Opinion
TC 4262.
Decision for Plaintiff rendered August 7, 2003.
Plaintiff appealed the assessed value of a federally subsidized low-income housing project constructed under section 515 of the Housing Act of 1949, 42 USC § 1485 (1990). The court held that to properly account for the federal government interest in low-income housing property, the government interest must be valued and then that value subtracted from the overall, unrestricted value of the property, thereby arriving at the value of the taxable private interest in the property.
Property Taxation — Valuation — Public Interest
1. Under the public interest rule, the taxable value of property must be reduced to the extent that federal low-income housing restrictions diminish the value of the property.
Property Taxation — Valuation — Government Restrictions
2. ORS 308.205(2)(d) does not change or modify the conditions of an income approach to value and government restrictions should be reflected in an income approach if those restrictions would not be removed by sale of the property.
Property Taxation — Valuation
3. A property interest may be valued by capitalization of payments made for the interest in the property.
Property Taxation — Valuation — Presumed Equivalency of Value
4. Where there is an arm's-length exchange and one side of the exchange can be valued with reasonable certainty, the other item in the exchange, although difficult to independently value, is assumed to have a value equivalent to the value of the item for which value can be calculated with reasonable certainty.
Property Taxation — Valuation
5. When valuing property using a band-of-investment method to establish a capitalization rate, it is necessary to establish what market participants would expect as to debt return and equity return, as of the date of assessment.
Trial was held April 15 through 19, and April 23 through 24, 2003, in the courtroom of the Oregon Tax Court, Salem.
Christopher Robinson, and W. Scott Phinney, Lake Oswego, argued the cause for Plaintiff (taxpayer).
Rochelle Nedeau and Marilyn Harbur, Assistant Attorneys General, Department of Justice, Salem, argued the cause for Defendant (the department).
I. FACTS
Plaintiff (taxpayer) owns a 24-unit apartment project built in 1985-86. The property is located in Wilsonville, Oregon. The property was constructed with financing obtained under section 515 of the Housing Act of 1949 (515 Program or 515 Project). See 42 USC § 1485 (1990).
Under the 515 Program, as it existed during the tax years at issue, 1992-93, 1993-94, and 1994-95, developers obtained financing for construction of apartment projects in rural areas on the following basic terms:
(1) The project owner contributed 5 percent of the project cost.
(2) The federal government either provided the remaining construction cost or guaranteed a loan from a private source in the same amount.
(3) The project loan was amortized evenly over a term of 50 years and was also secured by a mortgage on the project. The lender had no recourse against the project owner for any failure to make the payments required by the mortgage.
(4) The loan could not be prepaid before 20 years.
(5) The agreement between the project owner and the federal government (the regulatory agreement) imposed on the project owner numerous compliance requirements relating to who could be a tenant in the project, maintenance of the project, reporting on project operations, and other matters.
(6) Under the regulatory agreement, the project owner agreed to limit the rent charged for occupancy to levels that were often below the level of the unregulated rental-housing market.
(7) If the project owner complied with the regulatory agreement, the federal government paid the project lender the difference between the fully amortizing loan payment computed at the rate stated in the note (in this case 10.625 percent), and what the payment amount would have been if the loan interest rate were 1 percent. Cash flow of the project was used to make all other debt payments, which was the amount of principal and interest due on a loan amortizing at 1 percent. However, if the project owner failed to comply with the regulatory agreement, the government payments (interest subsidy) would end.
(8) Annually, the project owner could receive an equity dividend payment of no more than 8 percent of the amount contributed to the project by the owner. All other cash flow of the project was applied to the debt-service obligation of the project owner, and required maintenance or operating costs, including reserves. Any reserves were and always remained the property of the federal government.
(9) If cash flow was inadequate to fund operating costs and required reserves, the project owner might not receive the equity dividend for the year. That deficiency could be remedied only if, within the following year, the project operations produced surpluses adequate to fund the payment.
(10) So long as the 515 Program mortgage and regulatory agreement were in effect, the project owner could not sell or refinance the project.
(11) In no event could the project owner receive more than the agreed equity dividend on the project.
In the case of the property at issue here, construction financing was provided by First Interstate Bank under a note and mortgage with an interest rate of 10.625 percent. It appears that the federal government committed to a "takeout" loan and did, in fact, become the lender on the project upon completion of construction with a right to receive 10.625 percent interest.
Federally subsidized housing programs have existed for some time. Those programs provide a public benefit through making affordable housing available to low-income persons. In the late 1970s and the 1980s, the federal government became concerned with the fact that the regulatory agreements on many projects were soon to expire because of mortgage prepayment. Legislation was adopted seeking to extend the period of rent restriction and otherwise protect the inventory of low-cost housing. Substantial litigation has occurred regarding the ability of owners to prepay federal program loans and, as of the years in question, confusion undoubtedly existed among owners and potential buyers of 515 Projects as to whether or when any given project can be operated free of program-rent restrictions. For these or other reasons, there was no immediate market for the interest of a project owner in 515 Projects during the years at issue.
A federally subsidized low-income housing project will be referred to throughout this Opinion alternatively as, a project, a program, a federal project, or a government program.
