“The electoral college is a disaster for a democracy.” Donald Trump wrote this on Twitter on 6 November 2012, the day after Barack Obama’s re-election. Four years later, having lost the popular vote but won the presidency himself, he took the opposite view.
“The Electoral College is actually genius in that it brings all states, including the smaller ones, into play. Campaigning is much different!”
Donald Trump, tweet, 15 November 2016
Trump’s reversal is a useful way into Presidential elections in the USA, because it captures a pattern that recurs across every strand of the topic: how a rule looks tends to depend on whether it currently favours you. The same pattern appears in how campaigns fund themselves, in how incumbents defend their record, and in how the parties argue over reforming the Electoral College itself. This page draws the topic’s separate strands, the structure of American elections, the Electoral College, campaign finance and incumbency, into a single fifteen-minute read, using the 2020 and 2024 elections and the early stages of the 2028 cycle as running examples throughout. It closes by returning to Trump’s two tweets to test whether his 2016 verdict still holds up in 2026.
1. The Structure of US Presidential elections
Before looking at the Electoral College in detail, it helps to see where the presidency sits within the wider American electoral system. The Constitution creates only the presidency directly; every other office below is a separate contest with its own electoral system, several of which shape the path to the White House long before the general election itself.
| Level | Electoral system used | Relevance to the presidency |
| Local government | Mostly plurality (first past the post); a growing number of cities, for example New York, use ranked-choice voting | Traditional entry point for a political career and local party organisation |
| State legislatures | Single-member plurality in most states | Control congressional redistricting, shaping House competitiveness |
| Governorships | Plurality in most states; majority run-off in Georgia and Louisiana | A common launch pad for presidential candidacies |
| US House of Representatives | Single-member plurality, except Maine and Alaska, which use ranked-choice voting | Midterm results are widely read as a referendum on the president |
| US Senate | Statewide plurality; run-off in Georgia; ranked-choice voting in Maine and Alaska | Senate control shapes a president’s ability to legislate and confirm appointments |
| Presidential nomination | Primaries and a small number of caucuses; delegates allocated proportionally or winner-take-all depending on state and party | Determines each party’s nominee before November |
| Presidential general election | The Electoral College: 538 electors, winner-take-all in 48 states and Washington DC, congressional district method in Maine and Nebraska | An indirect election in which the national popular vote does not determine the outcome, as in 2000 and 2016 |
Source: adapted from AQA 7152 (3.2.1.5) and Edexcel 3A.5.1 specification content.
- The Constitution supplies only the skeleton of this process: Article II establishes the presidency, a four year term and the Electoral College, while the Twelfth Amendment (1804), Twentieth Amendment (1933) and Twenty Second Amendment (1951) fix how electors vote, when a term begins, and how many terms a president may serve.
- Everything that fills the gap is party practice rather than law: the invisible primary, the sequence of primaries and caucuses, and the national conventions appear nowhere in the constitutional text.
- As of July 2026, the country is in the invisible primary for 2028, roughly twenty seven months before polling day, with the shape of the field still heavily dependent on this coming November’s midterm elections.
| Key terms Electoral College: the body of 538 electors, allocated by state, that formally elects the president and vice president rather than voters doing so directly. Invisible primary: the informal period of fundraising, positioning and organisation before any candidate formally declares a campaign. Delegate: a party representative who formally nominates a candidate at a national convention, having been allocated through primaries or caucuses. Faithless elector: an elector who votes contrary to their state’s popular result; rare, and now closely restricted by law. |
| Check your understanding 1. Using the table, explain why control of state legislatures matters to the presidential race even though state legislators are not on a presidential ballot. 2. “The primaries and caucuses that decide each party’s nominee are constitutional requirements.” Explain why this statement is inaccurate. 3. Suggest one reason why the 2026 midterm elections are being treated as significant for the 2028 presidential race, even though the presidency itself is not being contested. |
2. The Electoral College: mechanics and controversy
The Electoral College was devised at the Constitutional Convention of 1787 as a compromise between direct popular election, distrusted by many delegates as vulnerable to demagogues, and selection by Congress, which risked making the president dependent on the legislature. Winning requires an absolute majority, 270 of the 538 votes, allocated to each state according to its total congressional delegation, so even the smallest states receive at least three. Forty eight states and Washington DC award all their electors to whichever candidate wins the state; Maine and Nebraska instead split most of their votes by congressional district, a method that gave Kamala Harris a single electoral vote from Nebraska’s 2nd district in 2024 even as she lost the state overall.
Because electors are pledged rather than directly bound by the Constitution, faithless electors, who break from their state’s result, remain a theoretical possibility: seven did so in 2016, the most since 1972, though the underlying result was unaffected. The Supreme Court closed off much of this gap in Chiafalo v Washington (2020), ruling unanimously that states may legally require, and punish breaches of, elector pledges.
The system has produced five elections in which the Electoral College outcome diverged from the national result: in 1876, 1888, 2000 and 2016 the College winner lost the national popular vote, and in 1824 no candidate secured a College majority at all, so the House of Representatives chose John Quincy Adams despite Andrew Jackson leading in both measures.
In 2000, Bush v Gore ended a 36-day Florida recount battle by a final margin of just 537 votes, handing George W. Bush a 271 to 266 College win despite Al Gore’s half-million vote popular lead.
In 2016, Donald Trump won the College while losing the popular vote to Hillary Clinton by almost three million votes. After Trump’s 2020 defeat, his campaign organised slates of 84 self-styled “alternate” electors across seven states who signed certificates falsely declaring he had won, a scheme that fed directly into the Capitol breach of 6 January 2021 and prompted the Electoral Count Reform Act of 2022, which clarified the vice president’s purely ceremonial role in certification. In 2024, by contrast, Trump won both the College, 312 to 226, and the popular vote, 49.8 per cent to 48.3 per cent, the first Republican to win the popular vote since 2004.
| Strength | Weakness |
| Reflects the USA’s federal structure, guaranteeing every state a voice | Small states are over-represented: a Wyoming voter’s weight is around 3.6 times a Californian’s |
| Winner-take-all sustains a stable two-party system | The same mechanism can turn a narrow win into a landslide, or hand the presidency to the popular-vote loser, as in 2000 and 2016 |
| Confines contested counts to a small number of close states, producing swift, decisive results | Concentrates campaigning on a shrinking number of swing states, leaving most voters with limited influence |
| No alternative has attracted the sustained, cross-party consensus needed to replace it | The National Popular Vote Interstate Compact holds 18 states and DC, 222 of the 270 votes needed, but reform has stalled largely along partisan lines |
Source: Gallup (September 2024); Pew Research Center (2024); Ballotpedia.
- Public opinion has shifted since the 1968 near-consensus for change: a September 2024 Gallup poll found 58 per cent of Americans favoured electing the president by national popular vote against 38 per cent preferring the current system, with a Pew Research survey putting support for change at around 65 per cent, and the partisan gap now standing at roughly 80 per cent of Democrats against 46 per cent of Republicans.
- Party self-interest tracks this divide closely: every state that has joined the National Popular Vote Interstate Compact has been Democratic-leaning, and in 2025 Maine’s Republican-controlled House voted 76 to 71 to withdraw from it, only for the Democratic-controlled Senate to block the repeal 18 to 16.
| Check your understanding 1. Explain, using named examples, how the Electoral College’s winner-take-all method can produce a result different from the national popular vote. 2. Using the evaluation table, assess whether the Electoral College still serves the purpose the Framers intended in 1787. 3. Why might a state’s decision to join, or leave, the National Popular Vote Interstate Compact depend more on party advantage than constitutional principle? |
3. Money in American elections
In 2003, the Supreme Court heard McConnell v Federal Election Commission, a challenge to the Bipartisan Campaign Reform Act (BCRA) of 2002. In upholding most of the Act, Justices Stevens and O’Connor observed that “money, like water, will always find an outlet”, a line that has come to summarise the whole history of American campaign finance: whenever the rules on political money have tightened, new channels have generally opened to let money flow by other means.
Modern rules date from the Federal Election Campaign Act (FECA) of 1974, passed after Watergate, which introduced contribution limits, disclosure and voluntary public funding. Buckley v Valeo (1976) upheld those limits but struck down mandatory spending caps, establishing that spending on political speech is itself a form of protected speech.
Presidential campaign spending has grown from $66.9 million in 1976 to $5.5 billion in 2024, with two turning points: candidates from 2000 onwards increasingly declined public funding, whose spending caps became a disadvantage once private fundraising could outstrip it, and Citizens United v FEC (2010), together with SpeechNow.org v FEC (2010), allowed unlimited independent spending and created the Super PAC. Barack Obama became the first major-party nominee to decline public funding for the general election in 2008; no nominee from either party has accepted it since.
| Key terms Hard money: donations raised and spent within federal contribution limits and disclosure rules; an individual may currently give up to $3,500 to a candidate’s campaign committee per election. Soft money: money raised and spent outside those limits for generic party-building activity, banned by BCRA in 2002. PAC (Political Action Committee): an organisation, often linked to a business or trade union, that pools donations to give directly to candidates, capped by law at $5,000 per candidate per election, a limit unchanged since 1974. Super PAC: an independent group that may raise and spend unlimited sums, provided it does not coordinate directly with a candidate’s campaign. Hybrid PAC: also called a Carey Committee, running two accounts side by side: one a capped PAC, the other an unlimited Super PAC. |
| Year | Reform or ruling | Effect |
| 1974 | Federal Election Campaign Act (FECA) | Created the FEC; introduced contribution limits, disclosure and voluntary public funding |
| 1976 | Buckley v Valeo | Upheld contribution limits and disclosure but struck down mandatory spending limits as a First Amendment violation |
| 2002 | Bipartisan Campaign Reform Act (BCRA) | Banned national parties from raising soft money; upheld the following year in McConnell v Federal Election Commission |
| 2010 | Citizens United v FEC and SpeechNow.org v FEC | Allowed unlimited independent spending by corporations, unions and individuals, creating the Super PAC |
| 2014 | McCutcheon v FEC | Struck down the aggregate limit on how much a donor may give across all candidates and committees combined |
| 2014 | Congress ends convention public funding | Left only unused primary matching funds and a general election grant no nominee has claimed since 2008 |
| 2026 | NRSC v FEC | Struck down federal limits on party spending coordinated with its own candidates, overturning a 2001 precedent; decided 30 June 2026 by a 6–3 majority |
Source: FEC.gov, OpenSecrets, Ballotpedia, Brennan Center for Justice.
- Each closed channel has reopened elsewhere: soft money after Buckley v Valeo, tax-exempt “527” groups after BCRA, such as Swift Boat Veterans for Truth, which raised over $25 million against John Kerry in 2004, and Super PACs and undisclosed “dark money” after Citizens United v FEC, and now unlimited coordinated party spending after NRSC v FEC.
- In 2024, Future Forward, a hybrid PAC backing Kamala Harris, raised over $950 million, more than the combined cost of every presidential election held between 1976 and 1996, while Make America Great Again Inc., a Super PAC backing Donald Trump, raised $410.5 million; the businessman Elon Musk alone gave an estimated $277 million to Trump-aligned groups.
- Money is not decisive on its own: Harris and allied groups raised roughly $2 billion in 2024 against about $1.45 billion supporting Trump, and still lost; Hillary Clinton outspent Trump in 2016 and lost too, suggesting money is closer to a necessary than a sufficient condition for a competitive campaign.
| Check your understanding 1. Explain what the phrase “money, like water, will always find an outlet” means, using one example from the timeline. 2. To what extent does the 2024 case study (Future Forward and Make America Great Again Inc.) suggest that the legal label attached to an outside spending group matters less than its scale? 3. Using named examples, evaluate the claim that “US presidential elections are decided by whoever spends the most.” |
4. Incumbency: a significant but conditional advantage
Incumbency advantage, the electoral benefit an officeholder derives simply from already holding office, rests on retrospective voting theory: the idea that voters judge an incumbent chiefly on the record of the preceding term, treating the election as a periodic referendum on the sitting administration rather than a forward-looking choice between programmes. Since George Washington established the norm of a voluntary two-term limit, 28 incumbents have stood for re-election; 18 won and 10 lost, a success rate of roughly two in three. Grover Cleveland remains the only president to lose a re-election bid, in 1888, and later regain the office, giving the United States its only non-consecutive presidency before Donald Trump’s return in 2025.
| Advantage | Disadvantage |
| Name recognition and free media coverage: Obama’s official activities in 2012 routinely led the news, coverage Romney had to buy | Blame for events beyond their control: Carter blamed for the Iran hostage crisis in 1980, Trump for Covid-19 and recession in 2020 |
| Easier fundraising and a united party, sparing a contested primary: Obama in 2012, Bush in 2004 | Heavy media scrutiny cuts both ways: Biden’s faltering 27 June 2024 debate crystallised doubts about his fitness and led to his withdrawal |
| Executive experience and the rally round the flag effect: Bush’s 2004 campaign benefited from the residual effect of 11 September | A fundraising advantage does not guarantee victory: Bush senior entered 1992 well-funded yet lost to Clinton amid recession |
| Voter risk-aversion favours the known quantity: Reagan’s 1984 “Morning in America” campaign | Primary challenges can weaken a surviving incumbent: Ford in 1976 and Carter in 1980 both survived challenges, then lost in November |
Source: historical election results compiled from the National Archives and 270toWin.
The advantage transfers poorly to same-party successors who have not themselves held the presidency, a quasi-incumbency problem visible across Richard Nixon in 1960, Hubert Humphrey in 1968, Al Gore in 2000, Hillary Clinton in 2016 and Kamala Harris in 2024, all of whom lost; the sole modern exception is George H. W. Bush’s 1988 victory. Harris entered the race as a quasi-incumbent after Joe Biden withdrew on 21 July 2024, running a compressed fifteen-week campaign that tried to combine continuity with the Biden administration and distance from its least popular positions, a balancing act several commentators judged incoherent; she lost all seven swing states and the popular vote.
Term limits guarantee that, roughly every eight years, the presidency is contested as a genuinely open seat, as in 2000, 2008, 2016 and, in practical terms, 2024. Structural advantage is also no protection against a large enough shock: Hoover in 1932, Ford in 1976, Carter in 1980, Bush senior in 1992 and Trump in 2020 each lost despite the usual incumbency advantages, undone by depression, stagflation, recession or pandemic respectively. The six year itch, the tendency for a president’s party to suffer heavy congressional losses in the sixth year of a two-term presidency, as in 2014 and 2018, shows that incumbency at the top of the ticket offers little protection further down the ballot, a pattern the 2026 midterms will test again.
| Check your understanding 1. Explain what is meant by “retrospective voting” and how it links to incumbency advantage. 2. Using named examples, explain why incumbency advantage transfers poorly to a quasi-incumbent such as a sitting vice president. 3. “Incumbency guarantees re-election.” Using the historical record (18 wins from 28 attempts), evaluate this claim. |
5. Bringing it together: 2020 and 2024
The 2020 and 2024 cycles show every strand above operating together in a single contest. In 2020, Donald Trump was a true incumbent, benefiting from the fundraising and unified-party advantages incumbency usually confers, yet the rally round the flag effect conspicuously failed to appear amid the Covid-19 pandemic; Joe Biden won 306 electoral votes to Trump’s 232 and the popular vote by 51.3 per cent to 46.8 per cent, on turnout of around 66.6 per cent, the highest in over a century. Trump’s subsequent legal challenges and the fake electors scheme exposed real gaps in the certification process, later closed by the Electoral Count Reform Act of 2022, passed after the Capitol was breached on 6 January 2021.
In 2024, incumbency worked differently again. Biden’s withdrawal left Harris as a quasi-incumbent, and despite Future Forward’s record-breaking operation on her behalf, she lost all seven swing states, becoming the third Democratic vice president in a row to fail in this position. Trump’s 312 to 226 College win, and his win of the popular vote for the first time by a Republican since 2004, meant the certification on 6 January 2025, with Harris herself presiding as vice president, passed off peacefully, a marked contrast with four years earlier.
- Both cycles were shaped by unexpected outside shocks: the pandemic and Black Lives Matter protests in 2020, an assassination attempt and a late candidate substitution in 2024.
- Both raised serious questions about a candidate’s fitness for office: Trump’s conduct and legal jeopardy in 2020, Biden’s age and capacity in 2024.
- The reforms passed after 2020, chiefly the Electoral Count Reform Act, appear so far to have held, though it remains an open question whether this reflects a genuine restoration of democratic norms or simply Trump’s 2024 victory removing his own incentive to contest the result.
6. The Presidential Election system now: the 2028 cycle
As of July 2026, the country sits at the very start of the 2028 cycle, in the invisible primary phase, around twenty seven months before polling day. On the Republican side, Vice President JD Vance remains the clear frontrunner in primary polling, at around 48 per cent among likely Republican primary voters in one recent survey, but Secretary of State Marco Rubio has recently overtaken him as favourite on prediction markets such as Kalshi, having trailed by around 25 points a year earlier, a reminder that an early polling lead is not the same as a settled nomination. On the Democratic side, no comparable frontrunner has emerged: Kamala Harris leads several polls of prospective primary voters, with Gavin Newsom, Alexandria Ocasio-Cortez and Pete Buttigieg all drawing measurable support.
Trump himself remains constitutionally barred from a third term under the Twenty Second Amendment, notwithstanding his own repeated suggestions to the contrary, including musing about a “Trump 2028” cap and a mooted plan to run as vice president alongside Vance before assuming office again; Trump has himself acknowledged the text is “pretty clear” that he is “not allowed to run.” The 2026 midterm elections, contesting the full House, roughly a third of the Senate and numerous governorships, will offer the first real test of whether the six year itch reasserts itself against the Republican Party, and will shape the field considerably before any 2028 candidate formally declares.
Evaluating the system as a whole
Taken together, these five strands support a single, cautious conclusion: the American electoral system’s formal, constitutional architecture has changed remarkably little since 1787 and 1804, while the informal system built around it, party primaries, campaign money and the conventions of incumbency, has absorbed most of the pressure for change instead. The Electoral College has survived over two centuries of criticism and two elections this century in which it diverged from the popular vote, not because reformers lack a workable alternative, but because reform requires either a constitutional supermajority no Congress has assembled, or a state-by-state compact that has stalled at 222 of the 270 votes it needs, with party self-interest, visible even in Trump’s own reversal from “disaster” to “genius”, blocking further progress in either direction. Campaign finance shows the same pattern in a different register: fifty years of reform since FECA have repeatedly relabelled the channels through which political money flows, from soft money to 527 groups to Super PACs to, as of 30 June 2026, unlimited coordinated party spending under NRSC v FEC, without ever succeeding in reducing the total amount moving through them.
Does Trump’s 2016 verdict still hold in 2026? On the Electoral College, the honest answer is that it has not yet been tested against him again: 2024 delivered him both the College and the popular vote together, so his “genius” verdict has not had to survive a second reversal of fortune, and there is little in the historical record, seven presidents have called the system a disaster only after losing under it, to suggest he would not flip back if 2028 or 2032 produced a College result at odds with the popular vote for a Republican candidate. On the money side, the picture is more clear-cut: the Stevens and O’Connor line has, if anything, been reinforced rather than tested since 2003, most recently by a Supreme Court that struck down one of the last remaining federal spending limits a month before this document was written. The water, in short, is still finding its outlet, and so, self-interest allowing, is Donald Trump’s own judgement of the rules that govern it.
| Check your understanding 1. Using evidence from at least two sections of this document, assess how far Donald Trump’s changing verdicts on the Electoral College illustrate a wider pattern in this topic. 2. “The US electoral system is essentially unreformable.” Evaluate this claim, drawing on both the Electoral College and campaign finance sections. 3. Looking ahead to 2028, explain which of the concepts covered in this document, incumbency, quasi-incumbency, the invisible primary, or campaign finance, you consider most likely to shape the outcome, and why. |
A fuller comparison with the UK’s capped, disclosed system of party funding and its first-past-the-post electoral system is developed elsewhere on histpology.co.uk.
Sources
Primary sources
U.S. Constitution, Article II and Amendments XII, XX, XXII and XXV https://constitution.congress.gov/constitution/
3 U.S. Code § 15, Counting electoral votes in Congress (Cornell LII) https://www.law.cornell.edu/uscode/text/3/15
Chiafalo v Washington, 591 U.S. ___ (2020) https://supreme.justia.com/cases/federal/us/591/19-465/
Bush v Gore, 531 U.S. 98 (2000) https://supreme.justia.com/cases/federal/us/531/98/
Buckley v Valeo, 424 U.S. 1 (1976) https://supreme.justia.com/cases/federal/us/424/1/
McConnell v Federal Election Commission, 540 U.S. 93 (2003) https://supreme.justia.com/cases/federal/us/540/93/
Citizens United v FEC, 558 U.S. 310 (2010) https://supreme.justia.com/cases/federal/us/558/310/
SpeechNow.org v FEC, 599 F.3d 686 (D.C. Cir. 2010) https://www.fec.gov/legal-resources/court-cases/speechnoworg-v-fec/
McCutcheon v FEC, 572 U.S. 185 (2014) https://supreme.justia.com/cases/federal/us/572/185/
Carey v FEC, 791 F. Supp. 2d 121 (D.D.C. 2011) https://www.courtlistener.com/opinion/2146537/carey-v-federal-election-comn/
NRSC v FEC, 609 U.S. ___ (2026) https://www.fec.gov/updates/supreme-court-finds-limits-on-coordinated-party-expenditures-unconstitutional-in-nrsc-v-fec-609-us-____2026/
Secondary sources
Federal Election Commission, campaign finance statistics and historical reports https://www.fec.gov/data/
OpenSecrets, “Cost of Election” https://www.opensecrets.org/elections-overview/cost-of-election
OpenSecrets, “Dark Money Basics” https://www.opensecrets.org/dark-money/basics
Ballotpedia, Federal Election Campaign Act (FECA), legislative history https://ballotpedia.org/Federal_Election_Campaign_Act
Congress.gov, legislative history of the DISCLOSE Act, the For the People Act and the Freedom to Vote Act https://www.congress.gov/
Brennan Center for Justice, research and analysis on campaign finance and FEC enforcement https://www.brennancenter.org/issues/reform-money-politics
Bipartisan Policy Center, “Trends in Campaign Financing, 1980–2016”, Zachary Albert https://bipartisanpolicy.org/wp-content/uploads/2019/05/Trends-in-Campaign-Financing-1980-2016.-Zachary-Albert..pdf
National Archives, U.S. Electoral College results by year https://www.archives.gov/electoral-college
270toWin, presidential election results by year https://www.270towin.com/
Pew Research Center, “Behind Biden’s 2020 Victory” https://www.pewresearch.org/politics/2021/06/30/behind-bidens-2020-victory/
Pew Research Center, “Voter turnout in the 2020 and 2024 elections” https://www.pewresearch.org/politics/2025/06/26/voter-turnout-2020-2024/
Gallup, “Americans Still Favor Replacing Electoral College System”, September 2024 https://news.gallup.com/poll/651353/americans-favor-replacing-electoral-college-system.aspx
Wikipedia, “2020 United States presidential election” https://en.wikipedia.org/wiki/2020_United_States_presidential_election
Wikipedia, “2024 United States presidential election” https://en.wikipedia.org/wiki/2024_United_States_presidential_election
Protect Democracy, “Understanding the Electoral Count Reform Act of 2022” https://protectdemocracy.org/work/understanding-the-electoral-count-reform-act-of-2022/
V. O. Key Jr, The Responsible Electorate: Rationality in Presidential Voting, 1936 to 1960 (1966)
Douglas A. Hibbs, ‘Bread and Peace Voting in US Presidential Elections’, Public Choice (2000) https://link.springer.com/article/10.1023/A:1005292312412
Ray C. Fair, Predicting Presidential Elections and Other Things (Stanford University Press) https://fairmodel.econ.yale.edu/index1.htm
Newsweek and Zogby Analytics, polling on 2028 presidential contenders, July 2026 https://www.newsweek.com/jd-vance-aoc-gavin-newsom-2028-poll-12194158
CNBC, “Rubio odds for GOP 2028 nominee close to overtaking Vance on Kalshi”, June 2026 https://www.cnbc.com/2026/06/01/rubio-odds-for-gop-2028-nominee-close-to-overtaking-vance-on-kalshi.html
Federal News Network, “Checking in on Presidential Election Odds for 2028”, July 2026 https://federalnewsnetwork.com/prediction-markets/2026/07/checking-presidential-election-odds-2028/