Month: September 2018

Crystal Bridges in NW Arkansas; ID’d by Sue Haskins.

Don’t Ignore Teacher Retirement Benefits is the actual title of a recent article on the website of the Oklahoma Council of Public Affairs (OCPA) by fellow Curtis Shelton.  He suggests that Oklahoma teachers receive generous retirement benefits that should be taken into consideration when discussing teacher pay, implying they can be paid less salary because the state is providing such a nice retirement for them.  After providing some important background I’m going to rewrite his article.

I agree that retirement benefits, just like health insurance and other fringe benefits, need to be considered when comparing teacher compensation with other labor markets.  $50,000 a year with no benefits is not necessarily better than $45,000 a year with paid health insurance and a 2 for 1 employer match of your 401(k) retirement plan contributions.   We’d have to do the math.  So let’s do it for Oklahoma teachers’ retirement.

Contributions to the Oklahoma Teachers Retirement System (OTRS) are actuarially divided into two parts:  first, a “normal cost” percentage that is the current value to the employee for whom the contribution is being assessed, and second, the balance of the contribution which goes to retire or amortize the pension plan’s “unfunded actuarially accrued liability” (UAAL) that is simply catching up, or paying for, past commitments that were made to retired and active employees.  To more fully explain I’ll use the information from Table 1 and Table 5 of the 2017 OTRS Actuarial Report.

For 2017 total payroll for active OTRS members was $4,115,686.767 and the system received contributions totaling $998,158,208 which was 24.25% of the payroll.  Here’s a helpful summary:

Don’t be distracted by the deviations from 7% and 9.5% contribution rates for employees and employers, respectively; the reasons are not important.  What is important is that the state and federal grants are contributing at a combined rate of 7.34%, over $300 million.  Those “contributions” are really just revenue flows to pay off the systems $6.5 billion UAAL which, legally (Baker decision), is the state’s obligation.  The “Grant” amount is a kind of “in lieu” assessment the state is allowed to take from federal grant payroll to do the same thing.

The normal cost rate is 10.24% but I’m going to use 10.34% which includes all the system’s expenses.  That’s the actuarial calculation of the combined contribution rate from the employee (7%) and employer (3.34%) needed to sustain the benefits promised, going forward, for active employees’ current year of service.  3.34% of 2017 payroll is $137.5 million which, together with employee contributions of $293 million, is the annual revenue needed to sustain the system’s current benefit structure for current employees.  Another way to say it that might help is if the OTRS were beginning from scratch this year, i.e. no past promises, only future promises, then that’s the amount it would take to support the year’s promises.  If the annually calculated “normal cost” is faithfully paid in each year, that’s what it takes to sustain the system and keep the promises made.  That is the value to current teachers/employees for the current year—10.34% of which teachers are paying in 7%, so at best enhancing their stated compensation by 3.34%.

The rest of the story, as summarized by this chart, is that the balance of contributions and state direct revenues, almost $568 million, went to pay down the UAAL of the system, being legal and contractual promises already made and mostly for teachers already retired and which is on track to be fully funded in 17 years.  Those payments are not new compensation to current employees.

Simply stated then, if average teacher salary is now $45,000, it would be fair to account for the OTRS benefit paid by employers as adding 3.34% for an average salary of $46,500.  Here are other articles I’ve posted about this.  What follows is his article with my edits.

Don’t Ignore Teacher Retirement Benefits

Teacher compensation debates tend to focus on average salaries. A recent report from the American Enterprise Institute shows that just 3% of media coverage during 2018’s teacher walkouts quantified pension benefits.      That’s probably a good thing because if they’d relied on the AEI, the OCPA, or a similar “conservative” think tank in their area, they likely would receive misinformation that exaggerates the current cost of pension benefits.  Now, after reading articles about pensions on OCPAThinker.org, I know how to correctly quantify pension benefits.

While annual wages make up the largest portion of teacher compensation, they are not the whole picture. Examining data from the Oklahoma Teachers Retirement System (OTRS) allows for a more complete look at teacher compensation.

First, a comparison of contribution rates between the average 401(k) account (according to Vanguard’s How America Saves) and the OTRS aggregate rate shows a considerablelittle difference. For private 401(k) accounts, the combined employer and employee contribution rate was 10.3% of the employee’s annual salary. OTRS recipientsmembers contribute a rate of 7% while the actuarially determined employer rate that benefits current employees is 9.5%. This gives OTRS participants an aggregate contribution3.24%, making the combined rate, adjusted for system expenses, also 10.3%. The balance of 16.5% of the employee’s annual salary.employer contributions, about 6.2%, goes to pay for past unfunded pension promises and is not compensation for current employees.

Second, data from the Vanguard report and OCPA’s data tool allow for a loose comparison of the total benefits between OTRS benefits and 401(k) account holders. The average retirement age for OTRS recipients is 59.7 while the median potential benefit is $278,963. The median 401(k) account balance for retirees of a similar age (55-64) is $71,105.     It appears that the savings discipline and the advantages of pooling teachers’ contributions into a large, professionally managed, retirement system that can invest for the long-term, generates a much better retirement outcome for teachers than do the individualized, high cost, do-it-yourself plans imposed on private sector workers.

