What do I have against the Seventeenth Amendment?

In the comments to my post yesterday on the electoral college, referencing my concern that the rhetoric surround the abolition of the electoral college is reminiscent of that surrounding the last amendment to work a major change to the structural constitution, I fretted that "[i]nstead of talking about removing the direct election of Senators, we are talking about commencing the direct election of Presidents. We have gone profoundly astray." Rafique asks me to elaborate. It appears that I never have addressed this question squarely here or elsewhere, and this seems as good a time as any.

I start from a position I take to be axiomatic: "Much of the Constitution is concerned with setting forth the form of our government, and its primary protections of liberty are structural ... First among these structural protections are the separation of powers, served vertically by federalism...." The structure was dramatically weakened, however, by the removal of the states from the federal legislative process. "The primary institutional role played by the original Senate was to protect the structure of federalism and state sovereignty, in response to concerns ... that an omnipotent federal government would swallow-up the state governments. Appointment of Senators by state legislatures gave the states a constituent role in the national government and a means to protect themselves from laws ... designed to subvert state sovereignty and independence." Todd Zywicki, Beyond the Shell and Husk of History: The History of the Seventeenth Amendment and its Implications for Current Reform Proposals, 45 Clev. St. L. Rev. 165, 169 (1997) (footnote omitted). Simply put, then, the fundamental problem with the Seventeenth Amendment is that it removed the primary structural check on the federal government's tendency to aggrandize itself vis-à-vis the states.

After the ratification of the Seventeenth Amendment, Senators continued to represent states in theory, but in practice, today represent their own interests first, their constituents second, and their state governments not at all. "Once Senators were no longer accountable to and constrained by state legislatures ... senators almost always found it in their own interest to procure federal legislation, even to the detriment of state control of traditional state functions." Jay Bybee, Ulysses At the Mast: Democracy, Federalism, and the Sirens' Song of the Seventeenth Amendment, 91 NW. U. L. Rev. 500, 535-6 (1997). Thus, the representation of the states qua states vanished from the federal legislative process, which is why I branded Justice Blackmun's pollyannaish opinion in Garcia v. San Antonio Metro. Transit Authority, 469 U.S. 528 (1985), the product of a "quaintly pre-17th Amendment mindset." Without a federal judiciary willing to step in to defend the structural constitution, the natural upshot was the evaporation of institutional checks on the power and size of the federal government.

To make matters worse, the incentives of Senators were actively thrown into reverse. Examining incentives is important for looking at separation of powers questions from a public choice perspective. If we treat "politicians as rent-seeking individuals using their positions to pursue private goals, and government as an institution that encroaches on individual freedoms and seeks to increase its grip on the private economy as much as possible," it is obviously vital that "institutional rules and arrangements forc[e] politicians to serve the public interest in the pursuit of their own goals...." Jürgen von Hagen, Fiscal Federalism and Political Decision Structures in Federalism in a Changing World 379 (Raoul Blindenbacher & Arnold Koller, eds. 2003).

The only incentive that men with fame and power feel more strongly than the desire to increase their fame and power is the fear of losing such fame and power that they already have. Thus, the incentive structure of politicians leads them to pander to their constituencies. If a Senator's constituency is a state legislature, his or her natural tendency is to resist measures before the Senate that tend to diminish or harm the power and interests of the states generally, and the prerogatives and powers of the state legislatures particularly. With institutional representation of the states in Congress, we should expect cases involving the power of Congress to waive state sovereign immunity, for example, to be few and far between. Senators would have strong incentive to deep six any such provisions poured out of the House, and so there would be nothing to litigate.

But when a Senator's constituency is the electorate, they have very different incentives. There is zero electoral cachet in telling voters that their concerns are being directed to the wrong person. That is why Presidential campaigns today are chock full of promises that whatever is on voters' minds is something the candidate will address when -- in American politics, always the conceit of "when" not "if" -- elected. Likewise, directly elected Senators have an incentive to increase federal power because doing so expands their own power and importance, and increases the sphere in which they can seek the lifeblood of electoral politics, credit. (President Truman is credited with the maxim that "it is amazing what you can accomplish if you do not care who gets the credit"; the remark's humor derives from the the golden rule of electoral politics that the only thing worse than an unsolved problem is a solution for a problem that voters don't care much about, or one they won't give you credit for.)

Thus, Madison's "great security" against the concentration of power, "giving to those who administer each department the necessary constitutional means and personal motives to resist encroachments of the others" such that "[a]mbition ... [is] made to counteract ambition," The Federalist, No. 51, is discarded, and the interests and ambitions of those in the Senate are realigned with, rather than against, federal power.

To make a bad situation worse, we did all this after adopting the Sixteenth Amendment. The latter not only provided the Federal government access to almost boundless power to obtain such resources as are necessary to carry out almost any design, but also, "[b]y abolishing the apportionment requirement ... gave the federal government the power to control 100% of the entire economy. It can tax income. It can not tax income, achieving its goals via tax expenditures, that is, by encouraging those things that aren't taxed. It can tax and then subsidize using the dollars that it's just collected from you, or it can grant the dollars back on condition. That combination of powers ... gives the federal government control over almost anything it chooses to control." Frank Easterbrook, Are Constitutional Changes Necessary to Limit Government? (2006) (emphasis in original) (hereinafter Easterbrook), recording at 44:54-45:40. (Justice Holmes' reply to Chief Justice Marshall's dictum that the power to tax is the power to destroy, "[n]ot ... while this court sits," has always been more of a quip than doctrinal landmark. See M'Culloch v. Maryland, 17 U. S. (4 Wheat.) 316, 431 (1819); Panhandle Oil Co. v. Missisippi ex rel. Knox, 277 U.S. 218, 223 (1928) (Holmes, J., dissenting)).

The result was a perfect storm. Congress gained the power to swell the federal sphere to Brobdingnagian proportions at the very moment that it was released from the primary structural restraint that might check its likelihood to use those powers to aggrandize federal power at the expense of the states, viz. the interests and ambitions of the states given force in the Senate. Thereafter, although "'[t]he Constitution, in all its provisions, looks to an indestructible Union, composed of indestructible States,'" and although "there can be no doubt that we have an indestructible Union," prevailing winds "cast[] some doubt upon whether those States are indeed 'indestructible.'" Fry v. United States, 421 U.S. 542, 549 (1975) (Rehnquist, J., dissenting) (quoting Texas v. White, 74 U.S. 700, 725 (1868) (first alteration in original)). The welfare of the henhouse is thus entrusted to the foxes - unless, as I said above, the judiciary can and does intervene, something it has shown at best episodic interest in doing. Compare, e.g., National League of Cities v. Usery, 426 U.S. 833 (1976); United States v. Lopez, 514 U.S. 549 (1995), with Garcia, supra; Gonzales v. Raich, 545 U.S. 1 (2005). The defense of federalism is reduced to the hope that the people will elect vegetarian foxes.

Still, as I noted here, while "the 16th and 17th Amendments collude ... (that is, the 16th amendment makes possible the funding for big federal government and the 17th amendment removes the primary check on big federal government)[,] ... I see the 17th Amendment as far more problematic than the 16th amendment" if each are taken separately. The Sixteenth Amendment by itself "merely makes possible a particular government action that is likely to be checked by a majority (the framers took care of that in their design of the House of Representatives and the appropriations process)...." By contrast, the Seventeenth Amendment is inherently dangerous because it strikes at the carotid artery of the Constitution's delicate balancing of powers. The results speak for themselves. Although Judge Easterbrook says, quite correctly, that the world has changed around the Constitution in ways that displace power to the federal government, he also acknowledges that this catastrophe of so-called progressive politics remains among the "principal causes" of the size of the federal government today. Easterbrook, supra, at 40:50; 39:23-50.

I acknowledge that the pre-Seventeenth Amendment regime was not perfect; not all of those who militated for it were driven by (as Pat well-described those pushing for the elimination of the electoral college) "a fetish for pure democracy." See, e.g., David Kyvig, Redesigning Congress: the Seventeenth and Twentieth Amendments to the Constitution in The American Congress 358 (Julian Zelizer ed. 2005) ("[f]requent state legislative deadlocks over the choice of a Senator furthered the image of a defective electoral system. In no fewer than forty-five instances in twenty states between 1891 and 1905, legislatures, unable to reach agreement on candidates, delayed filling seats. Fourteen seats remained empty for at least one entire Congressional session"); cf. Zywicki, supra, at 198-201. We cannot go back. What's more, in expressing my skepticism about our ability to improve on the Constitution, I do not mean to contradict Justice Story's observation that the framers understood that "time might develop[] many defects in [the Constitution's] arrangements, and [reveal] many deficiencies in its powers." Some alterations to "the fabric of government" may, as Story said, be suggested by "time and experience, or the new phases of human affairs ... to promote the happiness and safety of the people"; indeed, no lesser authority than Burke warned that "[a] state without the means of some change" - formal or intellectual - "is without the means of its conservation." Thus, although I may talk about repealing the Seventeenth Amendment as a shorthand, when I talk about my proposed 28th Amendment [See update, below], I'm careful to note that I am not proposing a direct repeal, simply returning to the status quo ante. What I propose seeks to avoid the worst structural problems of the pre-Seventeenth Amendment Senate, primarily that of open seats.

Of course, having said that "reformers who 'lustily eye the branch' ... [must] identify clear and compelling problems with current practice and then link them to a solution that offers substantial improvement on current practice and [that] seems unlikely to generate unforeseen consequences," I set a high bar for my own proposal to clear. I have also said that, in my view, "amendments should generally be in harmony with the general principles of the existing structure, unless dictated by profound and outstanding moral or practical exigency" (a test, I said, that the Seventeenth Amendment failed). Indeed, the bar is yet higher when a proposal is structural rather than a mere rights-bearing provision; placing one issue or one family of issues "outside the arena of public debate and legislative action," Washington v. Glucksburg, 521 U.S. 702, 721 (1997), greatly affects some, but tinkering with the process and domain of lawmaking itself affects all.

Still, as I said a year ago, "there is a clear and compelling case that as time has unfolded, it has become clear that the Senate that the Seventeenth Amendment has produced is institutionally dysfunctional, incapable of playing the mixed legislative/executive role envisioned by the framers, and worse yet, [for reasons discussed above] ... has fatally destabalized the delicately-balanced federal system." My proposal, I believe, meets these tests, and is fundamentally in harmony with the general principles of the Constitution.

In my view, the history of federal expansion in the Twentieth Century, and review of the Senate that electoral politics have wrought, make it clear that the Seventeenth Amendment is, on the most charitable view, a failed attempt to respond to legitimate concerns that has done far more harm than good. Some mistakes are water over the dam; we should live with them as best as we can. Others, however, are so fundamentally damaging that the best course is to acknowledge the mistake, and to correct it. For the foregoing reasons, I think that the Seventeenth Amendment is of the latter class.

Added: in the comments, Rafique draws my attention to Section 4 of my proposed amendment, and I should append what I have to say in reply to the main post, for sake of clarity. I wrote the amendment a few years ago, but I am now inclined to throw Section 4 and the related portion of Section 1 overboard.

The animating concern that lead to its inclusion was the so-called "permanent campaign" mode of governing that has been so corrosive over the last quarter-century. "We live in a time when the campaigning never stops, and the governing all too frequently never begins," as Senator Snowe said. I think that it's important and useful to look at people's incentives if you want to understand how a position is going to function and interrelate to a larger system, and you see that concern in several of my posts, including part I ¶¶3-4 of this one. That is why Federalist 51 has a particular resonance with me. We have to be realistic about human nature, and where possible, line up the natural incentives and inclinations of the office and its likely occupants (human beings, we can safely predict) such that they work for the general scheme rather than against it. The genius of the Constitution is that it does precisely that - it sets the ambition of Presidents against the ambition of legislators, compelling them to be rivals, which is is also why interbranch cooperation is generally something to be feared, not welcomed (cf. Easterbrook, The Chicago School and Exclusionary Conduct, 31 Harv. J. of L. & P.P. 439, 442 (2008) ("the main goal of antitrust is to compel firms to be rivals; cooperation is to be feared rather than welcomed")).

With that in mind, the idea behind Section 4 was that if neither the President nor the Vice-President were eligible to be reelected, the incentive structure that produced the permanent campaign would evaporate. The point of the permanent campaign is to secure reelection (or, in the case of the Vice President, election); if that goal were no longer possible, I thought, the President's incentives would be very different, and the focus would be on governing well. At best, I thought, we would squeeze the maximum amount of "governing" time out of the President, while at worst, we would cut directly to the "legacy" years, which is not a terrible result either.

I think that analysis still holds, for the most part, but I have become skeptical (reflected, for example, in my post here) that whatever new incentives would arise would be a sufficient improvement to merit a Constitutional amendment. They might be, of course: I think that it's very possible that Section 4 represents a real improvement that would greatly improve government. But I also think it's possible that it would cripple the executive and plough up any number of problems that I lack the foresight to imagine. As I said in reply to Susan yesterday, the point isn't whether I can devise something that seems like an improvement. To tinker with Article II, either Susan or I must do more than merely "show that [we] have a system that would be okay if we were designing a Constitution from scratch," but rather, a reformer "must overcome the inertia and presumption in favor of an institution that has functioned largely without incident for the entire history of the republic." For that reason, although I don't disown Section 4, I do now think that it fails the tests that I've articulated in this post and others for major structural reform.

Post facto:
A less perfect Constitution, 2 (10/18/08)
Continuity of government (1/28/09)
Feingold's amendment: the worst kind of pig-ignorant populist nonsense (8/20/09)