Front Cover
 Title Page
 Table of Contents
 The economic value of human time...
 Policies for equitable growth in...
 The fear of knowledge
 Measuring economic and social performance:...
 Appraising the economic performance...
 BAE-ERS alumni day
 World food problems and U.S....
 The political economy of natural...
 Agricultural production economics...
 Back Cover

Title: Lectures in agricultural economics
Full Citation
Permanent Link: http://ufdc.ufl.edu/UF00053788/00001
 Material Information
Title: Lectures in agricultural economics
Physical Description: ix, 165 p. : ill. ; 24 cm.
Language: English
Creator: Schultz, Theodore William, 1902-
United States -- Dept. of Agriculture. -- Economic Research Service
Publisher: Dept. of Agriculture, Economic Research Service
Place of Publication: Washington
Publication Date: 1977
Subject: Agriculture -- Economic aspects -- United States   ( lcsh )
Genre: federal government publication   ( marcgt )
bibliography   ( marcgt )
non-fiction   ( marcgt )
Bibliography: Includes bibliographical references.
Statement of Responsibility: Theodore W. Schultz ... etal.
General Note: Bicentennial year lectures sponsored by the Economic Research Service.
General Note: Issued June 1977.
 Record Information
Bibliographic ID: UF00053788
Volume ID: VID00001
Source Institution: Marston Science Library, George A. Smathers Libraries, University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: aleph - 000010707
oclc - 03110786
notis - AAB3035

Table of Contents
    Front Cover
        Front Cover 1
        Front Cover 2
    Title Page
        Page i
        Page ii
        Page iii
        Page iv
        Page v
        Page vi
        Page vii
        Page viii
    Table of Contents
        Page ix
    The economic value of human time over time
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
    Policies for equitable growth in developing countries
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
    The fear of knowledge
        Page 33
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
    Measuring economic and social performance: New theory, new methods, new data
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
    Appraising the economic performance of the food industry
        Page 81
        Page 82
        Page 83
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
    BAE-ERS alumni day
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
        Page 119
        Page 120
        Page 121
        Page 122
    World food problems and U.S. agriculture
        Page 123
        Page 124
        Page 125
        Page 126
        Page 127
        Page 128
        Page 129
        Page 130
        Page 131
        Page 132
        Page 133
        Page 134
        Page 135
        Page 136
        Page 137
        Page 138
        Page 139
        Page 140
    The political economy of natural resources policy
        Page 141
        Page 142
        Page 143
        Page 144
        Page 145
        Page 146
        Page 147
        Page 148
    Agricultural production economics in the future
        Page 149
        Page 150
        Page 151
        Page 152
        Page 153
        Page 154
        Page 155
        Page 156
        Page 157
        Page 158
        Page 159
        Page 160
        Page 161
        Page 162
        Page 163
        Page 164
        Page 165
    Back Cover
        Back Cover
Full Text


Year Lectures
By the

U.S. Department
of Agriculture

Theodore W. Schultz
Irma Adelman
Charles E. Bishop
Karl A. Fox
George E. Brandow
D. Gale Johnson
M.M. Kelso
Earl O. Heady

Economic Research Service

Natural Resource Economics Division
William Crosswhite, Committee Chairman
Tony Grano
Kathryn A. Zeimetz

National Economic Analysis Division
Wayne Rasmussen, Committee Secretary
William Hoofnagle

Information Division
Loretta Munari, Editorial Assistant
James Sayre, Editor

Economic Development Division
Calvin Beale
Clark Edwards

Commodity Economics Division
James Pearson

Foreign Demand and Competition Division
Arthur Mackie

Foreign Development Division
Roberta van Haeften

June 1977



Year Lectures
By the

U.S. Department
of Agriculture

Theodore W. Schultz
Irma Adelman
Charles E. Bishop
Karl A. Fox
George E. Brandow
D. Gale Johnson
M.M. Kelso
Earl O. Heady


Two hundred years ago, in the same year that the United
States declared its independence, Adam Smith published The
Wealth of Nations, probably the most influential economic
monograph the world has seen. The growth of the United
States over the past two centuries has been paralleled by the
development of economics as a profession. Today, the policies
adopted by.this Nation influence the entire world, while nearly
every major policy adopted is influenced by considerations
advanced by economists. It seems particularly fitting, then, that
the Economic Research Service should commemorate the bicen-
tennial of the Nation's independence by sponsoring a series of
lectures by some of the Nation's outstanding agricultural econo-
Agricultural economics has developed as a discrete branch of
economics only since the turn of the century. One reason for
this may be that from Adam Smith's time to then, agriculture
still dominated world economic activity. A review of Smith's
Wealth of Nations, Ricardo's Principles, or Mill's Principles
indicates that all three assumed that agricultural production was
a major factor in the economy, while Malthus' Essay reviews
the food-population problem in terms that are still used today.
In a sense, all economists were agricultural economists during
the 19th century.
As the century drew to a close, however, agriculture slipped
to a subordinate position in the United States and in the
industrial nations of western Europe. Farmers, now a minority,
organized and began to demand special consideration. The
development of agricultural economics as a separate discipline
was one indirect result of this situation.
As the discipline first developed, farm management came to
the front, followed by agricultural finance and marketing. Land
use also received attention in the early years. During the farm
depression of the 1920's and 1930's, price policy attracted many
able economists. Since then, use of natural resources, rural
development, and other areas have become important. Essen-
tially, as the Nation has changed over the past 200 years to
meet its citizens' needs, so agricultural economics has changed
over the past 75 years to meet the needs of farmers and of the
American public.
To help understand these changes and what they mean to

farmers and the American public as well as to agricultural
economics, the Economic Research Service sponsored a series
of lectures during 1976 by distinguished leaders representing
several areas in the discipline. In honor of our Nation's 200th
Year, we named the events our Bicentennial Year Lecture
Series. ERS' goals for these lectures were to learn from the
experience and foresight of the speakers, to give younger
USDA economists an opportunity to see, hear, and meet some
of the Nation's most innovative economists, and to call atten-
tion to the place of economists in our national life.
I would like to take this opportunity to thank our guest
lecturers: Ted Schultz, Irma Adelman, Charles Bishop, Karl
Fox, George Brandow, D. Gale Johnson, M. M. Kelso, and
Earl Heady. We also appreciate the participation in our ERS-
BAE Alumni Day of Don Paarlberg, O. V. Wells, Willard
Cochrane, Nathan Koffsky, John Schnittker, M. L. Upchurch,
and Walter Wilcox.

Quentin M. West
Economic Research Service


There is no reason why separate speeches by eight prominent
agricultural economists should add up to a consistent whole
when bound into a single volume. This collection of talks in
the Economic Research Service's Bicentennial Year Lecture
Series is no exception. Yet we do expect such a collection to
provide us with insights into some of the current issues agricul-
tural economists are facing. That expectation is fully satisfied.
In these papers, we find pertinent ideas to help ERS and other
agricultural economics research organizations bring into focus
current issues related to defining relevant problems, generating
useful data, identifying appropriate theory, and understanding
the social or political environment in which we work.
The social problems of concern in these papers encompass a
changing commercial U.S. agriculture, a struggling rural non-
farm economy, international trade, and growth in less developed
countries. Two papers were devoted almost entirely to height-
ening our awareness of problems. They both focused on the
world, rather than the domestic situation. D. Gale Johnson
emphasized the importance of stable, adequate world food
supplies and the need for both short-run and long-run solu-
tions. Irma Adelman addressed problems of growth in less
developed countries and emphasized that problems associated
with inequitable income distribution tend to increase as these
nations begin to grow.
Data were seen by several of the authors as a source of
difficulty. The quantity, price, and value measures developed
for U.S. agriculture and closely related sectors during 1922-53
by the former Bureau of Agricultural Economics were cited as
an example of good analysts applying good theory to good
data. The subsequent demise of some of the data, particularly
that related to typical farms and to enterprise budgets, was
lamented. Karl Fox addressed his paper almost exclusively to a
data problem. He pointed out that to measure the performance
of a social system we need direct counterparts to the quantities,
prices, and values required to measure economic performance.
Fox found the key in a complete accounting for the allocation
of time among alternative uses.
Economists have learned that once they have collected data
relevant to a problem, useful descriptions can precede analyses.
There are several examples of this in the papers, including a

descriptive history of natural resource development in the Unit-
ed States. Extensive descriptions are tabulated of the relations
of social and economic characteristics in less developed count-
ries to growth and equity.
Economic theory to explain and help us understand social
problems was of concern in most of the papers. Earl Heady 's
underlying theme is that good training in theory provides us all
with a basis for defining problems, information, and
relationships which need continuous re-estimation as markets,
technology, and institutions change.
Theodore Schultz finds that the explanation of the rising
economic value of human time is not to be found in the
theories of Ricardo, Malthus, Marx, or Keynes. He suggests it
is to be found in the investment of people in themselves, thus
creating human capital. George Brandow uses deductions from
economic theory as a basis for operational standards for evalu-
ating performance in the food industry. Maurice Kelso finds
conventional economic wisdom to be part of the problem,
rather than the solution, to natural resources development and
argues for incorporating the role of institutions in our descrip-
tions, analyses, and prescriptions.
Some of the analysis and processing of data evidenced in
these papers is informal in the sense that the human mind,
possibly aided by pencil and paper, is perhaps our most
important data processing system. Even so, there is extensive
evidence that these authors frequently turn to mathematical
models, large data banks, and extensive systems of hardware
and software to conduct cross-tabulations, simulations, factor
analyses, linear programs, and various econometric studies in
their efforts to interpret the data in the light of economic
Public policy prescriptions abound in these papers, even
though one of the authors explains that involvement in public
policy is controversial and that there tends to be a gulf
between the academics and the policymakers. Wrong-headed
policies are shown in these papers to be, sometimes, a part of
the problem. One example is natural resource policies which
assume that all one needs to do is to ensure that the channels
of communication are open in order to fine tune the economy
while assuming away the problems of setting goals, resolving

conflicts, and seeking productive combinations. Another exam-
ple of policies which complicate, rather than ameliorate, the
situation is in the policies of less developed countries which
discourage, rather than expand, food production. Some authors
used these lectures as a forum for expounding policy pre-
scriptions. For example, a grain reserves policy is outlined as
an insurance against short-run fluctuations in world food sup-
The environment in which agricultural economists work
receives explicit attention from a few of the authors. There are
references to the continual expanding demand for our services
and to the responsibilities that a large research organization
like ERS has in its influence on professional activities. Ed
Bishop addressed most of his lecture to the uncertain support
which academicians have received from the public during the
last 200 years.

Clark Edwards


Theodore W. Schultz 1 The Economic Value of Human
Time Over Time

Irma Adelman 25 Policies for Equitable Growth
in Developing Countries

Charles E. Bishop 33 The Fear of Knowledge

Karl A. Fox 47 Measuring Economic and
Social Performance: New The-
ory, New Methods, New Data

George E. Brandow 81 Appraising the Economic Per-
formance of the Food Industry

Don Paarlberg 101 BAE-ERS Alumni Session
Walter W. Wilcox
O. V. Wells
Willard W. Cochrane
M. L. Upchurch
Nathan M. Koffsky

D. Gale Johnson 123 World Food Problems and
U.S. Agriculture

M. M. Kelso 141 The Political Economy of Nat-
ural Resources Policy

Earl O. Heady 149 Agricultural Production Eco-
nomics in the Future

The Economic

Value of

Human Time

Over Time

Theodore W. Schultz

Lecture 1
March 23, 1976

What have our ideas and institutions contributed to the high
value we place on human beings and to the remarkable
increases in the value of human time? While it is fitting and
proper during this bicentennial year to celebrate our achieve-
ments as a Nation, it is much more important to examine
critically and to reassess the social, political, and economic
ideas embodied in our institutions. Because of the influential
books, pamphlets, and documents appearing in 1776, it is
convenient to date these ideas accordingly. Although these ideas
are a critical part of our social heritage, it will no longer
suffice to take them for granted. It is a heritage that is being
eroded for lack of support against strong alternative social,
political, and economic ideas.
The ERS bicentennial lectures afford an opportunity to
examine the interplay between social thought and the political
economy-presumably with special reference to the economic
history of agricultural productivity; the decline in the social,
political, and economic influence of the owners of land; and
the rise in the economic value of human time.
What people do privately and collectively over time is con-

Theodore W. Schultz

strained by the consequences of the interactions between ideas
and institutions consisting of observable responses of one to the
other. In this context, ideas are embodied in social thought,
and the social, political, and economic order is maintained by
institutions. The response of ideas to the institutionalized order
is of two historical types: those that rationalize and contribute
to the codification of the prevailing order, and those that arise
in protest to the established order which become embodied in
social thought and then become strong enough to induce a real
alteration in the prevailing institutions.
From the point of view of economics, the consequences of
the interactions between ideas and institutions differ markedly
from one period to another. It is useful to think in terms of
four periods: the mercantile period preceding 1776; the era of
economic liberalism that followed; the more recent forms of
neo-mercantilism ranging from centrally planned economic
development to a system of command economics; and, cur-
rently, there are signs of a nascent, neo-liberalism emerging in
protest to the various centralized and authoritarian institu-
tionalized orders.
The mercantile system that prevailed for decades, for exam-
ple in England, prior to 1776 was rationalized by economic
ideas. These ideas provided support for governmental
restrictions on trade, on internal prices and wages, and on
migration. The social and political order was buttressed by the
established Church and by the Law.
In protest to the adverse social and economic effects of the
then prevailing institutionalized order, 1776 was an extraor-
dinary year in terms of the various intellectual publications that
laid the foundation for liberalism. The open competitive econo-
my with people responding to market prices that are not
fettered by private or public monopolies in a political order in
which the functions of the state are greatly restricted-this was
the classic contribution of Adam Smith. Smith's economic ideas
complemented the more general liberal thought of his day
which over the decades that followed profoundly altered the
institutionalized functions of the government.
Liberalism, because of its accommodation of the economic
attributes of capitalism as seen in the adverse social effects of
industrialization, led in turn to a wide array of protests. The

ideas that emerged from these protests prior to those of Karl
Marx called for various forms of socialism. The contributions
of Marx, however, came to dominate the political and eco-
nomic foundations that are required for socialism. The response
to Marxian ideas has altered greatly the institutions of many
nation states. The economic functions of government are much
enlarged, and in this sense it is a new form of neo-mercan-
In protest to socialism there now emerges a neo-liberalism
which is still no more than nascent as a system of social ideas.
Because of the dependency of socialism on a vast increase in
the functions of government, which in some countries consists
of extreme authoritarian nation states, and because of the now
widely observed adverse effects of the governmental institutions
on personal freedom, protests akin to those of two centuries
ago are once again the order of the day.

As Professor of Economics at
the University of Chicago. T. W.
Schultz ranks as one of the Nation's
most distinguished and respected
economists. Outstanding credentials
include five honorary degrees;
author of II books and editor of
six: consultant to the Ford and
Rockefeller Foundations, Brookings
Institution, five U.S. Government
departments, and the President's
Economic Council: and advisor to
the World Bank and the Food and
Agriculture Organization of the
United Nations. His wide-ranging
interests include world agricultural
development, economic implications
of research and education. U.S.
Theodore W. Schultz farm policy, and the economic value
of human life.

Theodore W. Schultz

The core of the argument of this lecture does not deal
directly with the ideas and institutions of these various eco-
nomic systems. The historical evidence, however, which I am
about to present, does provide some indirect information that
gives rise to doubts about the adequacy of classical theory and
also of the economic theory of socialism. The economic history
to which this paper appeals supports four major propositions
that are not at home in either of two long established sets of
The first arises out of the fact that wage and salary workers
in high-income countries, and especially in the United States,
have become capitalists as a consequence of their large
investments in their own human capital. The opportunities and
incentives to which workers respond in their investment in
human capital is not a part of the theory advanced by Ricardo
or that of Marx. Nor is it an integral part of Keynesian
Secondly, contrary to the core of classical economics built
on the assumption of the dominating effects of Ricardian Rent
on personal income shares, the economic and the associated
social and political importance of landlords has in fact declined
markedly over time in high-income countries.
Thirdly, contrary to Marx and not anticipated in classical
theory, changes over time in the functional distribution of
income have had a major effect iii reducing the inequality in
the personal distribution of income.
Fourthly, given the imprint of Malthus, theory until very
recently has been silent on the economics of the decline in
fertility in high-income countries. There is still another major
implication that is not featured in this paper the extraordinary
decline in the labor force required to produce agricultural
products in high-income countries. Regardless of the source or
the mode of received theory, this important historical economic
development cannot be derived from the theory.
I first present the elements of an economic approach; I then
turn to the measurement of the increases in- price of human
time and to the secular trend in the prices of natural resource
materials and rent. Lastly, I consider some of the implications
of the high price of human time.