Leaving retirement benefits out of the conversation on teacher compensation in Oklahoma ignores an important selling point to those considering entering the cost to taxpayers.profession.  According to the fiscal year 2017 Comprehensive Annual Financial Report, taxpayers contributed $715705 million to Oklahoma’s teacher retirement system., of which $568 million went to fund the system’s unfunded liabilities for past promises.  In about 17 years those liabilities will be fully funded and that revenue can be used for other state priorities.  Just like discussing teacher shortages, having all the facts is crucial to finding the best way forward.

As always lunch is on me for the first to ID the photo location.

Subject is Picasso, on exhibit at Nelson-Atkins in KC; ID’d by Marianne Boshuizen.

The full picture:  401(k) plans vs. public pensions, a recent article by fellow Curtis Shelton at the Oklahoma Council of Public Affairs, illustrates only that he does not understand how pension plans work and are analyzed by actuaries.  Looking past his irrelevant swipe at retired Oklahomans who have faithfully served their state in many capacities, his announced effort to compare individual 401(k) plans to public pensions is a worthy task, but he isn’t up to it because he doesn’t know how to read a pension system actuarial report.  The relevant report is the 2017 OPERS Actuarial Valuation Report, the most recent, and available here for Mr. Shelton’s study.  If he wants to provide accurate information with sound analysis, he will do so before tackling the topic again; but he likely will not because the fellows at the OCPA rarely engage in thoughtful analysis since it distracts from espousing the predetermined script they’re hired to spout and echo about.

He uses a recent Vanguard study, How America Saves, for average data about individual 401(k) account contribution rates and performance.  I’m choosing to accept that data because what he reports is reasonable and Vanguard is my investment company of choice.  Here’s what he writes:

Starting with employee and employer contribution rates (measured as a percentage of the employee’s salary), you find significant differences between DC plans and the OPERS plan. According to the Vanguard analysis, the average DC employee deferral rate was 6.8% with an aggregate participant and employer contribution rate of 10.3%. The contribution rates for the OPERS plan must equal 20%. State employees have a contribution rate of 3.5%, which means the state contributes at a rate of 16.5%. Local government employees contribute between 3.5-8.5% and the employer then matches whatever the difference is to reach a combined 20% contribution rate.

Amusingly he goes on to ask, What do these numbers mean?, and then proceeds to demonstrate he hasn’t a clue.  Look at this table from page 6 of the OPERS report:

A perfect defined benefit pension system will always have assets (investments from past employee and employer contributions) equal to its accrued liability (how much is needed to assure payment of all future benefits previously earned by employees; better definitions are in the report).  OPERS as of July 1, 2017 was pretty damn close, lacking only $540 million against a need for $9.8 billion.  A perfect board of directors of a perfect system will always have a plan for closing that gap; here’s the plan from page 22 of the Report:

Translated so perhaps Mr. Shelton can understand, the board wants the Plan fully funded by 2027, just nine more years from now.  What that further means is that not every dollar being contributed is going to the benefit of the employees who are now actively paying in because part of the contributions are paying for the future benefits of former employees who are now retired, drawing benefits, but not making contributions.  Some of the unfunded liability may also be attributed to past promises to current employees that were not funded.  It’s my understanding that mostly the past sins had to do with COLA’s (cost of living increases) being granted without increasing the contributions.  Regardless, it is clearly stated in every actuarial report for a pension system that is not fully funded how much of contributions is going to fund the future benefits of current contributing employees and how much is going to pay off what is owed for past promises, largely for those already retired.  And here it is from page 7 of the Report:

None of the fellows at the OCPA have any business writing about Oklahoma’s public pensions until they understand what “Normal Cost” means.  I didn’t know until former Executive Director of OTRS James Wilbanks explained it to a group of school district CFO’s several years ago.  I’ve written about its importance for policy analysis in three earlier posts.  Simply stated it is the contribution, as a percentage of compensation, that is required to fully fund the additional pension benefit earned by all active employees for the valuation year, i.e. it’s the contribution rate needed to pay for the pension benefits promised for working the new year.  For OPERS that rate is 10.24%.  Combine that with the 0.40% above for “budgeted expenses” and you get 10.64% which I believe is the “apples to apples” comparable number to Vanguard’s reported 10.3% referenced by Shelton, not the 20% he uses.

He is correct that 20% is being paid in for each active employee, but to state, as he clearly does, that the 20% inures entirely to the benefit of the contributing employees is false as demonstrated above.  The Table uses a blended 4.14% employee rate, rather than 3.5%, which is consistent with Shelton’s note that employee rates vary from 3.5% to 8.5%, so its total is actually 20.64%.  Here’s how the Table divvies it up.  Normal Cost (the value to current employees) is 10.64% which is paid 4.14% by employees and 6.5% from employers.  The remaining 10% is going to pay off the UAAL, unfunded liability, which is a legal obligation of OPERS and the State regardless of whether there are active employees or not.  That 10% is clearly shown as the 3.5% listed which is estimated to eliminate the UAAL in the remaining ten-year target set by the board, and the 6.5% “surplus” at the bottom which is the amount of “overpayment” into the system meaning it’s not going to take anywhere near 10 years to fully fund.

