Public Education - What You Can Do

Follow the Money

August 12, 2018

I was motivated to write this four-part series of columns on education because I love public schools. Our public education system still provides an excellent service if you are able to get into a well-functioning school. My wife and I are both very pleased with the results of sending our children to San Francisco’s public schools for all their elementary and middle school years so far.

Nevertheless, too many of our schools are failing too many of our kids.

In the first column, we learned that the primary problem with California’s public schools is that they compare poorly to those in other states, and in particular are failing to provide a good education to our neediest kids. We also learned that San Francisco is doing an especially bad job in this regard (notably in its eastside schools), calling into doubt how “liberal” or “progressive” we really are as a city when it comes to education. A secondary problem is that public school teachers (especially the younger ones whose future pension benefits are highly likely to be less generous than what more senior teachers have already accrued) are insufficiently compensated for an increasingly difficult teaching job.

In the second column, we learned that inadequate funding is one root cause of the poor state of public education. Proposition 13 de-funded our schools, so that over the last four decades California has gone from being on par with other states to being near the bottom of the pack in funding as a percentage of state income. Over the same time period, our school population has become needier (e.g. a higher percentage of students do not speak English as their first language), and so logically should have received more resources, not less. Also over the same time period, we made substantial pension promises that we failed to fund; our schools now face a “silent recession” where a greater share of each future dollar of funding will go to fulfill past promises made – which means a smaller share of each future dollar of funding will go to teach the students.

In the third column, we learned that bad school management is another root cause of the poor state of public education. California’s Education Code has ballooned six-fold over the last forty years and is now stifling and bureaucratic. The results are in from a dozen years of analyzing the experience in New Orleans after Hurricane Katrina: a combination of radically increased autonomy and accountability can substantially improve education outcomes in public schools. Autonomy empowers teachers and administrators on the ground to make the decisions they believe are best for students in their schools. Accountability ensures that schools demonstrate their approach is improving outcomes; it also allows low performing teachers and administrators to be removed when necessary and low performing schools to be closed when necessary.

What can each of us do to improve our system of public education? I suggest three things.

First and foremost, we must own the reality that our schools are not nearly good enough. California is failing. San Francisco is failing. The status quo is unacceptable. The fate of our children’s education is literally our future. A healthy, robust society prioritizes its future by investing sufficient resources and holding the people responsible for managing those resources to account. We must acknowledge these truths, internalize their significance, and motivate ourselves to action.

Second, we must move beyond false debates. For example, all too often the politics of public education devolves into a debate over being “for” or “against” teachers’ unions. But Massachusetts is a union state with one of the highest performing public education systems, while Mississippi is a right to work state with one of the lowest. There is nothing inherently good or bad about unions per se; what matters is: first, whether the needs of students are being met and second, whether the interests of all teachers (and other school employees) are being reasonably served as long as those teachers (and other school employees) are doing an adequate job.

Another false debate to jettison is whether “more money” or “school reform” is needed; in fact we must have both, or neither will succeed. The Massachusetts Public Reform Act of 1993 is a useful model for us to study; this grand compromise combined more funding that was progressively distributed in return for increased autonomy and accountability. California’s recently enacted Local Control Funding Formula is only a baby step in comparison, but at least it is a beginning.

Finally, I urge you to educate yourself about the candidates running for office and ask yourself who is likely to fight for the smart, radical changes we need. One race in particular to focus on this year is the State Superintendent of Public Instruction. This is not a race most people are not even aware of, let alone focused on, yet it matters a lot. Some of the things the State Superintendent can do is:

1.      Interpret the Local Control Funding Formula to allow poorer schools within districts to receive greater funding over time – and intervene aggressively with failing schools

2.      Work with districts to get multi-year “master waivers” from unnecessary regulation to empower them to innovate

3.      Build a data system to capture and share best teaching and administrative practices across the state (and create a governance structure to support its use and implementation)

4.      Use the public megaphone to advocate with the governor and legislators for the funding and reform our public schools need

The two candidates running this fall for State Superintendent of Public Instruction are Marshall Tuck and Tony Thurmond. When I published the original version of this column in the Examiner, I was asked not to formally endorse either candidate. But here, I can tell you that I wholeheartedly endorse Marshall Tuck. I have met several times with Marshall Tuck and he has greatly impressed me with his knowledge, his passion, and his ideas. And Marshall Tuck has the full-throated endorsement of Arne Duncan, who was US Education Secretary under President Obama.

