Tag Archives: Governance

How Should We Govern Public School Systems?


A couple of impetuses for this post:

school district in Pennsylvania is considering contracting out all of their schools to one provider.

Robin Lake and friends have an Ed Next piece on why Detroit needs a portfolio manager to set performance standards for both traditional and charter schools.

Jay Greene thinks Robin and friends are socialists; he argues, parents, and not the government, are best positioned to drive quality over the long haul (as a side note, I found Jay’s argument pretty strange – to prove that Robin is wrong, he points to the Texas charter study, which showed the sector improved in part because the state portfolio managed the sector and closed ~50 schools!).

So what form of governance is best?

Here’s a rough taxonomy of how a community might govern their schools:

1. Solely Government Run: The government runs all of the schools. This is how most school district are governed.

2. Portfolio Model: A government entity puts every school operator on a performance contract, including district, charter, and voucher schools. The government is still allowed to run schools, it’s just not the sole provider. The portfolio manager performance manages the system: schools that perform well can expand; schools that do not must close.

3. Sole Provider Model: The government contracts out for the operation of its schools, but selects only one provider. It then holds this provider accountable for results.

4. Empowerment Model: It has been suggested that I rebrand Relinquishment as Empowerment, and I’m very open to this. So here you go. Under this model, the government is solely a regulator and schools are operated by non-governmental entities. Government performance manages the system: schools that perform well can expand; schools that do not must close. [Note: in many ways Empowerment is just a sub-strategy of the Portfolio principles.]

5. Demand Model: This is the same as the Empowerment Model, but instead of the government performance managing the schools, opening and closings are determined by parent demand. If you have kids enrolled to be financially viable, you can stay open. If not, you close.

Of course, one could come up with more variations on all of the above, but the aforementioned tease out most of the issues.

My rankings below:

1. Empowerment Model

The Empowerment Model ranks #1 for two reasons:

It’s more politically viable. The public is pragmatic: they will likely only support expansion of choice if we can show these options deliver better academic results.  The public may not wait for the “market to work it out.” Government performance management will likely lead to better short-term results.

It’s hard to transition from government run to non-government run in any sector. There’s numerous examples of deregulation going awry, including in education. More oversight at the outset may allow us to better build the regulatory regime we need for long-term success. Lastly, the model has been tried in New Orleans, and it’s working.

2. Demand Model 

I’m very sympathetic to the idea that parents, rather than government, are best positioned to drive school quality (in every aspect of the world). I think cities and states should experiment with this model.

But the fact is that the model has not adopted fully by any city or state in our country’s recent history, so we’ve don’t know how it’s implementation will unfold. My main worry is that it would be such a mess at the outset (in terms of performance and equity), that the public would turn against it. This, to some extent, occurred with Louisiana’s voucher program.

3. Portfolio Model

The only reason that Portfolio ranks below Demand is I get worried when the same government entity both regulates and operates schools. As practiced currently, most Portfolio districts do just this. Not surprisingly, there is generally pretty weak accountable over the traditional sector.

Granted, what Robin and her colleagues proposed in Ed Next is that a separate government entity (not the school district) be the portfolio manager. I think this would be much better, and it’s definitely worth trying. If this structure became the norm (and it gets early results in Detroit), I’d probably bump it to number two.

4,5 [Tie]: Government Run, Sole Provider

I’ve written a lot about why I worry about government run systems. So I’ll focus on my worries of the sole provider system. They are numerous and are as follows:

Instead of creating a local entrepreneurial market (deregulation), you’re simply contracting (outsourcing), which means that locally there is still one employer, one way of doing things, one type of option for kids.

The operation of schools is still highly politicized, as people would run on a platform of who should be the provider (instead of how to regulate the system).

There are high risks of corruption when one government entity is in charge of one large contract.

The national market for providers who would take on such a task is currently very weak; pragmatically I’d want to see more organizations that can do this well before I got too bullish on the idea.

That being said, it’s a tie for two reason: First, I think there are equally (if not more) problems with the government run model. Second, sole provider might be the only real non-governmental option for smaller districts with a few number of schools and a dearth of human capital. So it may be worth trying. I’d just go in eyes wide open.


All right – there’s my quick thoughts on governance.

But I’ll end by saying that the Empowerment, Demand, and Portfolio models are all just beginning to be implemented at any scale in our country. The world is a complicated place. Time will tell what works and what doesn’t. But I think all three should be tried a lot, as they all seem to have much greater potential to increase student achievement than do our existing structures.