A Critical, Yet Inexact, Science
Next Monday I get back to retirement, back from playing taxi driver all summer for my two teenagers. Between swim team practices and swim meets, then each with their own jobs, but no drivers licenses yet, I had little time off.
It was fun and interesting and, by this time next year, at least one will be driving on their own. But, like many others, the annual conversion to the school year means another round of coordination of schedules, flexibility with time and postponement of any number of "important" things I want to do.
During preparation for this annual conversion, I recently received a call to assist in updating some statistics for one of my favorite "hobbies" - forecasting the number of new students who will enter the doors of the Frederick County Public School system.
More specifically, what should be the "student yield factors" by housing type that the county and school system uses in applying for proper funding under impact fees.
For those of you, of which I suspect there are many, who do not know about student yield factors, here is a short primer. In order to determine the proper level of impact fee to assess against new home construction, government must have a reasonable method to calculate how many new students will be added to the public school system by each new home.
Student yield factors are the method by which this is done. By using reasonable data collected on the number of houses occupied and the number of students in the school system, algorithms are created for each level of school.
For instance, a single family detached (SFD) new home might have a .24 elementary school student, a .13 middle school and a .13 high school student. Each new SFD unit would be assessed an $8,378 impact fee. If 100 of these units were built the school system should see 24 new elementary, 13 new middle and 13 new high school students or an overall increase of 50 new students. Additionally the county would collect $837,800 in funds to be applied to new school construction.
Two main problems are and will continue to be: what are the proper student yield factors and what is the appropriate fee?
These two questions are to be answered every year. Unfortunately this is not always done.
Now back to the 2003-2004 school year and the telephone call I just received.
The second question, what is the appropriate fee, has been somewhat taken care of by applying a "recognized" average statistic for inflation and automatically applying it to the base dollar amount from the previous year. This makes a major assumption that the county costs figures mirror the "average statistic for inflation".
The first question above, what are the proper yield "factors," is a little more complicated. The biggest complication is in the constantly changing demographics of family size for those buying homes in our county. Notice I did not limit it to just new homes. With unusually high turnover rates for existing houses and a drastically altered price range for new homes, the county has been and will continue to experience significant shifts in household size, as defined by the Census Bureau.
The first indications of this shift in average household size has demonstrated itself in the past two years as lower than expected (per the impact fee methodology) student enrollment increases have come on the heels of two years of the highest number of units for new home construction. The example below is merely illustrative of the potential shift in demographics.
In all of 2000, 2,918 housing units were permitted for construction. In theory under the impact fee methodology approximately 1,430 new students should have graced our doors by September 30, 2001. Yet we only had 1,169. A year later there were 2,169 new housing units. That should have produced about 1,063 students. Yet again, on September 30, 2002, only 672 were there. (Please note: this is a simple comparison that does not include a number of variables needed to more accurately reflect the differences.)
One potential problem, once we figure all this out, is that we may be charging a higher than warranted amount under the impact fee for new school construction, based on student yield. A corollary may be that we are understating true costs by using the "average" inflation factor.
For the record, on October 20, 2002, in a published independent analysis of student enrollment increase, I projected 721 new students by this September 30, just over a month away. If I use the impact fee method on the 1,605 housing unit permits for 2002, the figure is 788. The school system, according to a document last updated on February 6, 2003, shows their estimate at 749. If the number is closer to 500, then maybe we have a firmer case on the demographic shift.
I can wait until this September 30 to see which method is closer to reality. But more important to me is the future of more accurate student yield factors and realistic impact fee dollars.