Monthly Archives: October 2014

The High-Performing Charter Sector Will Not Scale

growh

How’s that for click bait.

Here’s my take on three biggest weakness of the high-performing charter sector, and why the high-perfoming charter sector won’t scale unless these weakness are overcome.

#1: Teacher Pipelines

As charter sectors scale, they need more teachers. A simple rule of thumb for sizing teacher need is: total students in charter schools divided by 18 times .25.

So let’s take a school system with 100,000 students. At 10% market share, the charter sector needs ~140 teachers a year.

At 90% market share, the charter sector needs ~1250 teachers a year.

That’s a big difference.

Right now, the pipelines many charters rely on (TFA, TNTP, residencies, etc.) have not demonstrated the potential to scale in a cost effective manner.

Charters will only be able to scale if other pipelines are developed.

My current opinion is that there is only one way to scale cost effective pipelines: train future teachers when they are in college and are willing to pay tuition.

If I were a funder, I would invest heavily in organizations that were attempting to develop educators while they are still in college.

#2 Family Organizing

Charter schools have a built in constituency: their families. Given all the political obstacles preventing charter school growth, organizing families seems like an obvious solution to scaling the sector.

Right now, most charter schools are terrible at empowering their families to be effective advocates for educational excellence and equity.

If this does not change, high-performing charters will grow at significantly slower rate.

#3 New School Incubation

Most of the nation’s highest performing charter management organizations started off in the same way: a great entrepreneur launched a single school.

Right now, far little too attention is paid to new operator creation. There’s only a couple of organizations that incubate schools at a national level; and city based incubators vary significantly in quality.

My current thinking is this: what TFA did for the status of teaching, we need to do for the status of launching charter schools. The nation’s most talented entrepreneurs should be launching charter schools. Imagine if 10-20% of the nation’s top business school students applied to charter incubators after graduation. Students in this country would be able to attend so many more awesome schools.

In Sum

It’s worth noting that ten years ago this list would have been much longer. The sector is maturing in numerous ways.

Moreover, while I think funders should invest in the above areas, they should definitely not stop investing in all the current supports that are enabling high-performing charter growth.

But, ultimately, if we don’t solve these three issues, the high-performing charter sector’s growth will stall, and millions of children will be denied the rich educational opportunities they deserve.

Lastly, drop a note in the comments if you think I’m missing any big rocks. I’m sure that I am.

Conflict note: I have worked, will work, or am working with organizations on all the above issues. You can take that as a sign that I really believe that these issues need to be solved, or that I’m making all this stuff up in an effort to divert funds to clients.

Should You Be Able to Choose Your National Government?

italy

A group of citizens in Sardinia is petitioning Rome. Their request: sell their island to Switzerland.

The group’s founder states: “The madness does not lie in putting forward this kind of suggestion. The madness lies in how things are now.”

I’m sympathetic to his argument. From all I can gather (mostly by reading – but also in a recent visit to Italy), Italy’s national government is pretty bad at governance.

This is not the first time something (sort of) like this has been put forth: Hong Kong is perhaps the most prominent example of a city being ruled by a foreign country; and the charter city movement, led by Paul Romer, seeks to construct similar (but voluntary) arrangements elsewhere.

Of course, selling actual land goes beyond time bound legal jurisdiction, but the general idea of “competitive governance” intrigues me.

In the world of education reform, charter schools being able to select their authorizers is probably the most analogous situation – though this relationship is between corporation and regulator, not citizen and government.

With the lens in mind, some thoughts (by someone who is not an expert and only spent 30 minutes mulling this over):

Pros of Being Able to Choose Your Own National Government

1. This could provide a good incentive for more effective government behavior: tax revenue is lost when your citizens leave, so you better govern them well.

2. Innovation could increase: when you have to compete for your customers, you tend to generate more / better ideas.

3. This could lead to more start-ups: governments could form with the explicit purpose of trying to gain citizens through exceptional governance.

4. Liberty might increase: citizens could have the power of exit as well as the power of the ballot box.

5. Tolerance might increase: the world might be a little more cosmopolitan if government wasn’t tied to geography; also minority populations would have an escape hatch from terrible majorities, which could temper the worst instincts of majorities.

Cons of Being Able to Choose Your Own National Government

 1. Launching a race to the bottom: for example, governments with lax pollution regulations might be attractive to a region that pollutes a lot, especially if the costs of this pollution are born by other regions.

2. This could lead to an increase in authoritarian governments: if governments constantly felt threatened by their populations leaving, they might seek to exert tighter control on the activities of their citizens, so as to suppress succession movements.

3. Increased global instability: historically, annexing and succeeding territory has led to much bloodshed; the fact that this is on its face a peaceful process doesn’t mean it would lead to peaceful outcomes.

In Sum

As with much in public policy, one of the basic tensions here is between innovation and stability.

In general, I think the main function of government is to prevent absolute disasters from happening. Human history is littered with violence, and stable, democratic governments seem to be decent at mitigating our most violent tendencies (relatively speaking).

So my instinct is to tread lightly.

The main counterargument to this line of thinking is that most democratic governments, while in many ways better than those of the past, are still pretty terrible when it comes to existential threats. We should be spending way more public resources on the global catastrophic risks covered in this

So perhaps the real threat is that if we don’t get better governments we’re all going to die anyways.

All in all, I guess I’m in favor of small scale pilots. Let’s experiment with charter cities and the like. If it works, let’s try to scale them.

One last thought: I once worked with the Tibetan Government In-exile on their democratic structures. The most fascinating part of their democracy: exiles could vote for the home government offices regardless of where they lived. And, for their Congress, it was proportional representation based on how many exiles lived in each region; i.e., Europe got X seats, North America Y seats, and so forth. And the government existed in the shadow (the light?) of their living God, the Dalai Lama, who held incredible sway of the prime minister.

All of which is to say: there’s probably a million ways to run a democracy other than the ways with which we are most familiar.

Speak to the Tribe

birds

I just read this Mark Suster on PR, which I got me thinking about the subject.

Here’s some additional advice that, I think, especially applies to public sector work:

1. Public Sector Communication is Mostly About Overcoming Tribal Affiliation, Not Information Gaps

If you are correct on the merits, and you are trying to convince someone who disagrees with you, it’s best to assume that this disagreement stems from tribal affiliation rather than data. Why? Because that’s how humans work, especially with regards to politics. People listen to your message as members of tribes, and if someone doesn’t agree with you it’s likely in part because their tribe doesn’t agree with you.

2. The Best Way to Overcome Tribal Affiliation is to Tell Stories in Their Tribe’s Language

Basically, you’re trying to convince people that you’re really part of their tribe, even if it might appear to them that you’re not. The best way to do this is to tell stories in their language. Politically speaking, if you’re talking to conservatives talk about the continuation of our country’s history; if you’re talking to liberals talk about fighting for the oppressed; if you’re talking to libertarians talk about freedom. The hard part about this is you have to be authentic: you have to tell your story in their language in a way that rings true, and, substantively speaking, is true.

3. When in Doubt, Use the Story -> Data -> Story Method

Start with a story, then deliver the data (simply!), then end with a story. To the extent the people in the audience care about the data, you will have given it to them. To the extent the people in the audience are like most people in most audiences, you’ll have at least told two stories in their language.

None to the is very novel, but I’m amazed at how tone deaf people are when they are speaking to tribes that aren’t their own. Pay attention to this stuff!

Is DC PCSB the Best Charter Authorizer in the Nation?

Unknown

A strong argument could be made. The case for the DC Public Charter School Board would look something like this:

1. There are many authorizers, working in cities such as Boston, Newark, and New Orleans, that have created charter sectors with strong effect sizes.

2. However, there are few authorizers that inherited a weak charter sector (due in part to poor district authorizing), and then turned the sector around to the point where the sector now achieves ~.1 effect sizes when compared to the traditional sector.

3. PCSB did exactly this.

4. And they did it at scale: achieving nearly 50% market share in a major urban area.

If you want to learn more, Dell just did a case study on PCSB. You can read about it here.

One last note: the following table was also in the report:

Screen Shot 2014-10-27 at 8.21.19 AM

This data is from 2012-13, so not sure how things have changed, but I was surprised to see that only 61% of district students are eligible for free and reduced lunch. In the aggregate, DCPS is socioeconomically diverse district.

My guess is that this, masks three distinct subgroups: schools with very high free and reduced lunch rates, schools with mixed socioeconomic rates, and schools that primarily serve the wealthy.

In many ways, each of these subgroups is probably a district unto itself.

I’d be curious to see DCPS performance broken out by schools with below 35% FRL; 35-75% FRL; and 76%+FRL.

Does anyone have this data?

Get Your Government Hands Off My School District

gov off

Sorry for the over the top title of this post, but you might recall the Tea Party protestor who launched the rally cry: “keep you government hands off my medicare.

The irony here, of course, is that medicare is a government program.

Recently, I was reading about the eight states where charter schools remain illegal.

They are: Alabama, Kentucky, Montana, Nebraska, North Dakota, South Dakota, Vermont, and West Virginia.

Interestingly enough, at every political map I looked at, each one of these states except for Vermont (blue) and West Virginia (purple leaning red) was tagged as a red state.

So what’s going on here? Why are red states protecting government monopolies?

Well, first of all, we might want to revise the conventional wisdom that republicans support charter schools while the democratic party is split on the issue.

Republicans seems split on the issue too.

Second, this seems to re-enforce the adage that politics is about identity not about policy. Many republicans have two identities: a national identity based on more abstract principles (freedom, markets, etc.), and a local identity based on real relationships. On certain issues, the local identity (I like my child’s teachers) will trump the national identity (government monopolies are terrible).

Third, no politician is immune to politics. Schools boards and superintendents are often very powerful political actors, especially in smaller towns, where the school district can often be the largest employer. So it’s not surprising to me that many of the holdout states have large rural populations.

I don’t really have much more to say on the issue. I just saw the state list and found it interesting.

Sentences to Ponder

ponder

1. Testing in India

“A crowd gathers: parents, neighbors, sometimes the whole village. Children jump up and down, shouting, ‘Test me! Test me!'”

Note: Testing, used correctly, is a civil rights tool.

2. Choosing schools at any income

“But, just because low-income families haven’t had to choose schools for their children in the past doesn’t mean that they couldn’t learn to do so in the future.”

Note: as I’ve argued before, letter grades can help.

3. A dozen thing learned from Chris Sacca about venture vapital

“It’s people with these broader life experiences who have balanced relationships who come up with the cool shit.”

4. Fathers, sons, and the presidency 

“To look back through the years is to see presidents in rebellion against their fathers and presidents in thrall to them, presidents trying to be bigger and better than the fathers who let them down (Abraham Lincoln, Ronald Reagan) as well as presidents living out the destinies that their fathers scripted for them (John F. Kennedy, William Howard Taft). It’s to behold the inevitably fraught father-son dynamic playing out on the gaudiest stages, with the most profound consequences.”

Note: I want to dismiss this all as nonsense, but what if there is some truth to it?

What Kind of Bonuses Should We Give Teachers?

bonus1

I’ve never been to attracted to teacher performance bonus schemes.

See below for why, as well as what type of bonuses I think would be much more effective.

When Performance Bonuses Work

Generally speaking, I think performance bonuses work best in two cases:

(1) Rewarding employees for executing redundant, low complexity tasks; or

(2) Providing significantly large bonuses (20%+ of salary) to provide major incentives for high-level employees in companies or divisions where enterprise performance is really just the aggregate of a lot of individual performance (sales, trading, etc.).

The above is based on a mix of research (books like Drive) and personal observation. Of the two, I’m more confident in using performance bonuses in the first situation (low complexity tasks) than I am in the second (large bonuses to star individual performers).

When Retention Bonuses Work

Retention bonuses reward both longevity (you usually have to stay with the org for 3+ years) as well as basic levels of performance (you only get the reward if you perform well enough to still be employed).

Retention bonuses are best used when you’re more focused on retainining high-perfomers than  trying to increase their performance through financial incentives.

What Kind of Bonuses Should We Give Teachers? 

Most teacher bonus programs are designed to give relatively small bonuses (below 10% of salary) for excellent performance on very complicated tasks.

This is the worst of both worlds.

If we are going to give teachers performance bonuses, they should be of a significant size so as to at least really tap into individuals wells of achievement, financial, and status seeking.

That being said, I’m skeptical this is the right way to approach bonuses, especially in the high-performing charter sector, where teacher retention is a much bigger issue than teacher performance.

There are probably more gains to be had in keeping people longer than attempting to inducing them to perform at even higher levels.

How Might You Structure Retention Bonuses? 

This blog is probably not the best place to develop sophisticated financial models, but the following should give you the gist of how a very simple model might work:

1. Let’s assume a beginning teacher makes 45K.

2. I’d cut this to 40K and then try to attract people based on a (presumably) strong culture + deferred compensation down the road.

3. I’d say: We’ll give you a 30K bonus if you stay with us for four years and in the last two years of your teaching your are in the top 50% of teachers in our evaluation system. If you’re below, that we’ll still give you a 15K.

4. Financially, this will likely be pretty close to a break even. For the teachers who get the full bonus, you’ve lost 10K (20K saved in compensation, 30K spent on retention bonus). For teachers who get the lesser bonus, you’ve gained 5K (20K saved in compensation, 15K spent on bonus). For teachers who leave after two years or so, you’ve saved 10K.

Why I Think This is a Good Idea

1. It solves the right problem: retention rather than performance. If high-performing charters got four years out of their best teachers the performance of the sector would likely increase.

2. 30K is a lot of money! That’s a good portion of a down payment for a house. I think that much money at the end of the tunnel would affect behavior, especially when a second year teacher is thinking: that’s just two more years, and then I get 30K.

3. The performance component will likely not people off (few people predict that they will be in the bottom 50%). It also both acts as insurance (paying low performers less) and cost saver (makes them model work much better).

4. You could also iterate on it in a bunch of ways: (1) only offer it to teachers after X years of performance (2) shorten or lengthen the investment period (3) increase or decrease the amount of the bonus. Different combinations might work in different markets / different organizational cultures.

All told, I think this would be a much better way to increase performance of teachers and schools.

But Maybe I’m Wrong

Let me know why in the comments. Or publicly lambast me on or Facebook, if you’d like.