Finding a New Macroeconomics: (2) The Flawed Keynesian Model

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. – John Maynard Keynes [1]   


It was late in 2010 when I first saw a version of this graph showing changes in the share of total U.S. pre-tax income for the top 1% of incomes from 1913-2007. [2] It shows the top 1% share peaking in 1928 at 23.9%, one year before the stock market crash of October 29, 1929 ushered in the Great Depression.  Following WW II, that share gradually declined for three decades as prosperity grew, from 12% in 1946 to a low of 8.9% in 1976.  Then in 1980 the trend abruptly reversed and the top 1% share grew over the next three decades to 23.5% in 2007, one year before America’s worst-ever stock market crash on September 20, 2008 ushered in the “Great Recession.”

This kind of graph, and this kind of data, are new to economists, most of whom had not previously paid any attention to income distribution.  There has been a delayed response to the emergence of this data: It was not until 2012 that some of the most prominent U.S. economists even began to weigh in publicly on whether this correspondence between high income inequality and stagnation is more than just coincidental:  Nobel Laureate Paul Krugman, who did not address income inequality in his previous book on depression-era economics, [3] said this in his latest book, published in May of 2012: “The fact that a return to pre-Depression levels of inequality was followed by a return to depression economics could be just a coincidence.” [4] In his view, “rapidly rising incomes at the top” reflect “the same social and political factors that promoted lax financial regulation.” [5] Two months later, in his new book on inequality, fellow Nobel Prize winner Joseph Stiglitz remarked:  “It is perhaps no accident that this crisis, like the Great Depression, was preceded by large increases in inequality.” [6] Unlike Krugman, Stiglitz links inequality growth to aggregate demand and to overall income growth. 

The underlying question is whether the distribution of money within populations has macroeconomic significance, and the abundance of caution shown by economists in addressing that question shows how profoundly the emergence of detailed income distribution data has changed their world.  Phrasing these questions conversely may make them easier to answer: How could the amount of money most people possess lack macroeconomic significance? If income and wealth distribution does have material economic significance, how could the association of high inequality with depression be a coincidence?  

This graph reveals a close correspondence between growing income inequality and stagnation that was previously invisible: the growth of inequality was overlooked for years, because aggregate income (GDP) data hide income redistribution between the top 1% and the bottom 99%.  The income share data reveal a steady contraction of the bottom 99%’s share of the economy since 1980, suggesting that the increased concentration of income and wealth within the top 1% is related to the market crash in 2008 and the even worse decline that followed.  It further suggests that, unless the causes of this continuing inequality growth are reversed, the U.S. economy is headed inevitably for Great Depression II.  Hence, there is an urgent need to understand how inequality affects growth and prosperity.

As a career practitioner of regulatory economics, required by law to limit utility earnings to “reasonable returns” on investment, my initial impression was that excessive corporate profits have resulted in greater allocations of income and wealth to the top 1%, leaving too little money for everyone else.  Although that explanation is essentially correct, it has not been obvious to most economists.  

Those of us learning economics in the early 1960s were taught that Keynes had solved the depression problem, but as discussed in the previous post, Keynes’s General Theory took no account at all of wealth and income distribution, which he described in the above quotation  as “arbitrary.” Keynes’s “General Theory of Employment” included only determinants of demand and investment: (1) the interest rate; (2) the economy-wide “propensity to consume”; and (3) the “marginal efficiency of capital” (which is equivalent, over an entire economy, to the “cost of capital” used to set rates on regulated services).  

As noted in the previous post, Keynes expressly assumed that optimal demand would produce full employment, and that maintaining full employment would solve the “poverty” problem. But importantly, in Keynes’s model the levels of demand, employment, and investment do not depend in any way on the distribution of income and wealth; aggregate income (GDP) does not change with growing inequality. Accordingly, virtually all Keynesians have since ignored distribution presuming, as the Keynesian model effectively requires them to do, that growing inequality lacks macroeconomic significance.

