Last week, the Government Accountability Office (GAO) issued a misleading report on school segregation, which I discussed with NAACP Legal Defense Fund President Sherrilyn Ifill and others on the Diane Rehm Show.

The takeaway line of the GAO report was:

From school years 2000-01 to 2013-14, the percentage of all K-12 public schools that had high percentages of poor and Black or Hispanic students grew from 9 to 16 percent.

(When the GAO referred to “poor” students, it was not really speaking of poor students, but rather of those from families with incomes less than nearly twice the poverty line and who are eligible for subsidized lunches in schools.)

Not by coincidence, the GAO report was released on Tuesday, May 17, the 62nd anniversary of the Supreme Court’s Brown v. Board of Education decision banning school segregation. So it was not unreasonable for those who did not read the GAO report very carefully to conclude that it described a dramatic increase in racial segregation over the last 13 years.

But it did not, and could not.

There is no reason to doubt the report’s 9 to 16 percent estimate of growth in schools that had high percentages of both poor and black or Hispanic students. But such growth can be due to an increase in schools that had high percentages of poor students, or high percentages of black students, or high percentages of Hispanic students—or all three. Because the report does not present any data disaggregated in this way, it is impossible to tell whether the growth from 9 to 16 percent was due to a growth in poverty or a growth in minority concentration.

Data from the National Center for Education Statistics tell us that changes in school racial concentrations over the period the GAO report examines were complicated. Black students were less likely in 2013 than in 2000 to attend schools that were overwhelmingly black. Hispanic students were just about as likely in 2013 as in 2000 to attend schools that were overwhelmingly Hispanic. Both black and Hispanic students were more likely in 2013 than in 2000 to attend schools that were overwhelmingly minority, but that’s because the Hispanic population grew in absolute terms quite rapidly.

In other words, if you define the segregation of black students as their isolation from whites, their segregation did increase from 2000 to 2013. But if you define it as their concentration with others of the same race, their segregation declined from 2000 to 2013.

While greater segregation played a role, it seems that the main culprit behind the GAO’s 9 to 16 percent increase in the number of schools with high percentages of poor black and Hispanic students was not a change in racial segregation, but a growth in poverty. In 2001, 10 percent of all U.S. families had incomes below the poverty line. In 2014, it was 13 percent. For black families, the poverty rate grew from 21 percent to 25 percent. For Hispanic families, it grew from 20 percent to 23 percent (see Table 2, here).

These trends are supported by data on the growth in the share of schoolchildren eligible for free or subsidized lunches, whose parents are near poor, with incomes below 185 percent of the poverty line. In 2001, 38 percent of all children were eligible—in 2013, it had soared to 52 percent.

Simply put, if the racial concentration of students had been entirely unchanged from 2001 to 2014, the share of schools that had high percentages of poor and black or Hispanic students would still have grown, solely because of a growth in family poverty or near-poverty.

The GAO report expressed great concern about the 9 to 16 percent growth of such schools, but it made no recommendations regarding the most obvious way to reduce that growth—reduce the poverty and raise incomes by improving the ability of parents to earn a decent living from the hard work they perform.

Also last week, as readers of this EPI blog know, the Department of Labor issued new regulations increasing the salary threshold under which all workers are eligible to be paid time-and-a-half for overtime, from $24,000 a year to just over $47,000 a year. Many black and Hispanic children have parents who fall into this category, and who will receive a substantial increase as a result. And, as it happens, the poverty line this year is also $24,000 a year (for a family of four) and the lunch subsidy eligibility cut off for low-income school children is $45,000.

If the new overtime rule had been issued a few years ago, fewer black and Hispanic children would be in low-income schools, not because those children would be in different schools, but because they, and their peers, would have been less likely to have parents with low incomes.

Likewise, if minimum wage increases now being discussed nationwide had taken effect a few years ago, fewer minority children would be in low-income schools, not because they would be in different schools but because their parents and those of their peers would have higher incomes.

Nothing I have written above is meant to minimize the importance of racial segregation in schools as a national crisis. The fact that African American children are more isolated from whites than ever is a matter of very serious concern. This is mostly a consequence of our increasing residential segregation, since children attend segregated schools primarily because those schools are located in segregated neighborhoods. This can be solved only by housing policies that aim at desegregation.

The GAO report had nothing to say about housing policy, either. Its policy concerns, all of which had merit, were directed at improving the curriculum and services that schools provide, such as ensuring that disciplinary policies do not target minority children and that schools serving minority and low income children offer the full range of academically challenging courses. But by ignoring both income and housing policy, the report misses a bigger picture.