887 F2d 577 San Antonio Savings Association v. Commissioner of Internal Revenue

887 F.2d 577

64 A.F.T.R.2d 89-5695, 58 USLW 2295, 89-2
USTC P 9614

SAN ANTONIO SAVINGS ASSOCIATION and Subsidiaries, Petitioner-Appellee,

No. 88-4717.

United States Court of Appeals,
Fifth Circuit.

Nov. 2, 1989

William F. Nelson, Chief Counsel, I.R.S., Bruce R. Ellisen, Gary R. Allen, Chief, Appellate Section, Richard Farber, Richard Thornburgh, Atty. Gen., Dept. of Justice, Tax Div., William S. Rose, Jr., Asst. Atty. Gen., Washington, D.C., for respondent-appellant.

Richard L. Bacon, Peter J. Valeta, Sonnenschein, Carlin, Nath & Rosenthal, Washington, D.C., for petitioner-appellee.

Appeal from the Decision of Tax Court of the United States.

Before BROWN, WILLIAMS and JOLLY, Circuit Judges.

JERRE S. WILLIAMS, Circuit Judge:

view counter

The Commissioner of Internal Revenue appeals the tax court's grant of summary judgment to San Antonio Savings Association on its petition for redetermination of a deficiency. The tax court held that San Antonio Savings had realized a recognizable loss by exchanging 90% participation interests in first-mortgage loans in a triangular transaction with two other entities. The court held: (1) The transaction involved the exchange of participation interests in materially different first-mortgage loans. Because the mortgages were materially different, the exchange constituted a "realization event" for purposes of 26 U.S.C. Sec. 1001. (2) The transaction did not "lack economic substance" and hence did not fall under the non-recognition provisions of 26 U.S.C. Sec. 165 and Treas.Reg. Sec. 1.165-1(b). The loss was therefore recognized under the Code. We affirm.


This case presents the first review by this Court of the tax treatment to be given to "reciprocal mortgage sales" conducted pursuant to the Federal Home Loan Bank Board's Memorandum R-49.1 In dispute is a deduction of $14,956,898 in losses which San Antonio Savings Association ("SASA") claims as a result of a "reciprocal sale" pursuant to Memorandum R49. The sale consisted of a group of 90% participation interests in residential first-mortgage loans.


SASA is a Texas corporation operating as a savings and loan association. Since the late 1970s there has been a financial crunch in the savings and loan ("S & L") industry. This downturn was due (at least in part) to the fact that many S & Ls' mortgage loan portfolios contained numerous fixed-rate, long-term mortgage loans on residential property which had been issued at interest rates significantly lower than those charged on more recent loans. High market interest rates were experienced in the late 1970s and early 1980s which made many S & Ls' older, low interest loans worth less at the fair market value. SASA was among the many institutions experiencing this difficulty.


The regulatory body of the S & L industry, the Federal Home Loan Bank Board (FHLBB), 12 U.S.C. Sec. 1437 (1982), requires its member institutions to maintain certain liquidity levels for regulatory and financial accounting purposes. While many of the S & Ls wished to sell their low-interest mortgage loans, recognize their accumulated losses, and receive tax refunds for the losses, they could not do so without reporting the losses for the FHLBB's regulatory purposes. The reported losses in many instances would have brought the S & L's liquidity, at least temporarily, below the minimum required by the FHLBB.


Memorandum R-49 was devised by the FHLBB as a solution to this problem. Memorandum R-49 reads:




The purpose of this memorandum is to advise OES staff [the FHLBB's Office of Examination and Supervision] on the proper accounting for reciprocal sales of mortgage loans.

view counter

A loss resulting from a difference between market value and book value in connection with reciprocal sales of substantially identical mortgage loans need not be recorded. Mortgage loans are considered substantially identical only when each of the following criteria is met. The loans involved must:


1. involve single-family residential mortgages,


2. be of similar type (e.g., conventionals for conventionals),


3. have the same stated terms to maturity (e.g., 30 years),


4. have identical stated interest rates,


5. have similar seasoning (i.e., remaining terms to maturity),


6. have aggregate principal amounts within the lesser of 2 1/2% or $100,000 (plus or minus) on both sides of the transaction, with any additional consideration being paid in cash,


7. be sold without recourse,


8. have similar fair market values,


9. have similar loan-to-value ratios at the time of the reciprocal sale, and


10. have all security properties for both sidesof the transaction in the same state.




If a reciprocal sale does not meet all of the above criteria, the institution must record losses resulting from the sale.


Memorandum R-49, issued by the Director of the FHLBB Office of Examinations and Supervision (OES) to OES staff, June 27, 1980.


Thus, the admitted objective of Memorandum R-49 was to allow S & Ls to engage in transactions which would allow them to realize their losses on mortgage loans for federal income tax purposes, but which would not be treated as giving rise to losses for regulatory accounting purposes. A memorandum from the Director of the OES to an Executive Staff Director of the FHLBB explained the purpose of the ten criteria set forth in Memorandum R-49.


Our objective ... was to structure a transaction which was as close as possible to the IRS "materially different" definition which would still not change the economic position of the association after it engaged in the swap. It was and remains our opinion that Memorandum R-49 represents a transction which is on a fine line between "substantially identical" and "materially different."


These criteria represented our attempt to maintain the association's position with respect to three types of risks in a loan portfolio. These risks relate to credit (collectibility), rate (future earnings potential), and repayment (extent of principal repayments and prepayments). In our opinion, a change in any of these risks would change the economic factors underlying an association's loan portfolio and, as a result, require recording the resulting gain or loss.


On September 30, 1980, SASA and two other savings institutions, Farm and Home Savings Association ("Farm and Home") and Dallas Federal Savings and Loan Association ("Dallas Federal"), engaged in a three-cornered transaction designed to satisfy the requirements of Memorandum R-49. The agreements on each side of the triangular transaction were consummated by reciprocal conveyances of participation interests in first mortgage loans and simultaneous wire transfers of cash. SASA transferred to Farm and Home 90% participation interests in each of approximately 1,808 conventional first-mortgage loans owned by SASA. On the same date, SASA received a wire transfer of cash from Farm and Home to equalize the value. All of the transferred participation interests were in mortgage loans secured by residential properties located within the metropolitan area of San Antonio.


In the second leg of the transaction SASA acquired 90% participation interests in each of approximately 1,834 conventional first-mortgage loans owned by Dallas Federal. SASA transmitted cash to Dallas Federal to equalize the value. Residential properties located within the Dallas metropolitan area secured all the mortgage participation interests. On November 15, 1980, the number of participation interests acquired by SASA from Dallas Federal was slightly reduced so SASA's net payment for the participation interests was also reduced.


On the same date Farm and Home transferred to Dallas Federal 90% participation interests in another set of conventional mortgages completing the triangular transaction. The cash payment from Dallas Federal to Farm and Home and the remaining principal balances on those mortgages were roughly the same as the principal balances and cash amounts involved in the two other participation interest transfers.


The difference between the face value of the participation interests in the mortgages given by SASA to Farm and Home and the amount received for them was $14,956,898. This amount then formed the basis for SASA's claimed deduction. The IRS however did not credit this deduction and sent SASA a notice of deficiency. SASA then appealed to the tax court for a redetermination of deficiency and later moved for summary judgment on the issue.


$14,956,898 appears to be the actual reduction in value experienced by SASA on the 90% participation interests. In other words, if SASA had sold these participation interests in the secondary market for mortgages, $14,956,898 is the amount of loss it would have realized. The IRS does not claim that SASA did not experience a genuine depreciation on these participation interests in the amount of the claimed deduction. The IRS instead argues that SASA did not "realize" its loss through the R-49 transaction, because SASA merely swapped mortgages that were not materially different, leaving itself in the same economic position as prior to the transaction. Alternatively, the IRS argues that 26 U.S.C. Sec. 165 precludes "recognition" of the losses claimed because the transaction "lacked economic substance." The tax court rejected both these arguments, granted summary judgment for SASA, and entered a decision that no deficiencies were due. San Antonio Savings Association v. Commissioner, 55 T.C.M. (CCH) 813 (1988).


This Court has 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." 26 U.S.C. Sec. 7482 (1982). We therefore examine this decision as we do other summary judgment decisions. Because the dispute concerns findings of law, we review on a de novo standard.

I. The Material Difference Requirement


The IRS contends that before gain or loss can be realized on the exchange of property, the items involved must be "materially different.". Hence the IRS argues that no loss was realized by SASA in the R-49 transaction because the mortgages exchanged were not "materially different." The IRS depends heavily in this argument on Treasury Regulation Sec. 1.1001-1(a), associated with 26 U.S.C. Sec. 1001, which states:

Computation of gain or loss


(a) General Rule. Except as otherwise provided in subtitle A of the Code, the gain or loss realized from the conversion of property into cash, or from the exchange of property for other property differing materially either in kind or in extent, is treated as income or as loss sustained.


Treas.Reg. Sec. 1.1001-1(a). SASA takes the position that there is no requirement that the items being exchanged be materially different before such an exchange results in realization.


Before examining the merits of this argument, a conclusion must be drawn that the transaction is an "exchange". While the tax court below never explicitly ruled on whether the transaction was an exchange or a sale, it continually characterized the transaction as an exchange throughout its memorandum opinion. Furthermore, the court held that for summary judgment purposes, when the facts of the case were construed most favorably for the IRS, they were essentially the same as those found in Cottage Savings Association v. Commissioner, 90 T.C. 372 (1988). SASA, 55 T.C.M. at 814. In Cottage Savings, a case involving a similar R-49 transaction with four parties, the tax court held that the four-cornered transaction could not be regarded as a set of separate sales between the contracting parties. Each purchase and sale was interdependent, each party essentially using the money from its sale to finance its purchase. The court therefore held that while the sale agreements did not expressly make the purchases interdependent, the transaction viewed as a whole was a four-cornered exchange or swap. "Thus, we have adopted, for purposes of our opinion, respondent's characterization of the transactions as exchanges." Cottage Savings, 90 T.C. at 387. We conclude that the transaction properly is viewed for purposes of this appeal as a three-cornered exchange.

A. Realization


The key issue under Treas.Reg. Sec. 1.1001-1(a) is the well established concept of "realization" of income. We turn to this critical issue.


"[R]ealization is so basic to the taxing structure of existing law that the general principal is simply not challenged." 1 Bittker, Federal Taxation of Income, Estates and Gifts, p 5.2 (1981). The principle however is one which "emerges from the Internal Revenue Code largely by indirection." Id., p 40.1


The realization requirement takes shape when Sec. 61(a)(3) is combined with Sec. 1001(a).2 Section 61(a) of the Internal Revenue Code ("the Code" or IRC) states that "gross income means all income from whatever source derived, including (but not limited to) ... (3) Gains derived from dealings in property." 26 U.S.C. Sec. 61(a) (1982). The Code does not define the term "dealing" and it does not expressly state that gains and losses are disregarded until an act constituting "dealing" in property takes place. Section 1001(a) provides that:


The gain from the sale or other disposition of property shall be the excess of the amount realized therefrom over the adjusted basis provided in section 1011 for determining gain, and the loss shall be the excess of the adjusted basis provided in such section for determining loss over the amount realized.


26 U.S.C. Sec. 1001(a) (1982).


SASA argues that realization occurs for tax purposes if: (1) a genuine gain or loss occurs, and (2) that gain or loss is "fixed" by a disposition of the property in question through which the owner completely terminates his rights to the property producing the gain or loss. SASA therefore contends that a realization event occurred when each party completely terminated its ownership rights to the property exchanged even though it received back other property similar in type to the property transferred. The IRS insists upon a different concept of realization, arguing that termination of rights to a property is not a realization event if the party immediately replaces the property in question with similar property. In the context of an exchange, this contention requires the property exchanged to be "materially different."


SASA's position is reflected in one of the other two district court decisions before us pursuant to this consolidated appeal, First Federal Sav. and Loan Ass'n of Temple v. United States, 694 F.Supp. 230 (W.D.Tex.1988) ("Temple"). The government's position is adopted in the second district court opinion before us on consolidated appeal, Centennial Sav. Bank FSB v. United States, 682 F.Supp. 1389, 1397-98 (N.D.Tex.1988), as well as two tax court cases, Federal Nat'l Mortgage Ass'n v. Commissioner, 90 T.C. 405, 421-22 (1988) (hereafter "FNMA ") and Cottage Savings, 90 T.C. at 390-94. (1988). Each of these four lower court decisions examines the issue involved in this case, the deductibility of losses on mortgages pursuant to an R-49 transaction between two S & Ls. We therefore examine each of these cases in turn. Our conclusion is that the applicable authority establishes that in order for an exchange to constitute a realization event the items being exchanged must be "materially different" under the tax code.


