The previous page describes how unsustainable household borrowing stimulated the economy during the Consumer Age but ultimately led to recession when the housing bubble burst and credit dried up. This page links these macroeconomic trends to another important feature of the U.S. economy over recent decades: rising income inequality. The chart below shows the share of pre-tax income earned by the top 5% of the income distribution. Note that this share was remarkably stable from 1960 until the early 1980s. But then it began to rise sharply. Note also that the rise of income inequality corresponds almost exactly with the rise in household debt relative to income.
We discuss here our view that this correspondence is not a coincidence; rising income inequality was a central factor that led to increased household debt and therefore to the processes that ultimately led to the housing bubble and the Great Recession. Furthermore, we present the case that high and rising income inequality is an important factor in the sluggishness of the recovery from the Great Recession.
The analysis on this page is based on the research reported in Barry Cynamon and Steven Fazzari, “Inequality, the Great Recession, and Slow Recovery,” which is available here.
On the surface, the argument that higher inequality was responsible for a rapid rise in household spending might seem to get things backward. Going back at least to Keynes, economists have proposed that those with high income spend a smaller share of it than others, which suggests that rising inequality should raise saving and create a drag on consumer demand. But, as discussed on the previous page, the Consumer Age period prior to the Great Recession was a time of very strong demand for personal consumption and new houses. To understand the paradox of strong household spending with rising income inequality we need to dig deeper.
The inequality trend shown in the figure above resulted in large part from slowing income growth for the bottom 95% of the income distribution—that is, for just about everyone. These households could have responded by cutting either consumption growth or saving growth. During this period the data tell us that much of the response came from saving, consistent with the widely discussed fall in the American saving rate. Lower saving rates rescued the economy from demand drag that might have come from rising inequality during the Consumer Age. But it also put the ratio of household debt to income on an unsustainable path. Financial innovation in household lending—credit cards, new kinds of mortgages, home equity credit lines, et cetera—facilitated this trend for a long time. And falling interest rates delayed the day of reckoning, as households could refinance into lower cost loans that required less and less collateral (see a more detailed discussion of this process through the lens of Hyman Minsky’s financial instability theory here; **link to come soon**).
Data from the Federal Reserve’s Survey of Consumer Finance show that the rising debt trend was much more severe in the bottom 95% of the income distribution. The debt-income ratio rose 71 percentage points for the bottom 95% between 1989 and 2007, but just 6 percentage points for the top 5% (see the chart below). Eventually this behavior of the bottom 95% collided with limits on further borrowing when interest rates rose and the bubble in housing prices burst (as described on the previous page). With liquid savings depleted and much of new borrowing cut off, spending for personal consumption and new houses collapsed, and the Great Recession began.
Inequality and Income growth
From 1960 to 1980, inflation-adjusted income for the bottom 95% grew at an average rate of 3.9% per year while average growth for the top 5% was 4.0%. Between 1980 and 2007, average real income growth for the bottom 95% dropped to 2.6% per year while growth for the top 5% accelerated to 5.0% per year.
One can see the beginning of this “deleveraging effect” for the bottom 95% in the chart below. The debt-income ratio for the top 5% actually increases modestly.
How could the bottom 95% borrow so much? The chart below helps answer the question. It shows total net worth (assets minus debt) relative to after-tax income for the two income groups. The chart is indexed with the ratios set to 100 in 1989 so that you can compare the percentage changes across groups. (Of course, net worth to income is much higher for the top 5%, almost three times the value for the bottom 95% in 2010.) The net worth ratio for the top 5% grows somewhat faster than for the bottom 95% prior to the eve of the Great Recession in 2007. But the bottom 95% still appear to be doing reasonably well, which is perhaps surprising since they took on so much debt.
The next chart, however, shows what was really going on. This financial measure of net worth excludes the housing assets. It reveals big differences between the groups. The top 5% clearly accumulate lots of financial wealth until the Great Recession hits. The bottom 95% hold their own through 1998 (due in large part to rising stock prices). But when the unsustainable rise in home values no longer is counted as net worth, the measure contracts dramatically in the decade prior to the Great Recession.
It is clear that the balance sheet fragility that triggered economic collapse was concentrated in the bottom 95%, the same group whose share of income dropped dramatically in the years prior to the recession. Indeed, we can see this effect directly by looking as data on consumption and income that reveal important behavioral differences between the affluent and everyone else as the Great Recession unfolded. A recession, by definition, is a decline in output which causes a decline in income. Conventional economic models of consumption spending predict that when incomes decline during bad economic times, households will try to maintain a smooth path of consumption by drawing down savings or by borrowing more. This theory implies that in a recession we should see consumption relative to income rise, because consumption stays relatively stable while income declines.
