Show Summary Details

Page of

PRINTED FROM OXFORD HANDBOOKS ONLINE ( © Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Handbooks Online for personal use (for details see Privacy Policy and Legal Notice).

Subscriber: null; date: 18 January 2019

The Record of Economic Growth, Business Cycles, and Economic Policies in American Economic History

Abstract and Keywords

This chapter documents and analyzes the American historical economic record of growth and business cycles within the context of long-run and short-run economic policy changes. The United States is unique among all of the developed countries in terms of having a sustained and fairly stable record of economic growth. Given the large changes in various policies that have occurred over time, this record suggests that policy shifts have had almost no impact on the growth rate over the very long run. However, policy changes have had a significant impact on economic activity over shorter horizons, including the Great Depression and World War II. This chapter also argues that microeconomic policies, such as regulatory policies, tax policies, and labor policies, have had as much of an impact on aggregate economic activity as macroeconomic policies, such as monetary policy and government spending and transfer policies

Keywords: economic growth, microeconomic policy, macroeconomic policy, Great Depression, tax, labor, business

The impact of economic policies on America’s history of economic growth and fluctuations can be analyzed by general equilibrium macroeconomic models, often called dynamic stochastic general equilibrium (DSGE) models. This framework can be combined with traditional historical methods to analyze the US historical record, and how policies have affected that record.

This methodological development of combining the analytical forces of macroeconomics and economic history has created a productive interplay that has contributed to the development of both fields. Macroeconomic developments have brought a different methodology to economic history and cliometrics that is shedding new light on substantive historical questions, while macroeconomists are researching historical events integrating methods of economic historians. Exciting new progress is being made that integrates and exploits the specializations and skill sets offered by both disciplines.

Figure 28.1 shows per capita real gross domestic product (GDP) from 1889 to 2014. There are three distinctive features of these data. One is persistent and stable long-run economic growth. Over the post-1946 period, real GDP growth has averaged about 3.1 percent per year, with a standard deviation of about 2.4 percent. Note that real GDP growth has ranged from roughly zero to about 5 percent in fifty-four of the sixty-eight years since 1946. Decade-long averaged annual growth rates, calculated beginning in 1947 for each subsequent ten-year period, show decade-long growth rate averages of 3.7, 4.3 3.0, 3.4, 3.0, 3.3, and 1.2 percent per decade, in which the last statistic is calculated from 2006 to 2014. (p. 226)

The Record of Economic Growth, Business Cycles, and Economic Policies in American Economic HistoryClick to view larger

Figure 28.1. Per capita real GDP, 1889–2014

Sources: The data before 1929 are from Kendrick (1961) and Bureau of Economic Analysis, and the data from 1929–2014 are from the US National Income and Product Accounts, “Interactive Tables: GDP and the National Income and Product Account (NIPA) Historical Tables,”

The second feature is that there are fluctuations around trend, which are referred to as the business cycle. The third feature is that there are two historical episodes in which the deviation from trend is larger and more persistent: the Great Depression in the 1930s, in which real output declined substantially, and World War II in the 1940s, in which real output rose substantially.

It is common to divide economic policies into two groups: macroeconomic policies, such as monetary and fiscal policy, and microeconomic policies, such as regulatory policies and antitrust policies that directly impact specific industries or markets. An important theme is that both microeconomic and macroeconomic policies have had a significant impact on the United States’ macroeconomic record.

Macroeconomic Policies

There have been large changes in monetary and fiscal policy over time. In terms of monetary policy, both the inflation rate and the growth rate of money increased considerably following the founding of the Federal Reserve system in 1913. There was virtually no long-run inflation trend between 1789 through the 1930s.

Large inflations occurred around wartimes, including the Revolutionary War, the Civil War, World War I, and World War II. However, there was deflation after most of these wars, which ultimately returned the price level to about the level that prevailed before war. The longest period of deflation occurred after the Civil War, in (p. 227) which the price level fell by about 40 percent between 1869 and 1897. It is common today to associate deflation with depression, but this nearly thirty-year period of chronic deflation was accompanied by significant economic growth. The monetary base (currency plus bank reserves) doubled over this period, but real income and output nearly tripled. This is perhaps the leading example of long-run deflation resulting from monetary growth that is slower than real economic growth. In addition, there were two periods of severe deflation, which this author defines as an annual deflation rate of at least 10 percent that occurred during the early 1920s and during the early 1930s.

Monetary policy changed considerably after World War II. Between 1941 and 1947, the consumer price index (CPI) rose about 50 percent, and continued to rise afterward. Deflationary policy was not conducted after World War II, as was the case after earlier wars. Inflation became a chronic process after World War II, as the CPI grew at an average rate of 3.5 percent per year between 1947 and 2014. Inflation reached its peacetime high of over 10 percent per year by the late 1970s and early 1980s, which is a period known as the Great Inflation. The chronic post–World War II inflation has raised the CPI by a factor of nearly ten. Friedman and Schwartz (1963) and Meltzer (2009a, 2009b) provide detail on historical inflationary trends.

There are also large changes in fiscal policy that have occurred over time. Prior to the 1930s, government spending was a fairly small share of output, tax rates were low, and fiscal deficits were rare outside of wartime periods. Government spending (federal, state, and local), including transfer payments, was about 10 percent of GDP in 1900, whereas it was over 40 percent of GDP by 2010.

Tax rates also changed considerably. Barro and Redlick (2011) calculate average marginal tax rates based on federal income tax rates, state income tax rates, and social security tax rates. They find that average marginal tax rates were around zero in the early 1900s. Marginal tax rates rose substantially during World War II to about 27 percent, and continued to rise after that, increasing to around 40 percent by the early 1980s. Tax rates declined after that, but rose more recently. Today, marginal tax rates are close to 40 percent. The complexity of the tax code also has increased considerably. There are nearly 74,000 pages of tax rules in 2014, compared to about 500 in 1939.

The pattern of US fiscal deficits also changed substantially, and these changes are quite similar to changes in the historical pattern of inflation. Prior to the 1940s, wars were characterized by large deficits, and peacetime budgets were typically balanced, or were in surplus. Prior to World War I, the federal debt as a share of GDP was less than 5 percent, having declined from a level of about 40 percent at the end of the Civil War. Before that, debt as a share of GDP had declined following the Revolutionary War and the War of 1812. The pattern of large wartime debt continued during World War II, but deficits became fairly regular after that, just as inflation became fairly regular after this war. In the sixty-seven years between 1947 and 2014, a deficit was recorded in fifty-seven of those years, including a deficit of nearly 10 percent of GDP in 2009. Auerbach (2005) discusses the post–World War II fiscal history of the United States.

(p. 228) Microeconomic Policies

There also have been large changes in microeconomic policies. These policies have significant macroeconomic effects either directly, when the sectors involved are large, or indirectly, through general equilibrium effects, in which policies that directly affect one market impact other markets.

An important component of microeconomic policies includes changes in regulations, which are changes in the restrictions and/or requirements on what a buyer and/or a seller can do in a particular market. These include changes in financial regulation, environmental and land use regulation, consumer product and safety regulation, antitrust regulation, labor and immigration regulation, and transportation regulations.

One trend is that regulation of economic activity has increased considerably. While there is no canonical measure of the size or extent of regulation, some economists use the length of the Federal Register, which is the collection of federal agency rules and proposed rules, as a guide to changes in regulation. The Federal Register was first published in 1936, and contained around 2,600 pages at that time. Today, the Federal Register is around 80,000 pages, which suggests that regulation has increased considerably. Another trend is that regulatory expansion has broadly impacted many sectors, and some of these changes have plausibly impacted the aggregate economy, including financial regulation, antitrust and business regulation, and labor regulation.

In the financial sector, regulatory changes include the Federal Reserve Act (1913) which provided for the Federal Reserve system; the Glass-Steagall Act (1933), which separated commercial and investment banking functions; the Securities Act (1934), which created the Securities Exchange Commission; the Bank Act (1933), which provided for federal deposit insurance; the Federal Home Loan Bank Act (1932), which initiated federal intervention in housing markets; the Commodity Futures Trading Act (1974), which regulated futures trading; the Depository Institutions Deregulation and Monetary Control Act (1980), which allowed commercial banks to pay market interest rates; the Sarbanes-Oxley Act (2002), which created public accounting oversight; and Dodd-Frank (2010), which provided considerable new regulation of financial markets.

There have also been a number of major changes in business regulation over time. States were responsible for considerable regulation in the 1800s (see Fishback, Holmes, and Allen 2009) for a summary of state regulations. Changes in federal regulations include the Sherman and Clayton Antitrust Acts, which allow the federal government to break up private firms; the Federal Trade Commission, which was also aimed at preventing anticompetitive business practices’ the Interstate Commerce Commission, which regulated trade across state lines; and the Robinson-Patman Act, which restricts certain types of price discounts and which restricts price discrimination. Antitrust policy changed significantly in the early and mid-1930s with the National Industrial Recovery Act, which de facto stopped antitrust prosecution, and was followed by the resumption of antitrust in 1938. There were also significant Supreme Court rulings which provided interpretations of the antitrust laws, and that in many cases broadened or extended these regulations.

