Saturday, 3 August 2013

The right use of carbon tax revenue? Sorry, there isn’t one

It never ceases to amaze me that some people can be so utterly certain about the right answer to questions that don’t have simple answers. The answer to “What is one plus one?” is simple. The answer to “How should we use carbon tax revenue?” is not. And for good reason.

To reduce carbon pollution we need to price it or regulate the technologies and fuels that cause it. We could just regulate. We could just price. But we know we have to do one or both to seriously reduce the pollution that causes global warming. After that, the certainty dissipates rapidly as each jurisdiction tries to mesh its carbon pollution policies with its other policy goals, the preferences of its voters, and the actions of its major trading partners.

Some environmentalists tell me that all carbon tax revenue in all jurisdictions must be used to reduce carbon pollution via subsidies to building insulation, public transit, energy efficient appliances, low-emission vehicles and renewable energy. This is one possible use. But there are many others. And depending on priorities in a given jurisdiction, the approach can differ.

When British Columbia enacted a carbon tax, politicians were focused on political acceptability. This would be North America’s first – and it still is, which suggests that the concern was legitimate. Business and voter acceptance was deemed more likely if our political leaders could defend the tax as revenue neutral: not increasing net taxes. They therefore legislated that all carbon tax revenue be returned as corporate and personal income tax cuts, and as tax credit payments to lower income people – who otherwise would not be compensated for higher fuel costs because they pay little or no taxes.

Was this the right choice? It’s difficult to say. Some environmentalists point to surveys showing that some people say they would more strongly support the carbon tax if the revenue was used to subsidize the pollution reducing actions I listed above. But, in my view, these surveys are poor at predicting the anti-tax dynamic that would emerge through the media as stories surfaced of “carbon tax poverty” (since the offsetting tax cuts and tax credits would be gone) and “carbon tax boondoggles” (as some of the tax revenue would inevitably be misspent). You only need a modest percentage of people to be upset about a policy for politicians to balk.

And that reminds me of another benefit of the revenue-neutral approach. Real-world experience continually reminds us that the public and politicians are fickle. Initial enthusiasm can diminish as other priorities emerge. But the revenue-neutral carbon tax has a built-in poison pill for future politicians seeking a quick popularity gain by canning it: its elimination would cause a gaping budget deficit, unless they also reversed the associated income tax cuts. In the recent British Columbia election, I bet Premier Christy Clark – desperate for votes as she trailed in the polls – would have cut the tax if she could have, just as she cut the tolls on a new bridge. But since she also needed to appear fiscally responsible by balancing the budget, the best she could do was promise to freeze the tax for the next five years. Staying-power is an important design criterion.

This is the real world of politics in which carbon taxes, if they are to occur and survive, must be designed and defended against some pretty crazy arguments based on misrepresentation and false evidence, not some rational world in which everyone has the same goals, carefully studies the policy, agrees on the facts, and then comes to agreement on how it should be applied. In fact, a majority of British Columbians still refuse to believe that the carbon tax is revenue-neutral, and every week there is another incorrect article in the newspapers telling them they are right.

And this is why the carbon tax arguments of some economists can seem as out-of-touch as those of some environmentalists. These economists harp on economic efficiency as if that were the only thing the politician had to worry about. Sure, using almost all of the carbon tax revenue to reduce corporate taxes would be a big stimulus to investment and economic growth. But try selling that to voters, most of whom would only see this as a subsidy to fat corporations.

This is also why it’s misleading to compare the complexities of cap-and-trade with the “simple elegance” of a carbon tax. Yes, B.C.’s carbon tax was initially quite simple – the politician who implemented it, Premier Gordon Campbell, happened to be a policy wonk. But over the years, he and his successor have yielded to complaints from this industry or that region by introducing various exemptions and rebates – with undoubtedly more to come.

Just imagine the U.S. implementing a carbon tax in order to simultaneously address its climate and budget deficit reduction goals. Were this unlikely event to happen, we can be pretty confident the outcome would entail lots of complexity, both in the tax rate charged to different groups and in the use of the carbon tax revenue. Cap-and-trade – indeed even regulations like the proposed Clean Energy Standard – might not seem so inferior in terms of economic efficiency and administrative complexity after the political horse-trading played itself out.


  1. I think that the concept of a revenue-neutral carbon tax with the revenue returned equally to every citizen is the only fair system. Thus, people responsible for greater carbon dioxide emissions than the average citizen will pay more (in proportion to their respective emissions), and those responsible for fewer emissions than average will benefit financially.

    All other revenue neutral options will allow certain groups/individuals to benefit from the policy even though they are responsible for greater emissions than average.

