Why Bull Markets Make the Racial Wealth Gap Worse

In 2016, for every $100 of wealth held by White American households, Black households held $5.52. It was the highest ratio in 27 years of Federal Reserve data. Then came the equity surge: the 2017–2019 bull run, the post-COVID rally and the AI boom of 2023 and 2024. By the end of 2025, the ratio was $3.96. Below where it stood in 1989 when the data begins. Lower than at any point since the dot-com era.

Before getting into any of that, the raw totals need context. There are roughly 80 million White households in the US and about 17 million Black households. White households outnumber Black by nearly five to one, so of course the aggregate wealth figures look lopsided. The more useful comparison is per-household. Dividing the Federal Reserve’s total wealth by those household counts gives you averages of about $1.8 million per White household and $345,000 per Black household in 2025. In 1989 those figures were roughly $247,000 and $76,000. On a per-household basis, White households had about three times the wealth of Black households in 1989. By 2025 that had grown to more than five to one.

The equity market did most of it. That is not an oversimplification. Equity markets rise, White household wealth accelerates, Black household wealth grows more slowly. The ratio narrows when markets fall and widens when they come back. It has done this twice in 36 years.

The asset that runs everything

In 1989, White households held 9.7 percent of their total assets in corporate equities and mutual funds. Black households held 1.8 percent.

The S&P 500 was around 350 in late 1989. By the end of 2025 it sat above 5,900. A dollar in the index at the start would be worth roughly twelve in real terms by the end. Households with meaningful equity stakes compounded on that for 36 years. Households more concentrated in real estate and pensions did fine too, just at the rate real estate and pensions grew.

By 2025, White households hold 31.6 percent of their assets in equities. Black households hold 5.2 percent. The gap grew from 7.9 percentage points to 26.4. Both groups became more equity-exposed over time, but White households got there first with more money and had three more decades to let it compound. That is most of it.

Two crashes and what they show

The 1990s give you a clean test. The S&P 500 roughly tripled between 1994 and 1999. Over that same stretch the Black-to-White ratio fell from $4.71 per $100 to $3.63. Not because Black households lost money, they gained, but White equity wealth grew far faster and the ratio measures relative position.

When the dot-com bubble burst and equity markets fell roughly in half between 2000 and 2002, the ratio bounced back to $4.46. White equity wealth contracted and the math improved for everyone else. Crashes close the gap. Bull markets open it.

The 2008 crash hit real estate before equities, so because Black wealth was already more concentrated in housing you might expect a harder immediate knock. What happened instead was that White equity losses were large enough to move the ratio upward through 2009 and 2010, and it kept improving through a decade of moderate recovery. By 2016 it reached $5.52, its highest point in the dataset. Then the equity surge of 2017–2025 dropped it back below its 1989 starting point.

Real estate and the ceiling it sets

Black household wealth has always leaned heavily on real estate. In 1989, real estate accounted for 32.8 percent of Black household assets. In 2025 it is 32.6 percent. The same share. For White households it dropped from 28.4 to 22.7 percent as equity ownership filled more of the balance sheet.

Housing appreciates. But it does not compound the way equities do. A house that doubles over 20 years stays doubled. A stock portfolio that doubles tends to keep going because the companies inside it keep earning and reinvesting. And the leverage in a mortgage cuts both ways: it amplifies gains when prices rise and amplifies losses when they fall.

2008 illustrated the downside. Subprime mortgages, variable-rate loans and balloon payments were not distributed randomly. They were concentrated in Black and Hispanic neighborhoods. Black homeownership rates fell from roughly 49 percent in 2004 to about 42 percent by 2012. The households that lost homes also lost years of appreciation on them, and that does not undo itself when prices eventually recover.

Where the 401(k) helped

There is one corner of that chart that runs in the other direction. DC pension accounts, 401(k)s and similar plans, now make up 13.6 percent of Black household assets, against 8.0 percent for White. In 1989 those figures were 5.6 and 4.4 percent. Both small, both grew, and Black households grew faster.

Automatic enrollment pushed Black workers into equities through their paychecks at a time when most were not buying stocks directly. It is not a big enough effect to move the overall ratio. Total Black DC wealth is $1.0 trillion against $12.7 trillion for White. But it is the one place in the data where Black and White household asset composition came closer together rather than further apart, and that is worth noting.

The Black-to-White wealth ratio peaked at $5.52 per $100 in 2016 and had fallen to $3.96 by 2025. Both moves had the same driver.

The "Other" category

The Federal Reserve’s wealth data has four racial categories. The fourth, labeled Other, covers Asian Americans, Native Americans and households of mixed or unreported backgrounds. In 1989 it held $765 billion in net worth. In 2025, $18.9 trillion. That is nearly 25 times larger in nominal terms, against about 8 times for White and 7 times for Black.

In 1989, Other households held 5.4 percent of their assets in equities. By 2025 that figure is 28.1 percent, nearly matching White at 31.6. For every $100 of White wealth in 1989, Other households held $4.08. By 2025 they hold $12.91. The only group to substantially close the wealth gap with White households in this dataset did so by converging on equity ownership, arriving at something close to White levels of stock concentration in about three decades.

Technology probably explains most of it. Asian American workers moved into software, medicine and finance at high rates from the late 1980s onward, and those fields pay a lot of compensation in stock. Options and RSUs accumulate for years before vesting, often in companies that appreciate sharply in the meantime. The Federal Reserve bundles a lot of different households into “Other,” so this is inference, not direct measurement. But a 28 percent equity share in 2025 is consistent with a population that spent decades receiving equity as a form of wages.

From 1989 to 2025, total US household wealth went from $20.8 trillion to $175 trillion. Most of that growth happened in equity markets. The households that owned equities captured most of it. The households concentrated in real estate and pensions captured less. That is pretty much the whole story. The Federal Reserve has been keeping score every quarter for 36 years, and the score does not change much depending on which administration is in office or what the policy conversation is. What changes the score is what happens to stock prices.