Max Ehrenfreund, a writer at the Washington Post’s Financial Desk and Wonk Blog, recently wrote an article entitled: “White Kids are Winning San Francisco’s School Lottery, and the Data Proves It.”

The smugness of the headline aside, Ehrenfruend draws erroneous conclusions from the data he cites.

Ehrenfreund’s argument is as follows:

1. Unified enrollment systems are meant to give families equal access to schools.

2. However, white and (and wealthier) families are better able to navigate unified enrollment systems than minority (and poorer) families.

3. This results in white families getting into the best schools at higher rates than minority families.

4. To solve this problem we should abolish unified enrollment systems, build more affordable housing, and then create neighborhood attendance zones that draw in a socioeconomic diverse pool of public school parents.

5. If parents do no like the school they are assigned to, they can move to another part of town.

The argument has numerous flaws; specifically:

Poor Information Will Lead to Poor Choices

San Francisco does not publish uniform letter grades that accurately reflect school performance.

This lack of information is a dagger strike to equity. Barriers to information favor those with more resources.

In New Orleans, we place the letter grade of each school on the unified enrollment form. To select a higher-performing school, all a family needs to do is preference schools with an “A” or “B” rating. All families are quite capable of doing this.

Admittedly, our letter grade system is not perfect, and it should weight growth more than it does. But it’s directionally correct, and, as a result, academic performance is one of the strongest predictors of parent preference for poor, middle class, and wealthy families (poor families also strongly weigh other factors, such as availability of free after school services).

Before cities follow Ehrenfreund’s advice and get rid of unified enrollment systems, they should attempt some common sense fixes, such as providing easy to understand performance information to families.

Affordable Housing is Not a Near-Term Scalable Solution

Ehrenfruend admits that scaling affordable housing in San Francisco is a pipe dream.

He notes: “San Francisco’s famously restrictive zoning rules, however, have raised housing prices, making truly diverse neighborhood schools of this kind basically unimaginable today.”

I agree.

Moving is Not the Most Efficient Away to Get Better Public Service

Even if affordable housing could be built at scale, this would not be a reason to abandon unified enrollment

Ehrenfruend writes: “If parents were unhappy with a school, it would be easy for them to move to another part of town.”

This seems grossly incorrect.

First of all, I imagine that it is difficult for many citizens of San Francisco to move to other parts of town. For the wealthy, perhaps it is easy. But I doubt a family of limited means can simply scan Craiglist, find a good deal, and move.

Second of all, what if a family moves and then end up not liking their new school? It can be difficult to tell if a child will thrive in a school, and it might take multiple attempts to find a school that works. Should a family be forced to move each time?

Third, sometimes families have two children. Sometimes they have three children. Some of even have four. What if each of these children will thrive in a different school? Should a family buy four houses in San Francisco so they can secure attendance in each school?

There’s a general principal here that’s worth remembering: moving is a very high transactional cost.

It’s not equitable to force people to move to get access to high-quality public services.

In Sum

Unified enrollment systems are not perfect, but they are an improvement over connecting educational opportunity to zip code.

Common sense fixes, such as providing families easy to understand information on school performance, can mitigate many issues associated with unified enrollment systems.

One could go even further and alter unified enrollment systems to give preference to poor families in attending a district’s best schools; alternatively, a city could alter the system to ensure socio-economic diversity across its schools (every school could have ~40% of its slots reserved for free and reduced lunch students).

That being said, no fixes will make unified enrollment systems perfect; they will simply be better than any other alternative (that currently exist).

Unified enrollment will especially be better than solutions that have no chance of being implemented, such as scaling a massive affordable housing effort in the most expensive and regulated housing market in the nation.

Moreover, any solutions we attempt should be grounded in the realities that different children will thrive in different environments, and that moving is a very inefficient (and sometimes impossible) way to access better service.

A unified enrollment algorithm can incorporate a community’s values much more efficiently than residential planning.

Houses are tough to build. Moving is hard to do.

Tweaking an enrollment algorithm costs nothing; it can be adjusted over time; and, most of all, it is much more likely to deliver equity to families who need great educational opportunities the most.