Unions, the high-wage doctrine, and employment.

Gallaway, Lowell E.

In the more than 200 years in which formal organizations of workers (labor unions) have existed in the United States, there have been three distinct eras of policy toward them. Initially, in the late 18th and early 19th century, they were regarded as associations that came under the purview of the English common-law doctrine of conspiracy--that is, their very existence could be considered illegal, regardless of the objectives of the group.

The first significant departure from that doctrine came in 1842, in a Massachusetts court. In Commonwealth v. Hunt, the conspiracy doctrine was rejected. That decision generally was accepted in other state courts, and it ushered in a period of neutrality with respect to the very presence of an organization of workers. The specific acts of labor organizations were still actionable, but not the existence of the organization itself. However, the law did not provide protection or encourage labor organizations.

This posture became the status quo for nearly a century until the 1930s, when public policy shifted in the direction of offering an explicit mechanism to facilitate worker organization, while providing protection for such groups once they came into being. The changes were embodied in the National Labor Relations Act of 1935, popularly known as the Wagner Act. With the passage of this legislation, the present era of public protection of labor unions was created. Why this happened, and its impact on the American economy, are the subjects of this article.

The 1920-21 Recession and the High-Wage Doctrine

Our story begins with the last significant business cycle downturn before the onset of the Great Depression in 1929, the one embracing the years 1920-21. It is here that the stage began to be set for the sea changes in public policy that occurred in the 1930s. In the first year of this period, wholesale prices fell by nearly 10 percent, while money wage rates increased by almost 6 percent (Vedder and Gallaway 1997: 63). That combination led to a 17 percent rise in real wage rates and an 18 percent fall in factory jobs and industrial output. That was merely the beginning. In the first two quarters of 1921, wholesale prices plummeted to 65 percent of what they had been in the first quarter of 1920. Money wages also fell, but not nearly as rapidly, and, by the middle of 1921, real wages were more than 40 percent higher than at the beginning of 1920. At that juncture, both factory employment and industrial output had fallen by nearly 30 percent (Vedder and Gallaway 1997). Stanley Lebergott (1964) estimated the unemployment rate for 1921 at 11.7 percent.

During 1921, money wages fell by more than 18 percent. In the 18 months that followed, industrial output soared by about 50 percent, while the number of factory jobs rose by nearly 30 percent (Vedder and Gallaway 1997: 63). According to Lebergott's (1964) estimates, the unemployment rate fell to 6.7 percent in 1922, and to 2.4 percent in 1923, presaging the "Roaring Twenties." Over the next six years, the unemployment rate averaged about 3.6 percent.

Yet, there was discontent with this sequence of events. In particular, the secretary of commerce during this period, Herbert Hoover, was revolted by it, insisting that there must be a better way to deal with business cycles. To him, the recovery had been achieved by imposing a high cost on the "bones of labor." Thus, he convinced President Warren G. Harding to convene a Conference on Unemployment to discuss "better ways" to deal with the unemployment issue. Hoover controlled the agenda, selected the participants, and saw to it that the laissez-faire view of allowing markets to generate recovery got short shrift. Still, the actual recovery following the downturn of 1920-21 came so quickly that there were no policy impacts emanating from this conference. However, the conference did anticipate the arrival on the national scene of what became known as the "high-wage doctrine."

Many of the conference participants believed that unemployment is the result of a lack of spending and that raising money wage rates would alleviate this shortfall. A popular publication, The Nation (12 October 1921: 389) went so far as to declare, "If it [the conference] really succeeds in its commendable attempt to stimulate buying by forcing manufacturers, middlemen, and merchants to accept lower profits, it will have done better than could have been expected."

The overall tone of the conference was anti-market adjustment. In some ways, it anticipated John Maynard Keynes. The Report on the President's Conference on Unemployment (1921) recommended counter-cyclical government public works spending, and even referred to "the multiplying effects of successive use of funds in circulation." Moreover, it rejected the notion that movements in relative prices (particularly declines in money-wage rates) could alleviate unemployment. Hoover promoted this view throughout the 1920s. According to Hoover (1923: 5), "We are continually reminded ... that there is an ebb and flow in the demand for commodities that cannot ... be regulated. I have great doubt whether there is a real foundation for this view."

