is owned and operated by the Economic History Association
with the support of other sponsoring organizations.

Capitalist Development in the Twentieth Century: An Evolutionary-Keynesian Approach

Author(s):Cornwall, John
Cornwall, Wendy
Reviewer(s):Vedder, Richard K.

Published by EH.NET (August 2002)

John Cornwall and Wendy Cornwall, Capitalist Development in the Twentieth

Century: An Evolutionary-Keynesian Approach. Cambridge and New York:

Cambridge University Press, 2001. xv + 286 pp. $60 (hardback), ISBN:


Reviewed for EH.Net by Richard Vedder, Department of Economics and College of

Business, Ohio University.

As David Colander accurately states in his forward to Capitalist

Development in the Twentieth Century, John and Wendy Cornwall “are true,

unrepentant Keynesians.” In this tour de horizon of modern macroeconomic

history, aggregate demand is the leading actor — cycles in economic

performance are determined by the robustness of aggregate demand. The Cornwalls

more or less believe in the reverse of Say’s Law: Demand creates its own


The authors believe that the best single indicator of an economy’s

macroeconomic performance is the rate of unemployment. Unemployment was

relatively low or “full” in many Westernized nations in the 1920s, and

especially in the “golden age” of the 1950s and 1960s. The reason, they

believe, is that aggregate demand was growing by healthy amounts in those eras.

By contrast, the period since 1973 has been one of sluggish economic

performance, explainable in large part by institutional (often

government-imposed) restraints in the growth in aggregate demand. The slow

growth in aggregate demand, the authors opine, has led to reduced savings,

investment, and productivity growth. Of particular importance, nations where

labor, business and government reached “social bargains” (incomes policies)

were able to stimulate aggregate demand through government policy, but most of

these social bargains fell by the wayside after 1973.

While my overall impression of the book is not favorable, it nonetheless has

several strengths. Let me mention four. First, it is reasonably well written,

using enough symbols, jargon and econometrics to keep professional economists

satisfied, yet at the same time it is clear enough for the intelligent

layperson to understand the rudiments of the main points. In an era and

profession where writing incomprehensibly is considered to be a sign of virtue

and erudition, this is no small accomplishment. To be sure, the discussion of

such things as “hysteretic processes with exogenous origins” (p.102) is filled

with typical academic pretentious jargon that would put the most diehard

Keynesian to sleep, but on the whole this book is above average in clarity for

economist-written works.

Second, the book makes an important point, that many economic model builders

ignore, specifically that institutional arrangements and the structure of the

economy matter, and often matter a great deal. Moreover, as the Cornwalls

observe, institutional arrangements change over time with economic changes, and

this can impact economic performance.

Third, while the authors are truly militant Keynesians, they realize that a

1950-style old Keynesian story simply will not cut it in today’s world. In

particular, they eschew Keynes’s emphasis on the short run, and try to evaluate

the impact that aggregate demand has on intermediate to longer run economic

growth. With the decline in the importance of the business cycle, this is a

necessary adjustment. The Cornwalls also reject or downplay much of the New

Keynesian emphasis on microanalysis of wage and price rigidities (e.g.,

efficiency wages, menu costs, and so forth). Borrowing some from ideas of the

New Institutional Economics, the Cornwalls believe that evolutionary changes in

institutions and economic structure have an important role to play in

explaining changing economic performance.

Lastly, as EH.NET readers will applaud, the Cornwalls appreciate the importance

of history, and its usefulness in assessing economic phenomena. While not

economic historians, they have written what is a somewhat less than

comprehensive but still interesting macroeconomic history of the twentieth

century within the context of trying to explain what makes the macroeconomic

world work. Yet, despite all of these virtues, this is in my judgment a badly

flawed book for a simple reason: I think the authors are just plain wrong in

their assessments. Moreover, they are not merely sporadically wrong, but

persistently and unrelentingly mistaken. To borrow a favorite Cornwallian term,

this book suffers a bad case of misguided intellectual hysteresis. To be fair,

I am not a Keynesian (although I started out as one), so a priori one would not

expect a particularly positive assessment of this work from me. But I suspect

that more neutral observers on the Keynesian/non-Keynesian continuum would find

many of the same objections.

Before enumerating some problems with the Cornwalls’ analysis, I would make an

obvious point that the issue of whether economic progress is supply or

demand-induced is not a new one. For example, many trees have been destroyed

making books on the question of whether the Industrial Revolution is best

explained by emphasizing supply or demand. In an era where demand is

increasingly taken for granted, the Cornwalls’ book does make us at least

consider the possibility that the new (post-Keynes) conventional wisdom might

be wrong.

I would also note that in some respects Cornwall and Cornwall show deference to

an early, classical tradition that in some ways is the antithesis of Keynesian

economics as practiced in the original by Keynes himself. For example, the

authors stress the importance of capital formation in long-term growth, a view

far more akin to Adam Smith than to Keynes. Original Keynesian analysis

vilified savings, the funding source for capital formation, yet Cornwall and

Cornwall believe that investment is critical to the dynamic process of long-run

economic transformation. There is a bit of Adam Smith, and also a lot of Joseph

Schumpeter, in the Cornwall and Cornwall interpretation of history.

