Start with a statistic: In 2016, the net worth of the median white family was almost 10 times higher than that of the median black family. This doleful ratio had improved slightly since the depths of the Great Recession a decade ago, but a chasm remains between the two groups.

If we want to do something about the gap, we need to understand the cause, not merely gape at the effect. Causal explanations come in two flavors: structural or cultural. What “nature or nurture” is to biology, “structure or culture” is to public policy: Is the problem caused by forces external to the community or by its internal behavior?

In a recent article for Quillette, Coleman Hughes, who is himself black, looks at the data on spending and opts for the cultural explanation. The “fact that blacks spent more on cars, jewelry, and clothes explained fully 20 percent of the total racial wealth gap,” writes Hughes, an undergraduate at Columbia University but already a thinker to be reckoned with. Turning to a 2015 study by the Federal Reserve Bank of St. Louis called “The Demographics of Wealth,” he sees that black people are dramatically less likely to engage in healthy financial behaviors such as “saving any amount of money, paying credit card bills on time, having a low debt-to-income ratio.” Income or education didn’t play a mediating role: When the researchers looked at educated middle-class families, Hughes writes, “the racial gap in financial health-scores didn’t shrink; it widened.”

But before endorsing this argument, we should ask whether structure or behavior are really opposed — are they, in this case, just two sides of the same coin? If you dig deeper into one of the papers Hughes cites, “Conspicuous Consumption and Race,” a 2007 National Bureau of Economic Research study by Kerwin Kofi Charles, Erik Hurst and Nikolai Roussanov, you’ll find the suggestion that while spending may be an individual behavior, it’s strongly conditioned by the cultural context in which black Americans must live.

It can be personally costly to be visibly associated with a minority that is stereotyped as poor. Casual discrimination against black Americans adds up: Store managers may follow them around, cabs may refuse to pick them up, and then there is the asymmetric threat they face from the police. If you’re visible and vulnerable to discrimination, it is rational to invest in goods, such as jewelry and luxury vehicles, that send a signal of affluence to strangers. That signal reduces the likelihood that you’ll be hassled when you leave the house.

And what about the Fed study that looked not at spending patterns but at broader financial behavior? People’s budgets are finite. The extra money spent to reduce the frictional racism in everyday life must come from somewhere, and where it comes from is the sort of spending that isn’t immediately visible to strangers, such as education and health, and, quite possibly, savings. And it’s not surprising that the St. Louis Fed found this effect was strongest in the middle class; the more affluent you are, the greater the need to differentiate yourself from members of your group with low average incomes, because you are more likely to be spending time in places where your race stands out.

Of course, that isn’t the end of the discussion. Devoting more of your income to visible consumption, at the expense of investments in human and financial capital, is going to make the wealth gap persistent – and thereby help feed the very stereotypes that people are trying to escape with the visible spending. There is a case to be made that, to break this vicious cycle, the current generation should refocus its spending, at some cost in increased discrimination. That’s the ultimate implication of what Hughes writes.

He asks what you would do if a loved one was routinely making bad financial decisions. “Would you withhold criticism from them for fear that you’d be ‘blaming the victim’? Would you feed this person story after story meant to confirm their belief that society had caused all their problems? Or would you view it as your duty – a duty born out of your love for that person – to point out their self-defeating behaviors and encourage them to make wiser decisions?”

Yet where does that leave those of us who don’t know what it’s like to endure the routine hassles minorities face, and are therefore in no position to argue that they’re making the wrong trade-off?

When structural biases lead to behavioral changes, changing one side may not have much effect. The problem isn’t going to be solved overnight, but the solution may begin with people like Hughes arguing for cultural change to their peers, while conservative-leaning white people tell their own side: You’re right that culture matters, but this aspect of black culture did not spring full-blown from the head of Zeus — it was a rational response to racism. America has essentially imposed a tax on its black citizens, and like all taxes, it distorts behavior, making wealth hard to build. Lowering the structural biases faced by black Americans by addressing our racism would essentially give them a tax cut. And that’s one cut that everybody, liberal and conservative, should be able to get behind.