Businessmen claimed in with support for the idea. In 1923, Boston retailer Edward Filene defended a wide range of measures that served to increase wage costs, such as minimum wage laws. He was not unique in his views. Earlier, Henry Ford had instituted a $5 a day wage, a level that was significantly higher than the standard wage of that time. Thomas Edison also expressed profound skepticism about the efficacy of market adjustments in eliminating unemployment (Dorfman 1959: vol. 4, chap. 2).

Elsewhere in America, the high-wage doctrine had advocates (see Taylor and Selgin 1999). To no one's surprise, labor union leaders argued for it. But, so did church groups, such as the Federal Council of the Churches of Christ in America. Finally, writers William Foster and Waddill Catchings (1925, 1927, 1928) offered a popular version of the high-wage doctrine. The title of their second book, Business without a Buyer, encapsulated their basic thesis--namely, that workers are the largest group of buyers of business products and paying them higher wages will better enable them to purchase the offerings of the business community.

The 1929 Crash and Hoover's High-Wage Policy

This is how things stood as the United States entered 1929. Herbert Hoover, a firm believer in the high-wage doctrine, had just been elected president. At the time of his election, in November 1928, the doctrine appeared to be a minor aspect of his political record. Unemployment was not a significant issue at the time of his election or when he was inaugurated in March 1929. Yet, before the year was out, the earth had moved under his feet. The stock market crashed in October and, according to one estimate, the unemployment rate increased to 9 percent in December (Vedder and Gallaway 1997: 77).

This was similar to 1921. The one major difference was that Hoover was not just a cabinet member; he was president. Thus, he could pursue his high-wage instincts directly, and he did. In late November, he assembled a group of major business figures at the White House. He importuned them to forego any money wage rate reductions in response to the depressed economic conditions, At the conclusion of the conference, the press release stated:

The President was authorized by the employers who were present at this morning's conference to state on their individual behalf that they will not initiate any movement for wage reductions, and it is their strong recommendation that this attitude should be pursued by the society as a whole.... They considered that, aside from the human considerations involved, the consuming power of the country will thereby be maintained [New York Times, 22 November 1929].

When Henry Ford left the White House, he offered these comments concerning this issue:

Nearly everything in this country is too high priced. The only thing in this country that should be high priced is the man that works. Wages must not come down, they must not even stay on their present level, they must go up.... And even that is not sufficient of itself We must see to it that the increased wages are not taken away from the people by increased prices that do not represent increased values [New York Times, 22 November 1929].

On the basis of these statements, one might conclude that employers followed Hoover's call not to reduce wages. But did they? Perhaps they just went ahead and reduced money wages. Observers at the time do not think this was the case. For example, Business Week (1 January 1930) published an article entitled, "This Time They Didn't Cut Wages." The great economic historian Joseph Schumpeter (1931) also noted the same thing, as did Carter Goodrich (1931). Further, labor economist Leo Wolman (1931: 2-3) wrote, "It is indeed impossible to recall any past depression of similar intensity and duration in which the wages of prosperity were sustained as they have been in the depression of 1930-1931." Others agreed with this view. For example, Don Lescohier and Elizabeth Brandeis (1935: 92) stated:

   In 1921 wage cuts were advocated early in the depression to
   liquidate labor costs. In 1930-1931 they were opposed both by the
   government and leading employers, in the hope that the maintenance
   of wage earners' incomes would furnish a market for products and
   help business recovery. In 1921 they were inaugurated long before
   business had reached a dangerous position; in 1931 they became
   common only after a large number of businesses had taken heavy
   losses. Realization of the reluctance of a large number of
   employers to cut wages caused wage earners and the public to accept
   them calmly when they did come, perhaps too calmly.

