Show Summary Details
This content is available to you

Theories of stagnation in historical perspective

Roger E. Backhouse and Mauro Boianovsky

Keywords: secular stagnation; inequality; underconsumption; Hansen; Summers

This article traces some of the historical roots of current debates about secular economic stagnation, involving L. Summers, R. Gordon, and others. We focus on early contributions by Alvin Hansen and John A. Hobson. Although Hansen has been the main influence on the secular stagnation literature, Hobson's suggested link between economic inequality and stagnation is of interest in view of T. Piketty's recent claim about the connection between slower growth and inequality. E. Domar and P. Samuelson, Hansen's students at Harvard, elaborated economic growth and multiplier-accelerator models as follow-ups to Hansen's original treatment. Domar called attention to Hobson's insight about underconsumption, increasing capacity, and deficient aggregate demand. We conclude that the range of positions held today concerning stagnation and inequality is large, as it was earlier in the history of thought.

Full Text


Two old doctrines have recently been revived. Lawrence Summers (2014a; 2014b; 2015) has argued that we might be facing a situation of secular stagnation: a long period of low growth amounting to something beyond a regular cyclical downturn. Thomas Piketty (2014), in a book that attracted even more notice than Summers's article, drew attention to the trend towards increasing inequality and its consequences. These are both ideas that have generally been excluded from macroeconomics. Secular stagnation has for many years been considered empirically irrelevant and theoretically problematic, whilst economists have in general opted to refrain from discussing distributional issues, which raise welfare issues that are hard to treat without making political or ethical value judgments that few economists are comfortable making. The typical dynamic stochastic general equilibrium (DSGE) model used in modern macroeconomics abstracts from income distribution by analysing the behavior of a representative agent, and whilst imperfect competition may allow for unemployment, few macroeconomists took seriously the idea that long-term stagnation might be a serious possibility.

This literature raises three main issues: What is meant by the term secular stagnation? Does stagnation involve output lagging persistently below full capacity or does it involve a slowdown in the growth rate of productive capacity? Is there a connection between inequality and any tendency towards stagnation?

In the intervention that sparked off the most recent discussion, Summers offered a clear definition of secular stagnation: it was the outcome of a negative Wicksellian natural rate of interest, the rate of interest at which saving equals investment at full employment. 1 Demand for loanable funds may be low because slow population growth has reduced the demand for investment, because capital has become cheaper (thereby reducing the cost of any given level of real investment), or because leading companies are sitting on mountains of cash (Summers 2015). At the same time, the supply of savings is increasing for a complex of reasons. These include the reserves being accumulated by developing countries (particularly China), rising inequality, the effects of falling inflation, and changes induced by the financial crisis: increased demand for safe assets and an increased wedge between borrowing and lending rates as financial institutions seek to restore their profitability. 2 If the natural rate is negative, and inflation is very low, the existence of a zero lower bound for the market rate of interest would mean that the market rate must be higher than the natural rate, with deflationary consequences. Summers contends that recent experience makes such a situation more than a theoretical possibility. The EU and Japan are expected to experience zero interest rates for several more years and even if the US has rates above zero, the downward trend in interest rates over the past few years means that the zero lower bound is likely to bind in the next recession. Secular stagnation may not be an immediate problem for the US but it is likely to become important in the next few years.

Summers emphasizes the role of aggregate demand in creating the conditions under which stagnation is likely to be a recurring problem. Against this is the argument, represented by Gordon (2014; 2015), that stagnant growth is a supply-side phenomenon. He argues that the gap between actual and capacity output is falling, and that the problem is a slowdown in productivity growth. Taking a long-term perspective, he attributes this to a slowing down in the productivity gains associated with the third industrial ‘electronic’ revolution: investment in computing is encountering diminishing returns. He supports this with discussions of productivity growth in offices and retailing, and an argument that the declining share of employment accounted for by new businesses since the 1980s indicated a decline in business dynamism. 3 Looking ahead, he sees developments in education and social changes (fewer children living with both of their biological parents and rising numbers of young men with prison records) as signs that productivity growth may remain low. Thus, although he claims that ‘secular stagnation is evident in every measure of economic performance over the past five years,’ Gordon (2015: 58) sees it as a supply-side phenomenon, not brought about by the demand-side problems stressed by Summers. Whereas Summers's demand approach is taken largely from Hansen, Gordon's supply insight is reminiscent of Schumpeter (1942 [1947]), although he was less interested than Summers in tracing the historical roots of his approach and, therefore, did not refer to Schumpeter in that connection. 4

Although Summers cited rising inequality as one of the factors behind a rising propensity to save and the low natural rate of interest, the revival of this idea has been due to others. Because of their media profiles, Joseph Stiglitz (2012) and Paul Krugman (2013; 2014) have been the most visible proponents of the idea, along with Thomas Piketty, in his bestselling Capital in the Twenty-First Century (2014). They have, however, drawn on a wide range of research, from organizations such as the Organisation for Economic Co-operation and Development (Cingano 2014), the International Monetary Fund (Berg/Ostry 2011), and the International Labour Office (van Treeck/Sturn 2012), as well as from academic researchers who find a link between poor economic performance and rising inequality. There are differences about how this effect works, whether through reducing aggregate demand or through changing the structure of demand in ways that reduce employment and hurt anyone not in the top income bracket, but there is a wide spectrum of research supporting the view that the rise in inequality is dangerous. Krugman (2014) drew the conclusion that this ‘new view about inequality and growth’ had clearly become mainstream when it was endorsed by the rating agency, Standard and Poor's.