(p. 229) There have been enormous changes in labor regulation, including the Thirteenth Amendment, which prohibited slavery; the Railway Labor Act of 1926, which provided for independent unions in railroads, and which provided the foundation for broader forms of collective bargaining; the Norris-LaGuardia Act, which banned “yellow dog contracts” (a contract in which a worker agrees not to join a union) and prevented injunctions against labor during peaceful disputes; the Davis-Bacon Act, which provided for prevailing wages to be paid on government construction projects; the Fair Labor Standards Act, which provided for a forty-hour workweek, overtime pay, and minimum wages on jobs involving interstate commerce. The National Labor Relations Act (NLRA), also known as the Wagner Act, was a major change in labor policy that provided for collective bargaining and a governing board to enforce the Act. The Taft Hartley Act amended the NLRA to allow states to prevent union shops and allowed for “right to work states,” which allowed states to pass laws that prevent workers from being compelled to join a union or pay union dues. Other labor legislation includes the Civil Rights Act (1964), the Occupational Safety and Health Act (1970), and the Americans with Disabilities Act (1990).

These policy changes have coincided with relatively smooth long-run economic growth. This coincidence suggests that on net, policies have had relatively little quantitative impact on long-run economic growth. Stokey and Rebelo (1995) have made similar points about US fiscal policies and US long run growth, in that tax rates changed considerably over time, but long-run growth rates did not. Using a very different approach, Barro (1998) also finds relatively small effects of policy variables on growth in a cross section of countries. While this evidence suggests that policy changes have not had major effects on long-run growth, there are persistent deviations of economic activity from trend that are associated with large policy changes. The most striking of these episodes are the Great Depression and World War II.

Many policies changed substantially during these two periods, which represents a challenge for analyzing how these changes impacted the US economy. There were changes in monetary policy, in antitrust, in wage and price setting, in fiscal policies, and in regulatory policies. To assess which of these policy changes had the largest macroeconomic impact, this chapter applies some recent developments in macroeconomic methodologies. It provides some context for this approach by describing how these methods have more broadly become integrated in the study of macroeconomic history.

Policy Analysis and General Equilibrium Modeling in Macroeconomic History

The integration of the DSGE framework into economic history, which began in the 1990s, is similar to its integration into macroeconomics that occurred a decade earlier. Both adoptions of the DSGE approach were born from the need for alternative (p. 230) quantitative tools. The two quantitative frameworks for studying fluctuations in the 1960s and 1970s were based on the two competing macroeconomic paradigms of that time, Keynesian models and Monetarist models. The Keynesian models included the large-scale econometric models of Data Resources, Inc., Wharton Econometrics, and the Federal Reserve Board. Monetarist models included Anderson and Jordan (1968), which emphasized changes in monetary policy as impacting business cycles.

Both of these frameworks were falling out of favor in macroeconomics by the late 1970s. From a theoretical perspective, this reflected the fact that neither framework had sufficient theoretical foundations. From an empirical perspective, both frameworks were increasingly challenged to account for macroeconomic data, including the coincidence of high inflation and high unemployment of the 1970s, which was inconsistent with Keynesian models, and changes in the statistical relationship between money, inflation, and output, which was a challenge for monetarist models. Moreover, neither framework modeled expectations formation beyond the assumption of adaptive expectations, in which decision-makers predicted the future using only past values of the random variable. However, the rapidly growing rational expectations literature provided striking empirical examples in which agents formed expectations very differently from the adaptive approach (Sargent 1982). By the late 1970s, macroeconomists almost universally had adopted rational expectations modeling.

These theoretical and empirical challenges motivated the transition of macroeconomic modeling into the real business cycle (RBC) models of Kydland and Prescott (1982), which featured general equilibrium foundations and rational expectations. The benefits of the RBC/DSGE framework is that it features an explicit microeconomic framework that provides a consistent treatment of both growth and fluctuations. When aggregated, the framework provides model constructs that can be directly compared to National Income and Product Accounts (NIPA) data. Since the initial models of Kydland and Prescott, these models have been broadened and extended to include multiple shocks, including monetary shocks and fiscal policy shocks, heterogeneity, incomplete markets, nonconvexities, unemployment, demographic structures, learning, and other features (see Ohanian 2010).

A similar transition occurred in macroeconomic history. It became clear that alternative frameworks to the Keynesian and Monetarist models were required for addressing some historical macroeconomic questions. This was particularly relevant for the Great Depression. Some of the best-known studies of the Depression by economic historians did not rely on either the Keynesian or Monetarist frameworks. This included Romer’s (1990) research on uncertainty, and Bernanke’s (1983) research on financial intermediation. Neither the Keynesian nor the Monetarist models provided an adequate framework for analyzing the consequences of the destruction of banking information capital, as in Bernanke, or of uncertainty, which was the focus of the Romer study.

But the Keynesian and Monetarist frameworks faced broader and more basic challenges in assessing the Depression. These challenges are seen in the debate between Monetarist and Keynesian explanations of the Depression by Friedman and Schwartz (1963) and Temin (1976). Friedman and Schwartz argued that a large decline in broad (p. 231) measures of money contributed significantly to the Depression, and argued that the Depression would have been milder and shorter had the Fed prevented this decline. Temin argued that the money stock decline was not sufficiently large to generate the Great Depression, and instead proposed that an exogenous decline in spending was the key factor. But neither Friedman and Schwartz nor Temin provided a theoretical framework for advancing these ideas.

The Record of Economic Growth, Business Cycles, and Economic Policies in American Economic HistoryClick to view larger

Figure 28.2. Real consumption and hours worked per capita (Index, 1929 = 100)

Source: Kendrick (1961).

The RBC approach also faced challenges in understanding the Depression. This led Cole and Ohanian (1999, 2002) to use the RBC framework in a very different way. They applied a standard RBC model to identify the features of the Depression that were most strongly at odds with the model. They then used the deviations between the model predictions and data to develop hypotheses for understanding the Depression. Using deviations between model and data along these lines was essentially standing conventional growth theory on its head. The approach is now known as business cycle accounting, and is being used in a number of applications that intersect macroeconomics with economic history.

Before using this accounting framework, this section shows how these data, combined with market-clearing theory, can identify some important features of the Depression. Figure 28.2 shows the immediate declines of per capita consumption, which is measured relative to a 2 percent annual trend, and hours worked per capita. Friedman’s (1957) permanent income theory indicates that the large and immediate decline in consumption suggests that households perceived a large and immediate drop in their wealth. This further suggests that the economy was transitioning to a lower steady-state growth path. Moreover, permanent income reasoning indicates that households should also have reduced their demands for all normal goods, including leisure. However, per capita (p. 232) market hours of work declined substantially during the Depression, falling by 27 percent between 1929 and 1933, and recovering very little after that.

The Record of Economic Growth, Business Cycles, and Economic Policies in American Economic HistoryClick to view larger

Figure 28.3. Real consumption and hours worked per capita plus real wages (Index, 1929 = 100)

Source: Kendrick (1961).

Real wages are particularly puzzling during this period, given low consumption and low labor input. Figure 28.3 adds the real wage to the preceding figure, which shows that the wage is far above trend, despite the severe depression. Why didn’t the wage decline, given that consumption and hours worked were so low? These data suggest that a more accurate understanding of the depression results from identifying a quantitatively large and persistent shock that raised the market price of labor, and that depressed market hours of work and consumption.

Real Business Cycle Methods and Policy Analysis

The model for identifying the policies that have impacted the US economy uses a representative household and a representative business. The household chooses how to allocate its income, which consists of wage income and capital income, between consumption and investment. It also chooses how to allocate time between working in the market and other uses of time.

A representative firm operates a technology that combines labor and capital to produce output, decides how much labor and capital to hire in order to produce output and maximize profits. The technology grows on average over time, reflecting innovation and advances in knowledge. The model is constructed such that if the growth (p. 233) rate of technology is constant, and the markets for labor, output, and capital have no impediments, then output, consumption, and investment grow at a constant rate over time.

However, if technological growth is not constant, or if the markets for labor, output, and capital are distorted by regulations, taxes, information imperfections, transaction costs, or other factors that impede market function, then this economy will not grow at a constant rate. In contrast, it will deviate from smooth growth. We can observe these deviations from trend, but the specific factors that create these deviations will not be known. Therefore, Cole and Ohanian (1999, 2002), and Chari, Kehoe, and McGrattan (2007) have conducted research that shows how to use data and the equations of the model to measure the empirical deviations between the constant growth-rate predictions of the model and observations. These deviations are used as diagnostics for identifying what types of economic shocks or policy changes are promising candidates for understanding particular historical episodes.

There are four deviations from these first order conditions, which are also called wedges. These wedges are measured from the equations of the model that characterize the market clearing relationships in the markets for labor, investment goods, and output.

Given these specifications and historical data, the deviations in these equations can be measured directly. Note that if the deviations are zero in all periods, then the model variables would always be equal to their balanced growth path values. If any of these deviations are not zero, then either output, consumption, investment, and/or hours worked will diverge from respective balanced growth path values.

The wedges that are affected, and their relative size, are important for identifying the possible impact of policy changes. Cole and Ohanian (2015) and Chari, Kehoe and McGrattan (2007) provide a number examples that demonstrate how changes in labor market policies will arise as a wedge in the atemporal first-order condition governing time allocation. In fact, the labor wedge is identical to changes in a number of policies, including labor income tax rates, unionization, labor bargaining power, minimum wage rates, maximum wage rates, restrictions on working hours, and other restrictions on trading labor services. The labor wedge also is affected by changes in firm monopoly power, which in turn may result from changes in business regulation and antitrust policies. This is because changes in monopoly power impact the relationship between the marginal product of labor, which is the price of labor in a competitive economy, and the marginal revenue product of labor, which is the price of labor paid to workers by a firm with market power. As market power changes, the wedge between the marginal product of labor and the marginal revenue product of labor also changes.

