Note: All figures are in 2010 dollars unless otherwise specified.

For those in progressive policy circles, learning that income inequality isn’t increasing would be kind of like discovering the earth orbits the moon. It’s a finding that flies in the face of everything we’ve come to believe, a realization that would threaten our very senses of self.

Yet that, precisely, is the conclusion of a recent paper published by Cornell economist Richard Burkhauser, along with colleagues Philip Armour and Jeff Larrimore.

In fact, Burkhauser says, in the past two decades, the poorest Americans have seen their incomes grow the fastest, while the wealthiest Americans have seen their incomes shrink the most.

How can this be?

Much of it has to do with the way Burkhauser defines income: consumption plus changes in wealth. His major insight is that accumulated capital gains—appreciation in the value of assets like investments or property—have major implications for financial well-being, whether or not those assets are sold for a realized profit during any particular time period.

This income definition is comprehensive (it includes taxes, transfers, and in-kind benefits, like health insurance) and theoretically sound (few economists would argue against including accrued wealth in a measure of economic well-being).

Yet as my colleague Ben Landyhasdiscussed, Burkhauser’s approach is not without controversy, due to issues with data availability and modeling assumptions (Burkhauser himself acknowledges as much).

But Burkhauser is a serious economist, not a partisan hack. His conclusions are less important than his ideas – and his ideas raise an important question: is our conventional approach to studying economic inequality too simplistic?

When we talk about income inequality, we usually talk about the rich getting richer and the poor falling further behind. And we almost always measure it in terms of market income. That makes sense: the income distribution is the most natural way of talking about the income distribution. But being “low-income” is not necessarily the same thing as being “poor.” And economic inequality is not confined to income quintiles. Wealth matters, too.

To get a better understanding of the trends Burkhauser highlights, I took a closer look at the Survey of Consumer Finances, the Federal Reserve study of families he uses for many of his wealth calculations. Conducted cross-sectionally every three years since 1989 (and periodically supplemented with panel research), the SCF is the most comprehensive portrait of the financial condition of American families, including assets, liabilities, income – and their sources. In addition to being extraordinarily detailed, it has a sophisticated sampling methodology that takes great pains to ensure households across the economic spectrum are accurately represented.

However, as Arthur Kennickell, the Fed’s Chief of Microeconomic Surveys, points out, no accounting of income or wealth is perfect, and the SCF is no exception. Measures of inequality depend on how income, wealth, and inequality are defined, as well as on what data are collected. It’s important to be explicit about what’s being counted and how. In that regard, the SCF excludes the effect of taxes (and tax credits), non-wage benefits (like health insurance), non-realized capital gains, and defined benefit pensions.

These qualifications in mind, what I found surprised me. Whether or not you agree with Burkhauser’s perspective on inequality, the SCF shows that the last two decades have borne witness to an unmistakable trend in American society: there has been a major transfer of economic well-being from young families to older ones.

The intergenerational transfer of wealth has turned backwards. And I believe it is this trend – alongside the income distribution’s tendency to obscure it – that, in part, explains Burkhauser’s counterintuitive findings.

If, as Burkhauser suggests, the lowest quintile of the income distribution saw their incomes grow the most during the last two decades, this growth was not shared equally across the age distribution. While median family incomes for the bottom quintile increased by 25 percent between 1989 and 2010 (by the SCF measure), that of bottom quintile families headed by persons under 35 years of age decreased by 1 percent.

A similar story is true of younger families across the income spectrum. In the past twenty years, median family incomes of families headed by persons under 35 years old, persons aged 35–44, and persons aged 45–54 (across all income quintiles) saw their incomes decline by 2 percent, 13 percent, and 6 percent, respectively, while families headed by persons 55–64 years, 65–74 years, and 75+ saw their incomes rise by 25 percent, 52 percent, and 24 percent.

Figure 1.
Income and net worth by age
% change, 1989-2010

Source: Survey of Customer Finances

The story is even more skewed for wealth. Net worth declined broadly in real terms between 1989 and 2010 throughout the income distribution. However, the losses were most pronounced among young families. As a group, under-35-headed families had median wealth a third less in 2010 than they did in 1989. For families headed by 35–44 year-olds, median wealth declined by 56 percent; for families headed by 45–54 year-olds, it declined by 29 percent. But for families headed by 55–64 year-olds, wealth increased 8 percent. For families headed by persons 65–74 years-of-age, wealth increased 58 percent; for families over 75, wealth increased 77 percent.

In fact, if you look at the changes in median family wealth between 1989 and 2010, you’ll notice the growth of inequality across the age distribution has been markedly more than it has been across the income distribution. Families in the bottom three “age quintiles” (<35, 35–44, and 45–54) have seen their median wealth decline by more than families in the bottom three income quintiles (though, of course, there is some overlap). Similarly, families at the top of the age distribution have seen their wealth grow more than those at the top of the income distribution.

Figure 2.
Age inequality has outpaced income in equality
% change in median net worth, 1989-2010

Note: Age quintiles defined per 2010 population shares: <35 (20%), 35-44 (20%), 45-54(22%), 55-64(18%), 65+(20%).
Source: Survey of Customer Finances

While the share of wealth held by the top two income quintiles has remained relatively constant during the past two decades, the share of wealth held by the top two age quintiles has increased 20 percent, even after accounting for changes in the age distribution.

Figure 3.
It’s lonely at the top
Share of U.S. wealth, 1989 and 2010

Note: Top 2 age quintiles defined as 55-64 and 65+ (20%). Wealth shares expressed in 2010 constant-population terms.
Source: Survey of Customer Finances

Young families are falling behind both relatively (their parents’ wealth is growing faster than theirs) and absolutely (they have less wealth compared with the typical household than their parents did at the same age). In 1989, the median wealth ratio of “prime retiree” families (those 65–74) to “prime working” families (those 35–44) was 1.4—that is, typical prime retirees had $129,000 in net worth, compared to $95,000 for prime workers. But in 2010, the wealth ratio had jumped to 5: $207,000 for retirees versus $42,000 for workers.

Figure 4.
Relative and absolute wealth inequality are on the rise
Ratios of age group median wealth

Source: Survey of Customer Finances

Looking at it in absolute terms, families headed by 35–44 year-olds had median wealth slightly above the median for the population as a whole in 1989. Today their median wealth has plummeted to only half the value of the national median. During the same period, median wealth of households headed by 65–74 year-olds jumped to two-and-a-half times the national median.

This picture would be even more skewed were it not for the SCF’s idiosyncratic treatment of pensions. Traditional pensions, known as defined benefit plans, have been on the decline for some time and are disproportionately likely to be held by older persons. However, these plans are excluded from the SCF measure of wealth, thus understating the assets of older families. By contrast, younger workers are more likely to participate in defined contribution plans, like 401(k)s, and these are counted by the SCF.

Figure 5.
Contributing to measures of inequality
Pension participants by plan type (in 000’s), 1975-2010

Source: Department of Labor

The story of age-based comparative disadvantage is a familiar one to the under-40 crowd. The principle determinants of wealth accumulation—human and physical capital—are becoming more expensive. Like, way more expensive.

It doesn’t take a college education to point out the biggest problem is a college education. In today’s hypercompetitive information economy, nothing is more important for future earnings than human capital formation.

But rather than give grads a head start in the real world, college degrees are increasingly giving them a staggered one. Between 1989 and 2012, the average cost of a four-year degree increased 78 percent, to $92,000 (in 2011 dollars)—triple the rate of increase between 1969 and 1989. And the largest increases have come at public colleges, whose tuition has nearly tripled. This means that those students most impacted are very likely those least able to afford it.

Figure 6.
There’s no place like home (except college)
Trends in tuition, student loans, and homeownership

Source: Survey of Customer Finances

The inevitable consequence has been an explosion in student loan debt. Indeed, the story of college debt is largely a tale of the last decade. Between 2005 and 2012, total U.S. student loan debt tripled, from $363 billion to $1 trillion. Indeed, student debt is now larger than any form of debt besides mortgages—and it is the only category of debt to increase during the Great Recession. Some 38 million Americans have an average of $25,000 in student loans (2012 dollars).

Figure 7.
A mountain of debt
Total student loan debt, 2005-2012 ($ in billions)

Source: FRBNY Consumer Credit Panel/Equifax

The vast majority of student debtors are young: two-thirds is held by individuals under 40. Under-35 households today are more likely to have college loans than they are to own a home, a profound shift in investments from physical to human capital. The consequences borrowing young can be severe. Nearly one in three student borrowers in repayment status is more than 90 days delinquent.

Housing has become similarly unaffordable. Traditionally, homes have served as the primary asset for the middle class. However, between 1991 and 2013, housing prices have nearly doubled, according to the Federal Housing Finance Agency’s Housing Price Index. Since 1989, the median asking price for home sales increased from $60,000 to $140,000 (in 2013 dollars). It’s been great for people who own houses. Not so much for prospective buyers.

These increases would hardly be problematic if wages kept pace. But they have not—and the labor market has been especially brutal for younger workers. Between 1974 and 2011, median market income for 25–34 year-olds fell 7 percent in real terms. At the same time, older adults have been working longer and earning more. While typical families headed by 65–74 year-olds in 1989 earned only four-fifths as much as typical under-35 families, today the pattern has almost exactly inverted.

Figure 8.
Divergence of Fortunes
Median income, 1989-2010

Source: Survey of Customer Finances

Employment among young men has remained stubbornly at historic lows since the onset of the Great Recession. And the unemployment rate among 25–34 year-old women (7.9 percent) is actually worse than among 25–34 year-old men (7.2 percent). Both are significantly higher than the unemployment rate among 35–54 year-olds (6.4 percent) and those 55+ (5.3 percent). Once again, the juxtaposition between the fortunes of the young and old is stark: while the share of 25–34 year-olds employed has fallen by 6 percent in two decades, that of those older than 65 years has increased by 50 percent.

Figure 9.
Competition in the job market
Employment-population ratio, 1989-2012

Source: Bureau of Labor Statistics

As a result, young people are taking longer and longer to start their lives. A quarter of 20–34 year-olds live with their parents, up from 17 percent in the 1980s.

None of this is to say the older generation doesn’t have their share of problems. They do. Although wealth among older families is increasing, so is their cost of living. Chief among these expenses are medical bills. Average annual costs among Medicare enrollees increased 59 percent between 1992 and 2008, a rate exceeding the inflation rate used to calculate Social Security benefit increases by six percentage points. Economy-wide, health care costs have grown at a rate more than double the general level of inflation.

Figure 10.
The cost of living
Average annual health care costs for Medicare enrollees 65+

Note: in 2008 dollars
Source: Centers for Medicare and medicaid Services

Particularly problematic are prescription drugs, which represent seniors’ largest source of out-of-pocket medical expenditures. Such out-of-pocket drug spending increased 150 percent between 1992 and 2005, though it has dropped somewhat in recent years.

Figure 11.
Prescription for poverty
Average annual out-of-pocket prescription drug spending, Medicare enrollees 65+

Note: in 2008 dollars
Source: Centers for Medicare and medicaid Services

Further, although the decades-long appreciation in housing values has been an asset boon for seniors, this also means their primary source of wealth is highly illiquid. (To better understand the complex relationships between wealth, homeownership, and age, David Rosnick and Dean Baker’s research is a good place to start.) Roughly three-fifths of the wealth of senior homeowners is tied up in their houses. For the fifth of seniors who aren’t homeowners, the story is far worse: median net worth is just $5,600. Beyond Social Security, which comprises between 65 and 85 percent of income for the three lowest senior income quintiles, there are few solid financial flows for seniors; assets are a major source of income only for the very wealthiest. Higher living costs coupled with longer life expectancies is part of the reasons seniors are working longer and longer—and part of the reason younger workers are seeing fewer openings.

As a result, seniors are in a worse position that the official statistics would have you believe. According to the government’s official poverty measure, senior poverty is one of the success stories of the last half century. After falling to half its 1966 level by 1982, senior poverty has continued to decline, dropping to just 8.7 percent in 2011—the lowest of any age group.

Figure 12.
Poverty by age
Official poverty rate, 1966-2011

Source: Census Bureau, Current Population Survey

The official poverty measure, however, has numerous shortcomings, not the least of which is that it’s not so good at measuring poverty. According to the more nuanced supplemental poverty measure developed by the National Academy of Sciences in consultation with the Census Bureau—which, among other things, takes into account the cost of shelter and out-of-pocket medical care—the true senior poverty rate is 15 percent—double the official level.

Figure 13.
Defining Poverty
Official versus supplemental poverty measures by age, 2011

Source: Census Bureau

But on balance, older Americans are doing better than younger ones. Just as the top 1 percent is gobbling up an increasing share of America’s prosperity, so too are older households exerting outsized control over the nation’s financial resources. Partly of necessity, and partly thanks to the fortuitous timing of their life cycles, older Americans are working longer and living longer. As they do so, they are monopolizing jobs, wages, and real estate. All of this introduces a certain stickiness into America’s economic ladder, leaving less room for younger Americans to move up. Throw in increased global competitiveness and the decline of blue collar jobs and the upward climb becomes more difficult still.

And things are not looking up any time soon. As the baby boom generation ages, there will be fewer workers to support each retiree. Even those lucky enough to escape the morass of student debt, sluggish wages, and expensive housing will find their predecessor’s bequest bears suspicious resemblance to a bill. Consider, for example, that the amount of federal debt for each member of the working age population has increased nearly four-fold since 1980 in real terms, from $19,000 to $69,000. That bill will one day come due, it is more than likely today’s youth will pay for the excesses of their forefathers. For young Americans, it’s getting hot under the collar—and that’s without even discussing the consequences of global climate change.

So what does this have to do with Richard Burkhauser?

The SCF makes clear that all “low-income” families are not created equal. A retired millionaire living off a low-yield portfolio can fall toward the lower end of the income distribution, but this in no way means he is poor. That is a lot different from a single mom trying to raise two kids on a minimum wage job, while struggling to put herself through college. It’s also different than a thirty-something lawyer whose crushing debt makes aspiring to public service and aspiring to raising a family seem like incompatible goals.

The conclusion is inescapable: America is experiencing a major transfer of wealth from young to old. Inequality is on the rise.

On this point, NYU economist Edward Wolff—whose views on inequality are vastly different from Burkhauser’s—agrees. In his comprehensive analysis of wealth trends from 1962 to 2010, he found that, for young families, the usual hump-shaped pattern of life cycle wealth has become more of a plateau. For older Americans, it has become more of a steady ascent.

What can be done?

College costs are a logical place to start. We must get tuition growth under control while preserving affordable options—like public and community colleges. We also need to improve the ways students finance their studies. Adequately funding need- and merit-based grants, relaxing loan repayment rules, and capping interest rates would make college less crippling.

But perhaps the most important step is encouraging families to start saving early. The 529 college savings plans, which are operated by the states, offer major tax advantages and have few restrictions on contribution amounts, income, or where they are used. Overwhelming majorities of Americans express the intent to save for their children’s educations, yet only half actually do—and only a quarter avail themselves of 529 plans. Default enrollment for 401(k)s has dramatically increased retirement savings. Doing the same for college savings plans would make a lot of sense.

And while we’re at it, we ought to have a national conversation about “pretirement” benefits in general. American society does a pretty good job of supporting those deemed “deserving,” with programs such as Social Security and Medicare ensuring retired Americans can live out their golden years with dignity. Why not guarantee similar financial protections to young adults as they begin to tread the troubled waters of the real world?

We like to talk about America as a land of opportunity, a place of possibility, an engine of economic growth. We are troubled by our recent malaise and frightened by the prospect of rivals closing in on our heels. It seems strange, then, that we’ve become a country that systematically restrains our most productive assets right as they ripen. If youth are our future, we would do ourselves a favor by allowing them to pursue their own. Imagine the types of innovation and entrepreneurship that could be unleashed if only young Americans had the financial wherewithal. Imagine if young graduates took jobs not to pay the bills, but to pursue their dreams; if they worked not to make money, but to make their world a better place.

If we want to get serious about economic inequality, we need to understand that the old boys club really is the old boys (and girls) club. Diagnosing the intergenerational dimension of economic inequality is a step toward better remedies.

Previous generations of Americans made it to the moon. Perhaps we’re still capable of determining just who it is that revolves around whom.