An Economic Approach to the
Price of Human Time

Value and capital are the core of economics. Events and
human behavior alter the scale of values and the stock of
capital. Alterations that enhance the scope of choices are favor-
able developments. The various forms of capital differ signifi-
cantly in their attributes. Natural resources are not
reproducible; structures, equipment, and inventories of com-
modities and goods are physical entities that are reproducible;
and human beings are productive agents with the attributes of
human capital. Human beings are also the optimizing agents
and, in a fundamental sense, it is their preferences that matter
in the use that is made of the various forms of capital. The
concept of human capital, its development and its usefulness, is
a recent innovation. In high-income countries, the rate at which
human capital increases exceeds that of nonhuman capital.
Despite unfavorable events from time to time, human behav-
ior in high-income countries has brought about a secular
increase in the personal stock of human capital, and the
increases in it have gone hand-in-hand with the rise in the
value of human time. A part of the analytical task is to extend
the core of economics to explain the formation and the func-
tions of human capital in the changing context of secular time.
Our knowledge of the economics of the processes that alter
the supply of human capital has been advanced during the last
two and a half decades. Schooling, higher education, on-the-job
experience, migration, and health all contribute to the personal
supply. But changes in the supply of human capital are only
one of the two major parts of the story in explaining the
increases in the economic value of human time. The value that
is revealed in the price of human time is the intercept of the
prevailing supply of, and the demand for, the market and
nonmarket services and for the direct personal satisfactions that
people derive from their human capital. Our knowledge of the
economic processes that alter the demand for human capital is
not nearly as satisfactory as it is with respect to the supply.
To acquire an historical perspective on the unsettled issues
pertaining to the price of human time, it is necessary to go

Theodore W. Schultz

beyond current events and recent economic growth. It is useful
to recall that the early English classical core of economics
rested on the following propositions: a highly inelastic supply
of land (natural resources) resulting in Ricardian Rent, popu-
lation growth resulting in constant (subsistence) wages over
time, and the productivity attributed to additional nonhuman
capital and advances in useful knowledge (state of the produc-
tive arts) result in a larger population and in increases in
Ricardian Rent. All household activities were assumed to be
unproductive. Accordingly, the economic activities that mattered
were confined to market-oriented production activities.
We shall appeal mainly to seven decades of economic history
to establish the fact that the price of human time has risen
markedly in countries that have developed a modern economy
and to establish the additional fact that the secular trend of the
real price of the services of natural resources, of both renew-
able and nonrenewable natural resources, has been essentially
horizontal. As a consequence, the economic importance of the
price of human time has risen greatly relative to the price of
the services of natural resources.
In view of this historical development, how can the following
issues be resolved? Why has Ricardian Rent lost its economic
sting in these high-income countries? Why have real wages
increased so markedly despite population growth? Closely
related, why has the demand for human time increased more
than the combined quantity and the quality of the supply?
Although it is obvious that the high price of the services of
labor cannot be explained by a theory of subsistence wages,
nor by a theory of labor exploitation, and furthermore, in spite
of the fact that a very large share of our national income
accrues to labor, a labor theory of value will not suffice. The
extension of the economics of labor that is made possible by
the development of the concept of human capital explains in
large part the increases in the supply of valuable acquired
abilities including skills by means of investment. But, this
extension of labor economics, already implied above, does not
explain the increases in demand for the services of the human
capital that become embodied in our youth and our adult
We begin with an appeal to general theory and proceed to

build on the proposition that the value of the services of
capital holds the key to the explanations we are seeking. We
require for the purpose at hand an all-inclusive concept of
capital. Reproducible tangible wealth is only one category of
capital. Although natural resources are not reproducible, they
are treated here as another category of capital. Human agents
are the most important category in this all-inclusive concept of
capital consisting of the economic attributes of human capital.'
The value placed on services of these various categories of
capital is revealed in human behavior. A closely reasoned
economic theory to explain this behavior rests on two basic
concepts: the human being is an optimizing agent whose behav-
ior is governed by constraints that are in part peculiar to him,
and the market provides the auction at which all individual
offers are acknowledged and the terms of trade are established
for equalizing supply and demand.2
Our approach to the increases in the value of human time
concentrates on the changes in the supply and demand inter-
cepts over time. The variables that alter the supply of human
time are fairly clear: total hours devoted to market and non-
market activities, including hours allocated to consumption;
composition of the hours of the population that engages in
these various activities; and quality of these hours. These sup-
ply variables are interdependent in their responses to the hourly
price and to the income effects of that price.
Given the state of economics, the variables that alter the
demand for human time are not wholly clear. The amount and
the price of the services of natural resources and the effects of
this price on wages have been integral to economics since the
early classical period. They form one of the variables that
matters though not behaving in accordance with that theory.
Moreover, effects of the rent derived from natural resources on
the demand for human time are small in high-income countries.
The stock of nonhuman reproducible capital, the price of its
services, and the forms of this capital constitute a major
variable in their influence on the demand for human time. The
effects of the income from this category of capital on the
demand for human time are still substantial even though the
income from this source has declined much relative to the
income earned by labor. Another variable of increasing

Theodore W. Schultz

importance over time is the demand for the personal services of
professional and competent technicians by families who derive
their income from wages and salaries.
The approach advanced here requires an additional critical
variable. It consists of changes in the arts of production and
consumption that occur as a consequence of advances in useful
knowledge. This variable has two important specific economic
attributes. In the case of natural resources, it consists of new
man-made substitutes for such resources. In the domain of
reproducible physical capital, it consists of new forms of such
capital which are complementary with high human skills and
which for this reason increase the demand for such human
capital and the value of human time.
We now turn to the task of marshalling some of the historical
evidence on the extent of the secular rise in the price of human
time, concentrating on the period since 1900.

Measurement of the Price of Human Time

For most people throughout the world, economic conditions
are such that the value of their time is very low. Labor earns a
pittance. Work is hard; life is harsh. Countries with low earnings
cover most of the world's map. In a few countries, however, the
value of the time of the rank and file of people is by comparison
exceedingly high. The high price of human time that characterizes
these exceptional countries is from the viewpoint of economic his-
tory a recent development. In Ricardo's day, land rents were
indeed high relative to wages. During Marshall's period, real
wages were a small fraction of what they have become since then.
In the United States, for example, between 1900 and 1972, the re-
al hourly "wages" in 1967 dollars in manufacturing rose from
about 60 cents to $3.44.3
We begin with an overview of the secular increases in the eco-
nomic value of human time. The direct evidence consists of esti-
mates of real wages covering the period since 1900 by decades.
Starting with the United States, we use Rees' NBER (National
Bureau of Economic Research) estimates of the compensation per
hour at work for manufacture production workers as a proxy of
the price of human time in the market sector. The real hourly

price rose during each of the decades. By 1972, the real compen-
sation per hour had risen more than fivefold. As would be
expected, the annual estimates show several years when a decline
occurred. There were 3 such years before World War I and 8
since then.4 The marked upward trend, however, is shown in
table 1.
The upward trends in real wages in industry in France, Ger-
many, Sweden, and the United Kingdom (table 2) are, in general,
much like that of the United States. They differ somewhat, how-
ever, in that the rate of increase is higher in Sweden and the
United States than it is in the other three up to about 1960,
except for the United Kingdom which did not stay abreast of the
rest. There are also other country differences in the movement in
real wages that are noteworthy. France and the United Kingdom
show no increase between 1900 and 1910. As of 1925, Sweden
and the United States were substantially ahead of the other three
countries (was it a consequence mainly of differences in the effects
of the aftermath of the war?). Sweden and the United States
maintain their advantage over the others up to 1960 with the
United Kingdom losing ground relative to the rest. Lastly, during
the decade of the sixties, France and Germany join Sweden and
the United States in showing approximately a fourfold increase in
real wages over the period from 1900 to 1970, whereas the
increase for the United Kingdom is threefold.
The fact that real wages rose as much as they did implies
strongly that the various prices other than the price of human
time which enter into the deflator of actual wages declined rela-
tive to wages. Because of the stress placed on land rent rising as
capital is accumulated and population increases over time in clas-
sical theory, and the concern about the "unearned" income of
landlords in both liberal and socialist thought, we turn to the
trend in the prices of the services most dependent on natural
resources along with a comment on Ricardian Rent on farmland.

Prices of Natural Resource
Materials and Rent

Our aim here is to present evidence on the trends in the
prices of these materials. We shall concentrate on the com-

Theodore W. Schultz

Table 1-Total work compensation per hour of manufacturing production
workers, United States, 1900-1972

Year Rees' wages 1900=100
1967 dollar Index
1900 0.60 100
1910 .70 117
1920 .92 153
1930 1.06 177
1940 1.60 267
1950 2.15 358
1960 2.85 475
1970 3.27 545
1972 3.44 573

Source: Albert Rees, Long Term Economic Growth, 1860-1970. U.S. Bur.
Econ. Analysis, Washington, D.C.: 1973, App. 2, B70, pp. 222-223. Note:
Estimates for 1970 and 1972 are derived using a similar method. Rees' esti-
mates are adjusted from 1957 to 1967 dollars.

modities that are most closely identified with natural resources
because of the widely held belief that natural resources are the
critical limiting factor available to the economy. We consider
renewable natural resources (agriculture and forestry) and non-
renewable natural resources (mining including mineral fuels)
and we examine the trends in the prices of the commodities
from these sources. We shall refer to this natural resource
complex as the extractive industries. The commodity prices that
we present are of course not pure material prices; far from it,
for they embody in various combinations the productive service
of labor and of reproducible capital along with the material
Accordingly, our concept of the extractive industries encom-
passes mining (of which metals and mineral fuels are the
dominant components) and agriculture and forestry. Most of
the commodities produced by these industries tend to remain
quite constant over time in their physical, chemical, or biologi-
cal attributes. A bushel of wheat, for example, produced in
1900 differed little from a bushel produced in 1970. Similarly
constant are such commodities as lead, copper, or sulphur.

Table 2-Indexes of real wages in industry in France, Germany, Sweden,
United Kingdom, and the United States, 1900-1970

United United
Year France Germany Sweden Kingdom States

1900 112 108 110 104 110
1910 112 116 131 104 121
1925 135 127 158 113 160
1930 138 156 183 124 160
1938 142 155 190 133 203
1950 168 174 252 169 292
1960 290 282 343 219 381
1970 442 482 473 301 446

Source: E. H. Phelps Brown, "Levels and Movements of Industrial Pro-
ductivity and Real Wages Internationally Compared, 1860-1970," Economic
journal, 83, 58-71. Based on tables III and V of the appendix. Note: In
interpreting the increases in real wages shown in table 2, it should be borne
in mind that we are now dealing with real annual wages in industry. They
are not hourly wages. They are less complete in getting at the total com-
pensation of employees than the estimates by Rees. Accordingly, Rees'
estimates show a higher rate of increase than Brown's. Thus, for the
United States during the period from 1900 to 1970, Brown's real wages
show a fourfold increase and Rees' real hourly "wages" a strong fivefold

Quality changes occur, however, in such commodities as milk
and other livestock products. Historical records of these com-
modity prices are, in general, more reliable than that of final
and intermediate goods. For the United States, we have the
excellent study by Potter and Christy5 of commodities produced
by extractive industries covering the period beginning soon after
the Civil War with annual estimates up to the mid-fifties. The
Potter-Christy study has been updated by Manthy.6
The empirical story as told in table 3 is that the trend of the
deflated natural resource commodity prices over this period was
not upward but slightly downward,7 compared to the more
than fivefold rise in real hourly wages shown in table 1. Within
agriculture, the deflated prices of all crops declined about a
third despite various government price supports during parts of

Theodore W. Schultz

Table 3-Indexes of deflated commodity prices of the extractive industries,
United States, 1900-1972

All All All All
Year commodities agriculture forestry metals

1900 100 100 100 100
1910 99 126 99 76
1920 109 111 97 66
1930 76 90 56 45
1940 77 86 87 60
1950 108 131 99 68
1960 87 95 90 75
1970 79 88 74 76
1972 83 92 84 71

Sources: N. Potter and F. T. Christy Jr., Trends in Natural Resource
Commodities. Baltimore: Johns Hopkins Press for Resources for the Future,
1962. Actual prices are weighted by the value of output, updated using
1967 weights by Robert S. Manthy, Mich. State Univ. Indexes of actual
prices are deflated by the consumer price index, 1967=100.

this period. The index for all livestock closes out this period at
the level where it began (table 4). In general, the costs of
producing livestock products have been affected more by the
increase in the price of human time than have the costs of
producing crops. The deflated prices of mineral fuels shown in
table 5 tell us that, whereas the deflated price index for all
mineral fuels was about a fourth less at the end of this period
compared to 1900, the price of bituminous coal rose and that
of petroleum fell. It is undoubtedly true that the rise in real
wages accounts for a good deal of the increase in coal prices.
We have not used economic theory in mapping the course of
natural resource commodity prices. Nor have we used theory to
gain plausibility or to derive testable hypotheses, nor to vali-
date the reported behavior of these prices over time. Our
reason for this apparent non-theoretical approach to the factual
data is that there is no general economic theory that encom-
passes the type of development here under consideration. There
are conflicting theories of the substitution among capital, labor,

Table 4-Indexes of deflated agricultural commodity prices, United States,

All All All
Year agriculture livestock crops

1900 100 100 100
1910 126 127 118
1920 111 118 87
1930 90 99 73
1940 86 95 73
1950 131 141 110
1960 95 101 75
1970 88 100 66
1972 92 104 69

Sources: See table 3.

Table 5-Indexes of deflated commodity prices of metals and of mineral
fuels, United States, 1900-1972

All All mineral Natural ominous
Year metals fuels Petroleum gas coal

1900 100 100 100 100 100
1910 76 48 42 93
1920 66 118 131 118 146
1930 45 61 59 114 79
1940 60 59 57 80 104
1950 68 81 84 68 156
1960 75 79 79 119 125
1970 76 72 68 111 125
1972 71 73 67 112 143

Sources: See table 3.

Theodore W. Schultz

and natural resource commodities used by the industrial sector.8
There are some studies of the value of the amenity services
relative to the value of commodities from the same natural
resources.9 Going back, there is what was then considered to be
a general theory with built-in Ricardian Rent.
Received theory at the turn of the century predicted that the
price of natural resource commodities would rise relative to
wages and that the rental income accruing to the owners of
natural resources would become an increasing share of national
income. No less an authority than Marshall presents this pre-
diction cogently and strongly in the preface to the 8th edition
of his Principles of Economics. The observed course of natural
resource commodity prices, including the price of food pro-
duced by agriculture, is obviously not consistent with Marshall's
prediction. Furthermore, as we show presently, his prediction
with regard to rentals on agricultural land is also inconsistent
with the evidence since his day.
There is no general theory for the task at hand. Our
recourse is at this point limited to partial theories of the
demand for and supply of natural resource commodities and of
the amenity services derived directly from nature. A good deal
is known empirically about the income elasticity of consumer
demand for various farm goods, for material goods, and for
some of the amenity services of nature. Theory and evidence
about price elasticities of the demand pertinent here are not
especially useful in analyzing developments that occur over
In thinking about nonrenewable natural resources, the com-
mon sense perception of their eventual exhaustion or permanent
impairment as a source of amenities is not in dispute. With
regard to renewable natural resources, the ultimate limits of the
surface of the earth suitable for growing crops and trees on
which agriculture and forestry depend are also not being called
into question. The critical unsettled economic question in this
connection pertains to the changes over time in the substitution
possibilities among natural resources, labor, and reproducible
In dealing with this question, a specific theory of these
substitution possibilities for each of the following classes of
activities would be of some help: direct substitution of capital

"...economic theory is yet
incapable of predicting the sup-
ply of additional, new man-made
substitutes for natural

and labor for natural resources in the production of natural
resource commodities and in providing amenity services; substi-
tution of capital and labor for natural resource commodities in
industrial production; likewise in household production; and
natural resource-saving adjustments in what is finally consumed.
But to reckon fully with these various substitution possibilities,
we would also have to know the changes that occur over time
in the technical possibilities to substitute.
In retrospect, a wide array of man-made substitutes for
natural resources have been developed; namely, substitutes for
land in agricultural production, for natural resource com-
modities in industrial and household production, and in what is
finally consumed. These man-made substitutes of the past can
be identified and their supply effects can be determined, but it
is fair to say that economic theory is yet incapable of predict-
ing the supply of additional, new man-made substitutes for
natural resources. The lack of such a theory does not mean we
are wholly ignorant with regard to future prospects in this
Before considering the course of agricultural land rentals as
one of the prices of the services of natural resources, the state
of our knowledge about substitution within other classes of
economic activities calls for a brief comment. What is known
about substitution among capital, labor, and natural resource
commodities in U.S. manufacturing is presented succinctly by
Humphrey and Moroney.10 The two main sets of conflicting
views are reviewed followed by an appeal to new evidence
using two alternative research designs in estimating substitution
as it is revealed in manufacturing during 1963. They conclude
that regardless of the underlying causes of input substitution,
be it technology-induced or price-induced, "....the evidence sug-
gests that labor and, to a less degree, capital are substitutable
for natural resource products among most of the resource-using
product groups of American manufacturing."
There is no comparable study of substitution in household
production, in part, no doubt, because the extension of theory
to deal with household production and using it in undertaking
empirical analysis are very recent advances in economics. There
are some useful studies of substitution that occur in the choice
of the components that enter final consumption, including the

Theodore W. Schultz

consumption of natural resource amenities.
Our portrait of prices is pleasing to behold; more pay for
less work and enough more of nature's commodities at about
the same real price in 1970 as in 1900, although the population
had grown from 76 to 203 million and GNP had increased
nearly ninefold. But is it a true picture? Our deflator is obvi-
ously influenced by the rising cost of labor. It may not be the
appropriate deflator in the case of natural resource commodity
prices. Beyond this, there is an important defect in our treat-
ment of the price effects of natural resources. It arises out of
the fact that natural resource rents were a declining part of
actual commodity prices. Accordingly, the course of commodity
prices which we have presented is biased upward because of the
increasing share of the real costs of producing commodities
consisting of wages and of reproducible capital costs. Although
the output per man-hour of the extractive industries has
increased, we find that, in general, the more labor-intensive the
commodity the larger is this upward bias. Within agriculture,
for example, the deflated price of wheat declined by one-half
between 1900 and 1970-72, whereas that of fresh tomatoes
doubled between 1920 (the first date for which the price is
available) and 1972. I wish we could determine the true price
of the services of natural resources.
If we knew the true rent paid for the original properties of
farmland, we would have the most important price of the
services of natural resources, inasmuch as the value of the
output of agriculture dominates the total value of the output of
all extractive industries." But the rent we want is very elusive,
for here too the rent that is reported is definitely biased
upward and increasingly so over time. The farmland rent that
is recorded is not Ricardian Rent; only a part of it is rent paid
for the original properties of the soil.
A substantial part of the productivity of farmland is man-
made by investments in land improvements. There undoubtedly
have been circumstances and periods when some disinvestment
occurred. In the United States during the thirties, soil
depletion, including water and wind erosion, became a much
publicized issue. The New Deal provided government subsidies
for soil conservation programs. Although the actual soil
depletion was much exaggerated, the soil conservation programs

have been, among other things, land-improving public
To acquire a perspective of the extent to which the produc-
tivity of farmland is man-made, we turn briefly to other parts
of the world on this point. The original soils of western
Europe, except for the Po Valley and some parts of England
and France, were, in general, very poor in quality. As farm-
land, these soils are now highly productive. The original soils
of Finland were less productive than most of the nearby
western parts of the Soviet Union; today, however, those of
Finland are far superior. The original farmland of Japan was
vastly inferior to that of northern India. Presently, the differ-
ence between them is greatly in favor of Japan. Argentina has
excellent natural soils for growing corn and it has good wheat
land. But its productivity is far below that of Iowa and
Kansas, States in which farming began with raw land compara-
ble to that of parts of the Argentine. Harsh, raw land is what
most farmers since time immemorial have started with; what
matters much over time is the investment made to enhance its
In the United States, investments made to improve the raw
land with which farmers began are of various forms and the
accumulated amount is large. These investments, however, did
not occur at a steady rate because the incentives to invest
fluctuated widely over time. They were large during most of the
first two decades of this century, notably so in drainage in
what are now the best parts of the Corn Belt. From 1920 to
1929 and even more so in the thirties, investments to improve
farmland were at a low ebb. Since then, these investments have
gone on apace with the government paying much of the bill.
Modern earth-moving equipment reduced sharply the costs of
terracing, of improving the water run-off courses, and of level-
ing .land. Since the late thirties, the acreage under irrigation
doubled; it now covers 40 million acres and has cost billions of
dollars, much of which has been paid by U.S. taxpayers.
In the more arid and extensive farming parts of the Moun-
tain and Pacific States, the cash rent paid for grazing land,
which includes the good higher altitude pasture, averaged less
than $4.00 per acre in 1970. The market value of farmland per
acre in California provides a clue to the implied effects of

Theodore W. Schultz

investments on rents. In 1970, nonirrigated cropland sold on
the average for $560 per acre compared to $1,090 for irrigated
land used for intensive field crops. Irrigated land used for truck
and vegetable crops was priced at $1,670 per acre and irrigated
orchards and groves with their investment in fruit-bearing trees
were priced at $2,730 per acre.12
Lindert's study of land scarcity'3 includes a series of farm-
land rents extending back to 1900. It is the gross rent per acre
of land rented for cash, restricted to five Midwest States. The
actual gross rent is adjusted by consumer prices. These rents
rose nearly 15 percent from 1900 to 1915, declined 30 percent
by 1920, and drifted even lower as of 1940. Lindert's series
then shows an upward trend beginning in 1950. By 1970, these
rents were 6 percent above 1900 but less than in 1915.
Although farmland prices and agricultural commodity prices
fluctuate more over time than gross cash rents, the patterns are
in general quite similar. But none of these patterns provides
any direct evidence on the share of the gross cash rent that is
to be attributed to the original properties of the soil. The
indirect evidence indicates that the expenditures of landowners,
including the value of time they devote to the management of
this property, has risen substantially from 1900 to 1970. It also
indicates that land improvement investments and investment in
buildings and in other structures, especially so in a livestock
area such as the Midwest, have been large over the period
since 1900. My interpretation of this indirect evidence is that
the share of gross cash rent attributable to the Ricardian Rent
component declined between 1900 and 1970, probably as much
as a third for the reasons indicated. If this interpretation proves
to be valid, it implies that whereas gross adjusted cash rents
were about the same in 1900 and in 1970, the value produc-
tivity contributed by the natural land resource declined substan-
tially. Major factors accounting for this decline have been the
effects on farmland rents of the development and use of man-
made substitutes for farmland.
We have sketched the seven decades to show the marked
upward trend in the economic value of human time. In marked
contrast, as we have shown, the price of materials most dependent
on natural resources during these seven decades has not tended
upwards. It is also noteworthy, although we are not presenting

"The human agent becomes ever
more a capitalist by virtue of his
personal human capital..."

the data, that real wages rose somewhat between 1860 and 1900.
The increase was small, rising between one-half and onefold for
the several countries, including the United States, listed in table 2.
Going back in history a bit further, back to Ricardo's day, real
wages were rising, but here the comparison has to be in terms of
the cost of food. A clue is to be had in the change in the price of
wheat relative to that of wages. The estimates that follow tell a
good deal about economic history using wheat and wages.

