Soviet Growth &

American Textbooks
David M. Levy
Center for Study of Public Choice
George Mason University
Fairfax VA 22030

Sandra J. Peart
Jepson School of Leadership Studies
University of Richmond
Richmond VA 23173

December 3, 2009

Earlier versions were presented at the 2006 History of Economics Society, the 2006 Public Choice Society,
NYU and the 2008 Summer Institute for the Preservation of the History of Economics at which we
benefitted from the discussion. We thank Richard Ware for the Nutter-Rostow correspondence, Alex
Tabarrok for suggestions and encouragement and the Earhart Foundation for financial support. Adam C.
Smith and Margaret Albert provided valuable research assistance. We are responsible for the errors and
omissions.

Abstract
JEL A20, P17, P27

We examine the treatment of Soviet growth in successive editions of American economics textbooks
published between 1960 and 1980. What we find repeatedly is over-confidence in the potential for Soviet
growth and an asymmetric response to past forecast errors. More than this, the textbooks report faster Soviet
income growth combined with a constant ratio of SovietUS income. Textbooks that abstracted from these
institutional details (thin) offered a wider range of application than those which focused on one society
(thick). A simple way to distinguish these two traditions is whether the book used a productivity possibility
frontier [PPF] for cross-societal comparisons. Thick accounts did not while thin ones did. It was in the
institutional dimension that the account by Tarshis differed from that of Samuelson.

Electronic copy available at: http://ssrn.com/abstract=1517983

Department of State

22 October 1963

Dear Warren,

Many thanks for your kind invitation of September 24th. This autumn is
too tied up for me to take the trip to Charlottesville. Could we fix it for the
spring? Your name and your work are very much on my mind these days.
Our loyalty to the notion that key Soviet sectors were subject to
deceleration, a loyalty we both pursue against the views of the intellectual
establishment appears increasingly to be vindicated. Should you come this
way I should very much like to chat with you.

All the best.

Walt Rostow.

Electronic copy available at: http://ssrn.com/abstract=1517983

 Introduction

The collapse of the Soviet Union came as a surprise to many western economists in part because its

economy had long been portrayed in textbooks as a viable alternative to democratic capitalism. Textbooks

had shown that the Soviet economy was growing faster than the US and Soviet citizens would soon enjoy a

higher standard of living than Americans. Of course the Soviet economy was not growing faster than the US

economy.

In what follows, we examine the treatment of Soviet growth in successive editions of American

economics textbooks published between 1960 and 1980. What we find repeatedly is over-confidence in the

potential for Soviet growth and an asymmetric response to past forecast errors. More than this, the

textbooks report faster Soviet income growth combined with a constant ratio of SovietUS income. This

trust in the future and skepticism about the past was the basis of a standard Soviet-era joke: Under

Communism, the Poles are fond of saying, only the future is certain; the past is always changing (Nutter

1969). This fact, that accounts of Soviet growth emerged and changed over time in successive editions of

American economics textbooks, has passed almost unnoticed. More than this, at some key junctures textbook

authors disagreed about how to characterize the Soviet economy and, indeed, whether US-style economic

analysis might appropriately be applied to the analysis of Soviet growth.

A number of textbook reviews late in the 1940s and early in the 1950s complicate the account below.

These reviews surely altered the textbook landscape in the early 1960s. Textbooks that persisted in the early

1960s survived earlier attacks from libertarian writers, Rose Wilder Lane (1947), V. Orval Watts (1950)

and William F. Buckley (1951). One great textbook (Paul Samuelsons) passed through the attendant

political controversy, while another (Lorie Tarshis) succumbed. Why one survived and the other did not has

been a puzzle; their authors ideology is by all accounts very similar (Elzinga (1992), Colander and Landreth

(1995 & 1998) Samuelson (1997)).

1
 So there are two puzzles for what follows. Why were important textbooks of the 1960s and 1970s so

over-confident about Soviet economic growth that evidence of model failure was repeatedly blamed on events

outside the models control? Second, how did the earlier libertarian attack on particular textbooks affect the

discussion of Soviet growth by changing the mix of surviving textbooks? The counter-factual question is this:

absent a libertarian censoring would a Tarshis-influenced textbook universe offer the same analysis as what we

observe and describe below? On this question, our conjecture is that competition among differing viewpoints

was prematurely suppressed by the attacks. As a result, too few viewpoints persisted in the textbook

landscape, where by too few we mean too few to penetrate the non-transparent institutional arrangements

and economic outcomes of the Soviet Union.

A critical aspect for the Soviet-American comparison is whether the textbook offers a model by

which the economist can compare one society with another. Textbooks written in what might be described as

a thick, empirical institutionalist tradition focused on a particular society and as such they offered no basis for

comparison in the analysis. By contrast, textbooks that abstracted from these institutional details offered a

wider range of application using a thinner model. A simple way to distinguish these two traditions is to check

whether the book used a productivity possibility frontier [PPF] for cross-societal comparisons. Thick

accounts did not while thin ones did.

The difference between thick and thin models in the textbooks was replicated in the specialist

controversy over Soviet-American growth. Nutter (1958, 1962) insisted that growth comparisons preserve

institutional information by asking whether the Soviet Union was catching up with America. His many critics

saw no need for institutional details to enter into the accounts.

It was in the institutional dimension that the account by Tarshis differed from that of Samuelson.

Like Samuelson, Tarshis wrote in a reformist tradition. He was, however, immersed in the details of the

American economy. Consequently, he offered little by way of cross country comparisons. Samuelsons text, on

2
the contrary, pioneered the use of the PPF as the means by which to compare different economies. His model,

which collapsed societies into a two-dimension production possibility set, bought elegance by abstraction. Of

course, if an omitted variable was important in only one of the societies in the comparison, th.e models

predictions would be incorrect.

