Can CMOs Create Scale in Charter School Quality?

Private investors have put US$500 mln into funding Charter Management Organizations, but questions remain about whether quality can be scaled, or if, for instance, CMOs simply recreate the “pathologies” found in already existing public school districts. Robin Lake and Kevin Hall attempt to answer skeptics and address fundamental questions about the direction of education reform using CMOs.
This interview ran in Education Next and we are running it here for your review:

EN: How should we define a “quality” charter school? How does quality vary between those operated by CMOs and those that are not? What is the track record of CMOs to date, in terms of quality-conscious growth and replication?
Kevin Hall: At a minimum, a high-quality charter school produces a vast majority of students who meet or exceed academic standards regardless of their ethnic or socioeconomic background and who are well prepared for postsecondary success.
The Charter School Growth Fund invests in CMOs that operate networks of high-quality charter schools, providing grant and loan financing packages that enable these organizations to expand their capacity to serve more low-income and minority students. Over the past five years, CSGF has invested in more than 20 nonprofit charter-school operators. Among them are successful charter school networks across the country, such as Achievement First, YES Prep, KIPP, Rocketship Education, and IDEA Public Schools. In fall 2010, CSGF announced the launch of a new fund of $160 million to invest in the expansion of the best-performing charter schools and CMOs nationally over the next five years.
Approximately 95 percent of CSGF’s member schools enable students to outperform comparable district schools in both math and reading; nearly 70 percent of schools enable their students to outperform state averages in both math and reading, although they serve much higher than average percentages of low-income and minority students. Some of our CMO schools are beginning to close the achievement gap; their students perform better than affluent students who traditionally outperform low-income students by a significant margin. This is an extremely rare level of performance, particularly for organizations that run a number of schools.
While success stories reveal the potential of high-quality charter schools (and CMOs in particular), there are certainly many poorly performing charter schools across the country as well. It is important that those schools be closed in order to protect the integrity of the charter schools proposition: increased flexibility in exchange for performance accountability. As a balance to that strategy, however, there also needs to be an effort to expand the reach of the highest performers, particularly those that are able to scale their work to serve more students. Several of the CMOs in our portfolio are improving their performance as they get bigger, a historical rarity in the K–12 sector, though a phenomenon that is quite common in other sectors.
Robin Lake: The quality of any school, charter or not, has to be measured in terms of outcomes: are students better prepared for college, career, and citizenship than they would have been had they not attended that school? Rigorous research on charter school performance (studies that make true apples-to-apples comparisons) shows that there is tremendous variation nationally; charter schools often outperform traditional public schools, though not the majority of the time. When it comes to educating low-income students, however, charter schools do tend to outperform other public schools.
Many hope that by replicating high-performing schools CMOs will provide more consistent results than stand-alone charter schools have achieved, but there is no rigorous evidence yet to support that claim nationally. Many CMOs do seem to outperform their district schools, but there is also a lot of variation among CMOs and even within a particular CMO’s portfolio of schools. The CMOs we often point to as successes represent a very small portion of the 80 or so CMOs in the country. It’s not clear that the CMO model, as a rule, produces more consistent quality than does effective authorizing and oversight of “one-off” charter schools.
CMO founders are finding that large-scale replication with fidelity, especially at the high-school level, is a lot more difficult than they thought it would be. It’s also true that CMOs sometimes serve fewer special needs and ELL students and students with severe behavior challenges than their district counterparts, so achievement studies have to take that into account. The study we (CRPE) are conducting in partnership with Mathematica is the first nationwide apples-to-apples analysis of CMO effectiveness. “The National Study of Charter Management Organization (CMO) Effectiveness: Report on Interim Findings” is available on the CRPE website. A final report, with achievement results, is due out in 2011.
EN: How much private funding has gone into CMOs so far and have these investments delivered?
RL: Most estimates put the total philanthropic investment in CMOs at around $500 million, and most of the big foundations are no longer funding stand-alone charter school start-up. I don’t think anyone would disagree that some great things have come from those investments. Some CMOs are creating opportunities for low-income and minority students that people didn’t generally think were possible, and they have shown the results can be replicated. That’s a very important lesson.
But CMOs are growing slowly and exist in a very limited number of cities. The CMO model is typically highly centralized, with services akin to school districts. That model, so far, has produced new schools pretty slowly (the average CMO grows by one school a year), and many CMOs have built up very expensive central offices that could not exist without continued philanthropic support. Many of the well-known CMOs report annual spending of more than $13,000 per student. That level of spending may be necessary to serve our neediest students, but in no way represents an obvious cost saving over school districts and stand-alone charter schools. The CMO business model is so far proving impossible to sustain on public funding alone. A well-known CMO in California recently required an infusion of $700,000 in private funding to prevent financial collapse. More of these bail-outs may be on the horizon.
