Nancy STRICKLAND, et al., Plaintiffs, Appellees, v.
COMMISSIONER, MAINE DEPARTMENT OF HUMAN SERVICES, Defendant, Appellee, v. SECRETARY, U.S. DEPARTMENT OF AGRICULTURE, Third-Party Defendant, Appellant.
This suit questions the validity of a regulation promulgated by the Secretary of Agriculture in connection with his management of the Food Stamp Act, 7 U.S.C. §§ 2011-2025 (1988) (the Act). Answering the question requires us to explore the frontiers of Chevron deference. See Chevron, U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). Because we believe that proper respect for the Secretary's interpretation of the applicable statute validates the regulation, we reverse the district court's order barring its enforcement.
BackgroundAThe Food Stamp Act
The Act harks back to 1964. Congress passed it “to safeguard the health and well-being of the Nation's population by raising levels of nutrition among low-income households.” 7 U.S.C. § 2011. The Act creates a federally funded, but state administered, program designed to distribute food stamps according to income and family size. The recipient can use these stamps to purchase food at local markets. Participating retailers accept the stamps as if they were cash, for the government redeems them at face value. The Secretary of Agriculture is charged with overseeing the federal aspects of the food stamp program. See id. at § 2013. Agencies selected by the several states administer the state aspects of the program. The Department of Human Services (DHS) performs this function in Maine.
Congress originally restricted eligibility for food stamps to families of limited means but made no attempt to define income (leaving that chore to the states). In 1971, Congress directed the Secretary to establish uniform standards of eligibility. The Secretary then promulgated regulations that defined net income as gross income less “the cost of producing that income,” but excluded depreciation as a component of this deduction. See 36 Fed.Reg. 14102, 14107 (July 29, 1971) (enacting former 7 C.F.R. § 273.1(c)(1)(b)).
In 1977, Congress retrofitted the Act. In a comprehensive, detailed revision of the statute, Congress specified that, for purposes of program eligibility, income was not to include the “cost of producing self-employed income.” 7 U.S.C. § 2014(d)(9). Although the 1977 amendments did not define the term “cost,” the House Committee on Agriculture reported that “the Department would be expected to revise its regulations in this regard to allow some form of depreciation in arriving at ‘net’ business income.” H.R.Rep. No. 464, 95 Cong., 1st Sess. 25 (1977), reprinted in 1977 U.S.C.C.A.N. 1978, 2001-02. The Secretary revisited the topic in 1978 and promulgated regulations allowing depreciation as a cost in calculating self-employment income. See 43 Fed.Reg. 47846, 47912 (Oct. 17, 1978).
In 1980, a report produced by a joint House-Senate conference committee muddied the waters. The conference committee report accompanied the Food Stamp Act Amendments of 1980 (the FSAA), Pub.L. No. 96-249, 94 Stat. 357 (1980), which, among other things, decreased the aggregate value of non-excludable assets that a family might own while retaining food stamp eligibility. The report memorialized the conferees' “inten[tion] that the Secretary no longer permit depreciation to be subtracted in determining net self-employment income.” H.R.Conf.Rep. No. 957, 96th Cong., 2d Sess. 29 (1980), reprinted in 1980 U.S.C.C.A.N. 1057, 1070. Despite this statement of congressional intent, however, the FSAA made no change in the text of the statutory provision that allowed a deduction for the “cost of producing self-employed income,” 7 U.S.C. § 2014(d)(9).
Hard on the heels of this conference committee report, the Secretary proposed regulations aimed at eliminating depreciation from the computation of the cost of producing self-employment income. In the proposal, the Secretary stated:
The regulations implementing the 1977 Act included a provision allowing depreciation as a cost of doing business for self-employed households (§ 273.11(a)(4)(ii)). This was done in compliance with the legislative history, H.R.Rep. No. 95-464, p. 25. The Conference Report accompanying the 1980 Amendments suggests that the Secretary delete depreciation, H.R.Rep. No. 96-957, p. 29. Allowing such costs when determining net self-employment income results in an exemption of amounts not constituting “actual costs” to the household; households are, in a sense, given a deduction in advance for the cost of capital goods which is otherwise not allowed. Appropriate changes are being proposed to § 273.11(a) to correspond to the Conference Report's suggestion.
