The rise of the liberal Democrat
Is passing bolder policies worth holding slimmer majorities?
In June, Gallup released some interesting polling. They found that, over the last three years, the share of Republicans self-identifying as “conservative” or “very conservative” on social issues has increased from 60% to 74%. Independents shifted right as well, albeit to a lesser extent: from 2021 to 2023, the share who called themselves “conservative” or “very conservative” grew from 24% to 29%, while the share identifying as “liberal” or “very liberal” dropped from 27% to 23%.
This is all pretty banal. Views shifting somewhat to the right with a Democrat in the White House is exactly what Christopher Wlezien had in mind when he described public opinion as a thermostat in 1995. But it is interesting that the rightward movement is isolated to Republican and (to a lesser extent) independent voters; the share of Democrats identifying as liberal (62%) and conservative (10%) is unchanged.
This isn’t new: Gallup’s data show that Democrats have been getting more liberal for at least the last 30 years. And while Republicans have gotten more conservative over the same period, conservatives have always been a majority of the GOP — whereas a plurality of Democrats only started identifying as liberals within the last decade.
The shift cannot be explained by thermostatic opinion alone. We’ve had two Democratic presidents (Clinton and Obama) and two Republican ones (Bush and Trump) since the series began. And yet the number of Democrats who call themselves “liberal” has steadily ticked up.
So what’s behind the rise of the liberal Democrat — and why does it matter?
“I think we may have lost the South for your lifetime”
The best place to start is at the end of the Fourth Party System. After the Civil War and Reconstruction, the Democratic Party largely existed as a regional party that controlled the Jim Crow South while the GOP dominated national politics. Then in 1932, the Great Depression swept Franklin D. Roosevelt into the White House. As president, FDR got Congress to pass the New Deal: a sweeping set of programs including new banking regulations, public works programs, and additions to the welfare state like unemployment insurance and Social Security. Because many congressional Democrats of the time — especially those in positions of leadership or influence — were segregationists, some New Deal programs were written to exclude Black people from receiving equal benefits: the Civilian Conservation Corps, as one example, was initially segregated.
But as Matt has pointed out, enough New Deal programs were open or open enough to Black people that they improved the group’s economic fortunes. For example, the Works Progress Administration was crucial in providing jobs to Black workers during the Depression — jobs that paid better than what private employers offered. In fact, the material, economic advantages that the New Deal offered were enough to break Black voters’ loyalty to the Republican Party, despite the fact that Democrats were arguably worse on civil rights at the time.
Herbert Hoover was the last Republican nominee for president to win the Black vote. As Black voters entered the Democratic Party, they began to agitate for civil rights from within. That naturally created some tensions with the Dixiecrat wing; think Hubert Humphrey’s speech about states’ rights vs. human rights at the 1948 DNC or the protests in 1964 over whether to seat the Mississippi Freedom Democratic Party, an integrated delegation which argued that the official, whites-only delegation was in violation of party rules for excluding Black voters from participating in primaries and caucuses.
The tension between “the shadow of state’s rights” and “the bright sunshine of human rights” was resolved when Lyndon Johnson signed the Civil Rights Act of 1964 and the Voting Rights Act of 1965 into law. Unlike JFK, LBJ was a Southerner who had worked with and was trusted by Dixiecrat senators. With a great deal of maneuvering, the support of Republican minority leader Everett Dirksen of Illinois, and a dash of the “Johnson treatment,” he got the bills through Congress. On the morning LBJ signed the Civil Rights Act, while he was reading the paper on a plane, journalist Bill Moyers said to him, “Quite a day, Mr. President.” Johnson replied, “Well, I think we may have lost the South for your lifetime — and mine.”1
David Shor and Matt like to (correctly) point out that codifying civil rights laws was popular with voters in the 1960s. But that’s only true when you look at the national aggregates. While Dixiecrat support for Jim Crow was an albatross around the national party’s neck with Northern Black voters who could swing crucial battleground states, civil rights legislation was very unpopular among white Southerners. For precisely this reason, LBJ’s quip was prescient. In 1964, Johnson won a second term in a landslide, carrying every state save Arizona (Goldwater’s home state) and the Deep South — which swung heavily towards Republicans.