In terms of the other wedges, Chari, Kehoe, and McGrattan (2007) show how changes in capital income tax rates, investment tax rates, investment tax credits, depreciation allowances and schedules, and financial market regulations and restrictions will arise as a wedge in the equation governing the allocation of income between savings and consumption. Changes in government spending will arise as a wedge in the resource constraint. Finally, changes in regulation and policies that affect innovation, research and development, patent protection, intangible investments in organizational capital, and (p. 234) other policies that impact efficiency and productivity growth will arise as a wedge in the production function.

There have been large, temporary changes in the labor wedge over time, particularly a very large increase in the wedge during the Great Depression, and a large reduction in the labor wedge during World War II. An increase in the labor wedge is equivalent to a higher labor income tax, and a reduction is equivalent to a lower labor income tax. However, there are no permanent shifts, nor are there trend movements in this wedge.

There are large changes in the efficiency wedge over time, particularly a decline in efficiency during the Depression, followed by rapidly rising efficiency in the late 1930s and during World War II. However, there are no permanent shifts in efficiency after that.

The facts that economic growth has been reasonably stable over time, and that there are no trend changes in any of the wedges, suggest that long-run economic growth has not been substantially affected by the many large changes in economic policies that have occurred. The next section analyzes the impact of large policy changes that occurred in the Great Depression and in World War II.

The Impact of Cartel Policies on the Great Depression

This accounting method can be applied to study the Depression. The labor wedge emerges in 1930, as consumption falls, hours worked falls, but the real wage rises. This wedge then rises considerably after 1933, reflecting further decreases in consumption and in hours worked, and rising wages. To put this pattern in context, the labor wedge increase is equivalent in this model to a labor income tax rate increase of about 30 percentage points.

This pattern of the labor wedge stands in contrast to the efficiency wedge. The efficiency wedge declines considerably during 1930 to 1933, as it falls about 15 percent below trend. It then rebounds and returns to trend by the mid-1930s. Chari, Kehoe, and McGrattan (2007) show that these patterns in the Solow Residual and the labor wedge can account for nearly all of the movements in output, consumption, and investment in the Depression. The lower Solow Residual and the labor wedge generate a large drop in economic activity between 1929 and 1933. After 1933, however, the recovery of the Solow Residual generates rising output, while the continued deepening of the labor wedge accounts exclusively for the continuation of depressed labor and investment.

These diagnostics led Cole and Ohanian (1999, 2002, 2004) to identify cartel policies as a leading factor in generating the large and persistent labor wedge that is evident in the 1930s. This author defines cartel policies as those that restrict competition in product and labor markets. These policies include major legislation, such as the National Industrial Recovery Act (NIRA) and the National Labor Relations Act (NLRA), as well as unlegislated policies, such as Hoover’s wage maintenance program of the early 1930s.

(p. 235) Hoover’s Wage-Maintenance and Job-Sharing Policies

To understand how policies generated this labor wedge and how these policies impacted the economy, Ohanian (2009) develops an RBC model with Herbert Hoover’s wage maintenance and job-sharing policies. This chapter blends institutional detail regarding competitive pressure in labor and product markets of that time with dynamic macroeconomic modeling.

Hoover held very different views about competition in product and labor markets than is common today. In his memoirs, Hoover (1951) thought that there was too much competition in the American economy in the 1920s and believed that industrial cooperation and “codes of fair competition” among businesses in the same industry would generate superior economic outcomes than if firms engaged in what he called “destructive competition.”

Hoover’s initiatives along these lines helped industry develop collusive trade groups that fostered high industrial concentration and substantial monopoly distortions during this period (see Kovacic and Shapiro 2000, about the expansion of industrial cartels in the 1920s and 1930s). Hoover’s views about wages also differ considerably from standard reasoning based on competitive markets. Most economists interpret real wages as being determined by worker productivity that reflects human capital, physical capital, and the level of technology. But Hoover believed that increasing wages in and of itself would promote prosperity, as he believed that high wages were synonymous with strong purchasing power, which in turn would stimulate more production.

Hoover’s views set the stage for meetings he held at the White House with leaders from major industries, primarily from manufacturing, in late 1929. He advised industry leaders not to cut wages. Hoover told industry that maintaining wage levels would minimize the severity of a downturn and help him “keep the peace with labor.” Hoover also advised firms to share work among employees rather than relying exclusively on layoffs. Hoover then asked labor leaders not to strike and to withdraw requests for higher wages. Following these meetings, industry publicly acknowledged their compliance with Hoover’s program as they held wage rates fixed and shared work among employees. And Hoover’s program did keep the industrial peace.

But Hoover discounted the impact of high wages on business hiring decisions. Nominal industrial wages declined very little during the early stages of the Depression, as industry kept its bargain with Hoover. Ohanian (2009) shows that wage survey data from the Conference Board indicate nominal wages declining only by about 5 percent. Prices, however, fell much faster, and real wages rose by about 10 percent by the fall of 1931.

The coincidence of high real wages and much lower employment suggests that wages were above competitive, market-clearing levels, and that the Hoover program was responsible. Simon (2000) presents evidence that wages were above market-clearing (p. 236) levels by analyzing newspaper “situation wanted” advertisements from the 1930s, which were advertisements taken out by job seekers. Simon found that the supply price of labor—the wage rate being requested by job seekers in their ads—was well below the wage rates that were being paid to observationally equivalent workers during this period. But before the Depression, Simon documents that there was very little difference between the wage being requested by job seekers and the wage being paid. Moreover, agricultural wages, which were not subject to Hoover’s wage maintenance program, were falling substantially. These data indicate that industrial wages were above their market-clearing level, and that they would have been lower in the absence of the Hoover policy.

Declining prices and productivity, coupled with Hoover’s program of fixing wages, significantly increased industrial labor costs. As the industrial decline intensified, industry leaders asked Hoover if he would support wage cuts proportional to the deflation that had occurred. But Hoover declined, despite increasing criticism from various quarters that his program was keeping wages far above their market-clearing levels. Many large firms kept their pledge with Hoover until the fall of 1931, at which point industrial hours worked had declined by nearly 40 percent.

Ohanian (2009) developed an RBC model tailored to capture Hoover’s wage-maintenance and job-sharing program, and quantified the macroeconomic impact of the Hoover program during 1930–1932. From the perspective of integrating macroeconomics with economic history, the key was introducing the Hoover wage-maintenance and job-sharing policy into an RBC model. The three major departures from the standard RBC model was creating a multisector model, since the Hoover program did not impact all sectors, imposing a binding minimum wage in the industrial sector, and distinguishing between employment and hours per worker in terms of labor input.

The model preserves the same preferences and technologies of the standard model, but differs by featuring two sectors, an industrial (manufacturing) sector that is impacted by Hoover’s program, and an agricultural sector, which is not. The output of these two sectors is aggregated into a final good used for both consumption and investment. The impact of Hoover’s policy is analyzed by feeding the observed real manufacturing wage into the industrial sector of the economy as a minimum wage. The industrial wage is exogenous from the perspective of firms, and this author interprets this wage as the coincidence of Hoover’s nominal wage-maintenance program combined with deflation. The model also includes lower productivity in the industrial sector, which further increases unit labor costs. Lower productivity is specified as an exogenously reduced workweek, reflecting Hoover’s job-sharing program. However, this could be modeled as another exogenous factor that depresses productivity and it would have the exact same effect on hiring decisions.

Ohanian found that Hoover’s program generated a significant depression, reducing model real GDP by about 18 percent by the end of 1931, which accounts for about two-thirds of the Depression at that time. The analysis is also consistent with the observed large asymmetry across sectors, as the industrial sector declined much more than the agricultural sector. This asymmetry in sectoral responses in the model reflects the fact (p. 237) that the Hoover program directly impacts the industrial sector, which has an indirect impact on the agricultural sector through general equilibrium effects.

The Expansion of Cartel Policies and the Continuation of the Depression

The Hoover program ultimately was abandoned by industry, but Hoover’s agenda of limiting competition in product and labor markets was broadened and deepened by President Roosevelt. Cole and Ohanian (2002, 2004) discuss how the National Industrial Recovery Act (NIRA 1933) was designed to limit competition and raise prices and wages in most of the nonfarm private economy. The NIRA accomplished this by permitting industry to cartelize, provided that industry immediately raise wages and agree to collective bargaining with workers. The Act covered over five hundred industries, representing about 80 percent of nonfarm private employment. Each industry formed a “code of fair competition” that was approved by the National Recovery Administration and that became the operating rules for each industry. The code was enforced by a code authority, which typically comprised members from the industry.

The NIRA codes included textbook features of cartels: minimum prices within industry, restrictions on expanding capacity and production limits, resale price maintenance, basing point pricing, and open price systems, which required that any firm planning to reduce price must preannounce the price cut to the code authority, who in turn would notify all other firms. The announcing firm was required to wait a specific period before changing its price. The purpose of this waiting period was for the code authority and other industry members to persuade the announcing firm to cancel its price cut. Codes of fair competition also included explicit provisions for profits. Some minimum prices explicitly factored in depreciation, rent, royalties, director’s fees, research and development expenses, amortization, patents, maintenance and repairs, and bad debts and profit margins as a percentage of cost.