This paper traces some of the historical roots of these debates. Ours is a selective account, as we leave out of the picture secular stagnation theories that have not (directly or indirectly) influenced current debates – although they probably should have done so –put forward by J. Steindl (1952), P. Baran and P. Sweezy (1966), and other economists with Kaleckian and/or Marxian backgrounds (see Despain 2015; Hein 2015). Because his work was prominently cited by Summers, Alvin Hansen's role in developing the theory of secular stagnation is well known. However, it turns out that the story of the doctrine of secular stagnation is more complex: it did not disappear but, in the hands of Hansen's successors, it evolved into something quite different from the theory he proposed in 1939. However, we start not with Hansen but with the work of the British economist, J.A. Hobson, whose ideas on the relationship between inequality and growth influenced not only Lenin, but much American analysis of the Great Depression, the literature that forms the background to Hansen's work. Though praised by Keynes, Hobson was ignored by most economists. An important exception was Evsey Domar who found inspiration in his work and who provides a useful link between Hansen and more recent work. Hobson is occasionally cited in the modern literature on inequality and stagnation (see, for example, van Treeck/Sturn 2012).


In 1873 there was a financial crisis and a downturn in much of Europe and the US, with many countries, including Britain and the United States, entering what contemporaries believe to have been a long period of stagnation. Though historians have subsequently questioned this, arguing that national income continued to grow steadily, in Britain the period from 1873–1896 was known for many years as the ‘Great Depression.’ 5 A Royal Commission on Trade and Industry was set up in 1885 to look into the problem. Though this went against the prevailing free-trade orthodoxy, many of the businessmen called as witnesses argued that the problem was one of underconsumption: that consumers’ spending was too low to purchase all the goods being produced. F.A. Walker, the first President of the American Economic Association, believed that US output had fallen by a quarter between 1873 and 1879 and that there was nothing to stop it falling even further; industry might stagnate until something ‘from some unexpected quarter’ happened to revive expectations and produce an upturn. 6

This was the context in which Hobson first expounded his ‘heresy’ of underconsumption, in The Physiology of Industry (1889), co-authored with A.F. Mummery, the businessman who had converted him to the idea. According to their analysis, high savings were being invested in capital goods that were being used to produce more goods than could be sold, given the low level of consumption. It was an argument about the balance between the flow of spending and productive capacity. Mummery died shortly afterwards but Hobson went on to develop this argument, culminating in The Industrial System (1909). Here he developed the argument that excessive saving was the result of a maldistribution of income. Any income that was not a necessary cost of production (the cost of labor and the costs of producing and maintaining capital) constituted a ‘surplus.’ Part of this surplus was necessary to induce businesses to invest but the remainder constituted an unearned surplus that performed no economic function. This surplus, which accrued to the owners of whichever factor of production was scarce and hence had monopoly power, would typically be saved. The ‘unproductive surplus,’ Hobson claimed, represented ‘a failure of the competitive system to compete: it represents the powers of combination and monopoly’ (Hobson 1909: viii).

Hobson never developed this into a theory of long-term stagnation, though this was the implication of his argument. Instead he used it to explain imperialist ventures into Africa and elsewhere which provided opportunities for investing the surplus. His theory was never widely accepted in Britain, where free-trade ideology and the authority of Marshall were strong, but his ideas were well received in the US (see Rutherford 2007). During the 1920s there developed a wave of business cycle analysis. Hobson's ideas, emphasizing the flow of income and incorporating what came to be known as the ‘acceleration principle’ fitted well with the ideas being developed by John Maurice Clark (1917) and others. 7

After 1929, when American economists searched for explanations of the Great Depression (in which output really did fall massively), many of them turned to the idea that competition had broken down. 8 The growth of monopoly meant that the price changes necessary in a dynamic economy were disrupted and that economic power had become too concentrated. The early 1930s therefore saw much Hobsonian analysis of the relationship between monopoly, high savings, and depression. The problem was taken so seriously that in April 1938, Franklin Roosevelt wrote to Congress, claiming: ‘Among us today a concentration of private power without equal in history is growing. This concentration is seriously impairing the economic effectiveness of private enterprise as a way of providing employment for labor and capital and as a way of assuring a more equitable distribution of income and earnings.’ 9 This letter was the prelude to the setting up, by Congress, of a Temporary National Economic Committee (TNEC), which produced 33 000 pages of testimony and reports, investigating concentration of ownership and industrial power and its implications for saving, technological progress, and the operation of the financial system.

