Plaintiffs have appealed assessments for Oregon corporation excise taxes and Multnomah County Business *481Income Taxes for the years 1974 through 1981. The pleadings raised three issues, each of which will be discussed separately.
The first and most significant issue arises from a written agreement entered into between Kellogg Sales Company and defendant’s predecessor, the Oregon State Tax Commission.1 In order to understand this issue, it is necessary to review some history.
Plaintiff, Kellogg Sales Company (KSC), a Michigan corporation, is a wholly owned subsidiary of Kellogg Company (Kellogg). Kellogg, the well-known manufacturer of ready-to-eat cereal, is a Delaware corporation. KSC acts as a sales arm of Kellogg, purchasing stock from Kellogg and reselling it to customers. KSC has no storage or warehouses for its stock other than at Kellogg’s manufacturing plants. KSC’s agreement with Kellogg essentially has KSC selling Kellogg products in exchange for l/100th of one percent of the sale proceeds. (Exhibit 5.) The agreement is not an arm’s-length agreement and appears designed to produce minimal income for KSC.
KSC began doing business in Oregon in 1952. For the years 1955 and 1956, KSC filed Oregon income tax returns which combined KSC’s and Kellogg’s income and apportioned part of that income to Oregon. During those years, the law concerning state taxation of foreign corporations was in a state of flux. Businesses operating interstate were vigorously contesting state jurisdiction to tax when the only activity involved was the solicitation of orders. In 1959, the United States Supreme Court, in two cases, issued rulings favorable to the states asserting taxes against such businesses. See Northwestern States Portland Cement Co. v. Minnesota and T. V. Williams v. Stockham Valves and Fittings, 358 US 450, 79 S Ct 357, 3 L Ed 2d 421, 67 ALR 2d 1292 (1959).
Congress immediately responded by enacting Public Law 86-272 which attempted to establish objective standards for taxing companies doing business interstate. Prior to these cases and Public Law 86-272, there was some uncertainty as to the jurisdiction of the State of Oregon to impose corporate *482income or excise taxes on KSC. From the evidence, it appears that at least in one other instance defendant agreed to modify the sales factor as a compromise where jurisdiction to tax was doubtful. (Exhibit C.)
In 1959, defendant assessed KSC corporate income tax deficiencies for the years 1955 and 1956. KSC then petitioned defendant for administrative relief. (Exhibit 3.) In its petition KSC asserted that it was not subject to the corporate income tax and that the proposed excise tax deficiencies were in error because they were based on an apportionment formula which included all sales to Oregon customers. KSC contended that sales resulting from solicitation only in Oregon, with all other aspects of the transaction taking place outside Oregon, were not includable in the apportionment formula. KSC contended that since it did not have a local office present in the state, the state did not have jurisdiction to tax its solicited sales. (Exhibit 4.)
With this background, the parties entered into the subject agreement on December 11, 1962, to resolve the dispute between them for the tax years 1955 through 1959:
“[A]nd all future years provided the company’s method of operation in distributing its products to Oregon customers is conducted substantially in the manner such method of operation was conducted during the years herein.” (Exhibit 1, at 1.)
It is clear from the evidence and the document itself that the primary purpose of the agreement was to modify the traditional apportionment formula by reducing the sales factor one-half. The issue in this case is whether the agreement also established the “company net income” which was to be apportioned.
As indicated above, during the years 1955 through 1959, KSC reported apportionable income to Oregon based on a combined report with its parent, Kellogg Company. That combination, however, excluded any income of Kellogg’s foreign subsidiaries. KSC was the only domestic subsidiary of Kellogg until 1969 when Kellogg acquired Salada Foods, a Canadian corporation which also owned a United States subsidiary of the same name. In 1970, Kellogg acquired Fearn International, also a domestic corporation.
Later, defendant audited plaintiffs for the years 1974 *483through 1981, and asserted deficiencies based on the entire unitary operation of Kellogg, including foreign as well as domestic subsidiaries. KSC, after exhausting its administrative remedies, appealed to this court.