Confirmation of these views is provided by a formal statistical analysis conducted by my colleague, Richard Vedder, and myself in our book, Out of Work (1997: 95-97). We show that, in 1930, money wage rates were 8.3 percent higher than expected and, in 1931, the gap between actual and expected money wage levels widened to 10.5 percent. Apparently, Hoover's attempts at fending off money wage reductions were successful.

What were the results? In November 1929, the unemployment rate stood at 5 percent. One year later, the estimated unemployment rate was 11.6 percent, and, by November 1932, it rose to 18.6 percent (Vedder and Gallaway 1997: 77). Hoover's grand experiment in providing a more humane way to deal with unemployment appears to have been a failure. The law of unintended consequences reared its head. Ultimately, in March 1933, the unemployment rate peaked at more than 28 percent.

Despite a significant decline in money wages during 1932, real wages were nearly 10 percent higher than at the end of 1929. Given an almost 10 percent fall in the average productivity of labor, money wages were about 20 percent too high. These were the circumstances when Franklin Roosevelt was inaugurated. He had been swept into the presidency the previous November by an electoral coalition that included organized labor, which proved important in the famous "first hundred days," as Roosevelt's New Deal unfolded.

The New Deal, Labor Policy, and Unemployment

During this frenzied period, one of the major pieces of legislation being considered was the National Industrial Recovery Act (NIRA). As the particulars of the proposed law emerged, William Green, president of the American Federation of Labor, voiced organized labor's concerns about businesses being able to cooperate with one another in cartel-like fashion. (Vedder and Gallaway 1997: 138). As a part of the electoral coalition, this complaint was heard and the upshot was the inclusion in the final legislation of section 7a, which stated that workers shall have the right to bargain collectively.

The NIRA was passed at the end of the epic hundred days. In its final form, it included provisions for the establishment of industrial codes to spell out prices to be charged for products and floors on the wages (i.e., minimum wages) to be paid workers, which were set at high levels. This legislative act betrayed an almost studied obliviousness of the events that followed Hoover's ill-fated alternative approach to dealing with unemployment.

The impact of the minimum wage provisions of the NIRA was predictable. In the two quarters following its enactment, factory wages increased by about 20 percent (King 1938). In the process, a brief recovery was frustrated. Between March and July, the unemployment rate declined by 5 percentage points and then stabilized. Sixteen months later, it was virtually the same. Consistent with those data, factory employment declined in the fourth quarter of 1933 and did not return to its third quarter 1933 level until early 1935 (Vedder and Gallaway 1997: 134).

More importantly, there was a profound longer-term consequence of the NIRA. Although the law was held to be unconstitutional by the Supreme Court in 1935, section 7a lived on in a different form. Based on the experience under section 7a, a more detailed substitute for it was introduced in the Congress later that year--namely, the National Labor Relations Act of 1935, commonly known as the Wagner Act.

In the "Policy and Findings" section of the Wagner Act, there was an unabashed statement of the high-wage doctrine, stating that unequal bargaining power, in a nonunion milieu, "tends to aggravate recurrent business depressions, by depressing wage rates and the purchasing power of wage earners in industry."

What is to be seen in this statement is the siren-song quality of the high-wage doctrine. It has an appeal that clouds people's minds and makes them ignore any empirical evidence that might undermine the proposition. It is almost an "evidence be damned" attitude.

Initially, the Wagner Act did not have any significant effect on the level of money wage rates in the United States. The reason is straightforward. There was considerable uncertainty surrounding the eventual constitutionality of the law, and many employers adopted a "business as usual" attitude and declined to enter into collective bargaining agreements. Between the time of its passage and the end of 1936, money wage rates increased by a modest 4 percent, and real wage rates by even less.

