Constitutionality of Proposed Budget Process Reform Legislation

Constitutionality of Proposed Budget Process Reform Legislation Proposed legislation that would assign the Congressional Budget Office the duty to determine whether a spending bill would exceed current spending limits, thereby requiring a supermajority (two-thirds) vote in each House o f Congress for passage, is constitutional. Such a delegation would not raise problems under INS v. Chadha, because Congress may by rule require a supermajority majority vote in each House for passage of certain legislation under Art. I, § S, cl. 2. The proposed legislation may also subject spending bills passed in this manner to rescission by the President. W ith respect to entitlements, however, Congress must enact legislation specifi- cally making the expenditure of a certain percentage of the appropriated funds non-mandatory before such rescission authority may be exercised. May 26, 1987 M e m o r a n d u m O p i n i o n f o r t h e C o u n s e l t o t h e P r e s i d e n t At the request of your staff, this Office has considered the constitutionality of draft legislation, prepared by the White House Working Group on Budget Reform, entitled the “Budget Process Reform Act of 1987.” We are satisfied that the basic process that the bill would establish would be constitutional. The following comments suggest ways certain specific provisions of the bill might be changed in order to avoid or minimize possible constitutional issues. I. Determinations by the Congressional Budget Office A central feature of the draft bill is the assignment (in § 21) to the Congres- sional Budget Office (CBO) of the duty to determine, with respect to each spending bill, whether passage of the bill would cause the budget category within which the bill falls to exceed the spending ceiling established by the “budget law” enacted earlier in the year (or the previous year’s spending level, if no budget law is enacted). This determination has two important conse- quences under the draft bill: (1) under § 7, a supermajority (two-thirds) vote in each House of Congress would be required for passage of the spending bill if CBO determines it would exceed its spending ceiling (or previous year’s spending level); and (2) under § 25, any bill that would thus be subject to a supermajority vote requirement would also be subject to the rescission author- ity that would be granted to the President under that section. 44 This delegation to CBO of authority to make a determination that has such significant consequences gives rise to a possible constitutional question of whether that determination constitutes legislative action, and if it does, whether the constitutional requirements for legislative action would be satisfied. The Supreme Court has made it clear that any legislative action — i.e., any congres- sional action that has binding legal effect outside the Legislative Branch — must comply with the constitutional requirements of bicameral passage and presentment to the President. INS v. Chadha, 462 U.S. 919, 952 (1983). It seems clear that the first consequence of a positive CBO determination — requirement of a supermajority vote in each House of Congress — does not run afoul of these requirements. Its effect would only be on the internal legislative practices of each House of Congress, and would thus be limited to the Legisla- tive Branch. It would therefore not constitute legislative action within the meaning of Chadha. Moreover, because “[e]ach House may determine the Rules of its Proceedings,” U.S. Const, art. I, § 5, cl. 2, it is within Congress’ constitutional authority to adopt legislative procedures of this kind. We note in passing that, unlike a constitutional amendment, the draft legisla- tion would not have a truly binding effect on Congress. Clearly, Congress cannot by legislation prevent itself from enacting future legislation pursuant to whatever procedures it chooses to follow at that future time. A future Congress can always legislatively change what a previous Congress has done. In a legally enforceable sense, therefore, such future lawmaking would be regulated only by the requirements of the Constitution. Thus, notwithstanding the provi- sions of the draft bill, a future Congress could follow whatever procedures it chooses to apply with respect to a particular appropriations bill, including passage by less than a supermajority. Or it could choose simply to disregard the CBO determination. Although strong political pressures would certainly oper- ate against defiance of the budget process requirements, and the President could surely cite noncompliance as a basis for a veto decision, a subsequent appropriations law passed in compliance with constitutional requirements would be valid, notwithstanding any noncompliance with the procedures of this bill. We also believe that the second consequence of a positive CBO determina- tion — identification of appropriations that would be subject to Presidential rescission — does not violate the bicameral action and presentment require- ments, but we base this conclusion on different grounds from those applicable to the first consequence. The practical effect in this regard of the CBO determi- nation would indeed be to bind parties outside the Legislative Branch, because the President’s authority to rescind appropriations would extend only to appro- priations based on bills that are enacted under the supermajority requirement, which in turn is based on the CBO determination. Legislative action would thus be involved, but in our view the actual legislative action would be the enact- ment of the spending bill subsequent to the CBO determination and prior to the rescission authorization to the President becoming effective. The essential point is that the scope of the President’s rescission authority would be defined not by the CBO determination itself, but rather by the subsequent congressional 45 enactment of the spending bill. That enactment would satisfy the bicameral action and presentment requirements.1 We stress