2 weeks of wages
in bushels of wheat

Time of Ricardo (1817) England 1
Marshall's time (1890) United States 20
Eighty years later (1970) United States 200
State of being very poor (1975)
India plowman 3.5
India field laborer 2 or less

High Price of Human Time Implications

The social, political, and economic implications are pervasive.
The high price of human time is a clue to many puzzles. These
puzzles include the shift in institutional support from the rights of
property to that of human rights, the decline in fertility, the
increasing dependence of economic growth on value added by
labor relative to that added by materials, the increases in labor's
share of national income, the decline in hours worked, and the
high rate at which human capital increases. The human agent
becomes ever more a capitalist by virtue of his personal human
capital, and he seeks political support to protect the value of that
capital. These are some of the major implications of the high
price of human time. I shall comment on a few of them.

The rise in the value of human time makes new demands on
institutions. Some political and legal institutions are especially

Theodore W. Schultz

subject to these demands. What we observe is that these institu-
tions respond in many ways to the changes in demands of the
economy. The legal rights of labor are enlarged and in doing so
some of the private rights of property are curtailed. The legal
rights of tenants are also enhanced. Seniority and safety at work
receive increasing protection and discrimination in employment is
curtailed. Since I have dealt with these institutional issues and
with ways of bringing the analytical tools of economics to bear on
them elsewhere,14 I shall not pursue these issues further on this

Labor's Income Share
The interactions between the effective labor force at work and
hourly wages on the one hand, and the amount of nonhuman
capital and the price of the services of that capital on the other,
are exceedingly complex. Kuznets15 gives us an analysis of these
interactions in which he takes account of the increases in the
stock of wealth represented by land and of the stock of
reproducible producer capital and the changes in the prices of the
services of these forms of capital, along with the increases in man-
hours worked and the rise in the price per man-hour worked. His
analysis implies an increase in labor's income share.
The obverse of the increase in labor's share of national income
is the decline in the share accruing to property assets. Appealing
once again to the studies of Kuznets,16 where he takes a fairly
long view of the development in western countries, he finds that
the share of national income accruing to property assets declined
from about 45 to 25 percent, while labor's share rose from about
55 to 75 percent.
By 1970, about three-fourths of the official U.S. national
income consisted of employee compensation.17 The remaining
fourth is classified as proprietors' income, rental income, net
interest, and corporate profits. These four classes of "property"
income include considerable amounts of earnings'8 that accrue to
human agents for the productive time they devote to self-
employed work and to the management of their property assets.
A conservative estimate of the aggregate earnings that accrue to
human agents, as employee compensation plus self-employment
earnings, and for management of assets within the domain of the
market sector, accounts for more than four-fifths of the 1970 U.S.

national income.
Measured national income, however, is substantially less than
the full income that people acquire from the services of their
property and for their time inasmuch as the concept of national
income is restricted to the economic activities of the market sec-
tor. It excludes the economic value of all household production.
The additional income that is realized from household production
is in large part contributed by the value of the time of house-
wives. Also omitted are the earnings that adult students forego in
investing in their education, and the wages that members of the
labor force forego in acquiring on-the-job training. These and still
other income-producing activities are not included in the account-
ing of national income.
It is clear historically that labor's income share increased as the
price of time rose. The development of the U.S. economy from
1900 to 1970 strongly supports this implication. During 1900-
1909, using the official concept of national income, employee
compensation accounted for about 55 percent of national income
compared to 75 percent in 1970.19 Between 1900-1909 and 1970,
the changes in the shares of income other than employee compen-
sation were: proprietors' income declined from about 24 to 8 per-
cent, rental income from 9 to 3 percent, and net interest from 5.5
to 4.1 percent of national income, whereas corporate profits rose
from 7 to 9 percent. The latter two income components fluctuated
widely over this period as would be expected in view of the
uneven performance of the economy over time.

Farmland Rent
The reason why Ricardian Rent has lost its economic sting is
implied by the increases in labor's income share, in the declines in
the income share accruing to property assets, and in the marked
decline in the income share attributed to rental income, which fell
from 9 to 3 percent between 1900-1909 and 1970. Farmland rent
that is attributed to the original properties of the soil has become
an exceedingly small part of U.S. national income. As a con-
sequence, the social and political influence of farm landlords has
become minute.

Time Allocated to Work
Price and income effects of hourly earnings explain a wide

Theodore W. Schultz

array of changes in the allocation of time. When expected future
earnings from more education rise, the response of youth is to
postpone entering upon work for pay in order to devote more
years to education. The advantages of youth in acquiring the
additional education is of two parts, namely the wages that youth
forego are lower than they are at older ages and there are more
years ahead for youth to cash in on the expected higher earnings
from the additional education.
As wages increase people who earn their income by working
can afford to retire at an earlier age because of the larger retire-
ment income that they are able to accumulate during their prime
working years. A counter-effect of increases in earnings is in the
improvement in health that is purchased which extends the years
that individuals may opt for work. The rise in the value of time
of women is an incentive to substitute various forms of physical
capital for their time in household production, and inasmuch as
children are for women labor intensive, the demand for children
is reduced, and an increasing part of the time of women is allo-
cated to the market for labor.
The increases in earnings also explain the decline in hours of
work or the increase in "leisure" during this century. For the U.S.
civilian economy, the average weekly hours declined from about
53 to 37 hours over the period from 1900 to 1970, and the aver-
age annual hours per employee decreased from 2,766 to 1,929
hours. The interaction between annual hours allocated to work
and earnings shows a decline of 7 percent in hours and a 43-per-
cent increase in annual earnings between 1900 and 1920, for the
1920 to 1940 period these changes were 12 percent and 53 percent
respectively, and for 1940 to 1970 there occurred a 13-percent
decline in annual hours while real annual earnings increased by 73

Toward a Population Equilibrium
One of the important implications of the high price of human
time is its effects on fertility. The argument is that the high price
of time in high-income countries accounts in large measure for
the observed decline in fertility. Most of these countries have
arrived at birth rates that are at or below replacement rates. The
argument setting forth the economics of the process that leads

toward a population equilibrium is presented in two papers, one
by Nerlove and the other by me.20

I began with the question: what have our ideas and institutions
contributed to the high value we place on human beings and to
the remarkable increases in the value of human time? I have set
the stage with a historical view to help us in examining critically
our social, political, and economic ideas embodied in our institu-
tions, as we ponder this important question.

References and Notes

It will be necessary to distinguish between the concept of human capital
when it is restricted to the abilities that people acquire at some costs and the con-
cept of human capital that includes all innate and acquired abilities of human
agents. It is the latter concept that is relevant at this point.
2 See: Theodore W. Schultz, "The Value of the Ability to Deal with Dis-
equilibria," J. Econ. Lit. 13, Sept. 1975.
3 These are updated estimates of Albert Rees of the total compensation per
hour at work of manufacturing production workers in 1967 dollars. More will be
said about these estimates presently.
4 The upward trend of real hourly wages, as measured by Rees, rises weakly
from 1900 to the middle of the next decade; it then rises sharply during World
War I, after which it rises weakly through the twenties and early thirties. There
then follows a strong upward trend for two and a half decades, after which it rises
weakly once again from the late fifties to 1970. The years which show a decline
are 1904, 1907, and 1908, and then 1914, 1919, 1921, 1922, 1925, 1932, 1945, and
5 N. Potter and F. T. Christy Jr., Trends in Natural Resource Commodities.
Baltimore: Johns Hopkins Univ. Press for Resources for the Future, 1962.
6 Professor Robert S. Manthy, Michigan State University, has been most
generous in making offsets of his numerous tables available to me. I am much
indebted to him.
7 No doubt the reader, who is strongly of the belief that the 1973-75 upsurge
in prices of the services most dependent on natural resources is the beginning of a
new era, will be inclined to look upon the indexes of prices in these tables as
bygones that are no longer meaningful for the future. The argument against this
view is that the events, both natural and man-made, that accounted for this
upsurge in these prices are in large part transitory events. The economic processes
that account for the observed prices, say from 1960 to 1972, are approximately the
more permanent relative prices that are in general likely to prevail once again
instead of the very high transitory prices of 1973-75.

8 David B. Humphrey and J. R. Moroney, "Substitution Among Capital,
Labor, and Natural Resource Products in American Manufacturing," J. Political
Economy. 83, 1975, 57-82.
SJohn V. Krutilla and Anthony C. Fisher, The Economics of Natural Envi-
ronments. Baltimore: Johns Hopkins Univ. Press for Resources for the Future,
10 Humphrey and Moroney, op. cit.
"1 Agriculture accounted for 72 percent of the total output of the extractive
industries in 1900 and for 61 percent in 1970.
12 U.S. Department of Agriculture, Economic Research Service, Farm Real
Estate Market Developments. July 1972, table 8.
13 Peter H. Lindert, "Land Scarcity and American Growth," J. Econ. Histo-
ry. 34, 1974, pp. 851-84, app. table 1. States included in this series are: Iowa, Illi-
nois, Ohio, Wisconsin, and Minnesota.
14 See: Theodore W. Schultz, "Institutions and the Rising Economic Value of
Man," Am. J. Agr. Econ. 50, Dec. 1968. See also the useful paper by Vernon W.
Ruttan, Integrated Rural Development Programs: A Skeptical Perspective. Agri-
cultural Development Council, New York, 1975, reprinted from International
Development Review. 17, 1975.
15 Simon Kuznets, Modern Economic Growth. New Haven: Yale Univ. Press,
1966. Chapter 4, pp. 181-183, bears directly on this analytical issue. This part of
the analysis is restricted to the United States and to the period from 1909-14 to
16 Simon Kuznets' studies of economic growth and the distribution of income
are classic contributions to this subject. See: (1) "Economic Growth and Income
Inequality," Am. Econ. Rev. 45: 1-28, March 1955; (2) "Quantitative Aspects of
the Economic Growth of Nations: VIII Distribution of Income by Size," Econ.
Development & Cultural Change. 11 (11): 1-80, Jan. 1963; (3) Modern Economic
Growth. New Haven: Yale Univ. Press, 1966; and, (4) Economic Growth and
Nations. Cambridge: Harvard Univ. Press, 1971.
17 U.S. Bureau of Economic Analysis, Long Term Economic Growth, 1860-
1970. Washington, D.C.: 1973, p. 22. Compensation of employees includes income
accruing to persons in an employee status such as wages and salaries, tips,
bonuses, commissions, vacation pay, and payments in kind. Also included are sup-
plements and fringe benefits such as employer contributions to private pension,
health, and welfare funds.
'8 We shall restrict the concept of earnings to the income that accrues to
human agents as compensation for their productive services. The income accruing
to the owners of property assets for the productive services of their property will
be referred to as property income.
19 U.S. Bureau of Economic Analysis, op. cit.
20 In Economics of the Family: Marriage, Children, and Human Capital.
Chicago: Univ. of Chicago Press, 1975, edited by Theodore W. Schultz. See Marc
Nerlov, "Toward a New Theory of Population and Economic Growth." pp. 527-
545. Also: Schultz, "Fertility and Economic Values," part II, pp. 14-20, deals spe-
cifically with "the high value of human time: population equilibrium."

Policies for

Equitable Growth

In Developing


Irma Adelman

Lecture 2
May 17, 1976

This special adaptation of Dr. Adelman's Bicentennial Year Lecture was pre-
pared by Roberta van Haeften and originally published in the October 1976 issue
of Agricultural Economics Research (Vol. 28, No. 4).

After two decades of attempts at raising per capital GNP in
developing countries, the development community has recently
shifted its focus to the challenge of increasing the equity of
income distribution. This shift in emphasis was needed. Empirical
studies showed that benefits from economic growth did not trickle
down. More serious, economic growth has sometimes resulted in
a systematic worsening (both relative and absolute) in the position
of persons receiving the least amount of income.
The accompanying table tests major growth and equity hypoth-
eses generated by each of three recent studies:
1. A cross-sectional, statistical analysis (A-M in the table) of
the sources of differences among countries in the relative amount
of income received by the poorest 60 percent of households.'
2. An historical analysis (Hist in the table) of processes and

Irma Adelman

initial conditions leading to extreme poverty in 24 countries in the
middle of the 19th century.2
3. A model (A-R in the table) of the South Korean economy
developed and used to explore the effects' of various major (but
nonrevolutionary) strategies, policies, and programs on poverty.
In the following table, the interaction with income distribution
of each variable or influence is described; the studies on which the
description is based are noted; and the reasons for the impact on
income distribution are summarized.
All three studies differ completely in methodology and in the
settings in which the relationships between poverty and growth
are investigated. Yet, all three lead to the following remarkably
consistent and reinforcing set of policy conclusions:

Professor of Economics at the
Utnisersit of Maryland. Irma Adel-
man has focused her career on the
economics of developing nations.

Economist at the World Bank. Fel-
low at the Center for Ad',anced
Study in Behaioral Sciences. and
Professor of Economics at North-
western Uni.ersitt. Economic
(irovi th and Social Equity in Devel-
oping (Countrnes is one of four books
authored hi this Ph.D. graduate of
the Unisersity of Calitornia. %whose
research interests include income dis-
tribution and economic growth in
i [, developing countries.
Irma Adelman

Hypotheses concerning the impact of economic processes on income distribution

Variable or influence

Per capital GNP and level of
socioeconomic development

Short-term rate of growth of
per capital GNP

Economic innovations

Impact and supporting studies

For poorest 60 percent of population, their relative share
shows U-shaped relationship to per capital GNP (A-M).
Relationship between absolute per capital income and
income distribution complex; very large differences in
development levels positively related to both average
income and reduction in poverty; but rapid growth tends
to increase poverty even when average income rises (Hist
and A-M).

No simple association with income distribution; structure
and composition of the increase in income, not average
degree in income change, determines impact (A-M and A-

At low and medium-sized levels of development, net
impact of structural change of any kind on the poor is
systematically unfavorable, even when average incomes

Reasons for Impact

At very low levels of development, a small number of growth
points leads to concentration of benefits of economic change
in hands of oligarchy of merchants, industrialists, and planta-
tion owners; indirect effects of economic change tend to hurt
the very poor by displacingand marginalizing their activities.
Only beyond a threshold determined by extent and spread of
expansion ofeconomic opportunities is the trickle-down
effect sufficient to raise average incomes of urban wage earn-
ers and agricultural laborers; even then, extreme poverty is
reduced very slowly because the labor market is segmented;
human skills lack adaptability.

Nature, extent, and rapidity of structural change govern the
direction and magnitude of the net balance of the processes
of displacement, absorption, and social adaptation.

The poor do not have resources and skills to take advantage
of expanding economic opportunities; the operation of prod-
uct and factor markets tends to marginalize their earnings or
displace their skills. Once skills have been permanently mar-
ginalized or products displaced, the poor lack human and
financial capital for adaptation.

Socioeconomic dualism

Natural resources

Rate of population growth

Rural-urban migration

Sharply dualistic growth favors the rich and harms the
relative and absolute position of the poor more than do
lesser degrees of dualism (whether at very low or higher
levels of development).

A favorable ratio of agricultural land to population, with
family-size holdings prevalent, restricts extreme poverty;
abundant natural resources are associated with extreme
concentration of income (A-M and Hist).

Past patterns of population growth are an important
determinant of resource-population ratios and thus a key
initial condition determining extent of extreme poverty
(Hist and A-M). In the medium run, population growth
affects significantly the net balance of displacement and
absorption processes (Hist and A-R). In the short run,
population growth is insignificant to income distribution
(A-R and A-M).

Rural-urban migration reduces rural poverty and increases
urban poverty; up to a point, it reduces overall poverty
(Hist and A-R).

Concentration of growth in limited sectors or regions, in the
presence of segmented labor markets characteristic of under-
developed countries, aggravates displacement and mar-
ginalization of skills, thus contributing to pools of surplus

Availability of reserve of unappropriated agricultural land
restricts numbers of extremely poor; presence of abundant
natural resources fosters exploitation and appropriation of
the benefits by small elites, foreign and indigenous.

