News

Loading...

Wednesday, December 05, 2012

Nick Rowe Cleverly Demonstrates That Cantillon Effects Really Represent Fiscal Policy

What he does is show that for every effect we supposedly generate through monetary policy, we can generate the exact same effect minus monetary policy, using taxation and expenditures alone. So it is the taxation and expenditures, and not the monetary policy, that is generating these effects.

Now, he should do a similar post showing how all re-distributional effects in an Overlapping Generations Model are also due to taxation and expenditures, and not to the debt itself. Because the exact same technique he used for Cantillon effects demonstrates the exact same thing for an OLG model with government debt!

UPDATE: As rob noted in the comments, my original posted was not worded as well as it could have been. Therefore, where I originally wrote "fiscal policy," the post now reads "taxation and expenditures."

16 comments:

  1. Oh dang dawg! I admit I'm biased because my position emerges unscathed (since I am being consistent on the debt and Cantillon effects vis-a-vis your point here), but by Jove I think you've got him.

    ReplyDelete
  2. Would he not agree with you but say that govt debt is fiscal policy that has inter-generational effects ?

    ReplyDelete
  3. rob, your comment convinced me that I ought to re-word my post. I have done so, and acknowledged your contribution. Now see what you think.

    ReplyDelete
  4. He says '"Cantillon effects" are just another name for "the effects of fiscal policy".'. He thinks they are 2 names for the same thing. Whenever you see the effects of fiscal policy you could say "look - there is a Cantillon effect at work")

    While (I think) he agrees that you could simulate the effects of debt via taxation he (unlike you) would see these as 2 different things that just happen to have the same effect. He would not agree that when you see the effects of debt that you could say "look - there is taxation policy at work".

    ReplyDelete
    Replies
    1. OK, then, as I note in this post, he would be inconsistent.

      Delete
    2. Look at it this way, rob: to dismiss Cantillon effects, what Rowe does is slice "monetary policy" into two parts: pure monetary policy itself, and the taxation and expenditure effects that accompany monetary policy. And in doing so, he concludes -- correctly, I think! -- that what are called "Cantillon effects" are actually the consequence of the latter. And he does so by showing that he can duplicate any seemingly "monetary" Cantillon effect by pure taxation and expenditure, with no monetary expansion.

      But the same analysis can be applied, and should be applied, and *has* been applied, to his OLG models of the effects of debt, showing that what he claims is an effect of debt can be exactly duplicated by taxation and expenditure absent any debt. So he ought to admit that, if the argument against monetary policy having Cantillon effects is wrong, so is *his* argument against government debt having intergenerational effects wrong: the actual vehicle of these effects is taxation and expenditures.

      Delete
    3. "so is *his* argument against government debt having intergenerational effects"

      Oops, "so is *his* argument FOR government debt having intergenerational effects

      Delete
    4. Gene: the burden of the debt on future generations comes from the extra taxes that may be required to service that debt. Whether or not extra taxes will be required depends on whether the rate of interest is greater than or less than the growth rate.

      When did I ever say anything different?

      Delete
    5. Nick, I think we must be talking past each other a bit here. See my response to rob below where I try to clarify what I am saying.

      Delete
  5. Maybe I'm misunderstanding what Rowe is saying.

    I'm reading it as:

    - Monetary policy is all about changing the size of the money supply
    - One needs a way of doing this and the way you choose has distributional effects
    - These distributional effects are fiscal policy
    - Fiscal policy effects and Cantillon effects are synonyms.

    He seems to see fiscal policy and monetary policy as closely entwined but always it is the fiscal aspect of any policy that will have distributional (Cantillon) effects and the monetary aspects that will affect the money supply.

    When you say

    "every effect we supposedly generate through monetary policy, we can generate the exact same effect minus monetary policy, using taxation and expenditures alone."

    Do you mean that the fiscal implications of any monetary policy could actually be simulated without changing the money supply ? (for example if the fed increased the money supply by buying bananas and this made banana farmers richer, they could simulate this same effects just by taxing incomes and subsidizing bananas while leaving the money supply unchanged).

    If so then I don't think the parallel with the OLG model works. In that case you really are saying that 2 policies (debt and tax) can both be used interchangeably to achieve the same distributional effects while here Rowe is saying that fiscal and monetary effects can always be separated out from any set of govt actions and it will be always be the fiscal side that has the distributional effects.


    -

    ReplyDelete
    Replies
    1. rob: you understand me.

      Delete
    2. "If so then I don't think the parallel with the OLG model works. In that case you really are saying that 2 policies (debt and tax) can both be used interchangeably to achieve the same distributional effects while here Rowe is saying that fiscal and monetary effects can always be separated out from any set of govt actions and it will be always be the fiscal side that has the distributional effects."

      No, I am saying that debt also has tax and distributive effects that can be separated out from the debt itself, and it will always be those that have the inter-generational effects. So the parallel is exact.

      Delete
    3. Rowe and you agree that the statement

      "Any distributional effect of govt policy can be simulated by tax policy"

      is true. The re-distibutional effects of debt can be simulated by tax

      However the original debate was about burden-effects not distribution-effects.

      The statement

      "Any burden effect of govt policy can be simulated by tax policy"

      Would be rejected by Rowe who thinks that debt creates a burden but not tax, but accepted by you as you think neither govt policy can have such effects.

      Delete
  6. Hi Gene!

    Remember my very first post on burden of the debt? http://worthwhile.typepad.com/worthwhile_canadian_initi/2011/12/debt-is-too-a-burden-on-our-children-unless-you-believe-in-ricardian-equivalence.html

    Here is the fifth comment on that post, by my co-blogger and colleague Frances Woolley:

    "Your story is a perfect description of a Pay As You Go pension scheme. The first generation gets high consumption when young (don't have to pay premiums) and high consumption when old (get benefits). Subsequent generations get lower consumption when young (pay premiums), compensated for with higher consumption when old (get benefits)."

    My response to Frances:

    "Yep. Debt-finance, relative to tax finance, is just like a PAYGO scheme. The only difference is that the interest rate on debt is market-determined, and the assets are marketable, so it's like being able to sell your future CPP benefits when you want to."

    (CPP is Canada Pension Plan).

    Printing money is monetary policy. What the government spends it on is fiscal policy.

    But do the math (a ballpark estimate will do). Government revenue from seigniorage is about 1% of total government revenue from all forms of taxation.

    If Cantillon effects cause a discombobulation in the time-structure of production, regular fiscal policy must cause a discombobulation that is 100 times bigger.

    Cantillon effects are peanuts compared to the total effects of fiscal policy. The Austrians seem to have some sort of peanut allergy.

    ReplyDelete
  7. What he does is show that for every effect we supposedly generate through monetary policy, we can generate the exact same effect minus monetary policy, using taxation and expenditures alone. So it is the taxation and expenditures, and not the monetary policy, that is generating these effects.

    This is a non-sequitur. The mere fact that you can replicate the effects of doing A, by doing B instead, it doesn't mean A is not causing the actual effects when A is taking place!

    Just go the other way. By your logic, one can say that every effect of fiscal policy, because it can be replicated by monetary policy, means that monetary policy and not fiscal policy is actually causing the effects.

    Or in other contexts:

    Because one can replicate the effects of a store owner receiving increased sales revenues from his customers by taxing a group of people and then redirecting the proceeds to the store owner, it means that free market activity is really fiscal policy.

    Because one can replicate the effects of a rise in the stock market by fiscal policy, with monetary policy instead, it means that fiscal policy is really monetary policy.

    Because one can cause the effects (death) of a gun shot, with a bow and arrow instead, it means when someone gets killed with a gun shot, they're really getting killed with a bow and arrow.

    Etc.

    ReplyDelete
    Replies
    1. Petepetepete,

      Did you bother reading Nick Rowe's post before you made this comment? If not, I recommend doing so to see why the above is all irrelevant.

      Delete