How to Build a Supermajority
For the past fifteen years, American politics has been in a perpetual state of hyper-polarization. Our two political parties trade narrow congressional majorities, make incremental policy changes, and then are swiftly voted out of power. This creates a cycle of stasis and stagnation that causes Americans’ already-low faith in their government and institutions to further deteriorate. The solution is to build a supermajority that can pass policy change that is durable and improves the lives of the American people.
We aim to show what it looks like to build a supermajority coalition, one that can win in broad swaths of the country and expand the map of competitive states outside the handful of battlegrounds. That means going beyond traditional polling to give leaders and policymakers a more textured understanding of what divides the American electorate, what unites them, and how issues can be best prioritized to build a durable governing majority.
To do this, we ran an expansive survey designed to measure fundamental beliefs across various issue areas and values. As a result, we have come to a new and timely understanding of the American electorate—one that will chart a course for leaders to build a supermajority. Click here for a full overview of the project.
The most important takeaway from our work is that the vast majority of Americans hold a heterodox set of opinions that cannot be neatly categorized as liberal or conservative. Specifically, our data shows that only 13% of the electorate is completely ideologically conservative and 9% is completely ideologically liberal. We also found that voters have deeply held views on many issues, which influence the sort of candidates they are willing to support over and above partisanship.
The path to building a supermajority will involve deviating from party orthodoxy and running on a platform that matches the ideological diversity of the American electorate.
The 5 Axes of Public Opinion
Our expansive survey covered various topics, ranging from gender norms to immigration. After analyzing this dataset, we found that labels like Democrat or Republican and conservative or liberal only partially explained the variation in voter behavior. Instead, we found that the American electorate breaks down along 5 axes. These are ordered by how strong their effect is in dividing the electorate, but the differences are relatively small.
The axes are:
- Gender and gender roles
- America’s international role
- Parental authority and the family
- Economic role of the government
- Order vs openness
To build these axes, we aggregated survey responses on the following topics:
Gender and Gender Roles
Should the government enforce traditional gender roles, encourage marriage, or make rules based on gender at birth? While the most conservative voters take the fully pro-gender-role position, most other voters fall in the middle. Notably, all but the most right-leaning groups reject government promotion of gender roles, while only the most left-leaning groups want participation in sports to be based on gender identity.
International Role
How much should the U.S. focus on global affairs? What responsibility do we have to the rest of the world? The issues in this axis are less salient, but there is a large underlying split across the electorate that isn’t fully explained by partisanship. Isolationism is highly correlated with personal economic stability, as voters who report financial instability are 10 points more likely to say the U.S. should focus on problems at home. The less economically stable, more populist-leaning segments on both sides of the aisle share this view.
Parental Authority and Paternalism
Should the government make rules about how parents interact with their children, either in terms of discipline or privacy? While this axis is a social issue, it doesn’t perfectly correlate with views on gender, as many voters believe in parental authority, while also believing in gender equality. Views on this axis also relate to a general discomfort with the government telling you how to live your life. Liberals and the youngest voters are among the few who believe children have a right to privacy independent of their parents.
Economic Role of the Government
Should Americans be responsible for their own financial wellbeing and success, or does the government have a role? Is it worth limiting economic growth to protect the environment, or to cap the amount of wealth any American can achieve? Almost all voters are comfortable with robust government involvement, and the only segment that believes in limited government intervention in the economy is the most conservative segment.
Order vs Openness
Is it worth tolerating disorder to make sure people are free, or is it more important to promote law and order? To what extent should the government ensure that no one defrauds benefit programs or that people avoid choices that might harm them? This dimension combines libertarian ideas about freedom with more liberal leaning anxieties about abuse of power. Most of the variation is explained by partisanship, hence the low impact of a law and order policy position in our salience testing. But there is a section of the Democratic base that holds strong pro-order beliefs.
Dividing The Electorate
Using these 5 factors, we can divide the electorate into 9 groups. These groups reflect real underlying differences in values and policy opinions, and provide a more complete understanding of ideological division than party identification alone.
Liberal Arts Left: The most ideologically homogenous and progressive group. They are highly educated, younger, and report being financially unstable.
Lawn Sign Liberals: This group holds conventional progressive opinions on virtually all issues. They differentiate themselves from the Liberal Arts Left because they oppose wealth caps and are comparatively more favorable to a market-based economy.
The Neighborhood Watch: This group is the most racially diverse and supports mainstream progressive views on economic issues, but less so than the most liberal segments. They stand out among the more left-leaning groups in their low societal trust and support for order.
Trad-Digitial: This group is younger, online, and the most ideologically diverse segment — splitting their 2024 vote almost equally. They are relatively progressive on economics and traditional on culture.
