The Kentucky Department of Revenue entered an order increasing by fifteen percent the 1966 assessment of farm property in Oldham County, to bring the assessment up to fair cash value. See KRS 133.150 and Russman v. Luckett, Ky., 391 S.W.2d 694. The county through its fiscal court appealed to the Board of Tax Appeals, which upheld the order. The county then appealed to the Oldham Circuit Court, which entered judgment holding the order invalid. The Department of Revenue and Board of Tax Appeals have appealed here from that judgment.
There is no issue between the parties as to whether all classes of property must be assessed at full fair cash value; all agree that it must he. The main controversy is as to whether the farm property in Old-ham County was in fact underassessed by the local authorities so as to warrant the increase ordered by the department. The *388circuit court held that the evidence taken before the Board of Tax Appeals did not sustain the board’s finding that the farm property was underassessed; on the contrary it showed that the local assessment was at full fair cash value. While expressed in terms of an evaluation of the evidence, in reality the holding was simply a rejection of the department’s principles of computation and an acceptance of the principle advocated by the fiscal court.
The assessment in question undertook to value the property as of January 1, 1966. The Department of Revenue made a study of all farm sales in the county during 1964 and 1965. Those which did not appear to be bona fide arm’s-length transactions were eliminated from consideration. The actual sale price of each of the remaining sales was determined and it was compared with the 1964 and 1965 assessments (which admittedly were way below full fair cash value). This comparison showed that the ratio of assessments to fair cash value, for each of the years 1964 and 1965, was 21.6 percent. This meant that the aggregate assessments for those years had to be multiplied by 4.6 in order to produce an amount representing full value assessment. Applying this method and making appropriate adjustments, the department made a computation of what should be the aggregate full value assessment for 1966. The actual assessment was such an amount below that as to require a fifteen percent increase to bring it up to full value.
(It is significant, we think, that in applying the same procedure to the several classes of property in Oldham County other than farm property, the department found that they were assessed at almost 100 percent of value.)
The county used a different method of computation in support of its contention that the farm property was not underas-sessed. It took all sales of farm property in 1964 and 1965 (without attempting to weed out any that were not voluntary arm’s-length transactions) plus those occurring during the first six months of 1966. The total sales prices for each year were divided by the total number of acres sold, to-produce a per-acre price. This method indicated an average per-acre price of $327 in 1964, $224 in 1965, and $260 in 1966.. The combined average for the three years, was $273. The amount of the local assessment of farm land for 1966, divided by the total number of acres of farm land in the county, produced an average per-acre-value of $294. This, says the county, shows that the local assessment was above fair cash value.
As hereinbefore mentioned, the circuit-court held in substance that the department’s method was not valid and did not. produce reliable evidence of underassessment. We do not agree. Many things attest to the validity and reliability of the department’s method. First, there was evidence that the method, over many years of use, had proved dependable and reasonably accurate; that the method is used by some 35 other states and by the United States Census Bureau; and that it has been used with satisfactory results in the-allocation of school funds under the Minimum Foundation Program. Second, there-was evidence that the sales used in the method were sufficient in number and in character so that when compared with assessments a fair and reliable indication was-produced of the prevailing ratio of assessments to full value for the entire class of such property in the county. Third, the method is logically valid. Fourth, and we think particularly significant, is the fact that in the instant case the department’s, method will produce substantially the same result as the one advocated by the county,, if sales not truly representative are eliminated from consideration. This latter point: requires some explanation. The county used all sales in its computations. The department undertook to eliminate sales that were not voluntary arm’s-length transactions. For the year 1964 the sales used by the department were substantially the same as those used by the county, and the per-*389acre price was substantially the same ($328 for department, $327 for county). However, in 1965, there was a material difference in the sales used. Those used by the county averaged $224 per acre. Those used by the department averaged $328 per acre — the same as in 1964. The average per-acre value under the increased assessment ordered by the department would be $323. Thus the department’s method produces substantially the same result here as the county’s if consideration is limited to properly selected sales.
