In this appeal from a decision of the Tax Court, we are asked to determine whether certain costs incurred by banks for marketing, researching and originating loans are deductible as "ordinary and necessary expenses" as provided by section 162 of the Internal Revenue Code, 26 U.S.C. S 162 (1988), or whether these expenses must be capitalized under section 263 of the Code. Two banks that were predecessors in interest of appellant PNC Bancorp, Inc. deducted these costs as ordinary business expenses. The Internal Revenue Service disallowed the deductions and issued statutory notices of deficiency. PNC filed petitions for redetermination with the Tax Court. The Tax Court determined that the expenses in question were not deductible, but, instead, must be capitalized and amortized over the life of the subject loans. PNC now appeals from this determination.

We hold that the costs at issue were deductible as "ordinary and necessary" expenses of the banking business within the meaning of Internal Revenue Code section 162, and that these costs do not fall within the purview of section 263. Accordingly, we will reverse the judgment of the Tax Court.

The costs that the banks seek to deduct are the internal and external costs that they incur in connection with the issuance of loans to their customers. These costs, discussed in more detail below, are a routine part of the banks' daily business, and the services procured with these outlays have been integral to the basic execution of the banking business for decades.

The general contours of banks' involvement in making loans have not changed dramatically in recent years, and the relevant sections of the Tax Code have remained largely unchanged. Historically, the costs at issue have been deductible in the year that they are incurred; however, the Commissioner rejected this tax treatment by PNC. Why is the Commissioner now insisting upon capitalization of these costs?

There are two relatively recent developments that appear to have emboldened the Internal Revenue Service to pursue capitalization of such costs. One of these developments is the Supreme Court's opinion in INDOPCO, Inc. v. Commissioner, 503 U.S. 79 (1992), in which the Court held that expenses incurred by a target corporation in the course of its friendly acquisition by another entity were not currently deductible. See id. at 90. INDOPCO, which signaled that the Supreme Court's previously announced tests for capitalization were not exhaustive, may well have been viewed by the IRS as a green light to seek capitalization of costs that had previously been considered deductible in a number of businesses and industries. This phenomenon has not escaped comment from observers. See, e.g., W. Curtis Elliott Jr., Capitalization of Operating Expenses After INDOPCO: IRS Strikes Again, S.C. Law., Sept./Oct. 1993, at 29, 29, 30 (commenting on the IRS's "recently aggressive posture on capitalization" after INDOPCO, and noting that while the INDOPCO decision itself was not "necessarily troubling," the IRS's interpretation of it has stretched far beyond the scenario presented in INDOPCO); IRS Loses Battle in INDOPCO War: Advertising Remains Deductible, Taxes on Parade, July 16, 1998, at 1 (describing the "IRS's INDOPCO-fueled juggernaut"). Thus, INDOPCO ushered in an era of generally more aggressive IRS pursuit of capitalization.

An additional development may have prompted the IRS's assertive posture in the more specific case of the loan origination costs at issue here. This second development was the Financial Accounting Standards Board's promulgation of a new standard for financial accounting treatment of loan origination costs, Statement of Financial Accounting Standards No. 91 ("SFAS 91").1 Beginning in the late 1980s, SFAS 91 required for the first time that, for financial accounting purposes, loan fee income and the costs incurred in connection with loan origination should be deferred and recognized over the life of the loan, rather than being recognized in full in the year the loan closed.2 The FASB's authority extends only to financial accounting standards and not to tax accounting standards. For the first few years of SFAS 91's existence, the IRS did not require capitalization of the loan origination costs described in this financial accounting standard. However, the IRS apparently viewed INDOPCO as a reason to pursue capitalization of the costs that SFAS 91 requires to be deferred.3 Thus, the stage for this litigation was set.

PNC Bancorp, Inc. (PNC) is a bank holding company incorporated in Delaware. See A. at 102. Two smaller banking entities, First National Pennsylvania Corporation^PC and United Federal Bancorp, Inc. (UFB), were merged into PNC in 1992 and 1994, respectively, and PNC succeeded to the liabilities of both these companies. See A. at 103, 105. The activities at issue in this case occurred before the mergers and were performed by FNPC and UFB or their subsidiaries.4

The costs challenged in this appeal were incurred by both banks in connection with the origination of loans.5 There are two categories of such "loan origination costs," as they have been called.6 The first category includes payments made to third parties for activities that help the bank determine whether to approve a loan (credit screening, property reports, and appraisals) and for the recording of security interests when the bank decides to issue a secured loan. The second category consists of internal costs, namely that portion of employee salaries and benefits that can be attributed to time spent completing and reviewing loan applications, and to other efforts connected with loan marketing and origination.7 The Commissioner pursued capitalization of loan origination costs only when those costs were incurred in connection with a loan that was later approved; the Commissioner allowed the banks to deduct origination costs expended in connection with loans that were not successfully approved. See PNC Bancorp, Inc. v. Commissioner, 110 T.C. 349, 359, 362 (1998); see also Tr. of Oral Argument at 7.

