For the 2020-21 property tax year, all Class I residential properties in Nassau County were reassessed (the “Reassessment”) using a computer-assisted mass appraisal (“CAMA”) approach. CAMA uses state-of-the-art computer modeling for valuing property that ensures tax equity by creating uniform valuations. Work on the Reassessment initiative began in 2015.
For the Reassessment, the County’s Class I residential properties were divided into seven different market valuation areas and the CAMA system determined the market values for each of Nassau County’s approximately 386,000 Class I residential properties by using seven different mathematical formulas based on separate statistical multiple regressions , one for each of the seven market valuation areas.
A statistical regression is an industry-accepted method used to analyze and estimate the relationships between a number of independent variables (often called “predictors” or “features”) and a dependent variable (often called the “outcome variable”) as it relates to determining the value of a property. In general, regression analysis is a way of mathematically sorting out which variables have an impact on the outcome. It answers the questions: Which factors matter most? Which can we ignore? How strongly do those factors impact outcome? How do those factors interact with each other? And, perhaps most importantly, how certain are we about all of these factors?
Simply put, the independent variables used in determining values of the County’s Class I residential properties, included characteristics (independent variables) such as property size and location, property features, building style and size, building features, school district, neighborhood factors, and so forth. The dependent variables were the time-adjusted sale prices for all the Class I residential properties sold properties in the County during the period from 2013 through 2018. Time-adjusted sale price is the price at which a property was sold, adjusted for the effects of price changes in the market over time between the date of sale and the date of analysis. The regression analysis determined how the various property features affected the value of each of the County’s Class I residential properties.
To perform the CAMA, the assessment contractor hired by Nassau County – Standard Valuation Services (“SVS”) – recommended that the County’s Class I residential properties be grouped into approximately 326 neighborhoods based upon a variety of factors, including school district boundaries, municipal boundaries, construction type, style, quality, and size, property value ranges, streets and other physical boundaries, sales trends, and locational influences (such as proximity to commercial properties, railroad tracks/stations, major roads/highways, views of bodies of water or golf courses, quality of waterfront including linear feet abutting a body of water, and distance to “open bay” water). Based on Long Island since 1989, SVS is a highly experienced real estate appraisal and consulting firm. SVS provides real estate appraisal and expert services for all types of commercial and residential properties in many cities and towns from Manhattan to Montauk and the greater New York metropolitan area.
Before 2001, Nassau County did not create or use neighborhood distinctions for assessing Class I residential properties. At that point in time, since all such parcels were valued by a 1938 construction cost-based (not a market-based) system for improvements, there was no need for using residential neighborhoods to set assessed values. It wasn’t until Nassau County changed from a construction cost-based to a market-based assessment system (as agreed upon under a NYS Supreme Court supervised consent decree ) that neighborhoods distinctions were first used.
In 2000, Nassau County retained Cole-Layer-Trumble, now known as Tyler Technologies (“Tyler”), to conduct the County’s first market-based county-wide reassessment for the 2003-04 property tax year. Tyler divided Nassau County into 154 original neighborhoods. Tyler also used computer modeling for its property valuations.
Since 2001, Nassau County’s Department of Assessment has consulted with assessment contractors to improve the residential neighborhood delineations from time to time, starting with the original 154 neighborhoods in 2001, increasing the number to approximately 247 neighborhoods in 2002, and slightly modifying the neighborhoods over the ensuing years to approximately 254 neighborhoods when work began on the Reassessment in 2015. As the Reassessment progressed, SVS further refined and developed neighborhoods for each condominium complex which added to the total number of neighborhoods.
As part of its contractual work on the Reassessment, SVS reviewed the original neighborhood distinctions and the many refinements and modifications in Nassau County over a period of approximately five to ten years before work began on the Reassessment. SVS took those refinements and modifications into consideration to re-define the neighborhoods for the Reassessment.