The county argues that the department’s method is not proper or reliable because it utilizes 1964 and 1965 assessment figures whereas the law requires reassessment and revaluation as of January 1, 1966. This argument is wholly without merit. The assessment figures were used simply to determine a ratio as a test of the 1966 assessment. The county argues also that the assessment figures so used might not have been correct. We think there was sufficient proof of correctness in the testimony of employes of the department that they took the figures from the assessment records. The county further argues that the department has no statutory authority to use such a formula as it did use. This argument is •somewhat astonishing in the face of the •express directions in KRS 133.150 that the department compare the tax commissioner’s books with the records of sales of land and ■“determine the ratio of the assessed valuation of the property to the fair cash value.”
We are not convinced that the per-acre method advocated by the county is even acceptable, much less conclusive of valuation. It gives no consideration to differentials in improvements or in the amounts of acreage, or in the quality of the land sold. That there is something wrong with this method is quite clearly pointed •out by the fact that it produced a per-acre value of $327 in 1964, but only $224 for 1965. It is a matter of common knowledge that land values did not depreciate in Old-ham County or elsewhere by such a substantial amount between 1964 and 1965.
It is our conclusion that the evidence heard before the Board of Tax Appeals established the validity and reliability of the department’s method of computation of aggregate fair cash value sufficiently to give probative value to the computation and to support the finding of the board that the farm property was under-assessed to such an extent as to require a fifteen-percent increase.
Other questions are raised.
The circuit court held that KRS 133.150 does not authorize the Department of Revenue to equalize the assessment of classes of property by comparing the aggregate assessments of such classes of property with the individual assessments made by the county tax commissioner. It is sufficient to say that as we read the statute it plainly does so authorize. Related to this is the contention of the county that the statute authorizes only equalization among counties and not equalization of classes within a county. The statute expressly says that if any class of property in any county is not assessed at its fair cash value such assessment shall be increased or decreased to its fair cash value. Furthermore, this Court in Russman v. Luckett, Ky., 391 S.W.2d 694, made clear that the department has not only the power but the duty to see that all property is assessed at fair cash value.
This case was heard twice before the Board of Tax Appeals. At the conclusion of the first hearing the board remanded the case to the department with directions that the county submit a list of the specific properties that it claimed would be assessed at more than full value as a result of the increased assessment ordered by the department, and that the department make findings as to each of such properties. We construe this as a good faith move by the board directed toward achieving a fair result. However, the county seized upon it as *390an opportunity for a tactical maneuver, and proceeded to list 725 out of the 751 farms in the county. The county put the tax commissioner on the stand and he testified simply that he had assessed all of the farms at fair cash value. The department put on no evidence of its own, and at the conclusion of the hearing proceeded to find that there was no evidence presented before it that any particular piece of property would be overassessed. The circuit court held in effect that the blanket testimony of the tax commissioner showed that every one of the 725 listed farms would be overassessed, and since the department did not offer any evidence to the contrary the department was bound to make a finding that all of the listed parcels would be overassessed. We think it is obvious that the Board of Tax Appeals, in its order of remand, did not contemplate any such ploy as the county chose to engage in. The board obviously intended to give an opportunity for consideration of the effect of the increase on individual properties. The aggregate aspects of the assessment increase had been fully gone into before the board. However, the county elected to continue the battle of the aggregate assessments under the remand order. We think the department was not required to introduce any evidence to combat the blanket testimony of the tax commissioner, and that it properly found that no evidence had been presented on the issue of over-assessment of individual pieces of property.
At this point we should mention that no individual property owners are in this case and we do not have before us the issues that were involved in Fitzpatrick v. Patrick, Ky., 410 S.W.2d 143.
The Department of Revenue certified the increased assessment to the county clerk of Oldham County on July 29, 1966. KRS 133.180 provides that when the department has completed its action on the assessment of property in any county it shall, not later than June 20, certify the assessment to the county clerk. The circuit court held that the assessment increase for Oldham County was void because not certified by June 20.