Extremely long lag (50 years or more) occurs before changes
in rate of population growth affect extent of extreme poverty
significantly; medium-run effects depend on interaction of
population and migration given the availability of resources.

Rural-urban migration in labor market reduces urban wages
or increases urban unemployment, while decreasing rural
underemployment or increasing wages of agricultural labor;
in commodity markets, such migration shifts the terms of
trade in favor of agriculture, thus favoring the more numer-
ous and poorer rural low-income groups; extremely rapid
migration leads unfavorable impact on the urban poor to
dominate favorable rural effects.


Widespread education and literacy are associated with
larger share of income of middle quintile; no systematic
association occurs with other features of income distribu-
tion (A-M and A-R).

Impact of change on the poor critically depends on distri-
bution of landownership; prevalence of subsistence farm-
ing where land is abundant favors distribution of income;
parcelization of land or marked concentration of own-
ership with cultivation by either landless laborers or sub-
sistence tenants contributes directly to poverty; widespread
owner-operation of commercial farms favors distribution
of income, although it may not help the very poor (A-M
and Hist).

Increases in agricultural productivity tend to worsen the
position of the rural poor, while benefiting better off
farmers and the urban poor (A-R and Hist). No signifi-
cant cross-sectional relationship exists.

Total absence of commercialization of agriculture favors
distribution of income (Hist). Among countries with some
commercialization, no systematic relationship exists
between commercialization and distribution of income
(Hist and A-M).

Widespread education spreads ownership of human capital,
distribution of wage income,and increases rural-urban
migration; thereby shifting population to higher income areas
and improving agricultural terms of trade.

Concentrated ownership where cultivators face lack of alter-
native employment permits high rate of appropriation of sur-
plus product; opening up of new opportunities in presence of
unequal abilities to respond widens inequality; processes con-
ducive to dispossession of land or marginalization of eco-
nomic activities tend to be irreversible.

Access to complementary resources (held only by better off
farmers) is necessary for adoption of improvements; increases
in output worsen terms of trade, harming rural poor and ben-
efiting urban poor; no cross-sectional relationship exists due
to interdependence between impacts of agricultural tech-
nology and land tenure.

Impact on income distribution of commercialization of agri-
culture depends on land tenure and concentration of own-
ership. It also depends on the course of terms of trade; short-
term and medium- or long-run impacts may diverge,
depending on demand and supply elasticities.

Land tenure and holdings

Modernization of agricultural

Size of subsistence agriculture

Trade and industrialization

Level and change in industri-

Market socialism

Effectiveness of and
improvements in financial

Effectiveness of and
improvements in tax systems

More diversified labor-intensive exports are associated
with a larger share of income to both the poor and the
middle classes (A-M and A-R).

U-shaped relationship is evident in cross-sectional and his-
torical studies. Major differences in level of industri-
alization are negatively associated with extent of poverty,
but among more industrialized countries, poverty is
greater in those at higher levels which industrialized most
rapidly (A-M and Hist).

Share of income of top 5 percent is reduced; share of 15
percent is increased (A-M and A-R). Overall growth rate
and overall mean incomes are reduced but relative and
absolute incomes of lowest 70 percent are higher in the
medium run (A-R).

No systematic association with income shares of the poor
exists even though the effects are important for growth
(A-M and A-R).

No systematic association with distribution in cross-sec-
tional study exists (A-M). Little impact emerges in the
model (A-R).

Agriculturaltermsoftrade improveandrural-urban
migration is more easily absorbed.

The relationship iscomplicated: interdependence exists
between level and rate and threshold effect.

Profits and interest accruing to top 5 percent are reduced;
technocrat and bureaucrat incomes are increased. Given
nationalized firms which are less dynamic, absolute gains to
the poor evident in the medium run are eventually eroded,
but distributional gains increase.

Spread of financial institutions even in rural areas tends to
benefit primarily those who are better off. Improved access to
credit for poorfarmers is only beneficial when combined with
improved access to technology and knowledge. Increases in
investment, even in small-scale industry, tend to work
through their impact on economic growth (relationship of lat-
ter to income distribution is complex).

Structure of tax system influences who is poor and who is
rich rather than how many; tax base is sufficiently low so
that there is little scope for impact.

1. Intervention to improve the distribution of income is
extremely difficult. The Korean policy model experiments empha-
sized both the stability of the size distribution of income despite
policy interventions and also the ephemeral nature and ineffi-
ciency of most single-pronged antipoverty programs. The cross-
sectional study indicated how few potentially effective policy
instruments existed and what difficulties their very makeup posed
for persons intent on purposive change.
2. Within a given set of structural conditions, it is strategy and
process which determine the impacts of economic change on the
poor. The historical study indicated, for example, that countries
sharing a particular mix of poverty also shared a set of historical
processes of change. The Korean policy model experiments
showed that as long as policy interventions were tacked onto a
given strategy which remained unchanged, the distribution of
income tended to revert to the pattern it would have had in the
absence of interventions.
3. Successful antipoverty policy does not merely entail
choosing the right development strategy. In the cross-sectional
study, for example, some policy variables proved important only
when countries had achieved minimal levels of education and
spread of economic modernization. The historical study also
showed that the impact of economic change on the poor depends
critically on the type of social structure and social responses to
economic change-such as social constraints on population
growth, the response of fertility and migration rates to changing
economic opportunities, legal and customary barriers to the sub-
division of land, arrangements for land tenure and holding, and
the strength of extended family protection of the unemployed and
4. Unbalanced growth strategies are bad for the poor. Histori-
cally, the concentration of growth in a few sectors or regions has
had backwash effects that have accentuated the overall displace-
ments of economic activities arising from commercialization and
5. A systems or general equilibrium approach is required to
design a strategy which predictably improves the position of the
poor in the medium term. Both the historical and South Korean
modeling studies emphasize the importance of indirect effects and
dynamic interactions. Not infrequently, the indirect effects of an

Irma Adelman

initial impact of change swamp the direct effects and even reverse
their direction.
Taken together, results of the three studies underline the great
difficulty which planners face in finding policy instruments that
effectively bring about more equitable paths of economic growth.

References and Notes

I Irma Adelman and Cynthia Taft Morris, Economic Growth and Social
Equity in Developing Countries. Stanford: Stanford Univ. Press, 1973.
2 Irma Adelman and Cynthia Taft Morris, "Economic Growth and Impover-
ishment in the Middle of the Nineteenth Century," Econ. Jour. (forthcoming).
3 Irma Adelman and Sherman Robinson, Planning for Equitable Devel-
opment. Stanford: Stanford Univ. Press (forthcoming).

The Fear of


Charles E. Bishop

Lecture 3
June 15, 1976

The U.S. Department of Agriculture is one of the Nation's old-
est and foremost agencies in the generation and dissemination of
knowledge. Its Economic Research Service was a pioneer in the
application of economics to the problems of rural people. It
seems appropriate, therefore, that I should focus my attention in
this bicentennial lecture on the changing emphasis placed upon
education in our society with particular reference to the accept-
ance of intellectuals in society, and the changing relationships
between our institutions of higher education and our government.
Intellectuals have generally been held in relatively high esteem
in our country, and we have placed heavy emphasis upon provid-
ing opportunities for higher education for the public. The extent
of this support, however, has varied sharply over the past 200
Some of the greatest minds of 18th century Amer-
ica-including Jefferson, Hamilton, and Madison-once collabo-
rated in preparing a speech for George Washington that was
never delivered, but which was carried in various newspapers in
1796. In this address were the following sentences: "Promote then
as an object of primary importance institutions for the general

Charles E. Bishop

diffusion of knowledge. In proportion as the structure of a gov-
ernment gives force to public opinion, it is essential that public
opinion be enlightened."
Think of the significance of those words as we observe the
bicentennial of our country. They were the words of a group of
brilliant intellectuals intent upon founding a major new nation.
And two centuries later, the "institutions for the general diffusion
of knowledge" whose establishment these men demanded have
created a knowledge-based and knowledge-oriented society with-
out precedence in the annals of history.
It is therefore with considerable irony that we note that a soci-
ety founded and nurtured by intellectuals, a society which has
benefited so greatly from a spectacular development and utiliza-
tion of knowledge, also is a society in which a spirit of anti-intel-
lectualism has periodically slowed economic and social devel-
opment. In the recent past, anti-intellectualism has gained
renewed strength in our country, and I believe that this has par-
ticularly foreboding portents for a society so heavily knowledge-
based. I am concerned that this new surge of anti-intellectualism
is based not only upon skepticism of intellectuals but a growing
fear of their product: knowledge. I wish to address myself to that
fear on this occasion: to examine why and how it has arisen, and
to suggest some ways in which we should combat the nihilistic
tendencies it creates.
The early leaders of our country were remarkably wise in the
emphasis they placed on education and social enlightenment.
Their concepts and their public pronouncements exemplified a
spirit of American intellectualism in which freedom of thought
and inquiry contrasted sharply with the Puritan dogmatism of an
earlier era. Moreover, they were able to reconcile this wide-rang-
ing intellectualism with a practical approach to political activity.
The result was a blending of intellectual idealism and political
pragmatism which guaranteed a progressive form of government
for the fledgling republic. The founding fathers demonstrated that
men of learning were valuable to good government, so valuable,
in fact, that the fashioning of the bold new experiment of the
American democracy could be entrusted to them and to no oth-
ers. The fear, in that day, was of a lack of knowledge, not of its
acquisition and its practical use.
The initial golden years of intellectualism in American govern-

ment lasted until the Jacksonian era when anti-intellectualism
made its first serious appearance. While much was accomplished
under the Jackson presidency, and while his capable leadership
gave rise to a new and broader emphasis on democracy in our
country, the concepts of quality and equality were often blurred
in the government he headed. Intellectual development in Jack-
sonian America became resented as a boorish kind of excellence,
as an unjustified claim to distinction, as a challenge to egalitar-
ianism, and as a quality that somehow deprived a man or woman
of what was known as the "common touch." As stated by Rich-
ard Hofstadter, there was, in the Jackson administration, "an
estrangement of training and intellect from the power to decide
and manage."'
In the years between the Jackson and Lincoln administrations,
there was a general intellectual reawakening, a quickening of the
spirit in the pursuit of knowledge. Part of this was attributed to
industrial and agricultural expansion, and the call for new knowl-
edge to enhance this expansion. But there also was a spirit of

Now President of the University
of Arkansas. Dr. Bishop has been
Department Head and Reynolds Dis-
tinguished Professor at North Car-
olina State University: Executive
Director of the President's National
Advisory Commission on Rural Pov-
erty. President of the American Agri-
cultural Economics Association: and
Chancellor at the University of Mary-
land. This distinguished agricultural
economist's research career focuses
on economic development, labor
mobility,. human development, and
income distribution.

Charles E. Bishop

Charles E. Bishop

intellectual liberality exemplified by such men as Emerson, Long-
fellow, and Hawthorne. It was Hawthorne who identified and
promoted the intellectual flowering of his day with his statement,
"Nature's highest purpose for man is that of conscious intellectual
life and sensibility." And it was Emerson who said: "What is a
weed? A plant whose virtues have not yet been discovered."
These were the years also when the vast lyceum movement pro-
vided a stimulus to intellectual curiosity, and when the country's
leadership suggested the possibility of a nationwide, publicly sup-
ported system of higher education. It was from this movement
that the land-grant system emerged. Credit for the establishment
of the land-grant system rightfully belongs to Justin Morrill and
his colleagues in the Congress. But somehow it seems particularly
appropriate that the Land-Grant Act was passed during the
administration of a president who, like none before or after him,
exemplified the concept that intellectual development was demo-
cratic and egalitarian, and that it should be the universal aspi-
ration of all men and women, regardless of their stations in life.
The image of Lincoln, reading his books by candlelight, will for-
ever portray the intellectuality that springs from the uncommon
motivation of the common man. And the land-grant movement
provided the universally available opportunity for education that
Lincoln envisioned for all his countrymen.
From Lincoln's time until after the turn of the century, our
country underwent a great economic and industrial expansion and
transformation. During the latter part of this period, especially,
politics and government were oriented toward the acquisition of
wealth and power. During this period there also was a substantial
reorientation and growth in our educational institutions. Indeed,
the first thrust toward reform in scandal-ridden government in the
late 1800's began in the halls of ivy on the college campuses, just
as it did in the late 1960's.
One of the students at Harvard eventually became the second
great intellectual leader in the Republican Party. Early in life, he
developed a crusading spirit of reformation. Today, he is recog-
nized more for his exploits of physical leadership than for his
intellectual achievements. It is, therefore, sometimes forgotten
that Theodore Roosevelt was an intellectual who recognized that,
in his time, only a leader who appealed to the people on the basis
of his physical characteristics could be successful in bringing an

intellectual and rational approach into government. Indeed, there
was a general feeling in Roosevelt's day that intellectuals should
remain aloof from government, that the brighter minds in the
country tended to become fuzzy the moment they became
involved with the problems of government. However, Roosevelt
was exceedingly well-read and well-informed. And, as he rose in
political stature, he initiated reforms that strengthened the
Nation's economic development by changing its orientation
toward the common good, rather than toward the benefit of the
privileged few.
Ironically, the first modern-day Democratic intellectual in the
presidency-Woodrow Wilson-held a deep distrust toward other
intellectuals. This distrust probably came from the fact that he
was a university president for 8 years, and commitments of that
office prevented him from maintaining his own intellectual pro-
Following Wilson's administration, the country went through
the intellectual drought of the twenties and thirties. The Ku Klux
Klan gained strength. The Scopes Trial comprised an all-out
attack on intellectualism. And, the depression years discouraged
the development of higher education.
Then came the administration of Franklin Delano Roosevelt,
and the country for the first time since the days of Jefferson and
Hamilton saw a major influx of intellectuals into government.
While some antipathy was displayed toward such members of his
"brain trust" as Rexford Tugwell and Raymond Moley, FDR
nevertheless created an unprecedented harmony between the pop-
ular cause in politics and the dominant mood of the intellectuals.
The Ivy League schools provided much of the reservoir of talent
that went to Washington. Many of these bright young men were
the Harvard proteges of Felix Frankfurter. They aroused much
antagonism among the traditional political pros of the country,
but they brought the fresh winds of innovation into govern-
mental, economic, and social processes. Unquestionably, they
changed the shape and direction of government, and the country
itself was changed as it had never been before.
The pendulum then swung full cycle, and in the early part of
the fifties the country was subjected to the "great inquisition" of
McCarthy. It probably was the most intense surge of anti-intel-
lectualism since the Jacksonian era, and it represented a particu-

Charles E. Bishop

larly disturbing and damaging right-wing syndrome. Intellectuals
were automatically equated with a supposed communist con-
spiracy. Any intellectual in government was subjected to the most
frustrating kind of suspicion. Even a dedicated stalwart like Dean
Acheson, who was strongly supported by such patriotic cham-
pions as Harry Truman and George Marshall, became suspect in
the minds of many simply because he was classified as an intel-
The aftermath of the McCarthy era was so great that there can
be little doubt of its tragic effects on Adlai Stevenson, certainly
one of the great intellectuals in American political history. Most
of the great minds of the country flocked to the Stevenson banner
during his two unsuccessful campaigns for the presidency against
an immensely popular wartime leader. During the heat of the
campaign it is reported that an aide seeking to encourage Steven-
son said to him, "There is no cause for concern. You have the
support of all thinking Americans." Stevenson quipped, "That is
not enough. I need a majority." Stevenson's intellectual capacity
for leadership was unexcelled, but he was in the right place at the
wrong time and the Nation will never know the impact his lead-
ership could have had on its future.
The G.I. Bill had a major impact upon higher education fol-
lowing World War II, making opportunities for higher education
a reality for millions of young men and women. But the edu-
cational system in America did not progress as rapidly as it did in
some other countries. We became painfully aware of our com-
placency about education when the implications of the launching
of Sputnik became obvious. Immediately, President Eisenhower
moved to strengthen our educational system.
When John F. Kennedy entered the White House, intellectuals
hailed the beginning of an "American Rennaissance." Not only
was Kennedy an intellectual, but he was an aggressive and deci-
sive leader, giving strength and purpose to his intellectual ideals.
During the Kennedy administration, academicians who by nature
had been rather skeptical of politicians, and vice versa, eagerly
responded to the beckoning of a strong government committed to
bold new ideas. The stream of academics to Washington during
this period was so great that they soon turned from a feeling of
being unwanted and unused to the opposite extreme of feeling

Highly significant was the Kennedy administration's conscious
decision to expand the Nation's educational programs and to
commit vast new funding to scientific and technological research.
As confidence in science rose, the Nation sought to attain such
diverse objectives as conquering space, increasing human longev-
ity, removing discrimination toward minorities, developing the
underdeveloped nations, feeding the hungry, and abolishing dis-
ease and poverty.
Faced with the commitment to solve complex problems and
the necessity of having access to the best available knowledge, the
government turned increasingly to the universities and to mem-
bers of their faculties for advice. Among academicians there was a
growing feeling that knowledge was useful, that new knowledge
was needed, that the search for it would be supported, and that in
the process of generating it and making it accessible to the public,
the academician would gain enhanced stature in the society.
Yet, it is a characteristic of the American people that we are
impatient. As a society we are prone to leap into something with
tremendous gusto and great expectations without considering
important consequences. There is no doubt that during the 1960's
the public became overly imbued with research and education.
Expectations were developed that could not be fulfilled. Many of
the goals pursued by the Nation proved to be very elusive. As the
public recognized that the problems of society were acute, but
much more difficult to resolve than anticipated, enchantment with
research diminished and investment in research was increasingly
While the results of research formed the guiding assumptions
for many ventures by government, the ties between the ideas of
the scholars and public policy often proved to be very tenuous.
All too frequently the research information of the scientists was
fragmented, contradictory, and inconclusive. There also was a dis-
tinct difference in orientation of the scientist, serving as an
adviser, and the orientation of policymakers. The scientist is ori-
ented toward positivistic thinking, and is most effective when
dealing with measurable phenomena. He is accustomed to think-
ing and working within the protective confines of highly con-
trolled conditions. Public policy, on the other hand, is always
subject to the vagaries of politics, and normative considerations
weigh heavily on the calculus of the politician. The goals adopted

Charles E. Bishop

by the Nation during the early 1960's were so diverse that con-
flicts between political and academic considerations and the
necessity for harsh compromises soon became apparent.
The emerging strain between the academics and the poli-
cymakers was magnified by the fact that once the Nation had
committed itself to achieving goals that required the generation of
additional knowledge through research, a serious mismatch devel-
oped between the time required to carry out creative research and
the attention span of politicians committed to particular policy
objectives. Following the success of the space program and the
apparent failure of the poverty and manpower programs, the gov-
ernment sought to achieve a better matching of research products
and social needs by placing more emphasis upon "mission-ori-
ented research."
Faced with the complexities of achieving the pre-established
goals of social policy and the inadequacies of their research infor-
mation, the academicians advising government became frustrated
and their optimism rapidly dissipated into despair. At the same
time, government became more hesitant about seeking the advice
of the academician. As the decade of the 1960's drew to a close,
the ties between the intellectuals and government had weakened
During the Nixon administration the welcome mat, once
proudly displayed at the White House door for the President's
Science Advisory Council, was unceremoniously yanked away and
the Council was abolished. This had a devastating effect on the
morale of scientists who became painfully aware, of the extent of
their ostracism and alienation from government and its programs.
Increasingly, questions were asked about the utility of knowl-
edge. The rapidity of the expansion of knowledge in the past
three decades prevented adequate adjustment and adaption to the
new conditions it spawned. The result has been what we have
called a "culture shock," and it has compounded fears toward
intellectualism. Today, attacks against new knowledge are coming
from the left as well as from the right. The recurring resentment
against intellectuals which we have come to expect at various
stages in our history has been transformed into something much
more disturbing: a very significant fear of the new knowledge they
are creating. Part of this fear results from inability to retain the
traditional values of another time because of the value changes

"...retreating into a technological
Eden that never existed."

wrought during the sixties. Part of it stems from the foreboding
aspects of misguided technological development and its effects on
the environment. And part of it is the terror instilled in all human
hearts by the development of nuclear power. As a consequence of
these and other factors, the generation of knowledge is falling
into disrepute. Intellectuals are no longer ridiculed, as they once
were, but for a substantial proportion of the population they are
suspected and feared: resented for the changes they have created
and feared for the changes they will create in an uncertain future.
More knowledge suddenly is equated with the decline in moral
standards, the disasters of runaway technology, and the growing
complexities of all aspects of human life.
Perhaps even more significant is the growing concern of sci-
entists and academicians over the expansion of knowledge. The
awesome possibilities of high-energy research are widely recog-
nized in the academic community. Recently, knowledge in the
biological sciences concerning the nature of life and the produc-
tion of it has progressed so far that an increasing number of sci-
entists and academicians are questioning whether we wish to
know more. Underlying this questioning is a growing fear con-
cerning how additional knowledge would be used.
The fear of knowledge, however, is misplaced. Our fear is not
of knowledge. It is a fear of how knowledge will be used once it
becomes known. Underlying this fear, therefore, is a concern for
our own moral weakness and that our knowledge will not be used
for the benefit of society.
In the mood of our country today, fear of knowledge finds
expression through a wistful desire among large segments of our
population to erase some inevitabilities, and to return to the sim-
pler days of a bygone era. Thoughtful people like Lord C. P.
Snow point out the impossibility of "retreating into a tech-
nological Eden that never existed," but increasingly prevalent
reaction is that, while intellectual oracles of Snow's caliber may
be right, their messages should be shoved into the apathetic cor-
ners of our individual and collective consciousness. Thus, the cre-
ators of knowledge who once were the heroes of the American
dream are now increasingly being cast as the villains in a night-
mare of frightening intellectual development. Pleasurable wonder
with the beneficial miracles of scientific creativity is giving way to
the creeping dread of "future shock." And the fear snakes its way

Charles E. Bishop

deeply into the heart of our society.
What are the lessons to be gleaned from the changing
relationships between intellectuals and the public? First, what our
society expects of its universities remains unclear. The utilitarian
ideas of Franklin and Jefferson concerning the university's ser-
vices to the community and to the Nation have not been univer-
sally accepted either on university campuses or in government.
There is a general feeling that the interrelated functions of a
university, including the generation of knowledge through
research, the transmission of knowledge through instruction, and
the application of knowledge through public service, should con-
tinue to characterize the university. The importance of the univer-
sity as an idea-generating institution is recognized, but increasing
emphasis is being placed upon innovative ways of testing these
ideas and making the results available to the community.
Although this is an especially important function for a publicly-
supported university, there are widely divergent views with respect
to how this function should be carried out.
Second, the knowledge base of the university is woefully lack-
ing relative to the complexity of the problems faced by the mod-
ern society. Traditionally, our secondary schools, and to a lesser
extent our universities, have been cognitively oriented. But in our
universities we have been most comfortable when our teaching
emphasized basic concepts and when we were communicating to
students the scholarly disciplines of the natural sciences, the social
sciences, and the humanities. Our academic base has been
noticeably detached from the world around us, giving rise to the
outcry of students and the public for more relevancy of higher
education to the needs of society.
Most professors contend that the greatest service that formal
education can render students is to cultivate in them the ability to
think critically. Teaching them how to think is regarded as
infinitely more important than teaching them what to think
about. Formal education, therefore, has leaned heavily upon the-
ory in pedagogical practice. While heavy emphasis upon theory
continues to be necessary in teaching how to acquire knowledge,
hypothesize about it, assess it, and integrate it into our thinking,
it is increasingly apparent that students and the public generally
desire a better balance between theoretical and empirical analysis.
In many areas, empirical research is of recent vintage, and the

"Social scientists...should get out
of the business of huckstering
proposals for new policies."

findings to date have limited applicability. In some areas, and
especially in the social sciences, it is often necessary to conduct
research within a highly restricted framework of abstractions.
Extrapolation of the findings to the world of the policymaker
often results in shattered expectations and disillusionment.
The academic's frustration is exemplified by Moynihan's state-
ment as he left the White House that, "Social science is at its
weakest, its worst, when it offers theories of individual and col-
lective behavior which raise the possibility of controlling certain
inputs, or bringing about mass behavioral change. No such
knowledge now exists. Evidence is fragmented, contradictory and
incomplete." He went on to conclude that, "Social scientists
should stick to monastic studies of results, emphasizing which
programs work and which don't, and they should get out of the
business of huckstering proposals for new policies."
Clearly, more comprehensive descriptive and predictive anal-
yses are necessary in refining and improving theory and public
Third, we are at a stage in the development of our society that
is of unparalleled significance. While we are becoming more
dependent upon new knowledge, we are also becoming more
mindful of unanticipated consequences of the generation and use
of new knowledge that is developed under highly controlled con-
ditions but applied under general conditions. The consequences of
indirect effects of scientific and technological developments are
cause for concern to an ever larger spectrum of the population.
As a result, there is a growing concern throughout the land that
this be a moment when humanism must gain the ascendancy. We
are being challenged to guide and direct the geometric progression
of knowledge toward civility and humanity, not toward their
destruction. Clearly, we need a new humanism, written in a phi-
losophy of history that reflects the course of our past devel-
opment, but more importantly that will guide our future. In the
search for this new humanism, however, we must recognize that
there is no respite from the continued development of science and
technology. One of our most pressing needs is that intellectuals
chart such a humanism for the society.
Fourth, by nature, public policy is controversial. The essence
of public policy involves conflict, compromise, and consent. A
functionally effective democracy requires an intelligent and

Charles E. Bishop

enlightened citizenry capable of acting wisely in matters of public
policy. In an open society the free flow of new knowledge should
help to create citizen awareness of issues and citizen participation
in democratic processes. Such a flow of information should be
pursued resolutely, and the research and education programs of
universities must contribute greatly to this process.
But public policy is not determined by science. It is based upon
value judgments concerning what is good and bad and how soci-
ety should be structured and operated, as well as upon belief with
respect to fact. These valuations often are deep-seated, conflicting,
and powerful, and the hierarchy of valuations differs among indi-
viduals. Consequently, given the same data, analyses, and sci-
entific conclusions, individuals and groups may arrive at different
decisions with regard to the most desirable policies.
Fifth, we cannot assume automatic support of knowledge gen-
erating and disseminating activities. The universities must earn the
support, confidence, and public trust vested in them.
There is always the danger that programs of the university may
be out of phase with public support and public needs at any point
in time. For this reason alone, many pure scientists are concerned
that in playing the role of adviser to government on matters of
public policy, scientists can endanger public support of science
programs. The scientist knows that he must always reserve the
right to fail in his scientific investigations. The policymaker, on
the other hand, is not likely to be accorded this privilege. It is
important, therefore, that the role of the academician serving as
policy adviser be clearly differentiated from that of the poli-
As was indicated earlier, the current concern over research and
technological development derives in part from a growing feeling
that important secondary effects often are not given due consid-
eration in policy decisions. There also is a concern that scientists
have not been able to provide solutions to major problems that
continue to plague us. The low esteem accorded our institutions
of higher education also undoubtedly stems in part from the fact
that much of the anti-war sentiment of the late 1960's was cen-
tered in our colleges and universities.
Whatever the causes, it is clear that over the past 200 years
there have been rather pronounced vacillations in public atti-
tudes toward intellectuals and toward higher education generally.

"...our Nation can ill afford an
extended period of estrangement
between its scientists and its gov-

In a democracy, such changes in public opinion are not uncom-
mon. And, when anti-intellectualism has become strong or when
it has become clear that the Nation was under-valuing education,
corrective changes in public opinion occurred. Often, however,
the period of adjustment has been quite long. Faced with the mul-
tiplicity of complex problems of our time, our Nation can ill
afford an extended period of estrangement between its scientists
and its government.
Sixth, the academic must link the university to the other insti-
tutions in the American society. In the past a great gulf has exis-
ted between the academician and those in positions to make pub-
lic policy. American academicians have been particularly wary of
persons in positions of high authority in government. In like man-
ner, those in authority in government have been skeptical of aca-
Such a distance was accepted and probably was desirable in an
earlier era. However, the problems of the complex society in
which we live clearly require that the knowledge base be
expanded and that access to the best knowledge available be facil-
itated. If our society is to have this access, the trend toward alien-
ation of the academic community from the policy arena must be
halted and reversed. The initiative for this reversal must come
from those who are committed to the life of the mind. Only in
this way can the stream of knowledge coming from the univer-
sities be perpetuated and integrated into the central institutional
system of our country.

References and Notes

I Richard Hofstadter, Anti-Intellectualism in American Life. New York:
Vintage Books, 1962, p. 171.

Measuring Economic

and Social Performance:

New Theory,

New Methods,

New Data

Karl A. Fox

Lecture 4
June 22, 1976

To measure the performance of any economic unit, large or
small, we need to measure the quantities of its inputs, the quan-
tities of its outputs, and the prices at which they are (respectively)
bought and sold. Economic accounts for a nation are based on
the same principles as economic accounts for a firm.
To measure the performance of any social organization, or a
community, or an entire society, it seems to me we need direct
counterparts of the quantities, prices, and values that are required
to measure economic performance. The problems of concep-
tualizing and measuring these quantities, prices, and values will be
the subject of the last-and longest-portion of my lecture.
However, before I settle down to that task, I want to lay the
groundwork for it by covering three other topics rather briefly:
(1) The priority and excellence of the quantity, price, and value
measures developed for U.S. agriculture and closely related sec-
tors during 1922-53 by the former Bureau of Agricultural Eco-
nomics (forerunner agency of the Economic Research Service).
(2) The revolution in economic theory and econometrics in the
1930's and 1940's which led to the creation of national income
and product accounts.

Karl A. Fox

(3) The current revolution in sociometrics which is bringing
sociologists and economists closer together and which may lead to
a new synthesis of concepts, methods, and data covering the
social system as a whole, including the economy.

The Bureau of Agricultural Economics
Tradition and Development of the Data
Network for U.S. Agriculture 1922-53

In the 1920's, the Bureau of Agricultural Economics (BAE)
took world leadership in the application of multiple regression
analysis to the estimation of demand and supply functions for
farm and food products, the field in which econometrics achieved
its first practical successes. Mordecai Ezekiel and Louis H. Bean
of BAE were world famous, and several of their BAE colleagues
internationally known, well before the establishment of the Eco-
nometric Society in 1930 and the publication of the first volume
of Econometrica in 1933.
The Agricultural Estimates Division (now the Statistical
Reporting Service) was in BAE from (I believe) 1922 to 1953.
Charles F. Sarle of that division did some careful analyses of
sources and levels of measurement error in agricultural price and
production statistics in the 1920's and early 1930's. O. C. Stine
(head of BAE's Division of Statistical and Historical Research
from 1922 to the mid-1940's) was keenly aware of the limitation
of the data, both primary and secondary, from which the great
econometricians in and out of BAE were estimating their demand
and supply functions. Stine's pressure for more and better data
had little glamour in the short run, but Stine had the patience
and perspective of an economic historian, which indeed he was. It
is no accident that Stine's division was named the Division of
Statistical and Historical Research.
BAE's economists were more adventurous than those in most
other government agencies. Long before the Commerce
Department began to publish time series on national income,
BAE analysts had developed some proxies for it-industrial
workers income and nonagricultural income. In the late 1920's,
Norman J. Wall worked on an index of world industrial produc-
tion to measure shifts in the demand for U.S. cotton and other

export crops. One of Stine's lifelong concerns was to explain
changes in the general price level, a problem still with us.
During the 1930's and 1940's, under the leadership of Stine,
Frederick V. Waugh, and James P. Cavin and with pressure, no
doubt, from H. R. Tolley, O. V. Wells, F. F. Elliott, and others,
there were notable developments in BAE's secondary statistics
including farm income and expenditures, cash receipts from farm
marketing, the food market basket, the index of per capital food
consumption, and the supply and distribution tables. Ernest W.
Grove was primarily responsible for developing the farm income,
expenditure, and cash receipts estimates; Richard O. Been for the
food market basket; and Marguerite C. Burke for the index of
per capital food consumption and the supply and distribution
I was lucky enough to arrive in the Washington office of BAE
in February 1945, when BAE's data system was approaching a
peak of completeness, accuracy, and relevance to economic anal-
ysis and policy. Wells, Cavin, and Waugh were my mentors and

This Distinguished Professor of
Sciences and Humanities at Iowa
State University had an outstanding
early U.S. Government career,
including 9 years as Associate Head
and Head at the Bureau of Agricul-
tural Economics. Division of Statisti-
cal and Historical Research. The win-
ner of a USDA Superior Service
Award plus four awards for out-
standing published research, Karl
Fox has authored or coauthored nine
books. Presently a Fellow of the Eco-
nometric Society and a member of
the American Statistical Association,
the American Agricultural Economics
Association, and the American Asso-
ciation for the Advancement of Sci-
Karl A. Fox ence. his wide interests include
resource allocation in higher edu-
cation, theory of economic policy,
demand analyses, and econometrics.

Karl A. Fox

supervisors; C. Kyle Randall, Nathan M. Koffsky, Rex F. Daly,
Harold F. Breimyer, and Richard J. Foote were my close associ-
ates for the next 9 years, which were the happiest years of my
professional life. With our commodity and other specialists, our
statistical clerks, and our support units, we felt we were the best-
balanced economic research and service group in the public ser-
vice. And I think we really were!
In those days I felt I could work for BAE forever, but I awoke
one morning in 1953 and it was gone! I never left BAE. BAE left
So, the picture of BAE that stays with me is essentially this:
good analysts applying good theory to good data, and making
cumulative improvements in measurement and understanding.
From the 1920's into the early 1950's (at least), the U.S. food and
agricultural sectors were better supplied with economic data and
analysis than any other sectors of the U.S. economy and better
perhaps than the corresponding sectors of any other economy.

The Great Depression and its Consequences
for Economic Science: Revolutions in
Economic Theory and Econometrics and the
Development of National Income Accounts

Between 1929 and 1932, we in the United States discovered we
didn't have the slightest idea how to steer our economy out of the
deepest depression in our history.
In 1928, after several years of prosperity under Republican
administrations, Herbert Hoover ran for president on the slogan,
"two chickens in every pot and two cars in every garage." In
1932, when the national income had shrunk 50 percent in dollar
terms and 25 percent of the labor force was unemployed, a hostile
parody appeared: "Two chickens in every garage!"
Not only didn't we know how to steer our economy out of a
depression, we didn't even know how to measure how deep the
depression was.
How bad was the measurement problem? In September 1929,
President Hoover appointed a prestigious committee to study and
report on "recent social trends" in the United States. The chair-
man of the committee was Wesley C. Mitchell, President of the

National Bureau of Economic Research and Professor of Eco-
nomics at Columbia University. Another member, who also
served as the committee's research director, was William F.
Ogburn, Professor of Sociology at the University of Chicago. A
third member was Charles E. Merriam, Professor of Political Sci-
ence at the University of Chicago. As of 1929, these men were
among the most distinguished social scientists in the United
The committee's 1,600-page report' was printed at the very
depth of the Great Depression; it devoted only two or three pages
to the national income. Its main exhibit on this subject was a
table (p. 229) showing estimates of the U.S. national income
annually from 1914 through 1928, and the figures for 1926, 1927,
and 1928 were labeled "preliminary." The text beneath the table
says that the national income estimates depend on a multiplicity
of series, many of which are published several years after the
event, but "highly tentative estimates for 1929...show an increase
over the preceding year of more than 3 billion dollars." No news
yet about 1930, 1931, and 1932!
Apparently, prior to 1930, national income statistics were
viewed as having only historical and descriptive interest, and the
Republican administrations saw no reason to spend the taxpayers'
money trying to measure the taxpayers' national income. In con-
trast, a great deal of work was done during the 1920's on "eco-
nomic indicators" by and under the direction of Wesley C. Mit-
chell at the National Bureau of Economic Research (NBER).
Each economic time series-wages, prices, pig iron production,
interest rates, and so on-was grist for the mill, and was pro-
cessed as though it were an independent indicator of something
called "the business cycle."2
The conceptual framework for processing each time series was
denoted by the formula A=TS+CR, where A was the actual value
of an observation and T, S, C, R stood respectively for trend,
seasonal, cyclical, and random or irregular components. From
this, it followed that CR=A-TS; the effects of the trend and sea-
sonal factors were to be removed (subtracted) from the original
values of the observations in order to isolate, as a residual, the
combination of cyclical and irregular fluctuations. The CR com-
ponents of the individual series, many hundreds of them, were
charted and compared visually with a so-called "reference cycle"

Karl A. Fox

which was based on an average of the turning-point dates of
many of the individual series. Mitchell was a busy man, often
tapped for public service, and the final results of his approach3
were not published until 1946; their limitations were exposed in a
classic review article by Koopmans.4
Young Simon Kuznets came into the NBER in the mid-1920's.
In a brief span of years, he wrote three books on the analysis of
economic time series: one on cyclical fluctuations, one on sea-
sonal patterns, and one on secular trends. I believe Kuznets real-
ized that with these three books he had exhausted the potential of
the "economic indicators" framework, and therefore moved into
national income accounting (NIA) as an intellectually tougher and
more promising field.
During the early 1930's, economists in several industrial count-
ries were developing ideas about "public works" and "counter-
cyclical fiscal policy" with varying degrees of theoretical sophis-
tication. J. M. Keynes' General Theory of Employment, Interest,
and Money was the clincher.5 By making national income, con-
sumption, and investment the very core of his theory, he con-
vinced many that responsible economic policy was impossible
without good current estimates of the national income and its
Keynes (in 1939) set two bright young men-Richard Stone
and J. E. Meade-to work on national income accounts under
the auspices of the government of the United Kingdom, and their
results, the first official national income accounts for the United
Kingdom, were published in 1941. Simon Kuznets had begun
publishing his national income accounts for the United States in
the mid-1930's under NBER sponsorship.6 These were extended
by the U.S. Department of Commerce which began publishing the
accounts in substantially their present form, with emphasis on the
gross national product (GNP) in 1942.7
Almost simultaneously with Kuznets' national income estimates
and Keynes' General Theory appeared Wassily Leontiefs first
input-output model of the U.S. economy8 which amounted to a
special disaggregation of the national income accounts for a single
year, and Jan Tinbergen's econometric model of the United
States9 which tried to explain movements in the national income
and its components over a series of years, 1919 through 1932.
Tinbergen's model incorporated Kuznets' national income data,

and its structure was largely based on Keynes' theory.
The work of these great economists (Tinbergen, Kuznets, and
Leontief have all received the Nobel Prize in Economic Science
and Keynes would probably have been the first to receive it had
he lived until this honor was initiated in 1969) imposed new stan-
dards of quality and consistency on economic data systems.
National income accounts entered the Federal statistical program
in the 1930's and input-output analysis in the 1940's (shakily at
first, but rather steadily after 1961).
Econometric models of the Tinbergen type in the United States
have been developed mainly outside the government, but some
Federal agencies have been experimenting with them since the
mid-1960's. To some extent, I think the Federal Government has
defaulted in this particular field, which seems rather strange in
this the 30th anniversary year of the Employment Act of 1946.
The depression fiasco of 1929-32 demonstrated the impos-
sibility of steering the economy without road maps or headlights;
the Employment Act of 1946 (it seems to me) made it illegal to
drive the economy without a license!

The Social Indicators Movement and the
Sociometric Revolution

In 1966, during the Great Society period of Lyndon Johnson's
administration, the "social indicators" movement sprang to life
under impressive leadership, including such eminent sociologists
as Raymond Bauer,10 Bertram Gross," and Philip Hauser. The
need for this movement was succinctly expressed by Bauer: "For
many of the important topics on which social critics blithely pass
judgement, and on which policies are made, there are no yard-
sticks by which to know if things are getting better or worse."12
Albert Biderman13 studied the 1960 report of President
Eisenhower's commission on national goals.14 This report listed 81
specific goals. Biderman searched The Statistical Abstract of the
United States (1962) and Historical Statistics of the United States
from Colonial Times to 1957, not for series that would actually
measure performance or progress toward these goals but for series
that were at least remotely relevant to phenomena that were perti-
nent to each goal statement. He found 48 goals to which some

Karl A. Fox

time series in these publications were relevant and 33 goals to
which none of the included time series were relevant.
In my own thinking, the social indicators movement was asso-
ciated with two others: the "accountability" movement that hit
public universities in the mid-1960's and the PPBS movement in
the Federal Government (1965-68) which required each agency to
define its program elements and objectives and identify the
resources used to promote each objective. Both movements
attacked the traditional methods of resource allocation in non-
market institutions or systems.
The accountability movement evidently required careful defini-
tion and measurement of the inputs and outputs of universities
and a rationale for assigning equivalent dollar values to those
which were not bought and sold in conventional markets.15 A sat-
isfactory solution to this problem seems to call for the gener-
alization of national income concepts to include all inputs and
outputs of the social system.'6 Thus, we might visualize a gross
social product (GSP) which includes, but is much larger than, the
GNP; also, an accounting convention which specifies that gross
social product is equal to gross social income (GSI). I will expand
these ideas in the next section.
I should like to point out that the early leaders of the social
indicators movement consciously followed economic prototypes.
In 1967, they helped draft a bill which was introduced in the Sen-
ate by Walter Mondale as "The Full Opportunity and Social
Accounting Act of 1967." It provided for an annual Social Report
of the President, a Council of Social Advisers, and a Joint Com-
mittee of Congress to review the recommendations made in the
President's report and to issue its own findings and recommen-
dations as a guide to the several committees of Congress dealing
with matters covered in this report. Hearings were held in 1967
but no action was taken.
Much of the empirical work on social indicators during the
late 1960's and early 1970's was open to the charge of "mea-
surement without theory." By 1972, several mathematical sociolo-
gists were moving to rectify this situation. Kenneth Land17 pro-
posed that social indicators be identified as components in social
system models-i.e., as variables in structural equation models of
social processes. He adopted a terminology and formulation
which are almost identical with Tinbergen's18 "theory of economic

"...explore the feasibility of
implementing a system of social
accounts...which would include
the national income accounts
and add up to a gross social
product (GSP)..."

policy," one of the most fruitful ideas in economics since the
In Social Indicator Models,19 James S. Coleman (a sociologist)
and Richard Stone (an economist) show remarkable agreement on
the use of "transition accounts." Otis Dudley Duncan, like Cole-
man a distinguished quantitative sociologist, is also represented in
Social Indicator Models along with several of his younger associ-
ates. Duncan was also co-editor (with Arthur S. Goldberger, an
outstanding econometrician) of a volume entitled Structural
Equation Models in the Social Sciences20
These two volumes are landmarks in a sociometric revolution
which is closely comparable to the econometric revolution of the
1930's and 1940's, and which is bringing sociologists and econo-
mists closer together. Social Indicator Models is an encouraging
demonstration of what sociological research will be like under the
new methodological regime. The suggestion of a Council of Social
Advisers was premature in 1967; it may not be so in 1977.

Social Indicators and Social Theory:
Elements of an Operational System

I will proceed now to outline my own tentative integration of
concepts from the various social sciences (including economics),
the social accounting system to which they would logically lead,
and the data system that would be required to implement it.
What am I aiming toward? At the most general level, I would
like to contribute to an integration of theory, methods, and data
across the "social" parts of the social and behavioral sciences. My
specific objective for the next 2 or 3 years is to explore the fea-
sibility of implementing a system of social accounts, in dollars,
which would include the national income accounts and add up to
a gross social product (GSP) much larger than the GNP. It would
attach dollar values to all outputs from the social system to indi-
viduals and all inputs from individuals to the social system. The
outputs received by individuals may be called rewards and the
inputs supplied by individuals may be called contributions. Just
as gross national product equals gross national income in the eco-
nomic accounts, so the dollar value of total rewards to individuals
(the gross social product) would equal the dollar value of total

Karl A. Fox

contributions by individuals (the gross social income) in the pro-
posed system of social accounts.
This objective would require us to define an exhaustive set of
rewards, a unit for measuring the quantity of each kind of
reward, and a method for assigning prices (equivalent dollar val-
ues per unit) to each kind of reward. It would require us to do
the same things with respect to contributions.
It is clear that social accounts require a complete accounting
for time. There is a tradition of time budget or time use surveys
going back as far as the 1920's; there is also a 1965 article by
Gary Becker21 which places time and money in a common frame-
work. Time spent in household production is a major component
of time use; time spent in school is another, time spent at work
(producing GNP) is a third; time spent asleep is a fourth; and the
rest of the 24-hour day is the fifth and final component.
T. W. Schultz, Gary Becker, James Morgan, and other econo-
mists have done good work on assigning dollar values to vocation
and career education and to household production, extending
economic theory and measurement to adjacent areas. I have taken
a more radical approach, starting with Roger G. Barker's concept
of "behavior settings" and Talcott Parsons' concept of "gener-
alized media of social interchange."
What are the behavior settings? The concept was originated by
Barker22 in the late 1940's or early 1950's. Barker spent a good
many years observing the behavior of residents of a small mid-
western community of about 830 people. He early addressed him-
self to the question of how the environment of human behavior
was to be identified, described, and measured. He concluded that
the community environment could be divided into parts or units
which he called behavior settings.
Barker says:23

"Behavior settings are units of the environment that have rele-
vance for behavior. They provide the primary data of the study
to be reported here. We have dealt only with the settings that
occur outside the homes of the community, that is, the public
behavior settings. The number of public behavior settings in a
town is a measure of the size of the town's public environment.

"We must emphasize that a behavior setting coerces people and

things to conform to its temporal-spatial pattern. This is not
an incidental or accidental characteristic. The person or per-
sons who maintain and control the setting (the performers)
make a deliberate effort to insure that this is so, and that the
setting therefore fulfills its function. This aspect of a setting we
call its program. Two settings are said to have the same pro-
gram when their parts and processes are interchangeable.
When this is true, two or more settings belong to the same
genotype. Two grocery stores, for example, could exchange
stock, personnel, bookkeeping systems, shelving, and so forth,
with little interruption in their operation. They belong to the
same genotype. A Methodist and a Presbyterian minister
could, and sometimes do, exchange pulpits. The number of
behavior setting genotypes in a town is a measure of the vari-
ety of the town's environment."

Barker identified 198 genotype settings in his town of 830 peo-
ple. Examples include grocery stores, hardware stores, ice cream
socials, kindergarten classes, business meetings, religion classes,
hallways, bus stops, and many others.
When individual grocery stores, churches, and the like were
recognized as separate or specific behavior settings, Barker found
884 public behavior settings in his town during the 366-day period
September 1, 1963, through August 31, 1964. The durations of
individual behavior settings ranged from 1 hour to 8,784 hours in
that year, and totalled 286,909 hours. Multiplying the hours of
duration of each behavior setting by the average number of per-
sons participating in it at any given time, Barker obtained a
record of "hours of occupancy" of behavior settings, totalling
1,129,295 in 1963-64. As there were 8,784 hours in the 366-day
year (including February 29, 1964), the total hours of "life lived"
during the year by the town's 830 residents was 7,290,720. About
15.5 percent of these hours were spent in public behavior settings;
the remaining "hours of living" were presumably spent in private
Barker24 indicated that if his criteria for identifying public
behavior settings were extended into private households, they
would yield five behavior setting genotypes: home meals, home
indoors, home outdoors, home bathrooms, and home festive
occasions. Thus, he notes that his town had, as of 1951-52, some

Karl A. Fox

289 households and 1,445 specific household behavior settings.
However, Barker concentrated all of his research efforts on the
public behavior settings, those into which almost any resident of
the town could enter.
I met Roger Barker at a small interdisciplinary symposium in
October 1966, and the first aspect of my social accounting model
was inspired by that meeting. As Barker presented his data,25 it
occurred to me that an individual could be viewed as allocating
his time (24 hours a day) among behavior settings which col-
lectively define his life style; simultaneously, he allocates his
money (personal consumption expenditures) exhaustively among
these settings.
The second aspect of my model was based on a 1968 article by
Parsons.26 I may have read more into his article than he himself
intended. However, Parsons' concept of "generalized media of
social interchange" and his characterization of outputs from the
social system to individuals as "rewards" and of inputs from indi-
viduals to the social system as "contributions" play a major role
in my present approach.
Money, said Parsons, is only one of a number of media of
social interchange. Money circulates in the economy. In the polity
or political system the corresponding medium is (political) power,
and in the society the corresponding medium is influence. In more
specialized subsystems of the society, reputation (as among doc-
tors, lawyers, and scholars), faith (as among members of specific
religious denominations), and ideology (as among members of
doctrinaire left wing and right wing political organizations) also
serve as media of social interchange. Parsons listed several other
media including technological know-how and skill, affect (recog-
nition and response), and others.
So, a second feature of my present model is this:
At the beginning of any year, the individual has an endowment
of resources or capacities which he can use as inputs or con-
tributions to the social system. These capacities may be clustered
into five groups. The first three groups are inside the individual:
health (physical, emotional, and mental); skills (work-related and
other); and value commitments and character. Capacities in the
fourth group are socially validated in the sense that they reflect
the individual's evaluation by others: power in formal
organizations (backed by written or unwritten job descriptions),

prestige, and political power. The fifth group includes rights to
income from property and transfer payments, which again must
be socially validated.
Combining the two aspects leads me to the following model27
in which the individual seeks to maximize his utility by choosing
an optimal life style subject to the constraints imposed by his

Maximize U = f (x), subject to
Ax < b,
x 2 0.

In these equations, x is a vector of the proportions of the indi-
vidual's time that he spends in each of the n behavior settings he
occupies; b is a vector of his m capacities or resources; and A is
an m times n matrix whose typical element, aij, is the amount of
the ith resource used per unit of time in the jth behavior setting.
In general, the individual knows about how much of each
resource.he must put into a behavior setting per hour of
occupancy and about how much of each kind of reward he will
get out of it. So, he spends all of the resources available to him in
the course of allocating his time exhaustively among the n behav-
ior settings. When the individual maximizes U=f(x) by choosing
an optimal life style, x, a shadow price or marginal utility is asso-
ciated with each resource in his endowment vector, b. Since one
of these resources is money income (derived from his work-related
skills and/or his property and transfer payments), the individual
should be able to compare the marginal utility of an additional
unit of each other resource to the marginal utility of an additional
dollar of money income. In other words, the marginal utilities of
his other resources can in principle be translated into dollar
equivalents. When these dollar equivalents are multiplied by the
corresponding resource quantities, we obtain estimates of the con-
tributions of each resource to the individual's "total income,"
which, like the specific resource contributions, is expressed in dol-
Table 1 is taken from a survey of the use of time by adults in
the United States during 1966.28 In table 1, the time available to
each category of individuals (for example, "men, employed, mar-
ried" in column 1) is allocated among 27 primary activities; these

Table 1-U.S. time use survey, 1966: Time spent by respondents in primary activities, by respondents' sex, employment
status, and marital status

Men Men Women Women
Primary employed unemployed employed unemployed
activities Married Single Married Single Married Single Married Single
(1) (2) (3) (4) (5) (6) (7) (8)
All figures in hours per 24-hour day
N= 449 72 16 6 190 152 342 17
Weighted N= 945 127 41 17 398 243 724 41

1 Regular work
2 Second job
3 Nonwork breaks
4 Trips to and from work
5 Preparing food
6 Cleaning house
7 Laundry, mending
8 Other house upkeep
9 Gardening, pets
10 Sleep
11 Personal care
12 Eating
13 Resting
14 Child care
15 Shopping
16 Nonwork trips

6.1 6.2 0.2 1.6 4.7 4.6 0.1 0.7
.2 .1 .1
.6 .6 .4 .5 .1
.7 .6 .1 .5 .5
.1 .2 .5 .3 1.0 .5 1.6 1.1
.2 .1 .8 .1 1.2 1.0 2.0 1.6
.1 .6 .3 1.0 .7
.3 .1 .1 .2 .3 .1 .4 .3



7.6 7.5 8.5 8.4 7.7 7.5
.9 1.2 .8 1.0 1.2 1.5
1.2 1.0 1.6 1.7 1.0 1.0
.3 .2 .4 .2 .3 .5
.1 .4 .4 .2
.4 .3 .6 .4 .5 .6
.8 .8 1.5 1.4 .7 .8

17 Education
18 Organizations
19 Radio
20 Television
21 Reading
22 Social life
23 Conversation
24 Walking
25 Sports
26 Various leisure
27 Spectacles

.1 .5 1.5 1.9
.2 .2 .5

1.7 1.4
.7 .6
1.0 1.2
.2 .2

.2 .1
.2 .2
.1 .3

.3 .1
.1 .3 .4

2.6 1.9 1.0 1.2
.7 1.0 .5 .3
1.7 1.7 1.0 1.2
.4 .2 .3 .3

.1 .1 .1
.8 .2 .3 .3 .5
1.5 .1 .3 .1

28 Work-related 7.6 7.5 .3 1.7 5.6 5.7 .1 .8
29 Housework .6 .5 1.6 .6 3.0 1.9 5.1 3.8
30 Personal care 10.0 9.9 11.3 11.2 10.1 10.4 10.6 10.9
31 Family tasks 1.3 1.2 2.1 2.1 1.6 1.6 2.7 2.3
32 Education and organizations .4 .7 2.0 1.9 .2 .5 .5 1.3
33 Mass media 2.5 2.2 3.4 2.9 1.6 1.6 2.2 2.9
34 Leisure 1.7 2.0 3.2 3.6 1.8 2.2 2.8 2.0

35 Grand total 24.0 24.0 24.0 24.0 24.0 24.0 24.0 24.0
36 Free time 4.8 5.1 9.0 8.6 3.9 4.8 5.9 6.4

Source: John Robinson and Philip E. Converse, Sixty-six Basic Tables of Time Budget Data for the United States (Ann
Arbor, Mich.: Survey Research Center, Univ. Mich., May 30, 1966). 11 pages plus 66 tables (xeroxed). Adapted
from table 1.