The NIRA ended in May 1935, when the Supreme Court ruled in Schecter Poultry Corporation v. the United States that the NIRA was an unconstitutional delegation of legislative power. Roosevelt strongly opposed the Court’s decision, arguing that “[t]‌he fundamental purposes and principles of the NIRA are sound. To abandon them is unthinkable. It would spell the return to industrial and labor chaos” (Hawley 1966, 124). However, the Schecter decision did not end cartel policies. Policymakers continued cartel policies through new labor legislation and by largely ignoring the antitrust laws.

The main New Deal labor policy after the NIRA was the National Labor Relations Act, also known as the Wagner Act, which was passed on July 27, 1935. The NLRA gave workers the right to organize and bargain collectively. It prohibited firms from refusing to engage in collective bargaining, discriminating among employees for union affiliation, or forcing employees to join a company union. The act also established the (p. 238) National Labor Relations Board (NLRB) to enforce collective bargaining agreements and the rules of the NLRA.

The NLRA substantially expanded unionization and labor bargaining power. Union membership and strikes rose significantly under the NLRA. Union membership rose from about 13 percent of employment in 1935 to about 29 percent of employment by 1939, and strike activity similarly increased from 14 million strike days in 1936 to about 28 million in 1937. Moreover, the “sit-down” strike, in which workers forcibly occupied factories to stop production, gave workers considerable bargaining power.

The sit-down strike was the key factor in unionizing the auto and steel industries (see Kennedy 1999, 310–317). Sit-down strikes in General Motors (GM) auto body plants in late 1936 and early 1937 brought GM production to a virtual halt. There was no federal intervention to aid GM, despite support of GM’s position by Vice President John Nance Garner. General Motors ultimately recognized the United Auto Workers to end the strike. Later that year, US Steel recognized the Steel Worker’s Organizing Committee to avoid a sit-down strike.

In terms of industrial policies, Hawley (1966) describes how monopoly policies continued after 1935, as government tacitly permitted collusion, particularly in industries that paid high wages. He cites a number of Federal Trade Commission studies documenting price fixing and production limits in a number of industries after 1935. Post-NIRA collusion was facilitated by trade practices formed under the NIRA, including basing point pricing. Cole and Ohanian (2004) document that Interior Secretary Harold Ickes complained to Roosevelt that he received identical bids from steel firms on 257 different occasions (Hawley 1966) between June 1935 and May 1936. The bids were not only identical, but also 50 percent higher than foreign steel prices (Ickes 1953–1954).

This price difference was sufficiently large that Ickes ordered the steel from German suppliers. Roosevelt canceled the German contract, however, after pressure from both the steel trade association and the steel labor representatives. And despite wide-scale collusion, the attorney general publicly announced that steel producers would not be prosecuted. Hawley documents that the steel case was just one example of the failure of the government to enforce the antitrust laws. The number of antitrust cases initiated fell from an average of 12.5 new cases per year during the 1920s, which was in itself a period of limited enforcement, to an average of 6.5 cases per year during 1935–1938. Moreover, several of these new cases were initiated in order to prosecute racketeering.

The Impact of New Deal Policies on Wages and Prices

Cole and Ohanian (2004) present evidence on relative prices and real wages that suggest that New Deal cartel policies had significant impact. Tables 28.1 and 28.2 compare wage (p. 239) and price statistics between industries cartelized by New Deal policies and those that were not. Table 28.1 shows annual industry-level data for wages in three sectors covered by New Deal policies—manufacturing, bituminous coal, and petroleum products—and two sectors not covered—anthracite coal and all farm products. The farm sector was not covered by these industrial and labor policies, while anthracite coal was a de facto uncovered sector. Specifically, anthracite coal was to have been under the umbrella of the NIRA, but the industry and its coal miners failed to reach an agreement, and the industry never wrote a code of fair competition.

Table 28.1 Indexed Real Wages Relative to Trend























Bituminous coal











Anthracite coal





























Notes: Wages are deflated by the GNP deflator and a 1.4 percent trend, which is the growth rate of manufacturing compensation in the postwar period. They are indexed to be 100 in 1929, except for the wages in anthracite and petroleum, which are indexed to 1932 = 100 because of data availability.

Real wages in the three covered sectors rose after the NIRA was adopted and remained high through the rest of the decade. Manufacturing, bituminous coal, and petroleum wages were between 24 and 33 percent above trend in 1939. In contrast, the farm wage was 31 percent below trend, and anthracite coal was 6 percent below trend. It is striking that the bituminous coal miners—who successfully negotiated under the NIRA—increased their wages significantly, while anthracite coal miners—who did not successfully negotiate under the NIRA—were unable to raise their wages.

Table 28.2 shows monthly Conference Board data (see Hanes 1996), and indicates that all these industry wages significantly increased. The table shows real wages in eleven manufacturing industries in which there is also monthly price data. Table 28.2 shows significant increases in all eleven industries occurring after the NIRA was passed. In this table, the real wage is indexed to 100 in February 1933 (which is a few months prior to the passage of the NIRA) to focus on the effect of the adoption of the policies on real wages. All these industry wages were significantly higher at the end of 1933, six months after the act was passed. The smallest increase is 7 percent (farm implements), and the largest increase is 46 percent (boots and shoes). These wages remained high through the end of the NIRA (May 1935) and also after the Schecter decision ruled that the NIRA was unconstitutional. The average real wage increase across these eleven categories in June 1936 relative to February 1933 was 25.4 percent.

This section now turns to the impact of the New Deal on prices. By fostering cartels and collusion, New Deal policies also increased relative prices in several industries for (p. 240) which data is available from the Bureau of Labor Statistics. Where possible, this author has matched industries in which both wage and price data are available. Table 28.3 shows relative price data from several industries covered before the NIRA was passed and continuing through the 1930s after the Schecter decision. Prices for nearly all industries that were effectively covered by these policies rose shortly after the NIRA was passed, and remained high through the decade. As in the case of real wages, the table contrasts the two coal industries, as bituminous coal negotiated a code of fair competition under the NIRA, but anthracite coal did not. Note that the relative price of bituminous coal rose after the NIRA was passed and remained high through 1939. In contrast, the relative price of anthracite coal was unchanged after the NIRA was passed and then declined moderately over the rest of the decade.

Table 28.2 Monthly Wages Relative to the GNP Deflator (February 1933 = 100)










Leather tanning





Boots and shoes















Foundries and machine shops




















Rubber manufacturing










Farm implements





Note and source: The table shows monthly Conference Board data (see Hanes 1996), and indicates that all these industry wages significantly increased.

These wage and price data provide substantial evidence that New Deal policies significantly raised relative prices and real wages in the industries covered by these policies, while prices and wages did not rise in industries that either were not covered by these policies or in which the industry was unable to negotiate and collude. There is additional evidence, including studies by the National Recovery Review Board (NRRB), which was an independent agency charged with assessing whether the NIRA was creating monopoly. The NRRB produced three studies that assessed industries covering about 50 percent of NIRA employment, and concluded that there was substantial evidence of monopoly in most industries. The FTC studied a number of manufacturing industries following the Schecter decision and concluded that there was very little competition in many concentrated industries after the NIRA. (p. 241)

Table 28.3 Wholesale Prices Relative to the Personal Consumption Services Deflator (February, 1933 = 100 for all variables)

























All home furnishings





Anthracite coal





Bituminous coal





Petroleum products




















Nonferrous metals





Structural steel





All metal products









. . .






Auto tires










Farm equipment





All building materials










Source: Cole and Ohanian (2004).

(*) The average does not include rubber.

The Impact of New Deal Policies on Output and Labor

Cole and Ohanian (2004) developed a macroeconomic model that integrated a dynamic model of firm-union wage bargaining within a multisector RBC model. The bargaining model is unique in terms of capturing the main features of the NIRA and the NLRA, both of which fostered limiting competition. The model’s foundation is a multisector RBC model to capture the fact that not all of the economy was impacted by these policies. There is a single final good that is used for consumption and investment. There is an industrial (cartelized) sector, and a nonindustrial (competitive) sector, with (p. 242) many symmetric industries in each sector. Production technologies are constant returns to scale.

The model features a dynamic bargaining game between labor and capital that is similar to the actual bargaining process that occurred during the New Deal. There is an insider-outsider friction, as in Blanchard and Summers (1987) and Lindbeck and Snower (1988), in which incumbent workers in the industries that are covered by the cartel policies (insiders) receive higher wages in the form of industry rents than those in the competitive sector. However, the bargaining game developed in Cole and Ohanian advances insider-outsider modeling in that the insiders endogenously choose the size of their group. Thus, the model can quantitatively address the impact of these policies on employment within an optimizing framework.

There is a representative family household, in which family members work in different industries. Thus even though the cartel and noncartel sectors pay different wages, household members pool incomes so that each member has the same consumption level. In addition to working in a high-wage cartel industry, working in the competitive sector, or spending time outside market activities, the model allows household members to search for a high-wage cartel job. This allows the model to capture unemployment, as well as provide a model in which the rents associated with the high-wage industries are competed away.

