Last month Abe Naperstak of Forest City told Council the 22 figure could be lower because some of those children would attend private and parochial schools.
Earlier today Mayor Bramson told Council that retroactively applying the Rutgers Scale to Avalon would project 181 students and that today there are 178.
The Mayor says that since these two numbers are similar this means the 22 figure for Echo Bay must be accurate.
This notion was supported by Mr. Strome and Mr. Fertel.
But as they all know, 22 is the number of PUBLIC SCHOOL students projected to live in Echo Bay excluding PRIVATE and PAROCHIAL students.
They are talking about these numbers in terms of ALL STUDENTS when the 22 figure is for ONLY PUBLIC SCHOOL STUDENTS.
Mixing up these numbers appears to be an attempt to confuse the public in order to push this deal through without close scrutiny of the impact of Echo Bay on the school budget.
This same thing was done 15 years ago, and Mayor Bramson was a part of that effort, to confuse the public when residents expressed concerns that the Avalon projection of 58 was too low which it was by a factor of 250%.
Today, for the first time, we heard something entirely new from the Mayor. That the projection of 58 students for Avalon made 15 years ago was based on a method of blended models and was incorrectly tabulated due to an incorrect statement of the apartment mix.
Really?
If true, that little miscalculation will cost the school district $2.5 million this year, about 65% of the increase in the school budget and has, over all, cost tens of millions more.
I am not here tonight to advocate for or against the Echo Bay project.
I am here to say that these projections are not accurate and to hang the cost of your errors on the school district, as you have done in the past, is unfair and needs to stop.
I ran projections using an accurate apartment mix with today’s Rutgers Scale and came up with 187 public school students. There are 146. Off by 28%
125, 181, 146, 187, 178.
Somehow from all these inaccurate projections the Mayor concludes that the 22 for Echo Bay is not only accurate but likely too low.
Not even Forest City has made THAT argument.
According to the Mayor, Forest City is proposing to pay MORE in taxes than required. Is that even credible?
These projections can vary even more widely if you change the apartment mix to reduce the number of studios which count as zero students in the Rutgers Scale, or if you change from rental to owner units, or if the prices change, and so on.
Echo Bay as proposed is 22.
If they later eliminate studios that becomes 32.
If they switch from rentals to condos that becomes 42.
Rental or Sales prices are driven by market forces and can likewise change the projected number.
These wildly gyrating projections are what you would expect from understanding basic econometric modeling.
A very strong model will have a Confidence Interval of 99.
A strong model 95
The Rutgers Scale has a Confidence Interval of 90.
That is low and indicates the model is not terribly accurate.
To then rely on this low confidence model to set school taxes based on a fixed number of 22 students shifts all of the risks to the school district which can only result in fewer teachers and more crowded classrooms.
Noam’s Numerical Mummery
I have worked in the financial markets business for many years and have generally risk managed portfolios to a model using a 95% confidence interval. With this type of model there were many instances when the model failed to predict the performance of the portfolio, in some instances for an extended period of time. A number of years ago I was asked to manage a portfolio to a 99% confidence interval and that too had frequent periods where the model did not work. In my business if you provided a risk model based on a 90% confidence interval your business would simply be shut down by the regulators.
It is concerning that the Mayor is willing to bet “our last best chance” on projections generated by such a weak model. But honestly I am not surprised by his willingness to go down this path. For years I have listened to him use fictional terms like “statistical fact” and “net benefit” as part of his dog and pony show. It is refreshing to see that the community is finally catching on to his BS and is demanding real answers !
Keep up the good work !