McCleary Phase II

The Legislature has made significant progress toward fully funding basic education. In the last three years, we’ve reduced K-3 class sizes, funded all-day kindergarten, and provided kids adequate materials and supplies for their classrooms. But there’s one critical and final component we must deal with – teacher compensation and local levies. The chart below should start to explain the problem.

Average Salaries for TeachersThis chart shows how teacher salaries have been constructed since 1996. The gray portion at the top is the part provided by local levies, not by the state.

Teacher salaries aren’t the only cost. School districts also pay classified staff and administrators, and the split between state and local funding is even greater for these categories than it is for teachers.

There are lots of reasons for this. Some make sense given how the system works and some are the result of the legislature skipping cost of living (COLA) increases during the recent recession. When the Legislature doesn’t fund COLAs, but the local district wants to fund them (in order to actually be able to hire competent teachers) they use money raised in local levies to do so. If it was just extra, above and beyond what is needed to actually be able to hire, this would be expensive, but not a constitutional problem. However, every study that we’ve done shows that they’re paying just about what the market needs them to pay to be able to hire and retain competent employees. The court ruled this unconstitutional because it’s the State’s responsibility to adequately fund basic education, and we’re shirking that duty by foisting part of the cost on local districts.

How Big Is the Problem?

Compensation by School DistrictTo figure out the gray area in the above chart you have to figure out which teachers are doing “basic education” and which are providing “extras.” For example, the basic ed model (adopted in HB 2261 and HB 2276 in 2009 and 2010) assumes a six period day. The Bellevue School District has seven, providing more options to students. This requires more teachers. Then you need to know how much of their time is spent on basic education. Many teachers do extra stuff, like coaching or advising a club.

Most estimates of the problem assume that 90% of actual average statewide district compensation payments to employees that are part of the basic education staffing model go towards the State’s definition of “basic education.” The remainder is a local enhancement. This produces about a $3.5 billion total biennial cost that should be borne by the State. The details of this are mind-numbing, but the conclusion is pretty broadly accepted.

How Do We Figure Out How Much To Send To Each District?

The simple solution would be to figure out the average teacher salary, figure out how many teachers a given school needs based on its population and send the district that much money. Of course it’s more fun than that to figure out.

First, costs are higher in some districts than others. You can see this in the chart to the right. It shows a vertical bar for the salary of an average teacher in each district in 2012-13. The dark color is that state portion and the lighter color is the part from the local levy. The districts are arranged so that the total average salary goes up from left to right.

There is a huge difference between the districts on the left and the ones on the right. Mostly the higher paid districts are in urban areas and the lower paid ones are in rural areas.  Most proposals we’ve seen so far have some kind of localization factor applied to the base amount we give to each district to account for this difference. Figuring out a localization system isn’t rocket science, but it’s politically very challenging as some districts will get paid more than others. This is reality in the market, and a rational system has to account for it.

Goals for this system

  1. The localization factor should reflect labor market differences including cost of living and the relative attractiveness of the districts to potential workers.
  2. There should be minimal differences between adjacent districts, so you have to have lots of districts. If you do it by county you get big differences between adjacent districts that won’t work in practice.
  3. The data source for the localization should be broadly available and not arbitrary/political. This has to be perceived as fair by everybody concerned and not subject to

Today we have a very complicated system where we figure out exactly which teachers every district has hired and send them an amount of money that’s adjusted based on the actual experience and education levels of teachers. This turns out to not make much difference except for a handful of very small districts. The averages tend to balance out. Should we keep using this complex system, or should we just send the money to the district and let them work out how to allocate it? This seems fairer as well, as there is no reason one district should be able to hire much more experienced teachers than another district can.

Figuring out how much money goes to districts for classified employees (bus drivers, IT specialists, payroll clerks, etc.) and administrators (principals, superintendents, etc.) presents much the same problems and has to be added to the problem.

Local Levies?

If the state is paying the full cost of hiring employees, why do we still have levies? The example I gave above of Bellevue wanting to have seven periods is a real one. Districts want to do things that make sense for their students and their taxpayers. There is broad disagreement about what the levy system will look like after the transformation, but we all share at least one problem – what are the rules for the use of local levies by districts?

The goal is to prevent the use of local dollars to pay for compensation for basic education employees, which we define as employees in the staffing model. Districts should be able to hire additional employees, but there should be some mechanism to ensure that local funds are not used for basic ed. This will be complicated and will affect how districts and unions bargain compensation. There are wildly different proposals in this area.

How much time do we have to figure this out?

What is the timing of the implementation? In their August 2015 McCleary order the Supreme Court was very clear that the payroll problem (as they describe it) needs to be solved by the 2017-18 school year.

The deadline for legislative action is dependent on the tax solution used, as some tax sources have practical constraints about lead time. For example, property tax changes happen on January 1 of a calendar year, and there is significant lead time needed before this for taxing districts to set rates. Local bargaining typically takes place in the spring, so significant changes to either bargaining rules or salary structure would be difficult to manage in a short timeframe.

This requires all tax changes to be made prior to 1/1/17. Most of the things we’re talking about need three to six months of lead time before then, so the practical constraint is July 1, 2016. This does not lead itself to a phased-in solution. Doing a phase-in is computationally very, very difficult to figure out in advance.

So how does all of this get paid for?

Depending on the decision about total cost, about $3.5 billion per biennium needs to be raised to fund the new system. This has to get paid for, though there are ways to do it so that the impact on state taxpayers is significantly lessened. In most cases the payroll need is currently being met by local taxpayers. About 5 years ago I proposed a solution that “swapped” local levy revenue for state revenue that produced a solution that was close to revenue-neutral statewide. Time has crept by and there is no longer enough property tax available to make this solution work, but it can be a key element of whatever we choose to do. The end solution will have about $3.5 billion raised at the state level and significant reductions, but not quite $3.5 billion, in local levies. This produces something that is close to revenue neutral statewide, but with increases in urban areas and decreases in rural areas.

The key question is “what revenue sources should be used, and in what mix?” The state property tax is called the “Common School Levy” for a reason – it’s constitutionally dedicated to funding public schools. While staying below the constitutional limit of $10 per $1000 of home value and without significant disruption to local governments who also use the property tax the Legislature could increase the state rate from about $2.00 to about $3.11. This produces about $2.7 billion. You still need around a billion dollars and have to use some other source.

We’d like the McCleary solution to be durable over time, so the I-747 (1%) growth limit would need to be repealed to make this a sustainable source. Otherwise you recreate the problem in a few years when the revenue grows more slowly than the cost of providing public education.

Large increases in the state property tax cause large increases in per household property tax bills in urban areas reducing the interest of urban legislators to have the whole solution be property tax. There are other options that could be used in a variety of mixes.

  1. Capital Gains
  2. Carbon fee or tax
  3. Closing existing tax preferences
  4. Other new tax

The big question here will be how much property tax swap is done, how much other tax sources are used in the mix, and how close to revenue-neutral the complete solution is statewide. This will be very politically contentious, and will probably take quite some time to work out.

Related Problems

These do not need to be solved to comply with the constitutional problem posed by McCleary. They are related however, and in some cases a practical system cannot be created without solving them.

What is the structure of the levy system after implementation?

What determines how much a district can raise? Many options have been proposed here, including per-student amounts, limits based on property values, etc. There is broad agreement that districts will be able to continue to raise levies, and that we will fix the current system which has a lot of arbitrary and capricious rates left over from the last time a solution like this was attempted 30-40 years ago.

If we have a levy system, the state needs to provide some help to districts that don’t have lots of commercial property to tax. We call this help levy equalization. Currently over 220 of the 295 districts receive equalization, which seems too high. This system has grown at about twice the rate of the amount of levies collected by the richer districts, which also seems not sustainable and looking for a solution.

How are the costs of benefits calculated?

The state is required to adequately provide for both salary and benefits. There seems to be reasonable agreement to tie the total to the PEBB rate (the rate the state pays for benefits for other state employees,) at least for healthcare. A question related to this is “do we create a state-run healthcare system for school employees?” If so, what are the bargaining rules around it? There is some thought that this would be cheaper and provide better healthcare, but again, there are large disagreements about this.


What changes to reporting are required to ensure compliance with basic education compensation rules? Do we do audits?

Should we make other changes while we have the patient opened up?

  • Career and Technical Education (CTE) changes to MSOC? Possible to convert to an excess cost model like special ed. This would be simpler.
  • Small school factor. The new resources are changing the impact of the small school factor. Does it need to be changed?
  • Do we retain a state-mandated salary grid? If YES, what changes do we make to it? Are they just for new teachers, or do they apply to existing teachers as well? Is there a standard grid for classified employees? We do not currently have one. If NO, do we impose rules around local compensation bargaining beyond those required to ensure compliance with the “no local money for basic ed” rule?


This set of problems can be resolved. There are lots of parts that Republicans and Democrats disagree on. There are lots of parts that urban and rural legislators disagree about. There are just a lot of parts.

The changes will be wrenching, but in the end we’ll have a system that is fairer to all districts, pays teachers and staff adequately, allows local levies to be used for enhancements, not for making up compensation costs that the state doesn’t fund adequately, and the tax system will be more fair. This seems like a good outcome.

All sides will need to compromise to get a solution that makes sense and will last for the next 20-30 years.

Author: Ross

I am the Director of the Department of Early Learning for Washington State. I formerly represented the 48th Legislative District in the State House of Representatives, chairing the Appropriations committee and spent many a year at Microsoft.

One thought on “McCleary Phase II”

Comments are closed.