Skip to content
Publicly Available Published by De Gruyter July 25, 2015

Growing Apart: The Evolution of Income vs. Wealth Inequality

Michael Cragg EMAIL logo and Rand Ghayad
From the journal The Economists' Voice


The gap between the richest Americans and the rest of the nation has changed dramatically over the past three decades – becoming one of the most challenging political and economic trends for the nation. For decades prior to that, the distribution of wealth and income had been relatively stable, so much that a central problem posed in the economics literature was to explain this stability. But beginning in the early 1980s, inequality began to grow rapidly and has recently been attracting substantial attention from policymakers and researchers reflecting a widespread concern that reflecting a widespread concern that growing labor incomes of senior executives, finance professionals, and successful entrepreneurs is entailing large economic costs to society. The dominant paradigm in the media and Washington is that inequality is purely a matter of divergence in earned (labor) income inequality which can be ameliorated by making earned income taxes more progressive and shifting spending to help the poorer. However, this is not the story: wealth inequality, as it turns out, is much worse. This warrants emphasis for a variety of reasons: (1) a growing body of research that suggests that in the head-on comparison it is wealth inequality, rather than income inequality or poverty that has a negative, statistically significant effect on economic growth.[1] (2) Historically societies have failed when wealth has become overly concentrated; and (3) the wedge between earned and unearned income tax rates reduces progressivity as capital income rises. We offer a number of solutions which should generate debate amongst economists as they test conventional wisdom.

1 Rising Income and Wealth Inequality

Inequality in the US is a familiar phenomenon. Prior to the most recent recession, the bottom and middle class families were experiencing the ill effects of an unbalanced, top-heavy economy. Despite robust economic growth in the 1990s, followed by periods of relatively low unemployment, the US experienced income disparities that threatened middle class sustainability. Today, inequality is one of the main causes of our weak economy; and at the same time, our weak economy is leading to an increase in inequality, both of outcomes and opportunity. Unemployment is particularly critical. It has direct adverse effects, leads to lower wages, and induces public service cutbacks as tax revenues fall. It is thus not surprising that 95 percent of the gains in income in the first 3 years of America’s so-called “recovery” went to the top one percent.[2]

More broadly over the last 25 years, median income remained constant while the gap between median and mean income grew; and average income at best measures the level of income for the top 30 percent of the wage earners (see Figure 1). This gap has worsened considerably since the start of the Great Recession.[3]

Figure 1: Mean and Median Income.Source: Authors’ calculations.Data: SCF, Federal Reserve Board.
Figure 1:

Mean and Median Income.

Source: Authors’ calculations.

Data: SCF, Federal Reserve Board.

The distribution of net wealth (assets less liabilities) is even more inequitable. Median net worth rose steadily in real terms between 1992 and 2007 and then fell after the Great Recession. Mean net worth, on the other hand, has grown at a much faster pace than median wealth since 1989 and it is an even more top-heavy indicator than that of income as it is representative at best for top fifth and after the Great Recession is not even representative of this group (see Figure 2).

Figure 2: Mean and Median Wealth.Source: Authors’ calculations.Data: SCF, Federal Reserve Board.
Figure 2:

Mean and Median Wealth.

Source: Authors’ calculations.

Data: SCF, Federal Reserve Board.

The widening gap between the mean and the median is driven by increases in the top end of the distribution relative to others since the start of the Great Recession. Figure 3 depicts the average wealth for the bottom 90 percent of families (right y-axis) and for the top ten percent of families (left y-axis) from 1989 to 2013. Two results are worth noting. First, average wealth for the two groups moved together between 1989 and 2004. Second, average wealth for the bottom 90 percent has dropped at a faster pace than average wealth for the top 10 percent between 2007 and 2010; the Great Recession crushed all groups except the top few percent. The losses experienced by the bottom 90 percent are mainly driven by the substantial and persisting losses in the housing market. Because housing represents a larger share of wealth for those in the bottom half of the wealth distribution, their overall net wealth has been affected more by changes in home prices.

Figure 3: Top 10 Percent Wealth did not Suffer as much as Bottom 90 Percent.Source: Authors’ calculations.Data: SCF, Federal Reserve Board.
Figure 3:

Top 10 Percent Wealth did not Suffer as much as Bottom 90 Percent.

Source: Authors’ calculations.

Data: SCF, Federal Reserve Board.

The Great Recession entirely destroyed two decades of gains in income and wealth for middle income families (see Figure 4). Those in the middle (40–60 percent) of the distribution saw their average wealth falling more than five times faster than their income in 2007–2010 – mostly due to the decline in home prices. These wealth declines have been a persistent economic drag because growth in the 1990s and beyond reflected Chairman Greenspan’s explicit policies to encourage consumption through home equity loans. These loans cannot be repaid in the near term.

Figure 4: Wealth and Income for Middle Income Familes.Source: Authors’ calculations.Data: SCF, Federal Reserve Board.
Figure 4:

Wealth and Income for Middle Income Familes.

Source: Authors’ calculations.

Data: SCF, Federal Reserve Board.

2 Income vs. Wealth Inequality

In what follows, we divide the SCF sample into different population groups to quantify the substantial divide in income and wealth between the rich and the poor in the US. We measure the disparity by comparing the average amount of income and wealth owned by different socio-economic classes within the country (see Figure 5). On the right y-axis, we show the ratio of the mean income of those in the top one percent bucket versus the middle 20 percent (40–60 percent) of the distribution every year. The figure shows that in 2013 those in the higher bracket had 21.5× the income of those in the middle on average. This measure was 16× in 1989.

Figure 5: Top one Percent to Median.Source: Authors’ calculations.Data: SCF, Federal Reserve Board.
Figure 5:

Top one Percent to Median.

Source: Authors’ calculations.

Data: SCF, Federal Reserve Board.

The left y-axis shows the same statistic using net worth instead of income. As evident from the pattern, households in the top one percent of the wealth distribution had 118× the wealth of the middle 20 percent of the distribution in 1989. In 2013, the richer group had more than 237× the wealth of the middle 20 percent.

This increase in the wealth advantage enjoyed by the high income households has been argued to produce an unbalanced distribution of leverage among the income distribution – translating into higher debt leverage among poor and middle income households, and hence higher vulnerability to financial crises. Not only has the bottom groups have been holding higher debt leverage but it has also experienced a decline in its share of wealth which drove up their leverage ratio – the ratio of debt to assets-up. The Greenspan effects were highest for middle income families. Leverage ratios for them almost doubled as shown in Figure 6. As evident from Figure 6, households in the bottom end of the distribution have been more highly leveraged than those at the top – a factor that further amplified the impact of the recent crisis on wealth and wellbeing.

Figure 6: Leverage Ratio by Wealth Groups.Source: Authors’ calculations.Data: SCF, Federal Reserve Board.
Figure 6:

Leverage Ratio by Wealth Groups.

Source: Authors’ calculations.

Data: SCF, Federal Reserve Board.

3 Inequities in Horizontal and Vertical Tax Inequality have Increased

Summary data from the IRS show that TRA 86 dramatically reduced the degree of progressivity. Average taxes for the top one percent fell from 35 to below 25 percent while those for middle income tax payers fell from 16 to 15 percent (See Figure 7). Over the Clinton presidency, progressivity rose somewhat as the average rate on the top one percent rose to just below 30 percent. Since then the averate rates for the top one percent converged down to those of the top five percent and the middle tax payers. While middle income tax payers pay 15 percent of their income, the top one percent only pays 25 percent, despite the fact their income is 38 times higher and their wealth jumped from 114 times in 1989 to 220 times in 2013. The progressivity of the US tax systems has fallen dramatically over the last 25 years with disproportionately large gains accruing to the top one percent.

Figure 7: Average Tax Rates By Income Group.Source: Authors’ calculations.Data: IRS.
Figure 7:

Average Tax Rates By Income Group.

Source: Authors’ calculations.

Data: IRS.

The IRS does not provide the data necessary to estimate the distribution of taxes by income level. To examine the impact of taxes on various groups of the US population, we combine SCF income data with the NBER TAXSIM model to estimate federal tax liabilities for households of different incomes. We then estimate the federal tax rates for each household by dividing each household’s federal tax liability by its adjusted gross income.[4] Because different households can have different federal tax rates for any of several reasons, including the composition of income, the size of the household, and the amount claimed in tax-deductible expenses, we divide our sample into six groups based on their adjusted gross income and tabulate the federal tax rates within each income bracket. Figure 8 compares the 2010 cross-section to that of the 1989.

Figure 8: Distribution of Federal Taxes.Source: Authors’ calculations.Data: SCF and TAXIM.
Figure 8:

Distribution of Federal Taxes.

Source: Authors’ calculations.

Data: SCF and TAXIM.

The differences in rates within and across income groups have varied significantly between 1989 and 2010. In 2010, the 10th and 25th percentiles for tax rates for the top one percent of income earners lies below those for the top 10 percent.

While average rates generally rise with income, a significant portion of the highest income Americans were paying less in taxes as a share of their income than middle-class families. This pattern worsened over time particularly after the recent recession. The average tax rate masks the fact that some high-income individuals pay near their statutory tax rate, while others take advantage of tax expenditures and loopholes to decrease their tax payments. This variation in tax rates is particularly high among the very high income households.