The Emergency Low Income Housing Preservation Act of 1987 (ELIHPA), 101 Stat 1877, as amended, 42 USC § 1472(c) (1994 ed and Supp V), placed permanent restrictions upon prepayment of low-interest mortgage loans issued by the Farmers' Home Administration (FmHA) prior to 1979. For loans made between 1979 and 1989, prepayment may be made only after following a complex process. See Sheldon P. Winkelman, Low-Income Housing Preservation and Resident Homeownership Act of 1990, 73 Mich B J 1160 (1994); Richard Michael Price and Randall Kelly, Prepayment of Section 515 Affordable Housing Mortgage Loans 14-APR Prob Prop 15 (2000).
See Winkelman, 73 Mich B J at n 10.
For the tax years at issue, the positions of the parties as to the real market value (RMV) of the property are as follows:
Asserted in a motion to amend its Complaint to conform to the evidence at trial (that motion was granted April 29, 2003).
1992 $225,000 $700,000 1993 $225,000 $728,000 1994 $225,000 $757,120
II. REVIEW
The question of the proper methodology for valuation of federally subsidized low-income housing projects has occupied this court and the Oregon Supreme court for more than 10 years and is again presented in this case. A brief review of what has already occurred is helpful.
A. Bayridge
In Bayridge Assoc. Ltd. Partnership v. Dept. of Rev., 13 OTR 24 (1994) ( Bayridge I), aff'd, 321 Or 21, 892 P2d 1002 (1995) ( Bayridge II), this court held that the restrictions to which an owner voluntarily submits in a federal project are "governmental restrictions" within the meaning of ORS 308.205(2)(d). This court also concluded that federal income tax credits associated with the federal project in that case would not be taken into account as income in arriving at a value for the project using the capitalized income indicator of value. The basis of that conclusion was that the economic value of the tax credits, received by the initial developer of a federal project, was not transferable to a subsequent owner and would not be paid for in a market sale of the project.
For ease of reference, Bayridge I will refer to the Tax Court opinion, 13 OTR 24 (1994). Bayridge II will refer to the Oregon Supreme Court opinion, 321 Or 21, 892 P2d 1002 (1995). All references to Bayridge refer collectively to the opinions of both courts.
Unless otherwise noted, all references to the Oregon Revised Statutes (ORS) and Oregon Administrative Rules (OAR) are to 1991.
ORS 308.205(2)(d) provides:
"If the property is subject to governmental restriction as to use on the assessment date under applicable law or regulation, real market value shall not be based upon sales that reflect for the property a value that the property would have if the use of the property were not subject to the restriction unless adjustments in value are made reflecting the effect of the restrictions."
In affirming this court, the Oregon Supreme Court concluded that the restrictions agreed to by an owner in a typical federal project were governmental restrictions as to use within the meaning of ORS 308.205(2)(d), even though the restrictions were voluntary. The court also agreed that the federal tax credits produced by the project should not be added to the income of the property as no buyer would receive the benefit of those credits and they would be recaptured if the property was not maintained as a federal project. Bayridge II, 321 Or at 31-32.
The court review in Bayridge was de novo under ORS 305.445, prior to the amendment of that statute in 1995.
B. Piedmont
In Piedmont Plaza Investors v. Dept. of Rev., 14 OTR 440 ( Piedmont I), rev'd, 331 Or 585, 18 P3d 1092 ( Piedmont II) (2001), this court refused to accept both the conclusion of the taxpayer's appraiser and that of the government's appraiser in arriving at the value of a federal project. This court grounded its refusal to follow the taxpayer's appraiser on two stated reasons: (1) that appraiser's assumption that the burdens of the federal project cancelled out the value of the interest subsidy feature of the project, and (2) that appraiser's assumption that the property would be subject to federal project restrictions indefinitely. This court went on to find a value for the federal project based on the so called "Preservation Act transfer price."
For ease of reference, Piedmont I will refer to the Tax Court opinion, 14 OTR 440 (2001). Piedmont II will refer to the Oregon Supreme Court opinion, 331 Or 585, 18 P3d 1092 (2001). All references to Piedmont refer collectively to the opinions of both courts.
The Oregon Supreme Court, reviewing the case de novo, noted that only the taxpayer challenged the action of this court in discarding its expert's opinion. The court then analyzed the reasons this court gave for discarding the conclusion of the taxpayer's appraiser and, finding those objections not well taken, found values consistent with the valuation report of the taxpayer's appraiser.
C. Observation on Prior Cases
Some observations about the prior cases and the position of the parties in this case are appropriate.
1. First, with the focus on ORS 308.205(2)(d) in the appeal of Bayridge I to the Oregon Supreme Court, some other points from that opinion, not appealed, are important to keep in mind. One aspect of fundamental importance from Bayridge I, not discussed or questioned in Bayridge II or in the Piedmont litigation, was the nature of the governmental interest in a low-income housing project and the treatment of that interest under the settled law of property valuation in Oregon. As observed in Bayridge I:
"Generally, the entire value of property is assessed to the holder of legal title without regard to other interests. However, there are two exceptions: (1) easements appurtenant, which shift value from one tax lot to another, and (2) publicly held interests. Taxes are imposed on private property as a means of sharing the expense of public services. Publicly owned property is not taxed unless used for private purposes. Generally, property is taxed only to the extent of the value held for private benefit. It would constitute a form of double taxation to tax a private owner on interest held for public benefit. See, for example, Parkside Plaza Apartments v. Dept. of Rev., 10 OTR 132, 135 (1985), recognizing the effect of urban renewal restrictions on the value of property.
"There is no question low-income housing benefits the public. To the extent that low-income housing restrictions diminish the value of the property, they reduce its taxable value."
13 OTR at 28-29 (citations omitted; footnote omitted). The public interest rule is also discussed in Tualatin Development v. Dept. of Rev., 256 Or 323, 333, 473 P2d 660 (1970), in which the Oregon Supreme Court cited with approval the decision in Borough of Englewood Cliffs v. Estate of Allison, 69 NJ Super 514, 174 A.2d 631 (1961). That rule is also discussed in this court's Opinion in Willamette Factors v. Dept. of Rev., 8 OTR 400, 405, aff'd, 291 Or 568, 633 P2d 781 (1981).
2. Second, as observed in Bayridge I, ORS 308.205(2)(d) does not add anything or require adjustment to the traditional approach in determining market value. Specifically, ORS 308.205(2)(d) does not change or modify the conditions of an income approach to value and governmental restrictions should be reflected in an income approach if those restrictions would not be removed by sale of the property. Bayridge I, 13 OTR at 27; see also Douglas County Assessor v. Dept. of Rev., 13 OTR 448 (1996).
Third, there existed no immediate market in the property interests of the type to be valued in this case. What the Oregon Supreme Court observed as to section 236 projects during the same time period in Piedmont II is equally true here. Legal uncertainties or other factors have resulted in neither party submitting any evidence, in this case, of open-market sales of restricted properties. Piedmont II, 331 Or at 593. Accordingly, ORS 308.205(2)(c) applies. Id.; see also Truitt Brothers, Inc. v. Dept. of Rev., 10 OTR 111 (1985), aff'd, 302 Or 603, 732 P2d 497 (1987).
ORS 308.205(2)(c) provides:
"If the property has no immediate market value, its real market value is the amount of money that would justly compensate the owner for loss of the property."
III. ANALYSIS
A. Application of Settled Principles
In previous low-income housing cases, most attention has been on the effect government restrictions have on the value of the property interest subject to taxation, perhaps because of ORS 308.205(2)(d), which focuses on those restrictions. Under the public interest rule recognized in Bayridge I, the focus should be on the nature and value of the public interest, which can be subtracted from the overall property value. In this case, these considerations are two facets of the same underlying reality. First, the interest of the federal government is in having affordable housing, available to persons who qualify as tenants under the government program. Second, the restrictions on the property insure the government gets what it wants and, most importantly, what it is paying for.
The government interest, defined and protected by the restrictions on rent, operations, and other matters, is paid for through the interest subsidy and the undertakings of the government as guarantor or nonrecourse lender. Indeed, the economic interest of the government is chronologically coterminous with the restrictions on the project. If the rent and other restrictions end or are breached, the government no longer pays the interest subsidy.
The government interest in the project is a substantial economic interest, as demonstrated by the economic exposure the government undertakes to obtain it. At the outset, and well into the life of such projects, the federal government, either as guarantor or as the direct nonrecourse lender, has the dominant economic interest in the project. If the operating cash flow of the project falls below that necessary to meet obligations, the partners in the project have very little invested. They have very little incentive, and no legal obligation, to contribute to operating needs. As guarantor or nonrecourse lender, the government must ultimately be economically responsible, either by paying on its guarantee, foreclosing on its security (directly or as guarantor after subrogation) and becoming the owner of the project, or by further subsidizing the operations of the project.
The real and substantial economic interest of a nonrecourse lender is recognized in federal income tax regulations. The regulations recognize the nonrecourse lender bears the economic burden and risk-of-loss associated with assets financed with nonrecourse debt. See generally Treas Reg § 1.704-2(b)(1). In economic substance, until the government nonrecourse loan is repaid, the government has an economic interest approaching an equity interest in the project, which, in amount, approximates the outstanding balance of the mortgage loan.
Many 515 Projects operate not only with interest rate subsidies but also with rent subsidies. The rent subsidies involve federal government payment of a portion of the low-income tenant's rent. For purposes of the hypothetical stated in the text, if rents on the project were to be increased to meet increased costs, it is assumed that tenant contributions would stay fixed and the government subsidy would increase.
The existing case law requires that the value of the substantial government interest in a federal project be taken into account in the valuation process by reducing what would otherwise be the taxable value of the property by an amount equal to the value of the government interest in the property.
Defendant Department of Revenue (the department) repeatedly criticizes taxpayer's appraiser for failing to value all interests in the subject property. The department describes its approach as follows:
"Defendant's valuation evidence is the most reliable and credible to achieve the legal and appraisal requirements for the real market value of the subject. It values all the real property interests together as a whole."
The department's position is inconsistent with the law as to governmental interests outlined in Bayridge I. It is wrong precisely because it values all interests in the property and purports to value the taxable interest without reduction for the value of the public interest.
The department's error may derive from a focus on ORS 308.205(2)(d), which diverted it from the teaching of Bayridge, Willamette Factors, and Parkside Plaza as to the proper considerations in the case where public interests are part of the property to be valued. Those considerations predated the adoption of ORS 308.205(2)(d). As observed in Bayridge and Douglas County, the adoption of ORS 308.205(2)(d) has not affected the traditional RMV analysis; likewise, the statute does not displace or fundamentally alter application of the public interest rule. Indeed, the public interest rule and the statute are expressions of the same concern, that the value of public interests effected through use restrictions be taken into account in valuation.
In its briefing, the department has asserted that the decision of the Supreme Court in Bayridge II was in error in concluding that ORS 308.205(2)(d) applied to restrictions other than zoning restrictions. This court does not accept the department's invitation to call into question the opinion in Bayridge II. The public interest analysis of this opinion does not depend on the Bayridge II discussion of ORS 308.205(2)(d). Further, this court has indicated that voluntary restrictions on land use are similar to or the same as other zoning restrictions. See Willamette Factors, 8 OTR at 406 (holding that open space restrictions agreed to by a golf course developer were properly treated as zoning restrictions and serve to reduce value under public interest rule discussed in Tualatin Development, 256 Or 323, 473 P2d 660 (1970)).
To properly account for the government interest, it must be valued and be subtracted from the overall value of the property. In addition, because it is the only interest in the project other than the private taxable interest, it is mathematically true that the value of the taxable interest may be determined as follows:
For ease of reference, this approach will be set out as: VPWR/VGI/VTI.
Value of Project Without Regard to Restrictions (VPWR) — Value of Government Interest (VGI) _______________________________________________________ = Value of Taxable Private Interest (VTI)
Oregon case law recognizes that all interests are taxable to an owner, and the taxable interest is all value, except the value of a governmental interest or the value associated with an easement appurtenant. Bayridge I, 13 OTR at 28-29. If the subject property were also burdened with an easement appurtenant to another property, a further subtraction from VPWR would be necessary to arrive at VTI.
Use of the VPWR/VGI/VTI approach is particularly valuable in this case, because there is no immediate market for the private interest that is the subject of valuation. However, there is an active market in unregulated apartment housing of various types and sizes, information from which can be used to make reliable estimates of VPWR using the traditional cost, income, and market indicators of value. In addition, Oregon case law provides clear guidance on an appropriate way to calculate VGI. With VPWR and VGI determinable from approved methodologies, VTI may be calculated with reasonable confidence. That process will lead to a conclusion as to RMV as contemplated by ORS 308.205(2)(c) in circumstances, such as this case, where an immediate market value for the interest to be valued does not exist.
1. Determination of VPWR
The department's appraisers consistently refused to consider conventional indicators of value to arrive at an unrestricted value for the project. They repeatedly asserted that developing a value for comparable unrestricted properties would be to compare apples to oranges. That might be true if the VPWR, as computed, was used as an indicator of value of the restricted project. However, that is not the use of the VPWR proposed either by taxpayer or by this court. The VPWR is a starting point in arriving at a conclusion — it is not the conclusion.
Taxpayer's appraiser took VPWR and made adjustments to it for the effect of government restrictions. The court's approach differs in that the adjustment to VPWR is made by subtracting VGI. As stated above, the government interest is created by the restrictions.
Taxpayer's appraiser did compute a VPWR using conventional and accepted methods. The court finds that the value conclusions so determined are credible and adopts them for purposes of this analysis. Accordingly, the VPWR for the relevant years is found to be:
1992 $825,000 1993 $850,000 1994 $860,000
2. Determination of VGI 3. A property interest may be valued by capitalization of payments made for the interest. That method was employed in P.G.E. Company v. Commission, 249 Or 239, 437 P2d 827 (1968). In that case, the question was the value of an interest in Indian tribal lands that the taxpayer acquired in exchange for payments to be made over a fixed period of time. The Oregon Supreme Court overruled this court's valuation methodology and determined the value of the interest in question by reducing to present value the stream of payments that the holder of the interest was to make over time. Id. at 250. That approach was obviously founded on the premise that the present value of an interest can be equated to the present value of what is paid for the interest. Id. at 252.
In Joseph Hydro Associates v. Dept. of Rev., 11 OTR 49 (1988), this court followed the approach of P.G.E. Company where the interest being valued, a right to use water, was, like the interest in P.G.E. Company — in the nature of a lease. The interest of the government in a 515 Project is also in the nature of a lease. In substance, through the 515 Program, the government obtains rights to occupy rent restricted housing; those rights are in effect assigned to qualifying tenants through other provisions of the 515 Program.
Using the methodology of P.G.E. Company and Joseph Hydro, the value of the interest of the federal government in the project at issue, as of 1992, would be in the range of $550,000. That was the present value in 1992 of the future stream of interest subsidy payments.
The $550,000 amount was calculated by taxpayer's appraiser using methods the court finds reasonable.
In P.G.E. Company, the person acquiring the interest to be valued did not provide other consideration beyond cash payments in order to acquire the interest. In the case of 515 Projects, the federal government makes a series of payments. However, it also remains ultimately at risk on the entire unpaid debt on the project, which is initially 95 percent of the project cost. The federal government either guarantees the nonrecourse debt borrowed by the developer/owner to finance the project cost or, as in this case, actually becomes the lender of funds on a nonrecourse basis. The government credit support is critical; lenders would not advance adequate funds on a nonrecourse basis to construct the projects without the government guaranty. Consequently, VGI must reflect not only cash payments made by the federal government but also its credit support position. 4. The government credit support position is not an item regularly traded in a market and therefore its value cannot be easily ascertained. It can, nonetheless, be valued by employing the "presumed equivalency of value" concept. See Philadelphia Park Amusement Co. v. U.S., 126 F Supp 184, 189, 1954-2 US Tax Cas (CCH) ¶ 9,697 (Ct Cl 1954) ("the value of * * * two properties exchanged in an arm[']s-length transaction are either equal in fact, or are presumed to be equal"). That concept teaches that where there is an arm's-length exchange and one side of the exchange can be valued with reasonable certainty, the other item in the exchange, even though it is difficult to value independently, is assumed to be of a value equivalent to the value of the item for which value can be calculated with reasonable certainty.
Here the transaction between the owner and the federal government is at arm's length. Because of the lack of a market in government credit support, the value of the government side of the exchange is difficult to compute. However, the total value the project owner pays or surrenders to obtain the government participation is reasonably capable of calculation. It is the amount of rent lost by reason of the government restrictions on rent. That amount, expressed as a present value in 1992, was $620,000.
Again, as computed by taxpayer's expert witness. See footnote 13.
Accordingly, the implicit value of the government interest, including the interest subsidy and guarantee, can be said to be equal to the value of what the project owner surrendered to obtain government participation, a 1992 present value amount of $620,000.
The use of the present value of lost rents in valuing the government interest will reflect, over time, the changing conditions that are the reality of a federal project or any other real estate investment. As the years elapse for such a project, the period of the restrictions on rentals shortens. As time passes and the loan is repaid, the value of the remaining interest subsidy and the continuing "guaranty" position of the federal government also diminishes. Correspondingly, the value of the right to occupancy at restricted rents diminishes over time. Accordingly, other things being equal, the relative value of the government interest should decline and the relative value of the private taxable interest should increase as the day approaches when the project owner can operate free of the substantial restrictions on rents and returns that are the hallmark of the federal program.
The methodology of valuing the government interest by reference to the foregone rent, computed by comparing general market rents for apartments with comparable physical features, will also serve the important function of reflecting fluctuations in value as the gap between the project rentals and the market rentals widens or narrows. If, for example, market rents generally decline, and the rent restrictions remain unchanged, VGI, which is the present value of the smaller amount of rental gap, would decline. Further, the VPWR/VGI/VTI approach should, other things being equal, show an increase in the relative value of the taxable interest as time passes.
In cases where the interest of the landowner and lessee are in private hands, no such segregation of interests and calculation of premium or discount positions for leases occurs. But where the interest of the lessee is owned by a government, segregation does occur. See generally Willamette Factors.
The relative value positions of the government and private interests will always, of course, only be equal to the unregulated value of the property. If the property declines in condition, the absolute value of the positions will, other things being equal, decline as well.
By comparison, the methodology adopted by the department's appraisers is wholly inadequate. Two appraisal reports prepared by two different firms were submitted by the department. Both reports use only an income approach. They develop a net operating income (NOI) that is not materially different from that of taxpayer's appraiser. However, when calculating the capitalization rate, the department's appraisers look only to the actual terms applicable to this project, both as to debt/equity mix and as to return on equity and debt. They employ a direct capitalization method projecting any given year's experience into perpetuity.
That approach is reviewed in more detail below, but at this point it is appropriate to observe that the approach provides a value at the inception of the project, midway through the life of the project, and at the end of the project that does not vary. That result is obviously flawed because it does not take into account the decreasing value of the government interest or the impact of termination of the government interest, after which the project owner is able to charge unrestricted rents and otherwise use the property without restriction. The method of the department's appraisers also completely ignores whether the value of the owner's interest might change depending on the level of market rents generally or the condition of the property.
See footnote 2.
3. Value of Taxable Interest
Given the VPWR and VGI as set forth, the resulting VTI under the approach is as follows:
The department takes the position that the VTI cannot be as indicated because a lender made a loan of almost $700,000 on the property, and the property was insured for a comparable amount. Here again, the department confuses the value of the property with the value of the taxable interest. Once the value of the government interest is added to the value of the taxable interest, the total is in the range of loan and insurance values. Also, in federal projects, the position of the government as guarantor on public sector debt means the lender is, in economic logic, not particularly concerned with real estate collateral value. The lender has, instead, a classically risk-free federal obligation.
1992 $825,000 $620,000 $205,000 1993 $850,000 $620,000 $230,000 1994 $860,000 $640,000 $220,000
It appears taxpayer's expert witness may have mistakenly duplicated calculations for 1992 and 1993. This issue will be addressed in supplemental proceedings.
B. Alternative Approach or Indicator: Direct Capitalization of Income
Basic appraisal theory teaches that several indicators of value should be considered so as to test the value conclusion of any one indicator. As stated above, there is no immediate market for the taxable interest to be valued in this case. A comparable sales indicator is therefore not available and other indicators of value must be employed. The VPWR/VGI/VTI approach set forth above is based upon an active market in apartment housing (used to develop VPWR), an additive methodology derived from Oregon case law on valuation of public and private interests in one property, and a valuation methodology approved by the Oregon Supreme Court for interests acquired in consideration of a series of installment payments. The court will also consider what value is indicated by developing the traditional income indicator for the taxable interest, using a direct capitalization approach.
As stated, there is no comparable sales indicator for the interest in question in this case. A separate cost indicator has not been calculated. None of the appraisers placed significant weight on the cost approach created by the rent restrictions. In addition, that calculation would involve a deduction for external obsolescence that is essentially another application of the income indicator. Appraisal Institute, The Appraisal of Real Estate 413 (12th ed 2001).
1. Calculation of Private Interest/Direct Capitalization of Income
Both parties agree that the income approach can be accorded substantial weight. The parties also have little difference as to the income to be capitalized. That income is based on the actual — restricted — rents available to the owner.
The accepted approach to capitalized income as an indicator of market value recognizes two basic choices for a method of capitalization: direct capitalization and yield capitalization. Direct capitalization involves developing an estimate of a single year's income, dividing that number by an overall capitalization rate. Yield capitalization involves developing cash flows over time, including an ultimate remaining value (reversion) realized at the end of a predicted holding period. Those cash flows are then discounted to present value through application of an appropriate yield rate. Appraisal Institute, The Appraisal of Real Estate 494, chs 22, 23 (12th ed 2001).
Where a property is going through an initial lease-up or when income or expenses are expected to be irregular over time, the direct capitalization approach is not recommended. Id. at 529. However, the method is recognized as easy to use. Id.
In this case, both income and expenses may well be relatively stable during the period of the federal program, but thereafter the record shows that rents would undoubtedly change as would expenses incurred in managing the property. In the court's view, the direct capitalization method is appropriate for use, especially in the early years of a government program. As the termination of the program approaches, the major weakness of a discounted cash-flow (DCF) analysis, the need to project future revenues and expenses, diminishes. At the same time, the importance of post-restriction rents will increase. Therefore, as the remaining life of the government interest diminishes, the importance of the DCF analysis increases.
2. Capitalization Rate
Without regard to which capitalization method is used, a capitalization rate must be developed. That rate is a powerful element in the capitalized income indicator of value. Small changes in capitalization rate can make large differences in ultimate indicated value.
The capitalization rate developed should be a reflection of general-market realities. Such rates are typically developed either from examination of actual sales and income data or from methods such as the band-of-investment approach. Id. at 531, 534. In all cases, the capitalization rate should reflect what investors generally are expecting from an investment in a particular type of property. In this case, the investment is in multi-family rental housing. Therefore sales and income data or band-of-investment data for such properties must be considered. In this case, the department's appraisers relied solely on the band-of-investment methodology whereas taxpayer's appraiser made calculations based on both market derived rates and band-of-investment methodology.
5. Taking the band-of-investment methodology used by both parties, what capitalization rate is indicated? The parties take materially different approaches to the question. Taxpayer attempts to establish what hypothetical buyers in the market as of the date of assessment would use as a mix of debt and equity. Taxpayer then attempts to establish what the market participants would expect as a debt return and an equity return, again as of the date of assessment. That is the generally recognized approach. See ORS 308.210(1) (providing that the assessor shall assess the value of property as of January 1 of the assessment year); see also Appraisal of Real Estate at 534.
The department's appraisers, on the other hand, consult no market but only reference the facts of the particular project at issue, at the time of original construction of the project. They therefore conclude that the debt and equity components will be 95 percent and 5 percent respectively: the precise debt/equity mix of this project type. They set the equity return at 8 percent, the rate provided by contract for this project. Finally, and most importantly, the department's appraisers look to the project documents and conclude that the mortgage component should be calculated using 1 percent as the interest rate on the mortgage. Based on this approach to the band-of-investment, the capitalization rate computed by the department was:
.05 equity × .08 equity return = .00400 + .95 debt × .02542 = .02415 + adjustment for property tax rate = .02360 ________________________________ ____________ Adjusted capitalization rate = .05175
This mortgage constant is computed based on the specific terms applicable to this project: 50-year amortizing loan with 1 percent interest and the initial balance of the loan in this case.
This is the rate calculated by appraisers Cassinelli and Jackson. The rate computed by appraisers Miller and Zacha was .0536.
As discussed below, the approach taken by the department's appraisers in calculating a capitalization rate is not supported by any appraisal theory, is inconsistent with the department's published rules, and is incompatible with the facts in this case. The approach must, therefore, be rejected.
a. Appraisal Theory on Capitalization Rate
Accepted appraisal theory is that the debt/equity mix as well as the debt and equity rates used in the band-of-investment methodology are to be derived from market information as of the date of assessment. The goal is to estimate what terms and returns would be necessary to attract debt and equity capital from general market sources of such capital to a particular property. Id. at 534-37. A hypothetical buyer of the owner's income stream will use the usual market mix of debt and equity and will pay or expect usual market returns for debt and equity capital. Therefore, market debt/equity ratios and returns must be used to capitalize income into an indicator of value.
In focusing solely on the original terms relating to the debt and equity financing for this project, the department's appraisers not only departed from that principle, but also froze their focus on the time of the initial negotiation for debt and equity capital. Their approach prevents any fluctuation in value to occur based on shifting expectations of the capital markets over time. There is no support cited for that approach and it is fundamentally at odds with the fact that property values, do in fact, vary over time. That variation, even when the income stream from the property is relatively constant, is produced by market recalibration of financial expectations.
b. The Department's Rules on Capitalization Rate
The department's appraisers departed from the methodology that the department has publicly announced should be followed. In OAR 150-308.205-(C), the department addressed calculation or derivation of capital structure and discount rates for properties subject to central assessment. Neither in the rule, nor to the court's knowledge, is there any reason to limit the principles of the rule to only centrally assessed properties.
As to calculation of discount rate, identified in the rule as a synonym for capitalization rate, the rule states, in part:
"The cost of debt is the current market rate for new securities. The embedded rate on securities previously issued is not a proper measure."
OAR 150-308.205-(C)(2)(b). Similar statements are made for development of capitalization rates for preferred and common equity. See OAR 150-300.205-(C)(2)(c)(d). That rule, stressing reliance on current financial markets and rejecting embedded or historical capital costs, is described in rules applicable to centrally assessed property but is made applicable to valuation of industrial property by OAR 150-308.205-(D)(2)(h).
In the face of general appraisal theory and the department's own rules, the departures by the department's appraisers are unexplained in their reports. Cassinelli and Jackson simply observed that "the subject is not a typical market rate project. Therefore, the actual terms will be utilized in this analysis." No explanation was offered by Zacha and Miller for their reliance on the contract terms for the subject property. However, as to the historical or embedded rates they observe that "[i]n the subject's case, the mortgage requirement is known; it is established by contract and it constitutes most of the capital in the investment."
The department's appraisers insisted that the same financing package and terms available in 1985-86 would have been available to any purchaser of taxpayer's interest in the project in the 1992-95 time period. Their only support for that position consisted of hearsay statements made by the department's experts as to what federal program managers had said. It is not at all clear to the court if those managers specifically addressed the matter. In addition, taxpayer introduced testimony indicating such financing was not available for purchasers of existing projects.
c. Facts of This Case
The observation by Zacha and Miller that the mortgage requirement was established by the contract demonstrates another major defect in the approach of both of the department's appraisal reports. Zacha and Miller state that the mortgage requirement of 1 percent used by them is established by the contract. In fact, it is not. The reference to "contract" is clearly a reference to the regulatory agreement between the developer and the government and related debt instruments. However, even the department's appraisers acknowledge, "[t]he actual institutional lender was paid a market rate for the mortgage." The interest rate paid on the debt for this project was 10.625 percent, not 1 percent.
The logic of the department's appraisers, in using a 1 percent debt rate, appears to be that because the income from the project is used to pay debt service only to the extent of principal amortized at 1 percent, the cost of debt financing was only 1 percent. That approach is theoretically flawed, however, because it focuses entirely on the borrower's position and on the application of the borrower's operating cash flow. The generally accepted theory looks at the lender's demanded return for debt capital. Appraisal of Real Estate at 536-37. The approach of the department's appraisers, however, does not consider what lenders in the general market might have demanded for lending. Worse still, it ignores the fact that the particular lender that made a debt investment in this project did not agree to accept a 1 percent return on debt capital. In fact, the lender in this case bargained for and was entitled to receive a return of debt capital and an interest return of 10.625 percent.
As described above, the lender was initially First Interstate Bank and ultimately the federal government.
In using a 1 percent amount in the debt component of the band-of-investment analysis, the department's appraisers confuse the return required by the lender with the source of cash to satisfy that lender requirement. In projects such as the one at issue, principal and 1 percent interest comes, hopefully, from project cash flow. The lender's requirement, however, is 10.625 interest; the remainder of that interest requirement has its source in the United States Treasury. The government, in essence, makes capital contributions to fund the debt-service requirements for the project and also contributes its "guarantee" or nonrecourse debt position. It does so in order to obtain the interest it has in the project — the right to have submarket rate occupancy in highly regulated properties.
For this project, the lender demanded, and received, repayment of a loan amortized at 10.625 percent interest. Because the methodology of the department's appraisers in developing a capitalization rate ignores this basic fact, it must be discarded.
Only taxpayer's appraiser, Gardner, inspected and analyzed the capital markets and financing practices as of the assessment dates in question. The court has reviewed and adopts the conclusions arrived at by Gardner. In the direct capitalization band-of-investment methodology, Gardner concluded, for 1992, that a market capital structure would have been 75 percent debt and 25 percent equity. Gardner further concluded that the mortgage component would have been 10 percent and the equity component would have been 9 percent. From this, Gardner concluded a base band-of-investment rate of 10 percent.
To the base band-of-investment rate of 10 percent, Gardner added a property tax component of 2.36 percent for the 1992 year. Gardner then also added a component of 3 percent, which he felt was necessary to adjust the market rate to account for the following features of the subject property: lack of income increase; lack of property appreciation; and extremely long holding period.
Both parties agree that the proper methodology for dealing with property tax expenses in an income approach is to compute NOI without a deduction for property tax and add to the capitalization rate the applicable ad valorem property tax rate for the tax years in question. As to this point, the court notes that the years in question are prior to the adoption of Ballot Measure 50.
The record indicates that income for this federal project is rigidly limited to 8 percent of initial equity invested. Testimony of one experienced project owner indicates that savings or favorable developments cannot redound to the benefit of the project owner and, if rental rates produce surpluses, the rates will be reduced under the project agreements. Similarly, the term of the regulatory agreements is beyond the typical holding period for apartment housing, and during that term, the project owner cannot benefit from equity buildup or appreciation through a sale or refinancing of the project.
The department was highly critical of Gardner's method of increasing the capitalization rate, accusing him of being arbitrary. Gardner arrived at a minimum adjustment of 2 percent, a maximum adjustment of 4 percent, and recommended an adjustment of 3 percent without much explanation. The order of magnitude of the adjustment was, however, supported with the observation that a 1 percent adjustment was typical even for Class A/luxury apartments.
The court agrees with Gardner that an upward adjustment to general market capitalization rate is appropriate. That adjustment is also supported by the fact that in order to obtain the debt capital for this project, the project owner must also secure government credit or credit support. That is not obtained without cost. The court finds it appropriate to view that cost as an additional cost of debt financing reflected in the mortgage component of the band-of-investment analysis.
Although there may be no precise objective method for calculating an adjustment to the base band-of-investment rate, the court finds Gardner's estimate reasonable. Accordingly, Gardner's direct capitalization indicator of $280,000 is adopted by the court, with two important caveats. First, the conclusion represents the capitalization of calculated NOI, for 1992, in the amount of $42,711, even though the project owner has no access to NOI above the amount of the equity dividend and owner's portion of debt service, at most $20,146. Second, the adjustments made by Gardner to the mortgage rate do not reflect an upward adjustment for the cost of the government guarantee. Accordingly, the $280,000 figure is at the upper end of the range of indicated value.
This figure is the sum of the maximum equity dividend of $2,840 and the owner's portion of debt service in the amount of $17,306.
C. Alternative Approach: Yield Capitalization Indicator of Value of Economic Interest of Owner
Another method for testing the reasonableness of the outcome of the VPWR/VGI/VTI approach is to step back from much of the complexity in the theory and simply ask what a rational economic actor would likely pay for the economic interest owned by taxpayer.
This inquiry should be based on a present valuation of the stream of economic benefits inherent in the interest of the project owner. The project owner is to receive, at most, $2,840 each year for 43 years, the owner's portion of debt-service cash flow, and the reversionary interest in the property at the end of 43 years. The project owner cannot receive any other economic benefit. If cash flows from the project exceed expenses, either reserves belonging to the government are increased or rents are reduced.
The record demonstrates that although $2,840 is a limit on the cash return to the project owner, many events could cause the owner to receive less than that amount, even where the owner manages reasonably.
The uncertainty as to whether prepayment could be achieved at 20 years is significant. All of the appraisers tended to assume a remaining 43-year period for this program.
Throughout their testimony and reports, the department's appraisers argued that their approach took into account the benefits to the project owner of the government program. They identified one of those benefits as income tax depreciation on the projects. However, appraisal theory excludes consideration of the hypothetical buyer's income tax position. See Appraisal of Real Estate at 521. Quite apart from the fact that appraisal theory ignores the "benefit" identified by the department's appraisers, changes in federal income tax law after construction of this project and before the appraisal date significantly and adversely impacted what had once been the supposed "benefits." The department's appraisers also argue that the mortgage interest subsidy is a benefit to the property or its owner. As discussed throughout this opinion, the correct view of the subsidy is as a cost of the government interest that is not to be taxed to the owner.
This approach is a yield capitalization, and, as calculated by Gardner, resulted in a value of $195,000 for the project. The importance of that calculation is not its absolute outcome, but the fact that it provides a context of value in the area of $200,000. Similarly, the direct capitalization approach, as correctly applied, yields an indicator of $280,000 without fully discounting for limited income rights and costs of the government guarantee. Those values are to be compared to the result obtained under the VPWR/VGI/VTI approach of $825,000 minus $620,000, or $205,000. Making that comparison and reconciling the indicators, with special consideration that the $280,000 indicator is at the upper end of any range, the court finds that the value of the property is in the amount indicated by the VPWR/VGI/VTI approach.
The Gardner calculation is based on the reversion being bare land. Testimony about significant maintenance problems with projects of this type fully supports that assumption.
The court can place no confidence in the conclusion of the department's appraisers that a buyer would pay in the range of $700,000 to $750,000 for, at most, a $2,840 annual payment and a reversion after 41 to 43 years.
IV. CONCLUSION
The proper valuation of a federal low-income housing project must take into account, as a reduction from the unrestricted value of the project, the value of the interest of the federal government in the property. The department's methodology fails to do that and must be rejected. The approach of the taxpayer's expert witness provides the information necessary to apply the VPWR/VGI/VTI approach.
It is the opinion of this court that the RMV for the project for the years in question was as follows:
1992 $205,000 1993 $230,000 1994 $220,000
Subject to adjustment after supplemental review of calculations for lost rental for the 1993 year. See footnote 19.
Because of the fact that this court will not determine values outside the range of the pleading of the parties, Chart Development Corp. v. Dept. of Rev., 16 OTR 9 (2001), the value determined for each year will not be less than the $225,000 pled by the taxpayer for each year. Costs to Plaintiff.