It’s hard to say exactly how many schools have been created from private money invested in CMOs, but our estimate is around 400. That means that more than 1 million private dollars have gone into each of the existing CMO schools so far (though some of that money is likely intended to fund central office systems to support future growth).
Could that half billion dollars have been spent in ways that produced a far greater number of high-quality charter schools? I think it’s likely. Diverse investments in innovative approaches to scale, such as back office and data management services, leadership training, and technology platforms for promising stand-alone charter schools, might have vastly increased the quality and number of new schools throughout the country. The important question now is how to make the most out of new federal dollars intended to support replication and scale. It would be a shame and a great waste of money not to be honest about, and learn from, the first generation of CMOs to create faster and more efficient paths to scale. Our report suggests a variety of ways that that can happen, including experimentation with smaller and leaner management organizations like those that are cropping up in New Orleans and New York City.
KH: CMOs have demonstrated that they can create high-quality schools at a fraction of the cost of traditional school districts, but like any growing organization they need capital to expand. Until CMOs can benefit from the billions of dollars of school bonds raised by districts, they will need “equity-like” investments from philanthropy in order to expand and effectively serve more students. The schools operated by CMOs often receive less overall public funding on a per-pupil basis than comparable district-run public schools, with the deficit ranging from approximately 10 to 30 percent. The deficit varies widely by geography, due to differences in the cost and funding of facilities as well as other factors. Many CMOs utilize a model where new schools operate with a deficit for two to four years until the schools reach full enrollment capacity, at which point the schools will generate operating “surpluses” at the site level.
To date, CSGF has committed more than $75 million to more than 20 emerging CMOs that represent more than 55,000 seats and are on track to exceed 100,000 seats. At that point our portfolio’s total enrollment would place it in the top 30 school districts by size in the nation.
CSGF’s goal as a financial investor is to enable organizations to reach sustainability on public revenues. When the schools reach capacity, they pay management fees to fund central administrative costs, such as academic coaches, student data systems, and payroll.
The scaling of high-performing CMOs provides one of the highest levels of return and leverage for philanthropic funds, particularly when you consider that CMOs tend to deliver much higher student achievement than the local district; these schools will continue to serve students in a high-quality way over time; and there are few investments in K–12 that have consistently yielded this level of performance.
The long-term funding solution is twofold: equitable funding and access to publicly financed school facilities. The federal government has a critical investment role to play in 1) supporting the replication and scale-up of the best providers through its grant programs; 2) improving access to low-cost public facilities for charter schools through its own funds and by leveraging existing public-school space; 3) pushing states and local districts toward more equitable funding systems for all public school students, including those in charter schools; and 4) supporting efforts to create early-stage, innovative, and scalable models that incorporate greater uses of learning technology.
EN: What is the largest number of high-quality charter schools you could see existing CMOs creating over the next five years? What’s the theory that envisions how these schools will have an impact on the larger system?
KH: Over the past 10 years, the total number of students attending schools run by high-performing CMOs increased at least tenfold, from approximately 10,000 students in 2000 to more than 100,000 students today. Over the next decade, the opportunity exists for CMOs to continue this pace of growth and serve more than 1 million students by 2020.
Growth in the CMO sector will come in three main areas: 1) Existing CMOs will continue to scale up. Most CMOs are adding one to five new schools per year as well as filling out their existing schools, and a few over the next several years may begin to expand to new regions. Currently, the CMOs in our portfolio are averaging annual growth rates of about 30 percent. 2) New CMOs will emerge from outstanding single schools, particularly in regions where many high-performing schools were launched over the past five to seven years. 3) Next-generation models showing promising early results will take root. Rocketship Education, a CMO based in northern California, uses a “hybrid” model that combines learning technology with great teaching to deliver outstanding results at much lower school site–level costs.
High-quality CMOs will set the bar for the entire K–12 sector when it comes to educating disadvantaged students. Over the next decade, several successful operators will serve a significant share of their local market (at least 10 to 15 percent of all students). More importantly, based on continuing their historic levels of performance, these schools could double the number of low-income students going on to college in these communities. Even though charter school enrollment is relatively small, we expect charter schools to dramatically increase the number of low-income students graduating from four-year colleges, and, in many cases, exceed the number of college graduates emerging from the much larger local district schools. This level of performance at scale will have a deep impact in those communities on the expectations for what schools can accomplish.
About half of the public school students in the U.S. attend schools in districts that have fewer than 10,000 students. Many CMOs that started from scratch over the last decade will grow to be among the largest 10 percent of districts in the country. Aggressive smaller districts may adopt some of the practices of these CMOs, including how they recruit, select, and develop talent; the culture they build in their schools; and the way they manage multiple schools, more effectively pairing accountability and decisionmaking rights throughout the school system.
RL: The CMOs we’ve surveyed (about half of those in the country) say they plan to create just over 300 new schools in the next five years. If we double or even triple that number, we are still nowhere near the number of schools needed to replace or transform the 13,000 chronically low-performing schools in the U.S. So while CMOs are replicating as quickly as they can and are becoming a significant presence in some cities, they are clearly still “a drop in the bucket” when it comes to large-scale public-school improvement.
Some have suggested that the next generation of CMOs will produce far greater numbers of schools, but CMO growth projections have historically been overly optimistic. What’s more, many U.S. cities have no hope of attracting CMOs: a large majority of CMOs are either committed to staying in a particular city or state or cannot operate in states with lower per-pupil funding. Few existing CMOs are interested in turning around existing schools, one of the highest priorities of Secretary Duncan.
Some CMOs believe that by flooding a particular district with high-quality schools and by providing proof that it is possible to close the achievement gap, they will prod entire school systems into changing. Unfortunately, so far only a few school districts in the country are responding to CMOs in this way. It seems clear, then, that if the charter sector hopes to contribute to transformational numbers in high-quality public schools, the current CMO approach alone can’t get it there.
EN: How big a challenge is the replication of high-quality schools, even by admired CMOs? What measures, whether in terms of practice or policy, could help CMOs succeed at delivering more high-quality schools at scale?
RL: Finding ways to replicate successful schools has stymied public education for decades. CMOs have so far scaled faster and seemingly with more consistency than any of the many failed dissemination and replication strategies of the past, including comprehensive school reform. But the challenge is still immense. Most CMOs say that facilities costs, inadequate per-pupil funding, and limited access to high-quality teachers and leaders are barriers to growth. All of those issues could be addressed by changes to local, state, and federal policies.
But CMOs also suffer from many self-inflicted problems as they scale: many are dealing with very high teacher turnover, increasing standardization and bureaucracy, and difficulty maintaining consistent quality, especially in their high-school models. Larger CMOs are beginning to look a lot like the very school districts charter schools wanted to escape. The expense of CMO central supports means that few CMOs have shown they can replicate without the massive philanthropic subsidies they have enjoyed in the past. In our survey, CMO leaders said that uncertainty about continued philanthropic funding is second only to limited access to facilities as their most significant barrier to scale.
These challenges are not reasons to dismiss the CMO model, but they do point out that CMO problems will not be solved with simply more public funding or access to public facilities. CMOs were meant to help charter schools capture economies of scale, given expected lower per-pupil funding relative to school districts. Large CMO models have not achieved those economies (though they are likely achieving other goals), and it is possible that other initiatives might be better able to capture economies of scale and still maintain high quality. Some possibilities: CMOs might spin off schools once they are stabilized or they might only offer very limited central office services. New technologies might make it possible for stand-alone charter schools to “plug in” to remote services that provide CMO-like supports, such as data analysis or real-time teacher coaching.
KH: We are beginning to see CMOs succeed in delivering consistent high-level quality across their networks. For example, Aspire Public Schools operated 25 schools last year and averaged a 9.5 (out of 10) similar-schools ranking, meaning their schools are averaging a ranking in the top 5 percent of schools in California serving similar demographics. Organizations like Aspire, Uncommon Schools, IDEA, and YES Prep are proving beyond a doubt that school systems can deliver high quality at scale. The main barriers to replication are indeed building a quality talent pipeline, particularly at the school–site level; ensuring full per-pupil funding follows the student; access to publicly financed facilities, one of the largest barriers, particularly for early-stage CMOs; patient, long-term focused growth capital sufficient to enable CMOs to build out high-performing networks; and the regulatory framework, as state and district-level structures and policies are often at odds with scaling up high-performing and promising new operators.
There has been good progress over the past year on the policy front in many states, and the Obama administration and Secretary Duncan have been working on smart ways of lowering the barriers outlined above. Recently, a few states and cities have been working to create environments where the best-performing CMOs might be able to expand and thrive.