Rural electrification, tractors, trucks, and automobiles all reduced labor requirements on the farm. As the historian Wayne D. Rasmussen writes, “With electricity farmers could run useful devices of all kinds, including not only electric lights but also milking machines, feed grinders and pumps. It took the war, however, and the accompanying shortages of farm labor, high prices for farm products and an enormous demand for those products to convince nearly all American farmers to turn to tractors and related machines.”47 He also notes that most Americans, to whom better-paid jobs were available, did not want farm jobs, leaving those jobs to immigrants or machines: “Much of the tomato crop in California had been picked by Mexican laborers who entered the U.S. under the terms of the Bracero program. When the program was ended in 1964, growers reported it was not possible to recruit U.S. citizens to do the work. Some labor leaders disputed this view, but the controversy was effectively ended by the successful mechanization of the harvest.”48
Even when workers were forced to leave the farm, technology was rarely the reason. Natural disasters like the Great Mississippi Flood of 1927 prompted people to search for more stable employment in the cities, which induced farmers to mechanize. The Dust Bowl of the 1930s was another environmental catastrophe that eroded the livelihoods of many farmers on the Great Plains, leaving them with no viable option except to leave.49 There can be no doubt that some farm laborers struggled as work disappeared in the countryside, especially during the Great Depression. But overall, farm laborers were drawn to the cities by the job opportunities offered by mass production. The Great Migration from the rural South to industrial cities like Chicago and Detroit was a key event in U.S. economic history. Spurred by World War I, which simultaneously increased the demand for workers in manufacturing and interrupted immigration from Europe, many people left farms for factories.50 This, in turn, spurred mechanization on the farms. As Ivanhoe Whitted of the Iowa Department of Agriculture explained to the New York Times in 1919, “The Iowa farmer is turning to the tractor because it helps toward the solution of the vexatious problem of farm labor.” Four decades before, he pointed out, there were few large cities, and labor in the countryside was plentiful and cheap. But thanks to manufacturing, great cities had sprung up at the expense of rural areas, leaving no farm labor to spare. The tractor, he added, “saved the day.”51 But it was enthusiastically adopted only after cheap labor had dried up.
The smokestack cities and towns of the Second Industrial Revolution, which powered American growth for almost a century, continued to churn out more stable and better-paying jobs for semiskilled workers. Perhaps the best evidence that people were drawn into the cities is the absence of resistance to the mechanization of agriculture after 1879. As noted above, before the Second Industrial Revolution, there were several episodes of unrest due to employment fears over the mechanization of agriculture, but thereafter opposition to agricultural machines practically disappeared.
In the age of mass production, despite the endless churn in the labor market, workers could for the most part expect to come out ahead. As productivity grew, so did their wages. Indeed, when Nicholas Kaldor listed six famous “stylized” facts of growth in 1957, he essentially summarized what economists had learned from the steady growth over the century, noting that the shares of national income received by labor and capital had been roughly constant over the long run.52
Equal Gains
As America got richer, it also became more equal. What makes the growth in the early twentieth century truly exceptional is not just its pace but how widely it was shared. The period 1900–1970 has rightly been regarded the “the greatest levelling of all time.”53 Incomes were rising for virtually everyone, and they were growing even faster at lower ranks. As Americans in the middle and at the lower end of the income distribution became the prime beneficiaries of progress, inequality went into reverse. Along with every other industrialized nation, America saw the share of income accruing to people at the top, fall. It may be telling that unlike economists of the Industrial Revolution (like Thomas Malthus, David Ricardo, and Karl Marx) who were all fond of apocalyptic economic predictions, economists living in the aftermath of the Second Industrial Revolution were largely optimistic—perhaps overly so. In any event, the idea that industrialists grew rich on the misery of workers had evidently fallen out of fashion. In the 1950s, Robert Solow advanced a model of a balanced growth path, in which progress delivered equal benefits for every social group; Kaldor put forward his stylized facts of economic growth, showing that the labor share of income had remained roughly constant, at two-thirds of national income, despite rapid mechanization; and Simon Kuznets advanced his hugely optimistic theory of economic progress in which inequality automatically decreases, regardless of economic policy choices.54 Their optimism surely seemed warranted at the time. Schumpeterian growth did indeed make America both richer and more equal.
Like the doomsday economists of the Industrial Revolution, however, twentieth-century economists were unfortunately fond of developing iron laws of economics that could be used to explain the trajectory of capitalist development for every time and place, though it is not hard to understand their appeal. Kuznets’s theory, which shaped how economists thought about inequality for half a century, was intuitive and straightforward. It predicted that the early shift from the agricultural sector to the higher-income, higher-inequality manufacturing sector would drive up inequality in the early phases of industrialization. But as manufacturing employed a growing share of the population, a larger percentage of citizens would harvest the benefits from growth, and inequality would eventually diminish. Technological progress, in other words, inevitably brings about an episode of increased inequality, but all economies have to do to achieve shared prosperity is wait for the cycle to complete itself. This was the cheerful message that Kuznets brought to the annual meeting of the American Economic Association—of which he was president—in Detroit in 1954, where he first outlined his thesis, soon to become known as the Kuznets curve.