The rise of capital income as a share of total income has greatly contributed to the rising income share of the top one percent of households by income. Capital income is heavily concentrated at the top of the income distribution, with roughly 75 percent of the benefit of the preferential rates on long-term capital gains and qualified dividends accruing to the top one percent of households.[5]

Are tax incentives contributing to rising inequality? The 1986 Tax Reform Act, which equalized tax treatment of labor and investment income, created an incentive for many households to shift income away from wages and salaries toward capital income in order to minimize capital liability. This led to a significant reduction in the average tax rate among the very wealthy Americans over recent decades. Among the top one percent, the average tax rate, including federal income and payroll taxes, has dropped a stunning 40 percent since 1986 according to data from the IRS tax returns. Recent research reveals a strong correlation between cuts in top tax rates and increases in the top one percent income shares.[6] Hence, contrary to supply-side claims about efficiency loss from income taxation, raising top tax rates could yield large reductions in income inequality growth without substantially reducing productive economic activity.

4 What Can Be Done?

Even with equal opportunity, inequality in wealth and income will arise because of variations in skill, ability, and luck. Indeed, variations in outcomes are good when they encourage hard work, investment, and getting an education. However, the extent and continuing increase in inequality is alarming and requires immediate efforts to mitigate its adverse impact on economic performance.

Address Rent Seeking: Rents are nothing more than re-distributions from one part of society to another and at best unproductive and at worst distortionary if rent seeking is rewarded more than productive activity. Furthermore, to the extent that skills in rent seeking are correlated with high skill work that is a complement to the productivity of other skill classes, risk seeking is then economically undesirable.

There appears to be consensus that the financial sector has enjoyed both increased opportunities and increased incentives for rent-seeking. The regulatory and legislative changes that emerged both before and even after the financial crisis of 2008 have increased the range of activities in which financial firms could engage. The distortive compensation in financial services whether through carried interest in hedge funds, the retained compensation invested in specialized vehicles and the returns to private equity and investment management can all be addressed through tax policy.

In the property sector, rents accrue merely from scarcity and location. Tax policy has long served to exacerbate these rents through the perpetuation of like-kind exchanges where the government subsidizes effectively loans money to speculators who can defer taxes on gains from buying and selling property without paying taxes. Modifying these tax subsidies to rent seeking is straightforward.

Historic tax credits, renewable energy tax credits, low income housing tax subsidies and other targeted tax credits may all appear to be socially desirable but to the extent that these credits are purchased and sold through partnerships without changing economic investments, they these too are a form of encourage rent seeking. Such transactions that lack economic substance are historically common and require the IRS have the resources necessary to audit partnership returns. Inequities have emerged simply because the IRS does not have the resources to properly do its job and make sure the credits are used for their intended purposes.

Another form of rent seeking is the enormous returns that accrue to those who create and extend market power through first mover advantages and patents. The tech run-up at the end of the 20th century and the current tech boom can both illustrate the returns to such rent seeking. Politicians and policy commentators too quickly assume that these tech returns represent productivity gains as opposed to economic rents. In the absence of rents, it is hard to reconcile the compensation of those in the top one percent with their potential productivity and the cascade of compensation impacts that follow but even it were, it is even more difficult to see how high marginal tax rates on these rents would negatively impact productive efforts, risk taking and economic efficiency.

Address Tax Inequities: Tax reform over the last few decades dramatically changed how the rising disparity in income and wealth also drive a decrease in the progressivity of the US tax system. The lower tax rates on capital income relative to those on earned income together with the relative increase in the share of capital income for the wealthiest means that the average tax rate for the wealthiest has fallen.

In his State of the Union address, President Obama asked to increase the rate on capital income to 28 percent for families with incomes over $500,000 while Republicans call for an a reduction in the capital gains tax rate. Two arguments are typically offered for a low capital gains rate. The first is to create incentives for savings based upon a variety of theoretical propositions all of which exclude a critical fact: the propensity to save is correlated with earnings. The second is to avoid double taxation of dividends since dividends and share purchases are exposed to corporate taxes while interest payments are not. But the most direct means of addressing this is to treat dividends and interest payments both as deductions against taxable income. A bi-product of this approach is to remove distortions in corporate capital structure caused by the asymmetric treatment of interest and dividends and will incent corporations to disgorge retained earnings (currently in the $trillions). A direct benefit of raising the capital tax rate on individuals is therefore the lowering of US corporate tax rates which will make US corporations more competitive, and reduce incentives for corporations to export income to lower tax jurisdictions either through transfer pricing or corporate inversions. Finally, raising the tax rate on capital income to the level of earned income is that there is no value to playing shenanigans with the definition of income in the financial services sector (worrying about taxing carried interest in private equity is not then an issue).