The final report of the TNEC (US, TNEC 1941) outlined the rise of monopoly and then drew conclusions. Prices were increasingly fixed by managers, not in competitive markets; productivity gains were not shared with consumers but just raised profits; and savings had become more concentrated. One result was that investment no longer responded to profits, for businesses were able to finance their activities out of retained earnings without having to enter the capital market. The collapse in income and employment was the result of a breakdown in the financial machine that transformed savings into investment. One of the key witnesses in persuading the committee to take this view was Hansen, who had a few years earlier coined the phrase ‘secular stagnation.’


The notion that the Great Depression was the result of a breakdown in competition fitted well with American traditions, in which the uniqueness of the US lay in part in its open, competitive society. Hansen took this connection between depression and specifically American circumstances a stage further when he linked it to the ending of the frontier, proclaimed by historian Frederick Jackson Turner in 1921. 10 With the closing of the frontier at the end of the nineteenth century a major source of demand for investment ceased. In the early twentieth century savings had been absorbed by new inventions (electricity and automobiles), which also required large capital investments, and by the need to equip a growing population with capital goods. The problem now facing the US, he argued in his testimony to the TNEC and in his Presidential Address to the AEA (Hansen 1939), was that the flow of capital-using inventions had ceased and declining population growth was reducing the demand for investment. The result was that depressions became longer and recoveries weaker and more short-lived, the problem he called secular stagnation.

This view was a natural extension of his previous work as a business cycle theorist. During the 1920s, he had begun to see fluctuations in investment, driven by population changes and waves of innovations, as the root cause of the cycle. He took from Albert Aftalion the idea that the price level is determined by level of money income in relation to the quantity of goods and services being produced, and from Arthur Spiethoff the idea that there were certain investment opportunities available and once these were taken up, investment would drop off, causing a downturn. A free enterprise system tended towards full employment because price flexibility encouraged a healthy level of investment and a high level of spending. However, although there was a tendency towards full employment, the business cycle was an inevitable feature of a dynamic, growing economy with rapid technological change. Only if the economy matured and accumulation slowed down would the cycle become a thing of the past.

These ideas conditioned Hansen's response to the Great Depression: it was a particularly deep depression because it was the result of large monetary and technological shocks happening together. For a considerable time, he continued to believe that recovery would eventually come largely of its own accord. He was still analysing depression in terms of the business cycle. It was only after the 1937–1938 depression, which took place even though full capacity had not been reached and unemployment was still over 10 percent, that Hansen looked for a different theory of what was happening. Given views on the cycle, all he needed to add to derive his theory of secular stagnation was to take account of declining population growth. The central point of his AEA Presidential Address (Hansen 1939) was that population growth was declining and that this would lead to a large fall in investment unless there was a rise in technical progress. ‘We are,’ he argued, ‘rapidly entering a world in which we must fall back upon a more rapid advance of technology than in the past if we are to find private investment opportunities adequate to maintain full employment’ (ibid.: 10). It was a view that fitted well into a long view of American economic history. The immediate origins of the crisis might be short-term, but its severity was the result of long-term structural factors. 11


Hansen's secular stagnation thesis was derived largely independently of Keynes's General Theory (1936). He had obviously read the book, reviewing it twice and coining the phrase ‘under-employment equilibrium’ to describe Keynesian theory (Hansen 1936), but he did not consider Keynes's theory, which he perceived to be too static, to be the same as his own. Keynes's (1937) article in the Eugenics Review showed that Keynes had come round to Hansen's own, longer-term, way of thinking. However, though continuing to distance himself from Keynes, Hansen came to realize that Keynes's concept of the multiplier could make a useful addition to his own theory, working it into a theory of the cycle. This was made very clear by Paul Samuelson, who noticed that Hansen's theory could be formulated as a second-order difference equation to which a general solution could be obtained (Samuelson 1939a). Samuelson noted that the result was a sequence identical to those found by Erik Lundberg and Jan Tinbergen. In a follow-up paper he developed this idea, arguing that the multiplier determined the level of national income and that the accelerator determined fluctuations (Samuelson 1939b).

Remaining very close to Hansen, and expressing the same doubts about Keynes, Samuelson became pessimistic about the likelihood of full employment after the war. This pessimism was based on the assumption that, when the war ended, Federal government spending would fall sharply and, in the absence of careful planning, would produce a major depression. He and Everett Hagen (1943), in a pamphlet produced for the National Resources Planning Board, argued in thoroughly Hansenian terms that once wartime spending was over, demand for investment would be low and savings higher. They remained pessimistic even though Hansen had by then become more optimistic about postwar conditions.