Loan interest constituted the largest source of revenue for each bank during the relevant time period, and interest on deposits and other borrowing constituted the largest expense. See A. at 108.8 It is undisputed that banks generally can be profitable only if they successfully manage their "net interest margin" -the difference between interest earned and interest paid. A profitable bank's net interest margin plus its revenues from fees and other sources must exceed its losses on loans and investments. See A. at 108.

Bank personnel routinely undertook loan marketing activities in tandem with other marketing and customer service functions. Both tellers and "platform employees" (those bank employees who have desks apart from the teller windows) were encouraged to "cross-sell," that is, to sell multiple products to existing and new customers who came to the bank in search of a particular product or service. For example, if a new customer opened a checking and savings account, the bank representative might also suggest a certificate of deposit or a loan. Likewise, when a consumer applied for a loan, the employee taking the application was also expected to sell other bank products and services (such as checking accounts, credit lines, or ATM cards) during that same session. The banks provided financial incentives to their tellers and platform employees for each successful "cross-sale," see A. at 110, and such "crosssales" were a routine part of each bank's daily business.9

Before 1988, FNPC and UFB treated their loan origination costs in the same manner as their other routine expenses, both for tax accounting and financial accounting purposes. That is, they reported these costs for the tax year (and the fiscal year) in which the costs were incurred. This practice was apparently standard in the banking industry at that time. See A. at 169. In 1988, following the promulgation of SFAS 91, FNPC and UFB began to separate out their loan origination costs for financial accounting and reporting purposes in order to conform with SFAS 91's requirements. However, both banks continued to deduct loan origination costs for tax purposes in the tax year in which the loan closed. See A. at 124, 134. It is these deductions that the Commissioner and the Tax Court disallowed.

The Tax Court had jurisdiction over PNC's petitions for redetermination pursuant to 26 U.S.C. SS 6213 and 7442. We have appellate jurisdiction pursuant to 26 U.S.C. S 7482(a)(1), which states that "[t]he United States Courts of Appeals . . . shall have exclusive jurisdiction to review the decisions of the Tax Court . . . in the same manner and to the same extent as decisions of the district courts in civil actions tried without a jury." Thus, we have plenary review over the Tax Court's findings of law, including its construction and application of the Internal Revenue Code. See National Starch & Chem. Corp. v. Commissioner, 918 F.2d 426, 428 (3d Cir. 1990), aff 'd, INDOPCO, Inc. v. Commissioner, 503 U.S. 79 (1992). We review the Tax Court's factual findings and inferences for clear error. See id.; see also Pleasant Summit Land Corp. v. Commissioner, 863 F.2d 263, 268 (3d Cir. 1988).

There is a fundamental distinction between business expenses and capital outlays. The primary consequence of characterizing a payment as a business expense or a capital outlay "concerns the timing of the taxpayer's cost recovery," INDOPCO, Inc. v. Commissioner, 503 U.S. 79, 83 (1992): business expenses are deductible in the year in which they are incurred, whereas a capital outlay is generally "amortized and depreciated over the life of the relevant asset, or, where no specific asset or useful life can be ascertained, is deducted upon dissolution of the enterprise," id. at 83-84.