Workers and firms within an industry bargain each period over the wage and the number of workers that will be hired. The bargaining is conducted within a dynamic game in which workers make a take-it-or-leave-it offer. If the firms accept, then the government permits the industry to collude and act as a monopolist, which is consistent with the NIRA in which government approved a code of fair competition following a successful agreement with industry workers. If the firms do not accept the worker’s offer, then the firm pays the wage that prevails in the noncartel sector. With some probability, the government discovers that a wage agreement was not reached, and forces the industry to operate competitively. This latter feature captures the fact that New Deal cartelization was sanctioned only if industry reached an agreement with their workers.

Thus, workers must make an offer that industry weakly prefers to rejection. Worker bargaining power in the model thus depends on the probability that a firm that rejects a wage agreement is discovered by the government and loses its monopoly profits. Thus, workers will offer a wage and employment agreement that delivers a level of profits that is equal to expected profits if the firm rejected the offer and hired labor at the competitive wage. If the probability of the firm being discovered not adhering to a wage agreement is high, then the workers have considerable bargaining power. If this probability is low, then the firm has most of the bargaining power.

They most important parameters in this model are the fraction of industries that are cartelized and the bargaining power parameter. Cole and Ohanian assume that about one-third of the economy is cartelized, which represents the share of the economy at that time devoted to manufacturing, petroleum, and mining, and which is a conservative measure (Hawley 1966). They choose the value for the bargaining power (p. 243) parameter so that the cartel wage in the steady state of the model equals the observed real manufacturing wage in 1939, which is 17 percent above trend.

Cartel policies impacted employment and output substantially. Cole and Ohanian (2004) find that steady state output, consumption, and hours are all about 16 percent below trend. This accounts for about two-thirds of the extent that these actual variables remained below trend following the Depressions trough in 1933.

The End of New Deal Industrial and Labor Policies

By the late 1930s, Roosevelt acknowledged the impact of cartelization on the economy. Hawley (1966) reproduces part of a speech by Roosevelt, in which Roosevelt states that “. . . the American economy has become a concealed cartel system. The disappearance of price competition is one of the primary causes of present difficulties” (quoted in Hawley 1966, 412).

New Deal labor and industrial policies began to change around this time, and continued to evolve in the 1940s and during World War II. Thurman Arnold, who had the reputation of a “trust-buster,” was appointed as assistant attorney general and ran the antitrust division at the Department of Justice (DOJ). Arnold doubled the antitrust workforce in the DOJ and initiated a number of antitrust cases.

Labor policies changed considerably beginning in the late 1930s. The Supreme Court ruled against sit-down strikes in 1939. Labor policy continued to change during World War II, when the National War Labor Board (NWLB) refused to approve a large wage increase for Bethlehem Steel’s workers. From that point on, the NWLB’s “Little Steel” decision severely impacted bargaining power, as only cost-of-living wage increases were approved. Union strikes in response to the Little Steel decision pushed public opinion against unions, and the Taft-Hartley Act, which revised the National Labor Relations Act by reducing worker bargaining power, was passed in 1947. Postwar wage premia, which in this model is the major characteristic of union bargaining power, never again approached the levels of the late 1930s. Future research should be directed toward analyzing how this reversal of cartel policies contributed to the sharp expansion of the US economy in the late 1930s and early 1940s.

The 1930s is arguably the decade with the largest changes in cartelization in the history of the United States. Both unlegislated policy shifts by Hoover, as well as legislated policy changes in the New Deal, substantially changed the level of competition that prevailed in industrial product and labor markets at that time. Substantively, the research summarized here indicates that by impeding the normal competitive forces of supply and demand, these policy changes contributed to both the severity and the duration of the US Great Depression. Methodologically, this research shows how to model changes in policies that impact competition in product and labor markets through (p. 244) features such as excess labor supply, job rationing, and endogenous insider-outsider labor markets into an otherwise standard macroeconomic model.

The Great Depression and Other Policy Changes

Real business cycle methods have been used to study other aspects of the Great Depression. Christiano, Motto, and Rostagno (2003) analyze shifts in the money multiplier and how shifts in demand for liquidity and Federal Reserve policies impacted the Depression through wage rigidity, which is related to the cartel policies analyzed in Ohanian and Cole (2004) and Ohanian (2009). McGrattan (2012) studied how changes in capital income taxation in an RBC model, including the excess profits tax of 1937, substantially depressed business fixed investment. Crucini and Kahn (1996) studied the impact of tariff hikes, including their interaction with deflation, on the Great Depression.

Economic Policies and the World War II Economy

World War II represents the other large deviation of US economic activity from trend, and it is also a period of many policy changes. These policy shifts include higher government spending, military conscription, higher labor and capital income tax rates, wage and price controls, a large expansion of federal deficits, monetary expansion, and a larger role of government in the allocation of resources across industries. There has been considerable research in this area that has studied how these policy shifts impacted the World War II economy.

The size of some of these policy changes are unprecedented. Real government spending rose by more than a factor of six between 1940 and 1944. This increase in government spending was sufficiently large that real government spending in 1944 was nearly as large as real GDP in 1940. Conscription increased the size of military personnel from about 500,000 in 1940 to about sixteen million by 1944. The average marginal tax rate on labor income rose from about 9 percent to about 18 percent, and the average marginal tax rate on capital income rose from about 43 percent to over 60 percent. Factories and industrial capacity were converted from peacetime production, such as autos, to military production, such as jeeps, tanks, and warships, and the federal government invested substantially in the economy’s capital stock. These changes provide a unique opportunity to study how a variety of large policy shifts affected the US economy.

(p. 245) The Keynesian approach, which focuses almost exclusively on demand factors, was the traditional methodology for studying wartime economies. This approach was considered successful in understanding the wartime economy, because of the coincidence of higher government spending and higher output. However, the exclusive focus of the Keynesian model on demand factors led to some important inaccurate predictions. Specifically, the Keynesian model led some economists to predict a significant postwar depression when government spending returned to normal and millions of soldiers re-entered the civilian labor force. Samuelson (1943), in congressional testimony, urged policymakers to develop plans to deal with high postwar unemployment.

Samuelson’s fears regarding a weak postwar economy, however, were not realized. Real output, which was well above trend during the war, did decline, but remained near trend afterward, and unemployment remained under 4 percent. The post–World War II economy, despite the sharp decline in government spending, did not return to anything at all like the 1930s.

The RBC model provides an alternative approach to analyzing wartime economies that successfully captures much of the wartime and postwar changes in the economy. Applying business-cycle diagnostics to this episode identifies three key wedges: a large wedge in the economy’s resource constraint, which reflects the large increase in government spending; an efficiency wedge, which reflects rising total factor productivity; and a changing labor wedge. In terms of the labor wedge, key policies include a doubling of the average labor income tax rate, and a reversal of some New Deal unionization policies.

Early RBC studies of World War II began with Wynne (1990), Braun and McGrattan (1993) and Ohanian (1993, 1997). The Wynne and Braun-McGrattan papers were positive studies that focused on the ability of RBC to fit the data in response to government spending (Wynne 1990), and government spending and public investment (Braun-McGrattan 1993). Ohanian (1997) studied both positive and normative aspects of model fit in response to spending and tax policy changes, and also studied normative aspects of policies by conducting counterfactuals for World War II and the Korean War.

Wynne’s paper was motivated by earlier work that had analyzed the impact of government spending in a partial equilibrium environment, and that focused on the income and substitution effects of government spending. The substitution effects of government spending represent the economic effect of the temporary resource drain of government spending. This temporary reduction in output motivates households to supply more labor because of the relative scarcity of output. The income effect of government spending reflects the permanent impact of spending, as a large and long-lasting war reduces household wealth. This income effect also motivates households to supply more labor.

Wynne found that a standard RBC model can plausibly generate the World War II economic expansion in response to higher government spending, but for very different reasons than in the Keynesian model. Both the income and substitution effects from higher government expenditures were important factors in expanding economic activity in World War II. The very large resource drain of the war led households to supply more labor through the channel of intertemporal substitution, and the fact that war (p. 246) expenditures were so large and lasted sufficiently long meant that household permanent income was reduced. This led to an additional increase in labor supply. This paper demonstrated clearly that a market-clearing model naturally delivers higher output in response to fiscal policy.

Braun and McGrattan (1993) also studied whether an RBC model could account for wartime economic activity, and focused on the large expansion of labor supply and the fact that wage rates did not fall, which seems puzzling from the perspective of diminishing labor productivity. Their analysis focused on government-owned, privately operated capital investment. These government investments substituted for private investment during the war as the government funded substantial expansion in plant and equipment, particularly in manufacturing. Their study incorporates considerable historical detail about this process. They found that introducing this policy change into a standard RBC shed light on why total factor productivity—measured using just private capital—increased so much, why US wage rates did not decline, and why US labor supply expanded so much. These issues, raised by Braun and McGrattan, had not been studied in the context of the Keynesian model, presumably because of the focus on the supply side of the economy, which is absent from the Keynesian model.

Ohanian (1997) also analyzed the fit of RBC models applied to World War II as well as the Korean War. This research was the first to analyze the impact of labor and capital income taxes, as well as conscription during wartimes. Ohanian found a close fit to the data for both World War II and the Korean War. In particular, the model economy generates a smaller economic expansion during the Korean War, compared to World War II, as a consequence of a smaller increase in government spending and higher tax rates.

Ohanian also considered the normative question of the welfare costs of alternative war finance packages. This normative question was motivated by the fact that many US wars, including World War II, featured a tax-smoothing policy that used considerable debt to finance wartime expenditures, and then gradually retired the debt after the war. However, the Korean War deviated considerably from tax smoothing, as President Truman was committed to running a balanced budget, which included raising capital income tax rates to an all-time high. Ohanian found that welfare would have been higher had Truman and Congress continued with the typical tax-smoothing policy of previous wars.

McGrattan and Ohanian (2010) conduct the most detailed assessment of economic policy changes on the World War II economy from a positive perspective. They analyze the impact of a number of policy changes between 1941 and 1947, including changes in government spending, which was divided between government consumption, wage payments to military personnel, and investment. Their model also includes capital and labor income taxes, the draft, conscription, and productivity changes. In addition, the paper includes considerable detail about the post–World War II economy in order to assess how expectations about the postwar economy impacted wartime economic activity.

(p. 247) The McGrattan-Ohanian model provides the closest fit to World War II of any model studied. Output, consumption, investment, labor, wages, and the return to investment in the model are close to the actual data. They also decompose the model movements in labor into components due to each of the shocks. Table 28.4 shows this decomposition. The table shows that the most important shock impacting labor is the large increase in military purchases. In the absence of all other shocks, labor would have increased by 29 percent on average between 1939 and 1945 as a result of the massive increase in government spending. This increase in labor largely is driven by the very large resource drain of government spending. In terms of the other shocks, the draft, in which sixteen million were in military service by 1944, decreases labor by about 3 percent by reducing the supply of potential workers. The increase in income taxes decreases labor by about 11 percent. Overall, the model generates about a 10 percent average increase in labor over this period, compared to an actual 7 percent increase in the data.

Historical Business Cycles and Economic Policy

This section discusses some historical business cycles and the policy shifts that occurred around those times. These episodes do not involve changes in economic activity nearly as large as those in the Great Depression or World War II, but they are widely discussed in the literature and coincide with some important policy changes.

The 1907–1908 recession is well-known because it was large and because it occurred around the time of the Panic of 1907. It is common to connect the recession with the panic, given that they occurred around the same time. Odell and Weidenmier (2002) argue that this panic had its roots in the April 1906 San Francisco earthquake, which led to considerable insurance claims being paid from English insurers. They argue that this led to contractionary monetary policy by the Bank of England, which in turn contributed to the US panic.

The earthquake certainly devastated San Francisco, and it was accompanied by an immediate national decline in real output in the spring of 1906. The Miron-Romer (1990) index of US industrial production fell by about 3 percent in April. The index remained at roughly that level through much of the summer before recovering back to its previous level by August 1906, with production growing after that. Thus, the initial impact of the earthquake on the economy appeared to have passed well before the 1907 recession and panic.

The National Bureau of Economic Research (NBER) dates the 1907–1908 depression from May 1907 to June 1908, but the Miron-Romer index did not begin to turn down until October 1907, which was roughly around the beginning of the panic. Annual real GNP dropped by about 8 percent between 1907 and 1908, labor input dropped by more than 3 percent, and labor productivity dropped by more than 4 percent. (p. 248)

Table 28.4 Nonmilitary Hours During World War II

Model Predicted Hours Worked for the Following Scenarios


Government Spending Only

Add Technology


Add Military


Add Distortionary




















































Source: McGrattan and Ohanian (2010).

(*) All series are relative to average US nonmilitary hours for 1946–1960.

The coincidence in the timing of the two events suggests the possibility that the panic either caused the recession, or that it magnified and prolonged the recession. But it is challenging to quantify the impact of the panic on the broader economy, as the panic was over quickly, the panic did not appear to affect interest rates significantly, and other factors were present at that time that may have contributed to the downturn.

Most accounts date the panic as lasting about three weeks, occurring from roughly the middle of October 1907 to early November. In terms of the impact of the panic on credit markets, there was no substantial increase in either commercial paper rates or long-term interest rates around that time. Commercial paper rates in fact declined through late 1908, and real long-term interest rates rose by only about twenty basis points.

The stock market was another factor that may be relevant for understanding the recession. The real value of the stock market declined by 40 percent between late 1906 and October 1907. This may have reflected negative anticipations about future earnings, which in turn would have implications for the real economy in terms of anticipations about the profitability of future capital investment and the incentive to invest. To see this, note that real earnings had changed little from 1901 until early 1905. Earnings then began to rise, increasing by about 30 percent between early 1905 and late 1906. Share prices rose about 30 percent between mid-1904 and the fall of 1906. The 30 percent increase in share prices was sufficiently large as to suggest expectations for considerable and long-lived future earnings growth. However, earnings growth did not continue. Earnings peaked in September 2006, which was also the peak in share prices. Business fixed investment peaked in 1907, and then declined by more than 15 percent in 1908. With the exception of 1913, business fixed investment remained below its 1907 level until 1922.

(p. 249) Viewed from this perspective, the recession of 1907–1908 may have reflected a revaluation of the growth potential of the US economy and the profitability of capital investment. If investors lowered their expectations of future returns from investment, this in turn would reduce the incentive to invest in capital, which would reduce labor productivity and labor demand. While more research is needed to understand the 1907–1908 recession and the panic, this episode did lead to a substantial policy change with the Aldrich-Vreeland Act, which established the National Monetary Commission, and ultimately led to the Federal Reserve Act of 1913.

The 1919–1921 recession is striking in terms of changes in the composition of output, policy changes, and the speed of recovery. There were two major policy changes around the time of the 1920–1921 recession, both of which were related to World War I. One was a large decline in government spending. Even though World War I had ended much earlier, government spending remained high after the war. Between 1919 and 1921, real GNP declined by about $26 billion (1929 dollars), which represents about a 3.5 percent decline. However, government expenditures declined by more than the total drop in real GNP, falling by about $32 billion (in 1929 dollars), which represents more than a 30 percent drop in government spending. Thus, this was a recession in which private spending actually rose.

The second major policy shift was in monetary policy. The money supply and the price level had risen considerably during World War I. In 1920, the monetary base fell 9 percent, which is the largest annual decline in the history of the United States. The price level dropped by about 25 percent between 1919 and 1921, including a 15 percent deflation between 1919 and 1920. The cumulative deflation that occurred during 1919–1921 is comparable to the cumulative deflation of the Great Depression, though it took four years (1929–1933) during the Great Depression to accumulate a 25 percent drop in the price level.

Despite the fact that both of these episodes had the same cumulative deflation, the paths of output during these episodes are remarkably different. The peak-to-trough drop in real output during the Great Depression is about ten times larger than that during the 1919–1921 recession. The recoveries are also very different. While hours worked in the Great Depression recovered very little, the recovery following the 1919–1921 recession was very rapid. Real GNP was back to trend by 1922, and was 8 percent above trend in 1923.

By highlighting the substantial differences in downturn and recovery from 1929 to 1933, the 1919–1921 episode provides indirect evidence that cartel policies contributed to both the depth and the duration of the Great Depression. Hoover’s nominal wage maintenance policy provides an explanation for why deflation appears to be much more depressing in the early 1930s than in the early 1920s, and Roosevelt’s expansion of cartel policies in 1933 provides an explanation for why the economy did not recover after deflation ended.

Another implication follows from the fact that the 1919–1921 recession may partially reflect the shift of resources from military use to peacetime use. If the recession was partially the natural consequence of a return from a wartime to a peacetime economy, then (p. 250) the 1919–1921 episode suggests that a large deflation and high real interest rates, in the absence of other factors, such as nominal wage and/or price fixing, may not be nearly as depressing as widely believed.

Sometimes called the “Recession in the Depression,” the downturn of 1937–1938 is viewed by some economists as reflecting contractionary monetary policy and/or fiscal policy. Friedman and Schwartz argued that three increases in commercial bank reserve requirements were the key factor. In August 1936, the Federal Reserve increased the required fraction of deposits that member banks must hold as reserves from 10 percent to 15 percent. This rose to 17.5 percent in March 1937 and then rose to 20 percent in May 1937. These changes were implemented because banks were holding large amounts of excess reserves at that time.

However, this factor faces challenges as an explanation of the 1937–1938 episode regarding the timing of the changes in reserve requirements, and in the pattern of interest rates. The economy expanded around the time of these changes, as industrial production rose by about 12 percent between August 1936 and August 1937. The recession did not start until September 1937, which is fourteen months after the first and the largest increase. Moreover, interest rates did not rise after these policy changes. Commercial loan rates fell from 2.74 percent in January 1936 to 2.65 percent in August 1936. These rates then fell to 2.57 percent in March 1937 and rose slightly to 2.64 percent in May 1937, the date of the last increase in reserve requirements. Lending rates ranged between 2.48 percent and 2.60 percent over the rest of 1937 and through 1938. Interest rates on other securities showed similar patterns: rates on AAA-rated, AA-rated, and A-rated corporate debt were roughly unchanged. If higher reserve requirements depressed the economy, then presumably interest rates would have increased as a consequence of a decrease in the supply of loanable funds. (Interest rate data are from Banking and Monetary Statistics, 1914–1941, of the Board of Governors of the Federal Reserve System.)

Other economists have pointed to a decline in government spending, including Romer (1993), as a key factor in the downturn. This factor faces challenges in terms of timing, and also in terms of the size of the decrease in government spending relative to the size of the decrease in economic activity. In terms of timing, government spending declined in 1937, but grew considerably in 1938. However, the decline in real GDP did not occur until 1938. Thus, government spending would need to have the quantitative feature that only lagged spending impacts the economy, and not current spending. This author is unaware of any estimates of the impact of spending on the economy that have this property.

