OPINION and ORDER
LARKINS, District Judge:SUMMARY
This civil action comes before the court for the recovery of corporate income taxes alleged to have been erroneously or illegally assessed and collected by the United States. Jurisdiction is established in accordance with Title 28 U.S.C.A. § 1346 (a) (1). The cases are consolidated for the purposes of discovery, hearings, pretrials, trial and disposition.
Plaintiff complains, in Case No. 846, that it made an overpayment of its corporate income taxes for the year 1959 due to the erroneous or illegal computations of allowable bad debt reserves made by defendant’s agent. A timely claim for refund was made but disallowed in an alleged abuse of discretion by defendant’s agents who relied upon Section 166 (c) of the Internal Revenue Code of 1954.
The complaint in Case No. 847 raises questions substantially similar to those raised in Case No. 846 so far as the matters now being considered by the court are concerned. It complains, however, of defendant’s assessments for the year 1960.
Plaintiff has prayed for the recovery of the alleged overpayments, together with the interest thereon, and the costs of both actions. Plaintiff has also demanded a trial by jury.
Defendant has denied the substantial allegations of the complaints, although it has made some adjustments in its computations, all in plaintiff’s favor. Defendant also raised a preliminary issue of law which must be resolved prior to the trial of any issues which might be resolved by a jury.
This issue of law must be determined prior to trial, and to this conclusion both parties agree. It concerns the proper method of determining the computation of allowable additions to plaintiff’s bad debt reserves. Both sides resort to Internal Revenue Service Mimeograph 6209, and subsequent supplementary administrative rulings, to support their contentions.
The heart of the dispute developed from the merger of plaintiff, a commer*624cial bank, with five other banks during the period 1954 to 1960. At the time of each merger, the plaintiff and its respective merged components had substantially higher or lower loans outstanding,. and substantially higher or lower formerly established “bad debt losses to loans outstanding” experience.
Both the plaintiff and the defendant believe that the proper method of determining additions to bad debt reserves under a merger situation is computed by finding a “bad debt experience factor” in accordance with IRS Mimeograph 6209. In order to so compute a bad debt experience factor under the merger situation, the parties also agree that the experience of the parent and its merged components should be combined. Their agreement ends at these points, however, while the court has parted company with the parties from the outset, feeling that the formula stated in Mimeograph 6209 may not be applicable at all to the present situation.
At any rate, the parties and the court do agree that the predicted bad debt experience of the parent and of its components should be combined in some manner in order to more equitably establish future needs for purposes of the bad debt reserve deductions. The plaintiff insists the method of computing the combination of bad debt experience factors should be on the basis of a simple average of the parent and the appropriate components, while defendant contends for a weighted average basis. Both have submitted figures and computations to support their arguments, but the court is unwilling to accept either set of computations as correct in fact or theory.
FINDINGS OF FACT
A brief history of the events leading up to the present controversy is in order.
The parent bank, Peoples Bank and Trust Company of Rocky Mount, North Carolina, is a corporation organized April 1, 1931 with its home office in Rocky Mount, North Carolina. (Tr., p. 49). In 1954, a series of mergers were undertaken which resulted in the present controversy.
In 1954, the parent bank was thriving in the town of Rocky Mount, a town of some 27,000 people, and in some five neighboring towns where it had located branch banks. Rocky Mount has an economy based on textiles, tobacco, railroads and agriculture. It is a growing community showing a population increase of about 20% from 1954 to 1960, and it has also been industrializing during this period. The nature of the parent bank’s loan business ranged from large unsecured loans (in excess of $200,000.) to industrial loans, to consumer credit loans, to agricultural-mortgage loans. (Tr., pp. 53-56).
Prior to 1954 the parent bank had branches in five communities, as well as branches in and around Rocky Mount. The services offered at all these branches, as far as lending was concerned, conformed to that of the parent bank. (Tr., pp. 55-57). In spite of this conformity of services, the entire bank organization was of a “decentralized” nature. (Tr., p. 60).
In 1954 the Pinetops Banking Company of Pinetops, North Carolina, merged into the parent bank. Pinetops is an agricultural town of some 900 persons, and prior to the merger it made almost exclusively agricultural production loans. After the merger, the Pinetops bank began making consumer credit loans for the first time, as well as continuing making the agricultural loans. This change in lending policy was undertaken in spite of the fact that the town remained relatively unchanged in economic make-up and in size, and in spite of the fact that the managing personnel continued on in the merged component. (Tr., pp. 61-64).