    1. this is the scheme proposed by James Hansen in Ch. 9 of 'Storms', to me it has the advantages of being easy to understand & approximately bureaucrat-neutral

      but ... there is the issue of timing, the house is burning down, discussions around volunteer vs paid vs unionized fire departments may be moot - the science I am aware of ('2020 emissions levels required to limit warming to below 2°C' Joeri Rojelj et al, Nature Dec. 2012) puts the makes-no-difference-after-date at 2020, others (including Hansen) make it 2015


  2. Anonymous @14:29, a tax and per capita dividend does not change the incentive to reduce carbon emissions because it does not change the increased cost in carbon intensive products. It also has a greater distorting effect on the economy then, for example, a tax and income scaled dividend. That distorting effect may be worth it to counter other distorting effects implicit in the "free market" such as the impact of entities that can only exist as a consequence of regulation (corporations), and the distorting effects of limited liability (again a consequence of regulation that the "there should be no regulation ideologues ignore).

    A third alternative is a tax and dividend to power producing companies at a flat rate per kilowatt hour of electricity sold (and equivalently per passenger/km or tonne/km for transport). Properly constructed such a scheme should have the least economic impact of any carbon tax while still having the same effect in reducing emissions. Obviously there are multiple ways a dividend can be distributed.

    I disagree with Mark Jaccard, however. By design, carbon taxes decrease their revenue take over time if effective. That is because, if effective, the CO2 emissions should reduce to net zero at which point there is no further tax revenue. Therefore the carbon tax should never be used to fund recurrent expenditure. Doing so only sets you up with a future problem of how to replace the tax (or in the US, an ongoing budget crisis as Republicans refuse to replace the tax as a means to eliminate programs they disagree with).

    1. I have not seen empirical work to back up the assertions in Tom's comments and they don't jive with research I have seen. Not to get too far into this, but, for example, it is incorrect that the tax (with returns as dividend or tax breaks) does "not change incentives." The dividend (or tax cut or tax credit) is not based on the tax you paid individually. It is a constant amount per person or it is based on one's income tax rate. If I reduce my emissions more than the next person, I pay less tax and they contribute to my dividend. The incentive to reduce emissions remains as long as the dividend is not a dollar for each dollar of tax for each and every person, but is instead a pre-defined amount for each person.

      Finally, I am not sure what Tom is disagreeing with me about. I am not advocating anything in particular. I am simply pointing out the reasons why some people argue for revenue-neutral carbon tax, while noting that there can be other reasons to argue against it. One possible negative, according to Tom, is that the revenue source diminishes over time, meaning a declining government take to provide a given level of services. This is true. But our forecasts showed that a gradually rising carbon tax over three decades (with gradually falling revenues) - that would significantly reduce emissions in line with goals to prevent major climate change - is easily compensated for by very slow increases in other taxes, perhaps even other environmental taxes as we try to reflect the growing ecological and human costs of the human waste stream. Check out, for example, recent publications on environmental tax reform. Lots of economists work on this problem. Remember that the emissions falls, but the tax per unit of emissions rises, so the total revenue effect is not as great as it might seem if you forget this.

    2. My apologies, I expressed myself poorly. I should have said, "a tax and per capita dividend does not change the incentive to reduce carbon emissions relative to other dividend schemes because it does not change the increased cost in carbon intensive products." That is, the fact that the dividend is flat rate per capita neither increases nor decreases the incentive to reduce carbon emissions.

      With regard to declining revenues, while that can be limited by increasing tax rate in the short term, in the long term when emissions must be reduced to (effectively) zero, no amount of increase in the rate can gain revenue. Thus be funding recurrent expenditure from a carbon tax, we only set up for our descendants a potential political crisis as the low tax/low service party tries to manipulate the situation to obtain their policy ends. Even a gradually declining carbon tax return will be used to try and force a reduction in government services by refusing to vote for the new (or increased) taxes replacing the shortfall, and campaigning heavily on "no new taxes".

      It is not possible, in light of the actions of the US Congress in recent times to assume that low tax parties will act rationally in this regard. Therefore handing them a hostage to fortune is bad policy.

      Finally, you claim that there is no "right use" of the revenue from a carbon tax. On one point I disagree. By all means use it for dividends, for funding temporary renewable energy schemes or research, or for funding one of investment incentives. Just do not tie it to recurrent expenditure (for reasons already given).

    3. Sorry, but I am having to repeat myself in pointing out that you are still mistaken. The tax, which is per unit of carbon in fuels, increases the relative costs of carbon intensive products - directly (gasoline) and indirectly (electricity from coal). Unless, as I said, each person is always guaranteed a dividend that exactly equals the total carbon tax they paid in a year, the system creates a strong incentive to reduce emissions. All experts agree on this because the point is trivial.

      Also, with this next point I am forced to repeat myself. Assuming that the emissions fell to zero over the next 50 years (a very low probability event), this gives government plenty of time to find alternative sources of funding for the current level of expenditures supported by tax revenue. The transition is extremely gradual, and carbon taxes are unlikely to become more than a modest percent of total tax revenue. Also, there will be alternative candidates for revenue to replace carbon taxes over the next 50 years (higher value added taxes, other environmental taxes, user fees for roads and other infrastructure, maybe even - heaven forbid - lower government spending). Again, though, I am being forced to repeat.