1) Realization as formal disposition--the Temple interpretation of realization


SASA's interpretation or the requirements of realization is the same as the one used in Temple. It states that there are only two components to realization.


First, there is the economic (or what can be called the "common sense") aspect of realization. That is, "Has there been an actual economic gain or loss?" This is generally put in terms of an economic increase or decrease in a taxpayer's net worth. This aspect has often been summed up by the famous paraphrase from Eisner v. Macomber, "Is the taxpayer, in fact, richer or poorer?" Second, there is the practical accounting aspect of realization. Namely, that for accounting and valuation purposes the increase or decrease in a taxpayer's net worth (i.e. the value of his assets) will not be accounted for ("realized" as a practical matter) until there is some "event that freezes or fixes the gain with sufficient certainty so that it is proper to tax it." Bittker, supra at p 5.1 (quoting Lowndes, Current Conception of Taxable Income, 25 Ohio St.L.J. 151, 173, (1964)); see also Helvering v. Horst, 311 U.S. 112, 115, 61 S.Ct. 144, 146, 85 L.Ed. 75 (1940) (Holding that as a rule of administrative convenience, "realization" for tax purposes must be connected to a taxable event).


Temple, 694 F.Supp. at 239. Temple concludes that the items swapped in an exchange do not have to differ materially from each other before a realization event occurs. Id. at 239-42. Temple did not offer any specific authority, however, for its conclusion that realization is only a two-step process, not requiring the examination of the items being exchanged. Bittker and Horst both state that a realization event is required, but neither states that the realization event requires only termination of legal rights to the property earning the gain or loss. 1 Bittker, supra, at p 5.1; Horst, 311 U.S. at 115, 61 S.Ct. at 146.


Temple holds that the government's position is really better characterized as one claiming that the R-49 transaction should not be recognized. In Temple's view realization is only concerned with fixing gain or loss at a given time, whereas the non-recognition sections of the Code are concerned with analyzing the substance of a transaction. Temple, 694 F.Supp. at 242. A practical problem with this position is that it places no limits on realization through meaningless exchanges. Realization of gain or loss would occur if farmer A exchanged 1,000 bushels of Kansas spring wheat with farmer B for 1,000 bushels of Kansas spring wheat or if individual C exchanged his 1969 red Ferrari with individual D's 1969 red Ferrari (of the same model), both of which had been garaged and never driven. As SASA acknowledges, this doctrine would lead to deduction of the losses sustained on these transactions absent a specific Code non-recognition provision.


Furthermore, the case authority establishing the realization requirement and later applying it to exchanges does not hold that formal termination of a property right alone is sufficient to establish a realization event. The early corporate reorganization/stock swap cases, the bond/debt modification cases, and the substance-over-form cases all support the material difference requirement. The other R-49 cases which held that material difference was required in an exchange for realization to occur each relied on one of these different lines of authority. Hence we examine each R-49 case and its underlying authority separately.


2) Centennial and the material difference requirement


Centennial analyzed an R-49 transaction similar to the one at issue and concluded that a material difference between items exchanged is required before realization can occur. Centennial based this view on three older Supreme Court cases beginning with the premier case on realization in exchanges Eisner v. Macomber, 252 U.S. 189, 40 S.Ct. 189, 64 L.Ed. 521 (1920). See also Marr v. United States, 268 U.S. 536, 45 S.Ct. 575, 69 L.Ed. 1079 (1925); Weiss v. Stearn, 265 U.S. 242, 44 S.Ct. 490, 68 L.Ed. 1001 (1924). From these early realization cases emerges the principle that the receipt of something materially different ("a thing really different") from that which the taxpayer had previously is necessary for an exchange to be considered a realization event.


The parties in the cases before us rely heavily on Eisner v. Macomber and the other early cases. Eisner v. Macomber created the well established principle that the typical stock split does not create income tax liability. Recharacterization of these rights did not trigger realization.3


The case of Weiss v. Stearn further elaborates what constitutes a realization event. In Weiss, the assets of the original corporation were transferred to a newly formed corporation within the same state. Money was transferred from new investors back to the old shareholders and at the conclusion of the transaction the old shareholders and the new investors each had 50% of the stock in the new corporation. The result of the transaction was the same as if the old shareholders had sold half their shares, and they paid tax on that half but not on the entire transaction. The Supreme Court stated:We cannot conclude that mere change for purposes of reorganization in the technical ownership of an enterprise, under circumstances like those here disclosed, followed by issuance of new certificates constitutes gain separated from the original capital interest. Something more is necessary--something which gives the stockholder a thing really different from what he theretofore had.


Weiss, 265 U.S. at 254, 44 S.Ct. at 491-92 (emphasis added). Under Weiss, a formal change in ownership is not enough to trigger realization. In order for realization to occur, the party to an exchange must receive something really different than he previously had and the old stockholders did as to one-half of their investment.


Finally the third of the seminal cases considered by the parties, Marr, showed that the requirement of "something really different" depends on the circumstances, and does not require extensive economic analysis. In Marr a corporate reorganization took place in which stockholders of the General Motors Corporation of New Jersey were allowed to trade their stock for new stock in the General Motors Corporation of Delaware. The assets of the New Jersey corporation were all transferred to the Delaware corporation, and the New Jersey corporation was dissolved. Marr received 451 shares of preferred stock and 2,125 shares of the common stock in the Delaware corporation, worth considerably more than the purchase price of the 339 shares of preferred and 425 shares of common stock in the New Jersey corporation which he exchanged. Marr claimed that the principles of Weiss applied to his case, and that he should not have been held to have experienced a realization event in this stock swap. The Supreme Court, however, held that the two corporations were essentially different because they were controlled by different laws and also that the shares differed in their nature. Realization was found to have occurred.