One distinguished economist, however, believed otherwise.  When income inequality was falling in the 1950s, Nobel laureate Simon Kuznets, though hampered by an “extreme scarcity of relevant data,” argued insightfully that income distribution is a key determinant of growth:

Without  better knowledge of the trends in secular income structure  and of the factors that determine them, our understanding of the whole process of economic growth is limited; and any insight we may derive from observing changes in countrywide aggregates over time will be defective if these changes are not translated into movements of shares of the various income groups. [7] 

The data Kuznets (and Keynes) lacked is available today, however, and it offers a more complete understanding of how market economies work.

Instability and Growth

Keynes understood market economies to be inherently unstable, always tending toward decline and unemployment, and his explanation for this tendency revolutionized economics:  classical theory assumed that following a downturn an economy would always return to full employment “equilibrium” as savings and investment equilibrate; however, in a dynamic economy capital investment and job creation depend on expectations of future demand, and expected future returns on investment, which typically decline with falling consumption; thus, a decline in consumption leads to less, not more, investment and employment:

A decreased readiness to spend will be looked on in quite a different light if, instead of being regarded as a factor which will [all else equal] increase investment, it is seen as a factor which will [all else equal] diminish employment. [8] 

The central government could counter such slumps, Keynes argued, by stimulating demand and growth with monetary policy (lowering interest rates) or fiscal policy (deficit spending).

This “demand-side” explanation of how markets work, as the distinguished macroeconomist and Nobel Laureate James Tobin put it in 1997,  implies that “our market capitalist economy, left to itself, without government intervention,” does not “systematically return, reasonably swiftly, to a full employment state whenever displaced from it.” [9] Since the 1970s, however, mainstream economics has retreated back into the classical model, rejecting that conclusion:  “Most economists believe that a market economy is a self-correcting system,” Clifford W. Cobb wrote in 2009, [10] and James Galbraith recently concurred:

The deepest belief of the modern economist is that the economy is a self-stabilizing system. This means that, even if nothing is done, normal rates of employment and production will someday return. Practically all modern economists believe this, often without thinking much about it. [11]

This dubious faith in an automatic return to full employment after a typical fluctuation in spending on consumption or investment, such as a mild recession, denies that the instability of growth Keynes identified as resulting from such fluctuations in demand will be anything but temporary.  As such, this view, especially when motivated by a desire to minimize central government action through fiscal or monetary policy, retreats to the classical theory’s assumption of automatic adjustment to full employment equilibrium.  

According to Tobin, whether modern capitalist economies, left on their own without government supervision, “systematically return, reasonably swiftly, to a full employment state whenever displaced from it,” is one of “the central questions before economists of our generation.”  [12]  As Cobb and Galbraith suggest, Keynes’s negative answer to that question has apparently been rejected by most economists.

Introducing the Inequality Factor   

That is unfortunate, for the instability created by normal fluctuations in demand and investment, self-correcting or not, is only a minor aspect of the instability problem.  The entire debate over The General Theory ignores the far more substantial effects of growing inequality. Inequality results from structural changes in an economy — such as the development of monopolistic market power, the decline of trade unionism, or the effects of technological change — that affect income levels and wealth accumulation.  Unlike aggregate demand, these factors do not appear to have arguable cyclical or ebb-and-flow tendencies; the consistently unfailing purpose of capitalism is to maximize profits.  

Growing income inequality also causes aggregate demand to decline. As Joseph Stiglitz explained:

[W]hen money is concentrated at the top of society, the average American’s spending is limited. * * * Moving money from the bottom to the top lowers consumption because higher-income individuals consume a smaller proportion of their income than do lower-income individuals. [12]

Thus, when money transfers to people in the top 1% who have the lowest propensity to consume, the growing concentration of income itself reduces consumption and growth. People in the bottom 99% end up spending less not because they decide to spend less, but because they have less money to spend.  A depression, with its massive unemployment, poverty, and starvation, cannot be explained as an aggregate decision by the vast majority of households to reduce their consumption.  And because the distributional instability problem is declining income, not declining demand, its solution will be found not in Keynesian policies designed to stimulate demand and investment, but in measures that will reverse the growth of inequality.  