Then came one of the major events of the 1930s. To many people's surprise, in April 1937, the Supreme Court, in National Labor Relations Board v. Jones & Laughlin Steel Corporation, declared the Wagner Act constitutional (see Cushman 1958). Employer resistance to the burgeoning labor union movement collapsed almost immediately. Examples of the devastating effect of the Supreme Court decision abound. By mid-1938, both money and real wage rates in the steel industry were nearly 20 percent greater than at the end of 1936, and total man hours of employment were more than 50 percent lower (Vedder and Gallaway 1997: 136). At a more aggregate level--that of total factory employment--the story was the same. In mid-1938, money and real wages were about 15 percent higher than at the end of 1936, and employment was 15 percent lower (Vedder and Gallaway 1997: 132-33).

What about the economy as a whole? In the last quarter of 1936, the unemployment rate was at 15.6 percent, down by almost 13 percentage points from its peak in March 1933. By the second quarter of 1937, it was even lower, standing at 13 percent. One year later, in the second quarter of 1938, it was back to 20.3 percent, more than a 50 percent increase. Once again, efforts to stimulate purchasing power by increasing wage levels ravaged the American economy (Vedder and Gallaway 1997:143-44).

The Intellectual Climate

Meanwhile, there were rumblings in the intellectual community. In 1936, John Maynard Keynes published The General Theory of Employment, Interest, and Money, which contained passages that were congenial with the high-wage doctrine and, by implication, labor unions. Keynes (1936: 10) accepted the classical notion that increases (decreases) in real wage rates are associated with decreases (increases) in employment. However, he argued that increases (decreases) in money wage levels are associated with decreases (increases) in real wages. If true, this proposition means an increase in money wages will decrease real wages and increase employment.

Such an argument cried out for empirical evaluation, and that cry was answered by John Dunlop (1938), an American economist, who examined the British data, and Lorie Tarshis (1939), a British economist, who looked at the U.S. data. Their articles were published in the Economic Journal, which Keynes edited. Both authors concluded that changes in money and real wages move together, not in opposite directions. Keynes (1939) obfuscated and brushed off the Dunlop-Tarshis critique. He then offered another argument--namely, that movements in money wages normally are matched by changes in prices, leaving real wages and employment unchanged (Keynes 1936: chap. 19). Hence, Keynes concluded that relying on reductions in money wage rates to eliminate unemployment--the classical view--would be ineffective to restore full employment.

Other arguments suggesting that the high-wage doctrine is viable also surfaced. One classic example was Patti Sweezy's notion of the kinked-demand curve in oligopolistic industries. Eventually, that idea became a staple element in microeconomics, but Sweezy's motivation was purely macroeconomic. At the kink in his demand curve, there is a substantial discontinuity in an oligopolist's marginal revenue curve. Sweezy argued that, within the range of this discontinuity, wage rates (and marginal costs) could be raised without any direct effect on output and employment. However, he thought the increased wage rates would stimulate aggregate demand and, indirectly, increase output and employment (Hewins and Shelley 1979: 149-50).

A significant difficulty with this line of argument is that union wage increases may not translate into a redistribution of income from the nonwage to the wage sector. Rather, its effect may be to redistribute income within the wage sector. The mechanism through which this may occur is very simple. If higher wages in the union sector lead to reductions in employment, the displaced workers will increase the supply of labor in the nonunionized portion of the labor market, driving down wages in that area. A simple set of statistics suggests that this is exactly what happened in the late 1930s and early 1940s. The surge in union membership following the passage of the Wagner Act was concentrated in mining, construction, manufacturing, transportation, communications, and public utilities. When the compensation of full-time equivalent employees in those industries is compared with that in the remaining private sector employment areas--wholesale and retail trade, services, and finance, insurance, and real estate--an interesting pattern emerges over the 1921-41 period. Expressing the wage differential between the union and nonunion sectors as a percentage of the overall average wage, one finds that from 1921 through 1935, the differential is about 5 percent. However, after 1935, it began to explode, especially in 1940 and 1941, exceeding 23 percent in the latter year (Vedder and Gallaway 1997: 141-42).

World War II and Its Aftermath

By 1940, there had been some recovery from the 1938 downturn. Lebergott (1964) estimated the unemployment rate to be 14.6 percent, about half the peak rate of March 1933. Yet, during the fourth quarter of 1940, the unemployment rate of 14.2 percent was the 41st consecutive quarter of double-digit unemployment. All through the period, it seems that every time there appeared to be light at the end of the tunnel, the government immediately built more tunnel.

