(concurring). It does not seem that the assessed value of $400,000 placed on the investment property in question by the Special Term is justified in any aspect of the case before us. The property is located in the choicest business section in Albany, and is used for the conduct of a long-established, and one of the largest and most widely known department stores in upper New York State.
While all the credible evidence supports the assessment as originally made by the board, this is especially true of the actual leases now in force. The annual rent assured thereby to the owner for ten years to come is $48,000, and that too, with the obligation resting on the tenant to make substantially all repairs during that time. A yearly income, substantially net, of $48,000 capitalized at six per cent, indicates a market value of $800,000; capitalized at five per cent, it indicates a market value of $960,000; and at four per cent, $1,200,000. The payment of the rent and other performance of the lease may be fairly regarded secure, as much so as the best business probability. The reduction of the assessment from $800,000 to $400,000 would indicate a net return on the investment of more than twelve per cent. To put the case with exceeding mildness, argument to justify such reduction, for tax purposes, is, in its nature, extraordinary and most unconvincing. It is extravagant, and the resulting income unreasonable.
Again attention is called to the method of determining the rate of assessments generally in Albany. Parcels were selected by both parties for consideration by the Special Term, as representative of all the taxable property in the city. The rate of assessment was arrived at by dividing the sum of the values of the selected parcels by the sum of the proved values thereof. I regard this method, and the result obtained thereby, erroneous as a matter of arithmetical principle, and, therefore, of law. The corresponding error was pointed out in the dissenting opinions in at least two other recent cases in this court (People ex rel. Hagy v. Lewis, 255 App. Div. 916; People ex, rel. Woollard v. Lewis, 256 id. 872). As a result of the same method followed in the Woollard case in 1937, the rate of assessment generally in Albany was found to be seventy-one and four-tenths per cent of value, and in the Hagy case in 1937, sixty-four per cent. In a large number of appeals from judgments in certiorari proceedings for the review of similar cases heard by *482this court over a period of years, it was stipulated by the parties that the rate of assessment generally in Albany was eighty-eight per cent. In the present case the rate is found, by following the questionable method, at sixty-four per cent. These evident inconsistencies in themselves strongly suggest the application of a wrong principle.
The parcels of property of a tax district are not assessed on the basis of lump sum value, nor are the parcels assessed in batches; and a comparison of such aggregate assessments and values has nothing to do directly with the average rate of assessment. Each parcel is originally assessed as a unit, and to determine the average percentage, or average rate, which the assessments bear to the true value of the properties to be taxed, it is necessary to base the computations on the percentages, or rates, of assessments of the separate parcels; they cannot be determined on the aggregate values and assessments in Albany, nor on any group of such aggregate values and assessments. The assessments were not made that way by the assessors, in block. The various market and assessed values are necessarily different, in most cases; but the percentages, if correct, remain constant, as the tax rate itself remains constant. Accordingly, if the method pursued in this case is to be followed, of dividing aggregate assessed values by aggregate values as found, widely variant “ rates ” will result, depending on the fact whether parcels of large or small values (extremes) are used as a basis of computation — as they may be presented by relator on the one side, or the city on the other.
The arithmetic advocated here is clarified by reference to an illustrative table made by Mr. Justice Foster at the Albany Special Term on August 5, 1938, in two comparable cases (People ex rel. Ludlum and People ex rel. Flemming v. Lewis, not reported), in which it is assumed that there are only six parcels of real estate in Albany, and that they are of small value.
Table of Selected Parcels
Relator City
Assessed Market Rate Assessed Market Rate
value value % value value %
$5 $10 50 $40 $100 40
$10 $20 50 $60 $150 40
$20 $40 50 $80 $200 40
$35 $70 150% $180 $450 120%
150% + 120% -f- 6 = 45%, the average rate of assessment.
35 + 180 -T- 70 + 450 = 41 9/26%, the relation of lump sum assessments to lump sum values.
*483By adding the separate percentages, and dividing by the number of parcels, the average rate of forty-five per cent results. It is the average rate which is sought. But that is not true when the sum of assessed values is divided by the sum of the found values; as the “ rate ” of forty-one and nine-twenty-sixths per cent results, giving only the relation of lump sum values, but not an average on assessment percentages. By increasing the market and assessment values on one side or the other of the table, the percentage relation of those values will become higher or lower, even though the average rate percentage of the separate parcels remains constant.
A more extravagant demonstration of the error may be made to appear:
Property Assess-Assess Property Assess-Assess value ment rate value ment rate
$2,000 $1,000 50% $100,000 $10,000 10%
50% + 10% -T- 2 — 30%, the average of percentages.
10,000 + 1,000 v- 2,000 + 100,000 = 10.78%, the relation of lump sum values. Here is shown a difference in result of nearly twenty per cent.
The law intends that the rate of assessment to market value of each parcel shall be the rate at which every other parcel shall be assessed. The market values of properties will differ widely in many instances, and so will assessments; but the rate at which the separate properties are to be assessed must not differ; that must remain the same as to all parcels. We are not concerned with the mere values of properties, nor with the mere amounts of assessments; but the desideratum is that the rate at which one owner’s property is assessed shall be the same rate as that at which the property of every other owner is assessed. That must be determined by dealing with the various rates shown by the record, and the resultant average. There must be and there can be properly only one average rate. That must be found by adding the rates of assessment as to all of the selected parcels, and dividing that sum by the number of parcels.
The judgment should be reversed and a new trial granted before another referee to be appointed by this court, with costs.