President Obama concluded his press conference last Friday by arguing that liberals ought to think about reducing the federal deficit not simply as a concession they're forced to make because conservatives have a significant degree of political power. They should, instead, view it as a necessary step to create more fiscal and political space for the elaboration of the liberal agenda: 

If you are a progressive, you should be concerned about [the] debt and deficit just as much as if you're a conservative. And the reason is because if the only thing we're talking about over the next year, two years, five years, is debt and deficits, then it's very hard to start talking about how do we make investments in community colleges so that our kids are trained, how do we actually rebuild $2 trillion worth of crumbling infrastructure.

If you care about making investments in our kids and making investments in our infrastructure and making investments in basic research, then you should want our fiscal house in order, so that every time we propose a new initiative somebody doesn't just throw up their hands and say, "Ah, more big spending, more government."

Isabel Sawhill of the Brookings Institution has made a similar case, arguing that a nation whose government has "most of its budget on automatic pilot and a fifth of its expenses unpaid for" has abandoned "fiscal democracy." That is, all of the budgetary decisions that matter were made decades ago when social insurance programs were created. If those big programs are forever "off the table" then there will never be revenues for new programs to meet new challenges. Sawhill estimates that federal taxes will have to double or triple in the next 40 years just to fulfill the inviolable promises made by Medicare and Social Security. Not only would such taxes be intolerable in themselves, but they would obliterate all prospects for the government to do anything else on the liberal wish list.

The problem, of course, with sensible liberal arguments is that so many liberals end up rejecting them. Jonathan Cohn contends that opposition to the liberal agenda has less to do with the federal deficit than with suspicion about government in general, which goes back nearly half-a-century to the great disruptions of the 1960s. Joan Walsh argues that reducing spending to curtail deficits leads liberals in exactly the wrong direction. In her view, Barack Obama is making Bill Clinton's mistake, hoping in vain that fiscal rectitude will neutralize political opposition to new liberal initiatives. In fact, she says, the political space for American social democracy is already there, and liberal leaders would do more to expand it by closing "the gap between Democratic campaigning and Democratic governing," gratifying the liberal base rather than wasting political capital trying to placate liberalism's adversaries, in other words.

Walsh admits, however, that there's one teensy complication: According to public opinion surveys, there are large and durable majorities for all sorts of governmental expansions. "The problem is matching up the political beliefs measured by polling, and the political beliefs measured by voting, where lasting majorities on behalf of those priorities don't ever seem to materialize."

"I don't know exactly why that gap exists," she writes, but the gap is certainly the fundamental reason why liberals are so morose, always expecting and never experiencing the latitude and exhilaration of FDR's First Hundred Days in 1933. Many liberals are convinced that the public opinion surveys are accurate barometers of popular sentiment while the ballots Americans cast in elections are spurious. Much of the desire for campaign finance reform is to move the not-so-liberal election results in the direction of the considerably more liberal opinion survey results. The desire for more fundamental structural reforms - such as abolishing the filibuster, the equal representation of states in the Senate, or the Electoral College - partake of the same desire to make the Constitution a safe harbor for a permanent liberal majority drifting at sea.

The belief that polling results are valid and election results are irrelevant reflects the power of the "Howell Raines Fallacy," described by Mickey Kaus as the assumption, always consoling and usually lazy, "that one's righteous views are shared by the great and good American People." (The fallacy is named for a former editor of the New York Times, a journalist known in some quarters as Howl Reigns.) "HRF liberals are constantly calling in the American people as a cavalry (that never comes)," writes Kaus. The fallacy seduces people other than liberals, however: HRF conservatives had persuaded one another by 2005 that the electorate would insist on incorporating private savings accounts into Social Security if only a Republican president had the courage to propose it.

Beyond the Raines Fallacy problem, public opinion surveys are especially prone to overstate the attractiveness and political feasibility of expanding the welfare state. Intensity does and should matter in politics, and elections reflect intensity far more reliably and subtly than public opinion surveys. Democrats finally realized, for example, that polls showing majorities in favor of strict gun control laws were deceptive. The anti-gun majority contained few people who would vote for a politician on the basis of that issue, while the pro-gun minority contained a significant number of people who would vote, donate, and volunteer to defeat candidates who favored gun control and elect ones who opposed it. The "majority" position was, electorally, a loser. I suspect that a similar phenomenon explains part of the gap Joan Walsh laments: the people who tell Gallup they favor more spending on social welfare programs and higher taxes on the rich don't include many people who'll vote on that basis, while the people who take the conservative position on these questions in surveys are disproportionately likely to vote against government expanders and in favor of government restrictors.

Context also matters in politics, and is also better reflected by elections than surveys. It's easy to tell a pollster you're "for" more spending on noble sounding programs, when those sentiments don't cost anybody anything or require any hard choices about which taxes will be raised or which competing programs will be cut back in order to make budgetary room for the "favored" initiatives. As I've argued elsewhere, liberals like survey results because the polling context, artificially purged of scarcity or zero-sum dilemmas, is exactly the mindset liberals wish to bring to the enterprise of governance. It's when we consider pesky details - "How are we going to pay for all this stuff?" or "If the government does more of this - and this, and this - what, exactly, is it going to do less of?" - that the lasting majorities on behalf of the liberal wish list fail to materialize. There are any number of reasons to feel pessimistic about the future of conservatism, but the need to secure resources and set priorities won't soon disappear, which means it's likely to be a while before happy days are here again for liberalism.