05 Sep 2019

Murphy Triple Play

Shameless Self-Promotion 11 Comments

==> I get pedantic with John Cochrane, but I think the episode is instructive.

==> I defend David R. Henderson against his critics.

==> I’ll be debating again at the Soho Forum in April 2020.

11 Responses to “Murphy Triple Play”

  1. Transformer says:

    On the linked to IER article on David Henderson I’m not sure I follow the economic logic of of ‘Objection #2: “So Let’s Subsidize Trees and Tax Carbon!”’.

    I agree that other things equal if a tax had to be levied to pay for tree planting it would be better to pay for it via a poll tax on everyone than a specific tax on carbon producers on the grounds of ‘the wider the base the less deadweight loss from the tax’. But when carbon production itself produces a deadweight loss (a negative externality) this does not apply.

    Assume a ton of carbon production creates a $10 external cost in the form of environment daage, and $10 of tree planting can reverse that damage.

    Take 2 scenarios

    1. For every ton of carbon produced a poll tax is used to pay for $10 of tree planting
    2. For every ton of carbon produced the producer is taxed $10 to fund tree planting

    I am pretty sure that with option 1 there would be more carbon produced and trees planted that would socially optimal so 2 would produce the better outcome. The reason for this is that under option 1 the price of carbon would still not reflect its true cost.

    In a transaction-cost free world a likely Coarsian outcome would be that carbon producers would voluntarily agree to pay for tree planting in order to compensate those who are affected by the environmental damage caused by the carbon producers and this would also lead to higher prices for carbon-based goods and hence less production.

    • Transformer says:

      OK, I now see that once when takes transaction costs into account it is no longer the case that option 2 is necessarily better. If the transactions costs for 2 are greater than the transaction costs for 1 by an amount that exceeds the utility benefits of lower carbon production and tree planting then option 1 is optimal.

  2. Tel says:

    Slightly off topic, but potentially someone to interview for the Bob Murphy Show.


    She collects and publishes leaked information about US arms shipments to various militant groups … mostly the Sunni Jihadi. Interesting economic and moral question about buying and selling weapons. I mean, trade is good … right? We like trade round here.

    Q: What do you sell to the man who has everything? A: A gun, so he can keep what he’s got.

    Q: What do you sell to the man who has nothing? A: A gun on hire purchase, so he can go out and make his fortune.

  3. Tod says:

    “In other words, if a massive tree-planting campaign or other geo-engineering program worked out the way Henderson hopes, then in the new equilibrium when everybody reacted to the new scenario, the computed “social cost of carbon” would be close to $0/ton. And so at that point, if we had a bunch of statutory carbon taxes on the books penalizing emissions at the rate of (say) $30/ton or higher, that would be totally unjustified and would be causing massive amounts of economic damage for no environmental benefit at all.”

    I would have thought the real concern here would be that under the belief that “CO2 is the temperature control knob” that the scenario where planting trees creates net 0 carbon emmission then any imposed carbon tax (ignoring the political implementation costs) would then put us in a scenario where we are cranking the world temperature lower than ‘optimum’ and that would also have negative impacts…

    • Harold says:

      “that the scenario where planting trees creates net 0 carbon emmission”

      That scenario is not going to happen, nor anywhere close, so no need to worry about falling temperatures.

  4. Harold says:

    On Henderson
    “let’s first dispose of…the idea that taxing carbon is the same thing as “pricing carbon.” Carbon is already priced. Natural gas, oil, and coal all have prices and their prices are somewhat related to the amount of carbon they contain. ”

    I do not agree – the energy of the fuel is priced when people buy fuel. Nobody would care if the energy was from silicon or phosphorous if it worked the same. Taxing alcohol does price alcohol. A really expensive single malt has the same price added as the cheapest vodka, based only on the alcohol content. The price of the drink is not the price of alcohol. I expect to pay a lot more for a Chateau Margaux than a Diamond White cider, even if they contained the same alcohol. That is the price of the drink, not the price if the alcohol. It would be absurd to say what I pay for the expensive wine was an alcohol price. It does not sound ludicrous to me to say an alcohol tax is a price on alcohol. This is a semantic argument really, and I don’t see too much point in making it an issue.