SASA and the other S & Ls in this consolidated appeal maintain that the Centennial court misinterpreted these three early Supreme Court taxation cases, and argue that three other cases from the Eisner v. Macomber era show that formal disposition of property is all that is required for a realization event to occur. Those three cases are Cullinan v. Walker, 262 U.S. 134, 43 S.Ct. 495, 67 L.Ed. 906 (1923), Rockefeller v. United States, 257 U.S. 176, 42 S.Ct. 68, 66 L.Ed. 186 (1921), and United States v. Phellis, 257 U.S. 156, 42 S.Ct. 63, 66 L.Ed. 180 (1921).


The facts of Phellis parallel those of the transaction in Marr. A stockholder in a New Jersey corporation received a stock dividend of two shares in a Delaware corporation for every one share he held in the New Jersey corporation. The assets of the New Jersey corporation were transferred to the Delaware corporation. The New Jersey corporation continued as a going concern but, except for the "redemption of its outstanding bonds, the exchange of debenture stock for its preferred stock, and a holding of debenture stock to an amount equivalent to its own outstanding common and the collection and disposition of dividends thereon, it has done no business." Phellis, 257 U.S. at 167, 42 S.Ct. at 65.


The Supreme Court held that the transfer of assets to the new company and of stock from it to the stockholders of the old company "changed the former situation materially." Id. at 170, 42 S.Ct. at 66.


[W]e cannot regard the new company as virtually identical with the old, but must treat it as a substantial corporate body with its own separate identity, and its stockholders as having property rights and interests materially different from those incident to ownership of stock in the old company.


Id. at 173, 42 S.Ct. at 67. Distribution of the assets of the new company to the shareholders of the old company therefore constituted a realization event. Phellis clearly implies that if the exchange of stock resulted in ownership in entities which were "substantially identical" then a realization event would not occur.


In the Rockefeller and Cullinan cases, the Court held that the facts were essentially indistinguishable from those presented in Phellis. Both Rockefeller and Cullinan involved corporate reorganizations in which a corporation was split into two or three new entities, but in which the original stockholder received a percentage of the stock of the new entities identical to his percentage of ownership in the old corporation.


Phellis, Rockefeller, and Cullinan do not establish that an exchange of items which differ from each other only formally nevertheless constitutes a realization event.

3) FNMA and the debt modification cases

The tax court in FNMA held that


[i]n determining whether property received by a taxpayer in an exchange differs materially from the property exchanged so that the exchange results in realized losses and gains, courts compare the two properties to determine whether the taxpayer has acquired "a thing really different from what he theretofore had."


FNMA, 90 T.C. at 422. In stating this rule the tax court in FNMA relied on two cases involving the exchange of original municipal bonds for refunding bonds. Girard Trust Co. v. United States, 166 F.2d 773, 774 (3rd Cir.1948); Emery v. Commissioner, 166 F.2d 27 (2nd Cir.1948). In each of these cases the courts concluded that the refunding bonds differed materially from the bonds for which they were exchanged and thus a recognizable (and hence realized) gain had occurred.4


In several other bond/debt modification cases, the courts have held that the refunding bonds were not different in kind from the original bond, and hence no realization event occurred. Mutual Loan and Savings Co. v. Commissioner, 184 F.2d 161, 165 (5th Cir.1950); C.M. Hall Lamp Co. v. United States, 97 F.Supp. 481, 486-87 (E.D.Mich.1951); West Missouri Power Co. v. Commissioner, 18 T.C. 105, 111 (1952).


SASA argues that these debt modification cases are not good authority for the material difference requirement, because these cases involved "constructive exchanges" rather than genuine exchanges. Several of these cases however did involve actual exchanges of the old bonds for the new bonds. See Mutual Savings, 184 F.2d at 162; Emery v. Commissioner, 166 F.2d at 28; West Missouri Power Company, 18 T.C. at 109-111. Furthermore, SASA never offers an explanation for why a "constructive exchange" should be treated differently from an "actual exchange," other than asserting that because the borrower, debtor, and underlying property were all normally the same in these cases, other factors had to be considered.


Finally, it is argued that both these cases and the earlier corporate reorganization cases deal with "A for A" transactions in which a stock or bondholder trades in his asset for a new type of stock or bond. Hence, it is argued that realization occurs if a taxpayer (1) sells or disposes of his property or (2) exchanges his property for a different interest in the same property. In other words, the material difference requirement only applies if the taxpayer is exchanging his property for a different interest in the same property. This distinction is not the law. In Marr the court relied not only on the difference in the interest received but also on the difference in the corporations. In Phellis, the fact that corporations were not substantially identical was sufficient to create a material difference. Hence the material difference requirement applies to "A for B" transactions as well.


B. Treasury Regulation Sec. 1.1001-1(a) and the requirement of material difference


Contrary to SASA's contention and the Temple opinion, Treasury Regulation Sec. 1.1001-1(a), quoted earlier, supports the government's argument that before an exchange constitutes a realization event a material difference is required between the exchanged items. Thus, the government asserts that a material difference requirement found in the case law is also found in the regulations themselves.


It is true that Treas.Reg. Sec. 1.1001-1(a) does not in its terms create the requirement of material difference. The concept of realization is what creates the material difference requirement. Treas.Reg. Sec. 1.1001-1(a) states that "gain or loss realized ... from the exchange of property differing materially either in kind or extent is treated as income or loss sustained" (emphasis added). The semantic difficulties urged by SASA and the court in Temple are linked to the omission of a specific statement in Treas.Reg. Sec. 1.1001-1(a) conditioning realization upon a material difference between the items exchanged. Because the material difference requirement is already inherent in the concept of realization, however, the "differing materially" language in Treas.Reg. Sec. 1.1001-1(a) may be read as a reference to an established requirement rather than as establishing the requirement.


