Democrats Can Abandon the Center — Because the Center Doesn’t Exist

The center never held. Photo: Scott Eisen/Getty Images

By now, Donald Trump was supposed to be selling herbal Viagra and 1,000-count packages of Trump Ice to reactionary survivalists over Facebook Live; Jeremy Corbyn was supposed to be resigning from the leadership of the (moribund) Labour Party to spend more time with his sewer grates; and Bernie Sanders was supposed to be ruing the fact that, in America, “democratic socialist” is a synonym for “politically irrelevant.”

Alas, the pundits supposed wrong.

The Republican Party nominated a xenophobic insult-comic — who had campaigned in support of political violence, religious discrimination, and mass-murdering Muslim prisoners of war with bullets dripped in pig blood — and was punished with full control of the federal government.

The Labour Party fell into the hands of a radical leftist — who had called Tony Blair a war criminal and Hamas his “friends” — and was punished with its largest increase in vote share since 1945.

A septuagenarian, self-avowed socialist ran a protest campaign on a platform of unabashed class warfare, and ended up the most popular politician in America, a figure with so much grassroots support, he was welcomed into the leadership of a party that he refuses to join.

It is difficult to overstate how thoroughly these developments discredited the baseline assumptions of a certain strand of mainstream punditry. We’re living through a kind of Copernican revolution for the political universe: The old guard still insists that everything revolves around “the center,” but the data keep saying otherwise.

This is not to say that “moderate” politicians can no longer win elections, because voters demand radical solutions to the failures of neoliberalism, or what have you. Trump did lose the popular vote; Sanders, the primary; and Corbyn, the general election. Meanwhile, France appears to have made a socially liberal, business-friendly technocrat its emperor.

Rather, the point is that the relationship between electoral competitiveness and ideological “extremism” is far more complicated than conventional wisdom has suggested.

Your average Sunday-show or nonpartisan pundit paints a tidy political reality that’s governed by a set of fundamental laws:

1) The electorate is divided, almost evenly, between liberals and conservatives, with moderate swing voters marooned between them.

2) The ideological range of American public opinion is mirrored by that of the nation’s elected officials.

3) Thus, the more each party can move its platform in the direction of the other’s — without alienating its base of support — the better its chances of capturing the center, and with it, political power.

To be sure, most political commentary at highbrow publications offers a more nuanced portrait than this. But the spatial model of the electorate favored by the lowest common bloviator remains ubiquitous. The idea that there is an inherent tension between winning elections and moving away from the midpoint between the two parties’ ideological preferences continues to frame analyses of the Democratic Party’s future in the mainstream press.

But this idea has little basis in reality. It is contradicted not only by the developments of the past year, but by decades of political science research and much of recent American history. The political center’s reputed importance is a dangerous myth — one that obscures the genuine obstacles parties face when deviating from status quo policy norms; how those norms came into being; and the electoral benefits that parties might enjoy by willfully violating them.

Given the Democratic Party’s present weakness — and the severity of the threat that the modern GOP poses to America’s welfare state, climate, and democratic institutions — there has rarely been a greater need for a clear-eyed debate about Team Blue’s future. But to have such a discussion, we will need to accept that the center doesn’t hold.

In the middle of nowhere.
The notion that there is an easily identifiable, median political ideology in America derives from the “spatial model” of the electorate, which first gained prominence in the middle of the 20th century. Originally formulated by economists, the model imagines a voting public composed of ideologically diverse — but uniformly rational — citizens seeking to advance their coherent political philosophies. In their 2016 book Democracy for Realists, political scientists Larry Bartels and Christopher Achen offer this thumbnail sketch of the theory:

[T]he political “space” consists of a single ideological dimension on which feasible policies are arrayed from left to right. Each voter is represented by an ideal point along this dimension reflecting the policy she prefers to all others. Each party is represented by a platform reflecting the policy it will enact if elected. Voters are assumed to maximize their ideological satisfaction with the election outcome by voting for the parties closest to them on the ideological dimension, Parties are assumed to maximize their expected payoff from office-holding by choosing the platforms most likely to get them elected.


… [T]his framework is sufficient to derive a striking and substantively important prediction: both parties will adopt identical platforms corresponding to the median of the distribution of voters’ ideal points.

Those who subscribe to the spatial model acknowledge that this prediction doesn’t bear out: Real-life political parties are beholden to interest groups, donors, and hyperpartisan activists who inhibit their capacity to court the median voter. Still, the model dictates that whichever party can best neutralize those constraints and capture the center, wins.

But, as Achen and Bartels argue in their book, the problem with this theory is that none of its premises are true.

First, very few voters have uniformly liberal or conservative ideological views. Which is to say: Public opinion varies across more than one ideological dimension. There are a good number of people in America who support increasing Social Security benefits and banning Muslims from immigrating to the United States. Which shouldn’t be surprising, given that there is no inherent contradiction between those two beliefs. In fact, a recent study of the 2016 electorate by political scientist Lee Drutman found that most social/racial conservatives in the United States have broadly liberal economic views.

If voters’ views vary across more than one dimension, then there is unlikely to be any median voter for parties to target: One could calculate the median policy preference on each individual issue, and then find that virtually no one in the country holds all of them at once.

What’s more, the consensus position on any single issue in the United States is rarely located at the center of a left-right ideological spectrum. A 2014 study from Berkley political scientists David Broockman and Douglas Ahler surveyed voters on 13 policy issues — offering them seven different positions to choose from on each, ranging from extremely liberal to extremely conservative. On only two of those issues — gay rights and the environment — was the centrist position the most common one. On marijuana, the most popular policy was full legalization; on immigration, the most widely favored proposal was “the immediate roundup and deportation of all undocumented immigrants and an outright moratorium on all immigration until the border is proven secure”; and on taxes, the most popular option was to increase the rate on income above $250,000 by more than 5 percent. Meanwhile, establishing a maximum annual income of $1 million (by taxing all income above that at 100 percent) was the third most common choice, boasting four times more support than the national Republican Party’s platform on taxation.

When pundits implore Democrats not to abandon the center, they do not typically mean that the party should embrace legal weed, much higher taxes on the rich, and mass deportation. More often, such pundits call on Team Blue to embrace a combination of moderate fiscal conservatism, a cosmopolitan attitude toward globalization, and moderate social liberalism — in short, to become the party of Michael Bloomberg (minus, perhaps, the enthusiasm for nanny-state public-health regulations). The former New York mayor is routinely referred to as a centrist in the mainstream press, despite the fact that his policy commitments — support for Social Security cuts, Wall Street deregulation, mass immigration, and marriage equality — when taken together, put him at the fringes of American public opinion: In Drutman’s study, the percentage of the electorate that held right-of-center views on economic matters — and left-of center ones on “identity” issues — was 3.8 percent.

There are virtually no libertarians in the United States.

As Broockman told Vox’s Ezra Klein in 2014, “When we say moderate what we really mean is what corporations want … Within both parties there is this tension between what the politicians who get more corporate money and tend to be part of the establishment want — that’s what we tend to call moderate — versus what the Tea Party and more liberal members want.”

Identity trumps policy.
Let’s stipulate that pundits routinely overestimate the appeal of moderate, middle-ground policies. Perhaps parties still pay a predictable price when they diverge from the genuine center of public opinion on any given issue?

Alas, even this proposition withers under scrutiny.

The spatial model presumes that voters’ positions on any given issue are static and coherent. It does not allow for the possibility that a party could shift the views of the public dramatically left or right through messaging. But this is yet another baseless premise. While voters may have have strong, coherent stances on a handful of core issues, on most matters, their views tend to be hazy, and heavily dependent on framing. As Achen and Bartels note in Democracy for Realists, during the Reagan years, surveys consistently found more than 60 percent of Americans agreeing that the federal government spent too little on “assistance to the poor” — while only about 20 percent said that the government spent too little on “welfare.”

In the run-up to the 1991 Gulf War, nearly two-thirds of the public said it was willing to “use military force” against Iraq, but less than 30 percent said they supported “going to war” with that country. And on almost every issue, voters are much more inclined to “not allow” something than to “forbid” it: In 1989, 32 percent of Americans said “no” when asked whether “the law should allow marriages between blacks and whites” — but only 19 said “yes” when asked if the law should “forbid” such marriages.

The American electorate does have stable intuitions about a variety of perennial policy issues. But many of those intuitions are contradictory. A majority of voters want their nation’s distribution of wealth and income to be vastly more equal than it is — but also oppose taxing the inheritance of multi-millionaires. Thus, whether a party’s positions on redistributive policy attract or alienate voters likely depends less on how far “right” or “left” those stances are, than on how effectively candidates weave those policies into a story about what — and whom — they stand for.

Tell white voters a story about “welfare” and undeserving minorities, and you’ll see a rightward shift in their views on redistribution. Spin a tale about how “millionaires and billionaires” could afford to contribute more to “assistance for the poor” and other public goods, and you’re likely to see a different result.

It’s true that most voters appear to have stable policy priorities, in that they choose to back either America’s right- or left-wing party, year in and year out. But as UC Berkley’s Gabriel Lenz demonstrates in Follow the Leader?, his 2012 book, reams of political-science research suggest that this has less to with the stability of voters’ ideological preferences than with the strength of their partisan identities. In other words, most voters develop a preference for one of the major parties — typically, on the basis of the historic allegiances of their family, region, economic class, racial group, or religious community — and then take their ideological cues from their party’s leaders (when they don’t ignore the details of policy altogether).

You can see this in the evolution of Republican voters’ opinions on free trade, before and after Donald Trump’s candidacy. One month before Trump launched his campaign, 51 percent of Republicans told Pew Research that “free trade agreements have been a good thing for the U.S.,” while only 39 percent said the opposite. Fifteen months — and one anti-globalist GOP nominee — later, 61 percent of Republican voters said that free trade had hurt America, while only 32 percent said that it had helped.

And this kind of evolution is the rule, not the exception. Lenz’s work shows that in elections all across the world, voters tend to pick a candidate or party first — and then adopt the ideological positions that rationalize their choice.

Finally, the spatial model baselessly assumes that voters have an accurate sense of what, precisely, the policy commitments of each party are. Exit polls from November’s election found that Trump won 27 percent of white voters who wanted to see “more liberal policies” than Barack Obama had pursued. The mogul also won one-third of all voters who favored providing undocumented immigrants with a pathway to citizenship. These results may reflect another strategy that voters use to avoid any tension between their partisan identities (or candidate preferences) and their past policy views: Project the latter onto the former.

There’s little reason, then, to expect that a party will inevitably become less electorally competitive, the further it shifts away from the consensus views that voters express in opinion polls. In fact, the Republican Party’s domination of American government proves this isn’t the case. More than 80 percent of voters say that the economy is one of their top issues, but, in Drutman’s study, only 26.5 percent of the 2016 electorate espoused right-of-center views on economic policy. And yet, Trump lost to Clinton in the popular vote by only a couple of percentage points — while Republican House candidates received more votes than Democratic ones.

In sum: Almost no one in the United States has uniformly “moderate” policy views; on individual issues, the electorate’s consensus positions are often ideologically “extreme”; partisan voters are typically to drawn to one party over the other for reasons of group identity rather than personal ideology; such voters take their ideological cues, on most policies, from party elites; and many voters don’t know enough about each party’s policy commitments to vote on the basis of their ideological preferences, even if they wanted to.

If the center is dead, then everything is permitted?
In America today, the loudest fight about the costs and benefits of political moderation is being waged within the Democratic Party. Blue America’s ascendant progressive wing is convinced that the path back to power lies to the left of the one that Hillary Clinton pursued last year. But some centrists — and pessimistic liberals — worry that a sharp leftward turn will narrow their party’s appeal and make it easier for the GOP to retain its dominance.

One could read the preceding discussion as a brief for Team “Bernie Bro.” After all, one implication of the center’s nonexistence is that parties have a lot more freedom to move the poles of America’s political debate than is popularly believed. If swing voters aren’t actually ideological moderates, but relatively uninformed citizens who switch allegiances on the basis of identity appeals, economic conditions, and/or candidate charisma — while partisans take their policy positions from party leaders — then there’s little reason to believe that Democrats would inevitably lose votes by endorsing Medicare for All instead of the Affordable Care Act; free public college instead of tuition subsidies; or a federal job guarantee instead of infrastructure spending.

Then again, just because the Democratic Party’s agenda isn’t constrained by the need to appeal to a (largely nonexistent) mushy middle, that doesn’t mean that it isn’t hemmed in by other factors. Voters may not impose sharp ideological boundaries on the politically possible — but economic elites and monied interests still might.

Part of what makes the concept of “the center” dangerous is that it is used to legitimize the constraints that the powerful place on policy makers, by pretending that those limitations are actually being imposed by the will of the people. In truth, what’s politically tenable in Washington has relatively little to do with what’s acceptable to most voters. As Martin Gilens and Benjamin Page found in their famous 2014 study on the relationship between public opinion and policy-making in America, “economic elites and organized groups representing business interests have substantial independent impacts on US government policy, while mass-based interest groups and average citizens have little or no independent influence.”

Thus, for Democrats, one genuine risk of moving left is that doing so might invite the enmity of elite and corporate donors. Running on single-payer health care, for example, would surely gift the GOP a greater share of insurance and pharmaceutical industry dollars. And if voters’ views are malleable and sensitive to framing, then a well-funded ad campaign painting Medicare for All as the first step on the road to Soylent Green could plausibly erode the public’s current enthusiasm for socialized medicine.

And big money isn’t the only obstacle to social democracy in America. There’s also white racial resentment.

Few voters have sturdy ideological commitments, but virtually all have strong social identities. And in a nation founded with a melanin-based caste system, racial identity is bound to be a powerful political force. America’s rapidly changing demographics — and the right’s steadfast efforts to inflame and exploit anxieties about those changes — have made white identity politics increasingly salient in recent elections. This was especially true of last year’s contest, when Donald Trump stoked racial grievances more shamelessly than any major party nominee in recent memory, while his opponent campaigned on pledges to dismantle structural racism and grant citizenship to the undocumented. By November, there were few better predictors of a given voter’s preferred presidential candidate than her level of “black influence animosity.

Photo: Sandy Huffaker/AFP/Getty Images

Democrats have long relied on a flank of economically liberal, culturally conservative white voters, especially in the party’s old labor strongholds in the Midwest. The 2016 election was largely decided by the defection of such voters to Trump’s GOP. One could credit this development to the weakness of Clinton’s economic message or the failures of two decades of trade policy, or James Brien Comey. But given the broader context, the notion that the Democratic Party might pay an electoral price for continuing its leftward drift on immigration, police reform, and other so-called “identity” issues is not wholly baseless.

A separate, but related, potential constraint on Team Blue’s ideological freedom is the party’s growing dependence on the votes of the affluent. As the electorate has grown more polarized around attitudes toward multiculturalism — and less so, along class lines — the Democrats have lost support from white working-class voters, and gained it among suburban professionals. The upper-middle-class tax increases contained in Bernie Sanders’s plans for expanding the welfare state could, theoretically, undermine the party’s appeal in “Panera land” — and, thus, its prospects of taking back the House in 2018.

Finally, there is some evidence that ideologically “extreme” candidates in legislative races do more to mobilize the other party’s base than to turn out their own.

This rundown of the potential electoral headwinds facing progressivism isn’t intended to be comprehensive or dispositive. Rather, it is just to say: Abandoning the concept of “the center” doesn’t require us to deny that the wishes of a party’s most ideological activists — and the ambitions of its most single-minded campaign consultants — can come into genuine conflict. For the Democratic Party, moving sharply left would come with real risks.

But so would moving right — or staying put. The great benefit of forfeiting the commentariat’s favorite, incoherent metaphor, and the knee-jerk presumptions that it dictates, is that doing so allows us to weigh the risks of a leftward lurch against the potential electoral benefits of honoring the Sandernistas’ demands.

The left is right (enough).
Bernie Sanders, Elizabeth Warren, and their sympathizers have long argued that the Democratic Party must adopt a more adversarial posture toward corporate power — and a more robustly redistributive economic platform — in order to address the inhumane inequities of the American economy. They’re also confident that, had the party done so in 2016, Donald Trump would not be president.

Generally speaking, when a political movement describes its agenda as a moral necessity — and, coincidentally, an electoral panacea — it’s best to take the latter judgment with a speck of salt. But this particular political movement (whose ideological goals I just happen to sympathize with) is right. Or, right enough for Democrats to err on its side.

The strength of the populists’ case lies in a simple fact: Democrats have everything to gain by polarizing the electorate around issues of economics. In Drutman’s analysis, 73.5 percent of 2016 voters espoused broadly liberal views on economic policy. If people voted solely on the basis of their intuitions about how government should intervene in the economy (when interrogated about such matters by pollsters), Democrats would dominate all levels of government.

As we’ve established, that isn’t how most people vote. Rather, people generally cast their ballots on the basis of which candidate or party they identify with — which is to say, the one that seems to best represent people like them. The GOP has built a formidable coalition through its mastery of white middle class — and conservative Christian — identity politics.
But all voters possess myriad social identities, including latent ones. A single individual may understand herself in relation to a political party, religious group, line of work, ethnic heritage, gender, and socioeconomic stratum. And since Trump took office last January, the GOP has done (virtually) everything in its power to inform the vast majority of its voters that it does not stand for them, in their capacity as working people.

Congressional Republicans have spent the bulk of the past six months trying to make health care more expensive, and less accessible, for lower middle-class, elderly Americans, especially those who reside in rural areas — in other words, a wide swath of their own base. They have done this largely for the sake of carving out budgetary space for cutting taxes on the very wealthy.

Photo: Nicki Kohl/Telegraph Herald via AP

There are already signs that these betrayals are bringing some GOP voters’ class identities to the fore. A recent Vox/Survey Monkey poll found that one in seven Trump supporters feared that the GOP health-care bill would hurt them. These voters are, in the aggregate, poorer and less economically secure than the president’s other supporters. Critically, their fear of Trumpcare appears to be alienating them from the president: Compared to other Trump backers, they are less confident in the president’s economic management, and more concerned about the Russia scandal and the administration’s alleged ethical violations.

This presents Democrats with an opportunity. And there’s good reason to think that a populist turn — which is to say, a conscious effort to increase the electoral salience of economic class through rhetoric and policy — is the best way to capitalize on it.

Identities gain strength when defined against an other. Trump cultivated the white identity of his supporters by telling a story that pitted honest, hardworking Americans against undocumented welfare cheats, (implicitly nonwhite) cop killers, and Islamist militants.

Sanders and Warren, by contrast, aim to cultivate the class identity of the “99 percent,” by telling a story that pits them against an economic elite composed of billionaires and multinational corporations. This brand of storytelling is off limits to Democrats who seek to portray themselves as “centrist,” by definition. In 2016, Clinton sought to strike such a nonideological pose in her paid messaging, which largely ignored the progressivism of her platform, while framing her campaign around a narrative that put forward-thinking, broad-minded Americans in opposition to Trump, himself (and perhaps, implicitly, his “deplorables”).

It stands to reason that an electoral strategy aimed at polarizing the electorate around issues of economics would be better served by the populist frame. And Barack Obama’s 2012 campaign offers anecdotal support for this thesis. That year, the president pivoted away from the “centrist” politics of fiscal responsibility that dominated the latter half of his first term, and put a spotlight on inequality. His pitch to swing voters in the Midwest was, in essence: Mitt Romney represents all that’s predatory and exploitative in modern, financial capitalism.

This message secured the first African-American president enough votes from economically liberal, culturally conservative white voters to sweep the Rust Belt battlegrounds. And multiple focus groups of Obama-to-Trump voters since November 8 offer (limited, anecdotal) evidence that Clinton may have been able to win a larger share of that cohort, had she embraced more sharply ideological themes, and tied Trump to Paul Ryan’s plutocratic agenda.

Much postelection analysis (including my own) has presumed that, while a turn toward populism might make Democrats more competitive with working-class whites, it would simultaneously jeopardize the party’s ability to compete in the highly educated, affluent enclaves where Clinton actually outperformed Obama’s 2012 showing last year. But there’s reason to believe that this common wisdom was wrong. As more and more college-educated whites have drifted into Democratic tent, the share of the demographic with consistently liberal views — on matters both cultural and economic — has grown. In fact, recent survey data suggests that college-educated white voters actually have more progressive views on a variety of economic issues than (so-called) working class ones. Meanwhile, a Pew survey from April found a majority of voters who make over $100,000 a year saying that the wealthy and corporations pay too little in federal taxes.

Now, one could accept all these points, while rejecting the premise that Democrats would benefit from moving left on policy, instead of just rhetorically. After all, Obama won the Midwest with a more populist message than Clinton, but a less progressive platform.

The Democrats’ official 2018 pitch seems to reflect this insight. “A Better Deal” takes some of the more progressive pieces of Clinton’s 2016 agenda, and rebrands them in unabashed populist language (while putting more flesh on the bones of her anti-trust concerns). This is a reasonable gambit. And it’s certainly true that a party’s overarching message carries more weight than the details of its “issues” webpage.

But it’s also true that a campaign’s signature policies lend heft and credibility to the story it wishes to tell. It seems doubtful that Trump’s nativist rhetoric would have been as resonant, had it not been paired with calls for a border wall financed by Mexico. One advantage of the Sanders-Warren wing’s policy agenda, then, is that its relative boldness and simplicity is better suited to illustrating the populist narrative. Universal free public college, funded by a financial transaction tax on the rich, tells a clear story: Wall Street has too much money, America’s young people have too little opportunity, and something should be done to balance those scales. By contrast, allowing students to refinance their loans at a lower rate may be a worthwhile policy, but it does little narrative work.

Photo: D Dipasupil/2016 D Dipasupil

Similarly, Sanders’s plan to create millions of green-energy jobs through direct government investment offers a much more intuitive mechanism for expanding economic opportunity than does a tax credit for small businesses that invest in job training; Medicare for All is easier to grok than the Affordable Care Act; breaking up the big banks, more straightforward than aggressive enforcement of the Volker Rule.

A separate argument for leaning left on health care and education (as well as on the so-called identity issues of policing and immigration) is that doing so would strengthen the nascent progressive movements organized around those issues — movements that could help increase the woeful registration and turnout rates of young, Latino, and black voters. It’s worth remembering: The modern Republican Party did not gain dominance by spurning grassroots activists who challenged the status quo policy consensus. Abandoning “the center” on social issues, so as to nurture the mobilization of the evangelical right — a demographic that had, historically voted at lower-than-average levels — was one of best strategic decisions the GOP ever made. Trump’s election would have been impossible without it — last fall, white Evangelicals turned in high numbers, and backed him over Clinton by a margin of 80 to 16 percent.

Ultimately though, the argument for moving left on policy is primarily substantive: The potential benefits of enacting progressive policy goals are so great — and the argument for their electoral viability so plausible — it is worth erring on the side of (ideological) incaution.

America has more wealth than any nation in human history — and one of the highest rates of child poverty in the developed world. Tens of thousands of its citizens die each year for want of health insurance. The millennial generation’s $1.3 trillion student debt load is dampening demand throughout the economy. The richest top 0.1 percent of Americans own as much as the bottom 90 percent combined — and a growing body of research suggests that such grotesque inequality is depressing economic growth. Most people in this country have seen their average annual incomes decline over the past three decades, even as the richest among us have seen theirs shoot into the stratosphere. And all that while, humanity’s prospects of averting catastrophic climate change have grown rapidly dimmer.

On most of these issues, effective policy responses aren’t unknown — they’re just considered politically untenable. We know how to reduce inequality and eradicate poverty: you redistribute pre-tax income from the rich to the poor. When America expanded the welfare state, its poverty rate went down; when it scaled back the safety net, the opposite occurred. Nordic social democracies devote more resources to propping up the living standards of their most vulnerable citizens than most other countries, and their poverty rates are among the lowest in the world, as a result.

We know how to reduce student debt: You have the government directly subsidize the cost of higher education. And we know to reduce medical costs while achieving universal coverage — you let the state cap reimbursement rates, and subsidize the medical costs of the sick and the poor until everyone can afford basic medical care, (as they do in virtually every other developed nation on Earth). And while we can’t be certain about exactly what it will take to avert ecological catastrophe, we know that the more rapidly we transition our energy infrastructure toward renewable fuels, the better our odds will be.

But we won’t be able to do any of these things without moving our nation’s consensus on government intervention in the economy dramatically leftward.

There may be a compelling case for postponing that challenge, in the interests of near-term electoral pragmatism. But if there is, that argument makes no reference to the sensibilities of a “center” that does not exist.

Dems Can Abandon the Center Because the Center Doesn’t Exist