o For general discussion of issues concerning public education in DC, please refer back to the previous chapter.

o For more detailed information on relevant reports on DC public education and related topics please refer ahead to the next chapter.

o For a presentation of relevant statistics and projections on the DCPS system, continue here.

o Click here to jump ahead to NARPAC's somewhat critical commentary, recommendations, and conclusions

o This more recent section describes the extraordinary existence of 25 vacant or abandoned buildings marring neighborhoods around the city.

o And based on a 2003 GAO report dealing with DC's supposedly hopeless financial 'structural imbalance', NARPAC has done a superficial analysis of DCPS's own structural imbalance as a result of planning to continue modernizing too many too small schools for too many students unlikely to materialize.

o This, in turn, is re-emphasized by the fact that there have been no school closings despite a 14,000 student enrollment drop over the past six years. This results is substantial lost funds for DCPS.

o And the latest draft version of the 2003 DCPS Facilities Plan revised downward its expectations for future enrollment, but not enough to keep up with reality. It also shows how many kids 'migrate' from their neighborhood school to another preferred by their parents.

o Fast-forwarding two years from when this chapter was written and updated, there is news of a new 2006 Master Education Plan, and new hope that a substantial facilities downsizing program may become part of that effort. Keep tuned.


NARPAC has finally had the opportunity to study the full DCPS Facilities Modernization Plan (FMP) that was two years in the making and finally published at the end of 2000 with limited copies. The FMP is an essential element in the future of DCPS--and of the quality of life in the nation's capital as well. The plan is very thorough, assessing the condition of every DC school, and making recommendations heavily based on community inputs as to whether each school should be modernized or replaced in toto, and to what new size. According to School Board member Robert Peck, responsible to the Board for facilities oversight, the work was predicated on four ground rules:

1. The concept of neighborhood schools (not requiring buses) should be maintained, except, of course, for the large contingent of special ed kids (whose failed busing system is a city disgrace);

2. School buildings should be considered community assets and, where space is or can be made available, should provide for other community functions as well , such as health clinics, meeting places, or even police substations (NARPAC has strongly supported this concept);

3. The number of schools should not be decreased, despite declining public school enrollment, and present sites re-used. Existing facilities should be downsized if appropriate. Some 30 schools were shut down early in the '90s, and their disposition is not yet complete (NARPAC does not support this arbitrary rule); and

4. Many of the newer schools were built under the fleeting theory of 'open classrooms' which have proven unmanageable. Modernization should eliminate such open classroom configurations, making some of the older schools better candidates for continued modernization.

A 'Committee of 21' (C21) was assembled to oversee the development of the FMP, and they made certain decisions to standardize the estimates and the modernization process. For instance, the ideal school size for each level was established by polling and averaging. The obvious planning steps were followed: a) estimate the future school population; b) assess the condition of every school; c) select a nominal future enrollment level for each; d) decide whether each school needed a partial or full modernization, or complete replacement with community inputs; and e) make "planning estimates" of the costs and order for doing the work. They do not appear to have been in any way constrained by costs, and only marginally involved in overall community redevelopment.

The net upshot of the FMP includes the following:

o using 'high', 'moderate' and 'low' estimates for both the total kids in DC public schools and those choosing charter schools, the C21 "picked the one in the middle" and projected a slight growth in DCPS school enrollment from 68,600 to 72,600 kids in 146 schools by the end of the planning period. This is shown as the upper three projections on each of the charts reproduced below:
This is also the source of some of NARPAC's greatest disappointment with this plan. First, the combination of 'high' and 'low' alternatives leaves off perhaps the two most likely alternatives: continued decline in total kids, and continued increase in charter school attendance (the lower two lines). These two projections would size the eventual enrollment between 55,000 and 50,000 not 72,000. But it also assumes that existing trend lines can make abrupt changes (either up or down) in the immediate future. Unfortunately, the die is cast by recent birth rates, and the lower numbers of kids already started through the 13-year process (This will be addressed further on under NARPAC Comments).

o There are sizable uncertainties in the future growth of the charter schools, now proliferating in DC as in almost no other US city, but also in the shifting enrollment in the area's private schools, many of which, of course, are the outstanding Catholic Schools. Charter school enrollment has jumped to almost 10,000 in three years, while private school enrollment has dropped below 10,000 (data incomplete). In any event, defections from the DCPS are almost certain to continue for several more years until some of these long range DCPS systemic improvements become unmistakably clear Private school trends do not appear to have been addressed by the C21 or the FMP.

o The basic judgment to plan for 70,000 in DCPS indefinitely then (conveniently, some might say) permits elementary schools to be sized for 400, middle/junior highs for 500, and high schools for 900 while maintaining the 'proper' space per student and exactly the same number of schools (146) as currently in the DCPS system (see below). These students-per-school targets are low compared to most surrounding jurisdictions where school populations are growing, per-student costs are lower, and school achievement scores are uniformly higher as well.

o Based on 'ideal' space allocations per student, the current system has 3.7 million excess square feet of floor space some 25% of the current total of 14.8 million (lower left, above). With DC office space now renting for almost $40/sqft per year, this space has the hypothetical potential to bring in $150 million in revenues to the school system! But in the school system, this excess floor space is costly, and will be eliminated in newly built/fully modernized schools, or somehow re-used in schools due for only partial modernization. Note that the middle schools have the most excess space, and the elementary schools the least. It should also be noted that this discussion refers to school building floor space, not site land space which is estimated to be some 25.4 million square feet (over 580 acres).

To develop 'planning estimates' of construction costs, separate values per square foot have been chosen ranging from $216 for new high schools down to $115 for partially modernized elementary schools (lower right, above). These may well turn out to be low unless the current building boom abates.

For convenience, the FMP divides the city's schools into eight lettered planning areas, similar to, but not the same as, the school districts. And also for convenience, NARPAC has lumped them into three regional clusters: ABC, 'East of the Anacostia' (new Wards 7,8); DEF, 'the Middle City' (new Wards 1,2,5,6); and GH, the 'Northern Tier" (new Wards 3,4) as shown on this diagram.

o Based on these planning assumptions, and the judgments made on the conditions of various schools, some 63 elementary schools would be totally replaced and 42 schools would get a "full" or "partial" modernization over 10 years; 10 of the 23 middle schools, and 14 of the 18 high schools would also be totally replaced. The largest number of new schools (48) is planned for the Middle City, and the largest number of modernizations (40) would take place East of the Anacostia (upper right, below):

Several criteria were used for determining which schools to totally replace or modernize. One of course was age: DCPS has no schools younger than ten years old, and has eleven between 91 and 110 years old! (lower left, above). Another 59 schools were first built more than 50 years ago and expanded and/or modernized since. The real irony, however, is that many of the younger schools were built with "open classrooms" which are (justifiably) no longer acceptable, particularly in school systems with large number of often-disruptive special ed kids. But throughout the system, deferred maintenance has made some of the schools virtually uninhabitable, some with essentially no usable plumbing for kids or staff! This situation alone has been cause for discouraging any national pride in our nation's capital city.

The total cost of the 'planning estimate' is about $1.6 billion, with almost $900M going to the large number of elementary schools (lower right, above), and the 'Northern Tier' region getting a relatively small amount of the total capital investment (fewer schools, more modern now).

o Within the FMP documentation are some highly instructive data on changing school enrollment. The drop in DCPS school enrollment has been substantial over the past seven years from 80,000, to about 68,000, with the decline being proportionately larger in the lower grades (which thus portends the future). With regional grouping, it is clear that the decline in students is largest East of the Anacostia, and smallest in the relatively stable Northern Tier (see left). This is confirmed by population trends identified in Census 2000. It is perhaps noteworthy that the distribution of students between the grades also differs by region within the city. The higher grades suffer from more drop-outs in the poorer sections of town, while the numbers of pre-pre-Kindergarten and special ed kids (identified as "ungraded") are lower in the Northern Tier (at least in the public school system).

o It is also significant in projecting future enrollment requirements to note how the number of students in a particular year-class ("Class of '08",say) declines from entry through twelfth grade. The available data do not provide any reasons for this drop-off, but it is remarkably consistent across the years from the Class of '98 to the Class of '10 (so far). Although the big drop-off in the high school years may be primarily attributed to dropping out, the earlier years are more likely due to moving away, or shifting to other school options. One way or the other, however, this trend seems to be relatively immutable (upper chart, below). For the 'Class of 2000" in particular, (lower right, below) only 2800 of an entering class of 6200 will even be enrolled in 12th grade, much less graduate. Again, the drop-off rate varies with parts of the city and may in fact, represent some students transfers within the system:

o The lower left-hand graph (above) also shows in finer detail the change in enrollment in the various grade clusters over the past seven years, but in "schools-worth" of students (only from NARPAC...), rather than individual students. There is a gradual decline across the board except in the "ungraded" special ed kids (many of whom do not actually attend DCPS schools, but are nonetheless assigned to them). Though these drop-offs may not seem large, given the "ideal school sizes", the change is enough to warrant closing five pre-K schools, ten elementary schools, seven middle grade schools and three high schools! In other words, these gradual declines should impact on the FMP.

return to thetop of the page NARPAC Commentary:

NARPAC is often presented with the quandary of whether to describe a newly discovered half-glass of fluid as half-full or half-empty. To press forward, it is necessary to point out that the glass is not yet full, but not to discourage those who have the task of filling it further. In this particular case, we are guided by what we have long considered to be one of the major elements in the revitalization of our nation's capital: the development of a first class school system.

As we note at length elsewhere, the schools cannot produce world-class "output" without having attentive, receptive "input" in the forms of both the kids and their parent(s). It is incontrovertible that the "inputs" to the city's least successful schools come from the most impoverished, disadvantaged communities. We thus conclude that the schools must radiate their positive influence into their communities, and where possible, become beacons of hope unfettered by counterproductive political or religious demagoguery, poor health, or parental neglect. NARPAC wants nothing but the best public school system Americans can devise.

That said, NARPAC is delighted that a sensible planning process has begun, and in the spirit of positive thinking, points out several areas where it needs to be improved, particularly because the city is in flux, because funding for education is limited primarily by other needs of DC's many disadvantaged, and because so much is at stake. These individual areas follow:

DCPS Enrollment Projections Miss the Mark

NARPAC analysis of the data presented in the FMP indicates two kinds of shortcomings in the projection of student enrollment. Both errors favor a significantly larger than likely student population ten years hence.

1. Combining Scenarios

Future enrollment levels are determined by several factors, such as live birth rate (major driver, apparently), influx of families with school age kids (unlikely), change in attrition (drop-out?) rates, and the largest uncertainty: continued shifts to charter schools. FMP chose three scenarios which balance out around a continuation of current enrollment levels which conveniently supports the current number of schools at the most desired sizes (see above). However, as mentioned above, if the full range of negative scenarios had been displayed, a considerably worse 'worst case' would have shown up. Continued low (but not lower) inputs, combined with the higher estimate of transfers to charter schools, yields alternatives much closer to 50,000 than 72,000 ten years hence a difference of at least forty schools.

2. Including School Enrollment Dynamics

More serious, in NARPAC's view, is the failure to include the 12-yr dynamic. Be it a baby boom or a baby bust, the resulting "bow wave", or "bubble", rolls through the system slowly. Today's DCPS senior class entered the system six years after the peak of DC's last baby boom. Today's DCPS first graders were born when live birth rates in DC were 25% higher than they are now. The net impact is that today's G-1 enrollment will continue to decline predictably for five more years, and the graduating class will continue to decline for 15 years! Only a major and very unlikely--influx of DCPS-bent school-age kids can change this. The spread in combined scenarios is larger, but a plausible median is much lower (30-40,000 students by 2015). Hence NARPAC's projections, shown below, are substantially lower than those in the FMP:

It should also be noted, but not belabored, that the bookkeeping for, and the actual enrollment of DC's 8 "schools-worth" of special ed kids is not fully addressed. Many of these kids now attend special private schools at huge cost to the city. Hopefully the city plans to bring more of them back into its own care. For some information on the composition of this population, see characteristics of special ed kids.

Sample Demonstration

To illustrate how this happens, the graphs immediately below show the enrollment by grade in 2003 (in red) and the expected enrollment in 2016 (in green) if the number of K kids stays constant at 4000, and the year-to-year attrition rate of any given class stays at the average of the past ten years. The net result is a drop in enrollment from 55,000 to 40,000, a reduction in needed schools from 110 to 83, and a different ratio of schools at the three levels. As a rule of thumb, school enrollment from G1 thru G12 levels out at 9.5 times the size of the G1 class.

The Role of DC's Live Birth Statistics

According to the DCPS FMP, the leading indicator for how big the next first grade class will be entering the DC public school system is to look at the number of live births occurring six years prior. In fact, the correlation is very high indeed strongly suggesting that the primary "input" to the DCPS system is the offspring of people already living in the district and starting families. The overriding importance of this clue is that those live birth statistics provide six years advanced warning of the likely size of the DCPS school population for the next twelve years. This is clearly shown on the graphic below:

The question then arises (to analysts, at least) as to what is driving the live birth statistics, and NARPAC's answer came from an unexpected source: a recent report on recent changes in the TANF welfare-to-work caseload. It noted that DC had recently received federal awards for reductions in the rate of out-of-wedlock births. This clue led in turn to the statistics published in DC's INDICES showing the live births is DC since 1991 (now updated by NARPAC through 1999) and the portion of those to unwed moms. As is shown in the companion graph below, the reduction in single-mom births has not been replaced by an increase in married-mom births. In fact, the latter has remained essentially flat. NARPAC surmises that because single moms generally have less means than couples, their kids are more likely to enter the public school system and are, in fact, less likely to have the mobility to leave the District. According to the Urban Institute, 40% of all US kids raised by single parents are poor, compared to 9% of those with two parents available. In fact, the incidence of single-parent school kids in DC is extremely high. The latest available statistics for single-mom births are also shown below:

This leads inescapably to the sensitive conclusion that:

the primary driver in the future size of the DC public school system is the number of out-of-wedlock births, and that the more successful the welfare programs are, the smaller the DC school system will become!

Time Delays in the School Building Program

It should come as no surprise that the process of building a set of new schools is a long process perhaps five years from the decision to build, to the acquiring of the funds, to the completion of the jobs. The upper chart below shows the currently anticipated funding profiles starting with the program already funded and underway (Tier 0), through the next four planned tiers which will only be completed in FY2009. That program will require steady funding of at least $160-170M annually. Right below it is the FMP planning estimate for the time required to finalize the specs, design, and build a new school in each category. School availability thus lags plan approval by three to four years, and increases the premium on realistic projections several years in advance.

Finally, the next graph, below, shows the design enrollment level for the 30 brand new schools planned in Tiers 0, 1, and 2. While there is nothing absolute about these current plans, they are comparatively well along in the development cycle. The point is simply to show the relationship between the space available in the old schools, the actual enrollment in these schools five years ago and in 2000, and the projected enrollment in them when completed. In particular, it shows some reasonable correlation between past and future for the elementary schools, but seems increasingly unrealistic at the higher levels. The communities and the C21 appear to be forging ahead with plans to build more capacity in the new schools than are current enrolled in the schools they will replace. (Not shown is the fact that the average sizes will be very close to the agreed ideal: 411 for the elementaries; 520 for the middles; and 810 for the high schools.) But the plan seems far more a "wish list" than a realistic planning exercise. This does not bring about the end of the world, but it strongly suggests to NARPAC that the communities who get their schools early are far more likely to realize them than those further down the modernization priority list in Tiers 3 and 4. The end of universal community schools may then be blamed on poor planning, not intent or intentionally broken promises made to the communities.

Common School Designs for Economy

Related strongly to project duration, and at least secondarily to construction costs, is the matter of design originality. It must be somewhat cheaper and surely quicker-- to re-use standard school designs rather than starting each brand new one from scratch. Interestingly enough, there are no two schools of the thirty above that are designed to be same "ideal" capacity, and it appears that new school design will be primarily driven by community choice as to size, plus restrictions imposed by using current lots (see below). Nevertheless, if schools could be standardized to one of three designs in each of the three categories of schools, then NARPAC estimates that some $183M of the estimated $1600M could be saved, based primarily on design costs. Benefits from 'learning curves', competition among contractors, and common procurements could easily raise this to $250M

Economies of Scale in School Buildings

A favorite expression of cost analysts--and a complete bugaboo to humanists--is the notion of "economy of scale", which almost invariably leads to fewer units (of any sort) of larger size as a means of reducing operating costs. It is why office buildings, CostCo stores and Boeing 747's are as big as they are, even though they may dehumanize their occupants. Because each school has a fixed number of principals and assistant principals, and certain minimums in teaching and custodial staffs, it is possible to show as a ridiculous extreme that one big school for 40,000 students would be much cheaper to operate (per student) than 80 schools with 400 kids in each. Some colleges now support this view! But in between, there should be some room for compromise particularly if the outstanding estimates of enrollment requirements are uncertain.

Hence the chart below (using notional assumptions at best) is drawn from earlier NARPAC analysis on school sizing (look for school "pryamids" in a discussion of the rising costs of systems with shrinking school enrollment). This new chart shows four different solutions to the needs of a 48,000- student school system -- involving at one (left) extreme,120 schools of 400 each, and the other (right) end, 60 schools of 800 each. The larger schools are cheaper to run at design load on the basis of operating costs per student, and the penalties for operating 25% below that design occupancy are less. Hence a school system of 80 schools might cost $8000 per student at design load, or $9200 per student with a 25% reduced occupancy, but the smaller 400-student school system would see its per student costs rise from $9800 at design load to $12,000 if each one operated at 75% of capacity.

Public Schools as Community Centers

One potentially valuable use for excess capacity in the DC public school system would be to encourage the housing and cooperative use of other neighborhood functions, from the most obvious ones (adult education and health clinics) to the most controversial (day care centers and police substations). The sharing of staff members would be a very valuable plus (e.g., "school nurse tends senior citizens"; or "off duty cops coach high school sports"). Obviously, however, this cannot be at the school budget's expense. To the contrary, it becomes one potential downstream virtue of unintentionally building excess modern capacity. NARPAC has no idea whether current DC laws would allow/encourage these community uses, or whether DC's ever- separate bureaucracies could be coerced into working together to solve common neighborhood problems, but it would certainly be worth a try. NARPAC urged the new DC School Board Chairman in this direction in a letter late in 1999.

Land-Swapping for Fiscal and Community Benefit

Another innovative albeit bureaucratically-challenging--option is the seldom exercised potential to trade off existing school properties (with or without schools on them) for better suited modern school properties (generally smaller). The results can be more efficient modern schools, while benefitting either financially or in community revitalization by the better utilization (by someone) of the old property.

The outstanding current example of this creativity at work is admiringly known to NARPAC as the Great Oyster School Caper. In this instance, local activists realized that the Oyster School, 1926-era, decrepit, and shrinking in enrollment was sited on an oversized lot in a very high rent district of Northwest (Cleveland Park). With great intrigue, the activists found a developer who would be willing to construct a new school on one part of the lot ceded to him if allowed by the community to build high-density high-rise apartments on the rest of the lot. As a result, the community got a brand new, properly sized school house at no cost, the developer got a profit-making enterprise, and the city got a source of significantly increased tax revenues. A rare win/win/win situation for restoring pride to America's capital.

In a different sort of win/win/win situation, an innovative developer, in the process of building a number of up-scale town houses near Lincoln Park, SE, was approached with a request to take over and revitalize the large, now-surplus Bryan High School nearby. The empty 2-building facility had become a blight on its picturesque neighborhood, but the original part was deemed to be an historic site, not to be torn down. The developer is now in the process of turning the old school building and its demolished adjunct, into additional upscale apartments, using profits from the sale of the nearby town houses . He has been able to synchronize the style of the new town houses and the facade of the rebuilt school building with those of the period houses on neighboring streets. A useless property has become a community resource, in keeping with its surroundings, and at no cost to the city, or the school system.

These are two examples (there may not be many more) of transferring outmoded school-owned properties into productive business enterprises while serving the needs of the community and the school system. There are of course other reasons for considering land-swaps. In the more common case, there may be a substantial mismatch between the size of the future school and the size of the site on which it currently sits. In this case, it behooves the school system to dispose of the excess land so that it can be put to productive (revenue-producing) uses. In other cases, the school may be sitting on land whose value is now far greater (for other uses) than can be justified only for school use. In still other cases, there may be good and sufficient reasons to take over some other nearby tract of unused, blighted, or even "brownfield" land for the community's welfare while relinquishing a more valuable but ill-sized piece for other development.

NARPAC has done considerable analysis on the 'productivity' of DC's limited land and workforce in terms of tax revenues generated vs. city service expenditures required. For instance, for high-density office, apartment, or hotel buildings, the city should be able to product net revenues of from $4 to $5 million annually per acre of the development. Single-family residences, on the other hand, seldom break even in terms of the cost of their services relative to the taxes returned to the city. Schools, on the other hand, only generate revenues through the income taxes of their staff, while having gross operating costs near $10,000 per student. Hence a middle school designed for 500 students (85,000 sqft) may sit on 2.5 acres of land (108,900 sqft) and thereby consume $2 million per acre per year from the city's school budget. As a typical NARPAC rule of thumb, then, the city needs half an acre of high-density development to pay for each properly utilized acre of school property.

Here then are two quite different examples where school properties might be traded off to provide increased city revenues if, of course, the local community residents could be permanently anesthetized:

1. Wilson High at Tenley Circle: Wilson High is arguably one of DC's better schools with a full and stable enrollment of some 1500 students, many of whom go on to college. Nevertheless, it sits on a 10-acre site, one end of which is literally within sight of the severely underutilized Tenleytown Metro Stop. The school can almost certainly justify the use of seven acres, and its annual operating costs probably range about $14 million. The challenge then, is whether a three-acre parcel could be split off and sold to developers that could generate $14M per year. If so, the developer might well pay for the complete modernization of Wilson High!

2. Kingman Park 4-School Campus: Moving to one of the city's much less fortunate neighborhoods of public and subsidized housing, Kingman Park sits on what is to DC a mountain top a rise perhaps 400 ft above sea level, looking east down on the Anacostia River (across the Langston Golf Course), and looking west down the length of Maryland Avenue, NE to the US Capitol Building. Here on a little over 30 acres sit four severely underutilized schools: two high schools with less than 1200 between them, a middle school and an elementary school with about 900 students between them. At most, these schools need a total of ten acres and it by no means obvious that this is the best location for them, nor need they all be together. In any event, there is a reasonable case to be made for using 20 acres of land for mixed- if not high-density development which could easily generate total annual revenues between $40 and $100 million dollars for the city.

Awash in Surplus Land

One major reason to consider land-swapping may be to extricate a school from a wrong-sized lot. It may not be feasible--as in the case of Oyster School--to share the lot with a developer. In this case, it may be more productive to seek an alternate location of the proper size. As nearly as NARPAC can tell, the DCPS now owns substantially more land than it needs. If the site acreage can be scaled down in proportion to the reduction in school house floor space, then the system has 130 more acres than it needs even to support the FMP's design enrollment of 72,600 students. If NARPAC's estimate is closer to the mark that the design enrollment should not exceed 40,000 students then there could be as many as 330 surplus acres available for disposition. These are very large numbers: high-end office rental, or high- productivity development of these acres could essentially pay most or all of the bills for the remaining school system. Surely the sale of them would pay for the reduced-size school modernization plan. The chart below indicates what NARPAC estimates might be realized to the DC Treasury annually from either the rental of the buildings or the development of the excess properties. It compares those revenues to the annual budget of the DCPS at either the FMP-level or NARPAC-level of enrollment, using a simple estimate of $12,000 per student.

NARPAC Conclusions

NARPAC reluctantly concludes that the FMP as it exists bears no relationship to the eventual needs of the DCPS system even though its improvement might not change plans for near-term modernization at all. It seems to be a poll endorsed by a committee, not a plan endorsed by management. We therefore recommend that:

o DCPS should continue to take advantage of this initial FMP document with its excellent format, comprehensive data, and time-consuming community inputs and gradually convert it from what is in essence the output of neighborhood polling into a plan which includes the rigors of management discipline and realism;

o The first and most essential step is to come to grips with the real-world implications for the school system of real-world demographic changes in the city which will impact on the needs of the school system for the next decade. NARPAC would not be surprised if the DCPS school enrollment in 2010 dropped below 40,000 to perhaps exactly have of the projections in this FMP;

o The bigger the change from the "convenient" neighborhood-based enrollment projections, the more the DCPS will need to look at realigning its facility plan with much more emphasis on disposing of surplus and inappropriate lands, and taking into consideration operating costs, standardized designs, joint neighborhood use of the much fewer school centers; and creative land-swaps with imaginative developers.

In conclusion, and with no intent to offend, few if any other changes in our nation's capital city could be more beneficial in the long run than to fully meet the demands of DC's long-neglected school kids with an efficient and dedicated system half the size of the current one, but truly providing a world-class education to those in the city whose future most directly depends on it. NARPAC would like nothing better than to stimulate this transformation.


Not only has the DC Public School System been underutilizing its active school buildings (see preceding sections), its has failed to let go of some 25 aging school buildings, some of which have been closed for many years. Perhaps the most attractive is shown to the left here, the old Jackson School on R Street, NW, across from Montrose Park in the heart of Georgetown. It is not hard to visualize this site becoming used for a $10M home, or set of apartments

NARPAC has not done a detailed inventory of these buildings (some do not even appear to be at the addresses advertised), but estimates that they probably were built to house at least 15,000 kids, and occupy some 75 acres. The sale value of these properties should certainly exceed $100M, or bring in $50M in annual revenues if intelligently redeveloped. It is difficult to understand how DC can ignore these potential sources of city revenues at a time when it is claiming that its rapidly rising budget is inadequate.

Two seemingly typical vacant and underused school buildings include the Kingsman School in Ward 6 (below, left) and the Old Emery School in Ward 4 (below right).

NARPAC Commentary:

NARPAC's president testified before the DC Council Committee on Human Rights, Latino Affairs, and Property Management on October 13th, 2002. An edited version is provided below:

"Schools don't make dumb kids, dumb parents do"
"Poverty doesn't breed ignorance, ignorance breeds poverty"
(NARPAC editorial, 7/02)

The conditions of our schools, their kids, and their parents are matters of serious national concern. Let me make several assertions:

1. The output of DC's school system, like that of other inner cities, is a national disgrace. These kids' performance cannot be fixed just by adding monies to the DCPS budget. And wasting money on out-dated, half-empty facilities denies those funds to other city programs where they are sorely needed.

2. The DCPS Facilities Modernization Master Plan vastly overestimates future enrollment in DC public and charter schools (and is costing $50K per expected kid). The simplest predictor for total enrollment in a given year is DC's live birth rate 12 years prior. Total birth rates, teen birth rates, and unwed mom birth rates are all declining: so will total school enrollment over the next decade.

3. A typical DC public school now enrolls about half as many students as its suburban counterparts, and is hence more costly to run. Beyond doubt DCPS has too many facilities for its foreseeable student needs. Only a fraction of that surplus is on your current list. A further fifty schools (50) would be surplus if DC schools averaged as many kids as the rest of this metro area.

4. But improving the human output from DC's schools requires improving their human inputs. Many kids arrive each day from substandard physical and family environments. Assets must be added to blighted communities to break the cycle of ignorance. Otherwise, this self-perpetuating cycle will keep generating kids from failing parents (mostly unwed, drop-out moms) to fail in DC schools. It's a "closed-loop process": pregnant kids slink out the back door to produce kids to send back in the front door. DC needs to break the cycle of ignorance in the nation's capital, and the poverty it breeds.

Let me try to shock you, but not offend you. If DC's low-scoring kids-turned- parents were manufactured products, they would be recalled, taken off the market and reworked at the manufacturer's expense. DC and other cities have lacked the will and the means to replace the "faulty parts", re-cycle the "used materials", and remove the "toxic waste" of each trapped and helpless parent. Parents who unintentionally "pollute the environment" by spawning the next generation of failing kids, and relying for a lifetime on the city's taxpayers.

The Federal Government understands this problem. It just paid DC a $20M bonus for reducing out-of-wedlock births by 200 babies. Shouldn't that bonus go to some sort of "Cycle Breaker" program for the parents? The problem isn't just "faulty" kids, but the thousands of "faulty" parents already doomed to non-productive, high-risk lives, producing more "faulty " kids.

Today's challenge is how to dispose of 25 old schools on the current surplus list. NARPAC urges the Council to seek an innovative way to recycle these assets so they can help solve the disgraceful failure to educate thousands of recent DCPS drop-outs and marginal graduates who are now moms:

1. The Council should ordain a non-profit firm, or city agency, to break the cycle of ignorance that leads to the cycle of poverty, and worse. Transfer all surplus schools to this "Cycle Breaker Agency" for three types of disposition:

sell the most salable facilities to the highest taxpaying bidder and transfer the proceeds to this "Cycle-Breaker Agency";

rebuild a few (perhaps one each in Wards 1, 5, 7, and 8) as live-in "GED boarding schools" for young unwed moms with their kids;

lease a few of the most lucrative properties (like the Franklin School) to developers to provide a steady funding source for staffing and operating these "boarding schools" as (young) adult education centers.

2 The Council should require DCPS to provide a more rational estimate of future enrollment, and gradually transfer additional surplus schools to the Cycle-Breaker Agency to raise its output to at least 1000 young adults per year.

I urge you to be bold in this historic chamber. Give these old schools another chance to serve their city: provide a future for the least among DC's past and as yet unborn kids. Fix those parents. Then they'll fix their kids.

The Old Congressional Heights High School just south of the St. Elizabeth's site in Ward 8, shown below, is typical of a school building that might find a profitable new use as a "GED boarding school" for young unwed moms with their kids as part of a robust "Cycle-Breaker" program.



In May of 2003, the GAO released a supposedly authoritative report on DC's ever-growing financial imbalance, which, they claimed, now leads to an annual fiscal deficit ranging from $470M to $1.1B. NARPAC has spent considerable analytical effort refuting many of the exaggerated claims of DC's structural imbalance and in the GAO's recent report in particular. One of the most troublesome, though less well reported, aspects of this report deals with the potential of a $3 billion deficit in deferred infrastructure maintenance and renewal. NARPAC considers this to be a potentially more credible threat to the image of the national capital if true. It is also a shortfall for which federal assistance (already substantial) might well be rationalized. Our critique of DC's rather primitive six-year capital improvement planningis presented elsewhere.

The GAO report set out to demonstrate that an existing government-sponsored analytical methodology could be used to illustrate where any jurisdiction's government expenditures either exceeded or fell short of some reasonable national norms for that function. Those that lacked the resources to satisfy those norms would be designated as financially challenged. Unfortunately, GAO did not do a very credible job. Nonetheless, the basic approach appears sound and could well be expanded. In particular, there is no reason why it cannot be applied to school systems, and not just for their operating costs, but to their capital needs as well. NARPAC has taken a shot at doing this, and relies heavily on its earlier work in this chapter.

In short, NARPAC concludes that the largest share of DC's unfunded backlog of capital improvements falls within the outdated facilities of the public school system, and that much of it comes from three related sources:

1. DCPS is projecting an unrealistically large future school enrollment;

2. DCPS is continuing to project abnormally small schools compared to American inner city norms;

3. By hanging onto vacant and underutilized buildings, DCPS is denying both itself and the city the ability to underwrite much of its real unfunded infrastructure needs.

It is also clear that there are significant inefficiencies in DCPS operations which appear to contribute to relatively high operating costs and rock-bottom poor grade school test scores. However, there is nothing to indicate that fewer, larger schools would have any negative impact on this significantly below-normal performance, at least based on the available gross statistics, which are very extensive, if sometimes incompatible.

comparative educational and administrative performance

The rudimentary chart below indicates the state of public school education in DC relative to 12 states that have cities nearly the same size as the nation's capital. It compares the US "norm" in 8th Grade reading scores for: a) the US (in blue); b) those 14 states (in green); and c) DC (in red). While the 14-states very closely reflect the reading scores for the US as a whole, DC is significantly behind relative both to those states and to their 20 largest cities. NARPAC has shown in other studies that parental education levels and the poverty resulting from such lack of schoolinghave a first order impact on the educational prowess of their kids. In this area DC falls behind in all four categories: parents with less than a high-school educations; high school graduates; those with some college; and those with college degrees. Furthermore, boys seem to fare even worse than girls, Hispanics are doing worse than blacks, and only the very small fraction of whites in DC schools is outperforming 4th Grade national and state norms.

It is also clear that DC departs from the very relevant norms of twenty large-city school districts as shown on the summary charts below. At top left and right, DC is seen to have 17% less students per teacher and 30% less students per school staff member. Both of these reflect directly in abnormally high personnel costs. Below, it is evident that DC has 32% less students per school, with all the added overhead and operating costs implied thereby, and DC also has 34% more black and Hispanic kids than the 20-city average, which may introduce some problems in itself depending on their out-of-school environment.

One of the few areas where DCPS is right on the mark is in teacher salaries: within 1% of a 20- metro area average . The chart below was used previously in NARPAC's analysis of 'representative wages' to show that DC police, emergency, and school salaries were fully competitive, even though the GAO thought otherwise.

enrollment projections

With regard to future requirements for capital improvements, rather than operating costs, the major questions involve projections of probable school enrollment, and facility requirements. Over time, families are getting smaller, more of them are moving to the suburbs, and more kids are being sent to alternative charter schools or parochial schools with vouchers. More important, the birth rate for the kids most likely to enter DC's pre- and kindergarten classes has declined in each of the last six years, ensuring lower elementary school enrollment for many years to come. Despite these common trends, DCPS is projecting a substantial growth from its current 10-year trend of declining enrollment. As shown on the charts to the left, DCPS long-range facilities planning documents project an enrollment of 75,000 by 2013, even though there are clearly declining trends which are unlikely to be reversed. Even the seemingly unrealistic objectives of the mayor and the city's former Control Board Chairperson, Alice Rivlin, to attract 100,000 new residents will almost certainly not "include 25,000 kids" (which would cost the city a bundle!). The top chart shows the combined trends for elementary and secondary schools, the lower charts provide the detail by the school clusters used in analyzing the earlier facilities plan. NARPAC believes DCPS should be planning on no more than 55,000 kids (27% less), and be prepared to enlarge that smaller set of facilities should the unexpected need arise.

But equally important is the fact that DC is planning for a set of elementary and secondary schools that are significantly smaller in student capacity than the 20-city urban norm. The seemingly complicated chart below shows four characteristics for each cluster of school subdistricts (A-C, D-F, etc) for each the elementary and larger secondary schools. The broad background bars show the 20-city average capacities (519 for elementary and 962 for secondary). The red bars show DC schools' current capacity, and the blue bars next to them show current occupancy. Note that DC's current secondary schools are almost half empty, but that DCPS plans to modernize/rebuild to maintain the same number of schools!. The green bars indicate the increased enrollments that form the basis of the DCPS facilities plan, and the black bars, NARPAC's more conservative (realistic?) projections.

If one accepts NARPAC's "bleaker" view and average US urban capacities, then the top of the chart above tells the story: DC is planning to modernize 37 too many elementary schools and 14 too many secondary schools. At the very least this represents $800M in needless long-range modernization planning. But the average DC school occupies 3.2 acres of land that could otherwise be sold and put to "productive" use. Furthermore, DCPS still retains some 24 more vacant and abandoned school properties dating back to the '50s and '60s when black families typically had twice as many kids. Together these 75 schools probably occupy some 240 acres of scarce, eminently re-developable land. Sale prices of $2M per acre would not be excessive, particularly for 3-acre parcels, generating roughly $500M that could be applied against other unfunded capital improvement projects. Moreover, if those acres are developed for moderate density commercial/residential use, they should be able to generate at least $200M annually in city revenues. Perhaps those revenues could be earmarked for DCPS operating costs.

additional statistical insights

The four chart set below summarizes some additional trends derived from the new 20-city sample in 14 different states. Upper left indicates the not-surprising trend towards increasing per-student school costs associated with increasing median household income (MHI)of that jurisdiction. DC's relatively high school costs stand in stark contrast to its relatively low MHI. Upper right shows a little-discussed fact that school district demographics tend to become primarily minority students well before the majority of the community itself becomes non-white. This first came to NARPAC's notice in a book by David Rusk entitled "Baltimore Unbound" referenced under the web site's 'Suggested Reading."

Lower left indicates yet again that per-student school costs tend to rise as schools get smaller, in concert with established 'economy of scale' rules. And part of that increasing cost results from the need for additional teachers and staff per student as shown lower right. Although the trend- line arrows are no doubt fuzzier than the arrows suggest, it would not be unusual for cost per student to double and for school personnel per student to increase 50% when school size (or school occupancy for that matter) is halved.


More recent data allows additional comparisons between typical school scores and some of the factors that might impact on them. As has been shown from previous data, average school grades decline as the fraction of kids in poverty increases (top chart). NARPAC believes that this trend is primarily due to the lack of parents at home and (implicitly) those parents' education level (middle chart). Finally, NARPAC does not believe that there is any identifiable correlation between school performance and the total number of kids in the school (bottom chart). If such a trend were clearly evident, we would be reluctant to come out so strongly for increasing school size.

  return to the top of the page DRAINING THE SCHOOLS, PRESERVING THE EMPTIES

In the Fall of 2003, public school activists became alarmed by the failure of the DC Government to approve a $100M increase in the DCPS FY04 Budget (See school-related headlines). In response, DC's enigmatic CFO released a table shown that planned funding for DC public schools has increased 37% from $550.8M in FY99 to a proposed level of $754.8M in FY04. At the same time, the enrollment in DC public schools will have fallen by 17% or some 12,800 kids. That results in a per-student increase of 65% in six years. Surely that should be sufficient for even the most hearty public education advocate.

What the CFO table does not point out, however, is that DCPS will continue to operate essentially the same number of schools in FY04 as in FY99. Because there are only about 450 kids in the average DCPS school (way below regional and national norms), that means that DCPS is operating 28 more schools than it needed in FY99 to maintain the same occupational efficiency. This is shown in the two graphs below. The upper one shows how funding per student varies as a result of increased funding and decreased enrollment. The lower chart converts the lost student enrollment into 'schools-worth' on now excess schools.

The costs of low occupancy schools are difficult to quantify precisely, and NARPAC lacks the resources to explore the subject in great detail. However, we have made a stab at allocating all school system funding based on the DCPS FY02 Operating Budget on the table below. We divided various cost elements between "school-based spending" using the weighted student factor approach (see "WSF"), and the "centrally-based spending" of the central headquarters, common maintenance and other factors. Special ed funding is isolated and not included in the per-regular-student allocations that follow:

If the infrastructure stays the same size, and the student load is reduced, then at the very least, the unrealized savings will include school management and administration staff, custodian staff, cleaning supplies, and school operations and maintenance. Together, these items reach $1437 per student no longer there, or $18.4M for the 12,800 enrollment drop. This could essentially double the available funding for student and staff non-personnel support for the remaining school population, or permit a 37% increase in annual maintenance funding for the scaled down number of schools. These are both areas of substantial student and teacher dissatisfaction in the current shabby school system.

As is mentioned elsewhere in the preceding section on structural imbalance in the school system, there are also major savings involved in disposing of 28 schools with an average of perhaps 100,000 square feet of usable, if run-down, space, sitting on perhaps 2-3 acres of prime urban land. Substantial capital could be raised by the sale of the properties (at least $100M) and the subsequent commercial/residential redevelopment of those properties should bring at least $600K per acre, for about $40M in annual revenues. Applied back into the school system, these benefits would be far from trivial.

return to the top of the page 2003 FACILITIES PLAN LOWERS EXPECTATIONS..... SLIGHTLY

In December, 2003, DCPS posted their revised facilities plan in draft form. It has been in the making for some time, with full input from all local school proponents. From a cursory inspection, NARPAC sees some good news: the DCPS Facilities staff is now projecting a gradual reduction in enrollment based on "lower birth rates, modest housing growth, current out-migration trends, and assumptions about the continued expansion of charter schools". In fact, they are now "projecting that enrollment in DCPS will decline by 6,000 to 10,000 over the next ten years". 2000 to 3000 kids of that reduction results from "the effects of the lower birth rates in the '90s rolling through the grades". Nicely put. That "wave" had previously been ignored (see earlier section of this chapter). In fact, what is still rolling through the system is the bulge in births that took place in the late '80's, early '90s, peaking in '90. As shown in the graphic below, this bulge is just entering high school, and will not be gone until about 2010.

On the bad news side of the equation, the new Master Plan still avoids planning for the closing of any schools, but simply indicates that it will have to be considered in the future. It begins to present arguments as to why schools need more than the "standard" floor area per student (150/175/200 sqft.). Hence although they indicate they may have only 55,000 students a decade hence, the plan calls for almost 68,500. And even at that higher number, the average school population will be only 403 kids per elementary school; 533 per middle school, and only 859 per high school. If those numbers prove to be 20% too high, school populations will slide to 320, 425, and 690 respectively, and will have become ridiculously small. In fact, some 74 of DC's 107 current/planned elementary schools would enroll 400 kids or less, and 46 of them would require significant downsizing! And it should be noted that these enrollment numbers include both pre- Kindergarten (declining slightly) and pre-School (increasing slowly). It also assumes that as many as 85% of all DC's special ed kids will be accommodated in DC schools.

These latest projections were also based on a "leveling off" of the decade of declining live births in DC. This may not be confirmed by the newly available number for 2002, which resumes a downward trend to the lowest figure in decades, but additional years' data are needed. However, for the first time, NARPAC has identified the separate numbers for black and white moms. As shown on the chart below, the declining trend in black babies has continued more steeply, and may be more representative of the cohort bound for DC's public schools.

Finally, it is also instructive to look at the number of students of all ages that choose not to go to the school nearest where they live. It suggests that the concept of "neighborhood schools" is becoming less and less relevant. This in turn should make it easier to face up to closing the city's least productive schools (either in size or scores) and perhaps expand those schools that are proving to be the most productive. Surely it will weaken the case that neighborhood schools are essential, particularly if those neighborhoods are thoroughly blighted. As of 2003-4, only 53% of the parents choose the appropriate school closest to them, another 38% goes outside their local area, sometimes half way across the city, and the final 9% don't use the public school system: a number that could rise if the use of school vouchers is successful and expanded. The chart below shows the total, the subtotals by school grade, and the totals for each of the eight planning areas discussed in earlier sections.

As NARPAC reads these latest tea leaves, the new DCPS Facilities Plan includes 54 too many schools on 217 too many acres. This expensive, under-utilized, over-aged infrastructure is almost 40% larger than it need be. Not only does this mismanagement equal or exceed DCPS's academic inefficiency, it clearly contributes to that inefficiency. And it contributes significantly to DC's highly touted 'structural imbalance' as well. Any realistic plan to "fix" the embarrassingly poor DCPS school system must include a bold and near-term initiative to "right-size" DCPS .facilities.


Since that last paragraph was written in March of 2004, very little concrete progress has been made on downsizing the school system. In fact, not a school has been closed in the subsequent two years. However, fresh winds of change appear to be blowing, and NARPAC has added information on a new Master Education Plan. It appears to include a significant effort to belatedly "right size" DC's aging public school facilities. NARPAC looks at these new possibilities for downsizing DCPS in a new chapter following this one.

return to thetop of the page

Did you find this of interest?
Please give us your FEEDBACK

This page was updated on Apr 15, 2006

© copyright 2007 NARPAC, Inc. All rights reserved