Some counterfactual estimates indicate that New Deal programs in force in 1940, such as the Wagner Act, Social Security, and unemployment compensation programs, accounted for about 8.5 percentage points of unemployment with about three-fourths being attributable to unionization. In the absence of those programs, the Great Depression would have run its course. As it was, it took the onset of World War II and the drafting of many millions of men into the armed forces to bring an end to double-digit unemployment rates. Finally, in the third quarter of 1941, the unemployment rate fell below 10 percent and, for the entire year, it averaged 9.9 percent (Vedder and Gallaway 1997: 141-42). The unemployment rate plummeted as military conscription continued and, for a brief few years, there was no need for concern about unemployment. However, by 1943, it appears that the tide had turned in the war, meaning that it was appropriate to begin considering the postwar prospects for the American economy. In that year, one of the nation's economic doyens, Alvin Hansen (1943: 5), provided his prescription for the future, opining, "When the war is over, the Government cannot just disband the Army, close down munitions factories, stop building ships, and remove all economic controls." However, when the war ended somewhat precipitously, in 1945, within a year, that is exactly what the federal government did.

Remarkably, though, there appeared to have been no learning from the experiences of the 1930s. Indeed, less than a month after the Japanese surrender, President Harry Truman offered the following assessment of the postwar economic possibilities: "Obviously, during the process there will be a great deal of inevitable unemployment." He then argued for increases ha both the level and coverage of the minimum wage, explicitly espousing the high-wage doctrine by remarking, "The existence of substandard wage levels sharply curtails the national purchasing power and narrows the market for the products of our firms and factories" (New York Times, 7 September 1945). Truman also requested the passage of a Full Employment Act, a portent of things to come.

Such a law appeared in 1946, bearing the title "Employment Act of 1946." The Act contained an injunction to establish, "conditions under which there will be afforded useful employment opportunities, including self-employment, for those able, willing, and seeking to work, and to promote maximum employment, production, and purchasing power." It also created the Council of Economic Advisers and the Joint Economic Committee of Congress to assist the policy-making process, and required that the president submit annual Economic Reports to the Congress. In the first Economic Report of the President, submitted on January 8, 1947, the expression "purchasing power" took the spotlight. (1) Some examples are:

   It is ... of the utmost importance that at all times we be
   concerned as to the volume of purchasing power of the Nation and
   its relation to the volume of production of goods and services [p.
   2].

   It is plain ... that if employment is to remain high and if
   production is to increase in 1947, real purchasing power must rise
   sufficiently to take increased production off the market [p 10].

   Chief among the unfavorable factors is the marked decline in real
   purchasing power of great numbers of consumers.... Maximum
   production and employment this year would yield a substantial
   increase in the available supply of consumer goods and services,
   especially in the area of durable goods. This requires higher real
   purchasing power to take the goods off the market [p. 19].

According to the Economic Report (1947: 11), the higher purchasing power would be realized be reducing prices and increasing real wages: "A major approach to bringing real purchasing power of consumers into balance with productive capacity this year must be through reduced prices"--a back door way of achieving increased real wage rates. This was the rhetoric of the 1930s reprised. Politically, one can understand this approach; the labor movement was an integral part of the majority electoral coalition that emerged from the 1930s. As long as labor was a major political constituency, the high-wage doctrine had an undeniable appeal to elected officials.

The Disregard of Evidence against the High-Wage Doctrine

Far less understandable was the intellectual community's neglect of the evidence against the high-wage doctrine that accumulated during the 1930s. As pointed out earlier, all through that decade, increases in unemployment consistently were accompanied by rising real wage rates. However, the attitude among many influential intellectuals seems to be one of saying, "Don't bother us with the facts, it's our theories that are important." A classic example of this attitude appeared in a 1947 book written by future Nobel laureate economist Lawrence Klein, titled The Keynesian Revolution. Klein (1947: 107) examined Tarshis's (1939) analysis and considered the possibility that "Keynes was backing the wrong horse." But he dismissed that possibility: "Our main concern is not with the empirical problem but with the theoretical relation of wage cuts to employment." Has there ever been a clearer statement of the "don't bother me with the facts" idea?