We turn next to the textbooks after Sputniks October 1957 launch. The texts discussed in most

detail below were selected for attention in Kenneth Elzingas study of market leaders (Elzinga 1992). 1 The

first two constitute the most successful multi-edition post-World War II texts, Campbell McConnells

Economics: Principles, Problems and Policies first published in 1960 and Samuelsons Economics: An

Introductory Analysis first published in 1948. The other successful text noted by Elzinga is George Bachs

Economics. Elzinga also offers a careful discussion of Tarshis 1947 Elements of Economics. We use

Tarshis little-known 1967 Modern Economics to address the counter-factual question of how a

Tarshis-dominated textbook market might have considered Soviet growth. Three others are selected to

illustrate related issues. Robert Heilbroners influential book speaks to the question of ideology and

institutional thickness. Rendiz Fels first edition seems unique in recognizing the consequence of the two

ways specialists measured the US-Soviet competition.

Royall Brandis textbook suggests that the PPF gained

popularity because of its use in US Soviet comparisons.

McConnell

McConnell published the first edition of his

textbook, Elementary Economics, in 1960. There,

apologizing to the reader for the brevity of the survey of a

1
Documentation of the treatment of Soviet growth and the use of the PPF in all principles textbooks catalogued in the Library of
Ccngress under HB 171.5 between 1948 and 1970 is available on request.
3
very complex economy (1960, p. 718), he offered a snapshot view of the US and the Soviet economies. He

did so using a pie graph which appeared with only minor variations in McConnells chapter on the Soviet

economy through the 1990 edition.

McConnells second edition (1963), now entitled Economics: Principles, Problems and Policies,

added a discussion of the growth record of Soviet Russian controlled economy (1963, vi). The 50% ratio

reappears in McConnell for most of the editions that follow. In the 1990 version McConnell and his

co-author presented the same Soviet-US output ratio of 50%. Twenty-seven years have passed in which the

Soviet economy has purportedly been

characterized by greater investment, which

equates to faster growth, and yet remained

at half the size of the US economy (1990, p.

857). The discussion below begins with

McConnells use of pie charts and then

turns to his chapter summary.

The pie chart cross country

comparison is a staple of McConnells

textbook. It always appears in the last chapter. From 1975 onwards, US investment is said to be about half

that of the Soviet Union as a share of GNP and yet the ratio of Soviet to US GNP never changes. The table

below presents the contents of the McConnell pie chart over all editions of the textbook.

4
 The US v USSR in McConnell
Edition Figure US Investment USSR Investment GNP Ratio
(% of GNP) (% of GNP) US to USSR

1960 38.1 19% 27% 5:2

1963 40.1 19% 27% 2:1
1966 42.1 20% 33% 2:1
1969 44.1 19% 31% 2:1
1972 46.1 19% 31% 2:1
1975 45.1 15% 30% 2:1
1978 45.1 15% 30% 2:1
1981 45.1 15% 28% 2:1
1984 46.1 15% 33% 2:1
1987 44.1 14% 33% 5:3
1990 42.1 14% 33% 2:1

In McConnells pie chart, consumption and investment always sum to 100% of GNP; there is no

room for inefficiency.2 Consequently, investment in the two countries is equally effective and, without some

disturbing cause from outside the economic system, higher investment in one country manifests itself in

higher growth. But there is no catching up. Thats what the chart seems to be saying.

McConnells explanations also refer to higher Soviet investment and growth without catching up. In

1963 he writes that Soviet GNP is roughly one-half that of the United States. (1963, p. 754) and the rate

of economic growth is two or three times as great as that of the United States. In 1975 we read: Although

the Soviet GNP is only one-half as large as that of the United States, the Soviet GNP has grown more rapidly

than ours (1975, p. 905). In 1984, Although the Soviet GNP is only one-half as large as that of the

2
Our reproduction above is from the 1963 edition. McConnell cited the same sources for the pie charts in 1960 and 1963.
However the ratio of reported Soviet to US output changed from 40% to 50% in the three year period.

5
United States, the Soviet GNP has grown more rapidly than ours (1984, p. 837). In 1987 the text notes

that Although the Soviet GNP is only about 60 percent as large . (1987, p. 911). It comes as a relief of

sorts that the sentence is removed from the 1990 edition, leaving only a statement about the high historical

growth rates in the Soviet Union (1990, p. 865).

McConnell also discusses Soviet and US growth rates separate from the issue of catching up. Starting

with 1963, McConnell states that the annual rate of growth in the Soviet Union is two or three times as great

as that now achieved in the United States (1963, p. 750). In 1966, he writes in italics that the Soviet GNP

had been expanding at about 6 to 7 per cent per year as compared to 3 to3 per cent per year for the United

States (1966, p. 766). From 1966 until 1990, McConnell cites the lack of cyclical unemployment in the

Soviet Union, something which makes the US less efficient, as an additional explanation for the Soviet

expansion relative to the US (1966, p. 768, McConnell and Brue1990, p. 856).

In the 1966 edition McConnell adds a discussion of Sources of Soviet Growth (1966, pp. 767-68)

in conjunction with Possible Retarding Factors (1966, pp.768-70) and Possible Accelerating Factors

(1966, pp. 770-72). In the retarding group he includes the possibility of an increase in consumption (p.

768), the changing nature of investment from the aging of the capital stock, a reduced ability to profit from

western technology, labor shortages, a variety of problems in agriculture and planning problems. Each of these

is matched by possible accelerating factors, e.g., Against the tendency of Soviet planning to become less

efficient in the face of an increasingly complex economy must be set the likelihood of significant

breakthroughs in the techniques and mechanics of central planning (p. 771). The 1969, 1972 and 1975

editions contain the same discussion of sources of growth and retarding and accelerating factors (1969, pp.