46 Fed.Reg. 4642, 4646 (Jan. 16, 1981). The final regulation, 7 C.F.R. § 273.11(a)(4)(ii) (1994), mimicked the proposal and instructed the states to disregard depreciation in calculating net self-employment income.
Nancy and Lyle Strickland reside in Belgrade, Maine. They ran a successful construction business until 1990, when the recession forced them to downsize. Although they remained in business, their profits dwindled. At about this time, they applied for, and were granted, food stamp assistance. On their 1992 federal tax return, they reported a business loss of $4,686, largely due to claimed depreciation ($24,380) on construction equipment.1 In 1993, the DHS informed the Stricklands that they would no longer receive food stamps because, based on their tax return, they had annual self-employment income, without regard to depreciation, of $19,694. This equalled net income of $1,641.16 per month-more than twice the food stamp eligibility limit for a two-person household. See Stipulated Record at 5 (confirming that the eligibility ceiling for the relevant period is $766 per month); see generally 7 C.F.R. § 273.9 (1994) (linking income standards to the federal poverty level).
Disappointed by the finding of ineligibility, the Stricklands sued DHS in Maine's federal district court. They challenged the Secretary's amended regulation, 7 C.F.R. § 273.11(a)(4)(ii) (1994), which excluded depreciation on business equipment from the allowable “costs of doing business” in determining a household's eligibility for food stamps, as offensive to the mandate of 7 U.S.C. § 2014(d)(9). DHS filed a third-party complaint against the Secretary of Agriculture. The plaintiffs then obtained leave to amend, and asserted claims directly against the Secretary.2 The parties stipulated to the relevant facts and the district court certified the Stricklands as representatives of a class comprising all Maine food stamp applicants or recipients adversely affected by the amended regulation on or after July 1, 1992.
On April 8, 1994, the court granted the plaintiffs' motion for judgment on the stipulated record. See Strickland v. Commissioner, 849 F.Supp. 818 (D.Me.1994). The court framed the decisive legal issue in the following way: “Can Congress change the law, simply by directing that it be so in legislative history, without amending the pertinent statutory language?” Id. at 818. Judge Hornby answered this loaded question in the negative. He then ruled that the amended regulation could not stand because the Secretary had promulgated it in response to a perceived congressional directive, not embodied in a duly enacted statute, rather than in the authentic exercise of administrative discretion. See id. at 820. The Secretary now appeals.
Applicable Legal PrinciplesAStandard of Review
Interpreting a statute or a regulation presents a purely legal question subject to de novo review. See McCarthy v. Azure, 22 F.3d 351, 354 (1st Cir.1994); Liberty Mut. Ins. Co. v. Commercial Union Ins. Co., 978 F.2d 750, 757 (1st Cir.1992). Nevertheless, the availability of plenary judicial review “does not obviate the devoir of persuasion in a food stamp case in which a plaintiff challenges the validity of the regulatory mosaic.” Massachusetts v. Secretary of Agric., 984 F.2d 514, 521 (1st Cir.), cert. denied, 510 U.S. 822, 114 S.Ct. 81, 126 L.Ed.2d 49 (1993). An inquiring court-even a court empowered to conduct de novo review-must examine the Secretary's interpretation of the statute, as expressed in the regulation, through a deferential glass. See id.
The Chevron Doctrine
Judicial review of an agency's construction of a statute that it administers involves two separate but interrelated questions, only the second of which furnishes an occasion for deference:
First, always, is the question whether Congress has directly spoken to the precise question at issue. If the intent of Congress is clear, that is the end of the matter; for the court, as well as the agency, must give effect to the unambiguously expressed intent of Congress. If, however, the court determines Congress has not directly addressed the precise question at issue, the court does not simply impose its own construction on the statute, as would be necessary in the absence of an administrative interpretation. Rather, if the statute is silent or ambiguous with respect to the specific issue, the question for the court is whether the agency's answer is based on a permissible construction of the statute.
Chevron, 467 U.S. at 842-43, 104 S.Ct. at 2781-82 (footnotes omitted).
In performing the first part of a Chevron analysis, no deference is due. Instead, courts must look primarily to the plain meaning of the statute, drawing its essence from the “particular statutory language at issue, as well as the language and design of the statute as a whole.” K Mart Corp. v. Cartier, Inc., 486 U.S. 281, 291, 108 S.Ct. 1811, 1818, 100 L.Ed.2d 313 (1988); accord Dunn v. Secretary of Agric., 921 F.2d 365, 366-67 (1st Cir.1990). Beyond this point, it remains unclear whether, and if so, to what extent, a court engaged in the first stage of a Chevron inquiry may use other tools of statutory construction, such as legislative history, in searching for Congress' unambiguously expressed intent on a particular issue. See Dunn, 921 F.2d at 367 n. 2 (citing conflicting cases but not resolving the point).
Legislative history is subject to many and varied criticisms, and the uncertainty about its value in general parallels the uncertainty about its value in relation to the Chevron doctrine.3 Respectable authority indicates that it is appropriate to employ all the “traditional tools of statutory construction” in the first part of the Chevron analysis when the statutory language itself is not dispositive. See INS v. Cardoza-Fonseca, 480 U.S. 421, 432-43, 446, 107 S.Ct. 1207, 1221, 94 L.Ed.2d 434 (1987) (examining legislative history to confirm the validity of an interpretation suggested by the statute's language) (dictum); Massachusetts v. Lyng, 893 F.2d 424, 429 (1st Cir.1990). But there is also respectable support for the proposition that the Chevron analysis, in its initial phase, does not look beyond the statutory text. See, e.g., National R.R. Passenger Corp. v. Boston & Me. Corp., 503 U.S. 407, 416-17, 112 S.Ct. 1394, 1401, 118 L.Ed.2d 52 (1992) (stating that deference is due so long as “the agency interpretation is not in conflict with the plain language of the statute”); K Mart Corp., 486 U.S. at 292, 108 S.Ct. at 1818 (“If the agency regulation is not in conflict with the plain language of the statute, a reviewing court must give deference to the agency's interpretation of the statute.”); NLRB v. United Food & Commercial Workers Union, 484 U.S. 112, 133-34, 108 S.Ct. 413, 426-27, 98 L.Ed.2d 429 (1987) (Scalia, J., concurring) (criticizing dictum in Cardoza-Fonseca ); Stowell v. Secretary of HHS, 3 F.3d 539, 543 (1st Cir.1993) (approving deference where “statute is silent with respect to a specific question”).
We think that the difference between these two views may, as a practical matter, be more apparent than real.4 In any event, we do not find the legislative history in this case determinative. Thus, we need not precisely define the function, if any, of legislative history under Chevron. Rather, we assume arguendo, but do not decide, that an inquiring court may look in that direction during the initial stage of a Chevron inquiry.
On this assumption, the question whether Congress has spoken on a particular question involves two smaller steps. We look first to the statute's language. If the text, given its plain meaning, answers the interpretive question, the language must prevail and further inquiry is foreclosed. If no such readily apparent meaning springs from the statute's text, we next examine the legislative history, albeit skeptically, in search of an unmistakable expression of congressional intent. And if, at that stage, the statute itself, viewed in connection with the statutory design and the legislative history, reveals an unequivocal answer to the interpretive question, the court's inquiry ends.
Thus, it is only when a court cannot discern an unmistakably clear expression of congressional intent that the Chevron inquiry moves into its second stage. Until then, deference is not a consideration-but from that point forward, deference looms large. The court must examine the agency's interpretation to see how it relates to the statute. This examination involves a high degree of respect for the agency's role. The agency need not write a rule that serves the statute in the best or most logical manner; it need only write a rule that flows rationally from a permissible construction of the statute. See, e.g., Cohen v. Brown Univ., 991 F.2d 888, 899 (1st Cir.1993) (noting that it is unimportant to the Chevron analysis whether the court, if writing on a pristine page, would prescribe a different version of the regulation). In other words, an agency's interpretive regulations must stand “unless they are arbitrary, capricious, or manifestly contrary to the statute.” Chevron, 467 U.S. at 844, 104 S.Ct. at 2782-83.