Keynes’s General Theory and demand-side model did provide part of the explanation for instability and stagnation, and that was an important first step.  But the second step — recognition of the inherent tendency of unfettered capitalism toward an unstable distribution of wealth and incomes — is far more important. Paul Krugman’s characterization of distribution issues as “social and political” is more semantic than real — all economic questions are fundamentally social and political in nature. But to dismiss distribution issues this way begs the questions posed by growing inequality: What happens when more and more of an economy’s money — its income and wealth — is transferred into the hands of a progressively smaller group of people? Why has so much money been transferring up over the last three decades? What are the transfer mechanisms, how are they created, and how can they be controlled?   

The answers to these questions reveal a major, and stunning, flaw in Keynes’s model. Recall this premise to Keynes’s model, cited in the previous post:

[T]he national income depends on the volume of employment, i.e. on the quantity of effort currently devoted to production. . . [T]here is a unique correlation between the two.  [14]

As discussed, Keynes intended to specify in his model the most important determinants of aggregate income, and his system of recommendations for government action is based upon using his model’s three determinants to maximize income (GDP).  As this quotation reveals, his approach depends on the assumption that “the national income” represents “the quantity of effort currently devoted to production.”  In other words, his model assumes that all income is compensation for actual production, for the creation of real value. This amounts to the assumption that there is no collection of excessive income or accumulation of extraordinary wealth by people at the top for which no real value has been created — i.e.,  that there is no collection of “economic rent.”  As discussed in a later post in this series, the collection of massive quantities of economic rent lies behind the inequality crisis.  This is the primary engine driving inequality growth and stagnation, and it explains why the Keynesian model fails to work.     

Not surprisingly, the newly available income share data reveal that traditional Keynesian demand stimulation offers both an inadequate level of response and the wrong kind of response to rising income inequality.  As income growth concentrates more and more at the top, even carefully targeted deficit spending may do little to counter the upward redistribution of wealth, and it will likely have minimal effect on the continuous and accelerating concentration of U.S. income.  

To determine the best policies for the United States today, it is necessary first to review the scope and nature of American inequality growth over the past thirty years.  The next post in this series provides such a review.

JMH – 4/27/2013


[1] John Maynard Keynes, The General Theory of Employment, Interest, and Money, 1935, New York and London, Harvest/Harcourt, Inc., 1953, 1964 ed., 1991 printing, p. 372. (For a fuller  understanding of Keynes’s General Theory  and his full employment model, see especially: pp. 30-32; Book 3, “The Propensity to Consume”; Book 4, “The Inducement to Invest,” esp. Chs. 11-14, and 18, “The General Theory of Employment Restated”; and Ch. 24.)

[2] Robert B. Reich, Aftershock: The Next Economy and America’s Future , NYAlfred A. Knopf (2010), p. 21.  The graph is constructed from the database of Thomas Piketty and Emmanuel Saez: See, e.g., Journal of Economic Perspectives, Vol. 21, No. 1, Winter 2007, pp.3-24; and with Stephanie Stantcheva, DP No. 8675 (CEPR, November, 2011).

[3] Paul Krugman, The Return of Depression Economics and the Crisis of 2008, W.W. Norton & Company, NY (1999, 2009).

[4] Paul Krugman, End This Depression Now!,  W.W. Norton & Company, NY (2012) p. 82.

[5] Ibid.

[6] Joseph E. Stiglitz, The Price of Inequality: How Today’s Divided Society Endangers Our Future, New York and London, W. W. Norton & Company, 2012, pp. 84-85.

[7] “Economic Growth and Income Inequality,” by Simon Kuznets, The American Economic Review, Vol. 45, No. 1, (March, 1955), p. 27.

[8] The General Theory, supra, pp. 184-185.

[9] “An Overview of the General Theory,” by James Tobin, Cowles Foundation Paper (here),  947 (1997) (here), p. 4.

[10] Clifford W. Cobb, Introduction, After the Crash: Designing a Depression-Free Economy – Selected Works of Mason Gaffney,  MA, Wiley-Blackwell (2009), p. 3.

[11] James Galbraith, Presidential Address, Association for Evolutionary Economics (January 5, 2013).

[12] James Tobin, op. cit.

[13] Joseph E. Stiglitz, The Price of Inequality: How Today’s Divided Society Endangers Our Future, supra at 84-84.

[14] The General Theory, supra, p. 246.

This entry was posted in - FEATURED POSTS -, - MOST RECENT POSTS -, Economics, Wealth and Income Inequality. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s