Turning to the objections, it is argued that there are swings in economic

performance explainable by changes in the robustness of aggregate demand

influenced by institutional changes. In particular, the 1950s and 1960s were

the “golden age” of modern economies, and the era since 1973 has been something

of a disaster because of declining growth in aggregate demand.

Virtually the sole criterion used to evaluate economic performance is the

unemployment rate. Unemployment is higher in the last three decades, so

economic performance has worsened. I would suggest this is a highly

questionable basic premise as it pertains to the U.S., although it is certainly

more defensible for Europe. While average unemployment rates in the 1980s and

1990s were higher than in the 1950s and 1960s in the U.S., by most other

measures the economy in the latter period either approximately equaled or

surpassed the earlier record. Real per capita GDP grew 57 percent from 1950 to

1970 – and 55 percent from 1980 to 2000 – hardly an important distinction. Real

household wealth rose faster in the latter period, and real per capita

consumption rose by almost the same amount in both periods. Job creation was

actually greater in the latter period — the number of new jobs per 100

incremental population over 16 was 64 in the 1950-70 period, compared with 81

in the 1980-2000 era.

The authors assert that increased unemployment was involuntary in nature,

citing the rising duration of unemployment as evidence. I would argue that most

the rise in unemployment, especially in Europe, reflected onerous new labor

regulations and the impact that increasingly generous welfare state benefits

had on the desire to work. Reservation wages rose sharply as the alternative to

work — long-lived generous welfare benefits — became a viable option. Why is

the duration of unemployment more than twice as high in Germany as in the U.S.?

Germans can collect generous unemployment benefits for three to four times as

long as Americans without any adverse consequences. These unemployed are hardly

“involuntarily” out of work. A secondary factor in the unemployment rise in the

1970s and 1980s was demographic: an increase in the proportion of workers in

young age cohorts that are typically more unemployment-prone.

The Cornwalls assert that governmental macro fiscal and monetary

policies can reduce unemployment through heightened aggregate demand. It is

argued that political constraints limited the use of demand stimulus after

1973. The evidence shows otherwise. In the U.S. the federal government ran far

greater fiscal deficits on average in the two decades after 1973 than in the

two decades before. For example, in the midst of the “golden age” of the 1960s,

the federal deficit was less than one percent of GDP in eight of ten years,

while the smallest deficit in the 1980s was nearly three times that amount.

Monetary growth on average was greater in the latter era as well (the median

annual growth rate of M1 in the 1960s was 3.5 percent; in the 1980s, it was 7.0

percent). The same pattern generally is true in Europe. The Cornwalls simply

refuse to admit the problem may have been the impotency of macro stimulus, and

they claim fiscal/monetary constraint in the latter period prevented full

employment, despite the evidence that such constraint was simply not present.

The discussion of the Great Depression is also wanting. Other than the

Friedman-Schwartz monetary explanation, there is no mention of other

non-Keynesian explanations of the Depression, including ones stressing

international monetary disturbances (e.g., Barry Eichengreen), Austrian

business cycles, or the Hoover high wage policy. The Keynesian argument

explaining the Depression was made better, in this author’s judgment, by

earlier writers such as E. Cary Brown.

Moreover, there is not a scintilla of hard evidence relating to the “social

bargains” (incomes policy) allegedly common in the 1950s and 1960s compared

with later years. There is no description of how these policies worked in

specific countries, for example. We are supposed to take on blind faith the

repeated assertion that income policies worked in producing the golden age of

the 1950s and 1960s, but broke down somehow after 1973. Somehow a single

regression equation (p. 91) with no social bargain variables is construed to

support the Cornwalls’ incredibly weak argument.

The book is full of absolutely wild assertions. A few samples: “The view that

an increase in aggregate demand will not reduce involuntary unemployment

because it is unable to reduce the real wage contains the implicit assumption

that the real wage is determined in the labor market. This assumption has been

shown to be unrealistic…” (p. 46) A single unpublished paper from 1990 is

used to back up this assertion. Better yet, “Over two decades of neoliberalism

have revealed its similarities to the laissez-faire regimes of earlier times —

prosperity for the few and insecurity for many” (p. 268). To argue that in,

say, the 1990s, few had prosperity but many were economically insecure in the

U.S. or Europe is simply fiction. Speaking of the era after the golden age, the

authors claim that “the role of government in domestic and international

economic affairs has been greatly reduced, social bargains no longer dominate

labor market outcomes and price stability has become an overriding economic

goal” (p. 242). It is a fact that government spending as a percent of GDP has

risen, not fallen, in nearly every major western industrialized country in the

era since the so-called golden age, and regulatory activity has increased as

well. To say that government’s role has been “greatly reduced” simply defies

the factual evidence.

The possibility that rising unemployment and sluggish growth in Europe reflects

the debilitating effects of high taxation, regulatory rigidities, and the

disincentive effects of the welfare state is virtually ignored. There are a

variety of plausible explanations for economic changes that have occurred in

the past several decades, but the Cornwalls have not presented them. Save your

money: don’t buy this book.

Richard Vedder is co-author of Out of Work: Unemployment and Government in

Twentieth-Century America (New York: New York University Press, 1997).

Subject(s):Macroeconomics and Fluctuations
Geographic Area(s):North America
Time Period(s):20th Century: WWII and post-WWII