812-17, 1972, pp. 812-17, 1975, pp. 897-902).

The editions of 1978, 1981 and 1984, however, retain the retarding factors but omit the possible

accelerating factors in the growth section (1978, pp. 935-38, 1981, pp. 887-89, 1984, pp. 831-33).

6
McConnell adds a discussion of the recent growth slowdown to the 10th and 11th editions (1987, pp.

905-07, 1990, pp. 857-59).

All editions contain estimates of Soviet and American growth. In 1969, McConnell writes in italics

that the Soviet GNP had been expanding at about 6 to 7 per cent per year as compared to 3 to 4 per cent per

year for the United States. (The previous edition, as noted above, had the US growing between 3 and 3.5 per

cent.) In 1972 this becomes: But Soviet growth performance slackened to about 5 percent per year in the

1960s, and projections for the 1970s suggest possible further deterioration to 4 to 4 percent. The latter

figures are quite close to the full-employment growth rates in the United States. In short, the substantial

growth rate advantage which the Soviet Union enjoyed in the 1950s and early 1960s has tended to diminish

and disappear. (1972, p. 812). The same passage appears in 1975 (1975, p. 897).

In 1975 McConnell added a discussion of the hypothesis of institutional convergence, something

which continues for several editions. In 1978, McConnell finds evidence of [growth] convergence in recent

years: Soviet annual growth was 4.9 percent in the 1960 period while the United States enjoyed a 3.8

percent rate. Both economies performed poorly in the first half of the 1970s as reflected in a 3.5 percent

annual rate for the Soviet Union and a 2 percent rate here at home (1978, 933-34). The same passage is

found in the 1981 edition (1981, p. 886). In 1984, a new table shows Soviet growth falling to 2.7% in the

late 1970s (1984, p. 830) with the text telling us that Experts also agree that the Soviet growth rate has

generally exceeded the United States in the post-World War II period as a whole (1984, p. 830). The table

does not offer United States growth for comparison. In 1987, the question of the recent growth slowdown

is raised with Soviet growth having now fallen to 2.6% per annum for the 1981-83 years (1987, p. 905).

Yet none of these variations influences the visual presentations in the pie charts.

In 1990 McConnell and Brue provide estimates of US to the USSR growth from 1961-1965. Soviet

growth of 4.8% now differs only slightly from the US growth of 4.6% (1990, p. 856); the pie chart, however,

7
remains unchanged. Less dramatically, Soviet growth in the 1990 table has been revised downward from the

1987 table. What had been a 5.0% growth rate from 1971-65 in 1987 is now revised to 4.8%; a 3.7%

growth rate from 1971-75 is changed in 1990 to 3.1%; what was 2.7% for 1976-1980 is now revised to

2.1%. The 1990 text reports that from 1983 to 1988 the Soviet growth rate of 2% was overwhelmed by the

US growth rate of 4.0%.

Samuelson

In 1961, Samuelson took a step beyond McConnell and devised a graph which provides a rough and

ready forecast of Soviet and American growth trajectories. Using different assumptions about Soviet and

American growth rates, Samuelson projects when the Soviet economy will overtake the US economy. His

first projection (a max-min overtaking point) is based on the maximum respectable Soviet growth assumption

and the minimum respectable American growth assumption. The second date, a max-max prediction, is more

cautious about when the overtaking will occur: it uses the maximum Soviet growth assumption and the

maximum American growth assumption.

In the 1961 graph reproduced below the max-min year seems to be 1984; the max-max year is about

1997. So the optimistic forecast of time before

the Soviet overtaking is 23 years; the more

pessimistic time to overtaking in the max-max

world is 36 years. The non-overtaking

trajectory is constructed on the specification

that something reduces Soviet growth in out

years below what simple extrapolation would

have it. Presumably this possibility is an

exogenous shock outside the scope the theory.

8
 Like McConnells pie chart, Samuelsons overtaking

graph is always the first graph in the last chapter of the book.

There are however two editions, the 7th of 1967 and the 8th of

1970 (reproduced here), in which the projection is the first

graph students see. Perhaps the projection is offered as a key

motivation to study economics.3

The table below presents the year of Samuelsons

editions, the placement of the graph, the two forecasts obtained

from reading the graphs and the ratio of US to USSR starting

points. There are two regimes in the history of the graph. One

which is shaded preserves the 1960 starting point. The second

regime rebases the forecast to the year of the edition.

The table also presents the starting ratio of American

and Soviet real outputs used in each graphical representation.

From 1961 through the 1973, this is constant at 100/50. Only

in the last two graphs does the ratio change. The claim of

considerably faster Soviet growth made by the graph is juxtaposed with an unchanging ratio of levels of real

output notwithstanding the obvious fact that sustained faster growth would have to imply a change in the

ratio.

3
From the 6th through the 10th editions Samuelsons inner lining has line graphs comparing many different countries. McConnell
has a table of National Income Accounts.
9
 Samuelson Growth Forecast
Year Figure Max-Min Max-Max Start GNP
Overtaking Overtaking Ratio: US to
Time Time USSR

1961 37.1 23 36 100/50

1964 38.1 20 33 100/50
1967 1.1 & 40.1 10 28 100/50
1970 1.1 & 42.1 18 35 100/50
1973 43.1 17 35 100/50
1976 43.1 16 32 100/57
1980 43.1 22 32 100/55

As noted above, the graphs appeared twice in Samuelsons 1967 and 1970 editions. The following

information accompanied the second appearance of the graph:

In the decade preceding 1970, the United States grew toward the top of its projected range of growth
rates. But the U.S.S.R., because of bad weather and crops and shortening of the workweek, seems to
have moved lower down on its projected range of growth rates (1970, p. 831)

Exogenous elements such as bad weather and an unforeseen political decision to increase the consumption of

leisure intervened to throw the forecast off.