Klein's attitude was not unique. As a general proposition, the intellectual community became a handmaiden for the high-wage doctrine, and, inferentially, for the labor movement. Most economists accepted Keynes's (1936: 268) statement that "it can only be a foolish person who would prefer a flexible wage policy to a flexible money policy." Meanwhile, the work of scholars such as Wilford King (1938) and Benjamin Anderson (1949), both of whom observed the wage disequilibria of the 1930s, was ignored. A new conventional wisdom was created that echoed the first sentence of the concluding chapter of Keynes's General Theory: "The outstanding faults of the economic society in which we live are its failure to provide for full employment and its arbitrary and inequitable distribution of wealth and incomes" (Keynes 1936: 390). That idea translated into the proposition that market failure was the story of the 1930s.

In truth, the problem was not market failure but intellectual failure on a grand scale. Many intellectuals overlooked the incorrectness of the "doom and gloom" forecasts for the postwar period and ignored the unprecedented decline ha federal government spending. Little attention was paid to the significant decline in real wages or to the decrease in wage compensation as a share of national income. Moreover, the relatively smooth transition after the sharp reduction in the size of the armed forces--nearly 10 million people left the military between June 1945 and June 1946--was inconsistent with the high-wage doctrine and the Keynesian weltanschauung. Instead, Keynesians assumed there was an unusual burst of consumption spending (a product of pent-up demand), although there was scarcely any evidence to confirm that presumption (Vedder and Gallaway 1997: 164-67).

Largely ignored was a reality that contradicted the high-wage doctrine. The declining real wage levels were symptomatic of the operation of a labor market adjustment process that established a new set of equilibrium relative commodity and resource prices by about mid-1948. The new relative prices incorporated the market's evaluation of the numerous structural changes in the economy that occurred in the 1930s. What followed was a series of relatively mild business cycles, including recessions in 1949, 1953-54, and 1958. In the first of these cycles, unemployment peaked in the fourth quarter of 1949 at 7 percent. For the second cycle, the peak occurred in the third quarter of 1954 at 6 percent. Finally, not quite four years later, in the second quarter of 1958, unemployment reached a cyclical high of 7.4 percent (Vedder and Gallaway 1997).

Signs of Intellectual Discontent

By the standard of the 1930s, these were minor episodes. Nevertheless, by the end of the decade, there were signs of discontent on the intellectual scene. For example, in 1957, Hansen (1965: 53-64) argued that we can do even better if we just relax and accept some additional price inflation. Meanwhile, in Britain, A. W. Phillips (1958) argued that there is a negative relationship between the rate of change in money wage rates and unemployment. Lurking in the wings in his paper was Keynes's version of the high-wage doctrine. The idea is that increases in money wage rates produce declines in real wage rates, causing employment to rise and unemployment to fall.

Phillips's paper became the basis for Paul Samuelson and Robert Solow's (1960) argument that there is a permanent tradeoff between the rate of price inflation and the unemployment rate; the "Phillips curve" was born. With it came an inflationary mind-set among economic policymakers: they were willing to accept a little inflation if it meant lower unemployment. The long-term viability of such a tradeoff, however, depended on workers' having a permanent money illusion--that is, being unaware of the effect of price changes on their level of real wages. Armed with the Phillips curve and high-speed data processing technology, policymakers believed they could fine-tune the American economy--and, for the better part of the 1960s, it appeared to work.

Alas, though, workers cannot be fooled permanently. Toward the end of the 1960s mad into the 1970s, the Phillips curve began to shift rightward, something that Milton Friedman had predicted in his 1967 presidential address before the American Economic Association (Friedman 1968). A new era was beginning for the U.S. economy.