In the Reassessment, SVS also utilized geographic information system (“GIS”) technology to map and plot relevant sales to judge the accuracy of the existing neighborhood breakdown. SVS’s in-house real estate appraisal experts were given individual maps in their areas of expertise to review. Maps were produced to include school district and municipal boundaries, construction type, style, quality, and size, property value ranges, streets and other physical boundaries, sales trends were reviewed. Using GIS, SVS also considered locational influences, both negative and positive (such as proximity to commercial properties, railroad tracks/stations, major roads/highways, views of bodies of water or golf courses, quality of waterfront including linear feet abutting a body of water, and distance to “open bay” water) to assist in determining accurate neighborhood boundaries. SVS also considered standard valuation guidelines, such as those published by the International Association of Assessing Officers (“IAAO”), for dividing communities into neighborhoods.
Based upon the recommendations of SVS’s respected and knowledgeable in-house appraisers, neighborhood boundaries were changed as necessary. After extensive review of those factors, SVS revised the neighborhood count by approximately 75 additional neighborhood delineations to approximately 326 neighborhoods. A new neighborhood map was then created using GIS and ArcMap software to develop a polygon layer depicting the boundary for each individual neighborhood. Increasing the number of neighborhoods was a major improvement over the neighborhood distinctions that were produced in 2001. The additional neighborhoods allowed the Reassessment to draw finer, more accurate neighborhood boundaries. Edit checks were performed utilizing GIS to ensure that the physical inventory was complete and as accurate as possible.
After considering the many factors that were described previously and exercising real estate valuation judgment and using local knowledge, SVS recommended that Nassau County use the neighborhood delineations they created and that they be used in the Reassessment. In turn, Nassau County, after consulting with and reviewing the SVS’s recommendations (taking into consideration the prior neighborhoods, the factors that SVS used in redefining the neighborhoods, the refinements and modifications made in the county over the approximately five to ten year period prior to the commencement of work on the Reassessment, and the real property valuation experience and local knowledge of SVS, exercised its own assessment judgment and local knowledge and accepted SVS’s neighborhood delineations that would be used in the Reassessment.
Information on the factors used to assign a specific property to a specific neighborhood was made available to the public through a Nassau County website titled “Land Record Viewer,” beginning on or about November 1, 2018. Members of the public were additionally given the opportunity to ask questions about neighborhood factors during the many public outreach meetings held by Nassau County. The Land Record Viewer website provided information on the neighborhood factors. In addition, neighborhood factors were also made available to any member of the public who properly requested them pursuant to Section 6-7.3 of the Nassau County Administrative Code beginning in 2019.
Next, the neighborhoods were grouped into seven homogeneous real property market areas for purposes of the CAMA analysis based upon several factors. Homogeneity is the degree to which the neighborhoods within each real estate market area are like each other in terms of valuation characteristics. A higher degree of homogeneity creates more accurate and reliable statistical modeling; that is, better CAMA results. Market Area Seven comprises all the condominiums and cooperatives in the County. The neighborhoods were then grouped into six additional market areas based upon a number of factors, including (a) geographical proximity; (b) school districts; (c) locational features; (d) average selling price per square foot; (e) housing construction characteristics and size; (f) property characteristics and size; and (g) infrastructure (such as transportation, electric, water, sewer review trends, and locational influences).
Once all of Nassau County’s neighborhoods were divided into the seven market areas, the CAMA system computed the market value for each of the County’s approximately 386,000 Class I residential properties based upon approximately 180 different property features and attributes. The property features and attributes were used to perform a CAMA analysis of the market value of all the Class I residential properties in each of the seven market valuation areas. Each of the seven statistical analyses was done by means of a multiplicative multiple variable regression, which is a powerful methodology for valuing real property. The kind of statistical analysis performed by Nassau County is recognized as the most accurate and reliable way to value real property for a taxing jurisdiction as large and complex as Nassau County.
In the Reassessment, the independent variables were the property features and attributes and the dependent variables were the time-adjusted sale prices of the Class I residential properties sold between 2013 and 2018. Some of those features and attributes affect time-adjusted sale prices more than others, and the effects may differ from market valuation area to market valuation area. The regression formula calculates a coefficient for each of the independent variables. Regression coefficients measure the strength of the relationship between each independent variable and the dependent variable. A higher coefficient means that a change in the independent variable has a stronger impact on the dependent variable. In the Reassessment, the regression coefficients measured the strength of the relationship between each property feature or attribute and the time-adjusted sale prices for each of the seven market areas.
