Brad DeLong reckons that the relative merits of carbon taxes and cap-and-trade "roughly offset each other". "To first order cap-and-trade and carbon taxes are the same," he says, but there are second- and third-order differences. Among the second-order differences are these, and you can see how he ends up with the "roughly offset" conclusion:
Cap-and-trade runs the risk that the cap will be set at the wrong place and so the price will go damagingly above its social optimum value.
Carbon taxes run the risk that the tax will be set too low and so the quantity emitted will go damagingly above its social optimum value.
These two considerations do not offset each other. The second risk is high and real; the first risk is low and politically much more unrealistic.
Given the hysteria over energy prices in general and gasoline prices in particular, it’s easy to imagine how a carbon tax would be set too low. And it’s true that no one really knows what the elasticities are in the energy market, which means that an aggressively-low emissions cap could indeed send prices into the stratosphere.
But, realistically, what would happen in such an event? Would Congress sit idly by as fuel-oil costs exceeded monthly mortgage payments? Would they tell their constituents that, sorry, nothing they can do about $15-a-gallon gasoline prices, we set our cap and now we have to stick with it? Of course not. They would tweak the cap-and-trade system in one of any number of ways: they might allow companies to borrow emissions credits from future years, or they might implement a "safety valve" allowing the government to auction off new emission credits at a certain price, or they might simply raise the cap. Alternatively, of course, they could take the unexpected excess revenue from the cap-and-trade auctions and start mailing large checks to everybody in the country, thereby helping to cancel out the ill effects of higher energy prices.
The one thing you can be pretty sure would not happen is that Congress would happily take the cap-and-trade windfall revenue and use it to, say, pay down the national debt. Although even that would have social value which would offset the negative social effects of higher energy prices.
But what of the scenario where emissions permits are auctioned off at a relatively low price, and then suddenly skyrocket in the secondary market, giving no windfall to the government? Well, for one thing, the value of next year’s permits has just gone up, so the windfall will come. And for another thing, given that most people bought their emissions permits at a relatively low price, and that it’s only the marginal permits which are expensive, the effects on actual energy prices would likely not be huge.
In other words, as I’ve said many times in the past, cap-and-trade is flexible. Once you’ve installed the mechanism, it can and will be tweaked over time. Changing tax rates, by contrast, is much harder. Which is why cap-and-trade is superior to a carbon tax.