I encourage you to learn about both Marshall Tuck and Tony Thurmond, so you can make up your own mind. But whatever you do, don’t sit this race out! Your engagement in November and beyond is critical for the future of our public schools.

Follow the Money

August 5, 2018

Last week we looked at the funding side of the root causes of the poor state of the public schools in California (and San Francisco). This week we look at school reform. Next week we will conclude by proposing an agenda for change.

Analyzing the need and the potential for school reform with analytic rigor is a challenge. It is one thing to describe the degree and character of bureaucratic bloat in the public school system, but how can one objectively quantify it? Even more challenging: How can one test and observe what changes to how schools are organized and managed produce better outcomes, given all the confounding variables in the real world?

One way to answer the first challenge is literally to measure the growth of the California Education Code from 1978 to 2018. The first photo shows that the California Education Code in 1978 was a mere two volumes, when there were 4.0 million students in California’s public schools. The second photo shows that the California Education Code in 2018 had ballooned to 12 equivalently sized volumes to serve the needs of 6.2 million students in the state’s public schools.

Was this six-fold expansion of the education code necessary to serve a student population that had only increased by 50% over the same time period? Or was it instead a function of the increased state oversight (versus local control) that occurred after Prop 13 led to the majority of school funding shifting from local property taxes to state income taxes? Regardless of the cause, I know from conversations with teachers, administrators and parents that the existing thicket of rules and regulations are widely perceived to be stifling and bureaucratic.

But perception, even when based in experience, is not always reality. We need a laboratory to test different ways of running schools. This is the motivation behind charter schools: allow new public schools to open and operate under a public charter with more freedom to set their rules, then observe over time which schools provide better student outcomes and – by extension – what methods and organization structures work best.

However, there is a key structural problem with the charter school test and learn approach: cherry picking. A charter school may end up attracting a non-representative sample of students (e.g. those who have either more personal initiative or a more supportive family environment), so that any positive effects the charter school produces might be mainly due to its student base. Furthermore, a charter school engaging in such cherry picking – even if wholly unintended – might destabilize the existing public schools by leaving the existing public schools with an even needier student base, and/or siphoning off badly needed public funding.

What we would like to analyze is a large-scale, well-tracked experiment with robust results that are free from the distortions and negative side effects of cherry picking. And fortunately, we have it in the post-Katrina New Orleans school reforms. The results of these reforms were written up in a July 15 New York Times column by Pulitzer Prize winning Opinion columnist David Leonhardt.

Here are the background facts, as reported by Leonhardt. After Hurricane Katrina devastated an already struggling community, “The state of Louisiana took over the school system in 2005, abolished the old bureaucracy and closed nearly every school.” Then, “Rather than running schools itself, the state became an overseer, hiring independent operators of public schools – that is, charter schools – and tracking their performance.” Crucially, because the changes operated over the entire school system, there was no cherry picking. An entire population of students who had previously been poorly served was transformed into a laboratory where different schools operated with substantially more freedom to set their rules – and then the results were then rigorously tracked.

We now have a dozen years of data from the experiment. (If you have the interest, I suggest reading the more detailed technical report in addition to the summary policy brief.) Performance has substantially improved. Not only are test results much better, but those results are persisting in post-test outcomes: high school graduation, college attendance and college graduation are all significantly higher.

Leonhardt visited New Orleans to interview “students, teachers, principals, community leaders and researchers” in order to understand the insights we should take away. Here is what he learned:

"People here point to two main forces driving the progress: autonomy and accountability.

"In other school districts, teachers and principals are subject to a thicket of rules, imposed by a central bureaucracy. In New Orleans, schools have far more control. They decide which extracurriculars to offer and what food to serve. Principals choose their teachers – and can let go of weak ones. Teachers, working together, often choose their curriculum. …

"Crucially, all of this autonomy comes with accountability: Schools must show their approach is working. They are evaluated based on test scores, including ACT and Advanced Placement, and graduation rate – with an emphasis on the trend lines. Schools that fail to make progress can lose their contract. Over the past decade, the district has replaced the operators of more than 40 schools in response to poor performance. … [The] research has found that much of the city’s progress has stemmed from closing the worst charter schools and letting successful charters expand."

In a follow-up column, David Leonhardt notes how poorly we are served by the typical debate between, on the one hand “staunch defenders [of school reform]” who “tend to be conservative” and “see market competition as a cure-all” versus on the other hand “the harshest critics of reform – who are largely progressive – [and] oppose nearly any alternative to traditional schools.” I agree wholeheartedly with Mr. Leonhardt. In our debates over education policy, we should never lose sight of two critical points: (1) Our current public education system is serving us inadequately, and is most especially letting down our neediest students. (2) If taxpayers will have to contribute more money to our public schools – and we will – then we have a responsibility to do everything possible to improve how our schools are run so that our higher taxes will be put to the best possible use.

Next week, I will summarize what we have learned from our inquiry and propose an agenda for change.

California Public Education, Part Two

Follow the Money

July 29, 2018

Last week we laid out the key problems with California (and San Francisco) public education. The primary problem is that educational outcomes are unacceptably poor, particularly among the neediest students. A secondary problem is that teachers – especially younger teachers – face a similarly tough work situation as teachers in many other states. We also noted that some people say our public schools need reform, while others say we have failed to fund public schools adequately, and both sides are right.

Now we analyze root causes. This week we look at funding. Next week we will look at school reform.

Although our focus is California, it is useful to start by comparing US spending to other nations, which you can find here (or Google “OECD public spending on education”). Compared to our peers (as of 2014, latest data available), the US is middle of the pack at 3.2% of GDP spent on “primary to post-secondary” education. (This is not quite the K-12 range we want to consider, as it misses kindergarten and captures college, but it is close enough for our purposes.) Many countries spend more, but some spend less and there is no clear correlation among countries regarding spending and outcomes.

Okay, so that’s how the US compares worldwide. How does California compare to other states? There is terrific data here (or Google “EdSource States in Motion”). The key facts are as follows:

1.      Before Prop 13 passed in 1978, back when California schools were among the best in the nation, California’s K-12 expenditures as a share of total state personal income dropped from over 4% (close to the national average) to roughly 3% (near the bottom), where it has remained since. Similar results are found comparing per student funding adjusted for the high cost of living in California.

2.      Over the same time period, California’s childhood poverty rate rose from well below the national average to somewhat above the national average, suggesting that our student base probably became harder to educate. (A key wonky point is that this childhood poverty rate does not reflect the higher cost of living in California, i.e. it is not double-counting point #1 above.)

So the funding argument goes like this: Before Prop 13 California had excellent educational results and funded schools adequately. Then after Prop 13 we de-funded the schools at the same time that a more diverse student population required more resources, and student outcomes dropped precipitously, especially for our neediest students. Hence, the problem (and the solution) is the level of funding.

This argument is correct insofar as it goes. But nothing in the argument addresses the need or potential for reform. Remember, the US spends an average amount on education for subpar results, and there are countries that spend the same amount or less for better results (e.g. Canada, Germany and Japan). We will address the issue of school reform in detail next week, but we can start to tease it out by turning to a key piece of school budgets we have so far overlooked: pensions (and healthcare benefits).

All of the numbers above count education expenses only as the cash is spent. That might seem sensible, but consider this: What if part of how someone is compensated (and by “someone” read “teachers”) is with promises that they will be paid more money in the future? In that case you must expense the promises when those promises are made, and raise sufficient taxes when those promises are made so that when it’s time to make good on the promises all the cash needed is there. If you don’t do that, then for many years it looks like you’re spending less money than you really are. Then one day it’s time to pay the cash for all those promises, and the next generation foots the bill the last generation rang up.

That, in a nutshell, is the situation in California’s school system. More money needs to be raised now and for many more years to come to pay the promises made for teacher pensions (and healthcare benefits) over the last generation, particularly during the Gray Davis administration.

(Before we dive into the numbers, one might ask whether California’s education spending really fell so much relative to the national average if we adjust for pensions promised to teachers. That question is too complicated to analyze here, but because most other states were similarly irresponsible, I believe adjusting for pension promises shouldn’t significantly alter the comparison of California to other states.)

The California State Teachers’ Retirement System (CalSTRS) runs the pension system for all California public teachers, including San Francisco. As of June 30, 2017, by their calculation, the unfunded pension liability was $107.3 billion, up from $22.5 billion ten years earlier, making it only 62.6% funded. (Retiree healthcare benefits add another $24 billion to the unfunded liability, according to the Legislative Analyst’s Office.) Unfortunately, although CalSTRS recently lowered the long-term investment return assumption to 7%, a more realistic rate is probably closer to 4%, which means the true value of the liability is significantly higher.

In 2013, the state legislature put in place a set of measures to start filling this gigantic hole, which you can read about in this outstanding article. (Or Google “Ed100 Lesson 3.11.”) Teacher contributions have increased from 8% of salary to 10.25%, while district and state funding (i.e. taxpayer contributions) will be increasing steadily for years. People in education policy call it the silent recession: more money is going to pay for past promises, leaving less for current services. For example, the latest budget assumptions for San Francisco Unified School District (SFUSD) has the employer contribution rate to CalSTRS rising from 14.43% to 18.13% over the next two years. (See page 31 of the SFUSD Second Interim Report for FY 2017-2018.) These changes mean that a greater share of each dollar of taxes will go to fund past promises, leaving less to educate today’s kids. Even on current assumptions, these changes may not be enough, as explained in this excellent story. (Or Google “CALmatters teacher pension debt.”) And if I’m right that the current liability is understated (because it should be discounted at a lower rate than 7%), then much more money than is currently forecast will eventually be needed to fill the hole.

Another key point is that the structure of teacher salary and pension benefits (shown in the graphs) misaligns incentives. 

Source: https://ed100.org/lessons/pensions; assumes the teacher hired at age 29, works without interruption until age 65, draws a pension until age 86

A teacher’s salary level and pension benefits are based solely on tenure instead of merit (being a great teacher) or difficulty (teaching needier students). And once teachers have been teaching continuously for a long time they are highly reliant on reaching beyond 26-27 years of teaching, both because that is when they earn most of their pension benefits and because unlike most workers they do not accrue future social security benefits during their career. The teacher’s union is therefore motivated to fight for teacher tenure above all else, taking away the promotion and firing authority a principal needs to ensure a school functions well for everyone.

The numbers show that we have insufficiently funded our schools for many years. At the same time, there is a powerful argument that it is necessary to reform our schools to ensure that any increased funding we need to provide is put to good use. Next week we will examine the case for school reform.

California Public Education, Part One

July 23, 2018

I’m passionate about public education. My wife and I were both educated in public schools, and our two kids go to public schools (Jefferson Elementary and A.P. Giannini Middle). My wife was President of the Jefferson Elementary PTA for two years and is currently Vice President. I started and continue to run an annual charity investor conference (see our website at https://excellencesf.org/), which has so far raised over $1.4 million that largely helps Bay Area organizations improve educational opportunities and life outcomes for underserved youth. When it comes to public schools, I’m all in.

So it makes me very sad to report that the public school systems of San Francisco and California are in poor shape, and have been for many years. There is no sugarcoating it. We are failing to fulfill our most basic responsibility to invest adequately in our future through our children’s education.

The primary problem with our public schools is that the educational outcomes are poor. To start with, the overall level of American education outcomes is unimpressive compared to other rich countries. According to the OECD’s “Programme for International Student Assessment” (PISA) 2015 Results in Focus, the United States scores roughly average in science and reading, and below average in mathematics. Specifically, we ranked #15 in reading, #16 in science, and #25 in math.

Yet even compared to a lackluster American benchmark, California falls short. Back in the 1960’s and 1970’s, California was widely regarded as having one of the nation’s best public school systems. Today we are subpar. Since 1990, national and state-level education performance has been tracked by the “nation’s report card,” administered by the National Assessment of Educational Progress (NAEP). You can read the results at www.nationsreportcard.gov. (For those reading this online, I like this page.) As you can see from the graph comparing 4^th and 8^th grade student reading and math proficiency, California has long lagged the national average and most other states. (Online readers can find the data here.)

There are some caveats and silver linings. Comparison between states is imperfect because states set their own rules for how the NAEP test is administered, e.g. Texas exempts English learners with fewer than four complete years of English instruction, while California administers the test to all students. (See https://edsource.org/2015/states-in-motion-school-finance-naep-child-poverty/83303, notes under the graph of 4^th and 8^th grade proficiency.) California has improved its abysmal performance from the 1990s and 2000s, and 2017 saw a welcome uptick in reading and math scores, most notably in grade four reading. Nevertheless, California’s performance is below average, and has been for many years.

While even the best public schools could improve – e.g. like many public school parents I would like more opportunities for gifted math students to learn at a faster pace – where our public schools truly fail is in serving our neediest kids. As the table shows, SF public schools significantly underperform in student proficiency for low income African American and Latino students compared to the statewide average for each subgroup (as well as the statewide average for all students).

Source: A Dream Deferred (https://reports.innovateschools.org/wp-content/uploads/2017/10/A-Dream-Deferred-Report-on-SF-Schools-.pdf), data from pp. 14-15

Another way to see the opportunity gap is geographically. The map (see page 31 of the linked report, “A Dream Deferred” by Innovate Public Schools) shows the “GreatSchools Rating” for San Francisco’s public schools. Look how the eastside/westside divide leaps out! And for those reading online, pages 32 and 32 of the linked report shows additional maps that highlight even more starkly the dearth of opportunities available to African American and Latino students.

Source: A Dream Deferred (https://reports.innovateschools.org/wp-content/uploads/2017/10/A-Dream-Deferred-Report-on-SF-Schools-.pdf), p. 31

Is San Francisco a liberal, progressive city? You wouldn’t know it from how we treat our students.

A secondary problem with our public schools is the working conditions for teachers and administrators. To be clear, the kids come first: schools exist to serve our students, parents and citizens. Still, I’m sure we all agree that competent teachers should be reasonably paid for their vitally important work.

You have probably read about teacher strikes in Oklahoma and West Virginia. Before adjusting for cost of living, the average annual teacher salary in California looks much higher by comparison: Oklahoma and West Virginia are roughly $45,000 and $46,000 respectively, versus nearly $73,000 in California. But once you adjust for California’s much higher cost of living, compensation in the three states are much more similar: Oklahoma and West Virginia are roughly $51,000/year and California is a little more than $57,000 (which ranks it 19^th out of the 50 states plus DC). These cost of living adjustments are tricky and imperfect. Still, in high cost California cities like LA and (especially!) San Francisco, it’s undeniably challenging to live a normal middle class life on a teacher’s salary.

(An enormous caveat is necessary here: a significant component of teacher compensation is earned via pension benefits, which are not counted in these salary numbers. We will revisit pensions in a later column. For now, here are a few key points to keep in mind. First, more senior teachers who have accrued a lot of pension benefits are in a much better situation than more junior teachers who will have to pay more into the system to receive less generous benefits. Second, while pensions can be a good tool, they can also lead to misaligned incentives if mismanaged. Third – and this will be no surprise if you have read previous Follow the Money columns – the pension liabilities are large and underfunded.)

California teachers also have very high class sizes. California is tied for first (with Arizona and Utah) for the highest student-teacher ratio at 24:1, substantially higher than #4 Nevada at 21:1. (The average student-teacher ratio is 16:1.) The research is indeterminate to what degree a difference of this magnitude is worse for students; what is clear, however, is that it entails more work for teachers, especially given the high percentage of students for whom English is not their native language.  

So let’s recap the problems. California public school educational outcomes are poor. The neediest students are hurt the most by this, especially here in San Francisco. And the teachers? When you adjust for cost of living, larger class size and the high proportion of English language learning students, California’s teachers (especially the younger ones whose future pension benefits are highly likely to be less generous than the ones more senior teachers have already accrued) face a similarly tough work situation as teachers in many other states.

The next post will look at root causes. Here’s a hint. Some people argue that our public schools need reform, while others say we have failed to fund our public schools adequately. Both sides are right.