The People Who Show Up

Section 1The question before us

Stand on the sideline of a Saturday morning soccer game. Sit through a shift change, a Sunday service, a parent-teacher conference, a bowling league. The Americans you meet in those places are, overwhelmingly, reasonable people. They hold opinions without holding grudges. They can work beside someone who votes differently and never think about it.

Now look at the people who represent them. The gap between those two groups is enormous, and it is the puzzle this paper exists to answer. If Americans are as ordinary as they appear at the ball field, why is their government not?

The answer this paper proposes is not that Americans have become worse people. It is that the rules governing American elections systematically overweight the most motivated citizens and underweight everyone else — and that the most motivated citizens are, by definition, not a cross-section. A person angry enough to attend a Tuesday special election in March is not representative of his neighbors. He is simply the one who came.

Ross Perot put the general principle better than we can. He argued that American problems are more often structural than moral: that if you open the hood and repair the mechanism, a self-governing people will handle most of the rest themselves. One need not have voted for the man to think he was right about that. This paper takes the screwdriver to six things.

1.2   Our own share of the failure

The complaint that government is broken is very often made by people who did not vote in the election that produced it.

That is the uncomfortable half of this argument and it belongs at the front. The reasonable, busy, decent Americans described above are also, in large numbers, the Americans who skipped the school board race, could not name their state representative, and did not know a municipal election was being held. The extremes did not seize anything. They showed up, and the rest of us did not, and the results are what results always are under those conditions.

This paper argues that the structure makes showing up unreasonably hard, and we believe that. But structure is an explanation, not an excuse. Every reform proposed here would be worth less than an hour of a citizen's attention if that attention never arrives. If we fix the mechanism and still do not vote, we will have earned exactly what we get.

Section 2Who actually decides

The claim that low turnout favors the organized is not folk wisdom. It has been studied directly.

Sarah Anzia's Timing and Turnout, published by the University of Chicago Press, examines the timing of American elections and its consequences. The finding: public policy in the United States is made by more than 500,000 elected officials, the great majority of whom are chosen on days other than Election Day — which means that most American officials are selected by a politically motivated minority of Americans.1

The mechanism is straightforward. Organized groups turn out at high rates whatever the date, because their members have a concentrated stake and someone whose job is to remind them. Ordinary citizens have a diffuse stake and no one calling. When total turnout falls, the organized share of the electorate rises — not because anyone cheated, but because arithmetic. Anzia's account traces how interest groups and parties have historically fought over election timing precisely because they understood this.

Note what this argument does not require. It does not require that any group be malicious, or that any election be improper. A teachers' union, a police association, a developers' PAC, and a county party committee all behave rationally by turning out in March. The problem is the calendar that rewards them for it.

Section 3Reform I: One election, four days long

Every election in the United States, from president to water district, should occur during a single annual voting period: Friday through Monday, with the Friday and Monday designated federal holidays. No special elections at other times except to fill a vacancy, and those consolidated to the next scheduled window.

3.1   Why Tuesday

The current schedule is a fossil. Congress fixed Election Day as the first Tuesday after the first Monday in November in 1845. November followed the harvest and preceded hard winter. Tuesday avoided Sunday, widely observed as a day of worship and travel to the county seat took a full day, and avoided Wednesday, which was market day in many towns.2

Every one of those reasons has expired. We are not an agrarian society, we do not travel by horse, and no one plans a week around market day. The rule remains because rules remain.

3.2   What everyone else does

Twenty-seven of the thirty-six OECD member countries hold their national elections on a Saturday or Sunday. Two more — Israel and South Korea — vote on weekdays but declare those days national holidays so that economic hardship is not a barrier to participation.3 By turnout among the voting-age population, Pew ranked the United States 31st of 49 countries studied, most of them OECD members.7

Among registered Americans who did not vote in 2016, being too busy or having a schedule conflict was the third most common reason given — roughly 14 percent, some 2.7 million people.7 International IDEA's global review of turnout estimated that moving election day from a working day to a rest day raises turnout by five to six percentage points on average.8

3.3   The consolidation is the more important half

The holiday gets the attention; the consolidation does the work. Under the present system a city, a school district, or a special authority can place a measure before voters in March, in an odd-numbered year, when nobody is watching except the people with a direct financial interest in the outcome. That is not a loophole in the system. Under Anzia's evidence it is the system's most reliable feature.