If there was a persistent tendency for savings to exceed private investment, full employment could be achieved only if the gap were filled by government spending. This might necessitate continued deficit financing, raising the question of whether the resulting government debt would become unbearable. This problem was tackled by another of Hansen's students, Evsey Domar (1944). Measuring the burden of debt by the fraction of national income required to pay interest on the debt, Domar showed that the crucial variable was the growth rate of national income. However, though his algebra showed that there would be a limit to the burden of debt, even with year-on-year budget deficits, Domar was pessimistic. He considered a constant growth rate unlikely. He argued, in Hansenian terms, that ‘[t]he rate of growth achieved in the United States in the period 1879–1928 was due to technological improvements, growth of the labor force, and the discovery of new resources’ (ibid.: 819). Given that population was growing more slowly, would be stationary or falling by 1980, and there might be some technological progress, a growth rate of 3 percent per annum was not sustainable, though it might be possible to achieve 2 percent for 50 or 100 years.

He followed this up in his better known paper, ‘Capital expansion, rate of growth and employment’ (Domar 1946), in which he developed his version of what later came to be called the Harrod–Domar growth model. It was this paper that provided a theoretical model of how secular stagnation could arise. Investment had two effects: it generated aggregate demand and increased productive capacity. Because investment was related to the level of aggregate demand and to the growth of productive capacity, there would be one growth rate at which aggregate demand would grow in line with productive capacity. Stagnation would occur if growth fell below this rate.

Using exponential functions, Domar showed that the limit of the capital–output ratio tends to the ratio of the propensity to save α to the rate of growth of income r. 12 By assigning various magnitudes and modes of behavior to the variables α and r, Domar used this apparatus of dynamic accounting relations to investigate problems involving capital accumulation, growth of public debt, rate of profits, etc. It follows from Domar's expression for the capital–output ratio that the rate of profits tends to equal the share of profits in income multiplied by the ratio of the rate of growth of income and the propensity to save. It may also be expressed, as Domar stated, as equal to the share of profits divided by the capital–output ratio, both of which had remained stable in the US. Interestingly, Piketty (2014: 52 and 166) has called Domar's expressions for the rate of profit and the capital–output ratio the first and second ‘fundamental laws of capitalism’ respectively, though he does not refer to Domar in that context (see also Boianovsky 2015).

Having derived these results, shifting the focus away from the burden of debt to the problem of economic growth, Domar (1947) related his ideas to the previous literature, most of which omitted one of these two aspects of investment instead of considering them together. One of the early writers he praised most highly was Hobson: ‘Hobson's writings contain so many interesting ideas that it is a great pity he is not read more often’ (ibid.: 51). Hobson had recognized the two aspects of investment and, though he had not talked in terms of secular stagnation, he had analysed the problem of persistent over-saving.

Hobson was fully armed with the σ [accelerator] effect of investment, and he saw that it could be answered only by growth. His weakness lay in a poor perception of the multiplier effect and his analysis lacked rigor in general. He gave a demonstration rather than a proof. But the problem to which he addressed himself is just as alive today as it was fifty and twenty years ago. (Ibid.: 52)

Domar concluded, contrary to Keynes's (1936) assessment, that Hobson's claim that underconsumption arose through savings being automatically invested in productive capacity expansion, thereby creating excess capacity, was not a mistake. The problem tackled by Hobson, Domar argued, was possibly a deeper problem than the one addressed by Keynes, namely what would happen when savings were not invested. Moreover, argued Domar (1947: 52), Hobson did not advocate reducing the propensity to save as much as possible, but ‘to reduce it to a magnitude commensurable with requirements from capital arising from technological progress–an interesting and reasonable idea’ not far from Hansen (1939). Like Hansen and Samuelson, Domar found Keynes too static. A decade later Domar (1957: 6 and 14) was even more forthright in his criticism of Keynes for suggesting that the economy tends to the ‘desert of the stationary state,’ a mistaken perspective that led him to endorse pyramid building and to anticipate the euthanasia of the rentier.

Hansen (1951: 258, fn 78), however, disagreed with Domar's positive assessment of Hobson's contributions and originality. In chapter 14 of his overview of the history of business cycle theories, Hansen (ibid.: 254–258) placed Hobson together with Lord Lauderdale and T.R. Malthus as early formulators of the role of aggregate demand in the determination of the level and changes of income. In Hansen's (ibid.: 255) opinion, ‘after reading Lauderdale and Malthus, one gains relatively little from Hobson's work.’ Hobson's suggested argument was that investment outlets are determined by the requirements of growth (technological progress and population growth). If the propensity to consume brings about an amount of full employment saving in excess of such growth requirements, the economy is beset by insufficient aggregate demand. From Hansen's standpoint, that was just a suggestion, as Hobson did not succeed in showing why the propensity to consume does not change automatically so as to provide precisely the amount of saving that matches the requirements of growth; this squares with Domar's 1947 remark that Hobson lacked the multiplier concept. Hansen acknowledged, though, that Hobson's analysis of the positive effect of a ‘greater equalization of incomes’ on aggregate demand and employment was a step in the right direction.

Samuelson realized, as early as 1946, that his forecasts of postwar depression had proved wrong. However, he continued to anticipate long-term problems. His most prominent discussion of the topic of secular stagnation came in his textbook, Economics: An Introductory Analysis (1948). It is probably little exaggeration to say that this was the book through which most students first encountered economics during the 1950s. Though he did not commit himself to the notion, presenting arguments for and against, he ensured that anyone studying his text would be familiar with the notion of secular stagnation. It was not until the 1985 edition, the first co-authored with William Nordhaus, that the topic was dropped (Samuelson/Nordhaus 1985). Rather than presenting it as a historical thesis about the United States, Samuelson presented it as a theoretical possibility: there might be secular stagnation if demand was too low or secular exhilaration if demand was too high. In the 1960s Samuelson presented it as an empirically relevant possibility, for in his sixth edition (1964: 353) he had a section titled, ‘A decade of sluggish growth and rising unemployment’ in which he pointed out that, since 1953, the unemployment rate had been higher at each successive cyclical peak (1952, 2.5 percent; 1957, 4 percent; 1960, 5 percent). He concluded: ‘Although the picture does not add up to one of stagnation, it does seem to have elements of sluggishness in it’ (ibid.: 353). This was the justification for the policy of the Kennedy and Johnson administration of running a budget deficit, and in later editions was described as the New Economics at work. Like Domar, Samuelson linked the problem of stagnation to problems of growth and public debt.


Though Hansen had derived most of his ideas independently of Keynes, not making any connection between underemployment equilibrium and secular stagnation, and though Samuelson was far closer, intellectually as well as personally, to Hansen than to Keynes, it was Keynes with whom the so-called ‘New Economics’ came to be associated. 13 Secular stagnation came to be associated with the last chapter of The General Theory rather than with the dynamic arguments favored by Hansen, Samuelson, and Domar. For example, Robert Lekachman, in a widely read book (1966), saw secular stagnation as lying at the heart of Keynesian economics, and Harry Johnson (1971), in an AEA Presidential address, argued that secular stagnation helped turn Keynesian economics into a new orthodoxy. However, with the decline of Keynesian economics, support for the idea of secular stagnation declined dramatically. Part of the reason was economic events. Whilst the idea of an economic crisis had traction in the 1970s, the idea that it was due to long-run flaws in capitalist economies was attractive only to a small group of heterodox economists. There were ample potential causes of the crisis that did not require invoking a general tendency towards stagnation. European economies stagnated in the 1970s but the United States recovered relatively quickly and there was an appeal to factors specific to Europe such as welfare-state spending or imperfectly competitive markets.

However, the main reason why the doctrine of secular stagnation fell out of favor in the 1970s was the rise of macroeconomic theories based on individual optimization and competitive markets. In what came to be called the Age of Keynes, from the Second World War to the early 1970s, economists accepted that many markets were imperfectly competitive and oligopolistic, and they also took for granted that economic agents were highly heterogeneous, making aggregation treacherous. Macroeconomic theories, therefore, had to rest on foundations that were empirical as much as theoretical. For example, labor markets, dominated by unions, were seen to be characterized by complex bargaining practices that defied formal theorization (game theory did not take off till the 1970s). As a result, much macro was based on assumptions such as mark-up pricing, cost curves that were horizontal up to full capacity, and wage equations postulating adjustment to exogenously given targets for wages that were estimated empirically.

In the 1970s economists increasingly abandoned such modelling practices, moving towards individualistic explanations of social phenomena. Searching for ‘microfoundations’ increasingly meant searching for formal models that related macro relationships to the behavior of optimizing individual agents (see Hoover 2012). It became acceptable to ignore heterogeneity and to adopt the device of the ‘representative agent,’ whereupon the macro behavior was simply micro behavior scaled up. It also became common to assume perfect competition. In such a world, modeled by Edmund Phelps, Robert Lucas, and others, the only source of unemployment was errors in expectations. In such a world it was hard to generate stagnation for any longer than it took people to learn that their expectations were systematically incorrect. Secular, long-term stagnation made no sense.

The introduction of market imperfections and phenomena such as hysteresis, whereby the equilibrium rate of unemployment, determined by supply-side factors, rose in response to persistent unemployment. But whilst this helped explain some features of European experience in the 1980s, it was a long way from the notion of secular stagnation, as the term had been understood by an earlier generation. Thus when the question of stagnation did rear its head, in Europe in the 1980s or Japan in the 1990s, it did not make sense to see these countries as exhibiting a problem underlying capitalist economies in general. Their experiences were explained in specific, local circumstances. To do otherwise would have been to question the foundations on which macroeconomics had been built.

