Aggregate supply really does matter

A couple years ago I did some posts strongly criticizing the Krugman/Eggertsson argument that the AD curve might slope upward when the economy is stuck in a liquidity trap.  If this “paradox of toil” were correct, then moves to increase AS (say by payroll tax cuts or an increased desire to work) would reduce output.  Higher minimum wages might increase output.  Here’s Tyler Cowen on February 3, 2012:

Another big loser is those liquidity trap theories which tell us that positive real shocks are bad for the economy because the AD curve has a perverse slope, etc., and that negative shocks might help spur recovery.  That theory is looking very weak, again.  I consider it the weakest economic theory that has any currency in the serious economics blogosphere.

I’m not sure it has much currency today, as three respected bloggers attacked this view within a 24 hour period.  Here’s Karl Smith:

Sometimes I think I am detecting a sense that folks are concerned that Labor Force participation will affect the unemployment rate. Eh. Its complicated but there is no lump of growth nor a lump of job creation even in a recession. To put it terms now common place, the number of people who want to work affects the natural rate of interest.

If lots of people are trying to work they will drive up demand for capital and durable goods in the attempt and this will drive up the natural rate. In the face of a constant Fed Funds rate this will mean a faster growing economy.

And here’s Ryan Avent, responding to Jared Bernstein’s call for a higher minimum wage:

That last, substantial rise in the minimum wage looks strangely familiar. In fact, it mirrors quite closely the sharp rise in unemployment that occurred over a similar time frame.

Now, I wouldn’t begin to suggest that the rise in the minimum wage caused all of that unemployment. And, generally speaking, I’m prepared to accept that in most cases, small increases in the minimum wage are less harmful than straightforward micro would have us believe (although it also seems to me that as a means to fight poverty, minimum wage rises are a far worse idea than other alternatives, like wage subsidies or, hey, tight labour markets).

Still, this is terrible timing for a proposal like this. The unemployment rate for workers without a high-school diploma is currently 13.1%. For workers between 16 and 24 it’s 16.0%, and for those between 16 and 19 it’s 23.2%. These are not high marginal productivity workers. I’m trying desperately to think of a dynamic in which raising the cost of employing these people increases their employment, but I just don’t see it.

And here’s Matt Yglesias:

It’s been suggested that recent events have refuted the Eggertsson analysis, but I think that if you look more closely you’ll see that all this work on Retro Keynesianism was predicated on the economy being in an actual state of deflation that the central bank is unable or unwilling to reverse. What actually happened in the United States, however, is that we had a brief deflationary episode followed by an extended period of sub-trend NGDP growth. That’s produced some of the ills you might associate with inflation, but it means that the conditions under which Retro Keynesian ideas were allegedly supposed to apply don’t currently exist and haven’t existed for some time. Under the circumstances, the theoretical questions are perhaps interesting to model-builders and academics but we (thankfully) aren’t living in the kind of universe that would put them to the test.

Yup.  If there truly are an infinite number of universes then it’s possible there’s one where “Retro Keynesian” is true, but we aren’t in that one.  I’d add that the standard Keynesian model of the fiscal multiplier is also based on the notion “that the central bank is unable or unwilling to reverse” demand shortfalls caused by a lack of fiscal stimulus, although I’d guess Smith, Avent, and Yglesias wouldn’t go quite that far.

Peter Laan sent me the following, from a VoxEU post by Jonathan Portes:

Finally, and returning to what I originally learned at the Treasury, there still remains the view that if we think demand is too low, then the right response is always through monetary rather than fiscal policy. Again, there is a vigorous debate among blogging economists on this topic (see Economist 2012 for an introduction to the debate). And here my perspective has indeed changed; I no longer subscribe to the Treasury View of the last two decades, described above, that fiscal policy never has any role to play in demand management, even though I don’t think it should be the tool of first resort. (See the excellent discussion in Simon Wren-Lewis 2012a, especially the penultimate paragraph).

