Wednesday, December 28, 2011

Who looks at the financials? Any financials?

I suspect that most people who complain about White Hat have never examined the financials of their own public school district.  This isn't really about disclosure, or accountability.

It's about competition.

Monday, December 12, 2011

How Much is Enough?

On December 9, 2011, Colorado State District Court judge Sheila Rappaport, in the case of Lobato v. State of Colorado, issued a court order declaring Colorado's funding system for public schools to be unconstitutional, going on to say:
"Evidence establishes that the finance system must be revised to assure that funding is rationally related to the actual costs of providing a thorough and uniform system of public education. It is also apparent that increased funding will be required."
The Court has given the Governor and the state legislature until the end of the 2012 legislative session to come up with a fix.

Without even attempting to read or understand this case (I'm not a lawyer), I'll assume that that the arguments in this Colorado case have little difference to those raised by DeRolph v. State of Ohio, which was filed almost exactly twenty years ago.

DeRolph was the school funding case that made it to the Ohio Supreme Court, which declared the public school funding system then in use to be unconstitutional. Most people mistakenly believe the Supreme Court declared the use of property taxes to fund schools to be unconstitutional, but that's not at all what they said. Regardless, DeRolph is now moot, as the General Assembly has enacted two  different funding systems since then, and neither has been test by a lawsuit.

It looks like Colorado might be the next state to go down this road. Lobato will undoubtedly be appealed. If it isn't, and their legislature doesn't satisfy the Court's order, it will still raise the constitutional question as to whether the legislative branch in their state can be ordered to do anything by the judicial branch.

Both states have similarly vague, and not very useful constitutional standards in regard to public education. Ohio's Constitution talks about a "thorough and efficient system of common schools"  (Article 6.02), while Colorado's says "thorough and uniform system of public education."

It will be interesting to watch the Colorado case as it progresses. Lots of people would like to know how much a public education should cost and how it should be funded.

So how would we go about figuring that out - what Judge Rappaport called "actual costs" - and where should the money come from?

It seems to me that we first have to go back to the fundamental fact that 80-90% of the money spent on public education (88% in our district) goes to pay for the compensation and benefits of the teachers, staff and administrators who are employed by the school district, and most of that is spent on the teachers.

I'm not saying that's a problem. As long as educational services are delivered primarily by teachers in a classroom, that's where as much of the money as possible should be spent. Everything else should be viewed as support for what happens in the classroom, or as "extras" above and beyond the State requirements.

But how much should teachers be compensated, and how many of them should we employ (in relationship to the number of students)?

The notion of Governor Strickland's "Evidence Based Model" funding approach was that there are clear answers to those questions. The EBM specified how many teachers there should be, and how much they should be paid - at least at a minimum. It also specified how many folks we should have in key administrative and support staff roles, and also what they should get paid.

Here is a copy of our district's PASS report, the document which shows how a school district was funded under the EBM approach. Page 2 of the report spells out how many people we should have in specific job categories. For example, lines 6-12 detail the minimum number of teachers we should have in various categories. The State would then give a district $56,902 of funding per teacher, adjusted by the "Education Challenge Factor," which is essentially another measure of how affluent a district might be. Our ECF was .98509, meaning that the State's contribution would be adjusted to  $56,902 x .98509 = $56,054.

This $56,054 is supposed to cover both compensation and benefits. In our district, the average teacher compensation is $69,369 plus 35% in benefits, or $93,648. So the EBM would cover only 60% of the cost of one of our teachers. The other 40% was our local contribution beyond the State funding.

Of course, Gov. Strickland's EBM was never fully funded, nor was the system it replaced. In the case of Hilliard Schools, and most school districts like ours, the changeover from one system to the next had virtually no impact on the amount of funding we received from the State, due to the effects of the so-called Transitional Aid Guarantee.

Any change in the school funding algorithm creates winners and losers. Since the objective of such a change is always to get more money to the poorest of Ohio's school districts, the winners are usually those districts while the losers are districts like ours. The Transitional Aid Guarantee came to be to prevent there being any such losers - a matter of political expediency rather than good policy.

So if 80-90% of the spending in a school district is on compensation and benefits, is there any data which helps us figure out how many folks we need on staff, and what we should pay them?

How about if we go back to the philosophy of the Empirical (Augenblick) Method, which was the approach used from the early 1990s until it was replaced by Strickland's OEBM in FY2010?  Only this time, let's use only the 81 districts rated as "Excellent with Distinction" in 2009 as the sample set.

The Augenblick Method is to list the per-student spending of the school districts in the sample set, throw out the lowest and highest 5% or so to eliminate outliers, then average the rest. With this algorithm and using 2010 per student spending from the CUPP report, we get $10,026/student as the benchmark number, .

In other words, we don't know how exactly a district should or would spend the money, but it seems like with $10,026/student, any district should be able to achieve a rating of "Excellent with Distinction."

Yet the actual range is quite large. Throwing out the 4 high and 4 low districts (5%), the per-student spending still ranges from $14,733 at Sycamore Schools (Hamilton County) to $7,949 in Wausean Schools (Fulton County). So why can't Sycamore Schools get the job done at $8,000/student/year like Wausean - close to half the cost?

As said before, it's all about how many people the school district employs, and what they are paid.

At Sycamore, the ratio of Regular Education Teachers to Pupils is 17/1. At Wausean, it's 21/1. The average Wausean teacher has 24% more students in the classroom than a Sycamore teacher, yet they both achieve the same rating. So do the Wausean teachers get paid more for achieving the same result while having larger classes?

The average teacher salary in Sycamore is $71,137, while in Wausean it's $55,668, meaning Wausean teachers have 24% more students in their classes than Sycamore teachers, but get paid 22% less. So it's not class size or teacher comp which determines the rating on the State Report Card.

How about teacher experience?  The CUPP report gives the percentage of teachers with 0-4 years of experience, 4-10 years, and 10 or more years. For Wausean, those statistics are 14%, 19% and 67%, respectively. For Sycamore, it's 14%, 17% and 69% - nearly identical, and both skewed to the higher experience band. That would certainly suggest that experience makes a difference in terms of results, but doesn't explain compensation differences between districts (Hilliard's numbers are 12%, 22% and 66%).

I should stop here and remind folks that correlations are not the same thing as "cause and effect," and that merits remembering here. There are within this dataset school districts which achieve Excellent with Distinction yet have fewer than 25% of their teachers with 10+ years of experience. Twinsburg for example achieves this rating while having 60% of its teachers with 0-4 years of experience.

So is there anything we measure which seems to be a strong predictor of outcome? I'd suggest that there is, at least within the constraints of the data published in the CUPP report. That predictor is the level of poverty in a school district. Among the 81 Excellent w/ Distinction districts, the average and median number of students living in poverty conditions is 22% (stdev=13). As poverty goes up, performance goes down, and it doesn't matter that much what the per-student spending might be.

For example, Columbus City Schools spends $14,904 per student, second in our area only to Grandview Heights, which spends just $74/student more. Yet the Performance Index for Columbus City Schools is only 80 (2009 data) vs Hilliard's 101.5, which we achieve while spending $3,400/student (23%) less. The difference: 80% of Columbus kids live in poverty, while 21% of ours do (I suspect that number is surprising large to many in our community).

So how do we determine how many folks we employ in our school district, and what we pay them?

The best answer I can come up with is that it's all a matter of local choice.

Across Ohio, teachers are paid according to what their local union negotiates with the local school board, and it seems clear to me that local school boards in more affluent communities tend to settle the labor contracts at higher salaries than do those in less affluent communities. That's about it.

So it behooves a young teacher to get hired in an affluent district like ours, or one of the many suburban districts around Ohio. Presumably those districts can be very selective in their hiring, so only the most promising young teachers get a job. In our district, new fulltime teachers typically serve as substitutes for a year or more before being hired. It's an effective way to give them a 'try-out' before being invited to join the team, and I've been impressed with the new crop every year when we get to meet them. It must be a valued opportunity, as we have literally thousands of applicants each year for dozens of openings.

Does that mean the tens of thousands of teachers in the less affluent districts are all duds, because they couldn't find jobs in districts like ours?  Certainly not!  Many things will cause a young teacher to live in a less affluent community: family, spouse's job, lifestyle preferences, etc. And there are districts that achieve Excellent w/ Distinction in spite of having very low teacher salaries. Bloomfield-Mespo (Trumbull) is such an example, with an average teacher salary of $37,751, less than the starting salary in our district of $38,362.

And the 'how many' part of the equation is a matter of local choice as well. We choose locally the overall student-teacher ratio. We decide locally the breadth of programming we wish to offer, and how those programs will be staffed.

You may respond that YOU don't get to make those decisions. But that's a cop out. You elect the school board members, and you should hold us accountable for making decisions consistent with your wishes, understanding that there will never be 100% agreement on anything about anything.

Your input makes a difference. Input from a lot of you makes a big difference, and can change the direction of important decisions. We've seen that a couple of times this year.

It will seem like we're entering a 'quiet' time right now. We have a workable agreement with the teachers and support staff through 2013. With your passage of the levy, we are committed to not putting another levy on the ballot before 2014 - provided the State doesn't hammer us with another significant funding cut.

Indeed we can take a little breather. But I feel strongly that we soon need to start the community education and communications effort to prepare us for 2013, when we'll next be negotiating with the teachers and support staff, and for 2014, when we'll be voting on the levy to fund that new contract.

The basics of school economics aren't difficult to understand, but we have lots of people to educate.