In terms of the size of the decline, government spending declined by $0.6 billion (1937 dollars) in 1937, and then rose by $1.0 billion in 1938. Real GDP declined by $4.1 billion (1937 dollars) in 1938. Abstracting from the issue of timing, this would require a government spending “multiplier” of about seven, which is well above standard estimates (see Ramey 2011 and Ohanian 1997).

An alternative theory for the 1937–1938 decline is based on an expansion of unionization policies. In April 1937, the Supreme Court upheld the National Labor Relations Act in Jones v. Laughlin Steel Corporation. This led to a large increase in strike activity (p. 251) and unionization. The unionization rate increased from around 13 percent of the private workforce to about 27 percent between 1936 and 1938. The sit-down strike, in which workers forcibly occupied factories in order to stop production, was used to unionize GM in 1937, and the threat of a sit-down strike was used to unionize US Steel. Real manufacturing wage rates rose by about 10 percent.

The manufacturing sector, which experienced these large wage increases, declined substantially in this period. Industrial production dropped by about 33 percent. Rapid productivity growth in 1939 and 1940 reduced unit labor costs, and industrial production began to recover at that time. More research is needed on this episode, but these statistics suggest that wage setting and unionization policies played a significant role in depressing the economy at this time.

The Recessions of the 1970s and Early 1980s

The recessions of 1973–1975, 1980–1981, and 1981–1982 were large, and also were associated with large policy changes. These recessions coincided with higher oil prices, slow productivity growth, and monetary policy responses to those oil price increases. Several economists note that the Federal Reserve tended to vacillate between reducing inflation and attempting to increase employment and output during this period. Kilian (2014) summarizes recent research in this area. Kilian argues that the Fed was attempting to reduce inflation in the early 1970s, but then began to pursue expansionary monetary policy to try to raise employment and output as economic activity weakened. Romer and Romer (1990) argue that the Fed tried to reduce inflation beginning around April 1974, then again in late 1978 and late 1979. The growth rate of money during the 1970s was substantial, as the monetary base roughly doubled between 1973 and 1981. Not surprisingly, inflation was high in the 1970s and early 1980s, and the price level more than doubled between 1973 and 1981.

These recessions were all large, which partially reflects the fact that trend productivity growth slowed around this time. Specifically, total factor productivity grew only at a 0.5 percent annual rate during this period. Moreover, the Federal Reserve initiated a large shift in monetary policy beginning in October 1979, designed to reduce inflation, which had increased to a peacetime high of around 13 percent. Long-term interest rates, including mortgage rates, were as high as 18 percent, and the Fed funds rate was as high as 20 percent. Contractionary monetary policy is often cited as a key factor in the 1980–1981 and the 1981–1982 recessions.

These recessions would have been even more severe, however, if the typical recession pattern of a growing labor wedge had been present. Ohanian (2014) and Chari, Kehoe, and McGrattan (2007) describe that the labor wedge actually narrowed, beginning around the mid-1970s. This narrowing, rather than a widening of the labor wedge, (p. 252) which is typically observed during a weak economy, likely reflects a trend increase in women’s labor supply, as the share of women working in the market rose from around 40 percent in the early 1970s to around 58 percent by 2000. Ohanian (2014) finds that real GDP per capita would have been 15 percent below trend by 1981 in the absence of the narrowing of the labor wedge.

The Great Moderation, the Great Recession, and Its Aftermath

The period following the 1981–1982 recession up to 2007 is known as the “Great Moderation.” This period is characterized by smooth aggregate economic growth, with just two recessions, one in 1990–1991 and another in 2000–2001, both of which were comparatively mild. The Great Moderation ended with the Great Recession. This recession, which NBER dates from December 2007 to July 2009, was the most severe recession the United States had experienced since the 1981–1982 recession. Ohanian (2010) describes how the US Great Recession differed considerably from previous US recessions, as well as the Great Recession in other countries. The drop in US output was almost entirely due to lower labor input, rather than lower productivity. In contrast, lower outputs in the other G-7 countries were primarily due to lower productivity. This means that there was a large labor wedge in the United States during the Great Recession, and large efficiency wedges in the other G-7 countries.

Perhaps the most striking feature of the post-2008 US economy is not the Great Recession per se, but rather the failure of the US economy to recover from the Great Recession. The employment-to-population ratio, which peaked at 63.5 in mid-2007, declined sharply to 59.3 at the NBER-dated recession trough of July 2009, and was at 59.3 in June 2015.

While US productivity did not fall much during the recession, productivity growth has been remarkably low since the recession ended. Business sector productivity has grown at a 0.7 percent annual rate since mid-2009, compared to its historical average growth rate of 2.5 percent (see Ohanian, Taylor, and Wright 2012). Consequently, real GDP relative to the adult population is about 8 percent below its pre-2007 trend.

There also have been many policy changes during the Great Recession and its aftermath. There was a large fiscal expansion that included the American Recovery and Reinvestment Act (ARRA), which increased federal spending and provided transfers to states in order to increase demand; Cash for Clunkers, which provided subsidies for auto purchases; Homebuyer Tax Credits, which provided subsidies for home purchases; the suspension of the Payroll Tax, which was intended to increase consumer demand; and a number of Federal Reserve and Treasury initiatives, including the Troubled Asset Relief Program and quantitative easing, in which the Fed purchased securities that traditionally had not been part of its portfolio. There were other major policy shifts, including (p. 253) the Affordable Care Act, which substantially impacted health-care markets, and Dodd-Frank, which has substantially impacted financial markets.

Peacetime Wage- and Price-Control Policies

One of the most extreme macroeconomic policies adopted in the United States is wage and price controls. Widespread controls typically have been adopted only during wartime periods (see Rockoff 1984). However, President Nixon imposed peacetime wage and price controls in August 1971 that were intended to last ninety days to limit price increases following Nixon’s decision to stop gold convertibility. The controls ultimately lasted nearly three years. Several of Nixon’s advisers, including Labor Secretary George Schulz, opposed the controls, arguing that this would create shortages and inefficiencies. Federal Reserve Chairman Arthur Burns and Treasury Secretary John Connally were the main proponents of the policy, and they convinced Nixon that controls would be necessary to keep inflation in check after the United States decided to leave gold convertibility of the dollar.

The controls, however, appeared to be ineffective. Between January 1968 and August 1971, the CPI grew at an average annual rate of about 4.8 percent. CPI inflation averaged about 7 percent during the period of price controls, however, and inflation then approached 10 percent after that. Inflation during the control period was likely understated due to the fact that there were shortages of key goods, including gasoline and some foodstuffs. The controls were largely considered to be a failure, and the interest in “incomes policies,” as controls were known at that time, faded quickly.


This chapter has reviewed the US historical macroeconomic record and has assessed how economic policies have affected that record. With the exception of the post-2007 period, real GDP has typically been close to its trend growth of about 3 percent per year, and it has always returned to trend growth following periods of either above-normal or below-normal levels of economic activity. At the same time, economic policies have changed remarkably. This suggests, with perhaps the exception of the last few years, that policy changes on net have had little impact on very long-run US economic growth rates.

While policy shifts may have not had a large impact on very long-run growth, there is considerable evidence that policies have had large macroeconomic effects over other periods, including the Great Depression and World War II. Moreover, microeconomic policies, such as antitrust policies and labor market policies, may have had as much, or (p. 254) perhaps even more of an effect, on the US economy than traditional macroeconomic policies such as monetary and fiscal policies.


Anderson, Leonall, and Jerry Jordan. (1968). “Monetary and Fiscal Actions: A Test of Their Relative Importance in Economic Stabilization.” Federal Reserve Bank of St. Louis Review (November): 11–24.Find this resource:

    Arias, Andres, Gary Hansen, and Lee Ohanian. (2007). “Why Have Business Cycle Fluctuations Become Less Volatile?” Economic Theory 32(1): 43–58.Find this resource:

      Auerbach, Alan. (2005). “American Fiscal Policy in the Postwar Era: An Interpretative History.” Available at

      Baker, Scott R., Nicholas Bloom, and Steven J. Davis. (2012). “Has Economic Policy Uncertainty Hampered the Recovery?” In Government Policies and the Delayed Economic Recovery, edited by Lee E. Ohanian, John B. Taylor, and Ian J. Wright, Chapter 3, 24–40. Stanford, CA: Hoover Institution, Stanford University.Find this resource:

        Barro, Robert J. (1998). Determinants of Economic Growth: A Cross-Country Empirical Study. Cambridge, MA: MIT Press.Find this resource:

          Barro, Robert J., and Charles J. Redlick. (2011). “Macroeconomic Effects from Government Purchases and Taxes.” Quarterly Journal of Economics 126(1): 51–102.Find this resource:

            Bernanke, Ben S. (1983). “Non-monetary Effects of the Financial Crisis in the Propagation of the Great Depression.” American Economic Review 73(3): 257–276.Find this resource:

              Blanchard, Olivier J., and Lawrence H Summers. (1987). “Fiscal Increasing Returns, Hysteresis, Real Wages and Unemployment.” European Economic Review 31(3): 543–560.Find this resource:

                Braun, R. Anton, and Ellen R. McGrattan. (1993). “The Macroeconomics of War and Peace.” In NBER Macroeconomics Annual 1993, Vol. 8, 197–258. Cambridge, MA: National Bureau of Economic Research.Find this resource:

