During the financial crunch, the state abandoned some formula elements that were designed to better match aid to community ability to pay. The result over the past four years has been an erratic pattern of aid increases which does not treat communities of similar wealth in similar ways. An additional feature of the erratic pattern is that, for communities receiving smaller aid amounts, it includes major spikes in aid that are not proportionate to spikes in need.
Overview of the Response to the Crunch
In the 2006 reform, downpayment aid was designed to cover an increment towards the “Fully-phased-in Foundation Aid”, which in turn was the difference between target local share in the new wealth model and the foundation budget. For each community which, under the old model was receiving an amount less than it was entitled to under the new model, the down payment aid was set to close the gap in five roughly equal annual steps. After three years of basically following the new formula, in Fiscal 2010, when the financial crunch came, the state stopped paying downpayment aid.
A fundamental element of the old Chapter 70 construct had been preserved for transition purposes in the computation– “required local contribution” or RLC. Foundation aid (as opposed to “fully-phased-in foundation aid”) continued in the new model to be based on the gap between the RLC and the foundation budget, not directly on the new-model target share. The very problem that the reform was intended to address was that many communities of middling resources had old-model required local contributions that were well above their foundation budget. Even in the new-model, these communities received no foundation aid; instead these communities received incremental downpayment aid. When downpayment aid was stopped in FY10, we partially reverted to the old model and many of these communities stopped receiving aid increases.
The reversion to the old model was only partial because we continued to make phased adjustments to the RLC that were designed to cause old-model RLC’s to converge to the new-model target shares over time. Since the beginning of education reform RLC’s have been adjusted up each year by a few percent for each community by a “Municipal Revenue Growth Factor” which basically reflected rising property tax levies (as constrained by Proposition 2.5) and general municipal aid. The MRGF is computed by the Department of Revenue and historical values can be viewed here. Under the 2006 reforms, the RLC is annually adjusted in two ways to cause a long-term convergence towards target shares:
- For communities with RLC’s more than 5 percentage points below the new-model target the RLC is pushed up by one additional point and one further point if the RLC is more than 10 points below target.
- For communities with an RLC above the new-model target share the RLC is reduced by an “effort reduction” percentage — the effort reduction percentage (15% in 2013 H2) is applied to the excess of the preliminary RLC (prior year plus the MRGF) over the target.
Both of these computations appear in the townwide contributions tab of the DOE complete formula spreadsheet.
Annual Details of the Response to the Crunch
- The kinds of aid available in each fiscal year:
- F = Foundation aid — gap between old-model Required Local Contribution (defined above) and foundation budget (to the extent not covered by previous year’s grandfathered aid level).
- DP = Down Payment aid — the down payment percentage applied to the gap between new-model target local share and foundation budget (to the extent this quantity exceeds incremental foundation aid).
- GR = Growth aid — new-model target local share applied to growth in the foundation budget (to the extent this quantity exceeds Down Payment plus Foundation Aid).
- MI = Minimum Increase — a per pupil amount applied to the number of people, granted to the extent it exceeds the sum of other aid forms.
- OG = Only grandfathering — a code used in the table at the end of this page to indicate that none of the other aid components were available to a community in the particular fiscal year.
- A list of the additional major parameters being varied to match total formula aid to fiscal constraints:
- The down payment aid percentage used to compute down payment aid — this was increased on a schedule that would have led to full phase-in in roughly equal steps over five years (20%, 30%, 33%, 50%, 100%), but only the first three years were implemented.
- Effort reduction percentage is the amount of the excess of the old-model RLC over the new-model target share which is reduced in each year. This started on the same basic five year schedule as the down payment aid, but was slowed, although not stopped during the crunch. As shown in the chart further above, convergence to target share has been considerable, but the low recent reductions have left some communities still above target share and are not necessarily sufficient to maintain communities at their target share if reached, depending on foundation budget growth and the MRGF.
- The increment towards target is the upward adjustment for communities with low RLC’s — this has been maintained consistently at 1 percentage point for communities over 5 points below target and 2% percentage points for communities over 10 points below target. In FY2010, a special further adjustment was made to pull communities up towards their new-model target — the RLC was increased to the lesser of new-model target or 95% of their FY08 actual contribution (based on spending records).
- The minimum increase per pupil is the parameter set to determine a minimum aid increment.
- A base reduction was made in FY10 — a cut in the prior year base used in the computation — start with FY09 aid, cut it 2%, then add back foundation aid using federal aid (to the extent foundation budget less RLC exceeds the reduced amount). In FY11, a 4% base reduction was used and then a further 1.919% cut was made to the total of base plus foundation aid, leading to a total cut for many communities of almost 6%. However, with federal jobs aid, added outside the Chapter 70 cycle, the cuts were restored and communities received a minimum $25 per pupil increase. In FY12, no aid cut was applied, but the jobs aid went away so that communities not benefiting from incremental foundation aid lost their previous minimum increase and further took the FY11 cut — for a total cut of roughly 7%.
|AVAILABLE AID CATEGORIES||F, DP, GR, MI||F, DP, GR, MI||F, DP, GR, MI||F||F, MI||F||F||F, MI, TA||F, DP, MI|
|Down Payment Aid Percentage||20%||30%||33%||n/a||n/a||n/a||n/a||n/a||35%|
|Effort Reduction Percentage||20%||25%||33%||15%||30%||20%||15%||15%||50%|
|Increment Towards Target||None||1%/2%||1%/2%||See text.||1%/2%||1%/2%||1%/2%||1%/2%||1%/2%|
|Minimum Increase per Pupil||$50||$50||$50||n/a||$25||n/a||n/a||$25||$25|
|Base Reduction||2%||See text.||See text.|
|Target Aid||n/a||n/a||n/a||n/a||n/a||n/a||n/a||See text||n/a|
|Cap on RLC as % of Foundation||150%||n/a||n/a||n/a||n/a||n/a||100%||100%||90%|
|Foundation Budget Inflation||5.86%||4.66%||5.18%||3.04%||-2.2 %||1.78%||3.65%||2.58%||1.67%|
|Link to DOE Explanation||FY07||FY08||FY09||FY10||FY11||FY12||FY13||FY14||FY15|
All of the aid variations summarized in the table above are visible in the annual complete formula spreadsheets available under the heading Chapter 70 State Aid and Spending Requirement on the DOE’s Chapter 70 page.
NOTE: We updated this table in January 2014 to include the Fiscal 2014 and 2015 (Governor’s proposal) details. The legislature introduced the concept of Target aid in Fiscal 2014 in response to the concerns of below-target communites (e.g., Watertown). Target Aid provided was computed as 25% of the gap between target aid and actual aid. It was provided only to communities below a high wealth level cutoff.
Effects of the Crunch on Individual Communities
How do these variations affect individual communities? The chart below shows RLC as a percentage of foundation budget for ten communities in or bordering the Second Suffolk and Middlesex Senate District. Click here for data for communities not shown— the spreadsheet available includes data on all communities for all charts in this sequence of pages; view the “Comparative History” tab for data for the charts on this page.
For all 10 of the sample communities the target share happens to work out to 82.5% and rough convergence towards that level over seven years of the new model is apparent in the chart.
The chart below shows annual education aid increases from prior year for the 10 example communities. The figures include federal stimulus add-ons in FY09 through FY11 and were extracted from the disthist and aid436 tabs of the DOE complete formula spreadsheet — both of these tabs are hidden on download; to show them in Excel 2003, select format/sheet/unhide. FY13 is based on the Governor’s House 2 proposal.
|Aid Increase||FY07||FY08||FY09||FY10||FY11||FY12||FY13||FY 06 to FY13|
Two things immediately jump out in the chart above: (a) there is a wide disparity in cumulative increase over the last seven years; (b) from Fiscal 2010 on, after down-payment aid was eliminated, for the communities that did have increases, the increases came in spikes, rather than continuous growth. The communities fall into three groups:
- Somerville and Boston were below new-model target effort (so their RLC in the chart further above was increasing) and they see only minimal aid growth in the first few years and cuts in the later years.
- Arlington, Belmont, Brookline and Newton were above new-model target effort (with a dropping RLC) and do see substantial gains with a few big spikes.
- Cambridge, Lexington, Waltham and Watertown were also above target effort with a dropping RLC, but saw relatively limited aid increases.
Neither the contrasts between the latter two groups of above-target communities nor the spikes in aid are readily explainable by differences in foundation budget growth rates. The foundation budgets grew more or less smoothly at similar rates. The chart below shows annual changes in foundation budget for all 10 communities.
|FB Change||FY07||FY08||FY09||FY10||FY11||FY12||FY13||FY 06 to FY13|
Nor are the differences between the high aid growth group and the low aid growth group explainable with reference to valid new-model aggregate wealth differences. The next chart assembles several comparison variables, which are combined in the last three columns:
- Combined Effort Yield as % of Foundation shows how revenue raising capacity based on income and property wealth compares to foundation budget — first column divided by second column.
- The column headed Relative Per Capita Wealth displays an index of property and income wealth (third and fourth columns) compared to statewide per capita averages (combined into a single index). This measure is included only as information — it is the preceding column that best measures a community’s ability to fund its schools, which depends on enrollment, not population.
- The final column shows education aid from the Governor’s Fiscal 2013 Budget Proposal as a percentage of foundation budget.
|Combined Effort Yield||FY13 Foundation Budget||2010 EQV/Cap||2010 PCI||Combined Effort Yield as % of Foundation||Relative Per Capita Wealth (PCI and EQV/Cap combined vs. statewide average||Education Aid as % of Foundation, FY2013 House 2|
Inspection of the table above shows that, with the exception of Cambridge, the communities with lower education aid levels (Cambridge, Lexington, Waltham, Watertown) are not communities with higher ability to cover their education costs (based on combined effort yield as a percentage of foundation budget).
Understanding the Formula Modifications
Since the state stopped paying downpayment aid in 2010, aid increments depend directly on the RLC. For communities whose wealth was overestimated under the old model, the RLC has been annually down-adjusted by an “effort reduction” percentage to cause a convergence towards the 2006 reform measure of ability to pay. However, the convergence is not complete, so the RLC is still heavily determined by the discredited, pre-reform measure of wealth. Even as downwards adjusted for effort, the RLC leaves many communities with no access to foundation aid. The impact of the effort reduction percentage in any given year depends on three variables (a) the amount of the excess effort — if the community is just a little above target, a 15% reduction may not change the RLC much; (b) whether RLC has dropped below the foundation budget so that there is a positive foundation aid amount — if not, then the effort reduction has no effect on annual aid; (c) whether that foundation aid amount is greater than the community’s historical aid level — again, if not, then effort reduction does not increase annual aid. Spikes in aid arise when modest foundation budget increases coincide with scheduled cuts in RLC in a year that foundation aid is applicable to the community. The result is an unpredictable pattern of annual changes which bears too much relationship to the long discredited pre-reform measure of ability to pay.
The disparities are sharpest between communities that (a) have similar new model wealth, but (b) dissimilar old model wealth. So, in the charts above, compare Belmont and Watertown, which have very close new-model wealth levels — Belmont has a combined effort yield of 124% of foundation budget and Watertown has 129% (only slightly wealthier). However, in 2006, under the old model, Watertown had a required local contribution of 141% of foundation budget, whereas Belmont had only 99%. So, even after seven years of effort reduction towards the maximum target of 82.5%, Watertown’s required local contribution has not reached target, still at 88% in FY13. Watertown benefited from Down Payment aid that raised its aid from 12.8% of Foundation budget in FY06 to 14.9% in FY09. However, Watertown’s aid has drifted back down to 12.5% of foundation budget in FY13, with the general reductions. This is because Watertown’s RLC is still too high for it to benefit from foundation aid — i.e., Watertown’s foundation aid amount (foundation budget less RLC) has remained less than its grandfathered prior year aid amounts. By contrast, starting from a much lower old model level (for no good reason), Belmont’s RLC hit the 82.5% target in FY2010 and has bounced along just above it for the last few years. The result is that Belmont has had much sharper aid growth than Watertown, rising from 11.8% in FY06 to 17.5% in FY10 and remaining above 16% in subsequent years.
The chart below shows the categories of aid that happened to be applicable to the ten communities in each fiscal year as a result of the interaction of these quantities (using the aid codes in the first chart of this page). The best way to get a feel for the quirky ways that the parameters interact is by picking a community and following it through each year’s DESE aid spreadsheet — accessible from this table of contents.
|ARLINGTON||DP, GR, MI||DP||F||F||MI||OG||F|
|SOMERVILLE||GR, MI||GR, MI||GR||OG||MI||OG||OG|
|WALTHAM||DP, GR||DP, GR||DP||OG||MI||OG||OG|