that under the draft legislation the ultimate decisionmaker on defining the scope of the President’s rescission authority would not be an arm of the Congress, but rather would be Congress itself acting in compliance with the constitutional requirements for legislative action. The budget process role that is contemplated for CBO under this bill thus differs in a critical respect from the role the General Accounting Office (GAO) was given under the Gramm-Rudman-Hollings Act. See Bowsher v. Synar, 478 U.S. 714 (1986). Under Gramm-Rudman-Hollings, GAO was authorized to present binding budget reduction calculations directly to the President. In contrast, under the draft bill, CBO’s implicit instructions to the President concerning what pro- grams are subject to his rescission authority are presented through the Con- gress, pursuant to procedures that satisfy constitutional requirements.2 Although, for the reasons stated above, we believe that a strong argument can be made to sustain the role of CBO in defining the scope of the President’s rescission authority, that argument turns principally on whether the subsequent enactment of the spending bill may properly be viewed as congressional action that itself has the effect of defining that scope. Under the draft bill, it would appear that any such congressional action would have to be viewed as implied. We suggest, therefore, that consideration be given to requiring in the draft bill that the congressional action be express. Under one possible version of such a requirement, any spending bill enacted pursuant to a CBO determination would have to include, most likely in introductory language (such as the “whereas” section), a statement that a two-thirds vote was required on the basis of the CBO determination that the bill would exceed the spending ceiling. An alterna- tive approach would be to require that each such spending bill state that appropriations authorized under the bill would be subject to the President’s rescission authority. II. Rescission of Enittitlemeinit Appropriations Section 25 of the draft bill would add a new § 689 to Title 2, United States Code. Under that section, the President would be authorized to rescind any spending appropriations that are authorized by legislation enacted pursuant to the supermajority voting requirement. Thus, under the regime established by 1 An alternative way to analyze this second consequence of the CBO determination is to take the view that the subsequent appropriations law — which is passed pursuant to a supermajority vote triggered by the CBO determination — would amount to an implied congressional ratification or adoption of the CBO determina- tion. W e prefer the analysis taken in the text, because in our view it is based on a more accurate description of the process contemplated under the draft legislation. Under either analysis, however, the critical fact is that intervening between the CBO determination and the establishment of the President’s rescission authority is a legislative actiun effected in compliance with constitutional requirements. 2 An additional distinction — although of less significance for this analysis — is that Gramm-Rudman- Hollings involved a delegation to GAO of an executive function (determining how to implement spending reductions), while the draft bill involves a delegation to CBO of a legislative function (defining the programs with respect to which the President is being delegated rescission authority). 46 the draft bill, any such appropriations law would by clear implication provide that all appropriations are non-mandatory. Congress certainly may make expenditure of a particular appropriation non- mandatory. See Train v. City of New York, 420 U.S. 35 (1975). A fortiori, Congress may expressly grant the President the authority to rescind any appro- priation pursuant to a congressionally established procedure. In contrast to the non-mandatory appropriation situation, however, a Presidential rescission of a mandatory appropriation would amount to an unconstitutional unilateral amend- ment of the appropriations law. Congress may not authorize the President to circumvent the constitutionally required process for amending previously en- acted laws any more than it may authorize itself to do so. Cf INS v. Chadha, supra. Application of the draft bill’s Presidential rescission authority to entitlement appropriations presents a special situation. Unlike spending based on the usual appropriations bill, entitlement payments are generally made on the basis of two separate statutory enactments. The first statute establishes the entitlement and generally fixes a specified amount to which each person meeting the statutory requirements is entitled. The second statute is an appropriations bill that authorizes the expenditure of funds up to a given amount.3 Thus, if the President utilized the rescission authority granted by the draft bill to reduce entitlement payments below the statutorily prescribed level, he would, in effect, be amending unilaterally the previously adopted entitlement statute. However, entitlement statutes may be changed only by other duly adopted statutes; Congress may not delegate to the President unilateral power to do so himself. This conclusion does not mean, however, that it would be impossible for Congress to delegate to the President power to control expenditures under entitlement programs. To the contrary, the statute could be drafted so as to provide such authority. First, it is clear that Congress itself has the power to amend or reduce entitlements that it has previously granted. For example, the Supreme Court has held with respect to Social Security that “a person covered by the Act has not such a right in benefit payments as would make every defeasance of ‘accrued’ interests violative of the Due Process Clause of the Fifth Amendment.” Flemming v. Nestor, 363 U.S. 603, 611 (1960). The Court has also held that the “fact that social security benefits are financed in part by taxes on an employee’s wages does not in itself limit the power of Congress to fix the levels of benefits under the Act or the conditions upon which they may be paid. Nor does an expectation of public benefits confer a contractual right to receive the expected amounts.” Richardson v. Belcher, 404 U.S. 78, 80 (1971). Congress could utilize this power to effect a general cross-cutting amend- ment to all entitlement statutes that would make a certain percentage of the entitlement amounts subject to limitation or complete withdrawal either by Congress through the appropriations process, or by the President through the rescission process proposed by the draft bill. Thus, the draft bill could include a 3 In many cases, such appropriations bills set no absolute limits on entitlement expenditures, but rather state that the Executive may expend an amount sufficient to pay all individuals who qualify under the provisions of the relevant entitlement statute. 47 provision explicitly amending all entitlement acts so as to permit some Presi- dential control over entitlement expenditures in the same way as the draft bill would permit control over expenditures pursuant to appropriations bills. A cross-cutting provision would thus avoid the constitutional problem by making the expenditure of a certain percentage of appropriated funds non-mandatory.4 HII. Limiting the Reasons on which the President Can Rely Wlhen Exercising the Rescission Aratiooty Proposed 2 U.S.C. § 689(b) (see § 25 of the draft bill) would permit the President to rescind “excess budget authority” only for “reasons of economy, efficiency, or fiscal management of the Government.” The apparent purpose of this provision would be to indicate that the President’s authority is not intended to extend to situations in which the President’s primary reason for desiring to rescind budget authority is disagreement with congressional programmatic objectives. Although the provision does not give rise to an issue of constitu- tional law, you may wish to consider its separation of powers policy implications. The distinction that § 689(b) would draw might turn out to be illusory and unenforceable. It would be very difficult to separate motives of economy from policy judgments concerning the efficacy of a particular program. Moreover, although we believe that disputes arising under this section between Presidents and Congress would almost always involve only “political questions” that should not be resolved by the courts,5 the litigation potential created by such a provision should be recognized. Giving the courts an additional excuse to attempt to second-guess or inquire into the motives of the President could potentially give the courts an opportunity to seek to exercise significant “politi- cal” power, a role that is not contemplated under the Constitution and that they are institutionally ill-suited to exercise. IV. Technical Language Change to Avoid Amtlnoiriizimg Legislative Veto Proposed 2 U.S.C. § 689(d)(1) (see § 25 of the draft bill) is clearly intended to provide for congressional disapproval of a Presidential rescission by the constitutionally permissible means of a bill that is enacted in compliance with 4 We note that Congress has already effected such an amendment to a specific entitlement statute in the context of the food stamp program. The so-called Lugar Amendment authorized the Secretary of Agriculture to reduce the otherwise required food stam p allotments if insufficient funds were appropriated to fund the program at its full level, and additionally authorized the Secretary to change the allocation formula if such a reduction were necessary. 7 U.S.C. § 2027(b)-<d). 5 See Goldwater v. Carter, 444 U.S. 996, 1002-06 (1979) (Rehnquist, J., concurring); Barnes v. Kline, 759 F.2d 21, 57 (D.C. Cir. 1985) (Bork, J., dissenting) (“[I]t is absolutely inconceivable that Framers who intended the federal courts to arbitrate directly disputes between the President and Congress should have failed to mention that function or to have mentioned judicial review at all. The statesmen who carefully spelled out the functions of Congress and the President and the details of how the executive and legislative branches might check each other could hardly have failed even to mention the judicial linchpin of the constitutional system they were creating — not if they had even the remotest idea that the judiciary was to play such a central and dominant role ”), vacated on other grounds sub nom. Burke v. Barnes, 479 U.S. 361 (1987). 48 the bicameral action and presentment requirements of the Constitution. As written, however, the provision technically provides instead for a two-house legislative veto: a rescission would take effect unless within 45 legislative days of Congress’ receipt of the President’s rescission statement, “Congress shall have completed action on and sent to the President for his approvaP’ a bill disapproving the rescission. Id. (emphasis added). Thus, the disapproval would technically take effect upon presentment to the President, and the constitutional requirement that the President have an opportunity to veto the disapproval bill would be circumvented. See INS v. Chadha, supra. To accomplish the purpose that we assume is intended, we suggest that the above-quoted language be deleted and the phrase “is enacted into law” be added at the end of the sentence. Thus, under the draft bill as revised, a rescission would take effect “unless within 45 legislative days of the receipt of the President’s rescission message, a bill dealing solely with such rescission that restores all or part of such excess budget authority is enacted into law.” If you believe that 45 days would not be enough time to allow for a congressional attempt to override a Presidential veto,6 you might consider allowing instead for some longer period, such as 60 days. C h a r l e s J. C o o p e r Assistant Attorney General Office of Legal Counsel 6 A veto would be almost a certainty. Because the joint resolution would be a rejection of the President’s rescission, a veto would constitute a simple reassertion of the rescission. 49