This appeal arises out of Baltimore Gas & Electric Co.’s (BG & E) suit against Frank 0. Heintz, William A. Badger, Lilo K. Schifter, Wayne B. Hamilton, and Haskell N. Arnold, the commissioners of the Maryland Public .Service Commission (PSC),1 seeking both a declaratory judgment that Md.Ann.Code art. 78 § 24(e) (1980) is unconstitutional and an injunction prohibiting the future enforcement of that section. BG & E is a public utility organized under Maryland law with its principal place of business in Baltimore City, Maryland. It primarily supplies gas and electricity to customers in Baltimore City, Maryland. Its common stock is publicly traded on the New York Stock Exchange, and at the time of suit, 70% of its outstand*1412ing shares were owned by stockholders outside the state of Maryland.

BG & E was joined in its suit by BGE CORP, a Maryland corporation formed specifically to become the parent company of BG & E. BG & E proposed a share for share exchange of all the outstanding BG & E stock for stock in BGE CORP. The stock of Resource & Property Management (RPM), a wholly-owned subsidiary of BG & E, would also have been transferred to BGE CORP. As a result, BG & E and RPM would have become separate, wholly owned subsidiaries of BGE CORP; the holders of BG & E common stock would have become the holders of BGE CORP common stock. BG & E’s preference stock, First Mortgage Bonds, and other outstanding debt would have continued as obligations of BG & E. This share for share exchange of the stock of BG & E for the stock of BGE CORP was designed “to separate the utility and non-utility divisions of the company under the holding company structure, which would ‘facilitate the regulation of utility operations’ and allow the company’s non-utility divisions to diversify into activities other than the distribution of gas and electricity, without being held accountable to the PSC.” Baltimore Gas & Elec. Co. v. Heintz, 582 F.Supp. 675, 677 (D.Md. 1984). ■

BG & E and BGE CORP filed a joint application with the PSC requesting authorization for the latter company to acquire all the stock of the former. Pursuant to the Public Service Commission Law, Md. Ann.Code art. 78 § 20(a), the Commission appointed a panel to hold hearings on the proposed stock exchange. The People’s Counsel, an attorney appointed by the Governor to represent residential consumers in Commission proceedings,2 filed a motion to dismiss the Commission proceedings on the ground that art. 78 § 24(e) prohibited the proposed exchange. After various memoranda were filed and a hearing was held, the Commission held that the proposed transfer was prohibited by the statute. The PCS did not consider the merits of the proposed exchange because it interpreted section 24(e) to bar absolutely the exchange.

Despite the availability of state court review proceedings, see Md.Ann.Code §§ 89-98 (1980 & Supp. 1984), BG & E and BGE CORP challenged the constitutionality of section 24(e) in federal court. After BG & E filed a motion for summary judgment in its suit seeking declaratory and injunctive relief concerning the constitutionality of section 24(e) on the basis of the supremacy, commerce, due process, and equal protection clauses, the district court granted the People’s Counsel motion to intervene as a defendant. People’s Counsel and the PSC filed motions for summary judgment, and the Securities and Exchange Commission filed a memorandum as an amicus curiae. Various replies and responses were filed; and after a hearing on the cross motions, the district court held that section 24(e) unconstitutionally burdened interstate commerce in violation of the commerce clause. 582 F.Supp. at 681-82. Specifically, the district court ruled that “the ban resulting from the absolute prohibition of Section 24(e) is excessive in light of the less restrictive alternative — a law which would allow the formation of public utility holding companies, but only on the approval of the PSC.” Id. at 682. Relying on the amicus curiae brief of the Securities and Exchange Commission as “persuasive,” the district court held that section 24(e) was not pre-empted by the Public Utility Holding Company Act of 1935, (PUHCA) 15 U.S.C. § 79i(a)(2), which allows holding companies to be established and to acquire the stock of utilities with the approval of the SEC. The court determined that because section 24(e) was designed to protect rate payers, not investors, it did not conflict with the PUHCA, which was designed to control generally the activ*1413ities of holding companies by subjecting them to extensive SEC regulations. Id. at 679-80. Finally, the district court declined to address BG & E’s equal protection and due process claims.

From the judgment of the district court, the PSC and People’s Counsel appeal, arguing that the district court incorrectly characterized the state’s interest served by section 24(e) and that the court incorrectly balanced the state interest against the burden on interstate commerce. BG & E continues to rely on the commerce, equal protection, and due process clauses to argue that the statute is unconstitutional. In addition, it argues that section 24(e) violates the supremacy clause as it prohibits the execution of the full purposes and objectives of the PUHCA. Finding BG & E’s contentions unpersuasive, we reject its attack on the constitutionality of section 24(e). Accordingly, we reverse the district court.

I

Our first task when addressing the constitutionality of any statute is to determine the appropriate construction of the statutory language. Neither the federal courts nor the Maryland courts have construed this statute previously. When construing a statute for the first time, courts look first to the statute’s language, then to its legislative history, and finally to the interpretation given the statute by those charged with administering it. Brothers v. First Leasing, 724 F.2d 789, 792 (9th Cir. 1984). Section 24(e) provides:

(e) Holding, acquiring, etc., capital stock of public service company. —No public service company shall take, hold, or acquire any part of the capital stock of any public service company of the same class, organized or existing under or by virtue of the laws of this State, without prior authorization by the Commission. No stock corporation of any description (except, with the authorization of the Commission, a company of the same class) shall take, hold or acquire more than ten per cent of the total capital stock of any public service company organized or existing under or by virtue of the laws of this State, unless such stock is to be taken as collateral security and the Commission approves of its being so taken; and no such public service company shall be party to the violation of this subsection. For the purposes of this subsection, a company controlling a public service company shall be deemed a public service company of the same class as the controlled public service company.