Raising the personal tax rates on capital income, reducing the corporate tax rates while making dividends tax deductible certainly level the playing field between those with earned income and those with unearned income from holdings of public corporations. The far more difficult question is how to deal with private ownership of partnerships, trusts and corporations. The US tax system works on a realization basis so that any asymmetry in the tax treatment of earned and unearned income creates incentives to avoid realization of taxable income through the use of partnerships, trusts and corporations. The rules are arcane and illustrated well by the simple example of real estate taxation.

Real estate investors who experience a gain on a project can invest this gain in other real estate property within 180 days without paying tax on the gain. This rule applies to all exchanges of like kind property. Real property investors can avoid taxation of their gains by simply rolling over the gains from one real estate partnership to another. Ultimately they will pay taxes when money is received but until then, wealth accumulates tax free.

Finally, there are strong reasons to eliminate large charitable deductions. Generally large charitable contributions are a form of consumption for the wealthy as the rich give to such “charities” as ballets, operas, private schools and Ivy League universities. There is no reason for subsidizing the funding of these elite institutions by allowing the rich to choose how they want to move their tax obligations from the government to their favorite charities.

Address Education Financing: There are many other policy issues that need to be addressed in the short-term. The way the US finances higher education must be reformed. Cutbacks in public support for higher education have made it more difficult for young people to work their way through school. High youth unemployment and bankruptcy reform making student debt almost impossible to discharge even in bankruptcy dramatically raise the risks to investing in higher education. For many individuals and families, educational debt has contributed to the widening gap in inequality. College tuition has risen much faster than income for the large majority of households since early 2000 and has become especially arduous for those in low and middle income families. According to the Federal Reserve Board of Governors, outstanding student loan debt quadrupled from $260 billion in 2004 to $1.1 trillion this year. This, along with data on family wealth from the SCF shows that the relative burden of education debt has long been higher for families with lower net worth, and this gap has increased dramatically in the past 20 years. Figure 9 shows that the ratio of outstanding education debt to yearly income more than doubled between 1995 and 2013 for those in the bottom 50 percent. For those in the 50–95 percent of the income distribution, the ratio was less but also doubled. The ratio of mean education debt to mean income for the top five percent was much less than the other 95 percent of families. Without changes educational financing is emerging as a large obstacle to creating wealth.

Figure 9: Ratio of Mean Education Debt to Mean Income.Source: Board of Governors of the Federal Reserve System, SCF.
Figure 9:

Ratio of Mean Education Debt to Mean Income.

Source: Board of Governors of the Federal Reserve System, SCF.

5 Conclusions

There is no magic bullet by which inequality in the US can be significantly reduced. Traditional analyses by economists of optimal tax and the role of government typically ignore the interplay of various dimensions of the tax system. In this paper we document the main features of rising income inequality and rising horizontal vertical tax inequity. We tackle three dimensions of the tax system central to addressing the rise in these dimensions of inequality. The documented rise in addressing inequality focused on redistribution, or at the very least how a progressive tax system and a progressive system of taxes and transfers could reduce inequality produced by a market economy. In contrast, a broader perspective should be focused on addressing every aspect of our economic and legal system that contributes to growing inequality starting with efforts to discourage rent-seeking in the financial sector by instituting stricter regulations. In addition, other policy prescriptions such as reforming corporate governance to give shareholders more power over executive pay decisions, increasing the bargaining power of the low- and middle-income workers by raising the minimum wage, improving access to education and health benefits, and reforming labor law to make it easier for workers to form labor unions are effective tools that should be considered.

Corresponding author: Michael Cragg, 44 Brattle St., Cambridge, MA 02138, USA, e-mail:
aMichael Cragg and Rand Ghayad are employed by The Brattle Group.


Bagchi, Sutirtha, and Jan Svejnar. 2013. Does Wealth Inequality Matter for Growth? The Effect of Billionaire Wealth, Income Distribution, and Poverty. IZA Discussion Paper No. 7733.Search in Google Scholar

Piketty, Thomas, and Emmanuel Saez. 2013. Optimal Labor Income Taxation, Hand-book of Public Economics, Vol. 5. Amsterdam: North Holland.10.1016/B978-0-444-53759-1.00007-8Search in Google Scholar

Saez, E. 2009. Striking It Richer: The Evolution of Top Incomes in the United States. Retrieved May 2, 2015. in Google Scholar

Toder, Eric, and Daniel Baneman. 2012. Distributional Effects of Individual Income Tax Expenditures: An Update. Washington, DC: Urban-Brookings Tax Policy Center.Search in Google Scholar

Published Online: 2015-7-25
Published in Print: 2015-8-1

©2015 by De Gruyter

Downloaded on 6.12.2022 from
Scroll Up Arrow