I didn’t readily see an estimate in the Report but looks to me like it will be fully funded in less than half that time, maybe in four years around 2022.  After that happens OPERS could rock along with its existing benefits structure keeping employee contributions at their current levels and reducing employer contributions 10 percentage points to 6.5% which would free up about $170 million or so for other state priorities.  The simple math is four more years at $170 million each is enough to fund the $540 million UAAL; but simple math is outside the reach of the fellows at the OCPA.  They’d rather have you believe it takes a 20% contribution to fund today’s OPERS benefits when in fact it takes only a little over half of that.

The remainder of Mr. Shelton’s article is pretty worthless since it is based on a wholly inaccurate foundation.  The state pension benefit he claims has a value at retirement age 62 of $323,200 to a private sector worker does not require a combined contribution rate of 20% of that state worker’s compensation; it only requires 10.64%.  And he tells us the state worker’s private sector counterpart with a slightly smaller contribution rate of 10.3% ends up with a retirement nest egg at age 62 of only $71,105.  If I could follow his logic we’d probably find that’s not apples to apples either, but it’s kind of fun because it looks like to me the private sector pension world could learn a thing or two from the OPERS and other public systems.  More on that shortly, but one more demonstration of how Stink Tanks like the OCPA operate.  Here’s how Shelton finishes:

Oklahoma taxpayers paid more than $1.2 billion into public pension programs in fiscal year 2017. In all, Oklahoma public pension programs received $1.7 billion in contributions, with $443 million coming from employee contributions. Oklahoma legislators have done a good job ensuring these pension liabilities are adequately funded. According to a Tax Foundation report, Oklahoma’s public pensions had a funding ratio of 72% in fiscal year 2016. This ranked Oklahoma 20^th among all 50 states. Having a clear and full picture of public employee compensation is important when debating government employee pay and the tax burden on Oklahomans. 

I’m being a little picky but why does he go to the Tax Foundation for dated information when the Oklahoma State Pension Commission’s 2018 Actuarial Report was published six months ago and shows our state plans had a funding ratio of 78.6% in fiscal year 2017.  Is it because it makes government look a little worse?

By passing the Oklahoma Pension Legislation Actuarial Analysis Act (OPLAAA) the legislature did place our state’s public pensions on the path to full funding by stopping the practice of approving unfunded COLA’s (like Senator Inhofe does with Social Security).  At the same time the legislature moved new state employees out of OPERS assuring that many of our future public servants will face the same underperforming, high cost, individualized retirement plans that are failing workers in the private sector as Shelton’s corrected analysis above demonstrates.  That move was fully supported and advocated by the OCPA.  Here’s a comment from the OPERS Actuarial Report, page 6:

I think this comment is preparing the OPERS Board for the inevitable higher cost per benefit dollar of operating a pension plan that is going to steadily dwindle in size.  As the paper If Ain’t Broke, Don’t Break It published by the Oklahoma Policy Institute during the 2014 legislative debacle that ended OPERS for new state employees pointed out, a collective pension plan has advantages of higher rates of return, lower transaction and administrative costs, and greater annuity payouts than individualized, do-it-yourself, 401(k) style plans advocated by Stink Tanks like the OCPA.  While OPERS will continue on, it’s smaller size will reduce its efficiency meaning taxpayers pay more per dollar of pension benefit realized by our state workers.  Shelton’s $71,105 to $323,200 comparison above, rightly viewed, illustrates the result of pooling retirement contributions into a huge collective plan that can invest for the long term and in asset classes not available to individuals (think limited partnerships, commercial real estate and private stocks) for higher returns, demand lower transaction and administrative costs, and amortize higher benefit payouts.  Those are the advantages the OCPA pushed the legislature to throw out the window.

What is left is an opportunity for a real research group, or maybe the State’s Pension Commission, to test what I just said and plagiarized from If It Ain’t Broke, Don’t Break It.   New state employees are being forced into individualized plans where they have the “freedom” to participate in the “competitive markets” of the investment world, i.e. they get to be overcharged and misled by professionals who are compensated not based on performance, but based on often misunderstood fee structures.   I hope not, but it does happen.  Even if every state employee is honestly advised, still they will struggle to match the higher returns and lower costs inherent in a large, collective pension plan.  So Mr. Shelton, why don’t you figure out a way (not holding my breath) to measure over the coming decades just how well new state employees are faring with their retirement accounts.  It’s a state plan and the collective information is available.  I’ll buy you dinner each year their returns are greater, adjusted for annuity costs, than the Oklahoma Teachers Retirement System’s, if you’ll buy me dinner each year they are not.  Real research is on my side.

And out of the exercise might come a meaningful idea that I first heard from James Wilbanks years ago, namely that Oklahoma could have a very efficient, high performing, low cost retirement system for public employees that is operated as a Defined Contribution plan, but retains the significant benefits of a large, collective plan.  Maybe even ordinary, private sector citizens could participate.  Oh, but that wouldn’t serve the interests of the “investment advisers” who populate our legislature or members of the OCPA board who fear large, efficient public entities making investment decisions.

Unless Shelton corrects his more recent article Don’t ignore teacher retirement benefits soon, next time I’ll be forced to provide another lesson in Normal Cost demonstrating more clearly how the Limited Thinkers at the OCPA continue to fall short of a reasonable standard of competence.

As always lunch is on me for the first to ID the photo location or the subject of the “half picture” displayed.

 

 

 

Pete’s Place Restaurant in Krebs, OK ID’d by David Tinker

I had not looked at the Oklahoma Council of Public Affairs website for a couple of weeks while preparing for a presentation on school finance so was unaware they had again ventured into the arena of Oklahoma’s public pension systems with not one, not two, but three new posts.  This subject is near and dear to my heart because I receive an Oklahoma public pension system monthly retirement payment and because the OCPA’s past distorted advocacy for eliminating state public pensions is why I began vetting their so-called “research”, thereby discovering how they consistently distort or make up facts and use faulty reasoning to support the conclusions their funders and readers already believe.  These three posts are no exception and its amazingly easy to expose significant flaws in facts and analysis.

I’ll begin with the first which is Retirement Provision Boosts Superintendents’ Pensions by their President Jonathan Small.  It should be titled We Hope You Don’t Understand Compound Interest So You’ll Think You’re Being Ripped Off By Superintendents.  I’m not sure but possibly his article is the debut of another Data Tool.  Unlike their last Dat A Tool? I reviewed, this one appears to have accurate information, though I’m not sure for what public policy purpose.  He leads off by telling us that retired Union Public School superintendent Cathy Burden is expected to draw over $3.1 million during retirement but only paid in $342,000, the current employee contribution rate is 7%, while conceding that “the districts that employed her contributed somewhat more.”  The other facts he shares are that she retired at age 65, worked 41 years, and had an ending salary of about $270,000.  So how does $342,000 become $3.1 million unless she’s ripping us off?

First, in recent years as I’ve explained before and will again in my next post, much more than “somewhat more” is being paid in by school districts AND the state being about 17% or so in recent years.  So the state/employer “match” for her $342,000 is way more than double that amount, being over $47,000 just her last year.

Second, it doesn’t take $3,100,000 to provide her with 21 years (her life expectancy at 65) of benefits; at the OTRS current expected 7.5% rate of return the system’s trust fund needed about $2.16 million to secure her future payments when she retired, still a lot, but much less than the $3.1 million eventually paid out.  That’s how compound interest works.  If you need a dollar next year and can earn 7.5% investing, you only need 93 cents today; but if you can wait 21 years for that dollar, 24 cents today will get you there.

Third, her $342,000 and “somewhat more” than that in state/employer match has been paid into TRS and invested for over 41 years.  Without replicating her exact work history here’s a plausible scenario that illustrates how steady investing along the way can get to that $2.16 million.  I’m not sure but believe she and I are both TPS graduates from the sixties a year apart so she retired in 2014 and may have begun contributing in 1974.  If contributions by her, her district and the state were only $1500 (on about $10,000 income) that year then 41 years later, at 10% annual growth in her compensation as she earned her way up the ladder, her final total contribution is $67,900 which is darn close to the 24% share of her final salary that was collected by TRS from her, Union and the state that last year.  The total paid into the TRS trust fund on her behalf under this scenario over the 41 years is $732,000, including her $342,000 with the difference being Mr. Small’s “somewhat more” ($390,000) paid by her employers and the state.

Fourth, that $732,000 invested as it came in (nothing needs to be paid out) over the 41-year period at the long-time TRS realized rate of 8% yields a starting nest egg for her retirement of just over $2.13 million which is very close to our target $2.16 million (besides Oklahoma female life expectancy is falling).  That, Mr. Small, is how $342,000 becomes $3.1 million over a lifetime with promises kept and prudent investments made.  Here’s my calculations.

What has Mr. Small in a jazz is that the TRS retirement formula for us old timers is based on our highest three years compensation.  As he notes that has already been changed to five years but he seems to pine for the Social Security standard of highest 35 years.  He puts it like this:

OTRS members are awarded two percent of their average final salaries for each year of service. But unlike Social Security, which uses the average of one’s best 35 annual salaries to compute pension payouts, OTRS uses the top three years for Rule of 80 participants and the top five years for those under the Rule of 90.

There is an important fact here that Mr. Small omits, either because he is uninformed or because, wrongly, he doesn’t find it relevant.  Those 35 years of Social Security wages are indexed to inflation.  Something the OCPA loves to do when it fits their narrative that schools have too much money (a frank distortion) but conveniently omit if it shines a light on low teacher pay.  Here’s the Social Security calculation sheet.

According to my SSA earnings record my 1974 tax year compensation was $9,965 working as a classroom teacher for Tulsa Public Schools with a Masters and going from 4 to 5 years of experience.  The SSA inflation multiplier is 6.06 which results in that 1974 salary being $60,388 for calculating Social Security retirement in 2018.  I note that by the just improved Oklahoma Minimum Salary Schedule a classroom teacher with a Masters at 4 and 5 years of experience will be paid $39,728 moving to $40,200.  On the TPS current schedule that same teacher fares a bit better at $41,004 moving to $41,476.  So, Mr. Small, bring it on!

In effect the TRS, and other defined benefit plans, final three, now five, years highest salary is a simple proxy for compensation growth and inflation that is admittedly less precise and that can favor late bloomers over steady earners as Mr. Small alleges.  But he makes no serious effort to discern the actual impact because that would require real thought and research which seems beyond the capacity of the fellows at the OCPA.  Instead he simply scans the data for shocking numbers to cite, spanning over 60 years of time, without any attribution to the investment earnings that largely, if not entirely, close his implied “rip-off” gap.  Then he offers up Social Security’s methodology as being more fair without bothering to mention that it relies on inflation indexing that is absent from the TRS formula (and benefits!).

A personal note.  He also cites my former superintendent Lloyd Snow’s pension benefit as presumably another example of this shocking rip-off he’s discovered.  Lloyd Snow served the children of this state for 43 years, working every day with passion, energy and integrity.  39 of those years he served as superintendent so he was no late bloomer but was a steady performer making a difference for each school district he served.  If he made a mistake he would own up to it and try to correct it.  If he had employees who did shoddy work below recognized professional standards he worked to correct them.  If Mr. Small would do the same in his leadership role at the OCPA maybe it would perform useful, fact-based public policy analysis instead of rearranging facts and concocting arguments to fit its pre-determined conclusions—public school administrators are paid too much and public pensions are bad.

As always lunch is on me for the first to ID the photo location.

Children in Oklahoma are entitled to a free education from 4 year-old programs through 12^th grade.  With a few exceptions like charter schools, families avail themselves of this state service by attending schools operated by locally elected boards of education.  As we saw looking at OCAS data most of the cost is paid for with revenues generated by statewide taxes and by local property taxes levied within each school district.  This post is mostly about how the state revenues are allocated among school districts.

The revenues generated by property taxes vary wildly among school districts mostly because the value of taxable property within districts varies greatly depending on local home values and what business operations are located there.  Such disparities have been the subject of litigation here (Fair School Finance Council v State of Oklahoma) and elsewhere (San Antonio Independent School District v. Rodriquez ).  Oklahoma has in place a state aid formula that allocates state revenues to school districts in a way that equalizes most operational funds roughly based on student population.  The primary provisions are found at Title 70, Okla. Statutes, Sections 18-200.1 & 18-201.1.  Look at our State Revenues chart with some highlighting and a couple of additional columns for a simple explanation of how this works:

Look at the second column from the right highlighted and titled Formula.  It totals up the revenues that go into the formula, namely the 35 mills of local property taxes in the general fund, 75% of the county 4-mill tax, the four yellow highlighted state dedicated revenues, and the gorilla amount that is appropriated by the state legislature each year from the state’s general fund and some dedicated amounts (courtesy of HB 1017).  Each of these revenue sources has its own history and idiosyncrasies, like the county 4-mill was the original funding for segregated schools, like the volatility of gross production, and like both property taxes and motor vehicle being subjects of recent statewide litigation.  So there’s a lot of ifs, ands and buts in the process but the bottom line is those revenues all totaled together accounted for 76% of all school districts’ operating funds new revenue in 2016 and for 61% of all new revenue.  So if it in fact fairly allocates that much of the state’s school funding it’s doing pretty well.  Add to that the insight from the last column, highlighted in blue, which lists the revenues from all funds that either follow the employee (health insurance) or the students and we find that over 95% of the general/co-op funds and almost 82% of all funding is pretty fairly allocated among school districts.   Let’s see how the actual formula works by reviewing the FY 2016 State Aid Allocation sheet for the entire state; here it is (omitting some minor adjustments at bottom):

There are three parts, Foundation, Transportation and Incentive Salary.  Transportation is very small compared to the first and the last which allocate $1.8 billion, so we’ll ignore it.  Next notice the revenue sources listed under Foundation–ad valorem (property taxes), 4-mill, school land, gross production, motor vehicle and REA–are the same ones highlighted in yellow in the first table and are called “chargeables”.  The chargeable for Salary Incentive is a bit harder to discern being the $30.3 million at item 2 then multiplied by 20.  Let me explain why.

Our state constitution provides for levy of 35 mills in support of school districts’ general fund operations.  15 mills were always permanent while the remainder, up to 20 mills, had to be voted annually for most of the state’s history until districts were allowed to vote making it permanent also.  So when the Incentive Aid was put in place it was structured mathematically to require a district to vote the maximum 20 mills to receive the maximum available aid.  Notice on the statewide form line 4 the (00.0), that would be (20.0) for most all districts.  For right now suffice it to say that the only Salary Incentive chargeable revenue is 20 mills of local property tax; the other 15 mills is listed under the Foundation section.  Later on we’ll simplify this greatly.

Now let’s circle back and compare what shows on this State Aid Allocation sheet with the revenue highlighted in the summary chart above.  That total is $3,500,870,848 being the revenue actually received in FY2016 from the seven revenue sources that are part of the state aid calculation.  From the Allocation sheet these are the projections used:

$1,066,821,517 from the “Total Chargeables” line which is the total of the prior year collections from the dedicated sources and the current year projection for 15 mills of local property taxes.

$606,435,690 which is the current year projection for close to 20 mills of local property taxes and is the result of multiplying the Adjusted District Valuation of $30,321,784.516 X 20; this statewide calculation is not exact for reasons too in the weeds even for me, but is close enough to see how this all fits together.  This amount is also a “chargeable”.

And $1,826,404,722 which is the “Basic State Aid” total shown to be paid out according to the state aid formula; this money comes from the annual state appropriation and I think is the largest single line item of our state government’s annual budget.

$3,499,661,929 is the total of these three and is amazingly close to the actual collections of $3,500,870,848—as they say “pretty good for government work”. 

Now that we know what revenues are included and their total amount we turn to how they are apportioned among school districts.  Essential to understanding this part of the state aid formula is how school districts’ “weighted average daily membership” (WADM) is calculated, because it is on that basis, primarily, that the $3.5 billion is divvied up.  At the core of WADM is ADM, average daily membership, which is simply how many students were enrolled, on the average, over the 175 or so school days in the school year.  In the initial count, ADM, each student has a weight of 1, so for a district having an average daily enrollment over the year of 1,000 its ADM is 1,000.

Now comes the real fun.  Because it is believed that different students cost different amounts to educate, certain categories are given weights in an attempt to allow for the expected higher cost.  First there are Grade Weights, which include the 1 for each student, from 1.0 for grades 4-6 to 1.2 for grades 7-12 and 1.3/1.5 for Kindergarten and 1.5 for Out of Home Placement.  So a thousand students in the sixth grade start out weighted 1,000, but the next year as 7^th graders they are weighted 1,200.

The highest weights are for students receiving Special Education services which range from speech services only at .05, to Learning Disability at .4, to intellectual disability at 1.3, to severely emotionally disturbed at 2.5, to deaf at 2.9, and to blind at 3.8, as examples.  Therefore, a blind 7^th grader has a weight of 5.0 and a sixth grader receiving speech therapy is weighted 1.05.

Other weights added on are for Gifted at .34, Bilingual at .25, and Economically Disadvantaged (determined by qualification for a free or reduced price lunch under the federal school lunch program) also at .25.  To summarize, a ninth grade student with a learning disability who qualifies for a free lunch is weighted 1.85.  Here’s the statewide totals for our example year FY2016:

Divide the total WADM of 1,108,650.14 by total ADM of 686,281.45 and we see that the average Oklahoma student carries a weight of 1.62.  Anyhow the state aid formula provides that our goal is for every school district to receive from the seven revenue sources that totaled $3.5 billion in FY2016 the same amount per WADM.  The formula then generates a state aid factor which is the total expected revenue from the Allocation sheet, not actual because we don’t know that in advance, being our $3,499,661,929 total less the stuff like transportation that doesn’t go through the formula, so something like $3.47 billion divided by the WADM (higher than actual because districts can use highest of three years) shown on the Allocation sheet of 1,122,952.01 suggests a factor of about $3,090.  The factor actually shown on the sheet is $3,034.60 which is calculated thus:  $1592 + $72.13 X 20.  We’re not exact, but again close enough to know we see the forest without getting lost in the trees (I love trite banality).

Here’s how a school district can estimate its state aid in five steps:

1.  Calculate your WADM, example 5000

2.  Calculate the total state aid factor, $1592 + 20 x $72.13 = $3032.60

3.  Target Revenue then is 5000 x $3032.60 = $15,163,000

4.  From the Target subtract chargeables which are the sum of:

          35 mills local property tax   $2,000,000 (current assessment X millage)

          75% of county 4 mill             $480,000 (prior year)

          Gross production                  $220,000 (prior year)

          Motor vehicle                        $1,250,000 (prior year)

          Rural Electric                          $190,000 (prior year)

          State land earnings               $510,000 (prior year)

Total Chargeables then is $4,650,000

5.  Result is State Aid amount for that district calculated like this: 

$15,163,000 – $4,650,000 = $10,513,000 in state aid.

Let’s check out how this looks for three real school districts now at the beginning of FY2019, with WADM’s near this example of 5000.  Here are the Allocation sheets for Collinsville, Western Heights and Pryor.

For each, Step 1 uses its prior year WADM which will be adjusted mid-year.   Step 2 calculates the FY2019 initial state aid factor, being $1751.44 + 20 x $83.53 = $3,422.04, which is a huge increase over FY2016 due to the pay raise funding.  Step 3 multiplies Step 1 times Step 2 to get the Target revenue for each district which is in proportion, of course, to its WADM since the factor is the same for all.  Step 4 totals the chargeables estimates using approximately 35 mills times the district’s current valuation and the prior year’s actuals for the other five.   Finally, Step 5 is the difference between Steps 3 and 4, being the amount the chargeables total is shy of the Target Revenue, which is the amount that needs to be made up in state aid. 

Even though Collinsville is smaller than Western Heights by almost 1700 WADM, it will receive more state aid, $9.134 million compared to $5.605 million, because its chargeables are predicted to be much less per WADM.  Without state aid it cannot begin to pay its teachers or offer educational services anywhere near to what Western Heights is able to do.  But, if the projections are accurate, then they each will have about the same total revenue passing through the formula per WADM.  In this way state aid balances out the disparity in local and state dedicated revenues and makes it possible for teachers to be paid and students educated with comparable resources across the state.

Then there’s Pryor.  Because its chargeables total more than the Target Revenue (thanks to Google locating there) it will get no state aid, but still will have more revenue per WADM than the other two districts.  It is “off the formula”.   There are several districts that do not receive state aid and usually it is their property tax wealth, i.e. Valuation Per Student, that sets them apart.  To  find this statistic I go to the School Profiles Reports from Oklahoma Office of Educational Accountability

https://www.schoolreportcard.org/report-card

The statistic is Valuation Per Student, i.e. the district’s total ad valorem assessed valuation divided by its average daily membership.  Statewide the number is $49,471.  Here is Union’s report showing a very close to average $50,422. 

On the chart following is the amount for each of the three districts and my FY2016 example.

Next time we’ll see how to use what we’ve learned.

As always lunch is on me for the first to ID the photo location.

My friend Peggy, retired educator of many including a child of mine, invited me to talk with a group about Oklahoma school finance.  I agreed and decided to incorporate my preparation into three posts leading up to that discussion.  Hopefully this information will guide a patient reader, with further study, to more fully comprehend Oklahoma school finance.  Sadly, such a reader would be well ahead of the fellows at the Oklahoma Council of Public Affairs who seem destined to uselessly flail away without real progress toward understanding and being baffled by what is a pretty straightforward regime.

The first step, and the focus of this post, is to learn how the Oklahoma Cost Accounting System (OCAS) works.  The manual is available on the State Department of Education website here.  The system uses understandable codes to classify all school district revenues and expenditures.  Every school district is required to report its revenues and expenditures, properly coded, to SDE within two months after the end of each fiscal year.  You can find abbreviated versions here on the SDE site.  School districts are also required to have their revenues and expenditures, properly coded, audited by an outside auditor.  School district audits are posted online by the State Auditor and Inspector’s office.  Audits have their purpose and place but I recommend the OCAS data as the place to begin.

Initially I’m going to use statewide data, however even though complete summary data is presented for individual districts, SDE does not provide complete statewide summary data for FY 2017 expenditures (capital expenditures are excluded) so I’m going to use FY 2016 tables as examples for both revenues and expenditures.   Here’s the first of three pages for FY 2016 statewide revenue.

Across the top are separate funds, namely General, Co-op, Building, Child Nutrition and Maps, which organize revenues and expenditures for particular purposes.  General fund moneys can be used for most everything except capital expenditures.  Co-op funds account for activities jointly financed by more than one district.  The Building fund collects part of the district’s local property taxes and must be used for the construction, operation, repair and maintenance of facilities.  The Child Nutrition fund accounts for school lunch and breakfast operations.  Maps is one of five funds that are tiny, about 1% of the total, and can be ignored to keep our review simple.

On the left side are revenue sources listed with OCAS codes in the thousands.  The codes organize revenues broadly by the level of government from which they come.  1000s are Local District Revenues, 1100 is property tax collections; 2000s are Intermediate, usually through the County Treasurer; 3000s are State Revenue with three subgroupings:  3100s are state dedicated revenues that are paid directly to districts, 3200s are State Aid and direct payment for employee insurance, and 3300 and above are other state grants and revenues; 4000s are Federal Revenue; 5000s are “nonrevenue” flows, primarily transfers from other funds where the revenue is already accounted for; and 6000s are fund balances.

Look at the remaining two pages and then I’ll suggest how to greatly simplify the revenue summary.

The General (11) and Co-op (12) funds for school districts are available for general operations, i.e. teachers, principals, counselors, teacher assistants, custodians, bus drivers, maintenance workers, utilities, gas and diesel fuels, insurance, etc., but not construction of facilities.  So we can simplify by combining General and Co-op.  The Building Fund can also pay for operations, but we’ll leave it separate for now.

As I said above several funds are specific to only a few districts and are insignificant in amounts so we can set them aside for now, namely 24, 25, 26, 50 and 81-86; these are just 1.1% of total revenues.  Also the Student Activity Fund (60) is easily distinguished because it does not receive tax revenue and is limited to the activities that generated the revenue; it is 4.1% of total expenditures.  Removal of these funds decreases the total FY16 revenues by $419,078,744 to $7,642,007,347 including the fund balances. 

Another adjustment that makes sense is to move Child Nutrition revenues, namely 1700, 3700 and 4700, from the General and Co-op funds and add to the same sources in the Child Nutrition fund.  Districts have the choice to record child nutrition revenues and expenditures in the General/Co-op or Child Nutrition Fund.  There is no significant policy decision involved so we can simplify the overall picture by having all Child Nutrition financial activity in the same fund.  What is there are revenue sources 1700 for local (family payments for lunches), 3700 for state (the small state match that is required) and 4700 for federal (the federal funds that drive the school lunch and breakfast programs).

Next we can consolidate similar and/or minor revenue sources to further simplify our revenue summary.  Local, 1000, can be just property taxes (1100) and everything else (1200-1600); remember 1700 has been moved to Child Nutrition and 1800 and 1900 are removed with the Student Activity fund.  Minor County sources, 2300, 2400 and 2900, can be consolidated as can minor dedicated State sources 3111, 3150, 3160 and 3190.  Similar sources for Alternative Education 3310 and ICTE 3360 can be combined; remember 3700 amounts are in Child Nutrition now.  Lastly federal minor sources 4400, 4500, 4600 and 4800 can be consolidated, while 4700 also is now all in Child Nutrition. 

The result is a statewide school district revenue summary that fits neatly on one page and captures all revenue sources that should arguably be part of policy discussions.

I’ve included a few percentage calculations for each fund of its total new revenue.  Here are the essential take-aways:

Operational expenditures, such as paying for teachers, principals, bus drivers and diesel/gasoline, are supported from the General and Co-op funds primarily with State tax revenue (60%) and Local property taxes (26%).  The largest single entry is almost $2 billion for State Aid which will be the topic of my next post.

Additional local property tax revenues are available in the Building fund (92%) for operation, maintenance and acquisition of school facilities and equipment.

Child Nutrition (school breakfast and lunch) services are paid for with special Federal (75%) revenue and local charges (19%).

The Bond funds work in tandem with the Sinking fund.  Expenditures, as we will see, are recorded as Bond projects which are paid for with local property tax collections through the Sinking fund.

We can now turn to the expenditure side.  Here are the broad OCAS “Function” categories used along the left side of the following statewide FY 2016 expenditure tables for certain funds:

 1. Function 1000 Series – Instruction (teachers, teacher assistants, textbooks, etc.)

 2. Function 2100 Series – Support Service/Students (attendance clerks, counselors, nurses, OTs & PTs, speech path, etc.)

 3. Function 2200 Series – Support Service/Instruction Staff (library & staff, training, instructional labs and software, etc.)

 4. Function 2300 Series – Support Services/General Administration (superintendent & assistants, board, legal, audit, etc.)

 5. Function 2400 Series – Support Services/School Administration (building principals & assistants, school secretaries, etc.)

 6. Function 2500 Series – Central Services (financial and technology operations, insurance, etc.)

 7. Function 2600 Series – Operation and Maintenance of Plant Services (custodians, maintenance staff, utilities, etc.)

 8. Function 2700 Series – Student Transportation (school buses, drivers, mechanics, fuel, etc.)

 9. Function 3100 Series – Child Nutrition Program Operations (cooks, servers, food, kitchen equipment, etc.)

 10. Function 3200 Series – Enterprise Operations

 11. Function 4000 Series – Facilities Acquisition and Construction Services

 12. Function 5000-8000 Series – Other Uses and Repayments

Across the top are broad Object series codes; 100 is for employee compensation and 200 is for benefits like health insurance and Social Security.  100 and 200 added together are the cost of personnel. 

Let’s look for the largest dollar amounts in each of the major funds.

The largest function total for the General fund is upper right, $2.8 billion for Instruction.  Scan clear to the left and you see the 100 and 200 personnel costs, i.e. classroom teachers, account for almost all of the total.

The largest function total for the Building fund is toward the upper right, $140.8 million in 2600 Operation and Maintenance of facilities.  The largest Object column totals are 300-500 which include water/sewer/refuse utilities and contracting for custodial and maintenance work, 600 which includes electricity and natural gas utilities, and 100 & 200 personnel costs for custodians and maintenance employees.  All of these expenditures could, if there was enough revenue, instead be paid for in the General fund.  Note how little is actually used for construction, Function 4000.

Child Nutrition, as you would expect, is about Supplies, i.e. food, in Object 600 and the workers who prepare and serve the meals, again Objects 100 and 200.

Bond fund expenditures are dominated by Function 4000, construction and acquisition of facilities and building improvements, with most of that being paid to outside contractors as recorded in Object 300-500.  These expenditures are made with the proceeds from selling general obligation bonds, i.e. borrowing money.  The borrowed funds, plus interest, are paid back from Sinking Fund revenues.  Because the actual expenditures have already been recorded in the Bond funds, in the Sinking fund the bond payments are recorded as Function 5100 Debt Service and are NOT included in any traditional totaling of school district expenditures.  The value received by the district is through the expenditure of the bond funds; to include the bond payments, other than the interest cost, counts the same spending twice and exaggerates the overall total.

Next time we’ll see how the State Aid Formula affects the major funds.

As always lunch is on me to the first to ID the photo location (sorry about the graffiti, not).