There is no Maryland legislative history interpreting section 24(e). The predecessor to the current statute was enacted in 1910, 1910 Md.Laws ch. 180 § 35. That enactment was codified at Md.Pub.Div.Laws art. 23 § 449 (1911) and provided that

... No such corporation shall directly or indirectly acquire the stock or bonds of any other corporation incorporated for or engaged in, the same or similar business, or proposing to operate or operating under a franchise from the same or any other municipality, unless authorized to do so, by the Commission. Save where stock shall be transferred or held for the purpose of collateral security only with the consent of the Commission, no stock corporation of any description, domestic or foreign, other than a gas or electrical corporation, shall purchase or acquire, take or hold, more than ten per centum of the total capital stock issued by any gas corporation or electrical corporation organized or existing under and by virtue of the laws of this state. Nothing herein contained shall be construed to prevent the holding of stock heretofore lawfully acquired.

This statute was recodified several times and combined with similar statutes governing common carriers. In 1955, the Public Service Commission Law of Maryland was revised by 1955 Md.Laws ch. 441; section 23 of chapter 441 contained the same language as the current version, but no legis*1414lative history exists for any of these versions or the version under consideration.

In the absence of legislative history construing the statute, the interpretation of the administrative agency charged with enforcing the statute, the meaning of which is at issue, is entitled to at least “respectful consideration.” Schwartz v. Roanes, 495 F.2d 844, 853 (2nd Cir. 1974) (interpreting N.Y.Pub.Serv.Law § 107, McKinney’s Consol.Laws c. 48, which prohibited a public utility from using public service revenues for non-operating expenses without the PSC’s approval and reversing district court on basis of commission’s opinion) (quoting Law Students Civil Rights Research Council, Inc. v. Wadmond, 401 U.S. 154, 163, 91 S.Ct. 720, 726, 27 L.Ed.2d 749 (1971)). See also, Hart & Miller Islands v. Corps of Engineers, 621 F.2d 1281, 1290 (4th Cir. 1980) (following “venerable principle that the construction of a statute by those charged with its execution should be followed unless there are compelling indications that it is wrong”) (quoting Red Lion Broadcasting Co. v. F.C.C., 395 U.S. 367, 381, 89 S.Ct. 1794, 1801, 23 L.Ed.2d 371 (1969)). Despite the substantial deference accorded an agency’s interpretation, the courts retain the final say on the construction of a statute; and if the construction posited by the agency is inconsistent with the policy embodied in the statute, then the agency’s construction must be rejected in favor of one more reasonable. California v. Watt, 668 F.2d 1290, 1302-03 (D.C.Cir. 1981) (per curiam); United Food & Commercial Workers Int’l Union Local No. 576 v. N.L.R.B., 675 F.2d 346, 351 (D.C.Cir. 1982).

In this instance, the PSC’s interpretation 3 of section 24(e) in the regulatory proceeding that was the initiative for the suit below is eminently reasonable and consistent with the plain wording of the statute; and for that reason, we find its interpretation of the statute persuasive and adopt it as our own. A public service company may acquire the capital stock of a Maryland public service company of the same class if the transaction is approved by the PSC. A stock corporation that is not a public service company may acquire more than ten per cent of the capital stock of a Maryland public service company only if the stock is taken as collateral security for a debt of the public service company and only if the transaction is approved by the PSC. Under no other circumstances may a stock corporation that is not a public service company acquire more than ten per cent of the stock of a Maryland public service company. A stock corporation that controls a public service company is deemed to be a public service company of the same class as the controlled company and can acquire the capital stock of a Maryland public service company of the same class as the controlled company. With this interpretation in mind, we turn to BG & E’s challenges to the constitutionality of section 24(e).

II

Section 24(e) does not conflict with the PUHCA and thus does not violate the supremacy clause.4 The test for determining whether a state statute violates the supremacy clause is well established. Congress may pre-empt state authority to legislate within a certain area by explicitly so stating. Jones v. Rath Packing Co., 430 U.S. 519, 525, 97 S.Ct. 1305, 1309, 51 L.Ed.2d 604 (1977). Congress may also implicitly pre-empt state authority to regulate within an entire area by creating a federal regulatory scheme the structure and purpose of which is “so pervasive as to make reasonable the inference that Congress left no room for the States to supplement it.” Rice v. Santa Fe Elevator Corp., 331 U.S. 218, 230, 67 S.Ct. 1146, 1152, 91 L.Ed. 1447 (1947) (quoted in Pacific Gas & Elec. v. Energy Resources *1415Comm’n, 461 U.S. 190, 204, 103 S.Ct. 1713, 1722, 75 L.Ed.2d 752 (1983); Fidelity Federal Sav. & Loan Ass’n v. De la Cuesta, 458 U.S. 141, 153, 102 S.Ct. 3014, 3022, 75 L.Ed.2d 664 (1982)).

BG & E does not assert that Congress has completely pre-empted state regulation of public utility holding companies, and such an argument would be patently unconvincing in light of the provisions in the PUHCA that expressly contemplate coordinate state and federal regulation. “[T]he purpose of section 11 [15 U.S.C. § 79k] is simply to provide a mechanism to create conditions under which effective Federal and State regulation will be possible. It is therefore the very heart of the title, the section most essential to the accomplishment of the purposes set forth in the President’s message.” S.Rep. 621, 74th Cong., 1st Sess., p. 11 (quoted in North American Co. v. S.E.C., 327 U.S. 686, 704 n. 14, 66 S.Ct. 785, 796 n. 14, 90 L.Ed. 945 (1946)). By adopting 15 U.S.C. § 79k(b)(1), which permits the S.E.C. to require a public utility holding company to divest geographically and economically unrelated holdings, Congress “hoped to rejuvenate local utility management and to restore effective state regulation____” North Am. Co. v. S.E.C., 327 U.S. at 704, 66 S.Ct. at 795. See also 15 U.S.C. § 79h (states may prohibit, or otherwise regulate, a single company acquiring both a gas and an electric utility).

BG & E instead argues that the PUHCA preempts section 24(e) “because [section 24(e)] frustrates the accomplishment and execution of one of the purposes and objectives of the [PUHCA] — facilitating the continuance and creation of public utility holding companies within the regulatory framework created by that Federal Act when doing so is in the public interest — by making it impossible to establish a public utility holding company under Maryland law.” BG & E supports its assertion that the PUHCA was designed to facilitate public utility holding companies by pointing out that Congress rejected the initial version of the bill that eventually became the PUH-CA, which would have dissolved all public utility holding companies, and enacted instead a statute that regulated rather than eliminated public utility holding companies. Thus, asserts BG & E, because Congress decided to regulate rather than prohibit public utility holding companies, Maryland is prohibited from enacting a statute that would prohibit public utility holding companies. In other words, Congress considered the approach taken by Maryland and flatly rejected it; therefore, Maryland’s approach is prohibited by the PUHCA.

The flaw in BG & E’s argument is that Congress’ rejection of a statute that would completely outlaw public utility holding companies does not necessarily establish that public utility holding companies cannot be prohibited by the states. Furthermore, 15 U.S.C. § 79k(b)(2) explicitly authorizes the S.E.C. to dissolve registered holding companies if it finds their continued existence performs no justifiable function and is an unnecessary complexity enabling the holding company to perpetuate a pyramided system of ownership. American Power & Light Co. v. S.E.C., 329 U.S. 90, 104-06, 67 S.Ct. 133, 141-42, 91 L.Ed. 103 (1946). Thus, as the PUHCA contemplates that the S.E.C. may prohibit only certain public utility holding companies, the question becomes whether the authority vested in the Maryland PSC “stands as an obstacle to the accomplishment and execution of the full purpose and objectives of Congress” in prohibiting certain public utility holding companies and permitting others. Hines v. Davidowitz, 312 U.S. 52, 67, 61 S.Ct. 399, 404, 85 L.Ed. 581 (1941).

We hold that section 24(e) does not frustrate the purpose of the PUHCA. We find particularly persuasive the opinion of the S.E.C., presented in its amicus brief to the district court and incorporated in the PSC’s brief to this Court. The S.E.C. asserts that section 24(e) is not pre-empted by the PUH-CA because “Congress’ decision to regulate rather than eliminate holding companies reflects a legislative judgment in favor of regulatory restraint, and not an intent to guarantee the proliferation of utility holding companies.”

*1416Congress’ determination to do away with public utility holding companies unless certain conditions are met, see 15 U.S.C. § 79k(b)(l)(A)-(C) (holding company may continue if substantial economies will be secured; the controlled activities are geographically close; and localized management, efficient operation, and effective regulation will not be impaired), does not indicate a legislative intent that public utility holding companies shall be outside the reach of a prohibitory state statute. So long as the states are legislating constitutionally under other provisions of the Constitution, BG & E’s contentions about which we shall soon speak, Congress’ refusal to prohibit all public utility holding companies under the PUHCA does not preempt that legislation under the supremacy clause. Furthermore, the S.E.C. asserts that if BG & E were to apply to the S.E.C. for approval of the proposed transaction, it would deny the request on the basis of the PSC’s order. In the face of this interpretation by the agency charged with administering the PUHCA, the claim that section 24(e) is pre-empted is clearly baseless.5

Ill

BG & E argues that section 24(e) is unconstitutional under both the due process and equal protection clauses. Although we deem both arguments to be without merit, we feel compelled to address them briefly as the district court did not reach them. BG & E’s arguments are that section 24(e) violates the equal protection clause because it arbitrarily and irrationally defines who may and who may not acquire more than ten per cent of the stock of a Maryland public service company. BG & E also asserts that section 24(e) has been applied so arbitrarily and selectively as to violate the due process and equal protection clauses. We address these contentions seriatim.

The basis for the equal protection attack is that the statute allows a stock corporation that controls a public service company to acquire more than ten per cent of a public utility if the PSC approves, while prohibiting entirely a stock corporation that does not control a public service company from acquiring a public service company. The statute in that manner distinguishes between stock corporations that control a public service corporation and those that do not when establishing the conditions under which a public service corporation may be acquired. If the stock corporation does not control a public service corporation, it may not, under any circumstances, acquire more than ten per cent of a public service corporation. If the stock corporation already controls a public service corporation of the same class as that which it proposes to acquire, then it may acquire more than ten per cent of the public service company if the PSC grants its approval. We are not persuaded that the legislative distinction between stock corporations that presently control a public service corporation and stock corporations that do not control public service corporations is .wholly arbitrary and without a rational basis. Minnesota v. Clover Leaf *1417Creamery Co., 449 U.S. 456, 464, 101 S.Ct. 715, 723, 66 L.Ed.2d 659 (1981) (challenger bears burden of proving unconstitutionality).

Our review of the classification under the equal protection clause is two pronged: First we must determine whether the challenged legislation has a legitimate purpose. Then, we must determine whether it was reasonable for the state legislature to believe that the classification would promote that purpose. Western & Southern L.I. Co. v. Board of Equalization, 451 U.S. 648, 668, 101 S.Ct. 2070, 2083, 68 L.Ed.2d 514 (1981). Although BG & E does not distinguish between these two distinct questions, it argues that the classification is arbitrary (i.e., the statute has no legitimate purpose) and irrational (i.e., the Maryland legislature could not rationally believe that the classification chosen would achieve that purpose).

At the outset, we are mindful of the court’s limited review of state legislative decisions:

When local economic regulation is challenged solely as violating the Equal Protection Clause, this Court consistently defers to legislative determinations as to the desirability of particular statutory discriminations, [citation omitted] Unless a classification trammels fundamental personal rights or is drawn upon inherently suspect distinctions such as race, religion or alienage, our decisions presume the constitutionality of statutory discriminations and require only that the classification challenged be rationally related to a legitimate state interest. States are accorded wide latitude in the regulation of their local economies under their police powers, and rational distinctions may be made with substantially less than mathematical exactitude.

New Orleans v. Dukes, 427 U.S. 297, 303, 96 S.Ct. 2513, 2516, 49 L.Ed.2d 511 (1976). See also Western & S. L.I. Co., 451 U.S. at 670, 101 S.Ct. at 2084 (courts “not empowered to second-guess the wisdom of state policies”); Minnesota v. Clover Leaf Creamery Co., 449 U.S. at 464, 466, 469, 470, 101 S.Ct. at 723, 725, 726, 727 (the question for courts is whether the classification is a rational means of achieving state goal; not whether the classification in fact achieves that goal); Dandridge v. Williams, 397 U.S. 471, 485, 90 S.Ct. 1153, 1161, 25 L.Ed.2d 491 (1970).

There can be no doubt that the purpose of section 24(e) is legitimate. The primary justification for the state’s authority to regulate public utilities under the police power, see Bridgeport Hydraulic v. Council on Water Company Lands, 453 F.Supp. 942, 951 (1977) aff'd without opinion, 439 U.S. 999, 99 S.Ct. 606, 58 L.Ed.2d 674 (1978), is to ensure continued fiscal responsibility and the ability of the utility to continue to render service in order to protect consumers of the public service company’s product. See Pacific Gas & Elec. v. Energy Resources Conservation & Dev. Comm’n, 461 U.S. 190, 205-07, 103 S.Ct. 1713, 1722-24, 75 L.Ed.2d 752 (1983). Cf. Note, The Servicing Function of Public Utility Holding Companies, 49 Harv.L.Rev. 957, 978-82 (1936) (documenting potential for abuse in holding company structure by over-charging for services). Section 24(e) is clearly intended to limit the formation of public utility holding companies; that purpose is certainly legitimate in light of the extensive regulation by the federal government and Congress’ expectation in the PUHCA that the states would continue to exercise a certain amount of regulatory authority. Thus, the only remaining question is whether the state’s classification of holding companies and absolute prohibition of one of those classes is a rational means toward eradicating the abuses that are inherent in the holding company structure.

The legislature could rationally conclude that in order to eliminate holding company abuse of public utilities no holding companies should be permitted to acquire a public utility unless that holding company already owns a public utility. By limiting the class of holding companies that are eligible to acquire a public utility to those companies that own public utilities and by subjecting *1418those companies to public service commission review, the state ensures that it has an available record upon which to determine whether the proposed acquisition will create the potential for abuse. By prohibiting those holding companies that do not own public utilities from acquiring a public utility, the state further ensures that a holding company about which the state is unable to form a conclusion due to lack of information does not acquire a public utility. In addition, the legislative scheme recognizes that where a holding company already owns a utility of the same class, there are possible utilities of scale that may outweigh any detriment of the holding company structure. Thus the exception allows the PSC to consider whether those economies of scale outweigh the potential for abuse in each particular case. If a holding company owns no other public utilities of the same class, however, there are no potential economies of scale; therefore, the state acted rationally in imposing an absolute bar.6

BG & E also challenges the application of section 24(e) to bar the proposed share exchange on the ground that the PSC has not consistently applied the statute thereby violating due process. We reject that argument. BG & E has cited several examples of what it claims are acquisitions of a public utility by a holding company that does not own a public utility of the same class. Because the PSC did not bar those substantially similar acquisitions on the basis of section 24(e), it is argued, it is an unconstitutional deprivation of due process (and equal protection) to prohibit BGE CORP’s proposed acquisition of BG & E. The PSC urges us to analyze those other transactions and determine that they are not, in fact, substantially similar. Although the district court declined to rule on BG & E’s due process argument, it did delve into the intricacies of the various acquisitions and determined that only one other transaction — the acquisition of the Allegheny Power System, Inc. by the Potomac Edison Company — that the PSC permitted to proceed was prohibited by the letter of the statute. 582 F.Supp. at 678. We find it difficult to perch a deprivation of due process upon so slender a reed, but that the statute has been applied nondiscriminatorily does not form the basis of our holding.

We hold that BG & E has failed to satisfy the elements of a claim of selective enforcement. It is a well-settled proposition of administrative law that when an agency deviates from established precedent, it must provide a reasoned explanation for its failure to follow its own precedents. The leading case is Secretary of Agriculture v. United States, 347 U.S. 645, 74 S.Ct. 826, 98 L.Ed. 1015 (1954). This requirement does not mean that an agency may not change its policies, but rather that if the agency wishes to overrule or distinguish its precedents, it must explain why it has acted as it did. It follows from that general proposition that when an agency treats two similar transactions differently, an explanation for the agency’s actions must be forthcoming. Roadway Exp., Inc. v. N.L.R.B., 647 F.2d 415, 418-19 (1981); Contractors Transp. Corp. v. United States, 537 F.2d 1160, 1162 (1976).

But that general proposition is inapplicable to this case. That an agency is required to present a reasoned analysis when departing from established precedents is not required as a matter of constitutional due process, see W. Gellhorn, C. Byse, & P. Strauss, Administrative Law 393 (7th ed. 1979) but is required “so that the reviewing court may understand the basis of the agency’s action and so may *1419judge the consistency of that action with the agency’s mandate.” Topeka & Santa Fe Ry. Co. v. Wichita Bd. of Trade, 412 U.S. 800, 808, 93 S.Ct. 2367, 2375, 37 L.Ed.2d 350 (1973). One legal basis for this requirement is the statutory requirement in the Administrative Procedure Act, 5 U.S.C. § 557(c) (1984), that the agency provide “findings and conclusions, and the reasons or basis therefor, on all the material issues of fact, law, or discretion.” See Morton v. Ruiz, 415 U.S. 199, 231-34, 94 S.Ct. 1055, 1072-74, 39 L.Ed.2d 270 (1974). If the agency fails to fulfil this requirement, then it may be deemed to have acted arbitrarily. The problem with applying this analysis to the case sub judice is that this case is not a direct review of an agency action; BG & E declined to pursue judicial review in the Maryland courts. (Although the proceedings of the Maryland PSC are not subject to the Maryland Administrative Procedure Act, Md.Code State Gov’t Ann. §§ 10-202(a)(3)(vii), 10-302(a)(7) (1984), the Maryland Public Service Commission Law provides for judicial review. Md.Code Ann. art. 78 § 89-98 (1980 & Supp. 1984)).

In order to prove a claim of selective enforcement sufficient to invalidate the actions of an administrative agency as violative of constitutional due process, BG & E must show that the decision to bar its proposed stock exchange was based on some constitutionally impermissible criterion, such as race, sex, or exercise of first amendment rights. So long as an agency is not determining whether to enforce its regulations on the basis of some impermissible constitutional criterion, it is not violating the dictates of due process. The mere determination to enforce the statute or regulation in some cases and not to enforce in others, without more, may entitle one to an explanation from the agency designed to facilitate judicial review of the decision, but such an action does not constitute a deprivation of due process. That standard has “generated an almost unbroken line of cases upholding prosecutors’ powers to decide who and how to charge.” Vorenberg, Decent Restraint of Prosecutorial Discretion, 94 Harv.L.Rev. 1521, 1540 n. 71 (1981) (citing cases). See Wayte v. United States, — U.S. -,-, 105 S.Ct. 1524, 1531, 84 L.Ed.2d 547 (1985) (claim of selective enforcement requires “show[ing] that government prosecuted because of” impermissible factors).

We emphasize that we need not determine what showing is necessary to establish a due process violation by virtue of discriminatory enforcement in a civil, administrative proceeding. The PSC has asserted, and BG & E has expressly accepted, that in those cases in which the PSC permitted a transaction that might have violated the statute to proceed, the section 24(e) ban on the acquisition of more than ten per cent of a public utility by a holding company not owning a public utility of the same class was neither presented to nor addressed by the commission. We hold only that in the absence of some showing of impermissible intent or motive on the part of the administrative agency charged with enforcing the law, the mere decision of the agency to enforce the law against an individual cannot support a claim of selective enforcement in an independent action challenging only the constitutionality of the statute.

“An agency generally cannot act against each technical violation of the statute it is charged with enforcing.” Heckler v. Chaney, — U.S. -,-, 105 S.Ct. 1649, 1654, 84 L.Ed.2d 714 (1985). The agency possesses substantial discretion when deciding whether to enforce or not to enforce a particular law in a particular circumstance, and the proper forum within which to review the exercise of that discretion is an appeal of the agency’s decision to the appropriate court. See Dunlop v. Bachowski, 421 U.S. 560, 566-68, 95 S.Ct. 1851, 1857-58, 44 L.Ed.2d 377 (1975) (setting aside decision of Secretary of Labor not to prosecute); Environmental Defense Fund v. Ruckelshaus, 439 F.2d 584, 598 (D.C.Cir. 1971) (“Judicial review must operate to ensure that the administrative process itself will confine and control the exercise of discretion.”) When the exercise *1420of such agency discretion is challenged under the due process clause in an independent declaratory action regarding the constitutionality of the statute as applied in a particular instance, we are not reviewing the agency’s decision on the basis of the administrative “arbitrary and capricious” standard that governs agency action. In order to set forth a constitutional claim, BG & E must show some influence of a constitutionally impermissible nature that caused the PSC to decide to enforce in this case or not to enforce in others.7 BG & E admits that this is the first case in which the PSC has expressly considered the application of section 24(e); although other proposed transactions might have raised the issue, the question was never presented to or addressed by the PSC. That admission necessarily forecloses the showing of any impermissible motive, this being the first time the PSC has invoked the statute.

IV

The final issue that we must address is BG & E’s commerce clause8 challenge to section 24(e). Although the commerce clause explicitly grants authority only to Congress, the Supreme Court long has held that the commerce clause limits the authority of the states to interfere with interstate commerce. See South-Central Timber Development, Inc. v. Wunnicke, — U.S. -,-, 104 S.Ct. 2237, 2239, 81 L.Ed.2d 71, 76 (1984); Cooley v. Board of Wardens, 53 U.S. (12 How.) 299, 13 L.Ed. 996 (1852). Traditionally, the Supreme Court has applied a bifurcated test to determine whether a particular state enactment violates the commerce clause. For state statutes that directly regulate interstate commerce, the Court has formulated a virtual per se rule of invalidity. See Edgar v. MITE Corp., 457 U.S. 624, 640, 102 S.Ct. 2629, 2640, 73 L.Ed.2d 269 (1982) (plurality opinion of White, J., joined by Burger, C.J., Stevens, J., and O’Connor, J.); see also South-Central Timber, — U.S. at-, 104 S.Ct. at 2247, 81 L.Ed.2d at 85 (protectionist statute that burdens interstate commerce “falls within rule of virtual per se invalidity of laws that ‘block the flow of interstate commerce at a State’s borders.’ Philadelphia v. New Jersey, 437 U.S. 617, 624, 98 S.Ct. 2531, 2535, 57 L.Ed.2d 475 (1978)”). A state statute that seeks evenhandedly to effectuate a legitimate state interest and whose effects on interstate commerce are incidental has been generally upheld unless the burden on interstate commerce is clearly excessive in relation to the putative local benefits. Pike v. Bruce Church, Inc., 397 U.S. 137, 142, 90 S.Ct. 844, 847, 25 L.Ed.2d 174 (1970).

In a recent opinion, however, the Court has indicated that it may be no longer adhering to the direct/indirect distinction. In Arkansas Elec. Coop. v. Arkansas Pub. Serv. Comm’n, 461 U.S. 375, 103 S.Ct. 1905, 76 L.Ed.2d 1 (1983), the Court, when evaluating the constitutionality of a state regulation that asserted the authority to regulate wholesale rates for the sale of electricity, stated that “it is difficult to square [a prior line of cases,] based on a supposedly precise division between ‘direct’ and ‘indirect’ effects on interstate commerce, with the general trend in our Commerce Clause jurisprudence to look in every case to ‘the nature of the state regulation involved, the objective of the state, and the effect of the regulation upon the national interest in the commerce.’ ” 461 *1421U.S. at 890, 103 S.Ct. at 1915 (quoting Illinois Gas Co. v. Public Serv. Co., 314 U.S. 498, 505, 62 S.Ct. 384, 386, 86 L.Ed. 371 (1942)). The Court has justified this new approach on the ground that the interests of the nation and the states often conflict in varying ways and to varying extents and that the most principled method of reconciling the competing interests is through an explicit balancing process.

The arguments of the parties in this appeal present us squarely with the question whether the Court has abandoned completely the distinction between direct and indirect burdens on interstate commerce for purposes of commerce clause analysis. BG & E argues that section 24(e) directly burdens interstate commerce, and is therefore virtually per se unconstitutional, and that section 24(e)’s indirect burden on interstate commerce is excessive in relation to the state interest served by the statute. The PSC urges us to recognize that the direct burden strand of the traditional commerce clause analysis has been abandoned and to apply only the general balancing test of Pike v. Bruce Church, Inc.

We decline to hold that the Supreme Court has jettisoned an established analytical framework for commerce clause analysis on the basis of Arkansas Electric Coop. Rather, we leave that quandary to be resolved explicitly by the Court9. This case, however, illustrates the difficulties of applying the direct/indirect distinction. Section 24(e) does indeed burden interstate commerce by limiting BG & E’s ability to consummate a stock transaction having a substantial effect on interstate commerce. The vast majority of BG & E stockholders are not Maryland residents and would therefore transfer their shares in BG & E for shares in BGE CORP through the channels of interstate commerce. BG & E is traded on the New York Stock Exchange, and thus the PSC’s prohibition of the proposed share exchange has obvious consequences on interstate commerce.

Despite these obvious burdens, we are reluctant to characterize them as “direct” or “indirect.” The burdens may be called “direct” in that section 24(e) erects an absolute bar to the consummation of what would otherwise be a legitimate interstate transaction. The burdens may also be called indirect in that section 24(e) is not designed to regulate share exchanges in interstate commerce which partially occur in Maryland, but is designed to protect Maryland consumers from the abuses inherent in the public utility holding company structure. Any consequences that flow from that attempt at consumer protection are incidental to the purpose of the statute and therefore indirect. BG & E recognizes in its brief to this Court that the burden on interstate commerce resulting from the consequential prohibition of stock transfers may be characterized as either direct or indirect. After arguing that the statute imposes a direct burden on interstate commerce, it states: “Moreover, the previously described direct burdens on interstate commerce also constitute substantial indirect burdens____”

Of course, if we label the burden “direct,” the result is practically foreordained: the statute is virtually per se unconstitutional. If we label the burden “indirect,” then we must balance the state interest against the burden on the national interest in interstate commerce. Rather *1422than engage in what we regard to be an analytically unsound and result-oriented approach in this case 10, we hold that the test for the constitutionality of a state statute which is challenged under the commerce clause is as follows: “Where the statute regulates evenhandedly to effectuate a legitimate local public interest, and its effects on interstate commerce are only incidental, it will be upheld unless the burden imposed on such commerce is clearly excessive in relation to the putative local benefits.” Pike v. Bruce Church, Inc., 397 U.S. at 142, 90 S.Ct. at 847; see also South-Central Timber Development, Inc. v. Wunnicke, — U.S. at-, 104 S.Ct. at 2247, 81 L.Ed.2d at 85.

That we adopt this test does not necessarily imply that the direct/indireet distinction is no longer applicable to commerce clause analysis. When a state statute does not regulate evenhandedly, i.e., imposes on “out-of-state commercial interests the full burden” of pursuing even a legitimate state goal, or when “one State [attempts] to isolate itself from a problem common to many by erecting a barrier against the movement of interstate trade,” the statute will be found to be virtually per se unconstitutional. Philadelphia v. New Jersey, 437 U.S. 617, 628, 98 S.Ct. 2531, 2537, 57 L.Ed.2d 475 (1978).11 Cf. Bacchus Imports, Ltd. v. Dias, — U.S. -,-, 104 S.Ct. 3049, 3055-57, 82 L.Ed.2d 200, 208-11 (1984) (striking down state statute exempting certain local alcoholic beverages from liquor tax as economic protectionism on ground that it was designed to discriminate in favor of local commodities). As the language quoted in the margin makes clear, see supra note 11, the requirement of evenhandedness preserves, albeit in a different formulation, the prohibition on direct burdens on interstate commerce. “Over the years the Court has used a variety of formulations for the Commerce Clause limitation upon the States, but it consistently has distinguished between outright protectionism and more indirect burdens on the free flow of trade. The Court has observed that ‘where simple economic protectionism is effected by state legislation, a virtually per se rule of invalidity has been erected.’ [citation omitted] In contrast, legislation that visits its effects equally upon both interstate and local business may survive constitutional scrutiny if it is narrowly drawn.” Lewis v. B T Investment Managers, Inc., 447 U.S. 27, 36, 100 S.Ct. 2009, 2015, 64 L.Ed.2d 702 (1980).

We find that section 24(e) regulates evenhandedly and is thus subject to the Pike balancing approach. The statute, in neither design nor effect, discriminates in favor of local commerce or against out-of-state commerce. There is no doubt that discriminatory state statutes that are enacted for protectionist purposes are consistently struck down, but an examination of those statutes indicates the dissimilarity of section 24(e) to those unconstitutional statutes. In Minnesota v. Clover Leaf Creamery Co., 449 U.S. 456, 471, 101 S.Ct. 715, 727, 66 L.Ed.2d 659 (1981), the Court upheld a state statute banning the retail sale *1423of milk in plastic non-returnable, non-refillable containers but permitting the sale in other non-returnable, non-refillable containers, such as paperboard cartons. Because the statute prohibited the sale of the plastic containers regardless of whether the milk, the container, or the seller was from inside or outside the state, the Court found the statute to regulate evenhandedly. In Lewis v. B T Investment Managers, Inc., 447 U.S. 27, 44,100 S.Ct. 2009, 2019, 64 L.Ed.2d 702 (1980), the Court struck down as violative of the commerce clause a Florida statute that prohibited investment advisory services by bank holding companies with principal offices outside the state. Although the Court declined to find the Florida statute per se invalid, finding instead that the disparate treatment of out-of-state bank holding companies could not be justified by legitimate local concerns, 447 U.S. at 42-44, 100 S.Ct. at 2018-19, the Court found the chief constitutional difficulty with the statute to be that it discriminated against out-of-state bank holding companies. Finding that discrimination “highly significant” for purposes of Commerce Clause analysis, the Court found that “discrimination against affected business organizations is not evenhanded because only banks, bank holding companies, and trust companies with principal operations outside Florida are prohibited from operating investment subsidiaries or giving investment advice within the state.” 447 U.S. at 42, 100 S.Ct. at 2018 (emphasis in original). Because the statute “discriminates among affected business entities according to the extent of their contacts with the local economy”, it “displays a local favoritism or protectionism that significantly alters its Commerce Clause status.” Id.

The statute at issue here cannot be characterized as economic protectionism. It is not designed to favor local commerce over out-of-state commerce and imposes no burden on interstate commerce that it does not impose on intrastate commerce. Thus, we find that it regulates evenhandedly. Section 24(e) prevents the acquisition of more than ten per cent of the stock of only Maryland public service companies. Thus, as the district court found, “Maryland’s own corporations are subjected to more profound restrictions than are out-of-state corporations and economic protectionism is not an issue.”12 582 F.Supp. at 681.

Finding that section 24(e) regulates evenhandedly, we must now determine whether the statute effectuates a legitimate state interest13 and whether the burden imposed *1424on interstate commerce is excessive in light of that burden. We begin our analysis with an evaluation of the state interest protected by section 24(e). We have no doubt that the interest served is legitimate. The Supreme Court has stated that “the regulation of utilities is one of the most important of the functions traditionally associated with the police power of the States.” Arkansas Electric, 461 U.S. 375, 377, 103 S.Ct. 1905, 1908, 76 L.Ed.2d 1 (1983). In this case, section 24(e) is but one subsection of an elaborate public service corporation law, one of the primary purposes of which is to regulate the rates charged to the public by the utility and which involves a balancing of the investor and consumer interests.14 Potomac Edison Co. v. Public Serv. Comm’n., 279 Md. 573, 581, 369 A.2d 1035, 1041 (1977) (quoting Power Comm’n v. Hope Gas Co., 320 U.S. 591, 603, 64 S.Ct. 281, 288, 88 L.Ed. 333 (1944)). A necessary adjunct to ensuring the protection of consumers is the authority to regulate the corporate structure of public utilities, and this authority has been recognized as such by the Supreme Court. Great Northern Ry. Co. v. Washington, 300 U.S. 154, 160, 57 S.Ct. 397, 400, 81 L.Ed. 573 (1937).

The potential for abuse of public service companies, which are only “legal” monopolies, are familiar and well documented. The primary purposes of the PUHCA were “to curb abusive practices of public utility companies by bringing them under effective control and to provide effective federal regulation of the expanding business of transmitting and selling electric power in interstate commerce.” Gulf States Utilities Co. v. Federal Pwr. Comm’n, 411 U.S. 747, 758, 93 S.Ct. 1870, 1877, 36 L.Ed.2d 635 (1973). The legislative history of that statute recognizes, as do the commentators, that holding companies provide the occasion for deceptive financing practices, *1425nondisclosure of important corporate accounts, and the manipulation of various “service charges” by the holding companies which increase the utility’s costs and the ultimate charge to the consumer. See S.Rep. No. 621, 74th Cong., 1st Sess. 14; H.R.Rep. No. 1318, 74th Cong., 1st Sess. 5; Note, The Servicing Function of Public Utility Holding Companies, 49 Harv.L. Rev. 957, 978-82 (1936). Prior to the passage of the PUHCA, holding company systems routinely owned or operated subsidiaries in speculative non-utility enterprises, Ritchie, Integration of Public Utility Holding Companies 14 (1954), a particularly abusive practice at which section 24(e) is aimed. Against this background, it is clear that the state has a legitimate interest in regulating the structure of public utilities and their holding companies.15

BG & E argues that accepting the interest of the state in controlling the corporate structure of public utility holding companies, the state’s interest “no longer has any validity” because all the potentials for abuse traditionally associated with holding companies have been remedied by either federal or state legislation. That argument proves too much. The state is entitled to exercise its police power in the manner it sees fit, so long as it is legislating constitutionally. That the state has enacted a complex regulatory structure with overlapping provisions, more than one of which may serve ultimately to control suspect activity, does not vitiate the state’s otherwise legitimate interest in curbing such suspect activity. The state may regulate the rates utilities charge consumers either directly, by requiring commission approval of rate increases, or indirectly, by controlling certain investments and attempts at diversification by the utility. That is, the state may choose one or more means to a particular end, and the legitimacy of the end — the state interest served by the particular regulation — is not vitiated when the state chooses several means of achieving that end. To hold otherwise would permit a utility to choose which of several overlapping regulations with which it will comply and allow the utility to challenge as unconstitutional those that it finds unduly burdensome. Therefore, we hold that the legitimacy of the state’s interest in preventing holding company abuse is not affected by other regulations that may achieve the same effect as section 24(e).

Having addressed the legitimacy of the state interest served by section 24(e), we now turn to the burden that that statute places on interstate commerce. We find that the district court correctly determined the burden section 24(e) places on interstate commerce.16

The burden of Section 24(e) on interstate commerce includes all those effects which naturally spring from the prohibition on reorganizing the company into a holding company structure: preventing the company from diversifying into areas outside the control of the PSC, affecting BG & E’s ability to secure financing, [and] prohibiting BG & E’s stockholders from exchanging their shares of BG & E for shares of BGE CORP. 582 F.Supp. at 682.

At this point, however, our analysis departs from that of the district court. We find that this minimal burden on interstate commerce, necessitated by the very notion of control of a public utility’s corporate structure and investments, is outweighed by the state’s interest in protecting the consumers of the products of public utilities. The district court, “weighting] the effect on interstate commerce against the state interest served by Section 24(e) [,]____ [found] that the burden on interstate commerce is excessive, particularly in light of the Court’s finding ... that the *1426purported state interests could be ‘promoted as well with a lesser impact on interstate activities.’ ” 582 F.Supp. at 681-82. The district court characterized the burden on interstate commerce as “not so overwhelming ... that the state Public Service Commission could not prohibit the formation of the holding company after a thorough study and reasoned decision, [but] the ban resulting from the absolute prohibition of Section 24(e) is excessive in light of the less restrictive alternative — a law which would allow the formation of public utility holding companies, but only on the approval of the PSC.” 582 F.Supp. at 682. The evils associated by the PSC with the holding company structure could be addressed just as adequately through a permissive procedure in which PSC approval is required for the transfer of the assets from the regulated utility to the proposed holding company.” Id. at 679-80.

As authority for this analysis, the district court and BG & E rely on three cases: Edgar v. MITE Corp., 457 U.S. at 644, 102 S.Ct. at 2642; Telvest v. Bradshaw, 697 F.2d 576, 581 (4th Cir. 1983); and Martin-Marietta Corp. v. Bendix, 690 F.2d 558, 566 (6th Cir. 1982). Those cases all involve challenges to state anti-takeover legislation, and in each case, the statute was struck down as an impermissible burden on interstate commerce. Edgar v. MITE Corp,, 457 U.S. at 646, 102 S.Ct. at 643 (opinion of White, J. joined by Burger, C.J., and Powell, J.); id. at 654-55, 102 S.Ct. at 2647-48 (Stevens, J., concurring in part and concurring in judgment); id. at 655, 102 S.Ct. at 2648 (O’Connor, J., concurring in part); Telvest, 697 F.2d at 582; Martin-Marietta, 690 F.2d at 568. The cited portions of these cases all deal with virtually identical statutory provisions. The three statutes were purported to protect investors from diminution of the value of their shares in a forced take-over. All three courts held that although this was a legitimate state objective, that objective was “diluted by,” (Telvest, 697 F.2d at 581) “undermined by,” (Martin-Marietta, 690 F.2d at 366-67) or “at variance with” (Edgar, 457 U.S. at 644, 102 S.Ct. at 2642) an exception in the statute for a corporation’s acquisition of its own shares. That is, although the anti-takeover statutes were designed to protect shareholders, that protection was not provided when a corporation makes an offer for its own shares, because such corporation was not required to comply with any of the state statute’s formalities. Thus, the courts reasoned that if the statute did not protect shareholders from possible abuse in all the relevant circumstances, (that is, protect shareholders from an unfair purchase by outsiders and the corporation itself) then the statutory purpose was not truly served.

Contrary to the district court’s reasoning, the statutory scheme of the Maryland public service law does not undermine the purported state interest in protecting consumers. Section 24(e) prohibits absolutely the acquisition of more than ten per cent of the shares of a public service company but it excepts from that prohibition an acquisition of more than ten per cent of the shares of a public service utility by a public utility holding company that owns a public utility of the same class as that acquired so long as PSC approval is obtained prior to the acquisition. That exception, unlike the exceptions in the anti-takeover statutes, is not at variance with or contrary to the asserted state interest in protecting the consumer of the goods and services provided by public utilities, because the transaction that would otherwise be subject to the statute is not completely exempted from the regulatory scheme. Rather, by requiring PSC approval prior to the consummation of the transaction, section 24(e) ensures the protection of the consumer in much the same way as an absolute ban. What the exception does is to allow the PSC to determine whether the potential for abuse present in holding company ownership of a public utility is outweighed by economies of scale or production when the acquiring public utility already owns a public utility of the same class. In this way, the statutory scheme protects the public from the potential for abuse in the holding company structure and at the same time *1427recognizes that, in a limited context, the potential harm to consumers may be outweighed by other considerations — economies of scale or production. We are unprepared to hold that a statutory scheme that provides for consideration of countervailing factors when those factors may be applicable imposes an impermissible burden on interstate commerce.

An analysis of the final component — nondiscriminatory alternative — of the Pike balancing approach does not convince us that section 24(e) is an unconstitutional burden on interstate commerce. “When discrimination against commerce ... is demonstrated, the burden falls on the state to justify it both in terms of local benefits flowing from the statute and the unavailability of nondiscriminatory alternatives adequate to preserve the local interests at stake.” Hunt v. Washington Apple Advertising Comm’n, 432 U.S. 333, 353, 97 S.Ct. 2434, 2446, 53 L.Ed.2d 383 (1977). When as in this case, however, the challenged statute does not discriminate against interstate commerce but regulates evenhandedly, the court does not inquire strictly into whether alternate nondiscriminatory means of protecting the state interest are available. Hughes v. Oklahoma, 441 U.S. 322, 337, 99 S.Ct. 1727, 1736, 60 L.Ed.2d 250 (1979); Hunt v. Washington Apple Advertising Comm’n, 432 U.S. 333, 353, 97 S.Ct. 2434, 2446, 53 L.Ed.2d 383 (1977). The choice among nondiscriminatory alternatives is a decision to be made by the state legislature. So long as the alternative chosen by the state does not discriminate against out of state commerce, then the court will not find the state’s selection to be an unconstitutional burden on interstate commerce because other equally nondiscriminatory means of achieving the state’s purpose exist. Only when the challenged statute in some way discriminates against interstate commerce will the court engage in an analysis of alternate means. See Hughes v. Oklahoma, 441 U.S. 322, 336-38, 99 S.Ct. 1727, 1736-37, 60 L.Ed.2d 250 (1979) (invalidating statute prohibiting sale or transportation of Oklahoma minnows outside state, a statute “that most overtly discriminates against interstate commerce”); Hunt v. Washington State Apple Advertising Comm ’n, 432 U.S. 333, 350, 353-54, 97 S.Ct. 2434, 2445, 2446-47, 53 L.Ed.2d 383 (1977) (invalidating statute requiring apples shipped in North Carolina be identified by only federal grade or no grade, a statute that “has the practical effect of not only burdening interstate sales of Washington apples, but also discriminating against them); Dean Milk Co. v. Madison, 340 U.S. 349, 354, 71 S.Ct. 295, 297, 95 L.Ed. 329 (1951) (invalidating ordinance prohibiting sale of pasteurized milk unless pasteurized within five miles of city, an ordinance that “plainly discriminates against interstate commerce”).

For a court to analyze alternative nondiscriminatory means when the state has chosen a means of achieving the state’s interest that is itself nondiscriminatory would be to do no more than substitute our judgment as to appropriate means for that of the state legislature. The alternative suggested by BG & E does not appear to be any less burdensome on interstate commerce. In fact, BG & E does not suggest an alternative to prohibiting the acquisition of more than ten per cent of the stock of a public utility by a Maryland holding company, but. rather urges that before enforcing any such prohibition, it be given a hearing at which to argue that such an acquisition should be permitted. The ultimate burden in either case would be similar. Thus, we hold that BG & E has not shown an approach with a lesser impact on interstate commerce.

Accordingly, we conclude that the district court must be reversed. Section 24(e) is not pre-empted by the PUHCA. The statute does not violate the equal protection or due process clauses. And finally, the statute does not impose an unconstitutional burden on interstate commerce. The judgment of the district court is reversed and the case is remanded to the district court for entry of judgment in favor of the PSC and other appellants.

REVERSED and REMANDED.