Wilbur, Judge:
Respondent has determined the following deficiencies in petitioner’s Federal income tax:
Taxable period Deficiency
1972.$48,725
1973.79,284
1974.90,512
This case presents the following issues for our decision: (1) Whether petitioner, an automobile dealership engaged primarily in the purchase and retail sale of automobiles and trucks and which uses the dollar-value LIFO method of inventory valuation, may include all new vehicles in a single pool (or should include each model line of vehicle in a separate pool), and (2) whether respondent timely raised the issue of whether, assuming petitioner’s use of one pool for all new vehicles is proper, petitioner may treat all of its vehicles as a single item for purposes of computing a price index (or should treat each model of new vehicles as a separate item for such purposes).
FINDINGS OF FACT
Some of the facts have been stipulated. The stipulation of facts and the attached exhibits are incorporated herein by this reference.
Fox Chevrolet, Inc. (Maryland) (hereinafter referred to as petitioner or Fox) is a Maryland corporation having its principal office located in Baltimore, Md., at the time of the commencement of this suit. During the taxable years here in issue, petitioner timely filed its Federal income tax returns with the Internal Revenue Service Center at Philadelphia, Pa.
Petitioner is currently, and was during the taxable years 1972, 1973, and 1974, engaged in the purchase and retail sale of new and used automobiles and trucks under a franchise agreement with the General Motors Corp. (hereinafter referred to as G.M.). Under this agreement, Fox must stock and sell Chevrolet (a division of G.M.) automobiles and light trucks, service them, and carry their spare parts. Fox markets Chevrolets exclusively, carrying no other G.M.-made vehicles nor those of other manufacturers. Fox does not have a heavy-duty truck franchise.
The operation of petitioner’s business involved four departments during the taxable years here in question: a new vehicle department, a used vehicle department, a service department/body shop, and a parts department. A separate manager was employed to direct the operations of each department. These managers worked independently of each other, and their compensation was determined in part by reference to the relative performance of their respective departments.
The new vehicle department was responsible for the retail sale of all new automobiles and light-duty trucks. The department employed approximately 50 salespersons. Each salesperson had the right to sell any new vehicle carried by Fox. Selling commissions were based on a percentage of the profit on the unit sold. Since the gross profit was fairly constant regardless of the selling price of the vehicle (as will be explained shortly), the commissions were to a degree independent of the model being sold.
Pricing of the new vehicles for sale is initially done by the manufacturer at the factory. This is the “sticker price,” or manufacturer's suggested retail price, which is required to be posted on every new vehicle. These sticker prices are the same for every dealer of the particular vehicle. This price, however, is not necessarily the price for which the vehicle is intended to be sold.
Fox is not bound by the “sticker price” and often sells its inventory for a considerably less amount. Ignoring the suggested retail price, Fox determines what it hopes to eventually realize upon resale by adding a mark-up to its cost for the unit. This mark-up is not a percentage of the cost, but rather is a fixed dollar amount. An effort is made to maintain a constant mark-up which does not vary from unit to unit nor model to model. Thus this mark-up, or gross profit, ideally would be the same whether the vehicle being sold had a dealer cost of $4,000 or $10,000. In practice, however, deviations often occur as an additional profit can be realized when the model is in great demand or short supply. Conversely, a lesser profit must be accepted when the model is out of favor with the consumer. The following chart indicates Fox’s actual profit per unit by model line for each of the years here under consideration:
1972_1973_1974_
Model line Units Gross profit Units Gross profit Units Gross profit sold per unit sold sold per unit sold sold per unit sold
Regular 907 $355 885 $383 561 $404
Monte Carlo 167 450 227 713 243 469
Chevelle 425 $353 479 $361 440 $378
Camaro 35 387 75 427 169 482
Nova 425 346 376 378 383 365
Vega 358 400 608 372 447 426
Monza 1 627
Economy trucks 45 426 33 439 29 444
Light-duty trucks 176 344 213 417 291 457
Medium-duty trucks 18 370 18 463 84 461
Heavy-duty trucks 1 648 4 763 29 643
Total 2,557 367 2,918 407 2,677 419
What concerns Fox is how many units are sold — it is basically a volume business. The sales managers and the sales staff are ordinarily unconcerned about which models they sell and at what price, for both receive commissions based on a fixed percentage of a fairly constant amount — the gross profit on the unit sold.