The data in the chart below are with consistent this consumption-smoothing theory for the top 5%. Note the up and down cycles around the three recessions in the figure. When their income growth slows, the affluent spend a higher share of the money available to them, smoothing out fluctuations in their consumption. Then, when income growth resumes in the recovery, the ratio of consumption to income falls back to more normal levels. This pattern is especially striking for the Great Recession, during which the decline of income was more severe than in earlier recessions. But by 2012, the top 5% consumption-income ratio had fallen significantly, as it did following earlier slowdowns.
Things are much different for the bottom 95%. The consumption-income ratio for this group, shown in the chart above, was already high at the beginning of the data in 1989, as suggested by the decline in the aggregate saving rate and the rise of debt to income, starting at least five years earlier. From 1989 to the eve of the Great Recession, the consumption-income ratio for the bottom 95% rose further along a fairly smooth trend, compared to the dramatic up and down pattern for the top 5% (see the blue “Consumer Age” arrow in the chart above). This high spending of the bottom 95% was an important engine of demand growth for the U.S. economy over those decades. But when the Great Recession hit, the bottom 95% consumption-income ratio plummeted, exactly the opposite of the behavior predicted by the consumption-smoothing theory. Nothing like this had happened to this group in earlier recessions; what was different this time? Our answer is that the high and rising trend of spending for the bottom 95% during the Consumer Age was propped up by more and more borrowing. This way of generating demand was unsustainable (see further discussion here; link to Minsky’s financial instability page coming soon) and came to an abrupt end when the housing bubble burst, leading to the financial crisis of 2008. Therefore, what is unprecedented about the behavior of consumption in the Great Recession and its aftermath is the dramatic decline of consumption relative income for the bottom 95%, with more or less normal behavior for the top 5%. Although the consumption-income ratio rebounds somewhat for the bottom 95% in 2011, it falls again in 2012 and remains well below its pre-recession level with no indication that it will resume its climb. In retrospect this is not surprising when one recognizes that the behavior of the bottom 95% before the Great Recession was unsustainable.
According to the research behind the information presented here, rising inequality did not create much drag on demand when it started in the 1980s. Instead, it led the bottom 95% of households, as a group, onto a path of debt accumulation that would eventually cause economic collapse. Now that the borrow-and-spend era has ended for the bottom 95%, their consumption is no longer the engine of demand growth that it was during the Consumer Age.
Even though the demand growth that preceded the recession was financed in an unsustainable way, the economy needed that demand to maintain full employment. There is no evidence that demand was excessive prior to 2007—inflation was tame and interest rates were low. A robust recovery likely requires a return to the pre-recession trend of demand growth. But the financial crisis has now shackled household demand, and we doubt that the household sector can return to its previous spending trend. The chart below shows the profile of real (inflation-adjusted) consumption beginning with the onset of recessions since the mid-1970s and ending after 24 quarters (just shy of where the economy is in late 2013 relative to the beginning of the Great Recession). The data are scaled so that each recession begins at 100. Consumption tends to decline somewhat in a recession and, prior to recent experience, it recovered briskly after a few quarters. In the four recessions prior to the Great Recession, real consumption averaged about 18% higher 22 quarters after the recessions began. In the second quarter of 2013, 22 quarters after the Great Recession began, real consumption was just 6 percent higher than it was at the end of 2007 (the difference is highlighted by the blue arrow on the chart).
With personal consumption accounting for over 70 percent of demand, the lagging consumption recovery creates a huge gap in demand, with at least two thirds of the shortfall in consumption relative to the pre-recession trends concentrated in the bottom 95%. Indeed, real personal consumption of the bottom 95% is actually lower in 2012 than it was at the worst of the recession in 2008, while for the top 5% real personal consumption grew 17% between 2008 and 2012 (figures based on estimates from Cynamon and Fazzari, 2013).