Karl A. Fox

figures total 24 hours a day, and are averages for a 7-day week.
For example, the average time spent by married men, employed,
in regular work was 6.1 hours a day, or 42.7 hours per week.
Rows 28-34 of table 1 cluster these 27 primary activities into 7
groups: work-related, housework, personal care (including sleep-
ing, eating, resting, and personal care in a narrow sense), family
tasks, education and organizations, mass media, and leisure. The
time allocations vary considerably between categories of adults
differentiated by employment status, marital status, and sex.
The 1966 survey of time use contains three other tables, one on
secondary activities (such as listening to the radio while primarily
engaged in housework); one on the categories of places where
time was spent; and one on the categories of persons with whom
time was spent. To characterize any real human allocation of
time, we must employ the four principles of classification simulta-
neously. Symbolically, T=f(Pa,Sa,Pl,Pe), where T is time, Pa is
primary activity, Sa is secondary activity (if any), P1 is "place
where," and Pe is "persons with whom." Unfortunately, the 1966
survey applies each principle of classification separately and inde-
pendently of the other three. Symbolically, its four tables give us
T=f(Pa), T=f(Sa), T=f(Pl), and T=f(Pe); we cannot synthesize the
true time allocation function T=f(Pa,Sa,PI,Pe) from this infor-
mation with any degree of assurance.
Nevertheless, I have tried in table 2 to show what such a syn-
thesis would look like and what it would imply. Table 2 describes
the time allocation of a "man, married, employed." Each column
of table 2 is, or belongs to, one of the 27 primary activities of
table 1. However, certain primary activities have here been
divided into two or more different behavior settings, giving a total
of 46 behavior settings (hence 46 columns) in all.
For example, the first column indicates that the individual
spent 2.7 hours in Behavior Setting 1, in which the primary activ-
ity is regular work, the place occurring is "place of work" and
persons involved are "all alone;" the individual is alone in his
office and is not engaged in any secondary activity. The second
column indicates that the individual spent another 2.6 hours in
Behavior Setting 2, in which the primary activity is regular work,
occurring at the place of work, but involving interactions with
colleagues. Each of the remaining 44 columns can be interpreted
in a similar manner; the number of hours adds up to 24 per aver-

age day.
In table 3, I have allocated the time of a "man, married,
employed" among nine behavior setting aggregates. The central
portion of table 3 is a condensation of table 2. The "value per
hour (ri)" row represents a "total price per hour," and the four
bottom rows represent the proportionate allocations of time and
three other categories of resources among the various behavior
setting aggregates. The figures in the five rows just mentioned are
illustrative only.
Figure 1 is an example of the kind of data sheet Roger Barker
and his associates filled out for each of the 884 behavior settings
of Midwest in 1963-64. The occupancy time of town residents in
each behavior setting is disaggregated among seven age goups, the
two sexes, and four socioeconomic classes. The "maximum pene-
tration" of each subgroup into the behavior setting is recorded.
(Zone 5 is occupied by the officials who jointly control the set-
ting; if there were a single leader or official in control of the set-
ting, he would be the sole occupant of Zone 6. Zones 4, 3, 2,
and 1 are successively less central and less important to the oper-
ation of the setting than are Zones 6 and 5.) Each behavior set-
ting is rated on 11 action patterns and five behavior mechanisms,
plus certain other characteristics.
I will not try to describe Barker's system in more detail. The
best single description is Barker's 1968 book;29 however, to get the
full picture, the reader is also referred to his other works.30 To the
best of my knowledge, Barker has made the most complete and
detailed observations of the activities of all members of a small
but relatively self-contained social system that have been made in
the United States.
Barker compared the 198 behavior setting genotypes available
in his town with the entries in the classified telephone directory of
a midwestern city of 800,000 people. He concluded that, if manu-
facturing and wholesaling enterprises were disregarded, about 80
percent of the behavior setting genotypes available in the metrop-
olis were available in his town. About 85 percent of the consumer
goods used by residents of his town of 830 people could be pur-
chased from its various retail enterprises. Thus, if one could
develop a comprehensive system of social accounts for Barker's
community, as much as 80 percent of the task of conceptualizing
and designing a national system of social accounts might be

Table 2-Illustrative rearrangement of time budget data into behavior setting format ("men, married, employed")

Primary activity
Behavior setting
Number of hours
Value per hour (rl)
Setting structure (per
Primary activity
Secondary activity
Social life
Place occurring
At home
Place of work
Another's home
Streets, parks, etc.
Business places
Indoor leisure 2

1 1 1 1 2

1 2 3 4 5

3 4 4 4 5

6 7 8 8 9 10 11 11 12 12 12 12 13 14

6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

2.7 2.6 0.4 0.4 0.2 0.6 0.3 0.2 0.2 0.1 0.2

1 1 1 1 1

1 1 1 1 1

0.2 0.1 7.6 0.7 0.2 0.6 0.2 0.2 0.2 0.3 0.1

1 1 1 -

1 1 1 1

1 1 1

1 1 1 1 1

1 1 1 1 1

1 1 1 1

1 1

1 1 1

1 1 1

Restaurants, bars
Other places
Persons involved
All alone
Nuclear family
Friends, relatives
Neighbors and
Primary activity

Behavior setting
Number of hours
Value per hour (rl)
Setting structure (per
Primary activity
Secondary activity

1 1 1

1 1 1
15 16 16 17 18

25 26 27 28 29

0.4 0.6 0.2 0.1 0.2

1 1 1 1 1

19 20 20 20 21

30 31 32 33 34

0.1 0.7 0.9 0.1 0.2

1 1 1 1 1

1 1 1

1 1

21 22 22 22 22

35 36 37 38 39

0.5 0.5 0.2 0.2 0.1

1 1 1 1 1

1 1 1

23 24 25 25 26

40 41 42 43 44

0.2 0.1 0.1 0.1

1 1 1 1

26 27 Total
per day
45 46

0.1 0.1 24.0

1 1 24.0


Social life
Place occurring
At home
Place of work
Another's home
Streets, parks, etc.
Business places
Indoor leisure 2
Restaurants, bars
Other places
Persons involved
All alone
Nuclear family
Friends, relatives
Neighbors and

1 1 1 1 1

1 1

1 1

1 1

1 1

I 1 1 1

1 1 1

I I I 1

Source: Arbitrary rearrangement (by Karl A. Fox) of time use data for "men, married, employed" in column 1 of tables 1, 17, 33, and 45 from John P.
Robinson and Philip E. Converse, Sixty-six Basic Tables of Time Budget Data for the United States. Ann Arbor: Survey Research Center,
Univ. Mich. May 30, 1966. 11 pages plus 66 tables (xeroxed).
'There are 27 primary activities. Thus, Primary Activity 1 is Regular Work. Primary Activity 2 is Second Job, and so on.
Includes 0.1 hour of outdoor leisure (behavior setting number 46).

1 0.3
1 0.4

1 1 4.0

Table 4, based partly on Barker's data, illustrates the time allo-
cations of residents of a small town among behavior setting clus-
ters, by population categories. Approximating the distribution of
the U.S. population in 1963, I have assumed that employed adults
constitute 40 percent of the town's population; adults not
employed, 26 percent; adolescents and children in school, 24 per-
cent; and preschool children, 10 percent. The time allocations
among public (nonfamily) settings are based on Barker's data;
time allocations among family settings are based on the 1966 time
use survey (for adults) and upon personal judgment for adoles-
cents, school children, and preschool children.
Table 5 illustrates, for the same hypothetical small town of
1,000 residents, the distribution of the outputs of behavior setting
aggregates among the various population categories. The figures
in table 5 assume that the money income per employed adult for
current personal services is $8,000 a year and that 400 adults are
employed. Hence, the money income of the town's residents for
personal services would be $3,200,000; the nonmoney rewards or
outputs of the town's social system would be valued at
$14,480,000 minus $3,200,000, or $11,280,000. If I added another
$2,000 per worker for income from property and transfer pay-
ments (the national average share of such income in personal
income from all sources is about 20 percent), the community
would have a money income of $4,000,000 and a total income of
$15,280,000. The nonmoney income would be about 2.8 times as
large as the money income.
I hold no brief for these particular figures; considering that
only about 10 percent of total living time, or 15 percent of the
total waking time, of the town's population is directed con-
spicuously toward the production of GNP, the ratio of nonmoney
to money income does not seem outlandish. However, I must
stress again that these figures are illustrative only.

I am working intensively now with Barker's data, trying to
integrate it with data from the U.S. census on occupations, indus-
tries, education, and earnings, and data on Americans' use of time
in 1975-76 from the SRC (Survey Research Center, University of



Table 3-Man married, employed: Illustrative rearrangement of time budget data from table 2 into larger aggregates of
behavior settings and categories of activities, places, and persons

Behavior setting aggregates
1 2 3 4 5 6 7 8 9 Total
House- Educa- hours
Item hold Sleeping tion, spent
Work- pro- Family and Personal Mass organi- all
related duction tasks resting care Eating media Leisure zations settings
(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

Primary activity (hours)
Value per hour (rs)
Secondary activities (hours)
Place occurring
Place of work
At home or just outside
Another's home

7.6 0.6 1.3 7.8 0.9 1.2 2.5 1.7
1.0 1.0 1.5 0.2 1.5 3.0 0.7 2.0
1.1 2.5 1.7

0.6 0.5 7.8 0.9 1.0 2.5 0.6

Other places
Persons involved
All alone
Nuclear family
Other household adults
Other friends, relatives
Neighbors & children
Organization members
Proportion of total resource
Group 1
Group 2
Groups 3 and 4

0.6 0.4 3.1

7.8 0.9

0.6 0.5

.32 .03 .05 .32 .04
.30 .05 .05 .05
.50 .05 .10 .05
.30 .02 .08 .11 .07

1.0 1.2 0.6

0.2 0.2
0.2 1.0

.05 .10 .07 .02 1.00
.15 .05 .35 1.00
.05 .10 .10 .05 1.00
.20 .07 .10 .05 1.00

Source: Table 2.

Figure 1-Data sheet for high school boys' basketball game

Name: High School Boys' Basketball Game
Genotype # 1-3: 0 1 8 Genotype Commonality # 8: 9 Locus 16: 1

No. of
B S 4-6: 0 0 5 Authority System 13-14: 0 1 Occurr. 17-19: 0 0 8
Genotype Date 7: 3 Class of Authority Systems 15: 4 Survey # 20: 5

Occupancy Time of Town Subgroups Max. Penetration of Subgroups ACTION

No. P Hours Of Code


Town Child
Town Total





Inf. 21: 1
Presch. 22: 2
Y S 23: 2
OS 24: 4

Adol 25: 4
Adult 26: 5
Aged 27: 2
Grand Max 28: 5
Males 29: 5
Females 30: 4
I 31: 4

Aes: 53: 0
Bus 54: 1
Prof 55: 1
Educ 56: 1
Govt 57: 1
Nutr 58: 1

PersAp 60: 2

PhysH 62: 2
Rec 63: 8
Rel 64: 0


II 105 2236 45-46: 1 9
III 42 1014 47-48: 1 4
N-G 3 50 49-50: 0 5
Town Child 51-53: 0 4 3
Out Child 54-56: 1 8 7
Total Child 57-59: 2 3 0
Town Total 60-62: 1 8 5
Out Total 63-65: 9 3 7
Grand Total 66-69: 1 1 2 2
70: blank
Grand O.T. (code) 71-73: 0 3 1
Total Duration 74-77: 0 0 2 4

Average Attendance 78-80: 3 6 3

II 32: 5
III 33: 4
N-G 34: 4

Town Child 35-36: 0 1
Out Child 37-38: 0 0
Total Child 39-40: 0 1
Town Total 41-42: 5 3
Out Total 43-45: 2 4 9
Grand Total 46-48: 3 0 2
Perf/Pop 49-50: 2 7
Aver. No. 51-52: 8 4

Soc 65: 6
AffB 66: 9
GroMot 67: 7
Manip 68: 7
Talk 69: 9
Think 70: 4
GEN RICH: 71-72: 23
Children 73: 4
Adolesc 74: 2
Children 75: 0
Adolesc 76: 3

wtd 79: 7

Source: Reproduced from Roger G. Barker, Ecological Psychology. Stanford: Stanford Univ. Press, 1968, p. 99.

Table 4-Illustrative time allocations of residents of a small town among behavior setting clusters, by population category

Nonfamily settings Family settings tion
House- of
Busi- Govern- Social Traffic- work, Per- Mass Out-of- town
ness ment life, ways and family sonal media, town Total popu-
places Schools agencies Churches leisure hallways tasks care leisure settings time lation
Population category (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12)

A. Percent of total time of each population category

Town residents
Adults, employed 12.50 0.25
Adults, not employed 7.70 -
Adolescents and 0.80 10.90
children in school

Children, preschool
Total population

2.50 0.25 2.00 2.50 15.00 42.00 11.00 12.00 100.00
0.60 0.40 3.00 0.80 30.00 44.00 11.50 2.00 100.00
0.50 2.50 1.30 5.00 50.00 27.00 2.00 100.00

1.35 1.35 0.60 1.40 58.30 35.00 2.00 100.00 0.10
7.33 2.72 1.16 0.32 2.34 1.56 15.14 46.07 17.36 6.00 100.00 1.00

B. Percent of total time of entire population of town

Town residents
Adults, employed
Adults, not employed
Adolescents and
children in school
Children, preschool
Total population
Equivalent total occu-
pancy time of non-
family settings (by
residents and non-

5.00 0.10 1.00 0.10 0.80 1.00 6.00 16.80 4.40 4.80 40.00 -
2.00 0.16 0.10 0.80 0.20 7.80 11.44 2.98 0.52 26.00 -
0.20 2.62 0.12 0.60 0.30 1.20 12.00 6.48 0.48 24.00 -

0.13 0.14 0.06 0.14 5.83 3.50 0.20 10.00 -
7.33 2.72 1.16 0.32 2.34 1.56 15.14 46.07 17.36 6.00 100.00 -


3.06 1.06 0.34 2.27
5.78 2.22 0.66 4.61

Source: Columns 1 through 6 in "total population" rows and in "nonresident" row are based on data in table 5.3, in Roger G. Barker, Ecological
Psychology. Stanford: Stanford Univ. Press, 1968, pp. 110-116. All other figures are inferential estimates by Karl A. Fox, although figures
for adults are guided, in part, by data in Robinson and Converse (see footnote 28).

^ ^^


Table 5-Distribution of outputs of behavior setting aggregates in a small town to various population categories, based on
hypothetical data

Recipients of behavior setting outputs
Adults, Adults, not and children Preschool
Behavior setting aggregates employed employed in school children Total
in which outputs are produced (1) (2) (3) (4) population

Proportion of each resource of each population category No.
1. Firms
Time 0.225 0.077 0.008 0.013 0
Group 1 .225 .077 .008 .013 432
Group 2 .400 .150 .015 .013 804
Groups 3 and 4 .850 .150 .015 4,438
2. Schools
Time .003 .109 0

Group 1 .003 .109 110
Group 2 .015 .220 130
Groups 3 and 4 .015 .220 178
3. Social settings
Time .072 .048 .043 .020 0
Group 1 .072 .048 .043 .020 199
Group 2 .185 .250 .165 .020 637
Groups 3 and 4 .035 .250 .165 832
4. Family settings
Time .700 .875 .840 .967 0
Group 1 .700 .875 .840 .967 2,899
Group 2 .400 .600 .600 .967 1,649
Groups 3 and 4 .100 .600 .600 2,172
Value of behavior setting outputs
per capital of total population
(dollar equivalents: money Dollars
income per employed adult=
$8,000) 8,000 4,160 1,920 400 14,480


Karl A. Fox

Michigan) survey currently underway. (I participated in devel-
oping some of the questions for this survey, but as of June 22,
1976, the data were not yet ready for analysis).
A data system based on my approach need not intrude upon
individual privacy to a significant degree. The time series data
required could be based largely on sample surveys of behavior
settings rather than of individuals. A behavior setting is equiv-
alent to a small establishment; indeed, many of Barker's behavior
settings, such as grocery stores, barber shops, and service stations,
are identical with establishments as reported in the various eco-
nomic censuses. In measuring the characteristics of a behavior set-
ting, it is not necessary to identify the particular individuals who
are occupying the setting. And, in basing national accounts on
samples of establishments and other behavior settings, it would
not be necessary to reveal the identities of the individual estab-
lishments in the sample.
In addition, rather small sample surveys of the use of time by
individuals and households could be used to "true up" the basic
series derived from surveys of behavior settings.
My approach may be particularly useful in international com-
parisons of economic and social well-being. For example, I
grouped about 120 countries of the world as of 1970 into five
classes according to GNP per capital (in 1970 U.S. dollars). I then
regressed a number of demographic, economic, and social vari-
ables upon the GNP per capital categories and converted the
regression coefficients into a rough index of "total income per
capital made up of four subindexes, for the five groups of count-
ries. The subindex for life expectancy and health increases much
less rapidly than the index as a whole. In contrast, the subindex
for commitment to modernization (based on per capital use of
modem means of transport and communication) rose somewhat
more rapidly, and an index of GNP per capital much more rap-
idly, than the index as a whole. The ratio of GNP per capital in
the highest of the five GNP groups to that in the lowest was
about 24 to 1; the ratio of the index of total income in the highest
GNP group to that in the lowest (crudely calculated) was about 6
to 1.
Richard L. Meier has asserted that the real output of a social
system is substantive interactions among people. William Alonso
has reminded overzealous proponents of "new towns" that there is

"...there is a limit to the extent
to which bricks and mortar can
improve the human condition."

a limit to the extent to which bricks and mortar can improve the
human condition: the real city is people and organizations, and
the buildings are only their containers.
These observations, together with the "total income" concept,
suggest a new approach to the problem of design (of houses,
automobiles, cities, or whatever). What are the minimum amounts
of scarce materials and energy required to support a pre-specified
level of substantive and enjoyable interactions among people? Or,
what is the minimum amount of GNP per capital required to sup-
port the present U.S. level of "total income" per person?

References and Notes

SPresident's Research Committee on Social Trends, Recent Social Trends.
New York: McGraw-Hill Publ. Co., 1933.
2 Wesley C. Mitchell, Business Cycles: The Problem and its Setting. New
York: Nat. Bur. of Econ. Res., 1927. (See especially chap. 5, "Results and Plans.")

3 Arthur F. Burns and Wesley C. Mitchell, Measuring Business Cycles. New
York: Nat. Bur. of Econ. Res., 1946.
4 Tjalling C. Koopmans, "Measurement Without Theory," Review of Eco-
nomics and Statistics. 29, Aug. 1947, pp. 161-172. I do not mean to leave a dis-
torted impression of Wesley C. Mitchell. Mitchell was a very great economist who
in early and mid-career wrote classic works on a history of the greenbacks and
their contribution to inflation during the Civil War; on business cycles; and on the
making and using of index numbers. Mitchell was largely responsible for the
founding of the National Bureau of Economic Research, and served as its director
from 1920 until 1945. In addition to his own work on economic indicators and
business cycles, he had the breadth of vision to support the development of
national income estimates by W. 1. King in the 1920's and by Simon Kuznets in
the 1930's. Mitchell was correct in recognizing that large masses of data would
have to be carefully and systematically organized to provide an empirical basis for
business cycle research. The data Mitchell developed as "economic indicators" also
proved essential to Kuznets' estimates of national income and to Tinbergen's eco-
nometric model; otherwise, Keynes' theory might have gone unsupported by mea-
surement for a good many years.
5 John Maynard Keynes, The General Theory of Employment, Interest, and
Money. New York: Harcourt Brace Jovanovich, 1936.
6 Simon Kuznets, National Income and Capital Formation, 1919-1935. New
York: Nat. Bur. of Econ. Res., 1937.
7 John W. Kendrick, Economic Accounts and Their Uses. New York:
McGraw-Hill Publ. Co., 1972. (See especially chap. 2, "The Development of
National Income Estimation and Economic Accounting.")