The numbers in the chart above are derived from the graphics in the textbooks. Samuelson also

provides commentary about Soviet productivity and bad weather. In fifth edition (1961) he offers the

following discussion:

The decision of how to combine various productive factors land and labor, degree of
mechanization appears to depend on a mixture of purely technical considerations and adaptions to
the scarcity of various economic resources. A continual process of trial and error goes on. The
observer finds operations curiously uneven: on the one hand, he may see a military ballistic plant
which has achieved a precision of ball bearings and gyroscopes rivaling the best in the world; on the
other hand, he may find things being done an almost unbelievably primitive way, with the quality of
output practically worthless. (Example: A Soviet farmwoman may be assigned one cow to take care
10
 of; on a Wisconsin dairy farm, a man and wife may take care of 30 cows, in addition to performing
countless other daily chores. (1961, p. 826)

The same paragraph with 30 cows appears in sixth edition (1964, p. 803). The graph in the 1964 edition is

accompanied by this additional information

From 1960 to 1964 it would appear that the United States has moved at the very top of its projected
range. But the U.S.S.R., because of bad weather and crops and shortening of the work week, may have
moved at the bottom of her projected range. (1964, p. 807)

The productivity paragraph appears in the seventh edition (1967, p. 786). In 1967 as well Figure 40-1, but

not 1-1, appears with the ending date above changed from 1964 to 1967.

Samuelson rewrote the productivity paragraph in the 8th edition, updating the example from 30 to 50

cows and bringing mechanization to Wisconsin:

(Example: A Soviet farmwoman may be assigned one cow to take care of; on a mechanized Wisconsin
dairy farm, a man and wife may take care of 50 cows, in addition to performing countless other daily
chores. (1970, p. 827)

He accompanied the graphical information in 42.1, but not in 1.1, with this information:

In the decade preceding 1970, the United States grew toward the top of its projected range of growth
rates. But the U.S.S.R., because of bad weather and crops and shortening of the workweek, seems to
have moved lower down on its projected range of growth rates (1970, p. 831)

In the 9th edition the productivity paragraph is unchanged with 50 cows (1973, p. 879). The graph (1973, p.

883) is not qualified by bad weather although a footnote on the page before remarks In the last dozen years

both growth records were comparable. (1973, p. 882).

In the 10th edition, the ratio of US to USSR output changed to 100:57 (1973, p. 883). The

productivity paragraph contained a new piece of information:

Private allotments of land on the collective farm often have much higher, not lower productivity than
the collectivized sectors. (1976, p. 879).

The text mentions bad weather in a footnote and predicts that

Despite unfavorable weather, much-improved efficiency would seem technically feasible in the future.
(1976, p. 881).
11
In the 11th edition, the productivity paragraph italicizes higher (1980, p. 822); the ratio of US to USSR

output now falls to 100:55. (1980, p. 825). The edition drops the bad weather explanation and now simply

refers to the unfavorable past:

Despite this unfavorable past, much-improved efficiency would be technically feasible in the future
(1980, p. 824).

Bach

Not all the textbooks in the period assumed efficiency in the same manner that Samuelson and

McConnell did. We consider next two additional texts from the same era. The first is George L. Bachs

Economics: An Introduction to Analysis and Policy. Although this textbook was first published in 1954, it

descended from the 1940s text written jointly with Mary Jean Bowman. Instead of opening his discussion of

Soviet-American growth with a production possibility frontier, Bach discussed the details of the American

economy.

In the first edition, he paid tribute to Soviet growth.

No other nation in history has industrialized at anything like the rate of the Soviet economy since the
early 1920s. From a backward, rural economy, in less than three decades Russia has become the
worlds second greatest industrial power. Her current military force may be the worlds greatest
(1954, p. 698).

The ranking continued through the 10th edition (1980, p. 674). In the third edition Soviet aggregate growth

was said to be something like double the American rate in the aggregate (1960, p. 826).

Yet in contrast with McConnell and Samuelson, Bach mentioned (1960, p. 841) dissenting voices in

the profession, noting that the NBER studies by Nutter and Gregory Grossman suggest that the spread

between Soviet and American growth rates may not be as wide as has been commonly supposed. In the sixth

edition, Bach reported that Soviet growth had in the last decade just about equaled that of the U.S.A.

(1968, p. 572). He noted: Comparing U.S. and U.S.S.R. growth rates is one of the favorite statistical

games of the generation .... (1968, p. 573). Per capita gross national product, however, solidly indicated a
12
discrepancy:

Our $3,900 per capita g.n.p. in 1967 compared with only about $1,700 for the U.S.S.R. Russian per
capita output is about two-thirds that in the major West European nations, and during the 1960s she
has gained on them and us little it at all. (1968, p. 573)

As a partial explanation, Bach pointed to unemployment caused by a failure of planning in the Soviet Union:

The Soviet central planners do not program men or machines into unemployment. If there is
unemployment, it is because planning has gone awry or because someone is not behaving according to
plan. But there is still the problem of keeping total spending power roughly equal to goods available
for purchase. (1968, p. 571)

In 1968 Bach adds a new table of hypothetical growth rates (1968, p. 574), beginning with a base of

100 for the US and 50 for the Soviet Union. Over the period of 1965-1995, the US growth numbers are

allowed to be 2, 3 and 4%. The Soviet numbers are allowed to be 4, 5, 6, 7%. Bach sees little need for alarm:

Unless we fall short of our historical 3 to 4 per cent growth rate, the Russians cant catch us before
the 1990s unless they growth at 6 per cent or better, a very high rate. (1968, p. 574).