Before the election, South Carolina’s Senator Strom Thurmond switched parties because of Democrats’ new stance on civil rights. In 1968, Richard Nixon clinched the White House by splitting Dixiecrats from the New Deal Coalition, while George Wallace’s segregationist third-party campaign carried five Southern states.
The end of Jim Crow voter suppression allowed millions of Black Southerners to make their voices heard at the ballot box. But as white voters left the Democratic Party, the Solid South went from deep blue to deep red. This was about race, plain and simple. And it was still in motion as recently in 2008; the only areas where Barack Obama did worse than John Kerry were in the Deep South and Appalachia.
Of course, Gallup’s numbers on partisans’ ideology only go back to the 1990s. But it’s almost certain that the exodus of segregationists made the Democratic coalition more liberal even as it shrunk.
Everything is education polarization
Matt Levine likes to say that “everything is securities fraud.” Matt Yglesias might say that “everything is education polarization.”
We’ve talked about this phenomenon a couple of times at Slow Boring. The short version is that highly-educated people tend to score higher on a specific personality trait that psychologists call “openness to experience,” which turns out to be correlated with political liberalism. For example, people who went to college are much more likely to favor increased immigration, whereas those who didn’t are more likely to favor curbing it.
And over the last several decades, we’ve seen a rapid increase in the population’s overall level of education. FDR’s base wasn’t people who didn’t go to college; it was people who didn’t finish high school. In 1940, 76% of the population hadn’t graduated high school; 19% had, and less than 5% had completed college. Today, 53% of Americans have graduated high school or have some college experience, while 38% have a bachelor’s degree or higher and only 9% did not graduate high school.
I don’t think it’s a coincidence that society has become more liberal as educational attainment has risen. Compared to the 1990s, when Gallup’s ideology polling begins, public opinion has moved decidedly to the left on issues like gay and interracial marriage.
That’s not to say that these changes were entirely driven by increasing levels of education. They were partly driven by cohort replacement — old people with retrograde views passing away — rather than education. (I’d argue that this effect accounts for most of the recent shift in attitudes toward interracial marriage.) The other part was driven by persuasion; in my lifetime, support for gay marriage has gone from 37% to 71%. A large part of that increase was driven by activists winning hearts and minds.
Over the past decade, though, there has been a conscious factional effort to move the Democratic Party to the left. Trust me; I read the 2012 and 2020 DNC platforms in full so you don’t have to. I think that effort has been important — the Biden-era party has moved to the left of Obama’s Democrats on basically all issues — but the leftward shift is downstream of the rise of liberals within the Democratic Party. As overall education levels have gone up, college-educated voters have drifted into the Democratic Party while non-college voters have left. And survey data shows a noticeable “college degree divide” among Democrats when it comes to support for liberal policies. The upshot is that education polarization moves the party left faster than it moves the country.
Thus, the party’s leftward shift is primarily downstream of rising educational attainment and the end of Jim Crow in the South. The former could not really have been avoided; the latter should not have been put off for nearly as long.
The People’s Republic of Minnesota
We’ve talked before, at length, about how most Americans are moderates and about the relationship between moderation and electability. The implication of voters’ preference for moderate candidates is that parties pay a penalty for becoming more ideological in the form of narrower majorities — but they are rewarded with the ability to pass bolder policies.
Minnesota provides a useful case study. The last time the state went for a Republican at the presidential level was 1972. It’s been moderately but consistently left-of-center for a while now. And the Democratic-Farmer-Labor Party — the national party’s state arm — used to win elections by doing well in the Twin Cities’ core and the traditionally union-heavy Iron Range up north. But in the Trump era, the DFL has shed support in rural areas while gaining ground in the suburbs of Minneapolis-St. Paul. These shifts have mostly canceled out: Barack Obama won Minnesota by 7.7 points in 2012 with the old coalition; Joe Biden won by 7.1 points with the new one.
But for almost a decade, the DFL has lacked a trifecta in St. Paul until last year’s midterms, when they won a one-seat majority in the state Senate and a six-seat majority in the lower house. Despite razor-thin margins, Tim Walz has been able to pass a plethora of progressive policies: universal free lunch for K-12 students, legal recreational weed, codifying abortion rights into state law, stronger union protections, and higher taxes on the wealthy, among other things.
As Eric Levitz details, conservatives leaving the party left the DFL’s coalition both narrower and more liberal; lawmakers also inherited a record budget surplus. But a crucial factor was a shift in the state party’s strategy.
But the party’s leadership managed to cultivate a “you (might) only govern once” ethos, which stiffened swing-district senators’ spines. When Democrats last boasted full control of the state government in 2012, they chose to use their power carefully. As house majority leader Jamie Long told the Washington Post, “There were many things they decided not to do because they figured, ‘Well, we should win our reelections and then we’ll come back and do all those things next time.’” Then they lost power in 2014 and didn’t regain it for nearly a decade.
This time they chose not to take power for granted. “I’ve always said you don’t win elections to bank political capital,” Walz told the Post. “You win elections to burn the capital to improve lives.”
You can see this shift in the national party, too. A common criticism from the left is that Barack Obama wasn’t progressive enough as president. It’s also the case that by any reasonable metric, Joe Biden is the most left-wing president in history. My read of things is that Biden is a party man through and through — he tracked the Democratic Party’s tack to the right under Bill Clinton and then followed the party left post-Obama. It’s true that as president, Biden has signed bigger, bolder bills than Obama did. But it’s also true that Obama came into office with 60 Senate seats and Biden came in with 50. In 2009, Nancy Pelosi was working with a caucus of 257; in 2021, her caucus numbered 222.
One can argue, as Gov. Walz does, that political capital ought to be spent down when in power and that narrower majorities are a price worth paying for passing bolder policies. I think that the political scientists in the 1950s who worried about insufficient polarization would agree.
But if you’re aware of Betteridge’s law of headlines — which states any headline phrased as a question can be answered by the word no — then you’ll be able to infer from the subhead that I’m not so sure. My formative political memory is the 2016 election. I’m very averse to the risk of losing elections to Republicans who will take us backward on policy, so I’m more inclined toward strategically moderating on salient issues to keep the GOP out of power.
Next year’s Senate map looks pretty unfavorable for Democrats. Even if Democrats win every race in a state that went for Biden in 2020, they’d still lose three seats. This is the same Senate map that netted Harry Reid 53 seats in 2012. But since then, it’s become much harder for the party to win in states like Florida and Ohio, which you can see in the 2018 results. I think the fact that Florida and Ohio have become decidedly right-of-center states is related to the fact that they are less educated than the country overall. And that means that winning a Senate majority will be tougher, which makes it harder to confirm judges and cabinet nominees and pass legislation.
Again, one can argue that this is a price worth paying for the ability to pass more liberal policies when you do win a majority. But the point is that the tradeoff exists.
For more on the long, rocky road to securing the 15th Amendment’s promise (in conjunction with the 19th Amendment) of voting rights regardless of race, I cannot recommend Ari Berman’s 2016 book “Give Us the Ballot: The Modern Struggle for Voting Rights in America” enough. LBJ’s quip about losing the South is mentioned in the book, but I sourced the exact quote from Moyers himself via this Guardian article.
First, great writing, Milan. Really impressive.
I share your concern about the movement in the median Democratic position on a number of policy fronts and how that movement can result in more Republican wins. It would be nice if you would point to specific policies that should be either changed or dropped, though.
For me, it is around law-and-order policies that favor criminals more than victims and a toleration for disorder more broadly. These are mostly state or local policies, though, so national Democratic politicians still get my votes. But at some point, that might change if trends continue.
What policies are like that for you?
"But if you’re aware of Betteridge’s law of headlines...."
Are Slow Boring Readers Unaware of Betteridge's Law?