Proposed Katonah-Lewisboro School Closure: Data Analysis and Questions

I'm sharing here an important report researched and written by a group of fellow Lewisboro Elementary parents. Please feel free to comment on their report and share this link and continue the conversation.

For more on this issue, click here for a story on the first public hearing, and here for a letter to the Katonah Lewisboro School District Governance Team

**********

Proposed Katonah-Lewisboro School Closure: Data Analysis and Questions 

INTRODUCTION

Our community is being told that there is an overwhelming empirical case for closing Lewisboro Elementary School if we examine the available data in an unemotional way. However if the data is analyzed in detail an entirely different picture emerges which throws the rationality of the current proposal into substantial doubt. The purpose of this paper is to examine this data and present an accurate picture to the parents and taxpayers of the Katonah-Lewisboro School District of the questions the current proposal raises and which largely remain unanswered.

This paper will examine the data under 3 headings:

[1] The demographic projections

[2] The claimed cost savings of closing LES

[3] The overall budgetary context

[1] The Demographic Projections:

 A demographic study was prepared for the District in August 2012 detailing 10 year projections for enrollment in our schools. This study is being used as the basis of the claim that we will have sufficient excess capacity in our school in coming years to justify closing one of more elementary school buildings. However serious questions can be raised about the accuracy of the projections in this study.

The study is based on a pattern of historical data from the 5 years preceding the report. This is immediately problematic as that represents almost exclusively the period of severe economic downturn experienced by the USA since the financial crisis of 2008. As such there is a significant concern that this data represents a completely atypical period for the relevant data making it questionable to draw trends and make projections on this basis.

Birth rate projections:

The most significant driver of the decreasing long term enrollment projections are the birth rate projections. The report uses real and projected birth rates to estimate the future year K enrollment numbers and points to declining birth rates from 2007 as a driver of falling enrollment. The report claims that is principally due to a low percentage of females in the 20-34 age group in the District as compared to the national average.

This claim is immediately questionable. In order to justify this claimed correlation we would need to see evidence of a drop in the numbers of females in the 20-34 age group corresponding with the drop in birth rates. No such data is given by the report. Anecdotally it would seem that in a community with our demographic (as detailed in the report itself) including higher than national average educational attainment and median income, the 34-44 age group of females would also be a significant driver of birth rates. The data for this group is not examined.

More importantly there is strong evidence that the falling birth rate since 2007 is a result of the significant economic distress suffered during that period and should be viewed as cyclical rather than structural. Recent Pew Center1 research shows that the birth rate in the USA was actually climbing until 2007 and then fell off dramatically. This is consistent with the pattern seen in our district. Six out of seven of the economic indicators used in the Pew Center’s research correlated with this fall-off, suggesting a high likelihood of the recession being the driver of declining birth rates. Historical data on previous recessions support this. Moreover it is suggested in the Pew Center research that the drop in fertility rates which occurs during a recession is a postponement of childbearing rather than a decision not to have a child and that after economic conditions improve birth rates can be even higher than before as families ‘catch-up’.

The importance of the question mark over the birth rate projections cannot be overstated. Elementary school enrollment projection is extremely sensitive to such projections, as anything beyond a 5 year period from the date of study is based on children who are yet to be born. The current study itself states that after 2016 the data become significantly more uncertain as they are based solely on projected birth rates and that such projections should be used with caution2 . If the trend on birth rates differs from that suggested in the report, and the analysis above suggests there is a strong possibility of this, the entire enrollment projection model is called into question.

Other historical data While birth rates are the most significant factor in driving the enrollment model in the report, there are other factors which when looked at in terms of a 5 year historical period may present a distorted picture of future trends and thus future enrollment. The first of these is new home starts. The report states that despite the fact that as of the date of study there were some 118 potential new residential units in the District, due to the low number of new home starts in the District in the 2007-2012 period, the baseline projections were not modified to take account of any children which might enroll in the District in the future from new housing. However this data is taken from a period where new home starts were nationally at historically low rates and it seems imprudent to use them as an indicator of the potential for new home starts in the future. The national data show that new home starts are already beginning to increase from their historical lows and are predicted to continue to do so: for example according to industry data provider McGraw Hill, year-to-date as of 9/26/13, residential construction starts in the New York Metro area (which includes Westchester County) are up 34%

1 http://www.pewsocialtrends.org/2011/10/12/in-a-down-economy-fewer-births/
2 Demographic Study for the Katonah-Lewisboro School District, August 2012, p 40 over the prior year3 .

Therefore to assume no new children into the District in future years from new housing starts over and above the rate at which those occurred in the historical period also seems open to question.

 A similar picture emerges when we look at home resales. The report “assumes that the future home resale market would be similar to what has occurred historically” (p38), and yet we see here too that the historical period looked at by this report represents an atypical period in the area for home resales. It therefore again seems imprudent to try to draw a future trend from this data. It can be reasonably expected that due to improving economic conditions and evidence of improving property markets the future home resale market will be significantly better than it was over the historical period examined. This phenomenon is already being seen in neighboring and comparable school districts: the Chappaqua CSD for example saw its enrollment projections for 2013-14 significantly exceeded and had to add a number of class sections in its elementary schools at the last moment. This is principally being put down to housing resale improvement4 . It is reasonable to anticipate that a similar effect may be seen in our District in the near term. The most recent data from area realtor Houlihan Lawrence suggest an improving housing picture in the area, with 151 homes sold in the year to date as of September (comparable to 2006 numbers) and a year on year increase from September 2012 of 11.8%.5 We may already be seeing some evidence of this in our own schools - the LES K enrollment for 2013-14 was projected to be 41. It is in fact 48, 17% higher than the projected number.

Home starts and resales are particularly significant in a District such as ours where there is net migration into the District by families not native to the District seeking a quality education for their children. The enrollment model assumptions on these issues are again highly questionable and may be leading to a significant understatement of our future enrollment numbers.

Conclusions

The demographic report commissioned by the District paints a picture of ever- decreasing enrollment and half empty schools. However a close examination of the data in the report strongly calls these conclusions into question. The report fails to make the case that the trend drawn from the historical period is structural rather than cyclical. It is equally and perhaps more plausible to suggest that the chosen historical period is completely atypical and that the coming years may see climbing birth rates, recovering new home starts and a resurgent home resale market which would lead to blossoming school enrollment and a dramatically different long-term picture for our District. As will be seen in the next section, this is a key issue when looking at a school

3 per press release dated September 26 2013 4 http://chappaqua.patch.com/groups/schools/p/chappaqua-student-enrollment-hits-74-above-projection 5 http://www.houlihanlawrence.com/SSC/DB/hhl_wpm/MTGraphs/Westchester-Katonah-Lewisboro- School-District.htm

closure proposal as if the projections are indeed incorrect any closed building may have to be re-opened, which would be a significant cost for the District and would cancel out a substantial part claimed cost saving associated with the closure, especially if it was in the near term.

In addition to these issues with the historical data and the projections drawn from them, one very important factor which is not considered by the study in any way is the effect of the introduction of full day K programming. No study has been done on the effect on projected enrollments in comparable districts of the introduction of a full day K program even though such data must exist in the neighboring Mahopac School District, where anecdotal accounts suggest a significant increase in enrollment at all grade levels. The introduction of full day K could provide an immediate burst to our enrollment numbers and this should also be studied.

On the basis of this analysis it would seem that a much more prudent approach for the District would be to “wait and see” how the demographic picture changes as full day K rolls out and economic conditions continue to improve and so test the accuracy of the projections.

[2] The claimed cost savings of closing an elementary school

 The report of the School Utilization Committee details a number of options for school closings and re-organization and the cost savings estimated for each. A substantial part of the case rests on the assumption that closing 1 elementary school will save the District Budget $2.25m6. This number appears from the limited projections we have been given to be a static year on year increase, i.e. it is not projected to grow annually or with declining enrollment and it is not linked to the closing of any particular school.

This section will test the assumptions and data underlying this claimed cost saving. If this cost saving is over-estimated, the case for closure diminishes.

Research on school consolidation

Much of the research in favor of school consolidation as a driver of efficiency savings and improved educational outcomes relates to the consolidation of rural single-teacher schools in the 1930-1960 period. The post-1970 research on consolidation in modern school districts paints a drastically different picture on both counts. School closure has been associated with less educational attention for students, longer bus rides, lower participation in extra-curricular activities, and decreased participation in volunteering and school governance on the part of families7. Most significantly for this paper they have also often not been associated with the expected educational cost savings as the

 6 This figure increases if class sizes are also increased beyond the ‘goal class size’. It is presumed that there is not tolerance in the community for this step and so the $2.25m figure is being principally referred to.
7 http://nepc.colorado.edu/publication/consolidation-schools-districts

diseconomies of scale and other direct and indirect costs caused by consolidation are often not properly accounted for8. Assumptions about cost savings must be examined very closely on an individual basis to ensure their accuracy.

Claimed cost savings in our District

As noted above, the claimed cost saving in closing one elementary school in our District is given in the School Utilization Committee Report as $2.25m assuming class sizes stay at their present level. This represents approximately 1.95% of the current year school budget, or around $160 per District resident. The report does not provide a detailed fleshing out of how this figure was arrived at. There is no breakdown of where these cost savings come from, for example, we are not told how many teacher and other staff redundancies are involved in this and the cost savings involved in each of them. The community must be given much more detailed information in order to properly consider the veracity of these savings in the context of such a drastic step as a school closing. In addition various direct and indirect costs of consolidation are not considered or are underestimated. When we are talking about such a relatively small number in terms of savings the failure to fully consider these costs may have a significant impact on the overall projected savings.

Direct Costs

The report itself indicates that costs for renovation of existing buildings and necessitated changes in transportation have not been included. This represents a significant omission. School closure will have a significant impact on transportation in the District. The number of students bussed is obviously a driver of transportation costs and LES has a number of students who currently walk to school and will now require to be bussed. In addition bus times will lengthen which may also drive costs. Available studies show that school consolidation does not cut transportation costs and often increases them.9 There are also more indirect transportation costs. Longer bus times will lead to more parents driving their children to and from school and this will result in much longer car lines at school and may necessitate a staffing cost to monitor these lines. In addition no study has been done on the impact of school closure on traffic in the town. Existing traffic pressure points at the intersections of Routes 35 and 123 and Routes 35 and 121 will be put under increased strain by the closure of LES. This may necessitate an increased policing presence which could represent a cost directly to the school budget or indirectly to residents through town taxes. It should be emphasized that residents also pay town taxes and any saving in the school budget is therefore rendered meaningless for residents if town taxes go up as a result.

 8 http://nepc.colorado.edu/publication/consolidation-schools-districts 9 http://www.ruraledu.org/user_uploads/file/school_consolidation_and_transportation_policy.pdf; http:// nepc.colorado.edu/publication/consolidation-schools-districts

 The failure to account for renovations is also a potentially very significant omission. The report does not detail what renovations are anticipated, far less provide cost estimates for them. There are some immediately obvious potential costs in this area. With changes in transportation patterns additional parking may be needed in the remaining schools. Has any analysis been done of whether there is sufficient dining room space, or sufficient library space in the remaining schools? Will any extra specials classrooms be needed? Will any currently unused classrooms need to be repurposed?

The proposals which have been presented to the community envisage computer labs being re-converted to classrooms in order to make the space required to fit our students into 3 elementary schools: has this been costed, and have the costs of providing alternative arrangements for technology been factored in? The District Offices are currently housed at Increase Miller Elementary: will these offices need to be re-housed to make room for children re-districted from LES and if so at what cost? Where will these offices be relocated to and at what cost? A substantial potential cost heading has been completely unaccounted for in this area and it seems unacceptable for assumptions on savings to be made without any account being taken of this. More time is needed for this kind of study.

Also in the School Utilization Report it is noted that the expenses attached to labor reduction (i.e. the costs of laying off staff) are being assumed at 2013 levels. As these staff would be laid off in 2014, this cost may be under-represented.

Beyond what the report itself tells us is not included, it is very significant that no contingency is being included for the costs of reopening a school in the future. As we saw in the previous section, the questions which exist over the demographic projections indicate that this is a real possibility. This is likely to be an extremely expensive exercise which could wipe out much of the savings made by closure. If this had to be done in the near term, there is a possibility the whole exercise might end up costing the District money overall. This may be particularly significant in our case because the District is also proposing to introduce a full-day Kindergarten program in 2014-15. Anecdotal evidence from nearby districts who have recently introduced such a program show a significant increase in enrollment at multiple grade levels upon its introduction. If this was to be repeated in our District it is reasonably foreseeable that a re-opening might be required in the very near term which would be extremely damaging financially as well as a huge burden for our children and school staff in terms of disruption.

The numbers given in the School Utilization Committee Report are somewhat misleading in this regard: the ‘peak enrollment’ numbers for the elementary schools (with the exception of Meadow Pond) relate to the late 1990’s. These numbers suggest that there is significant excess capacity in the buildings and that therefore enrollment would need to increase massively in order to create a pressure for re-opening, and are also the numbers which have been used to support the talking point that “our schools are only half to two thirds full”. However changes to mandated services since that time have changed the demands on room space in the buildings, for example in requiring separate rooms for RTI services, and the maximum capacity of these buildings is now significantly less than these peak enrollment numbers.

Therefore any substantial change in projected enrollments due to either a different than predicted demographic trend or the effect of full-day K programming could lead very quickly to a choice between over-crowding and larger class sizes or an expensive re-opening. Therefore it is clear that a detailed plan and costing needs to be developed for this scenario and a contingency added to the projected savings to cover this. Again this possibility indicates that a ‘wait and see’ approach may be more prudent.

These are just some examples of the direct costs which have not been included in the projected savings and a much more detailed analysis must be undertaken before any claim can be realistically made about efficiency savings

Indirect Costs

As well as the types of direct costs noted above, there are indirect costs to the wider Katonah and Lewisboro communities which have not been considered. Research shows that closing a school is a considerable economic detriment to a community10. It can result in declining home values and lost property taxes, thus exacerbating any decline in enrollment or damaging the District’s revenue stream. It can also harm local businesses by decreasing the economic activity in the area of the closed school and cause infrastructural problems such as changing traffic patterns. As far as we are aware no analysis or study of these potential effects and costs for our community has been undertaken or is proposed and these issues again cast doubt on how real any savings would turn out to be. Detailed study of the economic impact of this plan on our town should be undertaken to reassure residents that what looks like a saving on paper will not end up costing the town much more in other areas.

Conclusions

It cannot be assumed that closing a school will result in cost savings. Modern research on this issue challenges this assumption and suggests that the costs of consolidation are often underestimated. Therefore a projection of cost savings on the basis of a school closure must be subject to detailed scrutiny.

Much of the data which has been provided by the District on the claimed savings to date is soft. Their projections must be spelled out in a great more detail for the community and a significant amount of further research will be required to fully account for some of the missing direct and indirect costs. The financial effect on the Town of Lewisboro must also be fully costed as it does not benefit the taxpayers in the Town if savings in the school budget lead to an increase in town taxes or a decrease in economic activity in the town.

A decision of the magnitude of closing a school requires robust financial data to support it. We have not been provided with this to date and the cost savings may well be

10 Lyson, T.A. “What Does a School Mean to a Community?”, Journal of Research in Rural Education, Winter 2002, Vol 17, No. 3, p131-137

significantly less than claimed. This would erode the case for closure substantially. No decision should be taken until a much more detailed and substantive analysis of the data is undertaken and a fully fleshed out and audit-able financial projection given. Only then can the community properly consider the necessary cost/benefit analysis.

[3] The overall budgetary context

 The final piece to consider is how the proposed closing should be considered in the context of the overall school budget. The first thing it is important to note is that falling enrollment does not increase costs to the District. In fact falling enrollment saves the District money as it educates fewer children on the basis of a similar tax take. This can be seen in the 2016-17 projections of the School Utilization Committee: under their own analysis, by doing nothing other than letting enrollment decline, we will save just under $1m in that year’s budget11. It is very important for the community to understand that falling enrollment is not the driver of costs in our District and to ask what those drivers of cost are, how they are being addressed and how this proposal should be understood in that context.

The current budgetary picture

The 2013-14 school budget is $114.9m. In his notes to that budget Superintendent Paul Kreutzer writes in strong terms about the significant structural budgetary challenges which the District faces and points to a $2.5m structural deficit in the 2014-15 budget with a worsening picture in the future. He rightly states that these challenges must be addressed.

These challenges must be understood in terms of both the revenue side and the cost side.

Revenue Side
 
Due to the NY State tax levy cap, the District’s ability to increase revenues though raising taxes has been limited and costs are growing at a faster rate than the District is permitted to raise taxes. Focus has therefore been placed on the cost control route to balancing the budget. While this is necessary, it is equally important not to neglect the revenue side and to understand that tax increases need not be the only way to increase District revenues. Making our schools increasingly attractive and therefore supporting our property values is an indirect route to stabilizing and improving the revenue picture. Conversely, if as noted above, school closure leads to decreased educational performance and a reduction in economic activity in the town then this may have the effect of harming the revenue stream further and actually worsening the budgetary picture. Cost cutting which significantly negatively impacts the educational and broader attractiveness of our towns is short-sighted in this regard.

 11 even including the cost of having a full day Kindergarten program.

 Cost side

The major driver of cost in our school budget is labor cost, which represents 80% of our total budget. While some of our labor costs are mandated and are outside the District’s control, many components of our costs are not and our labor costs are substantially higher than comparable districts. Because these costs represent such a large proportion of our overall budget, addressing these costs is the key issue for overall budgetary stability. A 1% saving on labor costs represents the same saving as cutting all other costs by 4%.

Superintendent Kreutzer recognizes the significance of controlling these costs in his notes to the budget, writing “the key component for closing [the structural budget] gap relies heavily upon securing contract concessions with organized labor units". The current plan viewed in this context

In the previous section it was strongly questioned whether the cost savings claimed for a school closing would be realized in practice. That analysis aside, even if we assume that projected saving to be correct, when it is viewed in the context of the overall budgetary picture, its limited impact can be seen.

 As has been previously noted the projected saving in the 2014-15 budget after closing an elementary school is $2.25m. This amount is the equivalent of just 2.4% of current budget labor costs. This is of significant interest because of the current status of labor cost negotiations in our District. There are active negotiations on-going with the unions. Healthcare costs represent a significant proportion of the labor cost in the District and a substantial future reduction in these was agreed subsequent to the publication of the 2013-14 budget.

In May Board of Education president Mark Lipton said that the new healthcare plan that has now been agreed by all parties will save ‘millions’ annually 12. In addition salary negotiations are ongoing and could represent a further reduction in overall labor costs. These future savings have not yet been included in the budget projections shown to the community and yet they could dwarf the claimed potential savings from shuttering a school. It is again not clear why the District is rushing to a decision on this rather than awaiting the final outcome of all these negotiations or waiting to see how the saving that will come when the new healthcare plans are fully implemented will affect future budget projections.

Moreover, focusing on school closure at this time takes attention and pressure off the necessity to control our labor costs. The problem with our labor costs is not just the overall amount and percentage of the budget they represent, it is also their rate of growth. For example the average teacher cost rose by 6.22% from the 2012/3 budget

 12 (http://bedford.patch.com/groups/schools/p/kl-board-union-back-insurance-change- retirement-plan)

 to the 2013/4 budget. If the rate of growth of labor costs remains unchecked and the rate of revenue increases remains capped, then given that labor costs represent the overwhelming majority of the budget, the gap between costs and revenues will quickly widen again irrespective of whether we close a school and we will be facing exactly the same structural problems again in a very short time. If we set the precedent that we can cut core educational services to feed the ever-growing cost-drivers in our District, then it must be asked what we will cut when this problem soon re-appears. Will we stop providing Kindergarten, or bussing, or art and music specials? This question must be addressed whether a school is closed or not and it would seem more sensible and fiscally responsible to tackle the real drivers of cost now, rather than later. Furthermore, the impact of the full budgetary picture on the broader cost/benefit analysis which our community will be asked to make should be considered. Residents may imagine the cost saving from a closing being applied to lower their taxes or to enhance the services provided to our students. However as noted above the reality of an unchecked growth rate in labor costs in a tax cap environment will be an ever decreasing slice of the pie for students and any savings will being sucked into supporting those ballooning costs.

Most significantly in the current time frame, it can be argued that the putting on the table of a school closure at this time actually harms the District’s ability to cut costs by weakening its bargaining position in the labor negotiations currently taking place. It is much more difficult for the District to argue for concessions to be made by the labor unions when they can point to cost cutting opportunities that exist elsewhere even though these are not the key drivers of our budget problems. It takes the focus of our community off these negotiations and their importance for the District’s future financial stability. There is a concern that with this focus diverted and less pressure being applied to the negotiations we could end up tied into new labor contracts which do not do as much as they need to control our inflated labor costs. The damage this would do to the long term financial picture cannot be overstated. The ongoing negotiation provides a unique opportunity for the District to begin to right its fiscal trajectory and this is where all our focus should currently be directed.

 [4] Conclusions

 This paper set out to question the data which has been presented to the community to date and the conclusions we have been asked to draw from it. Our community has been led to believe that there is an incontrovertible data-led case for school closure. This paper demonstrates that this is not the case. The demographic doomsday picture painted of our future enrollment is open to question, the claimed cost savings of school closure are ill-defined and inadequately researched and the effect of the closing on stabilizing our school budget in even the short to medium term is negligible. Closing a school is a decision of great magnitude and it can only be concluded that the data examined here are not strong enough to support it.