In case you missed it, there’s a good twitter / blog debate going on about the efficacy of the portfolio model.
At the heart of the debate is an argument made by Jay Greene.
Jay argues that it’s very dangerous to have one entity regulate traditional, charter, and voucher schools; decide which school should be opened, closed; hold schools accountable via testing regimes; and oversee functions such as transportation.
I generally agree with Jay.
I don’t think one entity should solely possess all these functions (thought it’s unclear to me that the portfolio proponents believe this should be the case).
While Jay and I agree on the importance of not putting all the power in one entity, my guess is that Jay and I differ on how these functions should be distributed across government entities, non-governmental entities, and families.
At minimum, I believe four major regulatory buckets need to be assigned out: operational, equity, information, and accountability. With this taxonomy in mind, see below for my ideal regulatory state (with major risks identified).
1. Operational Regulation should be Dispersed Across Numerous Entities
The primary operational regulatory power involves making decisions regarding which organizations open which schools in which neighborhoods.
I think it is very dangerous to put all of this power in one entity; if this entity is captured, all new school development could shut down.
As such, I think districts, non-profits, and the state should all be able to authorize new schools.
Risk: A lack of coordination and shortsighted authorizing can lead to supply saturation. All told, I would much rather attempt to mitigate this risk than have to worry about a total freeze on entrepreneurship.
2. Equity Functions should be Held by Local Governments or Non-Profits
Primary equity functions include enrollment and expulsion.
For enrollment, local government should operate unified enrollment systems that provide families with fair, transparent, and equitable access to all schools.
If each school or regulator is operating its own enrollment system, I think the practical burdens of choice become high enough to raise serious equity issues.
For expulsion, too many externality issues arise if each school or regulator can determine when a student is expelled from school. As such, one entity should enforce a unified set of rules – acting as sort of a court system for a set of decentralized school operators.
I think school district should only assume these functions if it is not operating schools. If the district continues to run schools, then these duties should be a delegated to a third-party non-profit or a non-educational government entity.
Risk: The regulator uses its enrollment and expulsion functions as a backend means to micromanage schools. I think this risk is best mitigated by ensuring that the entity that holds these powers does not also operate schools.
3. Information should be Provided by the State and Non-Governmental Entities
The primary information function is to provide families with information that will support them in making educational choices.
At a minimum, states should provide accurate and easy to digest school performance information to families.
Ideally, the state will measure each school by considering mostly student growth data, and then assign letter grades to each school based on their performance on this achievement metric.
Of course, this doesn’t preclude non-governmental organizations from creating their own performance rankings. Many states have, at best, mediocre labeling systems, and I think there is plenty of room for innovation in providing information to families.
Risk: Families over utilize mediocre state systems because they carry the legitimacy of government. I’d rather mitigate this risk than be in a world where families have no information due to an immature or non-existent private information market.
4. Accountability should sit at Multiple Levels, Including Families
The primary accountability function is closing underperforming schools.
This power should be held by the state, local districts, non-governmental authorizers, and families.
Local districts and authorizers would be the first line of accountability, while the state could intervene if these entities failed to do so.
Additionally, in choice markets, families themselves can close schools by choosing not to enroll.
Risk: Government will close down schools for performance despite family demand. This is either a feature or bug depending on your policy views. My current thinking is that government should intervene in the state’s worst schools regardless of parent demand.
I’m in favor of redundancy in regulatory functions, as well as utilizing family choice as a check against poor regulation.
That being said, when equity and externality issues looms large, I’m in favor of unifying regulatory functions into on entity, so as to prevent a race to the bottom amongst regulators.
Lastly, while the aforementioned model is a far, far, far, more deregulated model than our existing systems, it is still pretty highly regulated.
I’m open to the idea that a much more deregulated market could work. That being said, I feel that proponents of such systems often do not pay enough attention to the equity and externality issues.
In the coming days, I’ll try to sketch out what a vision for a more deregulated system might look like.