Which states screw the largest share of low income children? Another look at funding fairness

Here’s a little Friday afternoon fun with the updated Census Fiscal Survey data through 2009-2010. I’ve written recently about the national school funding fairness report card, which I work on with colleagues from the Education Law Center. The report card can be found here:


I also recently wrote a blog post about America’s Most Screwed City School Districts. It was clear to some readers that the most screwed city school districts happen to be concentrated in certain states like Illinois and Pennsylvania, and also in Connecticut which is often perceived as a reasonably well funded and fairer state (than the other two).

Par for the course, as soon as we release the School Funding Fairness report card using data from 06-07 to 08-09 (most recent available at the time we put it together), the Census Bureau releases their 2009-10 district level finance figures… leading to the usual flurry of misinterpretations of data (which I’ll get to another day). Not being able to resist the temptation, despite a heavy backlog of other work to do, I decided I had to play with the updated fiscal data. I also decided for fun to take an alternative look at the data, bridging the idea I presented on my blog about screwed city schools with the general idea of state school funding systems. I decided to ask which states screw the most low income kids.

Here’s my operational definition of screwed for this post. A district is identified as screwed (new technical term in school finance… as of a few posts ago) if a) the district has more than 50% higher census poverty than other districts in the same labor market and b) lower per pupil state and local revenues than other districts in the same labor market. As I’ve explained on numerous previous occasions, it is well understood that districts with higher poverty rates (among other factors) have higher costs of providing equal educational opportunity to their students.

I then tally the percent of statewide enrollments that are concentrated in these screwed districts to determine the share of kids screwed by their state. And here are the rankings… or at least the short list of states that screw the largest share of low income students:

Not much new here. The same culprits make up the list. Nebraska is elevated to its position of disgrace by its systematic underfunding of Omaha Public Schools, which seemed to improve for a fleeting few years, but recent data don’t look so good. Woonsocket and Pawtucket bring Rhode Island into the mix… but raising additional fun questions regarding placement of blame (another post, another day… but should the city managers/local officials have the authority to deprive children in their jurisdiction of state constitutional rights? under what circumstances and by what mechanism should the state step in? Can they?).

Here are a few graphs showing the distributions of individual districts in Illinois, Pennsylvania and Connecticut. On the horizontal axis is the relative poverty rate of districts compared to all other districts in the same core based statistical area. On the vertical axis is the state and local revenue per pupil relative to the average for all other districts in the core based statistical area.

Again, Allentown, Reading and Philadelphia are massively screwed (yep… a new school finance classification). Meanwhile… Lower Merion… in the Philly ‘burbs is not screwed at all. An intriguing contrast in Pennsylvania school finance is that Pittsburgh has long had far more adequate funding than Philadelphia for a variety of reasons. It is important to understand here that the highest poverty districts – those with 3x the average for their labor market – likely need FAR MORE revenue per pupil than their neighbors to get by – not just the same. So, while York and Harrisburg are decidedly less screwed than Allentown or Reading, they too are not in particularly good shape. They have about the same revenue per pupil as surrounding districts, and 3x the poverty rate.

Here’s Illinois:

Waukegan and Aurora East, along with Round Lake hold the coveted spots of “most screwed” but Chicago Public Schools isn’t far behind (with over 400k students). A multitude of smaller high poverty districts in the Chicago metro not shown here also have very low relative revenue per pupil.

Finally, here’s Connecticut once again:

Again, Bridgeport and New Britain, along with Waterbury (among others) remain substantially screwed. Recall from my previous post that Hartford and New Haven funding is somewhat distorted by magnet school aid.

So why does any of this matter anyway. Well, at face value it’s patently unfair to systematically deprive these districts of resources comparable to their less needy neighbors.  If money doesn’t matter to New Britain or Bridgeport, then why does it matter to Greenwich or Westport? Really, if money is so damn trivial for improving schooling quality, they why don’t all those districts in the upper right hand corner of these graphs just give all that useless money to those in the lower right hand corner. Oh, wait… perhaps money does matter…???…!!!

One thing about school finance that’s really important to understand is that the relative position of districts matters a great deal. It matters a great deal because education is a labor intensive industry. It is about getting a sufficient quantity of sufficient quality teachers in front of kids who need them. The spending behavior and negotiated agreements, and working conditions in districts like Westport and Greenwich matter for the teacher recruitment potential for Bridgeport.  The distribution of quality teachers across districts in a labor market depends on numerous factors, many of which tie back to available resources. And in these states, large numbers of children attend high need districts that simply lack resources to compete.

Notably, those districts sitting pretty in the upper left hand corner of these figures also have had traditional teacher contracts, tenure, seniority preferences and likely other policies that would make “reformers” cringe for years.  But most are doin’ just fine.  So too do the even higher spending and lower poverty elite private schools in the same labor markets! Most don’t use test scores as the basis for providing merit pay and I’m quite sure that few if any of them use test scores as the basis for firing the bottom 5% of their teachers every year. They haven’t been and aren’t being subjected to manipulative heavy handed takeovers, school closures and massive charter school expansion.

None of that reformy junk would likely do much good for the Westports, Greenwiches or Lower Merions of the US school system.  And none of that reformy junk is likely to be much good for the Bridgeports, New Britains, Allentowns, Readings, Philadelphias or Chicagos!

I find it particularly infuriating when I hear news of these “most screwed” districts being blamed for their own failure by the state officials who have deprived them systematically of resources for decades.

What these districts need as a baseline – a fair starting point – is equitable & adequate funding. Once that has been accomplished, then, and only then can we start having a reasonable conversation about how to best leverage that funding to improve student outcomes. But without the funding, there are no options for leveraging it.







  1. “…it is well understood that districts with higher poverty rates (among other factors) have higher costs of providing equal educational opportunity to their students.”

    Why is that?

  2. I don’t understand why you factor in relative per-pupil state and local *revenues*. Shouldn’t you be focusing on per-pupil expenditures, rather than revenues? They are not necessarily the same, since the state may make some effort to redistribute revenue for purposes of school spending.

    1. Indeed “state and local revenues” and current expenditures per pupil are not one and the same. State and local revenues tend to be a better indicator of the influence of state policies. State and local revenues best reflect the sum of the state policy context – how the state provides aid to counterbalance, or not, effectively inequities in local revenue raising. I will have a detailed report out on this stuff in the fall. Current expenditures include expenditure of federal funds and also reflect district decisions regarding such things as reserve funds/balances, etc. When I write about and/or critique state policies the best figures to use are state and local revenues. That said, in this type of analysis, I would tend to get the same states falling into the “worst” categories. Federal aid provides a slight moderation to the negative tilt in regressive state aid formulas, so there would be marginally fewer high poverty districts in the totally screwed zone if I used current expenditures but the relative position of states compared to one another would be similar.

      1. I’m sympathetic to the argument you’re making, but I’m still not understanding why you would look at overall state revenues, rather than how much the state is spending on education, or how much the state itself (as opposed to the federal government) is spending in any particular school district. Wouldn’t either of the latter two measures be a much better proxy for state educational policies than overall revenue? Or why not just look at overall per-pupil public spending in any given district? Is it just that those figures are harder to obtain?

        Overall state revenue seems like an awfully shaky proxy for “how the state provides aid to counterbalance, or not, effectively inequities in local revenue raising.” A state could raise lots of money and spend none of it on schools.

      2. I’m looking at combined state and local revenues – which appropriately represent the resources available as a function of state policy – policy which regulates local taxation and raising of local tax revenues and policy that defines the distribution of state aid. I’m not looking at state aid alone. It’s the combination of state and local revenues, or the vast majority of funds available to be expended by local school districts as a function of state policy. Also, I’m looking at combined state and local revenues allocated to local education agencies specifically. Not all state aid to all government functions, nor all local revenue for other function. This is state and local revenue to local public school districts, the majority of which translates over to current operating expenditures. Again, it’s about capturing the affect of state policy/governance over state and local revenues – setting aside federal revenues and/or local spending decisions. I can, and do often, separately evaluate the current operating expenditures of districts.

Comments are closed.