Karl A. Fox

8 Wassily Leontief, "Quantitative Input and Output Relations in the Eco-
nomic System of the United States," Review of Economics and Statistics. 18, Aug.
1936, pp. 105-125. Also: The Structure of American Economy, 1919-1939. (2nd
edition) New York: Oxford Univ. Press, 1951.
9 Jan Tinbergen, Statistical Testing of Business Cycle Theories. Vol. 1: A
Method and its Application to Investment Activity. Vol. II: Business Cycles in the
United States of America, 1919-1932. Geneva: League of Nations Econ. Intel-
ligence Ser., 1939.
10 Raymond A. Bauer (ed.), Social Indicators. Cambridge: M.I.T. Press,
Bertram M. Gross, "The State of the Nation: Social Systems Accounting."
In ibid.
12 Bauer, op. cit.
3 In Bauer, op. cit.
14 President's Commission on National Goals, Goals for Americans.
Englewood Cliffs, N. J.: Prentice Hall, 1960.
15 Karl A. Fox (ed.), Economic Analysis for Educational Planning: Resource
Allocation in Nonmarket Systems. Baltimore: Johns Hopkins Univ. Press, 1972.
(Includes contributions from Fox, Jati K. Sengupta, T. Krishna Kumar, and Bikas
C. Sanyal.) Also: "Practical Optimization Models for University Departments." In
Hector Correa (ed.), Analytical Models in Educational Planning and Adminis-
tration. New York: David McKay, 1975.
16 Karl A. Fox, Social Indicators and Social Theory: Elements of an Oper-
ational System. New York: John Wiley and Sons, 1974. Also: "Combining Eco-
nomic and Noneconomic Objectives in Development Planning: Problems of Con-
cept and Measurement." In Willy Sellekaerts (ed.), Economic Development and
Planning: Essays in Honour of Jan Tinbergen. London: Macmillan, 1974.
1" Kenneth C. Land and Seymour Spilerman (eds.), Social Indicator Models.
New York: Russell Sage Foun., 1975. (See especially chap. 2 by Land, "Social
Indicator Models: An Overview.")
18 Jan Tinbergen, On the Theory of Economic Policy. Amsterdam: North-
Holland Publ. Co., 1952.
19 Land and Spilerman, op. cit.
20 Arthur S. Goldberger and Otis Dudley Duncan (eds.), Structural Equation
Models in the Social Sciences. New York: Seminar Press, 1973.
21 Gary S. Becker, "A Theory of the Allocation of Time," Econ. Jour. 75, pp.
493-517, 1965.
22 Roger G. Barker and Herbert F. Wright, Midwest and its Children: The
Psychological Ecology of an American Town. New York: Harper and Row, 1955.
(Reprinted by Archon Books, Hamden, Conn., 1971.) Also: Barker, Louise S.
Barker, and Dan D. M. Ragle, "The Churches of Midwest, Kansas and Yoredale,
Yorkshire: Their Contributions to the Environments of the Towns." In W. J. Gore
and L. C. Hodapp (eds.), Change in the Small Community: An Interdisciplinary
Survey. New York: Friendship Press, 1967. Also: Barker, Ecological Psychology:
Concepts and Methods for Studying the Environment of Human Behavior. Stan-
ford: Stanford Univ. Press, 1968. Also: Barker and Phil Schoggen, Qualities of
Community Life: Methods of Measuring Environment and Behavior Applied to
an American and an English Town. San Francisco: Jossey-Bass, Inc., 1973.

23 Ibid., "The Churches of Midwest..." pp. 158-159.
24 Barker, op. cit., "Midwest and Its Children..."
25 The data were later published in Barker, op. cit., "The Churches of Mid-
26 Talcott Parsons, "Systems Analysis: Social Systems," International Ency-
clopedia of the Social Sciences. 15, pp. 458-473. New York: Macmillan and Free
Press, 1968.
27 1 arrived at a preliminary version of this model in 1968 and published it as:
Jati K. Sengupta and Fox, "Operations Research and Complex Social Systems,"
Economic Analysis and Operations Research: Optimization Techniques in Quan-
titative Economic Models. Amsterdam: North-Holland Publ. Co., 1969. In 1971-
72, Paul Van Moeseke and I produced a more rigorous mathematical formulation
of the model published as: Fox and Van Moeseke, "Derivation and Implications
of a Scalar Measure of Social Income." In H. C. Bos, H. Linnemann, and P. de
Wolff (eds.), Economic Structure and Development: Essays in Honor of Jan Tin-
bergen. Amsterdam: North-Holland Publ. Co. and New York: American Elsevier,
1973. It was also published with minuscule changes as: Fox, op. cit., Social Indi-
cators and Social Theory...(chap. 3).
28 John P. Robinson and Philip E. Converse, 66 Basic Tables on Time-Bud-
get Data for the United States. Ann Arbor: Inst. for Social Res., Univ. Mich.,
29 Barker, op. cit., Ecological Psychology...
30 Barker, op. cit.

Appraising the

Economic Performance

Of the

Food Industry

George E. Brandow

Lecture 5
September 8, 1976

After the turmoil about food prices during the past 4 years, lit-
tle documentation seems necessary to establish the economic
importance to the American people of the vast industry that pro-
cesses and distributes food. Collectively, questions asked by con-
sumers, farmers, political leaders, and others amount to a demand
for comprehensive information about the economic performance
of the food industry. Such questions state or imply concerns
about efficiency of the industry, the level of its profits, the pre-
sentation of products to consumers, and other matters I shall
describe in more detail later. Loosely speaking, performance
means how well an industry does the things that society might
reasonably expect it to do.
Economists asked to appraise the economic performance of an
industry have a difficult task. If they confine themselves to the
elegant abstractions of rigorous general theory, they find few han-
dles by which to grasp the inelegant real world and are wholly
unprepared for some of the institutional and dynamic character-
istics of the industry. If they adopt the approach of industrial
organization economics, they find standards imprecise, mea-
surement both conceptually and empirically difficult, and judg-

George E. Brandow

ment usually necessary to reach conclusions. Yet to insist on
elegantly derived results that are fully conclusive and leave no
room for judgment is to require the impossible. The facts are that
good performance is a set of sometimes conflicting goals; the eco-
nomic world is complex, changing, and never fully knowable; and
the economic results of interest range along continuous scales
from good to bad rather than being clearly one or the other.
Economists must accept this if they are to produce valid infor-
mation, and the public must accept it if effective use is to be
made of such information for policy purposes.
Appraisal of performance of the food industry necessarily
draws heavily upon ideas already a part of industrial organization
economics. Yet conventional versions of this area of economics
are somewhat deficient for the purpose in at least two respects.
First, heavy emphasis on market structure, especially on concen-
tration, tends to obscure important relationships within the verti-
cal economic sectors that perform the successive functions of pro-
ducing raw materials, processing them, and distributing raw and
finished products. Second, industrial organization economics ordi-
narily considers firms as the only economic units whose decisions
bear directly or indirectly upon performance. Yet it is obvious
that labor unions and government make important decisions
influencing performance. Industry, for the purpose of appraising
performance, should include all decisionmakers and performers.
This paper is limited to appraisal of performance, but under-
standing the determinants of performance and analyzing possible
ways of improving performance are also necessary for effective
policy formulation. Certainly, identification of an instance of par-
ticularly good or bad performance does not also explain it, and
understanding why a particular result occurs does not auto-
matically reveal all possible corrective measures or their full con-
sequences. Structure, conduct, and performance have become so
fused and holy a trinity in industrial organization economics that
I feel the need to point out that I am concentrating on one with-
out denying the importance of the others.

General Comments on Performance

Several broad assumptions lie behind the criteria for economic

performance to be discussed shortly. Some require explicit com-
ment to explain the choice of criteria and the difficulties met in
appraising performance.
Performance is defined from the standpoint of society, not of
participants in the industry or of any single group outside it. The
role of consumers is crucial, however. My test of what should be
produced by the food industry is the selection of products that
would be made by fully informed consumers given a wide range
of choices at prices just sufficient to cover necessary costs of mak-
ing the products available. Kind of product includes packaging,
food additive content, etc., as well as type of food. The mix of
products satisfying this test is not necessarily what the food indus-
try would like to sell, farmers would like to produce, nutritionists
would recommend, or consumer spokesmen think would be good
for consumers. Neither does the test give consumers all they want,
for prices must be sufficient to compensate all those who directly
or indirectly participate in supplying food efficiently. In a time of
general inflation or short domestic supplies of raw foods, most

As Professor of Agricultural Eco-
nomics at Pennsylvania State Univer-
sity, Professor Brandow has served
on many USDA committees, includ-
ing the Agricultural Committee of the
National Planning Association. Note-
worthy achievements include a posi-
tion in the Office of Price Adminis-
tration during World War II;
membership on the Joint Economic
Committee of the Congress; Execu-
tive Director of the Food Commis-
sion; President and Fellow of the
American Agricultural Economics
Association; and recipient of a 1975
AAEA award.

George E. Brandow

George E. Brandow

consumers would be unhappy even though the test were satisfied.
The test places much reliance on market behavior of con-
sumers but does not accept the actual market as a fully satis-
factory setting for the test. The main reasons are that consumers
are not fully informed buyers, strong efforts are made by sellers
to influence consumers' decisions, and prices are not always con-
sistent with necessary costs (including raw food prices). It seems
possible to set up better market tests experimentally, but only at
much expense. A more practicable procedure probably is to infer
at least the directions in which consumers' choices would differ
from actual patterns by methods suggested in a following section.
Despite its shortcomings for the test of ideal product mix, how-
ever, the market is an invaluable first approximation, and evi-
dence should be required before modifications of its results are
accepted as standards.
As already suggested, balanced appraisal of performance of the
food industry requires examination of the activities of all who
participate in the industry. This includes business firms, labor
unions, farmer and consumer cooperatives, and the government in
its various regulatory roles. Few points so forcibly call to atten-
tion the political context in which questions about performance
exist. Typically, each economic group wants the activities of other
groups to be scrutinized but expects its own virtue to be taken for
granted. Though studies of some aspects of performance arouse
little resistance, others meet intense political opposition. The diffi-
culties involved in getting sustained support through the political
process for thorough appraisal of the performance of the food
industry (or of any other industry) seem at least as formidable as
the difficulties of making the economic evaluations.
Appraisal of industry performance requires a distinction
between economy-wide goals and policies on the one hand and
criteria appropriate for individual industries on the other. For
example, even though high-level employment is a leading national
objective we should not require an industry experiencing declining
demand for its products, dwindling supplies of vital raw materi-
als, or rapid advance in labor-saving technology to maintain or
increase its labor force. Though overall price stability is an
avowed national objective, we cannot expect a particular industry
to maintain steady prices if economy-wide forces inflate costs and
money demand, nor should we require price stability of an indus-

"An individual industry
approach to (our economic
problems) would cause chaos
and be self-defeating."

try facing special demands or unique changes in costs. If the low
purchasing power of poor people for food is regarded as a prob-
lem, as it is, the appropriate remedy is somehow to increase their
purchasing power, not to ask the food industry to offer its prod-
ucts to the poor at special prices. In short, many of our economic
problems must be attacked on an economy-wide level. An individ-
ual industry approach to them would cause chaos and be self-

The Elements of Performance

The central task is to identify the principal criteria for per-
formance applicable to the food industry, to indicate how empir-
ical evidence might be obtained concerning each, and to suggest
standards that might be applied in appraising individual elements
of performance. A full-length book would be required to do these
things in detail for the highly complex, diversified food industry.'
No final word on standards seems possible. Here, it is necessary
to use generalizations, to indicate directions rather than specific
routes, and to forego debate with those who will not agree with
what is said.

Product Presentation to Buyers
The first criterion concerns the variety, quality, healthfulness,
and packaging of foods offered to buyers, together with infor-
mation about the products. Problems occur principally at the
retail level of sale, where buyers are consumers. Difficulties may
exist at wholesale and farm levels of sale, but the much greater
knowledge of commercial buyers, the frequent availability of
grades, and the feasibility of using technical specifications ease
potential problems there.
Wide variety is clearly necessary to meet consumer demand,
and is essential for providing a test of consumer preferences at
necessary prices. Variety can be overdone; if false impressions of
product differences lead to specious variety, the added inventory
and related expenses are not worth their cost. Food safety and
wholesomeness clearly require some public regulation and mon-
itoring of the food supply, but the appropriate boundaries of reg-
ulation and the tolerances to accept remain gray areas about

George E. Brandow

which I have little to say.
Information needed for effective consumer choice is of several
kinds. Again, the topic is familiar. The list includes such matters
as nutritive values, objective indications of quality, package and
unit prices, and open-dating. A crucial need is to substitute accu-
rate knowledge of intrinsic value for impressions created by pro-
motion and merchandising techniques. The high susceptibility of
consumers to such techniques is a basic reason for excessive sell-
ing costs and for inefficient distribution methods sometimes used
for highly promoted products.
Appraisal of this broad element of performance can proceed in
several ways. Food safety and wholesomeness often can be judged
by testing samples of products on the market and by examining
the production methods and inspection procedures to which foods
are subjected. Long-run health risks possibly associated with cer-
tain foods and food additives must be technically evaluated,
which can be a slow and uncertain process in borderline cases.
The adequacy of variety can be judged by observing whether mar-
kets offer apparently feasible choices to consumers. In particular,
when one product is displacing another, do consumers have alter-
natives that provide a valid test of consumer choice?
It seems possible to make objective tests of quality and taste
characteristics of many foods, to ascertain price differentials at
which various qualities and brands are offered, to learn the state
of consumers' information about the choices before them and to
put the findings together to identify instances in which consumers'
purchasing behavior differs from what it would be with full infor-
mation. Sufficiently comprehensive and integrated research of this
kind is virtually never done. Such research seems necessary for
identifying instances in which market choices are imperfect, for
producing data needed for consumer education, and for judging
the potential usefulness of consumer grades.

Efficiency in Processing and Distribution
The pragmatic criterion of efficiency is whether necessary func-
tions are performed at least cost in light of input prices and tech-
nical knowledge about methods and products. Efficiency ordi-
narily requires the use of up-to-date technology, achievement of
economies of size, productive use of labor and other resources,
effective use of transportation facilities, and so on. Efficiency can

be studied at the level of individual plants, of groups of plants
within firms, and of processes involving several firms. In the
broadest sense, systems analyses are required to identify the least-
cost ways of moving foods from the raw material stage to con-
sumers' shopping bags.
Estimates of least-cost levels can often be made from infor-
mation about the operating experience of firms performing func-
tions in different ways, on different scales, or with different man-
agement techniques. Or, more simply and therefore more
practicably, the effect on costs achievable by introducing an avail-
able but little used economy may be made. Establishing empirical
functional relationships such as between unit costs and levels of
inputs can be useful in judging achievable efficiencies. Obser-
vation of methods in other industries performing similar functions
or of the food industry abroad may be instructive. Economic and
engineering data may permit good estimates of costs under cir-
cumstances not found in actual practice. Analyses of costs of
alternative distribution channels, of transportation methods, and
of vertical coordination are especially pertinent to appraisal of
overall efficiency of the complete vertical system.
Judgments as to achievable minimum costs must be realistic.
All firms cannot be expected to be completely up to date when
new technologies become available at a rapid rate or when mar-
kets or sources of raw materials shift quickly. Good or bad luck
is often a reason why particular plants or firms have low or high
costs at a particular time.
An important consideration in the context of appraising per-
formance is the incidence of inefficiencies found in the system. If
high-cost firms are losing money, then not all (perhaps none) of
excessive costs are being transmitted to others, and a process is
underway to correct inefficiency. But if profits are good despite
clear inefficiencies, performance must be rated poor and reasons
for the situation investigated.
Work rules, terms of compensation, and other conditions insis-
ted upon by labor unions apparently cause numerous instances of
inefficiency in the food industry. But since equity is wanted as
well as efficiency, specific standards of performance are often
hard to state. The minimum work week in the meat packing
industry is an example of a cost-increasing provision that seems
warranted on equity grounds. Labor contract terms designed to

George E. Brandow

protect indefinitely the jobs of delivery truck drivers in some fluid
milk and bread markets, at the cost of perpetuating inefficient
distribution methods, seem unwarranted.
Appraisal of government regulations encounters similar diffi-
culties. Meat inspection increases costs but is thoroughly war-
ranted by its benefits. Some interpretations of the Robinson-Pat-
man Act that reduce efficiency in transportation and product
handling seem devoid of offsetting benefits. In both the labor and
regulatory fields, information on costs and benefits can have
important educational value because the impression is so com-
monly created that there are no costs or, alternatively, no bene-

Development of new methods and products is, of course, an
important means by which the food industry has improved its
performance. A large part of the technology embodied in new or
improved equipment is developed by manufacturers who sell to
the food industry and often represents rather simple adaptations
of science and technology broadly applicable to many industries.
One way of judging progressiveness, therefore, is to observe the
rate at which broad technology (e.g., automated warehousing) is
applied to the food industry. Processing methods unique to the
food industry and new products are more difficult to evaluate, for
knowing what might potentially be done is virtually impossible.
An absence of new methods or products is not a fault if the cur-
rent scientific opportunity for developing them is zero, and sub-
stantial but lagging adaptation of innovations being used else-
where is open to criticism. A flood of so-called new products is of
little value if most of them are trivial variations concocted as mer-
chandising devices.
Statistical series such as labor productivity data and farm-retail
price spreads often give, directly or indirectly, useful indications
of gains attributable to progressiveness and increased efficiency.
Judgments may be possible as to whether the industry is seeking
and is receptive to innovations. For example, are retailers and
packers experimenting with various forms of central meat cutting?
Expenditures on research and development are useful in this con-
nection but are not conclusive because scientific opportunity var-
ies from industry to industry and because much R&D in the food

industry appears to be for minor product differentiation of little
significance to society.

Selling Costs
The social usefulness of selling costs is highly controversial.
Some selling costs are warranted to inform buyers about product
availability, uses, and prices or to arrange buyer-seller trans-
actions. An additional consideration is that if enterprisers are to
be innovative and efficient, they need an opportunity to tell pro-
spective buyers about the results of their efforts. But when selling
costs average 15 to 20 percent of prices, which is above normal
but not unheard of in the food industry, no amount of ingenuity
can make a persuasive case for them.
It seems possible to devise rough estimates of how much adver-
tising is needed to make buyers well aware of sellers' offerings of
products. Possibly, one-half of 1 percent of sales would suffice for
food retailers and 3 percent of sales for food manufacturers of
genuinely differentiated products. Advertising costs greater than
these amounts and all giveaways (e.g., trading stamps, prizes)
would be considered excessive when appraising performance. Pro-
motions that are really price cuts would not be criticized on
grounds of selling costs. Crude as it is, such a procedure would be
better than considering all selling costs acceptable or considering
them all objectionable.
It seems likely that the effects of strategies of which intensive
advertising is a part are not limited to excessive promotion costs.
The National Commission on Food Marketing found that retail
prices of nationally advertised brands of common foods averaged
20 percent higher than retail prices of private label products of
comparable quality. The difference cannot be fully explained by
advertising costs and seller profits. Tracking down the difference
would be revealing about other marketing costs that often accom-
pany intensive advertising.

Returns to Factors of Production
Corporation profits are by far the most commonly discussed
aspect of this element of performance. The high sensitivity of the
public to profits apparently is due mainly to conceptions about
what is equitable and to often distorted impressions of the
importance of profits to prices. Resource allocation, which econo-

University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs