One of the things we Democrats use to rock ourselves to sleep at night in these politically perilous times is the hope that demographic trends are working in our favor. And the central source of that hope is the belief that the Latino population of the United States is growing so rapidly that the future shape of the electorate is morphing rapidly in a more progressive direction.Totally aside from the fact that Democrats would be foolish to assume our current performance among Latinos can be counted on in the future, there’s the troubling fact that the total Latino vote is a relatively small segment of the electorate, and will remain so for a while. That’s the important and sobering message provided by Roberto Suro, director of the Pew Hispanic Center, in today’s Washington Post.Suro nicely summarizes his argument in one sentence: “Because of a combination of lack of citizenship, a big youth population, and voter apathy, only one-fifth of Hispanics went to the polls in 2004. In other words, it took five Latino residents to produce one voter.” Of those three factors depressing the Latino vote, only the third is one we can theoretically do something about in the near term. So why all the excitement about percentage increases in the Latino vote?Here, too, Suro offers an important distinction in commenting on the “record turnout among Latinos” recently generated by Los Angeles mayoral candidate Antonio Villaraigosa: “[G]iven the low baseline, it wasn’t hard. When it comes to counting people in almost any category, Latinos break their own records every day.” But as my friend Mark Gersh, the number-crunching wizard of the National Committee for an Effective Congress, always points out, percentages don’t win elections; votes do. And small percentage increases from large groups generate more votes than large percentage increases from small groups. That’s why the little-recognized but central story of the 2004 presidential election was that a smaller percentage increase in ballots from non-Latino white voters more than exceeded the votes produced by near-record turnout among minority voters as a whole. This does not–let me repeat this–does not mean that Democrats should stop worrying about, or working among, minority voters. It specifically does not mean that Democrats should stop obsessing about now to reach Latino voters. Even if the Latino vote is growing less rapidly, in absolute terms, than some Democrats seem to assume, maintaining the current Democratic advantage is well worth every effort, and moreover, the Latino voting boom will definitely arrive in the relatively near future. What Democrats cannot do, however, is to comfort ourselves with the illusion that Latino voter growth will offset our ever-increasing weakness among white middle-class voters generally, or white married voters with kids specifically. (In fact, the upwardly mobile Latinos most likely to vote largely share the values and aspirations of middle-class non-Latino white voters). We need a strategy, a message, and an agenda that will make inroads into Republican majorities in those groups while continuing to attract and energize minority voters as well. We can’t simply wait for demography to save us.