In 1956, the newly-merged parent bank again merged, this time with the Bank of Enfield, located in Enfield, North Car*625olina. Enfield is a trading center for largely agricultural Halifax County, North Carolina. It has a population of approximately 3000 people, which has remained relatively constant. Prior to the merger, the Enfield bank made small business loans and agricultural loans. After the merger, consumer credit loans were added to the lending functions of the Enfield bank although the bank retained the same local management and employees. (Tr., pp. 64-67).
On February 1, 1958, a merger of the parent bank with the Bank of Edenton, Edenton, North Carolina, was consummated. Edenton is the county seat of Chowan County, and had a population of about 4300 people in 1958. It too, depends primarily upon agriculture as the base of its economic make-up, although it is also a trading center for the surrounding county, and has some textiles and historical attractions for passing tourists. Since the merger, the town has shown minor growth and influx of industry, but no material changes in the economic make-up of Edenton have occurred. Again, a consumer credit lending policy was adopted after the merger although the local management remained in charge of the newly-merged component. The business and commercial lending experience of the Edenton component materially expanded after the merger also. (Tr., pp. 67-70).
On February 23, 1960, the parent bank consummated two more mergers. One was with the Hertford Banking Company of Hertford, North Carolina, and the other was with the Commercial and Industrial Bank of Henderson, North Carolina.
The Hertford bank is located in the agricultural town of Hertford, which has a population of some 3000 people. The Hertford bank had previously engaged primarily in agricultural mortgage loans. After the merger a consumer loan function was adopted although the local management and employees were retained. (Tr., pp. 71-72).
The bank at Henderson was an industrial bank until 1955, when it began operating as a commercial and industrial bank. Henderson is a town of about 10,-000 people with an economic base of both industrial and agricultural make-up. The nature of the loans made were industrial, agricultural, and consumer credit loans. No new types of loans were added to the Henderson lending pattern, but an increase of loans of about 75% was noted in spite of the fact that the nature and size of the town did not undergo any material changes. (Tr., pp. 73-76).
It is necessary to further explore a few statistics presented to the court. Selecting at random from Defendant’s Exhibit “A” (which is a “Computation of Bad Debt Experience Peoples Bank and Trust Co. and Banks Absorbed via Merger”), will serve to illustrate. In the year 1935, the following computations are shown: the Rocky Mount bank (parent) had loans outstanding of $736,486.; with bad debts of $1867.; for a percent of bad debt experience of .2535%.
Pinetops shows loans of $60,253.; bad debts of $104.; for a percent of .1726%.
Enfield shows loans of $125,880.; bad debts of $1930.; a percent of 1.5332%.
Edenton: loans of $734,703.; bad debts of $13,138.; percent of 1.7882%.
Hertford: loans of $193,998.; bad debts of $6,356.; percent of 3.2763%.
There are no figures for Henderson in 1935, but the year 1936 shows loans of $104,233.; bad debts of $1,484.; percent of 1.4237%.
Turning now to a war year, from the great depression, the following appears for the year 1943:
Rocky Mount: loans of $694,577.; bad debts of $1,137.; percent of .1636%.
Pinetops: loans of $176,200.; bad
debts of $31.; percent of .0175%.
Enfield: loans of $315,411.; bad debts of 0; percent of O.
Edenton: loans of $1,156,110.; bad debts of $3,780.; percent of .2837%.
*626Hertford: loans of $164,234.; bad debts of $1,691.; percent of 1.0296%.
Henderson: loans of $120,599.; bad debts of $1,141,; percent of .9461%.
For the post-war year of 1947, the figures show:
Rocky Mount: loans of $2,929,375.; bad debts of $9,780.; percent of .3338%.
Pinetops: loans of $365,799.; bad debts of $374.; percent of .1022%.
Enfield: loans of $331,831.; bad debts of 0; percent O.
Edenton: loans of $1,618,933.; bad debts of $2,990.; percent of .1846%.
Hertford: loans of $414,923.; bad debts of $803.; percent of .1935%.
Henderson: loans of $396,694.; bad debts of $1,208.; percent of .3045%.
Looking next to Defendant’s Exhibit “B”, the last year of independent operation of each merged component, the bad debt factor and weighted factor are shown by defendant. Added thereto, for purposes of comparison, are the court’s computations of plaintiff’s simple average as the bad debt experience factor.
Pinetops Merger- — 3/2/54
Peoples Pinetops
Adusted loans = $7,553,458.23 $547,144.32
Bad debt experience = .36153% .47567%
Total adjusted loans = (7,553,458.23 plus 547,144.32) = $8,100,602.55
Defendant’s Computation of Peoples Experience = 7,553,458.23/8,100,602.55 x .36153% = .33709%
Defendant’s Computation of Pinetops Experience = 547,144.32/8,100,602.55 x .47567% = .03211%
Defendant’s Weighted Average = .36920%
Court’s Computation of Plaintiff’s Simple Average = .41860%
“Above figures based upon 12/31/53 call reports of each bank.”