In a multiplicative multiple regression, the relationships between the independent variables are also measured. In other words, changes in independent variables may have different effects on the dependent variable when coupled with changes in other independent market variables. For example, in the Reassessment, a change in livable square footage or lot size may have a different impact on market value in one neighborhood or school district than in another neighborhood or school district. The multiplicative multiple regression formula takes those relationships into consideration to yield a market value that is more accurate than if each property feature or attribute was considered separately.
Like livable square footage, lot size, or school district, the neighborhoods were one of the independent variables used in the CAMA regression modeling. The formulas for each of the seven market areas includes a coefficient for the neighborhoods, just as the formulas included coefficients for livable square footage, lot size, school district, and many other property features and attributes. The regressions computed the strength of the relationship between each neighborhood and market value, just as they computed the strength of the relationships between all the other property features and attributes (such as livable square footage, lot size, and school district) for each of the seven market valuation areas in Nassau County.
SVS retained Russ Thimgan and Thimgan & Associates (“Thimgan”) to assist Nassau County in performing the statistical regressions. Thimgan is a premier consulting firm that performs computer-assisted mass appraisal analysis for many state, county, and national governments. The firm also produces various publications and conducts workshops and seminars for IAAO. IAAO is recognized by the NYS Office of Real Property Tax Services as an authoritative source for valuation standards in the United States and worldwide.
Thimgan performed the statistical regressions using IBM SPSS® Statistics, a powerful and highly reliable statistical software program. To determine the seven valuation equations or regressions, Thimgan used SPSS® to model approximately 180 property factors against sales data for approximately 65,000 Class I residential property sales that took place in Nassau County over the five-year period from 2013 to 2018. In accordance with standard valuation methodology, the real estate sales data was time-adjusted to June 1, 2018, to account for variations in the real estate market over that five-year time period. Using a complex series of computer computations for each of the seven valuation markets, the SPSS® software calculated how much each property factor affected the time-adjusted sale prices for all the sales in that valuation market. The resulting seven equations, one for each of the market valuation areas were used to determine the market value for all 386,000 Class I residential properties in Nassau County. Each property was valued by applying the formulas generated by the computer-assisted analysis to the particular features of that property.
Nassau County then used proprietary software developed and owned Thimgan, called Prognose®, to help taxpayers understand and validate how the new assessed values were formulated. However, the regressions used in the Reassessment were performed using IBM SPSS® Statistics, not using Prognose®. Likewise, the Calculation Ladder previously provided for each Class I residential property together with a Comparable Sales Report showing several comparable sales for the subject property, showed “value adjustments based on particular property characteristics,” but the Calculation Ladder was not the means by which the estimated value of any Class I residential property was determined.
Taxes are levied on Class I residential properties based upon the budgets of the County, the three towns, all but one of the school districts and all the special districts in the County in proportion to the taxable assessed value of each Class I residential property. Here are some helpful terms to understand how taxes are levied on Class I residential properties:
- Estimated value is the estimate of market value assigned to each Class I residential property in the County.
- Level of assessment (“LOA”) is the percentage at which each Class I residential property in the County is assessed based upon estimated value. In the Reassessment, the LOA was 0.1% for all Class I residential properties in the County.
- Assessed value is the value for each Class I residential property on the assessment roll used to allocate taxes among all the Class I residential properties in the County (after giving effect to exemptions), and is determined by multiplying the estimated value for the property by the LOA. By way of illustration, if the estimated value of a property is $350,000, multiplied by 0.1%, the assessed value for that property is $350. A property worth $1,250,000 has an assessed value of $1,250. Under Section 1805 of the New York Real Property Tax Law, the assessed value of any Class I residential property in the County cannot be increased by more than 6% from the prior year’s assessment and cannot increase by more than 20% in any five-year period.
Property tax is calculated each year based upon the total property taxes that must be collected based upon the budgets of the applicable taxing authorities and each property’s share of that tax burden based upon its taxable assessed value and the taxable assessed values of other Class I residential properties.
Nassau County provided a comparable sales analysis for each Class I residential property to help property owners understand the market values of their properties, which is available on its website here. The comparable properties shown for each Class I residential property in Nassau County are ranked in order of similarity, with the most similar property shown first, the next most similar property shown second, and so forth.
Property market value calculations used the following coefficients: