Winner of the New Statesman SPERI Prize in Political Economy 2016


Wednesday 4 November 2015

Tax cuts vs spending vs helicopters

Some people still seem unable, or maybe unwilling, to understand the basic New Keynesian (NK) model. Should it be surprising in this model that cutting taxes on wages at the Zero Lower Bound (i.e. when nominal interest rates are fixed) are contractionary? Of course not. The basic NK model contains an intertemporal consumption function that implies Ricardian Equivalence holds, so consumers save all of the extra income they get from a tax cut. But cutting taxes increases the incentive to work, thereby increasing labour supply, which through a Phillips curve decreases inflation. With a fixed nominal interest rate that implies higher real rates, which are contractionary. QED.

Now the main thing not to like here is the consumption function and Ricardian Equivalence. Empirical evidence points strongly to a significant income effect, with a marginal propensity to consume around a third rather than zero. There are good theoretical reasons why you might get this result, even with totally rational consumers. But the implication that cutting taxes will lead to some increase in labour supply seems reasonable, and that will put some downward pressure on inflation. This is why pushing ‘structural reforms’ that expand the supply side in a liquidity trap can be counterproductive in the short term. (Things are more complex when you have a fixed exchange rate.)

Now you may quite reasonably believe that in the real world a positive income effect from a tax cut will raise demand by more than any increase in supply, so inflation will rise and real rates will fall. But it remains the case that as a stimulus measure directly raising demand through higher government spending does not generate this supply side offset. That the NK model has this feature seems like a virtue to me. The only point I have to add is that because helicopter money, as traditionally envisaged, is a lump sum transfer (everyone gets an equal amount, so it is independent of wages), you do not get this offsetting supply side effect. So for that reason helicopter money is more effective as a stimulus instrument in a liquidity trap than cutting taxes on wages.


38 comments:

  1. If "Empirical evidence points strongly to a significant income effect, with a marginal propensity to consume around a third rather than zero" then why is the presence of Ricardian Equivalence in NK theory an asset? Surely it's a classic case of trying to make an analysis - which could well lead to policy advice - in accordance with theory and with the premises of the theory rather than in accordance with the real world as measured. In other words, it's yet another example of the lamp post fallacy (the drunk searching for his keys near the lamp post rather than where he lost them 20 yards away "because the light's better there").

    I am all in favour of both public investment (with all the caveats of long lead times and so on) and 'helicopter money' so long as it can be distributed in some reasonably progressive, or at least neutral, manner, but to rule out labour tax cuts because theory says that the transfers will be saved, even while acknowledging in the next sentence that this is not actually borne out by the evidence, just seems weird.

    ReplyDelete
    Replies
    1. Its not an asset, as I think my second paragraph made clear. But the way to improve the NK model in that respect is to augment the consumption function, as many more complex NK models in practice do. Rubbishing the whole model on that account is silly.

      Delete
  2. "This is why pushing ‘structural reforms’ that expand the supply side in a liquidity trap can be counterproductive in the short term." (SWL).

    Before we have a go at this one, please could you explain your NK definition of "a liquidity trap". Meanwhile, we Post Keynesians, will spend the rest of the day trying to connect "structural reforms" to "supply side" to "liquidity trap" to "fixed exchange rate".

    PS. Don't tell anybody Simon but; there is no such thing as Ricardian Equivalence, that is why the government had to introduce PAYE taxation. Come to think of it, there is no such thing as "supply side" economics either. The mantra of the supply siders "Build it and he [that is, a customer] will come"; only occurs in Kevin Costner films.

    Good 'ere init ;-) ;-) .

    ReplyDelete
  3. I have to confess I had never thought it in this way, but then I'll also confess that I'm not an economist and therefore have an excuse not to know better. Still, it's a great thing to point out to my students when talking about fiscal policy effects.