SASA and the Temple opinion also argue that there could be no "material difference" requirement for realization because it would rob the Code's nonrecognition sections of their purpose. Treasury Regulation Sec. 1.1002-1 states the general rule that gain or loss which is "realized" must be "recognized" under the Code, unless a nonrecognition provision applies. The general principle of narrow construction of exceptions applies to such nonrecognition sections. The regulation then goes on to spell out exceptions in terms of differences in the property exchanged:


(c) Certain exceptions to general rule. Exceptions to the general rule are made, for example, by sections 351(a), 354, 361(a), 371(a)(1), 721, 1031, 1035 and 1036. These sections describe certain specific exchanges of property in which at the time of the exchange particular differences exist between the property parted with and the property acquired, but such differences are more formal than substantial. As to these, the Code provides that such differences shall not be deemed controlling, and that gain or loss shall not be recognized at the time of the exchange. The underlying assumption of these exceptions is that the new property is substantially a continuation of the old investment still unliquidated; and, in the case of reorganizations, that the new enterprise, the new corporate structure, and the new property are substantially continuations of the old still unliquidated.


Treas.Reg. Sec. 1.1002-1 (emphasis added).


The realization requirement of material difference does not undercut these Code sections. It only provides a minimal standard that items exchanged may not be identical, fungible commodities. Such nonrecognition provisions as the wash sales rule (Sec. 1091) and the like kind exchange rule (Sec. 1031) are bright line rules dealing with specific circumstances not necessarily addressed by the material difference requirement of realization. Furthermore, these nonrecognition sections are not coextensive with the material difference requirement. Section 1031, the like kind exchange rule, specifically omits coverage of evidences of indebtedness, such as mortgages, and hence is inapplicable to the present case. Section 1091 also fails to apply other evidences of indebtedness, only covering securities and stocks.


SASA contends that the material difference requirement leads to untenable and administratively impossible economic analyses of the positions of all parties and all things exchanged, and argues that this is illogical because realization is a concept based largely on administrative convenience. Helvering v. Horst, 311 U.S. 112, 116, 61 S.Ct. 144, 147, 85 L.Ed. 75 (1940). We do not read the material difference requirement of realization as demanding extensive analysis of the economic situations of persons involved in the exchange after such an exchange has occurred. The "material difference" inquiry does not examine whether items are "economic substitutes" or not. Such a determination is the realm of the like kind rule, which as noted above is inapplicable here.


We reject the administrative "parade of horribles" proposed by SASA on this point because the material difference requirement of realization is something that has always been required, and is essentially merely an extension of the "form over substance" principle into the area of realization. Thus, we see the prior interpretation of Treas.Reg. Sec. 1.1001-1(a)'s "differing materially" language as a nondetailed reference to the requirements of realization previously well developed in the case law. Since the Code provides no independent definition of realization, but refers to it in Sec. 1001, it is appropriate to employ judicial interpretations of the concept of realization. We conclude that there is a material difference requirement which has been developed by the cases and confirmed by the Treasury Regulations.


II. The Properties Exchanged in the R-49 Transaction are

Materially Different


Of the five lower court cases (other than the one at issue) which have addressed the question of R-49 transactions, the three tax court cases all held that the groups of mortgages exchanged were materially different from each other. Cottage Savings, 90 T.C. at 395; FNMA, 90 T.C. at 422-23; Leader Federal Sav. & Loan, 57 T.C.M. at 849-50. The two district court cases both held that the mortgage packages were not materially different from each other. Temple, 694 F.Supp. at 244-45; Centennial, 682 F.Supp. at 1398-99. We find the rationale behind FNMA and Cottage Savings is correct, and we affirm the tax court's finding in the instant case that the mortgages exchanged in the R-49 transaction were materially different.

A. The Cottage Savings and FNMA analysis


The tax court below relied heavily on the rationale employed in Cottage Savings. SASA, 55 T.C.M. at 815-17. Cottage Savings held that the participation interests in the mortgages exchanged were materially different because the collateral and obligors were different.


Although the loans were similar, there were important differences. Specifically, the loans had different obligors and were secured by different pieces of reality. The subsequent history of payments on the loans ... shows that the transactions were real, not feigned, and that the assets received were not the same as the assets given up. In the instant case (unlike Shoenberg and Horne ), when the smoke cleared away petitioner was left with assets that were different (and performed differently) from what petitioner had at the start. Thus, application of the principle of Horne and Shoenberg to the December 31, 1980, transactions leads us to conclude that the property is materially different, and petitioner is entitled to the claimed deductions.


Cottage Savings, 90 T.C. at 394.


The Cottage Savings court was further guided by Hanlin v. Commissioner, 38 B.T.A. 811, aff'd 108 F.2d 429 (3rd Cir.1939). Hanlin was held to support the taxpayers' position but not to be controlling, because it involved the wash sales provision, Sec. 118 of the Revenue Act of 1932,5 rather than the realization requirement of material difference. The tax court held in Hanlin that bonds of the St. Louis and Wichita Federal Land Banks were not substantially identical to bonds of the Louisville Federal Land Bank, because the bonds were issued by different obligors and the assets underlying the bonds of those obligors were also different. Hanlin, 38 B.T.A. at 819-20.


When affirming the tax court's decision in Hanlin, the Third Circuit downplayed the difference between the bonds' obligors, stating that "[t]hese institutions are all under the supervision of one central authority, formerly the Federal Farm Board, now the Farm Credit Administration.... They are at least secondarily liable on each other's bonds...." Hanlin, 108 F.2d at 430-31. The Third Circuit, however, did uphold the ruling that the bonds were materially different because the collateral security underlying the bond issues was significantly different.


It cannot be denied, however, that the collateral security underlying the bond issues of different Land Banks, is physically distinct. That collateral, consisting of United States bonds and first mortgages, is held in trust by the farm loan registrar for the issuing bank and the prospective holders of its bonds.... The statutes and decisions give no indication that such collateral is pooled indiscriminately behind all the outstanding bonds of all the Land Banks.... Consequently, the credit of the United States being what it is, we must, for example, discriminate, if at all, between first mortgages on farms in the vicinity of Louisville, and on those round about Wichita. Those mortgages are, to be sure, selected according to searching and uniform standards, 12 U.S.C.A. Secs. 723(a), 751-756. But can it be said that there is no substantial variance in agricultural and financial resource when Louisvillites are contrasted with Wichitavians? Though the average sense of obligation may well be the same, even an economist must recognize, by geographical definition, a salient divergence in, say, the type (and marketability) of crops produced--or, perhaps, the likelihood of dust storms. This difference, we think, deprives the bonds of one Land Bank of substantial identity with those of another.


Hanlin, 108 F.2d at 431 (citations omitted).


The IRS argued that the mortgages involved in Cottage Savings were distinguishable from those in Hanlin because the geographical variation in risks to farm property was absent in Cottage Savings, which dealt only with residential real estate. The Cottage Savings court held however that in Hanlin


each bond was secured by a mass of mortgages, so that only a regional disaster could affect the security of the bonds. Thus, in Hanlin, the courts focussed on regional differences. In the instant case, each obligation apparently is secured by one residential property, so that a misfortune affecting one family or one property could affect what happens to that obligation without affecting any of the other obligations.


Cottage Savings, 90 T.C. at 395-96. For this reason, the Cottage Savings court concluded that the taxpayers' position that the mortgages involved in the R-49 transactions were materially different was even stronger than the meritorious claim in Hanlin that the bonds of the federal land banks were substantially different.


The court in FNMA arrived at the same conclusion as in Cottage Savings, holding that different obligors and also different collateral properties each independently constituted a material difference. FNMA, 90 T.C. at 422-23. The FNMA court went on to note that the different rates of foreclosure, prepayment, and delinquency experienced after the transactions confirmed the mortgages' material difference. Id. The FNMA court also held that the overt identity of market value of the two pools of mortgages exchanged was not determinative. While the face amount and maturity of two obligations might be the same, the underlying securities could be radically different. FNMA, at 90 T.C. at 423.


In the instant case, SASA acquired 90% participation interests in residential mortgages in Dallas in exchange for 90% participation interests which it gave up in residential mortgages in San Antonio. In Cottage Savings, most of the properties underlying the mortgages were inside the Cincinnati "beltway," although there apparently were differences in the distribution of the in-beltway properties which were exchanged, and at least some of the properties ranged as far as 120 miles from Cincinnati. The court in Cottage Savings, found that although "diversification by geography and other factors may be reasons for reciprocal sales transactions generally, such factors were not considered in the ... transactions." Cottage Savings, 90 T.C. at 382. Because the ruling by the Court below was made on summary judgment, the facts were required to be construed in the best light possible for the nonmoving party, the IRS. The IRS conceded that if the facts were construed most favorably for it they would be essentially the same as those found in Cottage Savings. SASA, 55 T.C.M. at 814. Hence the geographic distinctness of the two groups of mortgages exchanged between SASA and its trading partners did not play a role in the summary judgment finding that the mortgages were materially different. Nevertheless, because the mortgages had different obligors and were secured by different pieces of reality, the tax court below concluded that they were materially different. SASA, 55 T.C.M. at 816.


We uphold the finding of the lower court that different obligors and collateral result in mortgage participation interests which are materially different.

B. The economic substitute analysis


In contesting the conclusion that the mortgages exchanged by SASA were materially different, the government urges that the R-49 transaction is specifically designed to leave the S & Ls in the same economic position as they were when they began. The participating S & Ls are supposed to receive new mortgages with the same risk of default and delinquency as the old mortgages. Hence, there can be no material difference between the things exchanged. This position may be summed up as follows: "In an R-49 transaction risk does not transfer." Risk consists of the best predication of potential future performance of the mortgage at the time of transfer, and when the mortgages meet the ten requirements of an R-49 transfer the risk associated with each mortgage is identical at the time of that transfer. The IRS therefore argues that because the risks associated with the mortgages involved in the R-49 transfer are indistinguishable, the mortgages involved are "economic substitutes" for each other. See Temple, 694 F.Supp. at 245.


Another way of phrasing this argument is that once the mortgages have been matched according to the ten criteria of R-49, the marketplace cannot distinguish between the loan packages. It cannot be predicted at the time of the exchange which set of mortgages will perform better, the secondary mortgage marketplace therefore assigns them equal value, and they are as fungible a commodity as coffee or wheat. Centennial, 682 F.Supp. at 1399.


The weakness in the government's argument is that the requirement of "material difference" does not go so far. Treas.Reg. Sec. 1.1001-1(a) when referring to (although not establishing) the material difference requirement alludes to the exchanged property as differing "materially in either kind or extent." It does not say "differing materially for economic purposes." The case law which establishes the requirement of material difference as a part of realization simply does not equate "materially different" with "not an economic substitute." The corporate reorganization cases did not rely on an analysis of whether the assets of the new corporation would perform in the same way as the old one.


The determination of whether two items are economic substitutes belongs to a broader inquiry; that of like kind exchanges. Essentially, the IRS is reading a universal like kind exchange rule into the requirement of material difference when it urged that "economic substitutes" cannot be materially different. Instead, the proper standard is that stated in Weiss; an asset given up must be really and truly different from the asset received in order for an exchange to be a realization event. See Weiss v. Stearn, 265 U.S. at 254, 44 S.Ct. at 492 (no gain is recognized unless the taxpayer received "a thing really different from that which he theretofore had").


Allowing the marketplace to determine whether or not two mortgages are materially different is flawed. Just because two mortgages are assigned the same value in the secondary mortgage marketplace does not mean they may not be different in a host of other ways. See FNMA, 90 T.C. at 423.


The IRS contends however that not only were the R-49 mortgage packages of similar value, but when matched with the ten R-49 criteria, the known risk or predicted economic performance associated with the mortgages was the same. If the government's "predicted economic performance" interpretation of the material difference requirement were employed, however, the results would be dire. Exchanges of items which were clearly different would not necessarily constitute realization events. For example, if two taxpayers exchanged two bonds of different corporate entities engaged in entirely different businesses, but which have the same rating by Moody's, the same interest rate, the same cash flow, the same maturity and the same value in the capital market then no change in their "economic status" would have occurred. The transaction, would, nevertheless be real, and the taxpayer should not be able to postpone losses and gains while switching bonds or other assets merely because they are "economic substitutes."


Administration of Sec. 1001 would be considerably complicated if the courts had to define the relevant marketplace and then determine exactly how that market perceived a reciprocal exchange of mortgage loans or any other exchange involving similar assets. The courts would have to determine which sources of information establish the market price, differentiate specific investment factors peculiar to individual assets, and specify what discrete information was or was not available to be perceived and acted upon by participants in that marketplace.


As a corollary to the "economic substitute" argument, both the Temple and Centennial cases stated that the S & Ls involved in their respective cases did not examine the mortgages and their prospects any further than by ensuring that they met the ten criteria of R-49. Temple, 694 F.Supp. at 245; Centennial, 682 F.Supp. at 1399. Indeed, the matching which put together the relevant mortgage packages in both Temple and Centennial was done by computer. This argument assumes that the subjective perceptions (or lack thereof) of the S & Ls involved must be taken into account in order to determine whether a transaction involved materially different items or not. Essentially, the Temple and Centennial courts were stating that because the R-49 transaction at issue was not a horse trade, with each party seeking to gain an advantage through the swap transaction, there was no material difference in the items being swapped. It must be the objective and not subjective factors, however, which determine whether a realization event occurred. Otherwise the Code could not be fairly applied to disparate individuals.


We reject the IRS's interpretation of "materially different," which would rely upon an "economic substitute" requirement and hold that while "material difference" requires more than purely formal differences, an "economic substitute" can still sometimes be materially different.


C. Centennial's "no richer--no poorer" analysis


Realization does not require that a taxpayer must be "richer or poorer" as a result of the exchange itself. Generally items exchanged will be of equivalent value. If the property exchanged in a straight swap was not roughly equal in value, either one party did not get the market rate or the exchange was a sham transaction. In order for realization to occur, a gain or loss must occur which leaves the taxpayer ultimately richer or poorer; the realization event fixes that loss or gain. The realization event itself, however, need not cause the loss or gain. In this case, the mortgages declined in value not because of the exchange, but rather as a result of market depreciation.

The Centennial court urged:


R-49 ... prevented the realization of a tax loss. Centennial was no poorer after the transaction than before. It reflected no loss on its books for regulatory and accounting purposes.... It appears to the Court that by means of a mortgage swap, Centennial hoped to create a loss without any real change in its economic status.


Taxation, however, is concerned with realities rather than appearances, Crenshaw v. United States, 450 F.2d 472, 475 (5th Cir.1971), and no loss can be claimed that is not real. The loss which Centennial claims it suffered as a result of the transaction is the paper loss it had been suffering all along because its old fixed-rate mortgages had declined in value. A decline in valuation without realization, however, is not enough to entitle an institution to deduct a loss.


Centennial, 682 F.Supp. at 1400 (emphasis added).


To the extent Centennial relies on its no richer/no poorer analysis, it is incorrect on the facts of these R-49 cases. The basis for Centennial's finding of no realized loss actually is no more than an assertion that no change in economic status occurred, and the Centennial opinion suffers the flaws of economic substitute analysis set out above.

D. Post transaction performance


The tax court in Cottage Savings and FNMA noted that its holding that the mortgages were materially different was confirmed by the fact that the performance of the mortgages after the exchange differed from each other. Cottage Savings, 90 T.C. at 394; FNMA, 90 T.C. at 422-23. Temple states that the post hoc performance of the mortgages cannot matter because the risk involved with each package of mortgages was the same at the time of the swap. Temple, 694 F.Supp. at 245. This conclusion misses a critical point. It is an assumption inherent in the Hanlin holding that different obligors and different collateral must result in different performance. Real property and individual human beings are in these cases the collateral and obligors. It is elementary in the law that real property, even residential property of the same general value, is unique. So is each person. Because the value of the mortgages depends on the performance of these unique entities (people in paying their obligations and real estate in retaining its value in the market) an exchange of residential mortgages is not like swapping identical fungible commodities. Post hoc performance is not necessary to prove the mortgages' uniqueness, but it is a result confirming it.


E. The weight given to the FHLBB determination of substantial identity


Admittedly at first glance the statement that two items can at the same time be "substantially identical" and "materially different" seems illogical. The critical consideration is, however, that the "substantially identical" language is employed by the FHLBB as an administrative accounting matter to carry out its own regulatory obligations. The "materially different" requirement is employed by a different agency, the IRS, in carrying out its separate regulatory responsibilities. Each of these phrases is a regulatory statement created for and used for a different administrative purpose.


The IRS urges that under Ford Motor Credit Co. v. Milhollin, 444 U.S. 555, 100 S.Ct. 790, 63 L.Ed.2d 22 (1980) the determinations of regulatory agencies are to be granted respect. So they are--but as to both the IRS and the FHLBB. The FHLBB is not in the business of interpreting the Internal Revenue Code. So also the IRS is not supposed to interpret the Federal Home Loan Bank Act. Here the IRS is arguing that deference should be shown in a tax determination case to a ruling by an agency which is not charged with enforcing the tax code. The FHLBB held that mortgages in R-49 exchanges are "substantially identical" for its own accounting. The IRS would have us defer to the FHLBB on the "substantial identity" but not on the "material difference" of the loans, and it would apply the "substantial identity" ruling (an accounting ruling) to the tax standard of "material difference." "[I]t has been consistently held that the accounting requirements of regulatory agencies are not controlling in the application of the revenue laws, which establish their own standards." Bellefontaine Federal Sav. & Loan Ass'n. v. Commissioner, 33 T.C. 808, 811-812 (1960). Hence, the FHLBB's determination that mortgages exchanged in an R-49 transaction are "substantially identical" for its accounting purposes is of no direct relevance when examining whether mortgages under R-49 are "materially different" for purposes of the Internal Revenue Code.


In conclusion, we hold therefore, as did the tax court, in this case and in Cottage Savings and FNMA that the mortgages exchanged are indeed materially different under the IRC.

III. Section 165


Finally, the government argues that SASA cannot claim the losses on the mortgages it exchanged because of Sec. 165 of the IRC and associated regulations. That section states


(a) General Rule--There shall be allowed as a deduction any loss sustained during the taxable year and not compensated for by insurance or otherwise.


26 U.S.C. Sec. 165 (1982). The Treasury Regulation associated with this section states


(b) Nature of loss allowable. To be allowable as a deduction under section 165(a), a loss must be evidenced by closed and completed transactions, fixed by identifiable events, and, except as otherwise provided in section 165(h) and Sec. 1.165-11, relating to disaster losses, actually sustained during the taxable year. Only a bona fide loss is allowable. Substance and not mere form shall govern in determining a deductible loss.


Treas.Reg. Sec. 1.165-1(b).


The tax court rejected this argument by the government. SASA, 55 T.C.M. at 815. The IRS conceded for summary judgment purposes that the facts in this case could be no more favorable to its position than those found in Cottage Savings. Given that acknowledgment, the court held:


We will therefore assume for purposes of petitioner's motion [SASA's summary judgment motion] that the transaction entered into by SASA was motivated solely by tax considerations, that the transaction was closed and completed and changed the flow of economic benefits from the loan participations, and that the measurement of the loss is not separately disputed. The transaction will not be found to lack economic substance in these circumstances.


SASA, 55 T.C.M. at 815.


The tax court in both Cottage Savings and FNMA when ruling on similar arguments based on Sec. 165 concluded that when the taxpayers participated in closed and completed transactions with the actual transfer of economic benefits from the loan participations, Sec. 165 was not applicable. Cottage Savings, 90 T.C. at 401-02; FNMA, 90 T.C. at 425-26.


We uphold the tax court's finding. The government's argument with reference to Sec. 165 is a different facet of its previous argument that the loss sustained by SASA was not economically real. Even if it is conceded (as it was for summary judgment purposes) that there was no purpose for the R-49 transaction other than tax reduction, nevertheless SASA suffered a real economic reduction in the value of the mortgage participation interests it transferred and the economic reality of that loss was fixed by an identifiable event, an exchange of materially different items.


The government in pursuing its argument that losses resulting from transfers lacking in economic substance are not deductible cites four cases as supporting this proposition. Three of those cases involve situations in which there was no genuine, arm's length, completed transaction which disposed of the property. Higgins v. Smith, 308 U.S. 473, 60 S.Ct. 355, 84 L.Ed. 406 (1940) (involving sale by a taxpayer of securities to a corporation wholly owned by the taxpayer); Fender v. United States, 577 F.2d 934 (5th Cir.1978) (involving sale of unrated municipal bonds to a bank over which taxpayers had sufficient influence so that the bonds could be later repurchased); Scully v. United States, 840 F.2d 478 (7th Cir.1988) (involving sale of real estate by trustees of family trust to other family trusts which had the same fiduciaries, beneficiaries, and remainderman). In contrast to these cases, in the R-49 transaction at issue a genuine completed transaction occurred between two legally independent entities which conclusively severed their rights to the items exchanged.


The government also cites Yosha v. Commissioner, 861 F.2d 494 (7th Cir.1988). In that case the Seventh Circuit held that individual investors were not entitled to deductions for losses incurred in trading options, straddles, and hedges on the London metal exchange, because the investors' losses lacked economic substance and did not change the individual's investment positions. The investors in Yosha placed a deposit with a broker with the understanding that they were not purchasing options with genuine market risks but rather tax deductions. The brokers guaranteed the investors that the only consequences of their trading would be tax consequences, and the brokers organized that trading to provide the taxpayers with an ordinary income loss and no capital gains in the first year, and long term capital gains in the second year. The investing taxpayers never faced the possibility either of actual gains or losses on the trading as the gains were retained and the losses absorbed by the brokers. Yosha, 861 F.2d at 500.


Yosha involved an entirely different circumstance from the one presented in the case at hand. Here the S & Ls involved in the R-49 transaction genuinely experienced a decline in the market value of their mortgages. They sought to realize that loss by the disposition of their mortgages in closed transactions. The loss was not created for tax purposes; its existence was entirely independent of the R-49 transaction. We conclude that 165 of the Code and Treas.Reg. 1.165-1(b) do not apply to the facts of this case.



The mortgages which SASA exchanged in the R-49 transaction with Dallas Federal and Farm and Home were materially different. Further, the losses experienced were genuine and the result of market forces independent of the R-49 transaction. We uphold the finding of the tax court allowing SASA the deduction of $14,956,898 under the Internal Revenue Code as loss realized through the exchange of materially different property.




Our research reveals only five cases (aside from the one at issue) which rule on tax treatment Memorandum R-49 transactions, none of them by appellate authority:

(1) First Federal Sav. & Loan Ass'n of Temple v. United States, 694 F.Supp. 230 (W.D.Tex.1988).

(2) Centennial Sav. Bank F.S.B. v. United States, 682 F.Supp. 1389 (N.D.Tex.1988).

(3) Leader Federal Sav. and Loan Ass'n of Memphis v. Commissioner, 57 T.C.M. (CCH) 846 (1989).

(4) Federal National Mortgage Ass'n v. Commissioner, 90 T.C. 405 (1988).

(5) Cottage Savings Ass'n v. Commissioner, 90 T.C. 372 (1988).

The two district court cases of Temple and Centennial are currently before this Court and are decided this day pursuant to a consolidated appeal with the instant case. Federal National Mortgage Ass'n is currently on appeal in the D.C. Circuit; and Cottage Savings is on appeal before the Sixth Circuit. While this opinion rules only on SASA's appeal, we refer throughout to the Centennial and Temple opinions because of frequent reference to these cases by both parties and because they represent differing positions on the issues involved in this case.


Unless otherwise indicated, all section (Sec.) references are to the Internal Revenue Code, 26 U.S.C. Sec. 1 et. seq as amended and in effect during the years at issue


The Centennial court focused on language in Eisner v. Macomber stating that because the stockholder holds the same percentage of the corporation after the issuance of stock dividends as she did prior to the issuance, "the corporation is no poorer and the stockholder is no richer than they were before." 252 U.S. at 203, 40 S.Ct. at 191-92. Centennial then concluded that Eisner v. Macomber endorsed an analysis of the change in the taxpayer's economic status resulting from the transaction to determine whether a realization event had occurred. See Section II, infra. Later on in the Eisner v. Macomber opinion, however, the Supreme Court stated that the stock dividend indicated "that the shareholder is the richer because of an increase in [her] capital, [but] at the same time shows [she] has not realized or received any income in the transaction." Eisner v. Macomber, 252 U.S. at 212, 40 S.Ct. at 195


Although Emery applied a predecessor of Treas.Reg. Sec. 1.1001-1(a) in determining whether material difference between the bonds existed, it relied on both the statute and the regulation as establishing the requirement of a material difference. Emery, 166 F.2d at 29


Section 1091 of the current Code is the successor to Sec. 118 of the Revenue Act of 1932, and has been conceded by the IRS not to be applicable to this case