The Center Square: This group is the least likely to be registered to vote and have few strong policy priorities. They are very pocketbook focused and oppose any new taxes.
Sticker-shocked Skeptics: This group is older, offline, and not politically engaged. They lean right on culture and vote mainly on pocket-book issues. They are the most likely to regret their vote for Trump because of inflation.
Stars, Stripes, and Steeple: This group is retired, religious, and southern. They are conservative on most economic issues, but stand out among conservatives for supporting humanitarian aid and more government involvement in providing healthcare.
No Trespassing Populists: This group is very conservative on cultural issues and the most supportive of aggressive deportation policy. But they break from conservatives because they support progressive economic policies like wealth caps.
Cul-de-Sac Conservatives: This is the only group to hold traditional conservative views on all issues, including government involvement in health care. They are the right’s parallel to the Liberal Arts Left.
For more on our segments, see our write up here (and take the quiz to figure out where you might land!)
The most important takeaway from this project is that very few voters hold consistently liberal or consistently conservative positions. The fully conservative and fully liberal segments are 13% and 9% of the electorate, respectively. These segments are also far more likely to say that almost all of their friends hold the same political opinions as them, particularly the most liberal segment, where 43% say “almost all” of their friends hold the same opinions. If you’re working in politics and punditry, you likely fall into one of these ideologically consistent segments. Your experience is not shared by the vast majority of the electorate.
The second most important takeaway is that public opinion isn’t a single linear scale. There’s no single slider between liberal and conservative and no magical ideal point. The issue space is multidimensional, and a politician who is very close to a majority of the electorate on one issue axis can be out of step on another in a way that prevents them from gaining votes. Effective positioning depends on the dynamics of specific races and the salience of the issue. The best way to categorize the most politically advantageous form of strategic positioning is heterodoxy. This means diverging from your political base (that tiny group with consistent opinions!) on issues where they are out of step with the voters you need in your coalition in order to win.
Do Issues Matter?
Yes, absolutely. Our testing is in line with other research which shows that issue positioning can significantly impact vote decisions, over and above partisanship. Voters have issue preferences, and are willing to make tradeoffs to prioritize the issues that matter to them the most.
We’re implicitly using a spatial model of electoral preferences, where being close to a politician in the “issue space” indicates agreement and likelihood of support. While we think this is a broadly accurate model, clearly some issues matter more than others. To help determine what matters the most, we ran a series of experiments to identify the most important issues where politicians should align themselves with voters.
Issue Salience and Impact
We assigned randomized stances to two hypothetical candidates, and asked respondents to choose between them, including the party identification of the candidate. This lets us measure how issues impact vote choice even when party is a factor. Partisanship is the largest driver of vote choice, as we might expect, but issue positioning still has large effects. Results of this testing are reported as a percentage point effect, that is, how many points more likely a candidate is to be chosen if they have the relevant position. A larger effect indicates that it is a higher salience issue.
We focused this experiment on issues that are either very partisan and likely to come up in 2028 campaigns (Abolish ICE), generally politically relevant (Social Security), or showed up in our earlier testing as a major predictor of political attitudes (Keeping government out of parental decisions). These results should be interpreted as directional, as we’d expect different combinations of issues with different wording to vary in outcome.
High Salience Consensus Issues
Issues around children are very salient to voters, and will likely impact vote choice. Voters also do not want to reduce Social Security benefits. Affirmative action is unpopular, and opposing it benefits a candidate among voters in both parties. Voters also support raising taxes on the rich, and react negatively to the opposing position. This is in line with additional evidence which shows that voters no longer favor conservative economic policies.
| Attribute | Position | Effect (percentage point impact on vote choice) | p |
| Children | Government not involved in parental discipline, except abuse or neglect | +9.1 | <.001 |
| Social Security | Reduce benefits to keep tax levels stable | −8.2 | <.001 |
| Affirmative action | Universities may not consider an applicant’s race | +6.8 | <.001 |
| Environment | Household product choices up to individuals, not government | +5.9 | <.001 |
| SNAP | No restrictions on what benefits can buy | +4.2 | <.001 |
| Taxes | Keep taxes lower on high-income individuals | −4.4 | <.001 |
Three Part Issues
While most issues can be phrased as a binary choice between two policy options, some current ones have three clear positions. This makes the findings more complicated, but still interpretable. These issues were military funding to Israel and Immigration and Customs Enforcement. We tested increasing, decreasing, or maintaining military funding to Israel and abolishing, significantly reforming, or increasing funding to ICE.
These two issues have a similar pattern in their results. Both are highly polarized, and the strong effects nearly cancel out to give a smaller overall effect. Strengthening ICE has a very large negative effect with Democrats (-18.7, when “reform” is the control condition), and a small positive one with Republicans (+5.1). Abolishing ICE has the opposite pattern. Reform still performs worse than abolition among Democrats, but this is counteracted by doing far better with all other voters.
On military funding to Israel, the same effect happens, but because it is a less salient issue, there is a smaller electoral impact. The option with the largest positive effect is maintaining current levels of aid, which has positive effects with both parties (+8.9 among Democrats, +1.7 among Republicans, with “increase” as the control condition). Reducing aid performs only slightly better than maintaining current funding levels, even among Democrats, where it shows less than a point of additional effect (+9.6 among Democrats, -0.5 among Republicans). Voters have the strongest reaction to increasing aid (negative among Democrats, positive among Republicans), and the difference between reducing or maintaining aid is minimally salient.
High Salience, but some partisan blowback
Enforcing rules around birth sex in sports and supporting legal abortion nationwide with some limits both have a positive overall effect, but there are some negative responses from the opposite party. Democrats have a weakly negative reaction to requiring children’s sports to be based on biological sex, contrasted with the extremely strong Republican reaction. While Republicans have a relatively strong negative reaction to legal nationwide abortion, this is counteracted by a large gain among Democrats and Independents. These issues are nearly mirror images of each other. It would be electorally beneficial for a Republican candidate to run on their support for legal abortion, and similarly beneficial for a Democratic candidate to run on making children’s sports based on biological sex. The complicated question for candidates is how to balance that general election impulse with backlash from their primary electorate.
| Attribute | Position | Overall Effect (percentage point impact on vote choice) | Among Democrats | Among Republicans | Among Independents |
| Children’s sports | Participation based on biological sex | +10.9 | -1.5 | +21.7 | +14.7 |
| Abortion | Legal nationwide under national law, with some limits | +6.6 | +18.0 | -6.9 | +8.9 |
Low Salience
The cultural issues were overall more salient than specific economic issues, although protecting Social Security and taxing the rich did move votes. Preventing fraud in government benefits was not salient. Saying that Americans should be responsible for their own financial situation did not change voting behavior either. These anti-safety net conservative positions are not salient, even to voters who might not be sympathetic to that view. Positions on law and order (i.e. how much disorder can be tolerated) also did not have much effect overall, though they had some effect on segments that prioritize this issue. Other low salience issues include standing up to China, and a more ambiguous position around being willing to make major changes to institutions. It’s likely that voters did not find the position on change specific enough to really affect their votes, so this may not replicate with stronger language or more specific ideas about institutional reform.
Issue Selection
We know that it’s hard to craft a specific policy agenda based on voters’ general beliefs over issues like toleration of disorder, gender norms, etc. So we crafted a set of issues that can help shape campaign platforms.
Easy Wins
There are some issues in our dataset that are so popular that they’re practically a consensus. Some are campaign issues, some aren’t. But they all provide meaningful insight into the consensus values of the American public concerning the government’s role in society that candidates should consider as they work to build their coalitions.
Consensus issues lean cultural, and include keeping the government out of parenting decisions and maintaining high educational standards for public school students. There’s also a clear desire for guardrails on technology, like AI and social media. Very few voters are willing to cut Social Security benefits, even if it means raising taxes.
Other high-consensus issues worth mentioning include the belief that the national debt is a big problem, at 75%, taxing the rich, at 73%, and protecting the environment at 75%. Although, this issue is complicated by voters’ clear preference for the government to not influence their consumer behavior on issues like purchasing gas stoves and general support for an “all of the above” energy policy in other surveys.
Wedge Issues
In order to build a supermajority, one party will always need to recruit voters from the opposing party. The best way to do this is to identify wedge issues, where one party is united, and the other is divided. If the issue is salient or pivotal to an election, it could move a significant portion of voters in the other party.
Not all of these issues are actually salient (see above). In practice, wedge issues will only work if voters are sufficiently attached to them. While our conjoint experiment can’t give us specific salience for every issue on this list, we’ve generally found that affordability, taxes, and healthcare are highly salient, while issues like gun control and climate change are not.
Wedges for Democrats
The Democratic position on economic policy is very popular and highly salient. Taxing the rich holds majority support in both parties, as does increasing the government’s role in healthcare. Increased background checks on firearms is also a wedge issue, but less influential on vote choice. The percentage of the electorate that supports a small government free market policy is marginal.
Wedges for Republicans
The path to a supermajority
Politicians have flexibility when it comes to building a campaign platform that appeals to the voters in their race. Our goal is to provide enough information that politicians and political strategists can make their own choices about how to advance their moral values and policy goals in a way that reckons with public opinion. We understand that the information presented here is dense and complex, so here are a few outlines of potential approaches.
The most straightforward path to a supermajority is to combine the most popular issue sets. This involves taking a very hands-off approach to personal decision making (allowing parents to decide how they raise their children, not restricting SNAP purchases, and not pressuring people to get married). This could be combined with a commitment to fiscal responsibility and an emphasis on popular economic issues like raising taxes on the wealthy and not cutting Social Security. Ensuring the government plays an active redistributive role in the economy while emphasizing conservative values around personal responsibility and law and order is a very effective combination. On the hot button issues, the best path is to allow multiple views in your coalition, but understand that the middle-ground is the best path to appealing to the largest swath of voters. That means reforming, not abolishing ICE, enforcing rules about sports participation based on birth sex while protecting trans rights, etc. Hard lines on these nuanced issues are not compatible with a supermajority project.
Neither party is currently taking steps to connect to a supermajority of voters. Republicans are most hampered by their anti-tax anti-safety-net approach, and could make significant gains by embracing a larger role for government in supporting citizens. Democrats struggle most with their liberal cultural positions and their association with aggressive government mandates on consumer choice decisions.
There are multiple paths towards a supermajority coalition. But the throughline is that these coalitions must include people with ideologically diverse viewpoints, especially on social issues. Only 13% of the electorate is completely conservative and 9% is completely liberal. There’s no secret coalition of highly ideological voters waiting to be driven to the ballot box; this is the heterodox American electorate.
Appendix
Battery #1
| Attribute | Position | Effect (percentage point impact on vote choice) | p | Marginal Mean | Win Rate |
| Children | Government not involved in parental discipline, except abuse or neglect | +9.1 | <.001 | 54.6% | 56.2% |
| Social Security | Reduce benefits to keep tax levels stable | −8.2 | <.001 | 45.9% | 44.6% |
| ICE | Significantly reform ICE from the ground up | +4.5 | <.001 | 54.4% | 56.0% |
| SNAP | No restrictions on what benefits can buy | +4.2 | <.001 | 52.1% | 52.4% |
| ICE | Strengthen and fund ICE | −3.8 | .002 | 45.9% | 44.7% |
| Regulation | Roll back federal regulations | −1.5 | .098 | 49.2% | 48.5% |
| China | Cooperate more with China | −0.7 | .43 | 49.6% | 49.8% |
| Border | Focus less on border security, more on other issues | −0.3 | .77 | 49.9% | 50.0% |
| Political party | Republican | +0.1 | .95 | 50.0% | 50.0% |
Battery #2
| Attribute | Position | Effect (percentage point impact on vote choice) | p | Marginal Mean | Win Rate |
| Children’s sports | Participation based on biological sex | +10.9 | <.001 | 55.4% | 57.2% |
| Israel | Maintain current levels of military aid | +5.5 | <.001 | 52.0% | 52.1% |
| Israel | Reduce military aid | +5.2 | <.001 | 51.6% | 52.1% |
| Political party | Republican | −4.1 | <.001 | 48.0% | 47.3% |
| Regulation | Roll back federal regulations | −2.5 | .009 | 48.7% | 48.1% |
| Change | Make major changes when institutions aren’t working | +1.6 | .096 | 50.8% | 51.5% |
| Order | Keep society open and free, tolerating some disorder | −1.5 | .12 | 49.2% | 48.5% |
| Immigrants | Don’t tie citizenship or services to English and assimilation | −0.9 | .37 | 49.5% | 49.5% |
| Government role | People mostly responsible for their own finances | +0.3 | .77 | 50.1% | 50.2% |
Battery #3
| Attribute | Position | Effect (percentage point impact on vote choice) | p | Marginal Mean | Win Rate |
| Affirmative action | Universities may not consider an applicant’s race | +6.8 | <.001 | 53.4% | 54.8% |
| Abortion | Legal nationwide under national law, with some limits | +6.6 | <.001 | 53.3% | 54.3% |
| Environment | Household product choices up to individuals, not government | +5.9 | <.001 | 53.0% | 53.7% |
| Taxes | Keep taxes lower on high-income individuals | −4.4 | <.001 | 47.8% | 46.7% |
| Health insurance | Mostly private, with programmes for seniors and the poor | +2.8 | .014 | 51.6% | 52.0% |
| Government programmes | Ensure everyone eligible receives benefits, even with some fraud | +1.2 | .19 | 50.6% | 50.7% |
| Health insurance | Mix of private insurers and government programmes | +0.9 | .42 | 49.7% | 49.9% |
| Regulation | Roll back federal regulations | −0.9 | .36 | 49.6% | 49.6% |
| Political party | Republican | −0.7 | .54 | 49.6% | 49.6% |