    The big problem with the other methods of reducing harm from global warming is that they are all speculative. There has been no study on how much it would cost to reduce CO2 by planting trees. The one mentioned did not examine costs but plucked one figure as an example. Intellectual Ventures schemes are wildly speculative from both a technological and legal perspective.

    “there might be more sensible policies than to tackle carbon dioxide emissions directly.” But there might not be, and we have no reason at the moment ti think that there actually are. All we have is speculation.

    The thing about planting trees is that it is removing exactly the same CO2 that the polluters are emitting. *If* it could be proved that planting trees effectively removed the CO2 permanently, then the best way to view it is that this reduces the emissions. If I put out 100 tonnes of CO2 and plant trees that soak up 100 tonnes of CO2 then I have emitted nothing and should pay not carbon tax. It is analogous to a factory installing pollution control equipment to prevent pollution leaving their site. As I suggested before, we could think of this as a reed bed to soak up pollution, to keep things as similar as possible. We tax the pollution, but if the factory does not actually emit any pollution it does not pay any tax.

    Pigovian taxes or Coasian exchanges do not necessarily work in this way. As Transformer says above, in a Coasian world the polluter causes some damage, then compensates the victims for the damage caused. This may be cheaper that reducing the pollution. An example may be that it pays people to move away from the pollution, but the river remains polluted – there is just nobody there to notice any more. Or my chimney soot makes your washing dirty, so I buy you a tumble dryer as it is much cheaper than stopping the soot. However, in the CO2 case, if the polluter plants trees they have removed the CO2 and they have not caused any damage to compensate anyone for.

    I think this way of pricing is easier to administer with a cap-and-trade system. Then anyone can plant trees and sell the credits to the power generators. Whether through a cap-and-trade or a tax/rebate system, if reducing CO2 levels by soaking up carbon is cheaper than reducing consumption, then that is what will happen. It doesn’t matter how it is arrived at, the atmospheric levels are all that matters. Is this distinction illusory?

    • Craw says:

      How sad, that even Harold can spank Bob.
      What is not priced is the externality. Taxes might not represent the actual price, but in a situation where no price can be set due to transaction cost, a tax can be a useful approximation of a price. We can either have a sensible discussion of the size, mechanics, and targets of that tax or we can play ideological word-games about “price”.

  5. Harold says:

    I was trying to sort this out in my mind and came up with this analogy. It may be obvious to others, but it helped me to work it through. At least I think it helped – let me know.

    A company (or industry) makes a really useful widget, which people like a lot.The market sorts out the price and the level of production is, say X units.  Given a perfect market this is the efficient level of production.

    It is now discovered that the stuff from the chimneys, call it crabon, causes decay in roofs.  The higher the level in the atmosphere, the more roof damage.  It is relatively easy to assess the cost of this damage.  It works out at $20 per tonne of crabon emitted.  What would happen in a world where  friction-less exchanges could occur, as per Coase, and a world with Pigovian taxes?

    Coase was not so much interested in who paid for what, but that the most efficient solution was implemented.  He showed that in a world without friction the efficient solution would be implemented, whichever party were held liable.

    Scenario 1. Lets assume for now that nobody can find a way to either reduce crabon emissions per widget or prevent the damage.

    In a Coasian, friction-less world, free exchanges would follow.  The industry would pay each person for their roof damage.  The price of the widgets would go up and fewer would be produced, say X1. This new level of production is now the efficient level, with any change causing people to be worse off.
    In a Pigovian world, without friction-less exchange, the Govt. steps in and taxes the industry exactly $20 per tonne of crabon emitted.  The price becomes the same as in the Coasian world and the level of production is efficient.  The only difference is that the compensation does not go directly to those who suffer. Roof owners foot their own bill.  What happens if there are alternatives?