Did American growth actually drive inequality along the lines that Kuznets suggested? Fortunately, we can explore his hypothesis empirically. Unlike the economists of the Industrial Revolution, Kuznets built his theory on a formidable statistical analysis—the first of its kind. Using newly collected data, Kuznets calculated that the share of annual national income captured by the top decile decreased by almost 10 percentage points from 1913 to 1948.55 Thus, he demonstrated that inequality had declined in the later stages of industrialization. Subsequent analyses by economic historians also traced the patterns of inequality all the way through the nineteenth century, allowing us to examine if inequality grew in the early stages of industrialization as the Kuznets hypothesis suggests. The most detailed account of the American experience is that of Peter Lindert and Jeffrey Williamson.56 Their findings show that both property and labor incomes became much more unequal between the dawn of the American Revolution in 1775 and the beginning of the Civil War in 1861—an unambiguous finding. Indeed, Alexis de Tocqueville, during his travels across America in the 1830s, found that the “the number of large fortunes is quite small.”57 America, at least relative to the Old World, still seemed to represent the Jeffersonian ideal of a nation consisting of independent and equal farmers. But Tocqueville also suggested that this ideal was gradually disappearing: “A manufacturing aristocracy … is growing up under our eyes.… The friends of democracy should keep their eyes anxiously fixed in this direction: for if a permanent inequality of conditions … penetrates into [America], it may be predicted that this is the gate by which they will enter.”58
By the end of the Civil War, the ascent of industrial “royals” had become all the more apparent. The new industrial America had become a Gilded Age, satirized in the 1873 novel The Gilded Age by Mark Twain and Charles Dudley Warner.59 The industries of the Second Industrial Revolution had not yet emerged, but steel, steam, railroads, and so on had already created unprecedented wealth. America, it seemed, was t
urning into a nation of the Old World, far removed from its Jeffersonian ideal and corrupted by an emerging industrial elite. Wealthy industrialists and financiers like John D. Rockefeller, Andrew Carnegie, J. P. Morgan, and Cornelius Vanderbilt were frequently labeled robber barons. In 1859, the journalist Henry J. Raymond compared Vanderbilt to a medieval nobleman, writing that he was “like those old German barons who, from their eyries along the Rhine, swooped down on commerce of the noble river and wrung tribute from every passenger that floated by.”60 To be sure, the size of the corporate giants that emerged must have been hard to comprehend. In 1893, for example, when the federal government collected $386 million in revenue, the Pennsylvania Railroad alone earned $135 million. And the combined railroad operations greatly exceeded the operations of the U.S. government, which was, of course, small by modern standards. Together, America’s railroad companies earned over $1 billion in revenues, and their combined debt was close to $5 billion—almost five times the national debt.61
While business historians have long debated to what extent the robber baron characterization is apt, contemporaries rarely pointed to American industrialists’ fortunes as the prime concern per se, instead of the methods by which they had been acquired. It goes without saying that how wealth is accumulated matters. Someone who successfully accumulates money by creating new sources of employment, comfort, and prosperity is to be regarded a public benefactor. In contrast, a person who profits from stifling competition, cheating his fellow citizens, and corrupting government is a public malefactor. Regardless of the means by which they acquired their wealth, however, the robber barons can surely account for a large share of the upsurge in the share of national income accruing to the top 1 percent. Yet to suggest that growing income inequality was entirely due to capital would be misleading.
Economists tend to rely on the Gini coefficient to measure levels of inequality across time and space. One of its virtues is that it is straightforward to interpret. If every citizen of a country had the same income, the Gini coefficient would be 0. If one person captured all the income, the income Gini coefficient would be 1. As Lindert and Williamson show, the Gini coefficient for property income was naturally higher, as property ownership is typically more concentrated, but labor income inequality rose even more rapidly: between 1774 and 1870 the property Gini grew from 0.703 to 0.808, while the earnings Gini increased from 0.370 in 1774 to 0.454 in 1860.62 A large part of this can be explained by the displacement of the blue-collar artisan, which led to the hollowing out of middle-income jobs and, thus, greater earnings inequality.63 Moreover, as Kuznets conjectured, inequality rose as a result of American workers shifting from low-income agricultural jobs to high-income manufacturing jobs in the city. Between 1800 and 1860, the urban-rural wage gap for a male laborer grew in both the North and the South. Because American urbanization accelerated between 1870 and World War I, while the Second Industrial Revolution spun out new industries and raised skill demand, it stands to reason that total inequality should have accelerated as well, all else being equal. It is therefore somewhat surprising that inequality (as measured by the overall Gini coefficient) plateaued around 0.5 and even fell slightly between 1870 and 1929, even though the top 1 percent share rose dramatically from 9.8 percent to 17.8 percent.64
Was the fall in inequality the result of industrialization’s reaching its intermediate stage? While the trajectory of American inequality between the Civil War and World War I does not disprove Kuznets’s hypothesis, it does suggest that there were also other factors at work. The upsurge in private wealth relative to total private incomes between 1870 and 1913 is consistent with the idea that capital played a larger role in shaping American income distribution, leading top incomes to continue to grow. In 1918, there were 318,000 corporations in America, of which the largest 5 percent captured 79.6 percent of the total net income. Thus, thirty-five years before Kuznets’s address to the American Economic Association, Irving Fisher, then the association’s president, painted a very different picture at its annual meeting. Fisher devoted his presidential address in 1919 to what he held to be the most serious challenge facing America at the time—namely, inequality, which he believed threatened the very foundations of American capitalism and democracy: “Whatever we may say of theoretical socialism of various types, and however much we may and ought, in my opinion, to favor in some form an increase of socialized industry, the great fact remains that the socialist group derives its real strength from class antagonism.… On the face of it, we should expect that all the evils mentioned would be relieved if we had more democracy in industry, that is, if the workman and the public felt that the great industries were partly theirs, both as to ownership and as to management.”65
In light of the concerns expressed by Fisher and others, it is not hard to understand why the idea of the Kuznets curve was received with such enthusiasm. It seemed to imply that redistribution was not necessary and would happen naturally if capitalism was allowed to take its course. Or would it? Just as Kaldor’s stylized theory of growth has recently been called into question, Kuznets’s assertion seems hard to reconcile with the post-1980 experience. Not only has the labor share of income consistently trended downward, but income disparities have trended upward in tandem (we shall return to this pattern in part 4). The reemergence of growing inequality seems difficult to reconcile with the Kuznets curve—a point that has been forcefully made by the economist Thomas Piketty.66
According to Piketty, the period observed by Kuznets was one of statistical abnormality. In the normal state of capitalism, Piketty argues, the return to wealth exceeds the overall growth rate of the economy, causing wealth-to-income ratios to rise and thus increasing income inequality, as wealth is highly unequally distributed. In Piketty’s world, there are no forces within capitalism that serve to drive inequality down. From time to time, however, macroeconomic or political shocks may disrupt the normal equilibrium. Two world wars and the Great Depression served to destroy the riches of the wealthy. The great leveling was the result of violence, economic collapse, and radical political change, not the tranquil process of structural change that Kuznets described. The historian Walter Scheidel has gone so far as to suggest that by destroying the fortunes of the wealthy, mass violence and catastrophes—like war, revolution, state collapse, or plague—have been the only economic levelers since the Stone Age.67
Needless to say, the violent shocks of the twentieth century struck America with much less force than they did Europe. But American fortunes were also buffeted. American private wealth decreased from nearly five times the national income in 1930 to less than three and a half in 1970, albeit this was mainly due to the Great Depression. Quite naturally, as Wall Street suffered from the depression, top incomes declined: the sharp fall in incomes of those employed in finance after the Great Depression almost exactly mirrors the drop in the top 1 percent’s share of income.68 This is not to suggest that policy didn’t matter. Indeed, the top income tax rate, which had been reduced to 25 percent in the administration of President Herbert Hoover, was hiked up to 63 percent under President Franklin D. Roosevelt in 1933 and continued to increase thereafter, reaching a staggering 94 percent in 1944.69 Yet inequality declined before as well as after taxes and transfers were adjusted for. And while violence, economic collapse, and policies designed to curb the influence of capital can all explain why top incomes fell, the leveling was not concentrated at the top. How macroeconomic or political shocks should have reduced inequality among the other 99 percent—particularly in the middle and lower ranks of the income distribution—is less evident, as most Americans’ incomes came from labor rather than capital.
Clearly, a number of short-run events are likely to have contributed to the leveling of American wages, but they do not provide the full story. The national minimum wage, first introduced in 1933, would have caused a compression in the lower ranks of the income distribution. Yet as Lindert and Williamson demonstrate, the great leveling was not confined to the lower ra
nks: compression was almost universal across occupational pay scales. Other government interventions, such as the tightening of immigration policy, provide another possible explanation, as immigrants were typically unskilled, and an influx of workers affects the growth of the labor force more broadly. It is plausible that immigration quotas caused the wages of unskilled Americans to rise, and indeed a slowdown in population growth might have caused the wage structure to become compressed. However, countries on the other side of the Atlantic similarly saw wage gaps narrow, suggesting that there were other forces at work. The fact that the wage compression was equally a European phenomenon also rules out other America-specific explanations, such as the National War Labor Board, which in 1942 was made responsible for approving all changes in American wages. Wage disparities remained quite stable after the board was dissolved in 1945, suggesting that the impact of wage-setting policies was limited.70
Just as the Kuznets curve was perfectly matched to its moment, Piketty’s treatment of inequality seems to have touched the zeitgeist. The Kuznets curve became a powerful political weapon in showing that America could stay true to its Jeffersonian ideal under capitalism. That surely seemed to be the case. The wages of ordinary citizens were growing as mechanization propelled productivity, and America was becoming a more equal place at the same time. The popularity of Piketty’s work, in contrast, reflects the post-1980 experience. Most Americans agree that inequality has reached unacceptable levels, and many quite rightly feel detached from the capitalist project: the trajectory of hourly compensation has been decoupled from productivity growth (see figure 9).
The Technology Trap Page 24