A single annual window ends it. Every officeholder faces the full electorate, on the same days, with the same attention. A bond measure that cannot survive a November electorate should not pass in March.

Section 4Reform II: Districts that stop moving

District lines should be fixed to permanent, publicly legible boundaries — county lines, municipal lines, or a geometric grid — and should change only when population requires and only by a rule stated in advance.

The case is not complicated and most Americans already accept it. When the people who benefit from the map are the people who draw the map, the map will be drawn to benefit them. There is no version of that arrangement that produces competitive elections, and competitive elections are the only thing that forces a representative to speak to the people in the middle.

In a safe district the only election that matters is the primary, and the primary electorate is the smallest and most ideological electorate in American politics. A member who fears only a primary challenge has no incentive to talk to anyone else and every incentive to avoid compromise. That is not a character flaw in the member. It is the rational response to the district he was given.

4.1   The courts have stepped back

In Rucho v. Common Cause, decided 27 June 2019, the Supreme Court held 5 to 4 that claims of partisan gerrymandering present nonjusticiable political questions beyond the reach of federal courts. Chief Justice Roberts described districting as an inherently political process and found no manageable standard by which judges could determine when partisan advantage becomes too much.4

The opinion is important here for a reason often missed. The Court did not hold that gerrymandering is acceptable; it acknowledged that the practice may be incompatible with democratic principles. It held that the remedy lies elsewhere — specifically in state constitutional amendments, state statutes, and congressional action under the Elections Clause.4

That is an invitation. It is also a warning: no one is coming to fix this from the bench.

Section 5Reform III: The Electoral College

The president should be elected by national popular vote.

The practical objection to the present system is not abstract. Under winner-take-all allocation, a campaign has no reason to speak to voters in states it cannot win or cannot lose, which is most states and nearly all of the country's largest population centers. A Republican in California and a Democrat in Alabama are equally invisible. The effect is to concentrate national attention on a handful of states and to teach tens of millions of Americans that their presidential vote is decorative.

That teaching has consequences beyond turnout. It sorts the country's political conversation into places that get visited and places that get written off, and it deepens precisely the urban-rural resentment that both parties then exploit.

5.1   This nearly happened, and conservatives led it

The history is not widely known and it changes the character of the debate.

After the 1968 election, in which George Wallace won 46 electoral votes and raised the prospect of a contingent election decided by deal-making, Representative Emanuel Celler introduced a constitutional amendment to abolish the Electoral College in favor of direct popular election, with a 40 percent threshold and a runoff if no candidate reached it. The House Judiciary Committee approved it 28 to 6. On 18 September 1969 the House passed it 339 to 70, with majorities of both parties. House Minority Leader Gerald Ford supported it. On 30 September, President Nixon endorsed it and urged the Senate to act.5

The New York Times reported in October 1969 that thirty state legislatures were certain or likely to ratify. Thirty-eight were needed. The Senate Judiciary Committee approved the measure 11 to 6 in August 1970. On the Senate floor it was filibustered, led by Strom Thurmond and a bloc of Southern and small-state senators of both parties. Two cloture votes failed, 54–36 and 53–34, and the amendment was set aside.5

It remains the only time a chamber of Congress has voted to abolish the Electoral College. It failed by a procedural margin, not a popular one, and it failed with a Republican president on the record in favor.

Section 6Reform IV: A majority, or a runoff

No one should take office having been rejected by most of the people who voted. Where no candidate reaches a majority, the top two should face each other.

Most states permit plurality winners, and many state constitutions affirmatively require the office to go to whoever receives the most votes rather than a majority.6 The results are routine rather than exotic. Maine elected a governor in 2010 with 38.2 percent of the vote. In nineteen presidential elections since 1844 — most recently in 2024 — the winner took office with less than half the popular vote. Those nineteen elections were won by fifteen different men.6

Two things follow. The first is consent: Americans accept outcomes they dislike far more readily when they can see that most voters chose them, and a winner opposed by two-thirds of the electorate begins with a legitimacy problem no amount of governing can fix. The second is quality of candidate. Under a majority requirement, a factional candidate with an intense base and broad opposition cannot win by splitting the field. He has to persuade somebody new.

The mechanism is negotiable. A separate runoff election, as Georgia and Louisiana use, achieves it. Ranked-choice voting achieves it in a single trip to the polls. We take no position on which. The principle is that a majority should be required, and the 1969 amendment — passed by the House with a 40 percent threshold and runoff — shows the idea is neither foreign nor new.

Section 7Reform V: Money with a name on it

The limit on what an individual may contribute directly to a candidate should be removed, and every dollar spent in an American election should be disclosed immediately and attributed to a named human being.

This is the reform most likely to be misread, so we state the reasoning before the proposal. We are not arguing that there is too little money in American politics. We are arguing that the money has been driven into the one channel where nobody has to answer for it, and that the law did the driving.

7.1   What the law actually says

Three facts, because this debate is usually conducted without them.

Corporations have been barred from giving money to federal candidates since 1907. In his annual message of December 1905, President Theodore Roosevelt asked Congress to forbid corporate contributions for political purposes outright. Congress did so in the Tillman Act of 1907, and Roosevelt signed it. That prohibition has never been repealed. It is the law today, and a Republican president put it there.9

Limits on what an individual may give a candidate came in 1974, in the amendments to the Federal Election Campaign Act passed after Watergate. The figure is now $3,500 per election, which for a candidate who runs in a primary and a general comes to $7,000 across the cycle.10

Citizens United, in 2010, did not touch either of those. What the Supreme Court struck down was the ban on corporations and unions spending their own treasury money on independent expenditures — communications that support or oppose a candidate without coordinating with his campaign. Two months later, in SpeechNow.org v. FEC, the D.C. Circuit reasoned that if independent spending cannot corrupt a candidate, then giving money to a group that does only independent spending cannot corrupt him either, and so limits on those gifts had to fall. Citizens United established a right to spend; SpeechNow established a right to raise without limit for the purpose. Together they built the super PAC.11

Anyone who says corporations may now write checks to candidates is mistaken, and the mistake matters. It sends reformers after the wrong target and leaves the actual arrangement undisturbed.

7.2   The inversion

Consider what an American may lawfully do today.

He may give a candidate $3,500 for an election. That gift is reported to the Federal Election Commission with his name, his address, his employer, and his occupation. It is searchable by anyone. The candidate accepted it knowingly and can be asked about it at any town hall in the district.

Or he may give ten million dollars to a committee that will spend it on advertising about the same race. The candidate did not solicit it, cannot legally coordinate with it, and can stand at that same town hall and say, truthfully, that he has nothing to do with it.

The law caps the channel where the money has a name on it, and leaves the channel where nobody answers for anything entirely open.

That is the whole of the problem. A reform passed to reduce the influence of money on candidates made the accountable dollar scarce and the unaccountable dollar unlimited. Nobody designed it that way. It is what happened when one channel was dammed and the other was left open.

The consequence for an ordinary candidate is not subtle. A man or woman who wants to represent a district, who has no fortune and no national list, raises money a few hundred dollars at a time. Against that, a single donor two thousand miles away may spend twenty million on the same race, and no amount of door-knocking closes the gap. You cannot answer ten million dollars three thousand five hundred at a time. Whatever that arrangement is, it is not representative government, and no honest description of it will make it sound like one.

7.3   Who is actually paying

Here the common account is wrong in a way that matters, and correcting it makes the case stronger rather than weaker.

Large corporations are not the main source of this money. Corporations gave roughly $301 million to super PACs and hybrid PACs across the cycles from 2012 through 2018 — about ten percent of those groups' funding at the 2012 peak, falling to roughly five percent by 2018. Only thirty-six companies in the S&P 500 gave $25,000 or more across that entire period. Most large public companies stay out, because the risk of alienating customers is not worth what they would gain.12

The money comes from individuals, and from very few of them. The largest individual donor of the 2023–24 cycle gave $291.5 million, almost all of it through outside spending vehicles. The next four gave $197 million, $148.3 million, $143.5 million, and roughly $100 million. Five people, near enough to nine hundred million dollars. In the 2022 midterms, twenty-one donor families gave $783 million.13 Set that beside the corporate figures above and the shape of the thing is plain: the ban Roosevelt won in 1907 still stands, and it has been rendered beside the point. Spending by individuals independently of campaigns was protected by the Supreme Court in Buckley v. Valeo in 1976, thirty-four years before Citizens United. Overturning the 2010 decision alone would leave the dominant source untouched.11

We say this plainly because it is the same argument this paper has made from its first page. The trouble is not a conspiracy of corporations. It is that a very small number of highly motivated people set the terms of American elections while everyone else is at work, at practice, or at church. In the earlier sections that few showed up at a March primary nobody knew about. Here they show up with a checkbook. Same few, same absence.

7.4   What the job has become

There is a reason capable people leave, and a better reason they never begin.

In January 2013 the Democratic Congressional Campaign Committee gave incoming freshman members a model daily schedule, later obtained and published by the Huffington Post. It allotted four hours a day to what the committees call call time, meaning telephone solicitation of donors. It allotted one hour to strategic outreach. For committee work, floor votes, and constituent meetings combined, it allotted about two.14

That is the official recommendation, from the party, to a person a district has just elected: more than half the working day on the telephone, and a fraction of it on the work. Members describe it as you would expect. One who ran the committee during that period said he had spent more than four thousand hours of his career soliciting donations. Another said he was told his job was to raise eighteen thousand dollars a day, and called the party call rooms a compromise of the dignity of the office.14

Two details in that account deserve more attention than they usually get.

The first is where the calls happen. It is illegal to solicit campaign money from a congressional office, so members walk from the Capitol to call centers run by the party committees, or to rowhouses nearby owned by lobbyists and fundraisers. The party owns the room in which a member of Congress spends most of his working day.

The second is that members owe the party money, and the amounts are set in writing. In October 2022 the House Freedom Caucus warned incoming Republican candidates that every committee assignment carries a specific fundraising quota, and that a member who does not meet it may find himself on a less desirable committee. Representative Ken Buck of Colorado put figures to it: chairs of the most powerful House committees were each expected to raise $1.2 million over two years for the National Republican Congressional Committee, and the Speaker $20 million. Internal Democratic documents reported the same year set the committee chair figure at $1.8 million for the Democratic Congressional Campaign Committee and the Speaker's at $31 million.15

Put those two facts together and the national talking points explain themselves. Nobody has to be threatened. A member whose committee seat depends on his fundraising total, who makes his calls in a room the party provides, from lists the party supplies, to donors the party cultivated, will find that the party's language becomes his language. Not through corruption. Through gravity.

The quotas are not symbolic. Across the 2021–2022 cycle the top Republicans and Democrats on the Appropriations, Energy and Commerce, Financial Services, and Ways and Means committees transferred more than $5.2 million of their own campaign money to the two House party committees. For several of them, roughly one dollar in every five their campaigns spent went not to their own reelection but to the party's effort in other people's districts.15

Character is not a substitute for structure. Ask enough good people to spend the majority of their working lives on a telephone repeating sentences written by somebody else, and most of them will eventually decline. The ones who accept will not be the ones we hoped for.

7.5   Why removing the cap is the provision that does the work

A reader may reasonably ask what stops a party from disciplining a member anyway. If the party controls the primary, and the primary decides the seat, does it matter where the money comes from?

It matters because of what a self-funded candidate can threaten to do.

A member who can raise without limit from people in his own state has somewhere to go. He can tell his party that if it runs a challenger against him for the offense of thinking for himself, he will run anyway, without them, and take his voters with him. A member capped at a few thousand dollars per donor cannot make that threat, because he has no way to fund the campaign it implies. He complies, or he retires. Those are the only two doors, and the country has watched a great many capable people walk through the second one.

The threat does not have to be carried out to work. It changes what the party is willing to attempt, what outside groups calculate is worth spending against him, and how much a faction that turns out in August can extract from a person who represents everyone in November. This is ordinary leverage of the sort every American understands from his own working life. A man who can afford to quit is treated differently than a man who cannot, and he does not have to quit for that to be true.

It has been done. Senator Joseph Lieberman of Connecticut lost his party's primary in 2006 after breaking with its base over the Iraq war, ran in the general election as an independent, and won. Senator Lisa Murkowski of Alaska lost her party's primary in 2010 to a challenger backed substantially by money from outside the state, ran as a write-in candidate, and won.16 In each case a sitting senator rejected by the electorate that shows up in a primary went to the whole state and was returned to office.

We should be honest about what this mechanism does and does not accomplish. It protects a representative who is broadly acceptable to his state but unacceptable to the faction that dominates a low-turnout primary. It does nothing for one whom the state as a whole has turned against, and it should not. That is the point rather than a limitation: the exit option restores the general electorate's judgment over the primary electorate's, which is the same correction this paper has argued for since Section 2, applied to money instead of to the calendar.

Two practical notes for anyone drafting the enabling law. Most states have sore-loser statutes barring a defeated primary candidate from the general ballot, which means the threat must be made before filing rather than after losing. And independent ballot access generally requires gathering signatures, a burden money can overcome but which should be set at a level a genuine candidate can meet. Note also that an independent candidacy produces a three-way general election, and Senator Murkowski won hers with a plurality rather than a majority — precisely the circumstance Reform IV exists to address, and a further reason these repairs belong together.

Section 8Reform VI: Money that comes from home

All money in an election should come from individual human beings residing in the district or state of the office sought — not corporations, not unions, not trade associations, not political action committees, and not party committees — and the same requirement should apply to independent expenditures.

A person who cannot vote in an election may still speak about it. He may not finance the candidate in it, and no organization may finance him at all. Representation means answering to the represented, and a representative who is paid for by strangers will in time speak to strangers.

The provision reaching independent expenditures is the one that makes the rest work. Without it, national money simply routes around the restriction into the one channel left open, and the reform makes the disease worse. Any version of this that reaches contributions but not independent spending should be voted down by its own supporters.

8.1   Why the party committees are included

This is the most contested line in the proposal and we will not soften it.

An organization does not have neighbors. It cannot be voted out, and it does not have to live under what it purchases. The party committee that supplies a member with a call room, a donor list, and a fundraising quota is not an abstraction in this argument. It is the specific mechanism described in Section 7.4 by which a district's representative comes to speak somebody else's sentences. A reform that leaves it in place has not reached the thing that is actually wrong.

We know this cuts against a serious finding, and we set it out against ourselves in Section 9.7 rather than hoping the reader misses it.

8.2   Oregon tried this, and lost

In 1994 Oregon voters passed Ballot Measure 6, which permitted candidates to accept contributions only from individuals residing in the district of the office sought. It is close to what we propose. In VanNatta v. Keisling the Ninth Circuit struck it down in 1998.17

The court's reasoning is what every supporter of this reform needs to understand. Oregon offered two justifications. The first was preventing corruption, which the court accepted as a legitimate interest but found the measure too blunt to serve, since it barred all out-of-district money regardless of amount or any indication of corruption. The second was that the measure protected a republican form of government by ensuring that representatives are chosen by those they represent. That is our argument, nearly word for word. The court rejected it, holding that a right to a republican form of government had never been recognized as a sufficiently important state interest. A state may prevent nonresidents from voting in a district, the court held, but not from expressing themselves about the election — and money counts as expression.17

We do not think that reading is obviously wrong as a matter of existing law. We think it means the existing law is inadequate to the country we now live in, and that the remedy is the one the Constitution provides for exactly this situation.

The Framers wrote in a country where a candidate's money and a candidate's audience were necessarily local, because nothing traveled faster than a horse. They did not contemplate a national advertising market, a media environment reaching every household continuously, or private fortunes large enough to fund a congressional race out of pocket interest. Article V exists because they knew they could not foresee everything. Using it is not a departure from their design. It is their design.

8.3   What nationalization actually is

Every reader has noticed that candidates for county office now campaign on national arguments. The observation has been tested.

Daniel Hopkins documented it in The Increasingly United States, which won the American Political Science Association's Converse Award. The correlation between how a person votes for governor and how the same person votes for president rose from about 0.6 to above 0.8 since 1990. The home-state advantage presidential candidates once enjoyed began declining after the mid-1970s. Americans now know more about Washington than about their own statehouse, and they vote accordingly.18

We owe the reader a caution here, because the honest answer complicates our case. Hopkins finds the leading cause is not campaign money. It is the collapse of the audience for state and local news. As Americans moved from local papers and local television to cable and the internet, they moved to sources that cover national politics and little else. Behavior nationalized before the money followed it.18

We therefore do not claim that fixing the money will restore local politics by itself. It will not. What we claim is narrower and still worth having: a candidate who must raise his money at home has a reason to be at home, in living rooms and union halls and church basements, asking people what they need. That is a partial substitute for what the local newspaper used to do, and at present nothing else is offering one.

8.4   Why we do not propose public financing

Arizona has run its Clean Elections program since 1998, and the evidence on it is genuinely divided. One study finds that public funding increases candidate polarization by reducing the influence of moderate donors, and another finds publicly financed candidates more ideologically extreme and less representative of their districts. Against that, two further studies find essentially no difference in how publicly funded legislators vote once elected. Readers should know the question is contested.19

What is less contested is that the program changed who runs. A review in the Arizona State Law Journal found that Clean Elections significantly weakened the control party elites held over candidate selection and strongly encouraged non-incumbents to enter primaries.19 Whether that is healthy competition or something else is a judgment each reader will make.

Ours is this. Asking your own neighbors to fund you is itself a test — not of whether you know wealthy people, but of whether anyone in your own community will back you at all. A system that lets a candidate skip that test will seat people who could not have passed it. We would rather remove the obstacles to raising money at home than have the government supply it.

Section 10What we are not claiming

Structural reform is not a theory of everything and we will not oversell it.

These six repairs would not end disagreement, and they are not meant to. Americans disagree about immigration, spending, and the use of force because those are hard questions, not because the ballot is badly designed. A country with perfect election rules would still argue.

What we claim is narrower. The current rules select for a particular kind of politician — one who answers to a primary electorate, in a district drawn to be safe, elected on a day most people did not know about, having persuaded a plurality rather than a majority. Change those four conditions and you change what a successful political career requires. You do not change what Americans believe; you change who has to listen to them.

That is Perot's point about the screwdriver, and it is the modest version of it. Open the hood, repair the mechanism, and a self-governing people will do a better job with the arguments than the present arrangement lets them.

Section 11The argument you can carry

The paper compressed to what a person can remember and repeat.

And the one that is ours rather than theirs. The complaint that government is broken is very often made by people who did not vote in the election that produced it. The extremes did not seize anything; they showed up, and the rest of us did not. Every repair in this paper is worth nothing without the hour of attention it is designed to make possible.

Section 12Conclusion

A self-governing people gets the government its rules select for. Ours select for the furious, the organized, and the unopposed, and then we are surprised by what arrives.

None of the six repairs proposed here is radical, and none is new. Three of the four have been law somewhere in the United States or in comparable democracies for decades. The fourth passed the House of Representatives in 1969 with a Republican president's endorsement and died to a filibuster. This is not a case for reinventing the republic. It is a case for maintenance that is eighty years overdue.

What such repairs would produce is not agreement. Americans will go on disagreeing about spending, borders, and war, because those questions are genuinely hard. What would change is who has to answer for the disagreement: whether a representative can win by satisfying a narrow slice of an off-year electorate in a district built for him, or whether he has to go stand in front of everyone else and make a case.

That is the whole of the proposal. Make the people who represent us face the people we actually are — the ones at the ball game, at the job site, in the pews. They are reasonable. Given a system that required someone to persuade them, we suspect the government would start to look a good deal more like them.