KSC contends that the purpose of the 1962 agreement was not only to modify the three-factor formula but also to establish the company net income to be apportioned. In the absence of any direct evidence on the latter point, both parties are left to the circumstances surrounding the execution and implementation of the agreement to prove their intent. The task is not made easier by the fact that the attorney who negotiated the agreement for KSC, the Director of the Income Tax Division who negotiated for the state and the state’s corporate auditor at the time are now deceased.
Evidence adduced at the trial indicated that there were a number of similar cases pending before the State Tax Commission in late 1962. In fact, KSC’s attorney, Jack Hay, represented a number of other clients in much the same position as KSC. The attorney for the state, A1 Thomas, testified that the basic agreement was drafted by Mr. Hay for a separate but similar case and that Mr. Thomas adopted the form for KSC, changing only the names and the years.2
After considering all of the evidence submitted by the parties and the arguments pertaining thereto, the court is of the opinion that the agreement does not fix the company net income to be apportioned. The purpose of the agreement was to modify the three-factor formula. The court finds no intent by either party to fix or modify the income that was to be apportioned. The witnesses for plaintiffs as well as defendant were consistent in testifying that the term “company net income” was never questioned or discussed. The net income to be apportioned appears to have been a fact which the parties simply assumed in their negotiations. This is understandable since at the time Kellogg had no other domestic subsidiary corporations. Also, at the time of the agreement in 1962, the state was not attempting to require worldwide unitary reporting. Thus, the existence of Kellogg’s foreign subsidiaries *484raised no question as to the company net income to be apportioned.
KSC contends that defendant’s acquiescence in its understanding is evidence that defendant had the same intent or, if not the same intent, understood KSC’s intent and is bound thereby. The basis for this argument is found in events subsequent to the execution of the agreement. KSC argues that after defendant began asserting the right to obtain worldwide combined reporting, defendant nevertheless acquiesced in KSC’s continuing to report only its combined income excluding Kellogg’s other subsidiaries, both domestic and foreign. Defendant’s answer to this argument is that it has limited manpower and resources. Even if defendant discovers such problems, it might not act if the increase in tax would not be sufficient to justify the expense of the audit process. The evidence also indicates that defendant takes an extraordinary amount of time to audit and effect changes with regard to foreign corporations. (Exhibit B.)
KSC countered with evidence showing that significant differences existed during the years 1969 through 1971. Thus, defendant had sufficient motivation for requiring worldwide combined reporting. Since defendant did not, the logical inference is that defendant understood “company net income” was limited. This logic, however, leads to one further step which crumbles KSC’s position. If defendant really understood the implications, as KSC contends it must have, it would have terminated the agreement, required worldwide reporting and increased the amount of taxes collected. Even KSC acknowledges that had the agreement been terminated, it would have paid more taxes. There is no question that defendant always knew it could terminate the agreement at will. The court cannot assume that defendant, fully understanding the significant differences in income that could result, would continue a course of action inconsistent with its statutory duty and administrative policies.
KSC also contends that evidence of its own conduct strongly supports its position since its interpretation was against its own interest. In support of this argument KSC submitted evidence showing that if KSC had used worldwide combined reporting under the agreement, it would have saved approximately $11,467 over a seven-year period. (Exhibit 21.) KSC contends that this demonstrates KSC’s intent and *485understanding that “company net income” was intended by the parties to be restricted to the combined income of Kellogg and KSC. This argument is weakened by the fact that if KSC had used worldwide combined reporting, defendant might have terminated the agreement, in which case KSC would have paid more in tax than it did under the agreement.
Plaintiffs’ brief strongly asserts that both parties intended “company net income” to refer to the combined net income of Kellogg and KSC. The court agrees. However, no real significance can be attached to that fact when, at the time the agreement was executed, that was the only income considered taxable. KSC is attempting to take an additional step and read into the agreement an intent that no change in corporate organization, tax policies or anything else would affect the apportionable income. There is no evidence to support such a finding. To the contrary, the terms of the agreement make it clear that the agreement was to continue for future years only where no changes in “the company’s method of operations” occurred. (Exhibit 1, ¶ 3.) The termination at will clause of paragraph V also adds to this view. It is highly unlikely that either party would intend to establish something so important as the definition of net income by implication alone.