The hypothetic table was removed from the 7th edition (1971) and replaced by a table showing Japan growing

over 10%, Soviet growth at 4.1% and US at 3.1% (1971, p. 691).

In the ninth edition Soviet growth is said to have slowed even more (1977, p. 650) and black markets

have arisen:

The Russians are human beings too, and when there isnt enough to go around, they apparently tend
to plan a little more resources for all the demands than there are to parcel out. The result is a demand
pressure that tends to bid up prices all along the line as shortages occur; black markets spring up
everywhere. (1977, p. 651)

Per capita incomes ($7,000 v $2,700) were offered as evidence of a worsening of relative US / Soviet

well-being 2.3 in 1968 to 2.9 in 1977. Bach again noted differences of opinion on this matter: Some

observers put Russian per capita incomes as high as half ours, but not more. (1977, p. 651). The

hypothetical growth table stays missing.

Comparisons of American-Soviet per capita income become more complicated by 1980: $9,600 in

the United States, $3,500-$4,500 in the USSR. (1980, p. 675). In this edition, a new piece of information
13
is added, a Soviet cartoon showing a planning confusion and an anecdote a standard Russian joke about

incentives: 4

The accompanying cartoon shows the Russian wryly observing the problem. In a market system, the
profit incentive continually pushes managers to avoid such inefficiencies. Under quotas and physical
planning, the managers goal is to meet the quotas, not to question why. A standard Russian joke has
a factory turning out one huge one-ton nail as the cheapest way to meet its quota of one ton of steel
nails. (1980, p. 676).

In the new appendix the unemployment which was recognized in 1968 has a name: disguised

unemployment.

The Soviet central planners do not program men or machines into unemployment. If there is
unemployment, it is because planning has gone awry or because someone is not behaving according to
plan. And there has been very little official unemployment in the USSR. But there is much evidence
of disguised unemployment in the form of excess labor allocated to some industries. (1980, p. 690).

The textbook no longer predicted the Soviet Union overtaking the US and the fixed ratio of Soviet/US

output disappeared. The table which showed Soviet growth exceeding US growth appeared in only one

edition. Bach provided no explanation to accompany the specification; this would have required a PPF, which

his textbook does not have.

Tarshis

Tarshis began his 1947 Elements of Economics with a clear statement that an unguided economy will

be inefficient:

It is, of course, not surprising that our economy sometimes fails to operate at peak efficiency.
Most of the economic institutions we have inherited were not designed by economists; certainly the
basic ones were not. In fact, they were not designed at all. Students of economic history can trace their
gradual evolution under the pull and tug of various interest groups: of the landlord and the
businessman, the merchant and the Church, the wage earner, the investor, and the bureaucrat. Anyone
who has studied the development of these institutions will not be surprised to find that they do not
always perform efficiently. After all, it is rarely enough that what we plan turns out as we planned it;
it would be remarkable indeed if something which grew without planning should perform in just the
way we want it to. If doctors had designed the human body, there would probably be no diseases.
Likewise, if economists had designed the economy, the chances are that there would be no economic

4
Peart and Levy 2005 discuss the anecdotal evidence aggregated in proverbial wisdom. Standard Soviet jokes are studied in Levy
and Peart (2006).
14
 problems to worry about. As it is, the economist believes his job is to understand the existing
economy in order that he can properly guide efforts to make it work efficiently. (1947, pp. 4-5).

This provoked some controversy. Rose Wilder Lane correctly read the book as advocating political solutions

in lieu of voluntary solutions (1947). Yet Tarshis reformist point of view may have immunized his analysis

from an assumption of economic efficiency. 5 Indeed, the Soviet section in his much ignored Modern

Economics of 1967 contained no such assumption. Tarshis brief discussion there addressed the implicit

assumptions behind the perennial issue of the relative growth rates of the US and the Soviet Union (1967,

p. 663). He began by asking what index we ought to use. Output per capita was problematical because it

assumed a roughly constant and equal labor input across countries (1967, p. 661).

Tarshis table (1967, p. 663), reproduced below, illustrates the problem with all US and Soviet

comparisons. A simple switch of weights from US to USSR prices cuts the relative size of the Soviet economy

in half.6

5
Harcourt (1995) emphasizes the importance of Tarshis assumption of imperfect competition. Tarshis employs the kinked
demand model to argue for price rigidity in oligopoly (1947, pp. 182-4). Samuelsons chapter on the firm treats perfect and
imperfect competition together (1948, p. 491-17) and does not consider the possibility of oligopoly-induced price rigidity.
6
This consideration is noted in Samuelson as a technical index number problem involved that need only be indicated
which gets solved by simply splitting the difference in all the editions; (1961, p. 828); (1964, p. 806); (1967, p. 790);
(1970, p. 830); (1973, p. 881); (1976, p. 882); (1980; p. 824).
15
 Tarshis then considered how to select time periods for the purposes of comparison, pointing out that

a short period analysis makes the choice of end points critical (1967, p. 661-62). Relying on Simon

Kuznets work, he compared growth rates of 19 countries. Since the time period spanned both Russian and

Soviet history (1870 1954), he chose to treat them as one country (1967, p. 666), a convention Nutter had

also followed that was widely and strenuously criticized (Brady 2008). Tarshis conclusion would not

surprise a reader of the 1st edition who remembered his emphasis on the economists role as reformer:

It is noticeable that most of these countries showed a decline in growth rate, comparing the two
periods. The exceptions were Italy and the Soviet Union, to which industrialization came late;
Sweden which succeeded in maintaining a high rate throughout an economist cannot refrain from
pointing out that Sweden; more than any other country in the group studied, has followed the advice
of its economists who incidentally have been exceptionally able. (1967, pp. 666-67)
This textbook, one notes with some regret, disappeared without a trace soon after it appeared.

The treatment of Soviet-American growth in Tarshis 1967 textbook suggests that the comparisons

depend very much on ones point of view. Tarshis makes no appeal to cross societal efficiency. He offers no

guidance for the future of US-Soviet competition.

Heilbroner

Robert Heilbroners 1968 textbook positions itself farther than any other from Samuelsons

neoclassical synthesis. Heilbroner presupposed that market economies require a culture of self reliance free

of direction either by tradition or command. He divided the world into regions of command, markets and

tradition in a map shown here from the first edition inner liner (Heilbroner 1968). For Heilbroner the

Soviet Union must be viewed in terms of economic development in which its institutions evolve (1968, pp.

599). He questioned both the assumption of constant growth and that of fixed institutions, remarking that

growth was decelerating (1968, p. 628) and the Soviet Union was moving toward markets (1968, pp. 599).

Heilbroner predicted a convergence of systems (pp. 629-30) as institutions developed toward some

advanced form.
16
 Samuelson captured the key features of economies using tradeoffs of guns and butter. By contrast,

Heilbroner discussed tradition, markets and command before he turned to guns and butter. For him, it was no

easy matter to compare economic activity in America with economic activity in the Soviet Union since one

model failed to encompass both economies.

Heilbroner occupies a distinct ideological position among textbook writers.7 Interestingly, Arnen

Alchian and William Allen shared Heilbroners reluctance to present an encompassing model for all

economies. Their University Economics comprised the Chicago-school alternative to Samuelson at this time.

Alchian and Allen explain why their book does not analyze any form of socialist economy: neoclassical theory
7
Heilbroner (1968, p. 3): no other branch of study holds such possibilities for the improvement of the human condition in a
world that is, in the main, still brutally poor. I do not mean that the rescuers of mankind must be economists, although I myself
believe that the appeal of economics is greatest to those who feel affronted at the miseries and inequities of the human
spectacle. Without minimizing Stalinist horrors he asks the student to reflect on the brutalities of capitalistic development
(1968, p. 595).
17
failed to characterize the Soviet economic system.8

Efficiency and Ideology in the Textbooks

Three major hostile reviews of the new generation of textbooks shaped the textbook landscape of the

1960s. Rose Wilder Lane reviewed Tarshis (Lane 1947), Orville Watts focused on a cluster of new

economics including Samuelson (Watts 1950), and William Buckley surveyed the textbooks used at Yale,

including those by Tarshis and Samuelson (Buckley 1951). Buckleys attack echoed those by Lane and

Watts but, as Elzinga (1992, p. 864) has noted, Buckley also asked why the positions of non-Keynesians were

excluded from the textbooks (Buckley 1951, p. 81).

Tarshis advocated reform within a particular economic framework. Samuelson, by contrast, took a

view that abstracted from any particular economy and so gave up the economists claim to the particulars that

might guide reform. Nor did Samuelson suggest that an economy cannot be efficient without direction by the

economists. Perhaps this is why the attacks on Tarshis ensured that the textbook would not survive into

immediately subsequent editions while the Samuelson textbook easily withstood the early criticism.

Between 1960 and 1980 it was widely believed that disinterested experts in the Soviet Union could

create a system of institutions such that the efficiency of market capitalism would be combined with the

ethical claims of socialism, the once-celebrated model of market socialism. The analysis assumed that those

directing the system were trustworthy so it was apparently plausible to write about the efficiency of

non-capitalist economies (Levy and Peart 2008). All of this implies that the economy which consumes least

will grow faster than the one which saves less.

Previous accounts of this periods textbooks have emphasized an ideologically-inspired

romanticization of planned economies (Skousen 1997). Our account, however, suggests that the treatment
8
The portion of economics comprising the theory of exchange is applicable to a wider class of problems in a capitalistic
private-property economy than it is in a socialist society. This does not mean there is no exchange in the latter; there is, of
course, a great deal of it. ... In a socialist system ... political power and exchange of non private-property rights are used much
more widely to solve the economic questions. If were to devote primary attention to the socialist system, we would investigate
much more fully the processes of political exchange and political decision making. Alchian and Allen (1964, p. 6).
18
was not driven by ideology alone. Tarshis ideology was similar to that

in Samuelson. Yet his very non-romantic view of the Soviet economy in

1967 was doomed to oblivion by the earlier assessment of the libertarian

reviewers. Ideological explanations also neglect how some researchers

such as Warren Nutter and W.W. Rostow, united across ideological

differences to oppose the CIAs use of the Soviet-US growth estimates

(Nutter 1958, pp. 231-32, Nutter 1964, Lipsey 2008). In his letter to

Nutter, reproduced above, Rostow referred the intellectual

establishment against which they were united (Rostow to Nutter 22

October 1963).

Samuelson pioneered the use of the production possibility frontier, reproduced here from his first

edition, to conceptualize production in different societies. Frank Knight asked What, how and for whom

and Samuelson, his student, combined Keynes and Knight in a simple snapshot that addressed Knights first

question, the fundamental problem facing any economy, what shall be produced.9 In Samuelsons

formulation, abstracting from the possibility of involuntary unemployment (Keynes concern), all economies

are efficient. Thus Samuelson wrote about the Soviet economy in terms of the production possibility frontier

in his first edition:

The Russians, having no unemployment before the war, were already on their Production-possibility
curve. They had no choice but to substitute war goods for civilian productionwith consequent
privation. Samuelson (1948, p. 20)

Tarshis and Samuelson believed in activist government policy to counteract what they saw as the

inherent instability and inefficiency in democratic capitalism; as a result, their books were attacked for being

9
What, How, and For Whom. Samuelson (1951, p. 14): These three questions are fundamental and common to all economies
The footnote credits the formulation of this approach to Frank Knights Social Economic Organization. The Knightian
foundation to Samuelsons text is discussed in Emmett (2008).
19
liberal and anti- market. They differed, however, on whether the economy might be represented by a PPF a

thin model or a thick description.

The following matrix captures the differences in textbooks:

Soviet Growth Overstatement: Ideology or Model Fragility?

Liberal Neutral
Thin model Samuelson McConnell
Thick model Tarshis, Heilbroner Bach

If the overstatement of Soviet growth was mainly driven by ideology we would expect Samuelson,

Tarshis and Heilbroner to overstate Soviet growth more dramatically than McConnell and Bach. They did

not. If the problem arose primarily because of the use of thin models, Samuelson and McConnell would

overstate Soviet growth more dramatically that the other texts. This is what we have observed above.

We noted above that Bach drew this students attention to the Nutter-Grossman research. Their

approach (Nutter 1957, 1962 and Rostow 1960), measured the years by which Soviet industry lagged

American industry. This required a belief that when Russia became the Soviet Union, its fundamental

institutions did not change. Tarshis skepticism about Soviet growth followed from a similar belief that,

institutionally, the Soviet Union remained similar to Russia.

The PPF

What drove the widespread adoption of the PPF by textbooks at this time? One explanation is that,

without the PPF, one cannot really model the Soviet economy at the level of a principles textbook. In

particular, the author cannot simply appeal to the historical record to illustrate the Soviet case because there

were two competing ways to read the historical record. The PPF allowed the author to accept one empirical

approach as opposed to the other. The implication of our account is that when the demand for Soviet analysis

in principles textbooks increased after Sputnik, textbooks came increasingly to rely on the PPF. An indication
20
of this transition are Rendig Fels 1961 purely empirical textbook, which contained no PPF, and Royall

Brandeis theoretical treatment, which relied heavily on the PPF.

The Challenge to the American Economy, whose very title refers to Khrushchevs boast that the

Soviet Union will bury American economically, dismissed Nutter and those who relied on Nutter's research.

After quoting Khrushchev and presenting standard growth comparisons, Fels turned to Henry Hazlitts use of

Nutters data:

The table appears to justify amply Hazlitts conclusion that the evidence has been
unmistakable that, far from there being any miracle of Communist production, the lands behind the
Iron Curtain are going through an economic crisis. These results are in striking contradiction to the
figures cited in the previous section. There it was proved that Communist output grows twice as fast
as American. Here it is proved that under Communism output lags farther behind. Plainly it cannot
do both at once. (Fels 1961, p. 11)

Fels (1961, pp. 11-12) criticized Nutter for using figures based in 1913 and Hazlitt for his blind

faith in the goodness of capitalism, a belief that reportedly biased his analysis:

Popular writing on economics is strewn with errors. Although it is perhaps unfair to expose
Hazlitt in one, since he generally maintains a high standard, nevertheless for educational purposes it
is valuable to do so. The fact that one of the best has fallen into a trap is a warning to read everything
including his book with a healthy degree of skepticism. The probable cause for his going astray is
likely to infect anyone. Hazlitt is a stronger believer in capitalism and a vigorous critic of any other
economic system (Fels 1961, p. 12)

Nutters data disappear from the second edition (Fels 1966).

Fels textbook was in some respect thick; it contained no PPF and in other ways it was less abstract

than Samuelsons textbook. But as a thick textbook, it contained no abstract model with which to coherently

predict continued rapid Soviet growth. He chose not to believe Nutters approach but that choice was

apparently theoretically unmotivated.

To illustrate why the PPF was so attractive at this time to textbook writers , consider how Royall

Brandis adopted the PPF midway through his textbook run as a way to model the Soviet economy. His was

perhaps the most elegant use of the PPF in the period. In the note to the instructor Brandis explained the

21
addition: the production- possibility curve

technique is developed and then used to

explore briefly a number of economic

problems generally familiar to the student.

These questions are not only important in

their own right but serve to fix the analytical

took in the students mind ... (1968, p. vii).

Brandis analysis is remarkable for making

explicit the supposition that investment goods

will shift the production possibility frontier

for the future more than consumer goods. Thus, higher investment brings about faster growth (Brandis 1968,

p. 34). The assumption of efficiency is critical here and is often viewed as the default assumption in the

absence of involuntary unemployment.

Conclusion

Thin models yield crisp implications. Models that supposed cross societal efficiency implied that the

Soviet Union must overtake the US economy because Soviet consumption was lower than US consumption.

Faster growth over time yields overtaking. Bad weather or some such secondary explanation then freed the

model from falsification. Nutters joke about the endogenous past captures the difficulty.

Yet one reason for the simplicity of thin models, of course, is that complicating dimensions are

missing. Those textbooks which did not assume efficiency avoided the problems associated with assuming

away complications. In 1980, Bach reported a standard Soviet joke about the quota-making nail. The same

joke appeared a decade earlier in the 1970 edition of Heilbroners book. A cartoon in the Russian satirical

magazine Krokodil depicts a nail factory proudly displaying its record output: one gigantic nail suspended

22
from an immense gantry crane (1970, p. 627).

That books with different ideological viewpoints could come to the same conclusion about the Soviet

economy suggests that more than ideology was required to get the story right. Instead, the application of

thin models with no account of institutional details led researchers into what we now know was an overly

optimistic account of Soviet growth. The moral we draw from the treatment of Soviet growth in American

textbooks is that multiple points of view are potentially useful when we study non-transparent institutions.