    ReplyDelete
  4. in the real world a positive income effect from a tax cut will raise demand by more than any increase in supply, so inflation will rise and real rates will fall. "
    In the real world, many working people do not pay any income taxes and many do not work so tax cut is only about helping a particular segment of population. In the real world tax cuts are givaways to wealthy and cost a lot and the help is concentrated to those allready well of that barely needs more help except if they are in burdensome debt. It can prevent debt defaults of some and that is all the real help tax cuts are to an economy with large unemployment.
    Hence, your 1/3 stimulative effect of tax cuts is really genwrous while at huge cost to government.
    Helicopter money on the other hand helps unemployed and their families the most and those working that do not pay any income tax while others will barely notice it.
    1/2 of helicopter money is generously giving to those that will spend it imediatelly hence the better stimulative effect.

    Spending will employ more then any other stimulus (or prevent firings) but will also increase imports more that any other stimulus.

    It is also a moral question not only economical question on what kind or what combination of stimulus is best.

    ReplyDelete
    Replies
    1. I do not agree with the penultimate paragraph. Typically government spending is less import intensive than the private spending that would result from a tax cut, and by design it could be made even less import intensive.

      Delete
    2. Jure Jordan, I think you’ve missed another, and perhaps more important flaw in the sentence you quote. It’s that SW-L’s article is concerned with a scenario where there is excess unemployment. I.e. unemployment is above NAIRU. Thus I have doubts about the phrase in the sentence you quote, namely “so inflation will rise”. Inflation won’t rise significantly till unemployment drops to NAIRU, seems to me.

      Delete
  5. In 2000 a large number of people in Denmark (a relatively well-informed country I should think) were asked if their government was running a surplus or a deficit. Less than half of them got it right. How Ricardian equivalence gets given the time of day in the light of facts like these is a mystery.

    ReplyDelete
    Replies
    1. "How Ricardian equivalence gets given the time of day in the light of facts like these "
      is a mystery."

      It's called ideology.


      Henry

      Delete
    2. The answer is that anything that makes economists' models more complicated is welcome because it helps keep economists employed....:-)

      Delete
  6. Market Fiscalist4 November 2015 at 14:43

    I'm also struggling with "But cutting taxes increases the incentive to work, thereby increasing labour supply, which through a Phillips curve decreases inflation. With a fixed nominal interest rate that implies higher real rates, which are contractionary. QED"

    A greater incentive to work increases the supply of labor, which will be expansionary and may cause real wages to fall.

    This means that any change in NGDP for that period will result in a relatively greater increase in RGDP and less inflation than without the tax cut. The fall in inflation means the the real interest rate rises which is contractionary.

    I find it hard to image a realistic scenario where the first effect is not greater than the second effect (even if full RE holds). Perhaps the NK model explains this, but its certainly non-intuitive.

    ReplyDelete
    Replies
    1. Market Fiscalist4 November 2015 at 16:54

      OK, let me try and say it in a different way.

      Assume we start with a rate of interest that is consistent with full employment.

      Other things equal a fall in inflation while the nominal rate stays the same will mean the real rate rises to a level above that needed for full employment.

      But, other things are not equal: The increased supply of labor at the current wage rate will cause upward pressure on the interest rate as firms borrow to take advantage of the increased supply.

      I find it hard to believe that the first effect will be greater than the second for following reason:

      If the rise in the real interest rate caused demand for labor to fall below its initial full employment level then that will prevent the "Phillips curve" effect from taking place. And as that is the the very thing that is causing real interest rates to rise in the first place I do not see how this outcome can be logically consistent.

      Delete
    2. I think the crucial part in Market Fiscalist's comment is his claim that "A greater incentive to work increases the supply of labor which is expansionary". However, the argument in the blog concerns what NK models imply would happen in the context of a period when the economy is at the Zero Lower Bound. During such a period, the output of goods is limited by the low aggregate demand for goods, not by the supply of goods. So an increase in the supply of labour will not in itself increase the real output of goods. It is therefore not expansionary on RGDP..

      Almar.

      Delete
    3. Market Fiscalist4 November 2015 at 19:46

      I suppose if we assume that demand for labor is relatively inelastic then an increase in supply will potentially cause a fall in the rate of wage increases and inflation, which will indeed reduce real interest rates and cause a contraction.

      But one could just as easily assume that wages are sticky and an increased supply at that price causes an increase in employment and output with no increase in inflation , which will therefore not change the real interest rate - so this will be expansionary.

      It is an empirical mater (or dependent upon one assumptions) what the effect will be.

      Delete
    4. An invigorating brain workout this morning as I tried to understand this post. Am I right in thinking that Anonymous here is correct and that the nub of the issue is that at the ZLB the rate of interest cannot be assumed to be 'consistent with full employment' precisely because, at the ZLB, the central bank is all out of means to make it so?

      Delete
    5. Market Fiscalist5 November 2015 at 17:42

      From the post: "Should it be surprising in this model that cutting taxes on wages at the Zero Lower Bound (i.e. when nominal interest rates are fixed)"

      This implies that it is the fact that rates are fixed , rather than that they are fixed at he ZLB, that is key to the the effect of the tax cuts - so I'm not clear why an assumption of being below full employment is now being assumed.

      Delete
    6. The one goes with the other. In the NK model the monetary authority can always achieve 'full employment' by setting the real rate to its natural level - unless it hits the ZLB.

      Delete
  7. I think a *permanent* cut in marginal tax rates would have the opposite effect in an NK model temporarily at the ZLB. Future employment, income, and consumption would rise proportionately, and consumption-smoothing implies current consumption demand would rise too in the same proportion. This would nullify the deflationary effect of the increased current labour supply, and provide a first-order increase in current output and employment.

    I *think* that's right.

    In NK models, a lot depends on our implicit assumptions about what is being held constant.

    ReplyDelete
    Replies
    1. Indeed. When macroeconomists talk about countercyclical policy, there is a presumption that it will be temporary - its just so obvious, to macroeconomists! Of course a permanent cut in tax rates would require a permanent increase in some other tax or a permanent decrease in government spending, so the welfare effect is ambiguous.

      Delete
  8. “…in this model [NK] cutting taxes on wages at the Zero Lower Bound (i.e. when nominal interest rates are fixed) are contractionary”

    Not from steady state. In fact, there is no steady state with interest rates pegged at zero assuming Ricardian equivalence (i.e. a passive fiscal policy) holds.

    Things change if we switch to an active fiscal policy (FTPL). Interest rates can forever be stuck at zero (or any level really) in steady state. A tax cut would then increase the amount of government liabilities, i.e. the amount of net nominal assets for the private sector. People increase their spending to smooth out their consumption over time. Output goes up. And so does inflation.

    From steady state, a tax cut is thus always expansionary in the standard NK model with a pegged interest rate.

    Now how long do we need to be stuck at the ZLB to consider it a steady state result? I think seven years is enough.

    ReplyDelete
    Replies
    1. Nice to see my old friend the fiscal theory of the price level, but I really hope we are not at a zero nominal rate steady state.

      Delete
  9. Simon, can you clarify how exactly the lump sums under helicopter money would be distributed? Apologies if I have missed this in an earlier blog. Thanks.

    ReplyDelete
    Replies
    1. The traditional idea is that every person (adult?) gets an equal amount.

      Delete
    2. This defines the problem rather than the solution. As well as the question you note over children, there is also an issue over citizens and residents. Then there would be the practical issues of distributing the money. 2 million UK adults lack bank accounts and some people have more than one, with that not always being easily identifiable, so just boosting all accounts would not be equitable. Similarly, not everyone falls pays income tax or receives benefits, so those routes would also miss some people out.

      If this is ever to become a practical proposal then these issues need resolution. It would not be politically acceptable for the BoE to pick and choose who gets the lump sum. One of the advantages of the Murphy/Corbyn variant of PQE is that it's clear where the helicopters land: on the roof of the National Investment Bank.

      Delete
    3. I agree those are all issues to be resolved, but they do not seem huge to me. (After all we had a poll tax.) Of course the government would have to cooperate with the central bank.

      I also agree that the Murphy/Corbyn variant of PQE also has merit, as long as the decision on when to fly the helicopters remains with the central bank. But there are practical issues there too. You would need to make sure that there was always a stock of projects ready to go which would not have gone if the helicopter had not arrived.

      No reason why you could not do both.

      Delete
    4. Agreed but on both options the practical issues could discredit a good idea if not worked through.

      I would argue strongly that the core decisions on entitlement (children, resident non-citizens) should be taken by government/Parliament for the sake of democratic accountability. The technical decisions on distribution mechnisms could be devolved but government should remain ultimately accountable in a similar way as for tax collection or benefit payments. Similar, rather than same, to acknowledge that HMRC is a government department while the BoE is not.

      Delete
  10. How do you feel about increasing world-wide low demand by countries running larger budget deficits financed by direct transfers from their central banks? (overt monetary financing) The point is political--to finance needed stimulus in a way that overcomes the fears of politicians and pundits to rising national debts.

    ReplyDelete
    Replies
    1. Have a look at what I have written over the last few months on helicopter money

      Delete
    2. Nick, why would running larger budget deficits need to be financed by "direct transfers" from a central bank? Where do you think the central bank will get the money from if it is not in "swap financial assets" mode. If it is in "creating new deposit (credit)" mode, just like a high street lending bank; its capital base would be nowhere near sufficient to do any useful credit creation, without the Treasury backstopping it. So why not have the Treasury just spend new "money" into existence, and cut out the middle men? Ah! ... that would be a fiscal operation and the responsibility of politicians, can't blame the designated whipping boy, the central bank for fiscal cock-ups.

      A central bank can't increase the net fiscal assets in a fiat currency economy, only its boss, the Treasury can do that. The modern classic example is the "Funding for Lending" scheme. A fiscal stimulus disguised as a central bank monetary operation, to protect the politicians from blame. (An excellent piece of financial engineering some might say.)

      Delete
  11. If tax cuts at the ZLB can be contractionary, might tax rises be expansionary? Would this be a good idea?

    ReplyDelete
    Replies
    1. No. Plus, tax cuts are always expansionary. Don't listen to this nonsense.

      Delete
  12. "Of course a permanent cut in tax rates would require a permanent increase in some other tax or a permanent decrease in government spending, so the welfare effect is ambiguous."

    Not with helicopter money! If the permanent tax cut is financed by the central bank purchasing perpetual bonds, there is no need to resort to compensatory fiscal measures, and therefore no Ricardian Equivalence effect is set in motion. That is in fact the major difference between helicopter money and non-monetary financed fiscal deficits. And this is also the reason why helicopter money is more effective as a stimulus instrument in a liquidity trap than cutting taxes on wages, not just because of the supply side effect that you mention. Helicopter money or overt monetary financing of deficits relieve the government budget constraint permanently.

    ReplyDelete
    Replies
    1. Gilts and money are the same thing in the round. Most of the debt is just "rolled over."

      Delete
    2. "relieve the government budget constraint permanently."
      There is no government budget constraint, except for real resources.

      Delete
  13. You claimed that in the real world a positive income effect from a tax cut will raise demand by more than any increase in supply, so inflation will rise and real rates will fall. But in the real world, there are lots of people who don’t have to pay any income taxes and others don’t work soa tax cut is only going to help some of the population. If more of these real world variations were put into place, could there be variances in real rates and inflation that were otherwise unaffected by these more real world influences?

    ReplyDelete
  14. Dear Prof.,
    I hope you will read this even if months have passed. I am thinking about an economy in which unemployment (especially the youth's one) is very high, such as Italy. In this case I am not sure if a labor tax cut (they are amazingly high in Italy) would only constitute a positive supply shock. It makes sense to me that lots of people would gladly spend more if only they had a job. Also considering that the less wealthy have a higher propensity to consume, bringing people living off benefits (for example) in the job market would boost demand a great deal as well. If that was the case, then the contractionary effect might even not occur.
    Does it make sense to you, in this specific case?

    Best,

    Fabio

    ReplyDelete

Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.