In summary, the court finds there was neither error nor misunderstanding by either party with regard to the term “company net income.” Both parties understood it to mean the combined income of Kellogg and KSC, excluding the foreign subsidiaries of Kellogg. However, there was no intent to freeze or fix the definition of “company net income.” That subject never came up. Consequently, the agreement does not prevent defendant from asserting that company net income should include subsequently acquired domestic subsidiaries of Kellogg or foreign subsidiaries of Kellogg. Plaintiffs’ position would require the court to read into the agreement an intent that such future changes would not affect the “company net income.” Such an interpretation is not supported by the evidence or the language of the document.
Two relatively minor3 issues remain.
*486Plaintiffs deducted payments of Michigan’s single business tax as a business expense. Defendant disallowed this expense on the grounds that it is an income tax paid to another state and therefore not deductible under ORS 317.265. That provision and its successor ORS 317.314 do not allow as a deduction in arriving at net income any “taxes upon or measured by net income or profits.” By an administrative rule effective December 31, 1984, defendant has specifically determined that the Michigan single business tax is a tax upon or measured by net income or profits. Plaintiffs challenge that rule and the defendant’s interpretation.
The leading and lone authority in Oregon on what constitutes a tax on net income is Keyes v. Chambers, 209 Or 640, 307 P2d 498 (1957). In that case the Oregon Supreme Court held that a Canadian tax of 15 percent on dividends received by Oregon residents from Canadian corporations was not a “net income tax” because no deductions or exemptions were granted. While that case provides some helpful guidance, it is not dispositive of the issue before this court. The emphasis here is not on whether the tax is gross or net but whether the nature of the tax is an income tax. Admittedly there are deductions and add-backs but are they designed to result in a tax upon “net income or profits”?
As defendant points out, “net income” is a generic term subject to many definitions. Undoubtedly this is because legislatures are allowed wide discretion in determining what credits, exemptions or deductions may be allowed. However, inherent in the common meaning of the words is the notion of a gain realized after payment of expenses necessary to earn income. While it may not be necessary that the gain be real in an economic sense, it does relate to a numerical measure of some net gain or profit. It does not appear that the Michigan single business tax purports to be so designed.
“The single business tax is designed to tax what a business has added to the economy, as distinguished from the income tax which taxes that which is derived from the economy. (Citations omitted). In that respect, it is the antitheses of the income tax which adds interest to the tax base but allows deductions for labor and services.” Town & Country Dodge v. Mich. Dept. of Treas. 118 Mich App 778 (1982), 325 NW2d 577, 580, 325 NW2d 577, 580, 118 Mich App 778 (1982).
*487Contrary to income tax principles, which allow deductions for expenses incurred in earning the net income or profits, such as wages, the Michigan single business tax adds wages to the “tax base” and deducts items which would normally constitute an element of income.
“Although federal taxable income is used as the starting point in computing the tax base, it is possible that a taxpayer may have no income and still be subject to the SBT. Other components of the tax base, e.g. wages, include expenses incurred, a theory not synonymous with income taxes.” Stockler v. State, Dept. of Treasury, 75 Mich App 640 (1977), 255 NW2d 718, 723-24.
The court agrees with plaintiffs that such a tax is not measured by or upon “net income or profits.” Defendant erred in disallowing plaintiffs’ deductions of the amounts paid.
In processing the audit of the separate tax returns of Kellogg and its various subsidiaries, defendant requested plaintiffs’ permission to “net the overpayments and underpayments.” Kellogg authorized defendant to do so. While not entirely clear from the statements and the parties’ briefs, it appears that defendant did not credit plaintiff with interest on the overpayments but did accrue interest on the underpayments. Defendant views this action as required by ORS 305.265(13), which required payments to be applied first to penalty, then to interest, and last to principal. Plaintiffs contend that:
“The Department’s interpretation has the effect of stopping interest accruing to the taxpayer on overpayments while permitting interest to continue to accrue on alleged deficiencies.” (Plaintiffs’ Opening Brief at 47.)
Plaintiffs further indicate that if they had known that this was the approach defendant would take, Kellogg would not have authorized the netting process.
Plaintiffs are correct. To net the over and underpayments as defendant did is to ignore the time differential. If defendant is to apply an overpayment without accruing interest, it should apply it to the underpayment before accruing any interest. If defendant believes that it must apply the netting process in accordance with ORS 305.265(13), then it must accrue interest in favor of plaintiffs on the overpayments and *488apply that accrued interest as well as the overpayments against the deficiencies and their accrued interest.
Judgment will be entered consistent with the above opinion. Costs to neither party.
OPINION ON RECONSIDERATION
Defendant has petitioned for reconsideration of the court’s decision on one issue. That issue concerns the proper procedure for accruing interest and crediting amounts when affiliated companies have authorized the defendant to “net the assessments and refunds” of such companies. After considering the oral arguments and memoranda of the parties, the court allowed reconsideration of this one issue.
The parties agree that the computations required under ORS 314.415 must be made prior to crediting amounts to a taxpayer account. This was done by defendant and plaintiffs do not question the amounts of interest credited for the overpayments or accrued for the underpayments.
After computing the interest on both the overpayments and the underpayments, defendant applied the KSC overpayments with interest to the underpayments of Fearn and Smith’s, applying the overpayments first to interest on the underpayments and then to principal. ORS 305.265(13). This left some tax (but no interest) owing which continued to accrue interest until paid by plaintiffs.
Plaintiffs contend that defendant is in error and that the proper “netting” procedure is to apply the overpayment of the KSC tax to the underpaid tax of Fearn and Smith’s, and *489the overpayment of the KSC interest to the underpaid interest of Fearn and Smith’s. This gives a different result because it retains the character of the amounts owing, i.e. the balance owing would consist of part tax, which continues to accrue interest, and part interest. The difference between the two methods is that under plaintiffs’ method, no interest accrues on that portion of the balance owing which constitutes interest because interest is not compounded.
Plaintiffs readily agree that the “netting” process they advocate is not found in the statutes or regulations. Rather, plaintiffs characterize it as a “consensual process” entered into by the parties for their mutual benefit and, therefore, in fairness ought to be “revenue neutral.” Although, as so characterized, the approach has much to recommend it, it is nevertheless contrary to the general rule pertaining to the satisfaction of liabilities. Under the general rule, amounts which are to be credited to an account are first applied against interest or other charges and lastly against principal. 45 Am Jur2d, Interest and Usury, § 99. First Nat. Bank v. Courtright, 82 Or 490, 158 Pac 277; 161 Pac 966 (1917). This is in recognition that interest generally does not earn interest (compound) and the party without possession of the principal is entitled to be compensated for its use by another.
Plaintiffs’ position attempts to read more into the process than reasonably should be. The department uses the same offsetting procedure where an overpayment from one year is applied against an underpayment of another year for the same taxpayer, i.e. interest is computed on both overpayments and underpayments and then all of the overpayments are applied first to interest on the underpayment and then to principal. See OAR 150-314.415(1)(d)-(A). When only one taxpayer is involved defendant is not required to obtain permission to make the offset. Here, defendant was required to obtain KSC’s permission to use its refund to satisfy the liabilities of Fearn and Smith’s. Plaintiffs may not read into such a request any new or unusual procedure not authorized by statute. This is so particularly in light of OAR 150-314.415(1)(d)-(A) which was in effect at the time plaintiffs agreed to “net” the assessments:
“When the combined report method is required (see ORS 314.363), penalty and interest under ORS 314.396 to ORS *490314.415 will be computed on the separate tax liability, or overpayment of each taxpayer included in the unitary group. There will be no offsets of overpayments and deficiencies between taxpayers in the group prior to computing penalty and interest. After computation of penalty and interest, an offset may be made by the department upon receiving authorization from the taxpayers, given the statute of limitations has not expired.”
In view of the above, the court must hold that plaintiffs are not correct and that defendant did correctly, after computing interest accruing on both the underpayments and the overpayments, “net the assessments and refunds” in accordance with ORS 305.265(13). As clarified, the decision of the court stands. Judgment will be entered as indicated by the opinion rendered October 15,1987.