This move towards individualistic explanations of social phenomena was not confined to economics: it was a much broader cultural phenomenon and had a strong political dimension (Amadae 2003; Rodgers 2011). Rationality, competition, and the efficiency of competitive systems came to be seen as defining features of capitalist democracies, marking their superiority over authoritarian, centrally planned socialist states. To entertain the possibility that capitalist economies might exhibit secular stagnation would be to question the superiority of the Western society over Communist regimes. Thus it was only after the fall of Communism in the Soviet Union and Eastern Europe that secular stagnation reappeared. It was used by Walt Rostow to analyse possible population trends in the twenty-first century, drawing on Japan as a country that had experienced these demographic changes earlier than other advanced countries (Rostow 1998; 2000). A phenomenon that, at the height of the Cold War, had been no more than a speculative possibility was now seen as a future that all countries had to confront.

Whilst Japan might face specific problems, which could easily be blamed on some of the many distinctive features of Japanese society, and whilst Southeast Asia and other parts of the world might have their own problems, there was great confidence in most Western countries as they entered the twenty-first century. The ‘dot-com’ bubble might have burst but the economic outlook was generally seen as benign, described in terms such as ‘the Great Moderation.’ Lending and real estate prices might have been high and inequality might have been rising but this was not taken to indicate structural problems. Stagnation seemed far away. When the crash came, there was a sudden spate of interest in Keynes, and central banks, supported by their governments, took decisive action. However, although Summers and economists who could be dismissed as being on the left (Stiglitz, Krugman, Piketty) might question whether structural problems were being overlooked, there was little inclination to question the foundations.


Although Summers's and Piketty's respective ideas about secular stagnation and inequality seem to be naturally close to one another, Summers's (2014c) critical review of Piketty's book indicates that the two sets of ideas do not necessarily go together. Whereas Summers accepts the Hobsonian perverse influence of inequality on growth–which is not discussed by Piketty–he rejects Piketty's claim that a lower growth rate leads to increased inequality. This reflects the contrast between a relatively high rate of return (or profit) in Piketty's scenario and a very low rate in that of Summers. Piketty's argument is based on the assumption that the wealth–income ratio increases without any substantial reduction in the rate of profit, which goes against established theory. The stability of the rate of return in Piketty's theory depends on the implicit assumption of a high elasticity of substitution between capital and labor. Moreover, as argued by some authors (see Pichelmann 2015), the saving ratio tends to fall as the rate of economic growth falls, meaning that the wealth–income ratio would increase only moderately (if at all) as growth falls.

Hobson's link between inequality and reduced growth came up in economists’ responses to the 2008–2009 crisis. However, as argued by Krugman (2008), it was difficult to reconcile that with the view that the American economy at the time featured overconsumption and overindebtedness instead. Economies with the biggest excess saving relative to investment, such as the euro area and (especially) China, export excess capital abroad by running large current-account surpluses, which drain demand from other countries. Hobson's (1902) discussion of capital export by Britain in the nineteenth century provides some interesting parallels. However, it is debatable whether capital exports helped lower British unemployment rates at the time, because of their potentially perverse effect on domestic investment (see O'Rourke 2015).

When these differences are taken together with the difference of opinion between Summers and Gordon over whether secular stagnation should be seen as a demand-side or a supply-side phenomenon, it becomes clear that the range of positions held today is very large. Looking back to the beginning of the twentieth century we find similar variety. Though Hobson's work was familiar to American economists in the interwar period, Hansen did not pick up on his arguments about distribution and stagnation, but constructed his own arguments rooted in American history. Among his students, Domar found much of value in Hobson, whilst Samuelson did not mention Hobson in the context of secular stagnation. There were other economists who wrote on stagnation, notably Steindl, but this literature remained substantially separate both in the arguments used and the literature on which it drew. The doctrine came to be tied to Keynes even though, in all its forms – Hobson, Hansen, Steindl – it drew as much on pre-Keynesian ideas as on The General Theory itself.

  • 1.

    This account follows Summers (2015: 60–62); some of the evidence is given in more detail in Summers (2014b).

  • 2.

    Given that tax is paid on nominal interest payments, falling inflation changes the relation between nominal and real interest rates.

  • 3.

    This account follows Gordon (2015: 55–56).

  • 4.

    The section titled ‘The stagnation thesis’ was added to the second 1947 edition.

  • 5.

    Morys (2014: 242–244) presents the statistics and explains why contemporaries might have viewed the period differently. The period was one of falling prices, though induced by falling costs, not by deficient demand.

  • 6.

    Walker's ideas are discussed in Backhouse (1987).

  • 7.

    Though the accelerator is usually attributed to T.N. Carver and Clark, the idea can be found very clearly, complete with a numerical example, in The Physiology of Industry (Hobson/Mummery 1889: 85–86).

  • 8.

    For a survey of this literature, see Backhouse (2015). Such ideas would play an important role again in Steindl's (1952) and Baran/Sweezy's (1966) interpretations of stagnation.

  • 10.

    This account of Hansen and some of the following material is a condensed version of the corresponding section of Backhouse/Boianovsky (2016), which owes much to Mehrling (1997).

  • 11.

    For a thorough account of the long-run facts behind Hansen's analysis, see O'Rourke (2015).

  • 12.

