The Economics of Recession: A Survey
Part 1/10, August 2022 [1]
Arturo Estrella

1. Introduction

          The pace of economic growth in industrial economies tends to change relatively smoothly from quarter to quarter. To be sure, there may be substantial differences in growth rates across several quarters or years, but in the short run we tend to see more continuity. Relatively long economic expansions are the norm, periodically interrupted by shorter periods of sustained economic contraction lasting from a few quarters to, in unusual circumstances, several years. This empirical regularity of alternation between extended periods of expansion and contraction led economic observers already in the nineteenth century to the idea that economic activity goes through recurring cyclical patterns, which came to be known as “trade cycles” or “business cycles.” Eventually, economists converged on the use of the term “recession” to refer to the contractionary phase of those cycles.

          This introduction to the economics of recession references a series of original articles that represent a substantial core of the professional literature on the topic. No single source can be totally comprehensive in its coverage of the field, but the articles selected here are intended to address some key fundamental questions and to strike a balance between depth and breadth of coverage. Specifically, we consider the following questions.

·        What is a recession?

·        What causes recessions?

·        How do recessions end?

·        What are the labor market effects of recessions?

·        What other systemic effects do recessions have?

·        How can we forecast recessions?

·        How can we identify recessions in real time?

·        How can we manage the individual risks of a recession?

          Subsequent parts of the survey focus on each of the above eight questions in turn, with references to the 60 readings contained in the 2-volume book The Economics of Recession (see note 1). References to those readings are identified in the text as [R#], where # is the reading number, and bibliographic information is provided at the end of each part of the present survey.

We begin with a historical overview of the development of the concept of recession. Business cycles and recessions seem to be standard features of industrial economies. Non-seasonal cyclical variation is perhaps not completely unthinkable in agrarian or pre-industrial economies, but the concept of the business cycle has been applied almost exclusively to economies that have undergone an industrial transformation. Beyond any conceptual reasons for this focus, it may result from practical factors such as the relative abundance of economic data and economic analysts, making it possible to track cycles systematically using quantitative measures of economic activity.

          An early reference to business cycles comes from the observations of a perceptive London banker just before the dawn of the Victorian era. Samuel Jones Loyd (1837), First Baron Overstone, published a monograph entitled “Reflections suggested by a perusal of Mr. J. Horsley Palmer’s pamphlet on the causes and consequences of the pressure on the money market.”[2] The Mr. Palmer in question was a director of the Bank of England who had published a defense of actions that the Bank had taken with regard to money and credit.

          In modern parlance, the Bank at the time was acting both as a central bank – issuing currency – and as a commercial bank – accepting deposits from and extending loans to the private sector. Lord Overstone argued that these two functions should be kept separate because they would be in conflict over the course of a business cycle. In his own words, “The history of what we are in the habit of calling the ‘state of trade’ is an instructive lesson. We find it subject to various conditions which are periodically returning; it revolves apparently in an established cycle.” The specifics of the cycle would be less familiar to readers today. “First we find it in a state of quiescence, – next improvement, – growing confidence, – prosperity, – excitement, – overtrading, – convulsion, – pressure, – stagnation, – distress, – ending again in quiescence.” It would take more than a century for the current terminology to evolve, but the basic ingredients of the business cycle were already there.

          Toward the end of the century, economist David A. Wells (1890) was using the term “depression” to signify an extended economic contraction in industry, commerce, and other sectors of the economy, something like a modern recession.[3] For example, he used expressions such as “general commercial depression” and “general depression of trade and industry” to describe essentially what we would call a recession today. In one particular case, he applied the term to a serious economic contraction, “a very great depression in the pig-iron industry” in the United States starting in 1873. That episode would soon afterwards become known as “the Great Depression” until another one even greater surpassed it in the 1930s. The 1873 event is now commonly known as “the Long Depression” in the United Kingdom and the United States.[4]

          The direct line to the current approach to recessions starts with the publication by Wesley Clair Mitchell (1913) of a tome entitled “Business Cycles.”[5] Mitchell was a university professor and researcher looking for a way to bring the conceptual framework of a business cycle to the data and to develop empirical techniques to date specific business cycles. His underlying premise was that a business cycle occurs when many sectors of the economy are simultaneously either expanding or contracting. He reasoned that if such a pattern actually existed in the data, it would be possible to analyze fluctuations in individual sectors, compare results, and ultimately identify the periods during which changes in activity in many sectors coincided. The analyst would then be able to give specific dates for the various phases of the business cycle.