Enfield Merger — 1/31/56
Peoples Enfield
Adjusted loans = $9,336,412.65 $702,698.49
Bad debt experience = .36920% 1.92888%
Total adjusted loans of both banks = $10,039,111.14
Defendant’s Computation of People’s Experience = 9,336,412.65/10,039,111.14 x .36920% = .34336%
Defendant’s Computation of Enfield Experience = 702,698.49/10,039,111.14 x 1.92888% = .13502%
Defendant’s Weighted Average = .47838%
Court’s Computation of Plaintiff’s Simple Average = 1.14904%
“Above figures based on 12/31/55 call reports of each bank.”
*627Edenton Merger — 1/31/58
Peoples Edenton
Adjusted loans = $11,196,406.30 $1,965,897.40
Bad debt experience = .47838% .89813%
Defendant’s Computation of Experience = .04691% .13418%
Defendant’s Weighted Average = .54109%
Court’s Computation of Plaintiff’s Simple Average = .68825%
“Above figures per 12/31/57 call reports of each bank.”
Henderson-Hertford Merger — 2/21/60
Peoples Henderson Hertford
Adjusted loans = $15,737,210.93 $2,296,891.21 $1,351,262.17
Bad debt experience .54109% .941585% .83063%
Defendant’s Computation of Experience .43926% .11158% .05789%
Defendant’s Weighted Average = .60873%
Court’s Computation of Plaintiff’s Simple Average = .77110%
“Above figures based on 12/31/59 call report of each bank.”
CONCLUSIONS OF LAW
The Internal Revenue Service has long recognized two methods of handling the bad debt deduction, whether it be by a bank or otherwise. One method is the “charge-off” method used where one’s accounting system is set up on the cash basis. The other method of handling the bad debt deduction is the “reserve method” used by plaintiff who is using the accrual accounting method.
In the “reserve method” the taxpayer is allowed to accrue an annual reserve against the probable debts which he predicts to go bad in the future. He thereby takes an annual deduction according to a formula designed to predict what percentage of his loans will ultimately be lost at the end of a tax year as bad debts.
In 1947, the Internal Revenue Service published Mimeograph 6209, which was directed at establishing the formula for computing bad debt reserves. It was entitled, “Reserve method of accounting for bad debts in the case of banks.” It is this Mimeograph 6209 which both parties insist, along with supplements thereto, control the case at hand. (For purposes of the case at hand, the most important supplement is Revenue Ruling 54-148, C.B.1954-1, 60; which provided plaintiff with “an alternative method of computing additions to a reserve for bad debts.” See Revenue Ruling 57-350).
The attorney for the plaintiff succinctly stated the formula generally recognized in the years 1959 and 1960, for determining bad debt reserves, as follows:
“ * * * beginning in 1928 and going through 1947, the ratio between the actual losses of the bank during that particular year and the outstanding loan of the bank as of December 31 of that year (is determined). * * * You add all of these twenty figures together, twenty ratios, and then divide by twenty, to determine an aver*628age ratio for the twenty-year period. The formula then says that you apply this average ratio * * * to the outstanding loans of the bank * * to determine what is the allowable addition to the reserve for the year 1959.” (Tr., pp. 7-8).
At this point it is necessary to recognize that a limitation is imposed by Mimeograph 6209 on the accruing of bad debt reserves.
Mimeograph 6209 states in pertinent parts the following:
«2. * * * However, such reserve can not be permitted to accumulate indefinitely * * *.
“3. The Bureau has accordingly approved the use by banks of a moving average experience factor for the determination of the ratio of losses to outstanding loans for taxable years beginning after December 31, 1946. * * * Such continued deductions will be allowed only in such amounts as will bring the accumulated total at the close of any taxable year to a total not exceeding three times the moving average loss rate applied to outstanding loans * * *.”
It is noted that no specific reference to the merger situation is made, although there is mentioned in Paragraph 5 of the Mimeograph some language which might be interpreted as applying. It has no language, however, which would be an aid to the problem at hand. Paragraph 5 states:
“5. A newly organized bank or a bank without sufficient years’ experience for computing an average as provided for above will be permitted to set up a reserve commensurate with the average experience of other similar banks with respect to the same type of loans, preferably in the same locality, subject to adjustment after a period of years when the bank’s own experience is established.”
The problem in this case is that there are no similarly situated banks from which plaintiff might draw information in order to base an experience. This bank is operating in rural, agricultural areas where few banks are located. As a result of the mergers now in discussion, lending policies on the part of banks were for the first time introduced into many of these communities.
The use of, or lack of use of Paragraph 5, therefore, serves to emphasize one basic feature of the purpose of computing the bad debt reserve, that is: to establish an equitable “method whereby there is an estimation in the current year as to * * * future uncollectibility of current loans. It is a prior deduction of loans that actually have not gone bad.” (Tr., p. 17, from discourse of Assistant United States Attorney.)
With this need for predictability in mind, the inherent weakness in both formulas advanced by the parties becomes apparent. The court concludes that the bad debt experience of the parent prior to the mergers will not necessarily remain a relatively stable computation subsequent to the mergers. In fact, due to the wide variances in statistical data presented in respect to size of loans, and previous bad debt experience, the parent bank’s post-merger data may reflect sizeable variations. This is especially true for the year or two immediately following the mergers because the parent bank will then be controlled, as to the merged component, by the loans made previous to the merger but still oustanding. Each component is likely to experience sizeable variations immediately after a merger such as this also because of the adoption of an altogether new element in its lending pattern.
All the components, at the time of each merger, were somewhat smaller in total amount of annual loans than the parent. Percentage-wise, however, their bad debt experiences covered a wide range. The influence of the parent was immediately felt on each banking component, and each underwent material changes in its lending pattern. Thus, the components need not have necessarily reflected similar bad debt experience after the mergers as that reflected by the parent either before or after the mergers. This latter conclusion *629is reached because of the wide variances in the economic structures and sizes of the communities in which all the components were located, and because of the retention of the established management and employees in the components in the decentralized organization. The bad debt experience of each of the components should be considered in trying to predict what the over-all bad debt experience of the new bank after the merger will be in the future, but it is not to be a controlling consideration. It is clear, an altogether new banking entity has been born and no component thereof will have a life experience in the future compare’d to that which it had in the past.
In this situation, total reliance on past experience, treating each merged component as an independent entity for purposes of predicting a future bad debt experience factor, totally ignores the economic facts of life. Obviously, an insistence on the formula set out in Mimeograph 6209, and supplemental Revenue Rulings, is erroneous. “Mim. 6209 does not have the force and effect of law, * * *.” See Correlator, Deductions, Bad Debts, CCH, Par. 1624.0334 at p. 25,096; discussing First National Bank at Wilkinsburg, 13 TCM 576, Dec. 20,418 (M).
In the years in question, 1959 and 1960, however, the Internal Revenue Service had published no other formula to serve as a guide in making the necessary computations. Today there is such a formula which has been found acceptable to both the Internal Revenue Service and the commercial banks of the United States. This formula has been established after a wealth of statistical data was analysed, computed and digested.
Advanced Revenue Ruling 65-92, IRB 1965-14, BAD DEBTS, has recently been published. The purpose of this ruling was to supplant Mimeograph 6209, and supplements thereto, in order to establish a “Uniform Reserve Ratio.” Section 6 of this ruling established a “Maximum Annual Reserve Addition” not to exceed a percentage of 0.8 of loans outstanding at the end of the taxable year. Allowable deductions for additions to its reserves for bad debts are also limited to not more than 2.4% of loans outstanding at the close of the taxable year. (This 2.4% being the three-year reserve limitation.)
Of course, Revenue Ruling 65-92 was not designed to be effective in a retroactive fashion, nor be implemented in all cases involving litigation concerning bad debt reserves and bad debt experience. But in a situation such as this, where the possibility of accurate predictability is seriously in doubt due to the numerous factors shown in the above quoted statistical data, some formula of broad and general application, which would also reflect general experience, is needed. This formula set out in Revenue Ruling 65-92 is such a formula; it is all-encompassing and establishes equitable guidelines fair to both sides.
The court recognizes that the information used in reaching the figures established in the Revenue Ruling 65-92 was not available to the parties at the time this dispute arose, and the litigation filed, but this does not mean that the court should not now consider the information and adopt this ruling as a guideline today, and no valid reason appears why the court can not now resort to the use of such advanced knowledge. No better formula has been shown the court, and no better one is anticipated, or the Internal Revenue Service would have surely adopted it.
For a discussion on problems similar to those raised herein, see North Carolina National Bank v. United States, 345 F.2d 544 (Ct.C1.1965).
ORDER
Therefore, it is ordered that the bad debt experience factor of the plaintiff, parent and merged components, be computed for the years 1959 and 1960 and recorded according to the formula stated in Revenue Ruling 65-92.
It is further ordered that the results of these computations be stipulated and agreed upon by counsel for plaintiff and counsel for the government, and submitted to the court.
*630It is further ordered that all other unresolved issues, if not waived, shall be determined by a jury.
It is further ordered that the Clerk serve a copy of this opinion and order upon all counsel of record.