    Scenario 2.  Someone invents a crabon scrubber to fit to smokestacks that removes crabon at a cost of $10/tonne.
    Coasian world: factories fit scrubbers instead of compensating for roof damage.  There is no roof damage.  New level of production is X2 to reflect the price of scrubbers, again the efficient level of production.
    Pigovian world:  Tax remains at $20/tonne.  Factories fit scrubbers as it is cheaper than the tax. Outcome is the same as in Coasian world.  No roof damage and production at X2.  The social cost of crabon has not changed, we just stopped emitting it.

    Scenario 3.  Someone invents a surface treatment for roofs that prevents damage.  It costs $5 per tonne of crabon.
    Coasian world.  Industry pays to treat all roofs as it is cheaper than fitting scrubbers.  No damage is done and level of production is efficient X3.
    Pigovian world.  The treatment has reduced the social cost of crabon to $5 per tonne, because this sum is enough to totally prevent damage.  To restore efficiency, the Govt needs to tax at $5/tonne.  As in the original Pigovian example, the roof owners bear the direct cost.  They each treat their roofs as this costs less than the cost of the damage. Industry pays $5 per tonne and production is again at the efficient X3.    Atmospheric levels of crabon can now rise and no harm is done. 
    Scenario 4.  Someone invents a crabon scrubber that removes the crabon from the air.  It cost $2.50 per tonne to do this* and can be fitted anywhere.
    Coasian world.  Industry installs air scrubbers as it is cheaper than the roof treatment..  Atmospheric levels of crabon are low.  No damage is caused, so no compensation is required.  Production level at efficient X4
    Pigovian world.  The true social cost of carbon in the air is $5/tonne (with treatment), since any crabon in the air will cause this amount of damage.  However, the social benefit of removing it is also $5/tonne.  If we only had a tax on crabon we would have an inefficient production level X3 and no incentive for anyone to install scrubbers (unlike the roof treatment, where each persons treatment benefits themselves directly).  If we add a $5 subsidy for the positive externality of installing air scrubbers, then anybody can do it and claim the subsidy.  We remove the crabon from the air, but we still pay a tax of $5 for producing it. We don’t restore production to efficient level X4.  This is because fundamentally we have not altered the social cost of a tonne of crabon.

    So how do we get to efficient production levels in scenario 4?  We offer the incentive as a tax credit to be used against the emission tax.  Only people actually emitting can benefit, so it is in their interests to install scrubbers (or buy the credits form someone else) to remove their emissions of crabon.  Each factory installs  said scrubbers (or buys credits) and we are back to efficient production level X4 and we don’t get any damage and don’t need to pay for treatments.
    Of course, if the scrubbers turn out not to work as well as thought, the correct tax of $5 per tonne still applies, and we have the efficient level of production however much crabon is removed.

    *As an aside, it might seem very odd to some that a scrubber that could remove crabon at a very dilute level in the air could possibly cost less than one removing it from flue gas, where it is at a very high concentration.  It strikes me as very odd from a thermodynamic perspective, which is why I am generally suspicious of chemical air scrubbing methods.  Whatever works for air will work much, much better up a chimney.

    • Transformer says:

      In the context of Bob’s commentary on David’s post its as if someone invents a crabon scrubber that removes the crabon from the air $2 for .50 per tonne. An economist (called Mruphy) then suggests that this means that as crabon scrubbing is so cheap it is best to pay for it by a general tax rather than a tax on widget producers as the Tax Interaction Effect shows a tax on crabon content hurts the economy more than raising a dollar from taxing labor or consumption.

      However as a tax on crabon content also reduces widget production closer to its socially optimal level its actually an empirical matter which kind of tax is better.

      • Transformer says:

        ‘from the air $2 for .50 per tonne’ = ‘from the air for $2.50 per tonne’

Leave a Reply