    See Appendix.

  • 13.

    There are many reasons for this, not least Hansen's (1953) decision to present himself as an interpreter of Keynes, but they go beyond the scope of this paper.


  • Amadae S.M. , Rationalizing Capitalist Democracy: The Cold War Origins of Rational Choice Liberalism , ( University of Chicago Press , Chicago 2003 ).

  • Backhouse R.E. , ' F.A. Walker's theory of hard times ' ( 1987 ) 19 ( 3 ) History of Political Economy : 435 - 446 .

  • Backhouse, R.E. (2015): Economic power and the financial machine: competing conceptions of market failure in the Great Depression, in: Marciano, A., Medema, S.G. (eds), Market Failure in Context (Annual Supplement to History of Political Economy, vol. 47), Durham, NC: Duke University Press, 99–126. (Early version available at URL: .)

  • Backhouse, R.E., Boianovsky, M. (2016): Secular stagnation: the history of a macroeconomic heresy, in: European Journal of the History of Economic Thought , forthcoming, available at URL: .

  • Baran P. & Sweezy P. , Monopoly Capital , ( Monthly Review Press , New York 1966 ).

  • Berg, A.G, Ostry, J.D. (2011): Inequality and unsustainable growth: two sides of the same coin?, IMF Staff Discussion Note, SDN/11/08, International Monetary Fund.

  • Boianovsky, M. (2015): Modeling economic growth: Domar on moving equilibrium, URL: .

  • Cingano, F. (2014): Trends in income inequality and its impact on economic growth, OECD Social, Employment and Migration Working Papers, No 163, OECD Publishing, URL: .

  • Clark J.M. , ' Business acceleration and the law of demand ' ( 1917 ) 25 ( 3 ) Journal of Political Economy : 217 - 235 .

  • Despain H. , ' Secular stagnation: mainstream versus Marxian traditions ' ( 2015 ) 67 ( September ) Monthly Review : 39 - 55 .

  • Domar E.D. , ' The ‘burden of the debt’ and the national income ' ( 1944 ) 34 ( 4 ) The American Economic Review : 798 - 827 .

  • Domar E.D. , ' Capital expansion, rate of growth, and employment ' ( 1946 ) 14 ( 2 ) Econometrica : 137 - 147 .

  • Domar E.D. , ' Expansion and employment ' ( 1947 ) 37 ( 1 ) The American Economic Review : 34 - 55 .

  • Domar E.D. , Essays in the Theory of Economic Growth , ( Oxford University Press , New York 1957 ).

  • Gordon, R.J. (2014): The turtle's progress: secular stagnation meets the headwinds, in: Teulings, C., Baldwin, R. (eds), Secular Stagnation: Facts, Causes and Cures , London: CEPR, 47–59, URL: .

  • Gordon R.J. , ' Secular stagnation: a supply-side view ' ( 2015 ) 105 ( 5 ) American Economic Review : 54 - 59 Papers and Proceedings .

  • Hansen A.H. , ' Mr. Keynes on underemployment equilibrium ' ( 1936 ) 44 ( 5 ) Journal of Political Economy : 667 - 686 .

  • Hansen A.H. , ' Economic progress and declining population growth ' ( 1939 ) 29 ( 1 ) The American Economic Review : 1 - 15 .

  • Hansen A.H. , Business Cycle and National Income , ( Norton , New York 1951 ).

  • Hansen A.H. , A Guide to Keynes , ( McGraw-Hill , London 1953 ).

  • Hein, E. (2015): Secular stagnation or stagnation policy? Steindl after Hansen, Levy Economics Institute Working Paper No 846.

  • Hobson J.A. , Imperialism: A Study , ( Nisbet , London 1902 ).

  • Hobson, J.A. (1909): The Industrial System , London: Longmans Green, URL: .

  • Hobson, J.A., Mummery, A.F. (1889): The Physiology of Industry , London: John Murray, URL: .

  • Hoover K.D. , ' Microfoundational programs ', in P. Garcia Duarte & G. Tadeu Lima (eds), Microfoundations Reconsidered: The Relationship of Micro and Macroeconomics in Historical Perspective , ( Edward Elgar , Cheltenham, UK and Northampton, MA 2012 ) 19 - 61 .

    • Search Google Scholar
    • Export Citation
  • Johnson H.G. , ' The Keynesian revolution and the monetarist counter-revolution ' ( 1971 ) 61 ( 2 ) The American Economic Review : 1 - 14 .

  • Keynes J.M. , The General Theory of Employment, Interest and Money , ( Macmillan , London 1936 ).

  • Keynes J.M. , ' Some economic consequences of a declining population ' ( 1937 ) 29 Eugenics Review : 13 - 17 .

  • Krugman, P. (2008): The underconsumption hypothesis, in: The New York Times , 27 July, URL: .

  • Krugman, P. (2013): Why inequality matters, in: The New York Times , 15 December, URL: .