Changes were also occurring for American labor unions. The wage differential between the unionized industries and the remainder of the labor force continued to expand rapidly. Between 1946 and 1960, it tripled while the overall level of compensation of full-time equivalent employees grew by about a third (Vedder and Gallaway 2002). The effect was dramatic. Employment in the unionized portion of the labor force grew more and more slowly. In 1950, employment in the unionized industries was about 57 percent of all private employment. However, between 1960 and 1970, union employment accounted for only 29 percent of the overall increase in private employment. After 1970, this figure declined to 17 percent between 1970 and 1980, and turned negative from 1980 to 1990. Job shifting from the unionized to the nonunionized sectors proceeded apace. All told, from 1950 through 1990, private sector employment increased by 52 million while the number of jobs in the traditionally unionized sector increased by only 9 million (Economic Report of the President 1998).

The phenomenon of job shifting had two profound effects on the American economy: it produced a significant volume of deadweight losses, and it increased the natural rate of unemployment. Building on a seminal paper written by Albert Rees (1963), Richard Vedder and I estimated that, cumulatively, deadweight losses amounted to approximately $50 trillion from 1950 through 1999 (Vedder and Gallaway 2002). We also found that as unions increased wage rates through the use of their monopoly power, job opportunities in the unionized industries decreased, increasing the supply of labor in the nonunion sector. This process pushed wages down in these areas and increased the number of lower-wage jobs available to workers engaged in the job-search process. This led the workers displaced from their higher-wage jobs in the union sector to increase the time spent searching for new jobs. The end result was an increase in the natural rate of unemployment. Using data from the individual states, we found that union density had a significant positive effect on the unemployment rate amounting to 0.074 percentage points of unemployment per percentage point of union density (Vedder and Gallaway 2002: 120-22). This result suggests that, during the post-World War II era, the bargaining activities of trade unions caused the natural rate of unemployment to be between 1 and 2 percentage points greater than it would otherwise have been.

Despite the negative effects associated with the implementation of policies based on the high-wage doctrine, there was no discernible lessening in its hold on political and intellectual figures. A simple anecdote illustrates quite well this phenomenon as it operated in the political arena. In 1992, when I was a senior economist for the Joint Economic Committee of Congress, file JEC held its Labor Day hearings in the midst of a recession. Morgan Reynolds, a well-known labor economist, who had been invited to testify, explained that the upsurge in unemployment was the result of money wage rates being higher than warranted, given the levels of commodity prices and labor productivity. Senator Donald Riegle (D-MI) responded to Reynolds by remarking, "I can't tell that to my constituents." Again, the attitude was, "Don't bother me with the facts."

Conclusion

In the intellectual world, the high-wage doctrine continues to have its appeal. Prior to his appointment as chairman of the Federal Reserve Board, Ben Bernanke, collaborating with Martin Parkinson, noted: "Maybe Herbert Hoover and Henry Ford were right. Higher real wages may have paid for themselves in the broader sense that their positive effect on aggregate demand compensated for their tendency to raise costs" (Bernanke and Parkinson 1989: 214). More recently, Patti Krugman reiterated this view in a New York Times op-ed (3 May 2009), arguing, "Many workers are accepting pay cuts in order to save jobs." He then asks, "What's wrong with that?' His answer refers to what he calls "one of those paradoxes that plague our economy right now.., workers at any one company can help save their jobs by accepting lower wages, but when employers across the economy cut wages at the same time, the result is higher unemployment." This is simply a reprise of Klein's (1947) views. Never mind the existence of more than a century of empirical evidence to the contrary. Krugman's concern is not with the empirical problem, but with the theoretical connection between wage rates and employment. The high-wage doctrine still lives. In 'all probability, this persistent adherence to an incorrect doctrine once again will prove to be detrimental to the U.S. economy, just as it was in the 1930s.

References

Anderson, B. (1949) Economics and the Public Welfare. New York: Macmillan.

Bernanke, Ben. S., and Parkinson, M. (1989) "Unemployment, Inflation, and Wages in the American Depression: Are There Lessons for Europe?" American Economic Review 79: 210-14.

