Government efforts to regulate campaigns, as apart from elections, suffer from the inability to develop meaningful or even non-arbitrary, judicially manageable standards. However, there is a further, more fundamental reason to be concerned about legislative efforts to regulate campaigns — they inexorably lead to the government placing its thumb on the scale to advantage certain candidates and interests.

Candidates enter campaigns with a wide variety of advantages and disadvantages. The same is true of political parties, and the various interests, citizen organizations and even viewpoints that will play a part in the campaign. Government regulation of campaigns inherently involves choices that benefit or harm particular types of candidates and interests, and as such they are inherently open to manipulation by the government itself.

One obvious group that may be routinely favored by campaign regulation is incumbents. Justice Scalia has been particularly skeptical of incumbent self-dealing over the years. As he wrote in dissent in Austin v. Michigan State Chamber of Commerce, “[t]he incumbent politician who says he welcomes full and fair debate is no more to be believed than the entrenched monopolist who says he welcomes full and fair competition.”

Of course, incumbents are not the only group that might benefit. The facts of Austin vividly demonstrate the problem. A Michigan law limited corporate expenditures, but not union expenditures, in political campaigns. Whatever “neutral” justifications may arguably exist, the law looked suspiciously like an effort to cement an advantage for one side in Michigan’s long-running political war between large manufacturers (especially auto manufacturers) and unions (especially the United Auto Workers). Even if the original motives were pure, such laws entangle the state in campaigns such that future decisions to amend, revoke, or expand the law will have partisan political consequences.

If the problem is substantial when government attempts to limit or prohibit certain types of campaign speech or tactics, it is even worse when the government seeks to directly intervene in the campaign, typically by directly funding candidate campaigns.

Howard Baker put it well in a floor speech opposing provisions of the 1974 Federal Election Campaign Act that authorized the government to fund candidate campaigns:

“There is something politically incestuous about the Government financing and, I believe, inevitably then regulating, the day-to-day procedures by which the Government is selected. … I think it is extraordinarily important that the Government not control the machinery by which the public expresses the range of its desires, demands and dissent.”

Once the government is involved in funding campaigns, it has a temptation to seek to benefit certain classes of candidates, and a vested interest in seeing those candidates win.

Current doctrine is, unfortunately, ill-equipped to deal with this problem. In Buckley, the Court accepted the legitimacy of direct government financing of campaigns, so long as candidate participation was “voluntary,” on the grounds that such financing did not suppress speech, but subsidized even more speech.

Thus three years ago, in Arizona Free Enterprise Club’s Freedom Club PAC v. Bennett, the Court struck down an Arizona scheme of government financing that provided a block grant to candidates who chose to participate in the public financing system. Under the plan, if a non-participating candidate was able to raise and spend more than the original block grant, the government provided the participating candidate with added funds to match that spending, up to three times the original grant. The Court held that the system unconstitutionally penalized the non-participating candidate for exercising his speech rights because each time he spent another dollar, the State provided a matching dollar to his opponent.

In a snarky but powerful dissent, Justice Kagan, writing for the Court’s four liberals, suggested that this was nonsense. After all, she reasoned, the State was clearly free, under Buckley, to simply increase the original grant, which presumably would put non-participating candidates at an even greater disadvantage. The matching funds the state provided simply allowed for more speech — supposedly a good thing — and assured that the big spending, non-participating candidate would not be unrebutted.

But the scheme in Bennett was intended to favor certain candidates. For various reasons — protection of the public purse, but also a general disdain for money in politics and, quite likely, for the benefit of incumbents — the initial grant was kept low (there is quite good evidence that incumbents benefit from low-spending campaigns). To make the plan work, however, candidates had to be assured that participating would not decrease their odds of winning. Hence the “matching funds.”

A history of the Act and the campaign for its enactment also shows that the system, though available to conservatives, was largely expected to favor the election of “progressive” candidates. For one thing, a candidate qualifies for the grant by raising a number of very small contributions. Democratic candidates such as former Governor Janet Napolitano have benefited as unions collected those qualifying contributions. Moreover, once the government had designed the plan to help participating candidates win, what candidates would not participate? Mainly, those who might object on ideological grounds, who are almost certainly likely to be candidates who favor limited government, lower taxes and spending and less regulation.

Consider another example, caused by indirect rather than direct government subsidies of campaigns. For the past decade, the idiotic phrase, “I’m John Doe and I approve this message” has appeared at the end of virtually every broadcast ad by a candidate for federal office. The information is of little use to voters — at a minimum less useful than “Give me liberty or give me death;” “We have nothing to fear but fear itself;” “I will not raise your taxes,” or “If you like your plan, you can keep your plan.” The phrase is not actually required by federal law, but virtually every candidate chooses to say it. The reason is because a candidate who uses that phrase is legally entitled to the lowest unit rate for broadcast advertising; omission of the phrase leaves one to pay higher rates.

This provision was part of the McCain-Feingold law of 2002, and there is no doubt as to how this came to be — incumbent officeholders believed that it would reduce the number of negative television ads, which are known to be more effective for challengers than for incumbents. The record is replete with testimony; anyone who doubts it and wants to avoid the 100,000-page record can read the brief filed by the National Rifle Association in McConnell v. FEC. As it turns out, the effort to suppress critical ads seems to have failed, but we are still saddled with a worthless “claimer” taking up 10 to 15 percent of the typical political ad. Next time incumbents may be more successful.

As these examples illustrate, such direct government involvement in campaigns is not wrong because it suppresses speech, but because it gives the government both a means and a further incentive to put its thumb on the scale. Separation of campaign and state provides the theory that explains why the Arizona plan was unconstitutional, and why “stand by your ad” should be.

Direct government involvement in shaping campaigns is incompatible with the American ideal of republican government. Once the government is dictating the format and content of campaigns, it is no longer beholden to the people, as intended in the Constitution, but the people are beholden to the government.

Dissenting in Buckley, Chief Justice Burger argued that “the inappropriateness of subsidizing … the actual political dialogue of the people-the process by which the Government begets itself-is as basic to our national tradition as the separation of church and state … or the separation of civilian and military authority.”

Tomorrow I’ll briefly address what the separation might look like.