Naturally, Fox would like to maintain an inventory consisting primarily of the models in the greatest demand. In practice, however, Fox has relatively little control over what it is required to stock. G.M. allocates its Chevrolets to its dealers based on each dealership’s past percentage of sales of the particular model being distributed. The number of orders placed with the factory is irrelevant to this procedure.
Naturally, this entails a situation whereby the dealer must accept the good with the bad. Of course G.M. is concerned with satisfying the consumer public, and to this end conducts extensive market studies. But demand for a certain model can shift abruptly, and it often takes a great deal of time for the factory to switch over production to a new model line. G.M. will not allow returns of models which are not selling, and Fox’s inventory therefore includes many units which are slow sellers. Consequently, each dealership has a special incentive to sell as many units of a popular vehicle as possible in order to receive a larger share of the next allocation.
Once a vehicle has been sold, it must be prepared for delivery to the customer. This preparation, service, and delivery procedure is the same regardless of the model or type of vehicle.
Fox maintained its books in accordance with the General Motors Dealers’ Standard Accounting System Manual. Under this system, retail sales of new cars were recorded in account numbers 400 through 405, and retail sales of new trucks were recorded in account numbers 420 through 423. Each sale was to be recorded at the selling price to the customer less discounts and allowances. Costs of retail sales were recorded in accounts 600 through 605 and 620 through 623 for new cars and trucks, respectively. Each vehicle was required to be costed individually as determined from the vehicle inventory records. Inventories of new cars and trucks were posted in accounts 231 and 237, respectively. The amounts in these accounts reflected the actual cost to the petitioner of the new vehicles it held for retail sale. For the sales and cost of sales accounts, each account category (e.g., 400, 401, etc.) corresponded to a separate model line (e.g., regular, Monte Carlo, etc.). However, the records are kept by account numbers and the model name is not required to be placed on the books (although Fox did so for its own convenience). No separate accounts by model line existed with respect to the inventory accounts, the only breakdown being between cars and trucks.
Under the terms of the franchise agreements that govern the relationship between G.M. and its dealers, all dealers are required to submit monthly financial reports to G.M. These reports are prepared from the dealer’s books and are thus in accord with the G.M. dealers’ standard accounting system. The financial reports break down the dealers’ operating results by profit centers: new car and truck department; used car and truck department; service department/body shop; and parts and accessories department. This breakdown corresponds exactly with the operational division of Fox’s business described earlier. These reports are further subdivided within the new car and truck department to show the number of units sold, gross profit and gross profit per unit by the G.M. standard account number, which in turn corresponds to each model line. American Motors Corp., Ford Motor Co., Chrysler Motors Corp., and Fiat all require their dealers to submit to them monthly financial reports in essentially the same form and departmental divisions as required by G.M.
During the taxable years here in issue, petitioner’s ending inventory consisted of the following:
Number of vehicles in ending inventory
1972 1973 1971
1. Regular (Impala and/or Caprice) O tH i — I 05 t-H CO to
2. Monte Carlo 05 CO rH O
20 3. Camaro 1
66 4. Chevelle 49 ^
78 5. Nova 10 CO
55 6. Vega 52
125 7. Light-duty trucks 18 CO
1323 463 Total 222
The first six categories represent the model lines of new automobiles carried by Fox. Number 7, light-duty trucks, was the only model line of new trucks included in Fox’s ending inventory during these years.
With its Federal income tax return for the taxable year 1972, petitioner filed a Form 970 properly electing to use the last-in, first-out (LIFO) method of valuing the inventories of new vehicles (cars and trucks) and parts. Prior to this time, Fox had properly valued its inventories utilizing the specific identification, lower of cost or market method. On the application, Fox elected to use the “dollar-value” method, the “double-extension” method of computing a price index, and the average acquisition cost method of valuing increments in its LIFO inventory. Petitioner also indicated that it was employing a single LIFO pool for its entire inventory of new cars and trucks and a separate LIFO pool for its parts inventories. Petitioner has employed these LIFO procedures consistently since the 1972 taxable year for both tax and financial reporting purposes.
Petitioner computed its LIFO inventories for new vehicles for the taxable years 1972,1973, and 1974 as shown in table I on pp. 714-715.