In Johnson v. Fordson Coal Co., 213 Ky. 445, 281 S.W. 472, under a predecessor statute, the State Tax Commission certified an increased assessment on August 8, whereas the statute required that it be done not later than June 1. This Court held that the statute was directory only, as to the date, and that the assessment increase was not void. To the same effect is Ross v. First National Bank, 213 Ky. 453, 281 S.W. 517. The county argues, however, that the above cases were decided prior to the 1942 revision of the statutes in which, in KRS 446.010(23), there was inserted the principle of construction that, unless the context otherwise requires, “shall” is mandatory. We find no force in the argument because this principle of construction was recognized in our case law many years before 1942 — at least as early as 1920. See Horning v. Fiscal Court of Caldwell County, 187 Ky. 87, 218 S.W. 989.
Evidence at the hearing before the Board of Tax Appeals showed that because of the problems arising in 1966 in connection with the transition to full fair cash value assessments it was impossible for the Department of Revenue to complete its equalization duties and certify all of its assessment orders by June 20. The county suggests that the delay in certification prejudiced the county by delaying its tax collection program. However, the county has not shown that any of its problems in relation to its tax collection program are attributable to the delay in certification.
We conclude that the delay in certification did not prejudice the county and did not invalidate the assessment increase.
A final question grows out of the fact that the Department of Revenue failed to notify the mayors of LaGrange and Pewee Valley of the contemplated assessment increase. KRS 133.160 states that the department shall notify the mayor of any city *391which is affected and which has adopted the county assessment. There were two parcels of farm land within the limits of Pewee Valley, and six parcels within the limits of LaGrange, both of which cities had adopted the county assessment. However, the department was not aware of the existence of these parcels so it failed to give notice to the mayors. The Board of Tax Appeals held that the failure to give such notice would have the effect only of making the increased assessment unenforcible with respect to city taxes on the parcels in question. The circuit court, however, held that the entire assessment increase was void, as tc all farm land in Oldham County, because of the failure of notice.
In support of the judgment on this point the county again argues that “shall” is mandatory. We think that is irrelevant, because the real question is what shall be the consequence of the failure to give notice, whether or not the giving of notice is deemed mandatory. We cannot rate the failure here, affecting just the city taxes on only eight parcels of land out of a total of 751 parcels, as being of such proportions as to require that the whole procedure be held invalid. The county points out that the ruling of the Board of Tax Appeals results in a preference or discrimination in favor of the eight parcels. This fails to impress us, because if we were to accept the county’s argument and hold the entire assessment increase void, the result would be not only that the preference or discrimination in favor of the eight parcels as to city taxes would be continued, but an additional preference would be created in favor of those parcels as to school taxes, county taxes, and state taxes, and a preference would be created in favor of the 743 other parcels of farm land as to school, county and state taxes. It is our opinion that the consequences of the failure to notify the mayors should be considered to fall within the maxim, De minimis non curat lex.
The discussion on the last point we think brings out the fact that the primary objective of 100 percent assessment is to achieve equality of the tax burden, not to increase the burden. If an increase in taxes results from an assessment increase it is because the locally elected representatives of the people have not adjusted the rate. To illustrate: A county’s budget is based upon an estimate of aggregate tax revenue, determined by application of a rate to the aggregate assessment; if the assessment of one class of property is increased, the aggregate assessment correspondingly is increased; thus the amount of income contemplated by the budget can be raised with a lower tax rate; the ultimate effect of the assessment increase is, then, that the aggregate tax burden is not increased; the burden simply is spread equally among the various classes of property. If there is a public demand that one class of property carry a lower burden than other classes, the remedy is through classification legislation under Section 171 of the Kentucky Constitution — not through unequal assessments.
The judgment is reversed with directions to enter judgment upholding the order of the Board of Tax Appeals.
WILLIAMS, C. J., and HILL, MILLI-KEN, STEINFELD and PALMORE, JJ., concur. MONTGOMERY and OSBORNE, JJ., dissent.