When the voices of fine scholars such as Lorie Tarshis and Warren Nutter were disregarded because of their

dissenting points of view, the profession became less able to penetrate Soviet non-transparency than it might

otherwise have been. We are all constrained by means of models: we gain insight in one dimension by

blinding ourselves to events in other dimensions. Competition among models may be necessary to insure that

the benefits of the models exceeds their cost.

23
 Bibliography

Alchian, Armen A. and William R. Allen. 1964 University Economics. Belmont: Wadsworth Publishing.

Bach, George Leland. 1954-1980. Economics: An Introduction to Analysis and Policy. Englewood Cliffs:
Prentice-Hall.

Bowman, Mary Jean and George Leland Bach. 1943-1949. Economic Analysis and Public Policy. New York:
Prentice Hall.

Brandis, Royall. 1968. The Principles of Economics. Homewood, IL: Irwin.

Brady, Gordon. 2008. The NBER Soviet Study. Summer Institute for the History of Economics. George
Mason University.

Buckley, William F., Jr. 1951. God and Man at Yale. Chicago: Regnery.

Colander, David C. and Harry Landreth. 1995. Editors. The Coming of Keynesianism to America:
Conversations with the Founders of Keynesian Economics. Cheltenhaum, UK: Edward Elgar.

Colander, David C. and Harry Landreth. 1998. God, Man, and Lorie Tarshis at Yale In Omar Hamuda.
Editor. Keynesianism and the Keynesian Revolution in America . Cheltenhaum, UK: Edward
Elgar.

Elzinga, Kenneth G. 1992. The Eleven Principles of Economics. Southern Economic Journal 58:4 (April):
861-79.

Emmett, Ross. 2008. Frank Knight and The Economic Organization. In Robert Leeson. The
Anti-Keynesian Tradition: Archival Insights into the Evolution of Economics. New York: Palgrave.

Fels, Rendig. 1961-1966. The Challenge to the American Economy: An Introduction to Economics. Boston:
Allyn and Bacon.

Gerschenkron, Alexander. 1978. Samuelson in Soviet Russia: A Report. Journal of Economic Literature
16:2 (June): 560-73.

Harcourt, G. C. 1995. 1995. Lorie Tarshis, 1911-1993: In Appreciation. Economic Journal 105:
1244-1255.

Heilbroner, Robert L. 1968-1970. The Economic Problem. Englewood Cliffs: Prentice Hall.

Lane, Rose Wilder. 1946. Review of Ludwig von Mises, Omnipotent Government. Economic Council Review of
Books 3 (August)

Lane, Rose Wilder. 1947. Review of Lorie Tarshis. Elements of Economics. Economic Council Review of
24
 Books 4 (August)

Levine, Isaac Don. 1932. Red Smoke. New York. Robert M. McBride & Company.

Levy, David M. and Sandra J. Peart. 2006. The Fragility of a Discipline When A Model Has Monopoly
Power. Review of Austrian Economics 19: 125-36.

Levy, David M. and Sandra J. Peart. 2008. Socialist Calculation Debate. The New Palgraves Dictionary of
Economics. Edited by Lawrence Blume and Steven Durlauf. Second edition. New York: Palgrave.

Lipsey, Richard. 2008. Assessements of the Soviet Union: Who Got It Right and Who Got it Wrong.
Presented at the Summer Institute for the History of Economics.
http://www.gmu.edu/centers/publicchoice/SummerInstitute/si_schedule08.htm

McConnell, Campbell R. 1960-1990. Economics: Principles, Problems, and Policies. New York:
McGraw-Hill.

Nutter, G. Warren. [1958] 1983. Economic Warfare. In G. Warren Nutter. Political Economy and
Freedom: A Collection of Essays. Edited by Jane Couch Nutter, pp. 222-36. Indianapolis: Liberty
Fund.

Nutter, G. Warren. 1962. The Growth of Industrial Production in the Soviet Union. Princeton: National
Bureau of Economic Research.

Nutter, G. Warren. [1964] 1983. The Soviet Stir: Economic Crisis and Response. In G. Warren Nutter,
Political Economy and Freedom (ed). Jane Couch Nutter, pp. 181-88. Indianapolis: Liberty Press.

Nutter, G. Warren. 1969. The Strange World of Ivan Ivanov. New York and Cleveland: World Publishing.

Peart, Sandra J. and David M. Levy 2005. The Vanity of the Philosopher: From Equality to Hierarchy in
Post-Classical Economics. Ann Arbor: University of Michigan Press.

Rostow, W. W. 1960. The Stages of Economic Growth: A Non-Communist Manifesto. Cambridge:

Cambridge University Press.

Rostow, W. W. 22 October 1963. Letter to G. Warren Nutter. In the possession of Richard Ware.

Samuelson, Paul A. 1948-1980. Economics: An Introductory Analysis. New York. McGraw-Hill.

Samuelson, Paul A. 1997. Credo of a Lucky Textbook Author. Journal of Economic Perspectives 11
(Spring): 153-69.

Skousen, Mark. 1997. The Perseverance of Paul Samuelsons Economics. Journal of Economic
Perspectives 11 (Spring): 137-152.

Tarshis, Lorie. 1947. The Elements of Economics: An Introduction to the Theory of Price and Employment.
25
 Boston: Houghton Mifflin.

Tarshis, Lorie. 1967. Modern Economics. Boston: Houghton Mifflin.

Watts, V. Orval. 1950. Review of New Economics Texts. Economic Council Review of Books 7 (August
& September.

26