To be sure, the Chevron doctrine has a protean quality. Under it, courts afford varying degrees of deference to agency interpretations in varying circumstances. See Stowell, 3 F.3d at 544; Sierra Club v. Larson, 2 F.3d 462, 468-69 (1st Cir.1993). To cite an example that possesses particular pertinence for present purposes, deference is “particularly appropriate in complex and highly specialized areas where the regulatory net has been intricately woven.” Massachusetts Dep't of Educ. v. United States Dep't of Educ., 837 F.2d 536, 541 (1st Cir.1988) (quoting Citizens Sav. Bank v. Bell, 605 F.Supp. 1033, 1042 (D.R.I.1985)). Accordingly, “[m]atters of accounting, unless they be the expression of a whim rather than an exercise of judgment, are for the agency.” Id. (quoting Cheshire Hosp. v. New Hampshire-Vt. Hospitalization Serv., Inc., 689 F.2d 1112, 1117 (1st Cir.1982)); accord American Tel. & Tel. Co. v. United States, 299 U.S. 232, 236-37, 57 S.Ct. 170, 172-73, 81 L.Ed. 142 (1936).
Ultimately, of course, deference depends on the persuasiveness of the agency's position. See, e.g., United States v. 29 Cartons of * * * An Article of Food, 987 F.2d 33, 38 (1st Cir.1993). Furthermore, an administrative agency's entitlement to deference is not limited to its initial interpretation of a statute. Agencies “must be given ample latitude to adapt [their] rules and policies to the demands of changing circumstances.” Rust v. Sullivan, 500 U.S. 173, 187, 111 S.Ct. 1759, 1769, 114 L.Ed.2d 233 (1991) (citations and internal quotation marks omitted). Consequently, an explained modification, even one that represents a sharp departure from a longstanding prior interpretation, ordinarily retains whatever deference is due. See id. at 186-87, 111 S.Ct. at 1768-69; Stowell, 3 F.3d at 544.
Taming the Oxymoron
“Subsequent legislative history” is, as others have noted, see, e.g., Continental Can Co. v. Chicago Truck Drivers, 916 F.2d 1154, 1157 (7th Cir.1990), an oxymoron. What is more, the hazards inherent in virtually all legislative history are magnified when congressional materials are created after the fact, and the views of a subsequent Congress are imputed to the earlier Congress that enacted a given statute. See Consumer Prod. Safety Comm'n v. GTE Sylvania, Inc., 447 U.S. 102, 117, 100 S.Ct. 2051, 2060-61, 64 L.Ed.2d 766 (1980); United States v. Price, 361 U.S. 304, 313, 80 S.Ct. 326, 331-32, 4 L.Ed.2d 334 (1960). Despite these problems, however, such materials occasionally have been found useful in decrypting an unclear statute. See, e.g., Seatrain Shipbuilding Corp. v. Shell Oil Co., 444 U.S. 572, 596, 100 S.Ct. 800, 813-14, 63 L.Ed.2d 36 (1980); United States v. Ven-Fuel, Inc., 758 F.2d 741, 758-59 (1st Cir.1985). We conclude that the value, if any, of such post-enactment materials should be decided case by case, see Liberty Mut. Ins. Co., 978 F.2d at 755 n. 7, but should always be ingested with a healthy dose of skepticism.
Finally, in evaluating an agency's response to subsequent comments by Congress, we must keep in mind not only the dubious value of such comments, but also two overarching constitutional principles. First, Congress cannot dictate the interpretation of a statute by a subsequent expression of one of its committees (even, as here, a joint conference committee); “it is the function of the courts and not the Legislature ․ to say what an enacted statute means.” Pierce v. Underwood, 487 U.S. 552, 566, 108 S.Ct. 2541, 101 L.Ed.2d 490 (1988). Second, and relatedly, Congress cannot amend a statute merely by inserting the proposed change in a congressional report (even in a report of a joint conference committee); the amendment must meet the various constitutional benchmarks, including bicameral passage and presentment to the President. See U.S. Const. art. I; see also INS v. Chadha, 462 U.S. 919, 944-51, 103 S.Ct. 2764, 2780-84, 77 L.Ed.2d 317 (1983).
Consistent with the methodology we have outlined, we begin by examining the plain language of 7 U.S.C. § 2014(d)(9) to determine if it speaks definitively to the necessity of including depreciation as a “cost” of producing self-employment income. We think it does not. Like many words, “cost” has more than a single, unitary meaning. While the plaintiffs proffer opinion evidence from an economist and an accountant to the effect that certain professional disciplines routinely calculate depreciation as a “cost” of producing income, that is only half the story; common sense argues that the word “cost” may also legitimately be restricted to cash outlays made in a given period to produce income. At bottom, then, the word “cost” is a chameleon, capable of taking on different meanings, and shades of meaning, depending on the subject matter and the circumstances of each particular usage. See 20 C.J.S. Cost (1940) (stating flatly that “[t]he term [cost] is one of equivocal meaning”). And among its varying constructions, “cost” assuredly can mean the price of, or the amount that must immediately be expended to purchase, an item. See, e.g., Webster's Third New International Dictionary 515 (1986) (defining “cost” in its primary sense as signifying “the amount or equivalent paid or given or charged or engaged to be paid or given for anything bought or taken in barter or for service rendered: CHARGE, PRICE”); Black's Law Dictionary 312 (5th ed. 1979) (defining “cost” as “Expense; price. The sum or equivalent expended, paid or charged for something.”); Funk & Wagnalls New Standard Dictionary of the English Language 591 (1934) (defining “cost” as “[t]hat which has to be given for a thing in order to procure it; especially, the price paid; outlay of any kind; expense”).
In these circumstances, a credible argument can be made that, as between two plausible meanings, a reader should give the word “cost” its ordinary meaning as opposed to the more specialized meaning preferred by accountants or economists. See Perrin v. United States, 444 U.S. 37, 42, 100 S.Ct. 311, 314, 62 L.Ed.2d 199 (1979) (recognizing that undefined words in a statute ordinarily should “be interpreted as taking their ordinary, contemporary, common meaning”); United States v. Holmquist, 36 F.3d 154, 159 (1st Cir.1994) (same), cert. denied, 514 U.S. 1084, 115 S.Ct. 1797, 131 L.Ed.2d 724 (1995). We need not go that far, however; it suffices to say that “cost” in this context can naturally be read as excluding depreciation on capital goods.
Since the statutory language does not resolve the issue, we must consult other sources for guidance as to whether Congress has spoken plainly on the question sub judice. The Stricklands asseverate that the legislative history of section 2014(d)(9) adds the requisite clarity. We do not agree. The lone reference to the subject in the House report accompanying the 1977 revamping of the Act, heralded by the plaintiffs as the linchpin of their asseveration, comprises but one paragraph in one report of one of the two chambers that passed the law. While a solitary reference may sometimes be enough to clear the mists that obscure a statute's text-the relevant inquiry, after all, is qualitative, not quantitative-the solitary reference to which appellants cling is too slender a reed to be accorded controlling weight under the totality of the circumstances that obtain here. See, e.g., United States v. Taylor, 752 F.2d 757, 764 (1st Cir.1985) (noting the hazards of relying on an isolated fragment of legislative history-there, a single paragraph in a 21-page committee report), rev'd on other grounds, 477 U.S. 131, 152, 106 S.Ct. 2440, 2454-55, 91 L.Ed.2d 110 (1986).
The actual statement-that the Secretary “would be expected” to include “some form of depreciation”-is couched in vague and precatory terms. The statement is neither precise in its content nor directory in its thrust. In itself, the statement implies the existence of agency discretion. And, moreover, whatever force it may possess is diluted because the 1977 amendments employ language that very closely parallels the language of a previous regulation, on the same subject, that excluded depreciation. When Congress codifies language that has already been given meaning in a regulatory context, there is a presumption that the meaning remains the same. See Commissioner v. Keystone Consol. Indus., Inc., 508 U.S. 152, ---- - ----, 113 S.Ct. 2006, 2011-12, 124 L.Ed.2d 71 (1993) (explaining that Congress is presumed to be aware of settled judicial and administrative interpretations of words when it writes them into a statute); cf. Greenwood Trust Co. v. Massachusetts, 971 F.2d 818, 827 (1st Cir.1992) (noting that when Congress borrows a word from a legal source, the word usually brings along its prior judicial interpretations), cert. denied, 506 U.S. 1052, 113 S.Ct. 974, 122 L.Ed.2d 129 (1993).
We add two related points. First, we think the traditional infirmities that attend legislative history generally, see supra note 3 & accompanying text, are accentuated where, as here, its proponents focus on a single, isolated, somewhat tentative statement contained in a single document produced by a single chamber of a bicameral legislature. Second, although we recognize the uncertain value of post-enactment materials, we deem the provocative statements in the 1980 House-Senate conference report competent to lend an element of opacity to already murky waters.
To sum up, the legislative history underlying section 2014(d), though suggestive, is simply not definitive enough to suck the elasticity from the word “cost” and convey an “unambiguously expressed intent of Congress.” Chevron, 467 U.S. at 842-43, 104 S.Ct. at 2781.
The District Court's Slant
Nor do we share the district court's view that the Secretary's change in position forfeits any entitlement to deference because it occurred as a knee-jerk response to jawboning that the Secretary mistakenly perceived as a congressional mandate. In the first place, we know of no sufficient basis for disregarding the Secretary's characterization of the conference committee's report, which accompanied the 1982 regulations, as a “suggestion.” Hence, we credit this characterization. By like token, we are obliged, in the absence of meaningful impeachment, to accept the Secretary's stated reason for making the change-that households were, in a real sense, being given an unwarranted anticipatory deduction for the costs of replacing capital goods, and that the practice ought to be stopped-at face value. This is exactly the type of “reasoned explanation” that, when accompanying an agency's change of position, can evoke the deference contemplated by the Rust Court. When an agency explicates a principled basis for revising an interstitial rule in a plausible way, judges should not simply shrug it off.
We make one final point. Courts must not lightly assume that the Executive Branch is untutored, or that cabinet officers are dolts. Here, for example, we are reluctant to presume that the Secretary either overlooked or misread an unbroken skein of cases, see, e.g., Bowsher v. Synar, 478 U.S. 714, 721-27, 106 S.Ct. 3181, 3185-88, 92 L.Ed.2d 583 (1986) (reviewing caselaw and history of separation of powers); Rhode Island v. Narragansett Indian Tribe, 19 F.3d 685, 699 (1st Cir.) (acknowledging that “[i]n our republican form of government, legislators make laws by writing statutes”), cert. denied, 513 U.S. 919, 115 S.Ct. 298, 130 L.Ed.2d 211 (1994), and concluded that he was duty bound to rewrite the rule simply because the conference committee groused about it. We think it is much more realistic to infer that the conference committee's unredacted comments served as a wake-up call, sparking the sort of reexamination that the Rust Court explicitly sanctioned. In our tripartite system of government, inter-branch communication and cooperation are not terrible diseases, to be avoided at all costs, but, rather, are a tested means of improving the health of the body politic. Thus, evidence that such a rapport exists, without more, does not cast doubt on the validity of agency action.
Once we have concluded both that Congress has not spoken authoritatively on the precise question and that the Secretary's change of heart is not outside Chevron's precedential orbit, the remaining pieces of the puzzle fall neatly into place. Fairly read, the amended regulation is reasonable in light of the Act's avowed purpose of supplementing the purchasing power of those unable to afford nutritionally adequate diets.5 The idea that excluding depreciation from income more accurately reflects the ability of a family to purchase food and, thus, better indicates the need for food stamps, is hardly heretical. Implementing such an idea merely shifts the emphasis of the relevant measure from an accountant's conception of profit and loss to a layperson's conception of cash flow. Hence, we cannot conclude that the Secretary's handiwork, as expressed in the amended regulation, is “arbitrary, capricious, or manifestly contrary to the statute.” Chevron, 467 U.S. at 844, 104 S.Ct. at 2782.
The Stricklands themselves are a good illustration of why the Secretary's second thought makes perfectly good regulatory sense. In 1992 the Stricklands had revenue in hand of more than twice the amount designated as the “maximum” income for a family of two receiving food stamps. We believe that the Secretary could reasonably conclude that households having this degree of cash availability are no more in need of food stamps than families with half as much take-home pay-who are, by dint of their income, ineligible for participation in the food stamp program.
To be sure, the plaintiffs muster a cavalcade of contentions that point in the other direction. They argue that excluding depreciation gives a somewhat arbitrary preference to self-employed food stamp recipients who elect to rent, rather than purchase, business equipment, for rental payments are deductible. They also argue that failure to recognize depreciation discourages food stamp recipients from undertaking certain business activities because, if an attributed self-employment income overstates the real profit they receive, they will forfeit food stamp eligibility. While these may constitute valid arguments against the wisdom (or, more aptly put, the lack of wisdom) of the amended regulation, they do no more than show that both the plaintiffs' and the Secretary's readings of the word “cost” as it is used in the statute are imperfect. By the same token, however, both readings are plausible. In that situation, it is up to the Secretary, not the courts, to balance the relevant policy considerations and formulate a rule. The implementation of a statutory term that can reasonably accommodate two or more interpretations must be left to the agency.
We hold, therefore, that the amended regulation is an entirely permissible interpretation of the statute, and, as such, is an entirely permissible exercise of the Secretary's authority. Accord St. Amour v. Department of Social Welfare, 158 Vt. 77, 605 A.2d 1340 (1992) (considering identical issue and upholding the exclusion of depreciation under section 2014(d)).
We need go no further.6 The Secretary's decision to exclude depreciation from the cost of producing self-employment income is not inconsistent with the language and history of 7 U.S.C. § 2014(d)(9). Moreover, the decision is grounded in a reasonable interpretation of the statute. Since the amended regulation must be upheld under Chevron principles, the district court's contrary judgment is
1. The Stricklands say that they could not sell their construction equipment because they would no longer have any means of producing income. They acknowledge, however, that they owed more on some pieces of equipment than those pieces were likely to bring on the open market.
2. From and after the time that the Secretary answered the plaintiff's first amended complaint, he has pulled the laboring oar in defending the regulation.
3. Critics say, for example, that legislative history is written by staffers rather than by Congress itself; that it is easily manipulated; that it complicates the tasks of execution and obedience; and that it often is shaped by members of Congress who cannot achieve passage of a desired interpretation in the actual text of an enacted statute. See Matter of Sinclair, 870 F.2d 1340, 1342-44 (7th Cir.1989); see also Stephen Breyer, On the Uses of Legislative History in Interpreting Statutes, 65 S.Cal.L.Rev. 845, 845-47 (1991) (describing various attacks on legislative history, but defending its use when judges are faced with unclear statutory language).
4. Courts that exclude legislative history during the first stage of the Chevron analysis may well decide to consider it during the second stage in order to determine whether the agency's interpretation is a permissible one. Thus, compelling legislative history probably will preclude a contrary agency position under either of the two views of Chevron.
5. Indeed, the amended regulation, which uses “cost” in its lay sense rather than in the specialized plutonomic sense, may better serve the Act's purposes. In any given accounting period, depreciation is likely to have scant effect on cash flow-even its proponents must admit that depreciation is, at best, an approximation (that is, a guess) that exists primarily, if not exclusively, on paper-and it may bear little if any relation to an actual decrease in the value of a capital asset. Moreover, although depreciation may account for money set aside to replace a piece of equipment, there is no guarantee that such equipment will in fact be replaced with similar equipment or with any equipment at all.
6. This case does not require us to decide whether self-employed food stamp recipients must be given some alternative deduction, such as a deduction for replacement costs, in recognition of either the cost of acquiring capital goods or their consumption in the course of producing income.
SELYA, Circuit Judge.