OPINION OF THE COURT
Cardona, P. J.The Town of Union Vale in Dutchess County has segments in three different school districts: Arlington, Dover and Mill-brook (hereinafter collectively referred to as the districts). In 1991, respondent lowered the equalization rate1 assigned to the Town from approximately 67% to approximately 44%. The lower rate, based on property valued as of January 1, 1989, resulted in a higher estimate of taxable real property and higher school taxes for residents in all three segments of *198the Town. We note that the lower rate remained in effect for two years. In 1993, the rate increased to 68.73%.2
The increase in school taxes prompted the Town to submit three applications (one for each segment of the Town) pursuant to RPTL 1314 (2) seeking higher, special rates for all three segments for the 1992-1993 tax year. The Town asserted that application of the Town-wide 1991 equalization rate to each segment was inequitable because the school tax rates borne by the Town taxpayers in each segment were higher than those borne by the taxpayers of the other towns within the districts. The Town conducted its own survey of sales of residential and vacant land within each of the three districts during the 1991 and 1992 calendar years and submitted, with its applications, a comparative study allegedly demonstrating that its taxpayers on the whole paid higher taxes than the taxpayers of other towns in the districts due to a disproportionately lower equalization rate.
Rejecting the Town’s sales study as a basis for its analysis, respondent denied the applications. The denial was challenged in a CPLR article 78 proceeding. Supreme Court found that respondent’s denial of the applications without making a determination (based either on the Town’s data or its own) of whether the Town-wide equalization rate was inequitable when applied to the three segments in the Town was arbitrary and capricious. On remand, respondent used data from its 1990 market value survey to calculate hypothetical equalization rates for each of the Town’s segments. Because the difference between the Town-wide rate and each of the three hypothetical equalization rates was less than 10%, respondent determined that segment special rates could not be established (see, 9 NYCRR 186-5.5) and, once again, denied the Town’s applications. The Town commenced this CPLR article 78 proceeding challenging respondent’s denial. Supreme Court dismissed the petition. This appeal ensued.
At issue is the interpretation of RPTL 1314 (2). That section provides in pertinent part as follows: "If it is made to appear to the state board * * * that the state equalization rate established by it for a city or town is inequitable as applied to real property within the school district in such city or town, it *199shall determine a special equalization rate for such city or town or for such real property, as the case may be, which shall be used for the sole purpose of equalization under this section.” Respondent interprets this provision as providing a remedy only where the equalization rate for a city or town as a whole is not representative of the portion of the city or town located within a school district (see, e.g., 9 NYCRR 186-5.3 [b] [3]). In other words, relief in the form of a segment special equalization rate is only granted where the over-all equalization rate of a city or town does not accurately reflect property values within a particular segment of a city or town.
Respondent is the agency charged with administering RPTL 1314 (2). Our reading of the statute leads us to conclude that respondent’s interpretation, i.e., to prevent inequities among sections of the same city or town, is not irrational or unreasonable and, therefore, its construction must be upheld (see, Matter of Johnson v Joy, 48 NY2d 689, 691; Mental Hygiene Legal Serv. [DeAngelo] v Cuomo, 195 AD2d 189, 190). We further note the absence of any language requiring a comparison of a town’s tax rates with those of neighboring towns when determining whether a segment special equalization rate should issue. Accordingly, we uphold respondent’s interpretation of the statute as it is neither irrational nor unreasonable.
We also find that it was reasonable for respondent to use its 1990 market value survey rather than the sales data supplied by the Town. Use of the Town’s data would have been inequitable because it would have meant using a different market value standard for the appraisals other than the January 1, 1989 valuation standard that was used to calculate the 1991 equalization rates for the other towns in the three school districts. Additionally, the Town’s proposed segment special equalization rates were based on sales of only two types of property, residential and vacant. Respondent’s equalization rates are based on estimates of full market value of all types of property, including, for instance, commercial property.
We conclude that respondent’s determination denying the Town’s applications for the establishment of segment special equalization rates was neither arbitrary nor capricious; Supreme Court therefore properly dismissed the petition.
Mercure, Casey, Peters and Spain, JJ., concur.
Ordered that the judgment is affirmed, without costs.
. Because each city, town and village throughout the State assesses its taxable property differently, usually at a fraction of market value, the different assessments must first be "equalized” before the tax burden can be fairly allocated among the various taxing units in proportion to the value of real property located within each taxing authority. To accomplish this purpose, respondent conducts periodic "market value surveys” of real property values in each taxing unit (see, RPTL 1200 [1]), appraising, as of a given "valuation date”, sample parcels selected from properties of different categories, locations and assessed values and calculating an estimate of full market value. The annual equalization rate assigned to a particular locality is the percentage of full value at which that locality assesses its taxable real property (see, RPTL 1202 [1] [a]; 9 NYCRR 186-1.2 [a]). Thus, if City A assesses its real property at only 75% of the full market value of such real property, the equalization rate for City A will be 75%. The total full market value of property in any given locality can be computed by dividing the assessed value of its property by its assigned equalization rate.
The use of State equalization rates facilitates the fair comparison of real property values in different localities for, among other things, the distribution of State aid, the determination of tax and debt limits, and the apportionment of taxes (see, 9 NYCRR 186-1.3).
. Respondent determined that the market value of the Town increased 48.87% from January 1, 1987 to January 1, 1989 and decreased 35.61% between January 1, 1989 and January 1, 1992. The Town has not presented evidence refuting this determination.