          Mitchell’s concept of the business cycle consisted of four phases, not as many as in Lord Overstone’s analysis, but certainly more than in the present approach. He labeled the four phases “prosperity, crisis, depression, and revival.” Of these, depression is most closely related to the modern concept of recession, as it was in the Wells analysis from 1890. In Chapter III of his book, Mitchell used “the annals of business, 1890-1911” in the United States, as well as in England, France, and Germany, to look for detailed evidence of co-movements in economic activity in each of the countries. The chapter concludes with a summary table that provides a rudimentary business cycle chronology for each country at an annual frequency, classifying periods into the four phases of the cycle that he had earlier laid out. Occasionally, the terms applied to the cyclical phases were qualified as to their intensity, for instance, as in “high tide of prosperity,” “deep depression,” or “mild depression.”

          Mitchell (1927) subsequently issued a substantially revised version of the book, this time entitled “Business Cycles: The Problem and Its Setting.”[6] A few years earlier in 1920, he had been one of the founders of the National Bureau of Economic Research (NBER) and was instrumental in making business cycle research a major focus for the new institution. The NBER, which itself published the new edition of the book, to this day remains the de facto official arbiter of business cycle dates for the United States.

          The empirical content of the 1927 edition was thoroughly revised from the 1913 version. The data sample was extended chronologically to 1925 and geographically to include 13 additional countries in Europe, North America, South America, Asia, and Africa. The conceptual framework remained basically the same, but Mitchell introduced the term “recession” to replace what he had previously identified as the “crisis” phase of the business cycle. He felt that the term “crisis” was imprecise and would have to be qualified each time it was used, and that “recession” was a better complement to the term “revival” at the opposite end of the cycle. In time, this usage would not stick. Mitchell’s 1927 term “recession” was more akin to the current concept of business cycle peak than it is to a modern recession.

          In the 1930s, Mitchell was joined at the NBER by Arthur Burns, one of his graduate students at Columbia University. For more than a decade, the two of them extended and updated the work on business cycles and by 1946 the NBER published their book “Measuring Business Cycles,” which was the culmination of their joint efforts in the field.[7] Although their research strategy was largely based on Mitchell’s earlier work, the results were presented in a novel way that included the first published statement of the NBER business cycle chronology organized in the form it is reported today.

          Instead of trying to identify all four phases of the cycle in their empirical analysis, Burns and Mitchell focused on reporting simply the high and low points of the cycle, which they labeled as “peaks” and “troughs.” In addition, they referred to the period between a trough and the next peak as an “expansion,” a label still currently in use, and to the period between a peak and the following trough as a “contraction,” which corresponds to what we know today as a recession. Following Mitchell’s early work, the 1946 book identified business cycles in France, Great Britain, and Germany as well as in the United States.

          The finishing touches in the development of the present concept of recession came in a 1958 monograph by NBER researcher Geoffrey Moore entitled “Measuring Recessions.” [8] The word “recession” appears in its modern acceptation in the title and also frequently in the text. However, in a historical nod to Burns and Mitchell, recessions are still called “contractions” in all of the tables and throughout most of the text. Terminology notwithstanding, the list of business cycle peaks and troughs given in the first table of the paper, covering the period from December 1854 to July 1957, matches the turning points currently identified by the NBER almost exactly. Only the last two dates in the table were later adjusted. In contrast to the earlier work of Mitchell and Burns, the Moore monograph focused exclusively on business cycles in the United States.

          In 1978, the NBER opted for a more systematic approach to the identification of U.S. business cycle peaks and troughs with the creation of its Business Cycle Dating Committee (BCDC). This group tracks multiple economic variables as values become available in real time and meets occasionally when circumstances suggest that a cyclical turning point is in the offing or may have occurred in the recent past. The timing of their public announcements strongly suggests that they prefer to identify turning points only when conclusions from the data are ineludible. In this way, they avoid having to revise their assessments later, especially since the data themselves are subject to subsequent revisions. Official public announcements of turning point dates usually take place about a year after the event.

          Initially, BCDC members were either NBER research economists or affiliated faculty members. For example, charter members of the committee in 1978 were Robert Hall (Stanford, BCDC chair), William Branson (Princeton), Martin Feldstein (Harvard), Benjamin Friedman (Harvard), Robert Gordon (Northwestern), Geoffrey Moore (NBER), and Victor Zarnowitz (NBER). More recently, the committee has been composed exclusively of affiliated faculty members. As of August 2022, they are Hall (chair), Gordon, James Poterba (MIT), Valerie Ramey (U.C. San Diego), Christina Romer (U.C. Berkeley), David Romer (U.C. Berkeley), James Stock (Harvard), and Mark Watson (Princeton).[9]

          Since the 1950s, the NBER has focused its business cycle dating activity exclusively on the United States. However, the NBER methodology may be applied in principle to any country for which sufficient data are available, as Mitchell and Burns showed in their research published between 1913 and 1946. A few public and private organizations have stepped forward to produce recession indicators for countries other than the United States. These indicators are carefully constructed, in some cases using methodology adapted from the NBER, and have become useful tools for economic researchers. It seems fair to say, however, that they are not yet perceived to be as authoritative and quasi-official as those of the NBER are for the United States.

          As an example, Geoffrey Moore himself founded the Center for International Business Cycle Research in 1979 at Rutgers University to apply the Mitchell-Burns-Moore methodology to economies other than the United States. The Center moved with Moore to Columbia University in 1983, where it continued to operate until 1996. At that time, Moore and some of his associates left the university to start a private firm, the Economic Cycle Research Institute (ECRI), which continued to operate after Moore’s passing in 2000 and currently reports business cycle dates going back to 1948. The data cover 22 countries, including the United States for which turning point dates coincide with the NBER’s.

          With greater economic and monetary integration in Europe, the London-based Centre for Economic Policy Research started in 2003 to produce a chronology of business cycles for the euro area, that is, for the aggregate economy of European Union member states that have replaced their national currencies with the euro. The chronology dates back to 1970. The CEPR-EABCN dating committee (now also associated with the Euro Area Business Cycle Network) employs a methodology based on the NBER’s but adapted for application to a geographical area that includes various national economies The Committee defines a recession as “a significant decline in the level of economic activity, spread across the economy of the euro area, usually visible in two or more consecutive quarters of negative growth in GDP, employment and other measures of aggregate economic activity for the euro area as a whole.”[10]

          Broader geographical coverage of business cycle turning points is provided by the Organization for Economic Co-operation and Development (OECD), which has a chronology of dates for each of the 34 OECD member countries as well as for 6 non-member countries and for various multi-country aggregates. Most of these chronologies start in 1960. In contrast to the NBER, ECRI, and CEPR, the OECD focuses on a single reference variable for the cycle. The reference variable was industrial production until April 2012, when it was replaced by real gross domestic product.

          The concepts of business cycles and recessions are ubiquitous in economic thinking about macroeconomic fluctuations. It may be possible to think about empirical macroeconomics in their absence, but the revealed preference of economists in academia, business, government, and finance since the nineteenth century has been to rely on some form of the business cycle framework for analysis of macroeconomic data, and in some cases for theoretical macroeconomic modeling. However, we have seen that the definition of recession has not achieved a uniform universal standard and that the term may not mean quite exactly the same to different economic analysts. For that reason, it seems important to take a systematic look at the economic literature on recession and to try to distill from that literature a consensus about the nature and consequences of recessions.

          The remainder of this essay considers each of the fundamental questions listed earlier and in the process connects the discussion to each of the 60 individual readings.

Readings referenced from book The Economics of Recession

R3    Arthur F. Burns and Wesley C. Mitchell (1946), ‘Working Plans’, in Measuring Business Cycles, Chapter 1, New York, NY, USA: National Bureau of Economic Research, 3–22

R4    Arthur F. Burns and Wesley C. Mitchell (1946), ‘Dating Specific and Business Cycles’, in Measuring Business Cycles, Chapter 4, New York, NY, USA: National Bureau of Economic Research, 56–114

R5    Geoffrey H. Moore (1958), ‘Measuring Recessions’, Journal of the American Statistical Association, 53 (282), June, 259–316