Earth Image Dot Net presents its analysis of the Greenwich Real Estate Tax:

   Is there systematic bias in assessment, either by Zone or Price Class??

    If so, then the "fair share" of the Town Budget is NOT FAIR.  But this is only a question.

                Investigative Findings:

           Sales in 2009 by price class ( Appraisal / Selling Price) :

          Fair Share of taxes born by sellers by price class:


2009 Sales of Greenwich Real Estate  Average Ratio
Selling $$ class Appraisal/Selling $
Over $10 Million 0.504
5 to 10 Million 0.661
4 to 5 Million 0.689
3 to 4 Million 0.684
2 to 3 Million 0.793
1 to 2 Million 0.787
500K to 1 Million 0.859

Source:  Tax Assessor's Master File 3/18/2010 and Land File 3/18/2010 Sale Price1 & Totval1

Terminology Note:  The old terminology was based on "appraisal" at a nominal fair market value - arms length- for highest and best use.  Then 70% of that value was the assessment.  Multiply the assessment times the mill rate and that was the tax.  Owners receiving Town sewerage were levied a surcharge in the mill rate to pay for treatment plant processing of effluent and dumping into Long Island Sound.  The others paid for their own septic systems.  The new terminology is "appraised value," purporting to be 100% of 2005 market value called TotVal1, and the tax value 70%, called TaxVal1.  In this discussion, appraisal is the setting of value and assessment is the levy of taxes - but used somewhat interchangeably for what the assessor does. 

  Conclusion:  Among sellers in 2009, the more your property was worth, the smaller percentage of the Town budget per $1000 you paid.  But this is only valid for the sellers.  What about the rest of the Town?  Is there a pernicious bias in appraisal methods by zone, neighborhood or property size?



Sales BY ZONE and By Year, based on the March 18, 2010 Town Assessor's Files :

The graphs are the list of sales in the category indicated in order of S/A ratio, to avoid averages and show the data set.  A perfect appraisal methodology would produce a nearly flat curve.

Selling Price to 

Tax Assessment Ratio

Sales For 2009:

R-6 Multi-Family 7,500 sf


Average Ratio Median Std.Dev.
1.2465 1.2178 0.3803
Min. Price Ave. Price Max. Price
$160,000 $873,689 $3,800,000




R-7 Single Family 7,500 sf


Average Ratio Median Std.Dev.
1.2018 1.2143 0.3297
Min. Price Ave. Price Max. Price
$1,000 $787,875 $1,675,000


R-12 Single Fam 12,000 sf

Average Ratio Median Std.Dev.
1.1906 1.1988 0.2608
Min. Price Ave. Price Max. Price
$202,000 $1,675,182 $5,800,000


R-20 Single Fam 20,000 sf


Average Ratio Median Std.Dev.
1.3264 1.2752 0.4729
Min. Price Ave. Price Max. Price
$250,000 $2,757,819 $8,250,000



Conclusion:  Sales may reflect a bias after the fact.  But that's life.  It's the win-place-show effect.  Only the top tier of horses and their riders pay off.

The winning ticket here after the race collects part of its reward through a lower share of taxes paid.  But before the race is run, the handicappers all get to look over the ponies together.    And at the butcher shop later on, all horse flesh is back to a basic commodity price.

So despite the logic of the sales rank bias after the fact, this does not follow through to bias by zone before the sale history is available.  Of course, it goes without saying that each sale should be examined by the assessor in a rigorous and systematic manner, to correct the appraisals.  The assessment data files reflect this effort.

However, the consistent even spread of selling price to assessment ratio of over over 100% in each class shows that appraisal is a very inexact science.

This could be due to may factors, including mindless or inaccurate formulas, and poor quality appraisal work in terms of perception of value.   But like the horse race (over the five year assessment cycle) three tickets will pay off - the rest are just in the pack.


Another experiment looked at grouping zones R-6, R-7, R-12 & R-20 (one-sixth to one-half acre min.) into one "size" class called the "R-Zones" for under one acre, and RA-1, RA-2 and RA-4 (1 acre to four acre min.) called the "RA-Zones" for the over acre size.

After sorting all sales into a progressive sequence of sale-to-assessment ratio, the scatter of actual prices paid was checked to look for correlation.  The conclusions below in graphic form show the result.

Between the two size classes, a similar effect to Zone was observed -  Size does not matter.  Other factors do.


The graph on the left of each pair is to test for flatness of the sale/assessment ratio, which is equality of appraisal by selling value  (the value is in the eye of the beholder -- in this case the buyer and of course the assessor.)

The graph on the right of each pair is the same sales in the same order, but graphed by price.  This tests whether the variance in ratio correlates to price, which is not apparent.  It does indicate a wide variance however, which is not good, but difficult to eliminate.

Part of the problem may be that in preparing appeals to the annual Board of Tax Appeal sessions, the applicants and their attorneys and consultants are looking at neighbors and neighborhoods locally, and not Town-wide.

This may be similar to the wide variance of the cost of airline seats on the same flight.  If you knew the whole range of ticket prices on the plane, you would likely want to buy more carefully.

The solution to both problems is the new public availability of data and assessment tools on the internet.  For travel we have sites like or  For assessment matters the Town GIS, now liberated and on-line, puts effective tools in the hands of the public.  While this data is free, the Assessment files are only sold at the Assessor's counter for $30 on CD or hard-copy, so freedom of information there is not yet free, unlike the GIS counter one floor below it.  

There has yet to develop a fruitful dialogue on what we ought to assess, and why, to raise the money to pay our Town budget, now $ 355.9 million without a new firehouse. The assessor has eliminated abatements for oversized acerage and current use.  Instead he focused on the yardstick of highest and best use, as valued 2005 to 2008 using runaway financing that fueled the buying frenzy. This assumed that value in the eye of the builder or buyer was real.

We now have opportunity during the playing out of the present financial breakdown to give some thoughtful consideration as a community to our rationale of what to tax. Full-rate taxation of vacant or open space land simply fuels 100% build-out, which in turn breaks the budget. It is taxation, shared and divided into 22,000 payer-accounts, that pays the bills that fall due.

We are over-zoned and under-planned. A "green" tax policy with principles of sustainable development that serves its population over generations should be the goal.

    (Technical assistance and consulting services available.  Email for inquiries.)

Conclusion:  Neither Zone nor Property Size is well correlated to "Fair Share" results, based on selling price to assessment ratio, when judged after sales take place.