Two sections of the Internal Revenue Code address the deductibility vel non of expenditures such as those incurred by FNPC and UFB. Section 162 of the Internal Revenue Code provides in pertinent part: "There shall be allowed as a deduction all the ordinary and necessary expenses paid or incurred during the taxable year in carrying on any trade or business . . ." 26 U.S.C. S 162(a). Section 263 of the Code states that capital expenditures, i.e., "amount[s] paid out for new buildings or for permanent improvements or betterments made to increase the value of any property or estate," cannot be currently deducted. 26 U.S.C.S 263(a)(1). It is true that these two sections are neither all-inclusive nor mutually exclusive. See General Bancshares Corp. v. Commissioner, 326 F.2d 712, 716 (8th Cir. 1964) (written by then-Judge Blackmun). For example, it is possible that an expense that might appear to be deductible under section 162(a) might instead be required to be capitalized because it also properly falls under the description provided by section 263(a). If an expense were to fall under the language of section 263(a), that section would "trump" the deductibility provision of section 162(a) and the expense would have to be capitalized. Thus, in order to be deductible, the expense must both be "ordinary and necessary" within the meaning of section 162(a) and fall outside the group of capital expenditures envisioned by section 263(a).10 Nonetheless, the two sections represent the archetypes of the two opposing alternatives for tax treatment of expenditures -deduction and capitalization -and, ordinarily, an expenditure will fall under one or the other section, not both.

The taxpayer bears the burden of showing that a given expenditure is deductible. See Interstate Transit Lines v. Commissioner, 319 U.S. 590, 593 (1943), quoted in INDOPCO, 503 U.S. at 84. In order to demonstrate deductibility under section 162(a) of the Code, the taxpayer must meet a five-part test. "To qualify as an allowable deduction under S 162(a) . . . , an item must (1) be `paid or incurred during the taxable year,' (2) be for `carrying on any trade or business,' (3) be an `expense,' (4) be a `necessary' expense, and (5) be an `ordinary' expense." Commissioner v. Lincoln Sav. & Loan Ass'n, 403 U.S. 345, 352 (1971). It is clear that PNC's loan origination expenses can satisfy the first four parts of this test.11 The question before us under S 162, then, is whether these expenses qualify as "ordinary" business expenses within the meaning of that section.

In determining what expenditures qualify as "ordinary," we must look to the particular facts of the case before us, including the particular puzzle posed by the circumstances of the banking industry. As Justice Cardozo stated nearly seventy years ago in interpreting an earlier version of this long-standing Code provision, ordinariness is "a variable affected by time and place and circumstance." Welch v. Helvering, 290 U.S. 111, 113-14 (1933). In interpreting the Code, we should not stray from the moorings of the "natural and common meaning" of the term "ordinary," id. at 114, and in doing so must examine the nature of the day-to-day operations of the particular business being considered. See also Deputy v. du Pont, 308 U.S. 488, 49596 (1940) (stating that each case "turns on its special facts," and that an expense that is ordinary-"normal, usual, or customary" -in one business may not be ordinary in another). Justice Cardozo's oft-quoted words regarding the heavily case-specific nature of this inquiry are no less appropriate today than they were in 1933:

Here, indeed, as so often in other branches of the law, the decisive distinctions are those of degree and not of kind. One struggles in vain for any verbal formula that will supply a ready touchstone. The standard set up by the statute is not a rule of law; it is rather a way of life. Life in all its fullness must supply the answer to the riddle.

Welch, 290 U.S. at 114-15; see also Commissioner v. Lincoln Sav. & Loan Ass'n, 403 U.S. 345, 353 (1971). Accordingly, we pursue a real-life inquiry into whether the expenditures associated with loan marketing and origination are "ordinary" expenses incurred in the day-to-day maintenance of a bank's business.

The Commissioner has conceded that loan interest was the banks' largest revenue source during the period in question, and that interest payments on deposits and other borrowing were their largest expense. There is no reason to suppose that this time period was any different from any other in this regard. Further, maximizing the "net interest margin" -the difference between interest received and interest paid out -is the principal manner in which banks earn their keep. As the Tax Court stated, "[t]he principal businesses of [the banks] consisted of accepting demand and time deposits and using the amounts deposited, together with other funds, to make loans." PNC, 110 T.C. 351. Modern banks are essentially dealers in money. See United States v. Philadelphia Nat'l Bank, 374 U.S. 321, 326, 327 n.5 (1963).

Given this context, the ordinary nature of the costs at issue, routinely incurred in the banks' businesses, would seem clear. In order to ensure deductibility, however, we must also ascertain whether these costs were expended for betterments to increase the value of property in a way that would require these costs' capitalization under S 263. We cannot conclude that in performing credit checks, appraisals, and other tasks intended to assess the profitability of a loan, the banks "stepped out of [their] normal method of doing business" so as to render the expenditures at issue capital in nature. Encyclopaedia Britannica, Inc. v. Commissioner, 685 F.2d 212, 217 (7th Cir. 1982). As we stated in National Starch, an important determination is whether given expenditures "relate to the corporation's operations and betterment into the indefinite future," indicating the need for capitalization, or are instead geared toward "income production or other current needs," suggesting deductibility. National Starch & Chem. Corp. v. Commissioner, 918 F.2d 426, 433 (3d Cir. 1990), aff 'd, INDOPCO, Inc. v. Commissioner, 503 U.S. 79 (1992). The facts before us demonstrate that loan operations are the primary method of income production for the subject banks. We have no doubt that the expenses incurred in loan origination were normal and routine "in the particular business" of banking. See Deputy v. du Pont , 308 U.S. at 496.

The Commissioner argues, and the Tax Court found, that the Supreme Court's decision in Lincoln Savings requires a different result. We disagree. In Lincoln Savings, the Supreme Court concluded that payments made by Lincoln Savings and Loan Association into a "Secondary Reserve" fund at the Federal Savings and Loan Insurance Corporation (FSLIC) were not deductible as ordinary business expenditures. See Lincoln Savings, 403 U.S. at 354, 359. In so holding, the Court upheld the IRS's distinction between payments that Lincoln made into the FSLIC's "Primary Reserve," which the IRS had found to be deductible as ordinary and necessary expenses, and those payments made into the "Secondary Reserve," found to be capital expenditures. The Court engaged in an extensive analysis of the nature of each reserve fund and the premium payments made into each by Lincoln Savings and other similarly situated FSLIC-insured institutions. The Court noted that the "only concern" was whether the premium payment to the Secondary Reserve "was an expense and an ordinary one within the meaning of S 162(a) of the Code." Lincoln Savings, 403 U.S. at 354. The Court noted that the fact that many institutions were required to make such a Secondary Reserve premium did not render that premium an ordinary expense, see id. at 358; nor did the fact that the premium could have some ensuing benefit to Lincoln Savings, in and of itself, render the premium a capital expenditure, see id. at 354 ("many expenses concededly deductible have prospective effect beyond the taxable year"). Rather, the Court focused on what the payment represented in the context of Lincoln Savings' business and of the "structure and operation of FSLIC's reserves":

What is important and controlling, we feel, is that the [Secondary Reserve] payment serves to create or enhance for Lincoln what is essentially a separate and distinct additional asset and that, as an inevitable consequence, the payment is capital in nature and not an expense, let alone an ordinary expense, deductible under S 162(a) in the absence of other factors not established here.

Id.12 Whereas Lincoln Savings and the other insured institutions had no property interest in the Primary Reserve, each did have an earmarked property interest in the Secondary Reserve that was carried as an asset on each institution's books, was enhanced by each institution's contribution, and was refundable to that institution in certain circumstances.

In the case at bar, the Tax Court concluded without any elaboration that the consumer and commercial loans "clearly" were separate and distinct assets of the banks, see PNC, 110 T.C. 364, and that the costs incurred in originating and processing the loans "created" these separate and distinct assets, see id. at 366. We believe that the Tax Court took too broad a reading of what Lincoln Savings meant by "separate and distinct assets," as well as an overbroad reading of what can be said to "create" such assets.

The Secondary Reserve fund in Lincoln Savings was a "separate and distinct asset" in two important ways that distinguish it from FNPC's and UFB's loans, the assets in question here. First, the Secondary Reserve fund was an asset that existed quite apart from Lincoln's main daily business of taking deposits and making loans; second, the fund was an asset that, although it existed within the FSLIC, was nonetheless separate from the FSLIC's other revenues and distinctly earmarked as Lincoln's property. The Tax Court's broad reading of Lincoln Savings essentially treats the term "separate and distinct asset" as if it extends to cover any identifiable asset. We do not subscribe to this reading of Lincoln Savings.

Furthermore, we do not agree that the marketing and origination activities actually "created" the banks' loans in the same way that the activities in Lincoln Savings created the Secondary Reserve fund. In the instant case, the Tax Court proceeded from the clearly accurate premise that the expenses in question were associated with the loans, incurred in connection with the acquisition of the loans, or "directly related to the creation of the loans," PNC, 110 T.C. 368, to the faulty conclusion that these expenses themselves created the loans. See id. at 364-68. We conclude that the term "create" does not stretch this far. In Lincoln Savings, it was the payments themselves that formed the corpus of the Secondary Reserve; therefore, it naturally follows that these payments "created" the reserve fund. In PNC's case, however, the expenses are merely costs associated with the origination of the loans; the expenses themselves do not become part of the balance of the loan. PNC argues persuasively that the Tax Court's interpretation of the Lincoln Savings language is inappropriately expansive:

While purporting to apply the Lincoln Savings language, both the Tax Court and the government effectively have transformed that language, by subtle but significant degrees, from a test based on whether a cost "creates" a separate and distinct asset, into a much more sweeping test that would mandate capitalization of costs incurred "in connection with" or "with respect to" the acquisition of an asset.

PNC Reply Br. at 4. We decline to follow the Tax Court's broad interpretation, for to do so would be to expand the type of costs that must be capitalized so as to drastically limit what might be considered as "ordinary and necessary" expenses. We conclude, therefore, that the loan origination expenses were ordinary expenses and that they did not "create or enhance a separate and distinct asset" within the meaning of Lincoln Savings.

A line of federal appellate opinions subsequent to Lincoln Savings, involving factual scenarios not that different from the one before us, supports our finding that Lincoln Savings does not compel a conclusion that FNPC's and UFB's costs should be capitalized. These cases, from the Fourth, Eighth and Tenth Circuit Courts of Appeals, address the deductibility of costs incurred in connection with credit card issuance. In Iowa-Des Moines National Bank v. Commissioner, 592 F.2d 433 (8th Cir. 1979), the Eighth Circuit Court of Appeals found that payments by banks to third parties who provided the banks with credit information on prospective credit card customers were deductible expenses under S 162. See id. at 436. The court found that "[p]erhaps the most significant factor is that the payments were for a service (credit screening) that could have been performed by personnel employed by the [banks]." Id. Because "[c]redit screening is a necessary and ordinary part of the banking business . . . not a capital expenditure," the Eighth Circuit Court of Appeals found that fees paid to third parties for credit screening were deductible. Id. The Iowa-Des Moines court seemed to assume that such expenditures would a fortiori be deductible if the bank's personnel were to perform the credit screens themselves.13 The Fourth and Tenth Circuit Courts of Appeals have reached similar conclusions. In First National Bank of South Carolina v. United States, 558 F.2d 721 (4th Cir. 1977) (per curiam), the Fourth Circuit Court of Appeals permitted the taxpayer bank to deduct start-up assessments paid to a nonprofit association formed to enable banks to combine their efforts in entering the credit card field. See id. at 723. The association was charged with centralizing billing and recordkeeping for the banks. The Fourth Circuit Court of Appeals characterized the credit card accounts that were the focus of this activity as being a type of loan. See id. at 722. In Colorado Springs National Bank v. United States, 505 F.2d 1185 (10th Cir. 1974), the Tenth Circuit Court of Appeals allowed the deduction of pre-operation expenditures for a nonprofit credit card related association that would cover expenses such as computer costs, advertising, credit screening, and clerical services. See id. at 1193. The Colorado Springs court found that these expenses did not "create or enhance . . . a separate and distinct additional asset," reasoning as follows:

The start-up expenditures here challenged did not create a property interest. They produced nothing corporeal or salable. They are recurring. At the most they introduced a more efficient method of conducting an old business. . . .

. . . [T]he use of bank credit cards in consumer transactions is a normal part of the banking business. The challenged expenditures were for the continuation of an existing business and for the preservation and improvement of existing income. Hence, they were ordinary expenses.

Id. at 1192-93.

In the case before us, the Tax Court distinguished these "credit card" cases, stating that they were inapposite to our fact pattern because in those cases, no "separate and distinct asset" was created, while in PNC's case, such an asset was created. As we have discussed above, we do not agree that the loan origination expenditures created distinct assets, any more so than did the expenditures incurred in entering into the credit card business. In fact, we find that PNC's situation presents an even stronger case for deductibility than do the credit card cases. In the credit card cases, the taxpayers were starting up new programs within their businesses, or at the very least, new methods of performing old tasks. In contrast, FNPC and UFB incurred the challenged costs in their routine selling and marketing of normal loans in the traditional ways that banks have been using for many decades.14 Thus, FNPC's and UFB's costs bear far more of the indicia of "ordinariness," and fewer of the indicia of "creating" something, than do the start-up costs described in the credit card cases.

The remaining question, then, is whether either the Financial Accounting Standards Board's adoption of SFAS 91 or the Supreme Court's pronouncement on deductibility in INDOPCO, both of which developments occurred after the decisions in Lincoln Savings and the credit card cases, would alter the calculus of deductibility versus capitalization in PNC's case. We conclude that the existence of SFAS 91 has little, if any, bearing on the appropriate tax analysis, and that the Supreme Court's decision in INDOPCO, while clearly relevant, does not change the result in the case at bar.

The Supreme Court has held that financial accounting standards such as SFAS 91 do not dictate tax treatment of income and expenditures. See Thor Power Tool Co. v. Commissioner, 439 U.S. 522, 542-43, 544 (1979) (discussing the "vastly different objectives that financial and tax accounting have" and stating that "[g]iven this diversity, even contrariety, of objectives, any presumptive equivalency between tax and financial accounting would be unacceptable" and would "create insurmountable difficulties of tax administration"). The IRS concedes that "financial accounting rules are not controlling for federal tax purposes," IRS Br. at 28 n.4 (citing PNC , 110 T.C. 364 n.15), and states that "[n]either the Commissioner's deficiency determination nor the Tax Court decision was based on the provisions of SFAS 91," id. Although SFAS 91 may have served as a catalyst for the IRS's desire to seek capitalization of the costs at issue here, and may have been considered by the IRS in determining where to draw the line between current-year and deferred costs, see supra note 3, the IRS disavows any argument that the financial accounting standards should dictate tax treatment, see IRS Br. at 27-28 & n.4. Further, as with the financial accounting standards at issue in Thor Power, it is clear that the reasons for SFAS 91's requirement that loan origination costs be deferred are reasons wholly specific to the realm of financial accounting,15 and thus those financial accounting standards do not affect our tax analysis.16

Nor do we view the Supreme Court's decision in INDOPCO as requiring a different result regarding the deductibility of the banks' costs. In INDOPCO, the Supreme Court required capitalization of the expenditures incurred by the target corporation during a planned friendly takeover by another company. See INDOPCO, 503 U.S. at 90. The Supreme Court was careful to emphasize in INDOPCO, as it had in Lincoln Savings, that the capitalization versus deductibility inquiry was heavily fact-based. See id. at 86 (citing Welch v. Helvering, 290 U.S. 111, 114 (1933) and Deputy v. du Pont, 308 U.S. 488, 496 (1940)). The Supreme Court in INDOPCO downplayed the importance of the "creation of a separate and distinct asset" described in Lincoln Savings, clarifying that it was not an exclusive test:

Lincoln Savings stands for the simple proposition that a taxpayer's expenditure that "serves to create or enhance . . . a separate and distinct" asset should be capitalized under S 263. It by no means follows, however, that only expenditures that create or enhance separate and distinct assets are to be capitalized under S 263.

INDOPCO, 503 U.S. at 86-87. The Court reasoned that, while in the Lincoln Savings setting the Court had seemed to attach limited significance to the concept of "benefit," see INDOPCO, 503 U.S. at 87 (quoting Lincoln Savings, 403 U.S. at 354), in the merger situation presented in INDOPCO a "future benefit" analysis was relevant and appropriate since the "resource-related benefits" to be reaped from the merger were of considerable importance, INDOPCO , 503 U.S. at 88. In the INDOPCO context of a friendly takeover, the Court found that one key inquiry was whether the money that the target corporation had spent on takeover-related expenditures was spent primarily for a "future benefit" extending beyond the tax year, rather than for the needs of current income production. The Court stated:

Although the mere presence of an incidental future benefit -"some future aspect" -may not warrant capitalization, a taxpayer's realization of benefits beyond the year in which the expenditure is incurred is undeniably important in determining whether the appropriate tax treatment is immediate deduction or capitalization.

Id. at 87.

As was recognized in several of the "credit card" cases discussed above, these circumstances are simply not presented by a bank's credit-issuing activities. The Eighth Circuit Court of Appeals in Iowa-Des Moines, anticipating the "future benefit" concerns later stated in INDOPCO, emphasized the "short useful life" of credit information as a reason for deductibility. Iowa-Des Moines, 592 F.2d at 436. The Iowa court stated that the prospective future benefit that could accrue beyond the taxable year as a result of credit screening was "very slight," and thus capitalization was "not easily supported." Id. In National Starch, the decision that the Supreme Court affirmed in INDOPCO, we found that these credit card cases contained the seed of the "future benefit" analysis, citing these cases as evidence that several Courts of Appeals "look[ed] to whether an ensuing benefit was created to determine whether the expense was ordinary and necessary," National Starch, 918 F.2d at 431, and that these courts found that future benefit was not substantial in situations similar to the case at bar. See id. (citing Iowa-Des Moines and Colorado Springs). We conclude that the credit card cases not only continue to have vitality after INDOPCO, but in fact anticipated some of the concerns addressed by INDOPCO.

We also conclude that the Tax Court erred in its interpretation of the "future benefit" analysis by relying on the fact that the loan itself was usually of several years' duration and by reasoning that the loan origination costs were, thus, essentially directed at future benefit. The Tax Court stated: "While the useful life of a credit report and other financial data may be of short duration, the useful life of the asset they serve to create is not." PNC, 110 T.C. 371. However, that analysis depends on the Tax Court's earlier assumption that the loan origination expenses actually created a "separate and distinct asset." Stripped of this assumption, the Tax Court's analysis is not supportable.17

In addition, we must remember that the "future benefit" analysis adopted in INDOPCO is not meant as a talismanic, bright-line test. See A.E. Staley Mfg. Co. v. Commissioner, 119 F.3d 482, 489 (7th Cir. 1997) ("[T]he Court did not purport to be creating a talismanic test that an expenditure must be capitalized if it creates some future benefit."). Rather, the INDOPCO analysis demonstrates the contextual, case-by-case approach to determining whether an expenditure better fits under the "ordinary and necessary" language of section 162(a) or the "permanent improvements or betterments" language of 263(a). We conclude that the loan origination expenses incurred by UFB and FNPC have the characteristics of the former, rather than the latter, statutory language.

As described above, the loan marketing activities at issue here lie at the very core of the banks' recurring, routine day-to-day business. The Commissioner has not been able to articulate a principled reason why these normal costs of doing business must be capitalized, while other ordinary banking costs need not be. Instead, the Commissioner relies on the line drawn by SFAS 91, a standard whose rationale we conclude is far removed from the concerns of the tax system. See, e.g., IRS Br. at 27 ("It should be noted . . . that SFAS 91 itself, which the banks have followed, expressly distinguishes direct loan origination costs, which must be deferred, from all other loan-related costs, such as advertising, soliciting potential borrowers, and servicing existing loans, which may be currently deducted for financial accounting purposes."); Tr. of Oral Argument at 27 (statement by IRS's counsel that "[w]here we draw the line is -actually the Financial Accounting Standards Board made it very easy for us."). Similarly, the Tax Court, while professing not to find the financial standards dispositive, see PNC, 110 T.C. 364 n.15, id. at 368 n.18, used SFAS 91 as the sole source of an explanation as to why these loan origination costs, but not other costs associated with the banks' lending business, must be capitalized, see id. at 368-69. We remain unconvinced that the line drawn by the FASB in SFAS 91 has any relevance here for tax purposes.

The Supreme Court has noted that "capitalization prevents the distortion of income that would otherwise occur if depreciation properly allocable to asset acquisition were deducted from gross income currently realized." Commissioner v. Idaho Power Co., 418 U.S. 1, 14 (1974). In the case of the costs at issue here, there need be no concern about a distortion of income because of the regularity of these expenses.

Finally, we emphasize that the key to the deductibility inquiry remains the statutory language of sections 162(a) and 263(a). See Erwin N. Griswold, Cases and Materials on Federal Taxation, at 15 (5th ed. 1960) ("There is no use in thinking great thoughts about a tax problem unless the thoughts are firmly based on the controlling statute."). The analyses set forth in INDOPCO and Lincoln Savings provide us with two applications of that statutory language. Like the Supreme Court in INDOPCO and Lincoln Savings, we do not here attempt to define once and for all a bright line between deduction and capitalization that will hold true for all factual situations. We can only heed Justice Cardozo's admonition that we should always keep the facts firmly in view, as well as Dean Griswold's advice that we remain cognizant of the language of the Code. Resorting to that language, we find the case before us today to be much farther from the heartland of the traditional capital expenditure (a "permanent improvement or betterment") than are the scenarios at issue in INDOPCO and Lincoln Savings. We will not mechanistically apply phrases from those precedents in ignorance of the realities of the facts before us. We see no principled distinction between the costs at issue here and other costs incurred as "ordinary expenses" by banks.

For the foregoing reasons, we find that the loan origination expenses are deductible as "ordinary and necessary business expenses" under section 162(a) of the Internal Revenue Code, and are not subject to the capitalization provision of section 263(a). Accordingly, we will reverse the judgment of the Tax Court.