Without reviving consumption spending of the bottom 95% we question whether the U.S. economy can sustain a robust recovery from the Great Recession. For a while, this goal could be accomplished by firing up another borrowing bubble, which would likely require another home price bubble. But that outcome is neither likely nor desirable. It’s not likely because the financial sector has been chastized by the mortgage-lending debacle that preceded the Great Recession. One would expect it will take some years to forget this painful experience before such risks are taken again. It’s not desirable because of the economic pain created by financial crises when the bubbles burst. We argue that a much better path to a strong recovery in consumption spendng is to restore high income growth across the wage and salary distribution so that consumption demand of all groups can grow along with the rising ability of the economy to produce in a sustainable way that does not require excessive borrowing. To see some ideas about how this might be accomplished, please go to “Income Inequality, Demand, and Economic Policy.”
We summarize the lessons to be taken from the dramatic events of the Great Recession here.
What can be done to fill the demand gap caused by the end of the debt-driven spending of the middle classes after the Great Recession? In principle, other sources of demand could take over the growth-generating role played by household spending prior to 2008. Businesses could invest more; but why should they with a stagnant economy? Households at the top of the income distribution could consume more, but that would further magnify the social tension from inequality. Governments at all levels could raise spending, but state and local governments remain financially constrained and the political debate at the federal level seems to be only about how much austerity to impose.
We believe that the best way to fill the huge and growing gap in U.S. demand would be if the trend toward greater income inequality is reversed, or at least stabilized. Tax policy could help to meet this goal, by providing more funds to the lower and middle parts of the income distribution. The cut in payroll taxes in 2011 and 2012 was a modest step toward restoring incomes that fell behind economic growth due to stagnating wages and salaries of the middle classes. But this policy fell victim to fear, excessive in our view, about government debt. Another approach could focus on strengthening the social safety net. Health care reform is a step in this direction, although it falls far short of what is needed.
Any redistributive policy is politically contentious. An alternative more consistent with the ideology of self reliance in the U.S. is for wage growth to keep pace with productivity growth across the income distribution. Basic principles of fairness and social justice among much of American society imply that the benefits of higher productivity should be widely shared: a rising tide should indeed raise all boats. The concept of the “American Dream,” which is close to a defining idea for the social outcomes we aspire to in the U.S., posits that people who work hard and play by the rules should be able to have a good and secure economic life as the result of the rewards from their work, not from a handout or forced redistribution. We agree with this principle, but what is somewhat different in the analysis presented here is that we argue for more widely shared benefits of prosperity not just because this goal is consistent with our concept of a just society. Our perspective implies that shared prosperity across the income distribution is also necessary for strong economic growth. Without sustainable demand growth in the middle class based on rising incomes, not rising debt, firms will not be able to sell all the output that they could produce at full employment. This problem not only compromises the economic life of those whose incomes fall behind potential growth, it also compromises the well-being of the affluent, when the businesses that generate the profits and capital gains fail to prosper due to inadequate sales growth to the middle class.
Wages and salaries across the income distribution did rise together with labor productivity in the decades of broadly shared prosperity after World War II. It is far from obvious how to implement policies to achieve this goal in the early 21st century, but there may be no other way to generate the demand necessary to escape the stagnant aftermath of the Great Recession in a sustainable way.
One step in this direction would be to increase the minimum wage substantially and adjust it for inflation, or possibly even to increase the minimum wage along with some measure of labor productivity. This policy would lift incomes at the bottom of the distribution and set a floor that would likely raise wages above the minimum as well. A rule that adjusts the minimum wage without political intervention as the overall economy prospers would help create stronger wage growth over time. In this regard, the indexation of Social Security benefits for inflation several decades ago provides a useful model. Not only has this policy improved the welfare of beneficiaries it has also helped fund sustainable demand growth from seniors and it is not subject to political bickering.
The Minimum Wage Over Time
The minimum wage in 2013 is $7.25 per hour. If the minimum wage had been indexed upward for both inflation and rising real labor productivity from 1980 (approximately when inequality began to rise) it would be about $15 in 2013. If one indexes for inflation and labor productivity back to 1960, the minimum wage would be about $22 per hour in 2013!
Perhaps the most important policy to encourage broad-based wage growth is a commitment to fiscal and monetary actions that keep demand growing at approximately full employment (see our discussion of fiscal policy and demand here). Low unemployment leads to fast wage growth both because it improves the bargaining power of workers and because it encourages firms to train scarce labor to move up the skill ladder.
Policy issues notwithstanding, a first step toward resolving the problem created for demand by stagnant wage growth in much of the income distribution is to have a clear understanding that inequality is more than an issue of social justice. The research discussed here shows that inequality also compromises the basic demand engine that the economy needs to grow at full employment.
Please continue to “Lessons from the Great Recession: A Summary.“