Cushman, R., ed. (1958) Leading Constitutional Decisions. New York: Appleton-Century-Crofts.

Dorfman, J. (1959) The Economic Mind in American Civilization, 1918-1933. Vols. 4 and 5. New York: Viking Press.

Dunlop, J. (1938) "The Movement of Real and Money Wage Rates." Economic Journal 48: 413-34.

Economic Report of the President (1947, 1998) Washington: U.S. Government Printing Office.

Edie, L. D., ed. (1923) The Stabilization of Business. New York: Macmillan.

Filene, E. A. (1923) "The Minimum Wage and Efficiency." American Economic Review 13: 411-15.

Foster, W. T., and Catchings, W. (1925) Profits. Boston: Houghton Mifflin.

--(1927) Business without a Buyer. Boston: Houghton Mifflin.

--(1928) The Road to Plenty. Boston: Houghton Mifflin.

Friedman, M, (1968) "The Role of Monetary Policy." American Economic Review 58: 1-17.

Goodrich, C. (1931) "The Business Depression of Nineteen Hundred Thirty." American Economic Review 21:183-201.

Hansen, A, (1943) After the War: Full Employment. Washington: United States National Resources Planning Board.

--(1965) "The Case for High-Pressure Economics." In A. Okun (ed.) The Battle against Unemployment. New York: W. W. Norton.

Hewins, D. C., and Shelley, J. B. (1979) "Sweezy's Kink: Macro Foundations for a Micro Theory." Economic Inquiry 17: 149-50.

Hoover, H. C. (1923) "Introduction." In L, D. Edie (ed.) The Stabilization of Business. New York: Macmillan.

Keynes, J. M. (1936) The General Theory of Employment, Interest, and Money. New York: Harcourt Brace.

--(1939) "Relative Movements of Real Wages and Output." Economic Journal 49: 34-51.

King, W. I. (1938) The Causes of Economic Fluctuations. New York: Ronald Press.

Klein, L. (1947) The Keynesian Revolution. New York: Macmillan, Krugman, P. (2009) "Falling Wage Syndrome." New York Times (3 May).

Lebergott, S. (1964) Manpower in Economic Growth: The American Record since 1800. New York: McGraw-Hill.

Lescohier, D., and Brandeis, E. (1935) Working Conditions and Labor Legislation. Vol. 3 of History of Labor in the United States, 1896-1932. New York: Macmillan.

Phillips, A. W. (1958) "The Relation between Unemployment and the Rate of Change in Money Wage Rates in the United Kingdom, 1861-1957." Economica 25: 283-99.

Rees, A. (1963) "The Effects of Unions on Resource Allocation." Journal of Law and Economics 6: 69-79.

Report of the President's Conference on Unemployment 1921) Washington: U.S. Government Printing Office.

Samuelson, P. A., and Solow, R. M. (1960) "Analytical Aspects of Anti-Inflation Policy." American Economic Review 50: 177-94.

Schumpeter, J. (1931) "The Present World Depression: A Tentative Diagnosis." American Economic Review 21: 179-82.

Tarshis, L. (1939) "Changes in Real and Money Wages." Economic Journal 49: 150-54.

Taylor, J., and Selgin, G. (1998) "By Our Bootstraps: Origins and Effects of the High-Wage Doctrine and the Minmum Wage." Journal of Labor Research 20: 447-62.

Vedder, R., and Gallaway, L. (1997) Out of Work: Unemployment and Government in Twentieth Century America. Expanded edition. New York: New York University Press.

--(2002) "The Economic Effects of Unions Revisited." Journal of Labor Research 23: 105-30.

Wolman, L. (1931) Wages in Relation to Economic Recovery. Chicago: University of Chicago Press.

(1) Edwin G. Nourse was the first chairman of the Council of Economic Advisers and played a key role in writing the 1947 Report.

Lowell E. Gallaway is Distinguished Professor of Economics at Ohio University.