Wealth of Nations: The Physics of Economic Growth 205
2007 in 2015. The federal share of total U.S. R&D is also now the lowest it
has been dating back to 1953, when NSF first began recording data on the
subject.
On another note, the findings of this chapter are consistent with the
overall literature, scholarly and popular, on the innovations that changed the
world. For example, the steam engine is widely seen as one of the defining
innovations that marked the start of the industrial revolution and the
dawning of the modern world. Another is electricity which survey after
survey show to be another defining innovation in the rise of modernity. At
the turn of the current century, a local newspaper surveyed its readers on the
question of 20th-century innovations, specifically of the most far-reaching
innovation. The results put electricity and electric power on the top of the
list. Other energy-related innovations include the internal combustion
engine which gave us the automobile and other forms of powered
transportation, and the jet engine which did likewise for flight. Both
increased the speed with which we as a species can displace ourselves in
space and time. That the general public would come to these conclusions is
not, in our view, surprising given the underlying science, namely the laws
of physics.
10.6 References
Henry, M. 2016. US R&D Spending at All-Time High, Federal Share
Reaches Record Low (aip.org/fyi/2016/us-rd-spending-all-time-high-
federal-share-reaches-record-low).
Schmookler, J. 1962 Economic Sources of Inventive Activity* The Journal
of Economic History, 1962, 22, vol.1:1-20
Schmookler, J, 1966 Invention and Economic Growth, Cambridge, MA:
Harvard University Press, 1966.
Thomas, R.P. and D. McCloskey” Overseas Trade and Empire, 1700-1820,”
Chapter 4 in Floud, R. and D. McCloskey, The Economic History of
Britain, 1700-Present (1984), vol. 1, 87-102.
SUMMARY AND CONCLUSIONS
1However, it is our view that what has been and continues to be bold are the myriad
engimas regarding production and growth that today constitute the science of
economics.
Science and the Wealth of Nations: The Physics of Economic Growth 211