  • Krugman, P. (2014): Inequality is a drag, in: The New York Times , 7 August, URL: .

  • Lekachman R. , The Age of Keynes , ( Random House , New York 1966 ).

  • Mehrling P.G. , The Money Interest and the Public Interest: American Monetary Thought, 1920–1970 , ( Harvard University Press , Cambridge, MA 1997 ).

  • Morys M. , ' Cycles and depressions ', in R. Floud, J. Humphries & P. Johnson (eds), The Cambridge Economic History of Modern Britain: II, 1870 to the Present , ( Cambridge University Press , Cambridge, UK 2014 ) 229 - 254 .

    • Search Google Scholar
    • Export Citation
  • O'Rourke, K.H. (2015): Economic impossibilities for our grandchildren?, Keynes Lecture, British Academy.

  • Pichelmann, K. (2015): When ‘secular stagnation’ meets Piketty's capitalism in the 21st century: growth and inequality trends in Europe reconsidered, European Commission, Economic Papers No 551.

  • Piketty T. Goldhammer A. , Capital in the Twenty-First Century , ( Harvard University Press , Cambridge, MA 2014 ).

  • Rodgers D.T. , Age of Fracture , ( Belknap Press of Harvard University Press , Cambridge, MA 2011 ).

  • Rostow W.W. , The Great Population Spike and After: Reflections on the 21st Century , ( Oxford University Press , New York 1998 ).

  • Rostow W.W. , ' Modern Japan's first challenge: the political economy of a stagnant population ' ( 2000 ) 144 Proceedings of the American Philosophical Society : 384 - 396 .

    • Search Google Scholar
    • Export Citation
  • Rutherford M. , ' Institutionalism and its British connections ' ( 2007 ) 14 ( June ) European Journal of the History of Economic Thought : 291 - 323 .

  • Samuelson P.A. , ' Interactions between the multiplier analysis and the principle of acceleration ' ( 1939a ) 21 ( 2 ) The Review of Economics and Statistics : 75 - 78 .

  • Samuelson P.A. , ' A synthesis of the principle of acceleration and the multiplier ' ( 1939b ) 47 ( 6 ) Journal of Political Economy : 786 - 797 .

  • Samuelson P.A. , Economics: An Introductory Analysis , ( McGraw-Hill , New York 1948 ).

  • Samuelson P.A. , Economics: An Introductory Analysis , ( McGraw-Hill , New York 1964 ).

  • Samuelson P.A. & Hagen E.E. , After the War: 1918–1920 , ( National Resources Planning Board , Washington, DC 1943 ).

  • Samuelson P.A. & Nordhaus W.D. , Economics: An Introductory Analysis , ( McGraw-Hill , New York 1985 ).

  • Schumpeter J.A. , Capitalism, Socialism and Democracy , ( Harper & Brothers , London 1942 [1947] ).

  • Steindl J. , Maturity and Stagnation in American Capitalism , ( Blackwell , Oxford 1952 ).

  • Stiglitz J. , The Price of Inequality: How Today's Divided Society Endangers our Future , ( W.W. Norton , New York 2012 ).

  • Summers, L. (2014a): Reflections on the ‘new secular stagnation hypothesis,’ in: Teulings, C., Baldwin, R. (eds), Secular Stagnation: Facts, Causes and Cures , London: CEPR, 27–38, URL: .

  • Summers L. , ' US economic prospects: secular stagnation, hysteresis, and the zero lower bound ' ( 2014b ) 49 Business Economics : 65 - 73 .

  • Summers, L. (2014c): The inequality puzzle, in: Democracy – A Journal of Ideas , 33, URL: .

  • Summers L. , ' Demand side secular stagnation ' ( 2015 ) 105 American Economic Review : 60 - 65 .

  • Turner F.J. , The Frontier in American History , ( Henry Holt , New York 1921 ).

  • US, TNEC (United States, Temporary National Economic Committee) , Investigation of Concentration of Economic Power: Final Report of the Executive Secretary , ( US Government Printing Office , Washington, DC 1941 ).

    • Search Google Scholar
    • Export Citation
  • Van Treeck, T., Sturn, S. (2012): Income inequality as a cause of the great recession? A survey of current debates, International Labour Office, Conditions of Work and Employment Series, No 39.


Define Y as output, K as the capital stock, r as the exponential rate of growth of output and α as the propensity to save.

Y T = Y 0 e r T

Savings are α Y , and the capital stock is the sum of accumulated saving up to time t:

K T = α Y 0 0 T e r t d t = a r 0 r ( e r T 1 ) .

Taking the ratio of these two, we obtain:

lim T K T Y T = α r .

Furthermore, if P denotes profits, the rate of profit is:

P K = P Y Y K .

From which it follows that

lim T P T K T = π r α ,
where π = P Y is the share of profits in output.


Backhouse, Roger E. - University of Birmingham, UK and Erasmus University Rotterdam, The Netherlands

Boianovsky, Mauro - Universidade de Brasilia, Brazil