Growing pains of a fiscal colossus

The study of economic history in the United States and some other countries has been revolutionised in the past 40 years by the analysis of large data sets using sophisticated statistical and econometric techniques. The explicit use of theory has been an even more important development. Other social sciences use theory explicitly, but economic history has been able to benefit from the major changes in the practice of economics. Hence, formal economic history in the US is now taught largely in economics departments and is largely a postgraduate subject, leaving sizeable parts of the subject that are not amenable to a theoretical and quantitative approach to history departments and business schools.

This book is the last of three chronological volumes, whose editors' individual contributions to many areas of US economic history have been impeccable. Although one of them (Robert Gallman) died in 1998, he did see the final versions of the chapters in the last volume. The Cambridge Histories are designed to summarise the state of knowledge, and most authors, who include many of the best-known scholars, have followed the editors' instructions. A more serious problem is the perennial one of late submission. Some contributors delivered their chapters late, which means that chapters are not as up to date as they could be, even accounting for publishing delays.

The chapters include discussions of the causes of long-run growth, the economic implications of war, overseas financial relations, trade and trade policy, the business corporation, population growth, northern agriculture, banking and finance, technical change, labour markets, government regulation of business and the public sector, the development of labour law, economic inequality, structural change and the regions, the Great Depression, and Canada. Naturally, there is some overlap but it is unobtrusive. Most chapters are of good quality, and those on war, labour markets, trade, agriculture, business corporations and Canada, excellent.

There is space to pick out only a few of many interesting questions. For example, why did western European productivity lag American for so long, and why is it still 20 per cent lower? Basically, in the early part of the century the uniquely abundant resource base of the US allowed scale economies that could not be achieved in Europe. But by the later 20th century, the US changed from a resource-dominated to a knowledge-based economy that was driven by education and applied research. On the other hand, this success was not always apparent. As Peter Lindert points out, by the 1970s and 1980s America caught up with Britain as a world leader in self-flagellation over lost markets.

Many of the contributors argue that growth has been delivered by a market that worked, with only a small role for government. One might wish to be a little careful here. Some of the analysis on which these conclusions depend is set in a neo-classical framework, which, in effect, assumes that factor prices were competitive. To take just one counter-example, in the early 20th century many types of steel that could be produced more cheaply in the US than in Britain were actually sold at a higher price. Economic inequality fell from 1929 to 1951 and was then stable until 1980. Despite rapid growth in the economy as a whole, inequality has subsequently risen. The turnaround is partly because the high-spending administrations of the 1960s and 1970s were followed by tax-cutting ones. More important, in the late 20th century, the productivity of those with skills, that is, the already well off, has increased relatively (human capital again), while many low-productivity industrial jobs have moved offshore. Interestingly, immigration in the late 20th century had little effect on inequality.

In trade, the so-called Leontieff paradox (1947) is shown to have been a special case. Leontieff showed, counter-intuitively, that US exports in the postwar period were relatively more labour intensive than US imports. But this was true only from the 1930s to the 1950s. Lindert, in an excellent chapter, gets to grips with the reasons for the relative decline of industrial exports, with the major exception of aircraft. (The aircraft industry is one of only a very few examples of a federal industrial policy, in this case dominated by military considerations.) In finance, federal regulation of the financial markets has not always been successful. For example, federal insurance of savings deposits created a serious moral hazard from the 1950s onwards. There was no incentive for the savings and loans institutions to lend wisely. They were said, according to Eugene White, to operate on a 3-6-3 principle. (Borrow at 3 per cent, lend at 6 per cent, be on the golf course at 3pm.) Hence, the savings and loan crash of 1985-86 probably cost the US taxpayer $150 billion - the present value of the support payments. This is a greater loss, in real terms, than that suffered by depositors and shareholders in the 1930s banking collapses.

In industry, the US was weak in science until the second world war. Like Japan in the 1960s and 1970s, the US applied science that had been developed elsewhere. (Chemicals are an obvious example.) The growth of in-house research and development in the 20th century, the rise of the research laboratory, may have been a reaction to federal antitrust policies. Corporations that could not increase their market share in one product developed new products themselves. But in the case of IBM, the opposite was probably true. Because of a fear of an antitrust suit, IBM invited an outside provider (Microsoft) to include its operating system alongside its own on IBM platforms, with devastating results for IBM.

In the labour market, it seems that many companies have been paying "efficiency wages" (wages above the market rate that tie workers to the firm) for a long time. This is related to the trend of wage differentials. In the second half of the century, the difference between the income of the top and bottom 10 per cent of earners has widened from three times to more than four times. Gender differences remain, but are largely explicable in economic terms. For example, the increase in female participation had the short-run effect of reducing the average level of inexperience in the female labour force, thus reducing the premium placed on experience. (In other words, cross-section data may obscure the changes within each cohort.) By the end of the century, the gender differential for white postgraduates aged 25-34 was nil and less than 10 per cent for college graduates.

The physical appearance of large parts of the US notwithstanding, agriculture has become a very minor part of the economy. In 1910, 35 per cent of the population lived on farms. In 1990, only 1.8 per cent did, of whom nearly half did not work in agriculture. The key was, obviously, productivity. Productivity growth in agriculture was close to nil from 1880-1920. Between 1920 and 1990 it increased about 15 times. In addition to the introduction of fertilisers, tractors and hybrid corn, the increased specialisation in the economy and transformation of the countryside (rural electrification, good roads and education) were critical. They provided the means for the agricultural workers (actually, their sons and daughters) to move out, in turn making farm investment profitable.

This is a substantial book on an important subject. It is beautifully produced, although there are some minor blemishes, such as its inconsistent referencing system. There are substantial bibliographical essays that are very helpful, particularly those that are in effect a literature review. In addition, large amounts of data are provided. The book summarises some technically difficult research but it is very accessible. However, few people will read the book from cover to cover. But that is not the intention of the book's editors. It is the essential starting point for the history of the American economy, for specialist and non-specialist alike.

Dudley Baines is reader in economic history, London School of Economics.