                  Cardia, Emanuela, and Paul Gomme. (2013). “The Household Revolution: Childcare, Housework, and Female Labor Force Participation.” Cahiers de recherche July 2013, Universite de Montreal, Departement de sciences economiques.Find this resource:

                    Chari, Varadarajan V., Patrick J. Kehoe, and Ellen R. McGrattan. (2007). “Business Cycle Accounting.” Econometrica 75(3): 781–836.Find this resource:

                      Christiano, Lawrence J., Roberto Motto, and Massimo Rostagno. (December 2003). “The Great Depression and the Friedman-Schwartz Hypothesis.” Journal of Money, Credit and Banking, part 2, 35(6): 1119–1215.Find this resource:

                        Cole, Harold L., and Lee E. Ohanian. (1999). “The Great Depression from a Neoclassical Perspective.” Quarterly Review, Federal Reserve Bank of Minneapolis (Winter): 2–24.Find this resource:

                          Cole, Harold L., and Lee E. Ohanian. (2002). “The Great U.K. Depression: A Puzzle and Possible Resolution.” Review of Economic Dynamics 5(1): 19–44.Find this resource:

                            Cole, Harold L., and Lee E. Ohanian. (2004). “New Deal Policies and the Persistence of the Great Depression: A General Equilibrium Analysis.” Journal of Political Economy 112(4): 779–816.Find this resource:

                              Cole, Harold L., and Lee E. Ohanian. (2015). “The Contributions of Cartel, Money and Productivity Shocks to the International Great Depression.” Working Paper, UCLA.Find this resource:

                                Conley, Timothy G., and Bill Dupor. (2013). “The American Recovery and Reinvestment Act: Solely a Government Jobs Program?” Journal of Monetary Economics 60(5): 535–549. (p. 255) Find this resource:

                                  Crucini, Mario J., and James Kahn. (1996). “Tariffs and Aggregate Economic Activity: Lessons from the Great Depression.” Journal of Monetary Economics 38(3): 427–467.Find this resource:

                                    Fishback, Price, Rebecca Holmes, and Samuel Allen. (2009). “Lifting the Curse of Dimensionality: Measures of the States’ Labor Legislation Climate during the Progressive Era.” Labor History 4: 313–346.Find this resource:

                                      Friedman, Milton. (1957). “The Permanent Income Hypothesis.” In A Theory of the Consumption Function, 20–37. Cambridge, MA: National Bureau of Economic Research.Find this resource:

                                        Friedman, Milton, and Anna J. Schwartz. (1963). A Monetary History of the United States. Princeton, NJ: Princeton University Press.Find this resource:

                                          Friedman, Milton, and Anna J. Schwartz. (1965). “Factors Accounting for Changes in the Stock of Money.” In The Great Contraction, 1929–33, edited by Milton Friedman and Anna J. Schwartz, 36–54. Cambridge, MA: National Bureau of Economic Research.Find this resource:

                                            Goodfriend, Marvin, and Robert G. King. (2012). “The Great Inflation Drift.” In The Great Inflation: The Rebirth of Modern Central Banking, edited by Michael Bordo and Athanasios Orphanides, 181–209. Cambridge, MA: National Bureau of Economic Research.Find this resource:

                                              Hanes, Christopher. (1996). “Changes in the Cyclical Behavior of Real Wage Rates, 1870–1990.” Journal of Economic History 56(4): 837–861.Find this resource:

                                                Hawley, Ellis. (1966). The New Deal and the Problem of Monopoly: A Study in Economic Ambivalence. New York: Fordham University Press.Find this resource:

                                                  Hoover, Herbert. (1951). The Memoirs of Herbert Hoover: The Great Depression, 1929–1941. New York: Macmillan.Find this resource:

                                                    Kendrick, John W. (1966). Productivity Trends in the United States. Princeton, NJ: Princeton University Press.Find this resource:

                                                      Kennedy, David M. (1999). Freedom from Fear, The American People in Depression and War, 1929–1945. New York: Oxford University Press.Find this resource:

                                                        Kilian, Lutz. (2014). “Oil Price Shocks: Causes and Consequences.” Annual Review of Resource Economics 6(1): 133–154.Find this resource:

                                                          Kovacic, William E., and Carl Shapiro. (2000). “Antitrust Policy: A Century of Economic and Legal Thinking.” Journal of Economic Perspectives 14(1): 43–60.Find this resource:

                                                            Kydland, Finn, and Edward C. Prescott. (1982). “Time to Build and Aggregate Fluctuations.” Econometrica 50(6): 1345–1370.Find this resource:

                                                              Lindbeck, Assar, and Dennis J. Snower. (1988). “Cooperation, Harassment, and Involuntary Unemployment: An Insider-Outsider Approach.” American Economic Review 78(1): 167–188.Find this resource:

                                                                Long, John B., Jr., and Charles I. Plosser. (1983). “Real Business Cycles.” Journal of Political Economy 91(1): 39–69.Find this resource:

                                                                  McGrattan, Ellen R. (2012). “Capital Taxation during the U.S. Great Depression.” Quarterly Journal of Economics 127(3): 1515–1550.Find this resource:

                                                                    McGrattan, Ellen R., and Lee E. Ohanian. (2010). “Does Neoclassical Theory Account for the Effects of Big Fiscal Shocks? Evidence from World War II.” International Economic Review 51(2): 509–532.Find this resource:

                                                                      Meltzer, Alan. (2009a). History of the Federal Reserve, 1951–1969, Vol. 1. Chicago: University of Chicago Press.Find this resource:

                                                                        Meltzer, Alan. (2009b). History of the Federal Reserve, 1969–1985, Vol. 2. Chicago: University of Chicago Press.Find this resource:

                                                                          Miron, Jeffrey A., and Christina D. Romer. (1990). “A New Monthly Index of Industrial Production, 1884–1940.” Journal of Economic History 50(2): 321–337. (p. 256) Find this resource:

                                                                            Odell, Kerry A., and Marc D. Weidenmier. (2002). “Real Shock, Monetary Aftershock: The San Francisco Earthquake and the Panic of 1907.” NBER Working Papers 9176.Find this resource:

                                                                              Ohanian, Lee E. (1993). “The Positve and Normative Effects of War Finance in the United States.” PhD diss., University of Rochester, Rochester, NY.Find this resource:

                                                                                Ohanian, Lee E. (1997). “The Macroeconomic Effects of War Finance in the United States: World War II and Korea.” American Economic Review 87(1): 23–40.Find this resource:

                                                                                  Ohanian, Lee E. (2009). “What—or Who—Started the Great Depression?” Journal of Economic Theory 144(6): 2310–2335.Find this resource:

                                                                                    Ohanian, Lee E. (2010). “The Economic Crisis from a Neoclassical Perspective.” Journal of Economic Perspectives 24(4): 45–66.Find this resource:

                                                                                      Ohanian, Lee E. (2014). “The Impact of Monetary Policy in the Midst of Big Shocks.” Journal of Economic Dynamics and Control 49(December): 35–48.Find this resource:

                                                                                        Ohanian, Lee E., John B. Taylor, and Ian J. Wright, eds. (2012). Government Policies and the Delayed Economic Recovery. Stanford, CA: Hoover Press.Find this resource:

                                                                                          Ramey, Valerie. (2011). “Can Government Purchases Stimulate the Economy?” Journal of Economic Literature 49(3): 673–685.Find this resource:

                                                                                            Rockoff, Hugh. (1984). Drastic Measures: A History of Wage and Price Controls in the United States. New York: Cambridge University Press.Find this resource:

                                                                                              Romer, Christina D. (1990). “The Great Crash and the Onset of the Great Depression.” Quarterly Journal of Economics 105(3): 597–624.Find this resource:

                                                                                                Romer, Christina D. (1993). “The Nation in Depression.” Journal of Economic Perspectives 7(2): 19–39.Find this resource:

                                                                                                  Romer, Christina D., and David H. Romer. (1990). “New Evidence on the Monetary Transmission Mechanism.” Brookings Papers on Economic Activity 21(1): 149–214.Find this resource:

                                                                                                    Samuelson, Paul. (1943). “Full Employment after the War.” In Postwar Economic Problems, edited by S. E. Harris, 27–54. New York: McGraw-Hill.Find this resource:

                                                                                                      Sargent, Thomas J. (1982). “The Ends of Four Big Inflations.” In Inflation: Causes and Effects, edited by Robert E. Hall, 41–98. Chicago: University of Chicago Press.Find this resource:

                                                                                                        Simon, Curtis. (2001). “The Supply Price of Labor during the Great Depression.” Journal of Economic History 61(4): 877–903.Find this resource:

                                                                                                          Stokey, Nancy L., and Sergio Rebelo. (1995). “Growth Effects of Flat-Rate Taxes.” Journal of Political Economy 103(3): 519–550.Find this resource:

                                                                                                            Temin, Peter. (1976). Did Monetary Forces Cause the Great Depression. New York: W. W. Norton.Find this resource:

                                                                                                              Temin, Peter. (1994). “The Great Depression.” NBER Historical Working Papers 0062.Find this resource:

                                                                                                                Wynne, Mark. (1990). “The Aggregate Effects of Temporary Government Purchases.” PhD diss., University of Rochester. (p. 257) Find this resource: