Drowning your sorrows will become more expensive if CSPI's proposal gets passed. (Astrid Riecken for The Washington Post)

It's basically a given at this point that any austerity crisis deal will involve new sources of revenue, be it in the form of higher rates for top earners, pared back tax expenditures, or a new tax altogether. Speculation around the third option has mainly centered on a carbon tax, but there are other options too. In particular, some are asking if we should make drinks foot the bill.

The Center for Science in the Public Interest, a watchdog group focusing on nutritional issues, is launching a campaign to address the fiscal cliff by raising the tax on alcohol and introducing one on sugary drinks like sodas. That's politically tricky -- two California towns rejected soda taxes in ballot referenda this year, just as voters in Washington and Maine have in the past.

They claim that a 1¢ per ounce tax on soda would raise about $16 billion a year, or $160 billion over 10 years, and cut consumption by about 12 percent, a major dent in a primary cause of the obesity epidemic. Raising the liquor tax, which hasn't increased since 1991, to where it would be if it had kept up with inflation, and equalizing the taxation of hard liquor and beer and wine, would raise $14 billion a year. Do these numbers hold up? Or are there hidden costs to this approach?

CSPI's revenue figures are solid. In 2008, the CBO found that a 3¢ per 12 ounces tax on soda would raise about $50 billion over 10 years. That's about a quarter the size of the tax increase that CSPI is proposing. The CSPI proposal probably wouldn't raise four times as much, as a high tax is likelier to spur people to reduce soda consumption and in turn revenue from the tax, but that does suggest that the CSPI number is in the ballpark. The Rudd Center for Food Policy and Obesity at Yale puts the number at $13.1 billion in 2012, for a 1¢ tax like CSPI's. That's lower than CSPI's figure, but in the same range.

More recently, the CBO found that equalizing the treatment of alcoholic beverages and increasing the tax from $13.50 per proof-gallon (the standard measure of how much alcohol is in a drink) to $16 would raise about $60 billion over 10 years. That's a much milder increase than CSPI is proposing. Returning to 1991 tax levels, when adjusting for inflation, would mean increasing the tax from $13.50 to more than $22 per proof gallon. That's likely to raise a lot more revenue than the option the CBO considered. The Center on Budget and Policy Priorities puts the revenue from CSPI's proposal at about $101 billion over 10 years.

Put together, the soda and alcohol tax changes CSPI is proposing would raise about $200-300 billion over 10 years. That's a significant step toward the revenue target that President Obama is demanding. But do the taxes work? Not necessarily. As Sarah has noted, past taxes have only had negligible effects on soda consumption. There's some reason to believe that's because the taxes were too small, with one study suggesting a 15-20 percent tax would reduce soda consumption by 8 percent. But passing a tax that big would require some heavy political lifting.

The evidence on alcohol prices is more solid. One study found that doubling the alcohol tax would reduce alcohol-related deaths by 35 percent, with another finding that the 1991 tax increase saved 7,000 lives. Taxes aren't the only way to accomplish that goal. A price floor for certain beverages — say, $25 for a bottle of whiskey — could also reduce consumption significantly, according to one study. But that doesn't raise revenue. An increased tax would, and, according to many economists, would be more efficient.

What about the distributive impact of taxes like these? Ay, there's the rub. The Obama administration has pledged to not increase taxes on those making under $250,000, which effectively rules out tax increases like these. That's especially true because lower-income people disproportionately bear the brunt of these taxes. According to the Institute on Taxation and Economic Policy, state excise taxes on goods like cigarettes, gasoline, alcohol and soda take 1.6 percent of the poorest taxpayers' income, and 0.07 percent of the richest's. In other words, the taxes hit the poor 22 times harder than the rich.

Then again, they help the poor more than they help the rich. Obesity and alcoholism are problems that disproportionately afflict low-income people and communities, and high soda and alcohol taxes could reduce the social costs of those problems. In any case, imposing a soda tax or an increased alcohol tax would require Obama to break his middle-class tax pledge. But it might be a pledge worth breaking.