The use of the LIFO method can, during a period of rising costs, increase the cost of goods sold with a resultant decrease in taxable income. By way of illustration, table II on pp. 716-717 compares Fox’s cost of sales using the dollar-value LIFO method with one pool; the dollar-value LIFO method with separate pools by model lines; and the identification, lower of cost or market method for the taxable years 1972,1973, and 1974:
The term “LIFO Reserve” refers to the amount of increase in Fox’s cost of goods sold resulting from its use of the LIFO
[[Image here]]
method rather than the specific identification, lower of cost or market method. As these figures indicate, the crucial distinction between these methods which leads to the varying results for cost of sales is the valuation determined for ending inventory (which becomes the following year’s opening inventory). Fox’s ending inventory for the years 1972-74 under the various methods considered herein is shown in table III on pp. 720-721.
Petitioner determined that it incurred a net operating loss of $54,033 for the 1973 taxable year. In June 1974, petitioner filed a Form 1139 for a tentative carryback adjustment to its 1972 taxable year. The carryback adjustment was allowed by the respondent, and a refund of $2,329.37 including interest was paid to petitioner. On its Federal income tax return for the 1974 taxable year, petitioner carried forward $15,748 of its net operating loss from the 1973 taxable year.
Responent, in his statutory notice of deficiency, determined that petitioner had incorrectly computed the value of its new vehicle inventories under the dollar-value LIFO method and substituted the use of the specific identification, lower of cost or market method thereby increasing petitioner’s taxable income by $97,544, $229,794, and $177,156, for the taxable years 1972, 1973, and 1974, respectively. Respondent has since conceded that petitioner properly elected and is entitled to use the LIFO method of valuing its inventories of new vehicles and parts. However, respondent now contends that petitioner must employ separate LIFO pools for its different model lines of new vehicles (Fox had already established a separate pool for its parts inventory). This use of separate pools for different model lines of new vehicles results in the following increases or decreases in petitioner’s taxable income:
Taxable year Increase (decrease) in taxable income
$46,468 1972
143,899 1973
(8,301) 1974
In computing its new vehicle inventories, petitioner included in the cost of its new vehicles the price of factory and dealer-installed options which were placed on, and therefore became part of, the new vehicles in inventory. Respondent concedes that including the cost of the options in the price of new vehicles is proper and that a separate pool for options is not necessary. No adjustments were made by respondent with respect to petitioner’s LIFO inventory of parts.
Respondent further determined that petitioner did not incur a net operating loss in its 1973 taxable year and denied petitioner’s net operating loss carryback adjustment to its 1972 taxable year and its net operating loss carryforward to its 1974 taxable year.
OPINION
Petitioner is an automobile dealership engaged in the retail sale of automobiles and trucks as well as the service of such products. Having previously valued its inventory by the specific identification, lower of cost or market method, Fox elected, beginning with its 1972 taxable year, to compute its inventory by means of the LIFO method. As part of the LIFO election, Fox chose to use the “dollar-value” method with a single pool for all new vehicles, the “double-extension” method of computing a price index, and the average acquisition cost method of valuing increments in its LIFO inventory. It has consistently used these procedures for both tax and financial reporting purposes at all times since that election.
Issue 1. Pooling
Fundamental to the “dollar-value” method of computing the LIFO value of inventory is the concept of pooling. Pursuant to its application to change to the LIFO method of computing inventory, petitioner utilized a single pool for all new automobiles and trucks in its inventory. The respondent later determined that petitioner’s method of pooling did not clearly reflect income and required petitioner to recompute its income based on a separate pool for each model line of new automobiles or trucks.
Section 4711 requires the use of inventories whenever necessary in order to clearly reflect income. The regulations seek to define “necessary” as being whenever the production, purchase, or sale of merchandise is an income-producing factor. Sec. 1.471-1, Income Tax Regs. When inventories are required, they must be maintained on a basis that conforms as nearly as possible to
the best accounting practice in the trade or business and that most clearly reflects income. Sec. 471.
The selection of a method of taking inventory is done with the objective of tracking the flow of goods through inventory, either actually or constructively. The accuracy of this process is important because ending inventory is subtracted from opening inventory plus purchases to determine cost of goods sold. Bearing in mind that beginning inventory for any given year is equal to the ending inventory for the prior year, it is easy to see that an error in any year carries forward to every subsequent year. Furthermore, the amount of taxable income is increased by every dollar that cost of goods sold is decreased. The lower the value of the ending inventory the greater the cost of goods sold and the less taxable income which must be reported.
In arriving at the valuation of inventory, two distinct factors must be recognized. The first is the criteria to be used for valuation. The “bases” of valuation most commonly used and which are permissible under the tax laws are (1) cost and (2) the lower of cost or market. Sec. 1.471-2(c), Income Tax Regs. The second factor is the identification of items which make up an inventory. When practical, the specific identification method can be used whereby sales proceeds are matched with the actual cost (or other valuation) of the specific item sold. The actual physical movement of the merchandise through the inventory thus controls the income tax results. Where the goods have been so intermingled that they cannot be identified with specific invoices, the first-in, first-out (FIFO) method is often used. Sec. 1.471-2(d), Income Tax Regs. The FIFO method operates under the assumption that the earliest goods purchased are the first to be sold. Thus ending inventory will be deemed to consist of the most recently purchased items. Prior to 1972, Fox utilized the specific identification method of identifying the items in its inventory, and valued these items at the lower of cost or market.
Section 472 provides another method for the identification of the items in inventory, the last-in, first-out (LIFO) method. This method reverses the FIFO presumption and assumes that the most recently purchased goods are those being sold. Therefore, ending inventory is deemed composed of the earliest purchased goods. During a period of rising costs, use of the LIFO method in most cases results in lower taxes since the valuation of ending inventory is less (consisting of the earlier, less expensive items) and thus cost of goods sold is greater.
Beginning with its taxable year 1972, Fox properly elected to change its method of identifying its inventories to the LIFO method. Under this method, inventories may be valued only at cost. Sec. 472(b)(2). The theory behind LIFO is that income may be more accurately determined by matching current costs against current revenues, thus eliminating from earnings any artificial profits resulting from inflationary increases in inventory costs.
Initially, it was thought that the only method permissible for computing quantities of inventory when using LIFO was the “specific goods” method. Under this method, inventory quantities are measured in terms of physical units such as yards or pounds. These physical quantities of homogeneous items of inventory at yearend are compared with the quantity of like items in the beginning inventory to determine whether there has been a change during the year. Wendle Ford Sales, Inc. v. Commissioner, 72 T.C. 447, 452 (1979).
In practice, it quickly became evident that a large number of taxpayers were either reluctant or unable to elect the LIFO method where the inventory was either composed of a large number of dissimilar items or where the quantity of items was not readily susceptible to measurement by standard units.2 A solution was devised which did not necessitate the matching of specific items. It was determined that the unit of measure would be the dollar.3 Thus the valuation of ending inventory expressed in terms of dollars would be compared with the dollar value of the beginning inventory, and after proper adjustment for the change in the value, any increase or decrease in inventory could be computed. Respondent at first resisted the use of this so-called “dollar-value” method. The problem was particularly acute in the retail store industry and it was from there that the first attack came. We held that the identification of specific articles in an inventory was not a prerequisite for the application of the LIFO method and use of the “dollar-value” method was sanctioned. The basis for this decision was that Congress intended the LIFO method to be available for all taxpayers not just for those few for whom specific matching is feasible. Hutzler Brothers Co. v. Commissioner, 8 T.C. 14 (1947). Shortly thereafter, we reaffirmed our holding in Hutzler in a case involving a retail grocery business. Basse v. Commissioner, 10 T.C. 328 (1948). Thereafter, the Commissioner accepted the propriety of the “dollar-value” method by promulgating regulations permitting its use.
At the outset, we must determine whether the method of pooling selected by the petitioner conforms to the criteria set forth in the regulations. We have selected the regulations as our point of departure since section 472(a) states that “The change to, and the use of, such method shall be in accordance with such regulations as the Secretary may prescribe as necessary in order that the use of the method may clearly reflect income.” Therefore, any pooling arrangement which is in compliance with the regulations must be presumed, in the Secretary's opinion, to clearly reflect income.
Section 1.472-8(c), Income Tax Regs., dealing with principles for establishing pools for retailers, provides insofar as is relevant:
Items of inventory in the hands of wholesalers, retailers, jobbers, and distributors shall be placed into pools by major lines, types, or classes of goods. In determining such groupings, customary business classifications of the particular trade in which the taxpayer is engaged is an important consideration. An example of such customary business classification is the department in the department store. In such case, practices are relatively uniform throughout the trade, and departmental grouping is peculiarly adapted to the customs and needs of the business. * * *
In dealing with the criteria of “major lines, types, or classes of goods,” petitioner chooses the term “classes” to argue that all new Chevrolet vehicles constitute a single class of goods thus entitling them to be placed into a single pool. Respondent, on the other hand, selects the term “lines” and equates each model line with a “major line” of goods.
Were we to accept each party’s contentions at face value, petitioner’s method would be permissible since the regulatory standard is written in the disjunctive and appears to allow pooling by lines, types, or classes. Therefore, petitioner is as much entitled to pool by classes as by lines. However, we do not agree that all new Chevrolet vehicles constitute a single class of goods.
There are many types of vehicles which can be sold for transportation. Petitioner here sells two types — automobiles and trucks.4 Although automobiles and trucks share some common characteristics, on balance they are sufficiently dissimilar that we believe each represents a separate and distinct class of goods. Each appeals to a different type of purchaser. The market for automobiles is comprised in the main of persons among the general public who desire to acquire a means of transporting themselves between locations, usually within their community and occasionally on extended outings to more distant locales. Trucks, on the other hand, are more often bought for business use. They are used principally for transporting property. See Md. Transp. Code Ann. sec. 11-171 (Michie Supp. 1980); N.Y. Veh. & Traf. Law sec. 158 (McKinney 1970); Ind. Code Ann. sec. 9-1-1-2(d) (Burns 1980).
The nature of the operation of trucks is also quite different ■from automobiles. Due to their size, weight, and in some instances mechanical complexity, greater experience or training is occasionally required to operate trucks designed to haul property. Furthermore, the evidence in this case reveals that trucks are a more expensive investment than is the average automobile. The registration and other legal requirements for the operation of trucks also tend to be much more stringent than those for automobiles. See, for example, Md. Transp. Code Ann. secs. 16-104, 22-301, 22-407, 22-408 (Michie 1977), and secs. 13-912,13-916,13-917 (Michie Supp. 1980); N.Y. Veh. & Traf. Law secs. 375, 385,401 (McKinney 1970); Nev. Rev. Stat. secs. 484.737 et seq. (1979); Ind. Code Ann. secs. 9-1-4-41, 9-4r-l-64, 9-8-6-10 (Burns 1980).
Petitioner bases its argument on light-duty trucks, claiming that they are acquired by the same class of consumers and are used interchangeably for the same purposes as automobiles. While economy and light-duty trucks are certainly closer to automobiles than are heavy-duty trucks, and it is always difficult to draw a fine line, we believe that smaller trucks have more in common with other types of trucks than they do with automobiles. In any event, if light-duty trucks are a hybrid, they are far from fully interchangeable with automobiles. The line is difficult to draw, but is more appropriately drawn between trucks and cars than between different classes of trucks.
Having determined that there should be at least two pools— one for automobiles and another for trucks — we now turn to respondent’s contention that further subdivision by model line is required. First we must reiterate that we believe that, to the extent petitioner included all of its new automobiles in a single pool, this was a proper method of pooling by class of goods under the regulations. On the facts of this case we believe as a matter of logic, and the parties appear to have assumed on brief, that petitioner’s method and respondent’s method are mutually exclusive. It follows that respondent’s proposed method of pooling by model lines is improper.
At the heart of the LIFO method is the principle that income is more clearly reflected by matching current costs with current revenues. Where, however, a pool of inventory is depleted because sales exceed purchases during the year, the LIFO reserve is invaded and older “historic” costs flow into costs of sales. It is self-evident that the greater the number of pools the greater the likelihood of such a liquidation occurring.
This is of great concern in the automobile industry. Model lines change very rapidly due in large measure to forces beyond the dealer’s control.5 Often virtually indistinguishable automobiles are marketed under different model names. We perceive no reason for liquidations of inventory pools to occur every time a model begins to sell slowly or is phased out to be replaced by a very similar product.
Assuming a period of rising costs, with separate pools (respondent’s position), the LIFO reserve is reduced and income is increased as the dollars invested in one model line are shifted to investment in other model lines. Conversely, where one pool for new cars is maintained (petitioner’s position), the dollar investment in one model line may be freely invested in another model line without decreasing the LIFO reserve or increasing income.
Admittedly, dollar-value LIFO was developed in order to effectuate the congressional purpose of making LIFO available to all taxpayers when some might otherwise be excluded due to the practical problems of accounting for complex inventories by matching specific items. Hutzler Brothers Co. v. Commissioner, 8 T.C. 14, 29 (1947). While it may be arguable that Fox’s inventory is small enough to match specific units, we find no limitation in respondent’s regulations concerning the types of inventories which can be valued by the dollar-value LIFO method. Nor does respondent deny Fox’s right to use it.
While we agree that a pool should consist of items of a similar nature or of the same general category, we believe that Chevrolet automobiles are, despite differences, sufficiently similar in nature to be of the same general category — particularly in the business before us.
In determining groupings for purposes of pooling, the customary business classifications of the particular trade are made an important consideration by the regulations. Sec. 1.472-8(c), Income Tax Regs. This part of the LIFO regulations, we believe, merely reemphasizes the requirement of section 471 that the manner of taking inventories must conform to the best accounting practice in the trade or business, with effect being given to trade customs. Sec. 1.471-2, Income Tax Regs. Departmental grouping is a relatively uniform practice throughout the retail automotive trade and such a grouping is particularly adapted to the customs and needs of the business.
From the evidence before us, we must agree with petitioner that a grouping by department with all new vehicles included in a single department is the customary business classification in the retail automobile industry. Particularly persuasive in this regard is the testimony of two expert witnesses, both accountants with substantial experience in the automobile industry. Both stated that automobile dealers typically employ the departmental structure used by Fox, including a separate department for all new vehicles. This departmental division is utilized for both financial and managerial accounting purposes. Furthermore, in their experiences with dealers who utilize the dollar-value LIFO method, most maintain a single LIFO pool for all new vehicles. Both stated that a single LIFO pool for all new automobiles and trucks most clearly reflects income for financial accounting purposes.6
Given the departmental grouping of new vehicles which has been generally adopted within the retail automobile trade, departmental pooling seems particularly appropriate. We can perceive no real distinction between this situation and the department store described in the regulations, save the fact that a single manufacturer and a single type of product is involved herein.7 It cannot be reasonably contended that the grouping of items within the new vehicle department of an automobile dealership is more diverse than that of a typical department in a department store. Certainly there is at least as much similarity between a Vega and an Impala as between a sofa bed and a dining room table.
Respondent makes much of the fact that Fox is required by G.M. to submit reports showing sales data on a model line basis. However, the data are first broken down into separate departments (including the new vehicle department) before being further subdivided by model lines. Additionally, this information is submitted for the manufacturer’s use, and there has been no showing that it serves any particular purpose of the dealer. Respondent also contends that the customary business classification is the model line because consumers to some extent identify with various model lines. The common error in both of respondent’s arguments is that he fails to focus on the critical party— the dealer.
The relevant consideration is whether model lines represent a standard classification of retail automobile dealers for their new vehicle inventories. Nothing in the record before us indicates that this is an important classification to Fox and the other dealers. To the contrary, the record is replete with indications that model lines are not a major factor from the dealer’s standpoint. Fox is a volume dealer — the emphasis is on the total number of vehicles sold.8
Furthermore, Fox in practice treats its new vehicle department as a separate business division not further subdivided by model lines. A single manager is employed for the department. Every salesman is permitted to sell all model lines of new vehicles carried by Fox. Compensation of the department’s employees is determined independent of the type of vehicle sold. Additionally, the preparation and delivery procedure following sale is a uniform procedure entailing approximately the same costs to Fox.
Respondent also criticizes any analogy to the department store contained in the regulations. He argues that departmental grouping is particularly adapted to the customs and needs of a department store but not of an automobile dealership. We are given no reason to doubt, however, that the new vehicle department serves an important functional role. Irrespective of model line, Fox orders, sells, and delivers each vehicle following a uniform procedure. Thus practicality and convenience, both of the utmost importance to a businessman, are valid purposes served by this functional separation.
Respondent appears to recognize the merit of the functional division of Fox’s operation, but contends that the department store example in his regulations is based on an accounting rather than a functional division. However, there is no language in the regulations to support this distinction. We believe that the functional division is as important a factor in determining the parameters of a pool as is the financial accounting division. Furthermore, respondent offers us no reason to believe that the two cannot be coextensive.
Finally, respondent contends that Thor Power Tool Co. v. Commissioner, 439 U.S. 522 (1979), requires that we decide this case for him. But Thor involved an inventory write-down, and the Supreme Court, in sustaining respondent, specifically stated: “Although the Commissioner’s discretion is not unbridled and may not be arbitrary, we sustain his exercise of discretion here, for in this case the write-down was plainly inconsistent with the governing Regulations which the taxpayer, on its part, has not challenged.” (439 U.S. at 533. Fn. refs, omitted.) This case involves an entirely different question — the proper number of pools for dollar-value LIFO. Additionally, petitioner’s selections are not “plainly inconsistent with the governing Regulations”; quite the contrary, for the governing regulations and the cases that interpret them permit the procedures petitioner has employed.9 We therefore, on the record before us, find that respondent has been arbitrary and has abused his discretion.10
We have determined that pooling of all new vehicles (except trucks) is the appropriate classification for LIFO inventory-purposes. For the reasons given earlier, we believe that a separate pool for trucks is required to clearly reflect income. In sum, we hold that two pools, one for automobiles and another for trucks, conforms as nearly as may be to the best accounting practice in petitioner’s business and most clearly reflects income.
Issue 2. Items
Having determined that petitioner is entitled to utilize one pool for all new automobiles and a second pool for all new trucks, we now turn to respondent’s alternative contention that, for purposes of applying the double-extension method of computing the LIFO value of a dollar-value pool, each model line of new vehicles constitutes a separate “item.” Petitioner, since its election of the LIFO method, has treated all new automobiles and trucks as a single item, and urges that it is entitled to continue this treatment. As our holding necessitates the fragmentation of automobiles and trucks into separate pools, petitioner’s argument must be understood to be that, within each pool, the various models should be treated as a single item.11
The regulations contain four different methods for computing the LIFO value of a dollar-value pool.12 The purpose of applying these methods is to determine whether or not there has been an increment or decrement in the value of the pool so that ending inventory may be calculated.
Petitioner properly elected to employ the “double-extension” method. Under this method, the quantity of each item in the inventory pool at the close of the taxable year is extended at both base-year13 unit cost and current-year14 unit cost. The extensions of the two costs are each totaled, the former yielding the amount of ending inventory at base-year cost and the latter providing the amount of ending inventory in terms of current-year cost. Sec. 1.472-8(e)(2)(i), Income Tax Regs.
In order to determine whether there has been an increment or decrement in the pool, ending inventory expressed in terms of base-year cost is compared with opening inventory also at base-year cost. Where ending inventory is greater, there has been an increment. To compute the LIFO value of this increment, the base-year cost of the increment (base-year cost of ending inventory minus base-year cost of opening inventory) is multiplied by the ratio of the total current-year cost of the pool to the total base-year cost of the pool.15 Conversely, if comparison reveals that opening inventory exceeds the ending inventory, there has been a decrement. This decrement (base-year cost of opening inventory minus base-year cost of ending inventory) or liquidation, is used to reduce the layers of increment starting with the most recent layer and working in reverse chronological order. Base-year inventory is reduced only when all of the layers of increment have been eliminated.16 Sec. 1.472-8(e)(2)(iv), Income Tax Regs.
The term “item” is nowhere defined in the regulations. The first time we were confronted with this problem we held that in the case of a retailer of goods an “item” refers to a finished product of inventory and not to its individual components. Wendle Ford Sales, Inc. v. Commissioner, 72 T.C. 447, 455 (1979).
Petitioner attempts to avoid this issue entirely by claiming that it was not timely raised. He claims that respondent first officially placed the question of what constitutes an item into issue during cross-examination of petitioner’s expert witnesses. The respondent, on the other hand, argues that the statutory notice was broad enough to encompass this question and that, in any event, respondent informed petitioner shortly before trial that he contemplated raising this issue.
After exhaustively examining the record in this case, we do not believe that the issue was properly and timely raised. No mention of it is made in any of the pleadings.17 The stipulations and exhibits prepared for the trial of this case suggest that respondent was satisfied with petitioner’s “double-extension method” computations, as they were used as the basis for the alternative calculations presented to the Court. In this regard, alternative computations were performed to demonstrate the relative impact of a single pool, multiple pools by model lines, and specific identification (Fox’s former inventory method). Yet no similar effort was made to show the distinctions and effects of one item versus multiple items within a single pool, even though this issue is more mathematically complex than the pooling issue and much could have been gained by such a display. And, in addition to all of this, respondent’s trial memorandum submitted to the Court on October 22, 1979, 2 days prior to the hearing, discusses only the pooling issue, although the computations involved in applying the double-extension method are discussed at length.18 The only clue to the existence of a dispute, respondent’s informal mention of the “items” question during pretrial discussions,19 hardly suffices to raise an issue where ample opportunity is available to do so formally in a judicially recognized manner. See Estate of Horvath v. Commissioner, 59 T.C. 551, 556 (1973); Rule 41, Tax Court Rules of Practice and Procedure.
Under the circumstances of this case, even the informal notification came late. In his opening brief, counsel for the petitioner admits having been informed by respondent during the week prior to trial. In his reply, respondent’s counsel does not dispute the time of the informal notice. An issue as complicated as this cannot reasonably be adequately dealt with in a week’s time. Those cases in which we have held that advance warning can mitigate the disadvantages of respondent’s failure to act formally either involved much earlier notification (Schuster’s Express, Inc. v. Commissioner, 66 T.C. 588, 592 (1976), affd. without published opinion 562 F.2d 39 (2d Cir. 1977), 40 AFTR 2d 77-5293, 77-2 USTC par. 9495 (apprised of respondent’s intention several months prior to trial); Rubin v. Commissioner, 56 T.C. 1155, 1163-1164 (1971), affd. 460 F.2d 1216 (2d Cir. 1972) (informed well in advance of trial)) or no claim of surprise by the petitioner. Nat Harrison Associates, Inc. v. Commissioner, 42 T.C. 601 (1964). The importance of adequate preparation time as a critical factor in this determination has been noted:
Although the most appropriate times to advise the taxpayer of the Commissioner’s theories to sustain an assessment would be first in the notice of deficiency and then in the Commissioner’s answer, we do not hold that the Commissioner necessarily loses his right to pursue a theory or Code section that is not specifically raised before or at trial. The basic consideration is whether the taxpayer is surprised and disadvantaged when the Commissioner has failed to plead section 482. * * * However, the longer the Commissioner delays in not expressly advising the taxpayer of the intended theories, the more reason there is to conclude that the taxpayer had not received fair notice and has been substantially prejudiced so as to deny the Commissioner consideration of theories raised for the first time in post trial briefs. The Commissioner may avoid this uncertainty and discharge his duty of informing the taxpayer by expressly notifying the taxpayer of the intended theories in the deficiency notice and the Commissioner’s answer. [Commissioner v. Transport Mfg. & Equip. Co., 478 F.2d 731, 736 (8th Cir. 1973). Citations omitted.]
Where the parties were at odds over a similar issue, this Court resolved the matter in petitioner’s favor, emphasizing the importance of petitioner’s conduct:
Nothing in petitioner’s actions indicates to us that she was aware that any issue other than the statute of limitations would be argued. Petitioner’s opening remarks to this Court and all of her evidence were directed to the statute of limitations. Indeed, from the pleadings herein, it was not unreasonable for petitioner to assume that the validity of the debt was not seriously in dispute. Further the facts required to face adequately the statute of limitations question presented are almost totally different from those necessary to prove the underlying validity of the debt. To require petitioner to argue the validity of the debt at the trial under these circumstances would severely prejudice her case and violate the fundamental fairness necessary in carrying out the judicial function. We hold that petitioner was surprised and that her ability to present her case was severely prejudiced by such surprise. [Estate of Horvath v. Commissioner, 59 T.C. 551, 556 (1973).]
The same is true here. Not only was petitioner’s conduct consistent with a lack of awareness, so also was respondent’s. The joint exhibits and stipulations were prepared without taking this issue into consideration. The trial memorandums fail to mention it. Finally, respondent submitted absolutely no proof of his own on this question but relied entirely on favorable responses from witnesses brought to court by the petitioner. We have no trouble finding surprise here. We do not believe that the petitioner had “fair warning” that the issue of what constitutes an item would be raised. Cf. Commissioner v. Chelsea Products, Inc., 197 F.2d 620, 624 (3d Cir. 1952), affg. 16 T.C. 840 (1950).
We also find that petitioner was prejudiced or harmed in its ability to prepare for trial. Evidence was adduced at trial concerning the pooling question. However, the considerations applicable to that issue are different than those which are required to answer the question of what constitutes an item within that pool. While we cannot be sure what petitioner may have done had the issue been timely raised, it does state on brief that it “would have prepared an entirely different case which would have focused on the nature of automobile models, the annual model change, the nature of accessories, inflation in costs by model, the substitution of replacement models over time, and the reconstruction of base year costs for new models.” Petitioner further notes that the record in Wendle Ford Sales, Inc. v. Commissioner, 72 T.C. 447 (1979), reveals a great deal of evidence as to the nature of technological changes which have occurred in the manufacture of automobiles, but is devoid of evidence of the type presented in this case on the pooling issue since the parties there had stipulated as to the proper pooling method to be employed.
The extent of the evidence petitioner would have actually produced, what it would have shown, and how it might have influenced the Court are all matters of speculation. This issue, however, is extremely complicated and the answer to it is likely to have a very pervasive impact on the automobile industry as well as many taxpayers who have adopted dollar-value LIFO. The Commissioner will always have other opportunities to attack this problem, but any prejudice to this petitioner would be permanent. We cannot decide such an important question on an insufficient record.
Decision will be entered under Rule 155.