But just as this approach was motivated by pragmatism more than theory – monetary policy was better suited to this task – my change of mind is similarly motivated. If monetary policy alone was indeed enough in practice, we wouldn’t be where we are now, with unemployment in the UK a million higher than the official estimate of the natural rate, and no prospect of it coming down in the immediate future. As I have argued previously (Portes 2012), any demand management policy that delivers that outcome is not one that policymakers should regard as remotely adequate.

So my views have indeed changed; not, I would argue, ideologically, but in recognition of the fact that life, and macroeconomics, is considerably more complicated than we thought. Again, this view is shared by Blanchard, who argues: “We’ve entered a brave new world in the wake of the crisis; a very different world in terms of policy making and we just have to accept it. … Macroeconomic policy [specifically fiscal and monetary policy] has many targets and many instruments.”

This pragmatic and questioning – but evidence-based – approach to macroeconomic policy is one I share. If he were here, I imagine Keynes would too.

When I read things like that I really want to pull my hair out.  He says his views on monetary policy changed, but gives no logical reason why.  Did the BOE fail to hit its nominal targets?  Didn’t Gordon Brown pursue one of the most aggressive fiscal stimulus programs in the world?  And hasn’t Britain had one of the most disappointing recoveries of any non-eurozone economy?  And the takeaway from this is that fiscal stimulus works and monetary stimulus doesn’t work?  This kind of argument drives me nuts.

Portes also claims that the argument against Keynesian fiscal stimulus is based on Ricardian equivalence, and doesn’t even mention the more important argument that in standard new Keynesian models the fiscal multiplier is close to zero when the central bank targets inflation.  Given the inflation rate in Britain over the last few years, I don’t think even the most fanatical Keynesian would claim that the BOE has been unable to keep inflation as high as they’d like.  But that’s the assumption you need (in the new Keynesian model) to make fiscal stimulus work.  Here’s David Romer:

As Robert Solow stresses in his remarks in this session, we should not be trying to find “the” multiplier: the effects of fiscal policy are highly regime dependent.  One critical issue is the monetary regime. Consider estimating the effects of fiscal policy over the period from, say, 1985 to 2005. Central banks were actively trying to offset other forces affecting the economy, and they had the tools to do so. Thus if they were successful, one would expect the estimated effects of fiscal policy to be close to zero.

David Romer doesn’t think this applies to the current situation in the US.  But does anyone seriously think the BOE has been unable to keep inflation as high as they’d like?  Imagine the laughter in the UK if a pundit made that claim on TV.

Marcus Nunes has a nice new post on fiscal stimulus, NGDP, and RGDP.  The graphs are quite illuminating.  Compare Britain to the other countries (in terms of government spending and the RGDP/NGDP split) and tell me why I shouldn’t believe Britain has serious supply-side problems.

PS.  Nunes’s graphs show the public sector in Sweden is now much smaller than in France.  Quite a turnaround from a few decades back.  Come to think about it a few decades back France was doing better than Sweden, and now the reverse is true.  Now I’m sure the G/GDP ratio will shrink as soon as France “recovers.”  But given that the unemployment rate in France has been stuck at 10% for 30 years, the recovery may take a little while.


Tags:

 
 
 

26 Responses to “Aggregate supply really does matter”

  1. Gravatar of Morgan Warstler Morgan Warstler
    9. February 2012 at 06:40

    That’s the bit the above crowd will all need to respond to…

    “Nunes’s graphs show the public sector in Sweden is now much smaller than in France. Quite a turnaround from a few decades back. Come to think about it a few decades back France was doing better than Sweden, and now the reverse is true.”

    Eventually, the worms turns.

    Monetary isn’t enough, you actually need Fiscal to get out of the way.

    We cannot just “accept” any size welfare state or structure AND expect Monetary alone to keep things going.

    The brutal effect of a level targeted NGDP is that IT PROVES the difference between RGDP and increasing public employee pay or entitlements.

  2. Gravatar of Britmouse Britmouse
    9. February 2012 at 07:08

    Portes: If monetary policy alone was indeed enough in practice, we wouldn’t be where we are now

    That is shameful. Just shameful. How can a macroeconomist even say something like that? He cannot possibly have looked at the data. The UK 2008-09 recession was almost over by the time we hit the ZLB in March ’09. But, Mervyn King said it he was trying really hard, so we must believe him. Shameful, the lot of them.

  3. Gravatar of marcus nunes marcus nunes
    9. February 2012 at 07:11

    If I understand Morgan´s argument correctly, Poland which has a much higher G/Y than Australia, for example, could not, but has nevertheless managed to “get things going” with “Monetary alone”.

  4. Gravatar of Peter Peter
    9. February 2012 at 07:22

    I was a bit surprised that France had passed us (Sweden) in government spending. I only remembered that we had higher tax revenue/GDP than France. I suppose our balanced budget compared to their deficit makes the difference. It seems like spending/GDP would be more interesting to look at than revenue/GDP.

    http://epp.eurostat.ec.europa.eu/statistics_explained/index.php/Tax_revenue_statistics

  5. Gravatar of dwb dwb
    9. February 2012 at 08:15

    reminds me of a debate way back where someone insisted that France’s then-mandatory reduction in work hours (from 40 to 37 or 35, i think) way back would reduce French unemployment (i guess under the theory the jobs still needed to be done so people would get hired to fill the hours – I can’t recall how the French govt justified this). ah, no no no no no no, and no. Hopefully the European experiment in rigid labor rules is coming to an end, for their own sake.

  6. Gravatar of Morgan Warstler Morgan Warstler
    9. February 2012 at 09:24

    nunes,

    “keep things going” = first world Reagan style growth rates

    Poland fails on the first world bit. Poland benefits from catch-up modernization.

    It sounds like you are missing the real point… “the brutal effect of a level targeted NGDP is that IT PROVES the difference between RGDP and increasing public employee pay or entitlements.”

    If you accept that as fact (which Scott accepts BTW), the corner is turned.

    Proper monetary doesn’t just trump fiscal in its effect, actually drives down the size of the state.

    This is why I think Austrians have to pay more attention to Scott’s position. He’s not a gold bug, but he’d deliver a lot less inflation historically than we had.

    Less inflation = less government. Generally speaking.

    You disagree?

  7. Gravatar of D R D R
    9. February 2012 at 09:56

    “Didn’t Gordon Brown pursue one of the most aggressive fiscal stimulus programs in the world? And hasn’t Britain had one of the most disappointing recoveries of any non-eurozone economy?”

    Hm. Well, RGDP did grow 3 percent from Q3 2009 to Q3 2010. Maybe not great, but far better than the 0.5 percent from Q3 2010 to Q3 2011. I wonder what might have changed?

    I’m not sure of the exact timing of Brown’s stimulus program, but I’m pretty sure it was supposed to be all over by mid-2010. By which point this Cameron character had taken over….

    *shrug*

  8. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    9. February 2012 at 10:39

    ‘When I read things like that I really want to pull my hair out. He says his views on monetary policy changed, but gives no logical reason why.’

    It’s Streisandnomics; Feelings, nothing more than feelings.

  9. Gravatar of Patrick R. Sullivan Patrick R. Sullivan
    9. February 2012 at 10:45

    ‘I’m not sure of the exact timing of Brown’s stimulus program, but I’m pretty sure it was supposed to be all over by mid-2010. By which point this Cameron character had taken over….’

    Back when I was learning Keynesianism it was called ‘pump-priming’. So, even if it was over by mid-2010 it should have effectuated, if not a boom, at least normal economic activity.

  10. Gravatar of Major_Freedom Major_Freedom
    9. February 2012 at 11:30

    Karl Smith gets it. Almost. When the price of labor falls to the market rate (which the state prevents, thus prolonging the agony), the demand for capital and durable goods goes up, which will increase net investment, and increases in net investment increases profitability and economic recovery. Net investment after all is the difference between revenues and depreciation costs, and revenues minus costs is profit.

    But Smith doesn’t go far enough. If there prevails generally low rates of profit (generated by time preference, not monetary deflation/inflation) in the economy, this fact alone stimulates more capital intensiveness to the extent that capital intensiveness saves more in future costs on capital invested than the prevailing general rates of profit of capital invested. The lower the prevailing rates of profit, the more attractive cost saving capital intensive investments become. The higher the prevailing rates of profit, the less attractive cost saving capital intensive investments become. Thus, if prevailing rates of natural profits were ever to become low, that fact alone stimulates more capital intensive investments and discourages less capital intensive investments. This has the long term effect of raising not only prevailing rates of profit, but it also raises everyone’s standard of living in terms of real wealth.

    Ryan Avent gets it.

    Matt Yglesias, and therefore Sumner, conveniently ignores the period of inflation PRIOR to the “sudden”, “inexplicable”, “animal spirits” drop in aggregate spending that was not due to the Fed soaking up liquidity, but “just happened.” Then, because the decrease in aggregate spending “inexplicably” occurred, we’re supposed to believe that an inexplicable increase of aggregate spending by the centralized counterfeiters, and/or the Treasury, is the answer, as if reversing the effect of a problem can solve the problem itself.

    Jonathan Portes is totally out to lunch.

    David Romer’s conclusion is right but for the wrong reasons.

    Marcus Nunes makes the oh so typical mistakes of conflating “RGDP” with actual real productivity, rather than what it is, which is just nominal spending “corrected” by an arbitrary baseline spending, and of the myth that NGDP is a causal force for, rather than an effectual result of, economic health. Those charts are just comparing one nominal spending quantity with another nominal spending quantity. Just look at Australia mid 2008 to mid 2009. No growth in nominal spending, and yet nominal spending corrected by baseline spending kept growing, and since mind 2009, nominal spending growth is once again increasing, and yet nominal spending corrected for by baseline spending has maintained trend.

    Even if you want to incorrectly view NGDP as causal, it is clear that in all 6 charts, there has been a gradual decline in the effectiveness of NGDP growth to generate RGDP growth. It’s taking more and more NGDP growth to generate RGDP growth. The slopes of NGDP are all steeper than the slopes of RGDP. This is consistent with the theory of fiat monetary breakdown.

  11. Gravatar of D R D R
    9. February 2012 at 11:53

    “Back when I was learning Keynesianism it was called ‘pump-priming’. So, even if it was over by mid-2010 it should have effectuated, if not a boom, at least normal economic activity.”

    Hm. $30 billion over 18 months to fill a hole of more than $600 billion per year. So when you were taught Keynesianism multipliers clocked in around 30? Impressive.

    Even if you counted the entire deficit of about $200 billion per year, that still would have required an overall multiplier of 3. No… there is no way “pump-priming” of this size gets the UK to “normal economic activity”.

  12. Gravatar of D R D R
    9. February 2012 at 12:01

    Check that… I’m not all that convinced of the numbers I’m seeing.

  13. Gravatar of ssumner ssumner
    9. February 2012 at 13:19

    Morgan, Your obsesssion with public employee pay is becoming unhealthy.

    Britmouse, Yes, people weren’t paying attention. I wonder how many Keynesians in the UK were criticizing the BOE for tight money back in 2008. You’d know better than me.

    Marcus, Good point.

    Peter, Yes spending is probably better to look at.

    dwb, Good point.

    DR, The conservatives didn’t take over until mid-2010, and didn’t change macro policy until 2011. Even in 2011 the deficit was third largest, trailing only Greece and Egypt. Obama can sort of blame Bush, but Brown’s policies were in effect for 2 years longer than Bush’s policies. I think Labour must accept a lot of the blame for the dismal performance of Britain during 2006-11. Not saying the Conservatives don’t also deserve blame, BTW.

    Patrick, Yes, it seems based on gut reaction, not careful analysis.

    Major Freeman, You said;

    “The slopes of NGDP are all steeper than the slopes of RGDP. This is consistent with the theory of fiat monetary breakdown.”

    Actually, it just means that most countries have a positive inflation target. That’s all.

    DR, Yes, you may have the wrong numbers. No way the hole was three times the deficit (which was 10% of GDP.)

  14. Gravatar of D R D R
    9. February 2012 at 13:36

    Scott,

    The later you start the clock on Cameron, the more “austerity” you get under Brown. (Nominal) net borrowing was lower in Q1 2011 than it was in Q3 2010 than it was in Q1 2010.

    http://www.ons.gov.uk/ons/rel/naa2/quarterly-national-accounts/q3-2011/rft-data-tables-q3-2011-month-3.xls

    You don’t get it both ways.

  15. Gravatar of Richard W Richard W
    9. February 2012 at 14:14

    I think we need to have clear definitions of what is meant by fiscal stimulus. I remember seeing in the IMF reports in 2009 that the UK fiscal stimulus as the IMF defined stimulus was one of the most modest in the developed world. Germany was vocal in opposing fiscal stimulus and in reality had one of the largest stimulus. Is the fiscal deficit really a stimulus? If an economy had a plunge of 20% in tax revenues, is it really accurate then to say that they are stimulating by the amount of the fiscal deficit?

    The UK deficit figures in the economist are not accurate.
    The UK public sector net borrowing was 8.3% of GDP in calendar 2011. However, the UK Treasury does not book net interest income from QE and other interventions as income. The U.S. Treasury does book QE income from the Fed in the public accounts. Therefore, comparisons between the two are distorted. There does not appear to be any reason why it should not be booked as income and if it was that puts the UK deficit at 6% of GDP in calendar year 2011.

  16. Gravatar of Major_Freedom Major_Freedom
    9. February 2012 at 14:20

    ssumner:

    I said: “The slopes of NGDP are all steeper than the slopes of RGDP. This is consistent with the theory of fiat monetary breakdown.”

    You said: “Actually, it just means that most countries have a positive inflation target. That’s all.”

    Sorry, I meant to say

    The increase in slopes of NGDP are all greater than the increase in slopes of RGDP.

    This cannot be explained by mere inflation targeting. Pre-2008, the difference in slopes was smaller than post-2008.

    If we start even earlier than what the graphs show, we’d see an even smaller slope difference.

    A relative lowering of the slope of RGDP compared to NGDP means that NGDP is having a smaller and smaller effect on RGDP.

    This gels with the following chart:

    http://i.imgur.com/IYScW.jpg

    Which shows the declining effect that debt is having on GDP.

  17. Gravatar of marcus nunes marcus nunes
    9. February 2012 at 14:42

    @ Morgan
    I basically agree with you: Too much government can be hazardous to your health!
    http://thefaintofheart.wordpress.com/2012/01/31/the-austerity-debate/

  18. Gravatar of D R D R
    9. February 2012 at 15:45

    “I think we need to have clear definitions of what is meant by fiscal stimulus.”

    You have a second.

    “Stimulus” should be exogenous and anti-cyclical. (“Anti-cyclical” meaning that which attenuates the business cycle, as opposed to “counter-cyclical”– that which runs counter to the business cycle.)

    An endogenous, anti-cyclical policy is an “automatic stabilizer.” Tax revenues are “cyclical” (as “pro-cyclical” should be reserved for that which amplifies the business cycle.)

    So if we plan for G=$100 this year and G=$200 next year and G=$300 the year after that, but Y falls suddenly from $500 this year to $400 next year and rises to $750 the year after that, then G is merely exogenous while G/Y (having risen from 20% to 50% and then down to 40%) is “counter-cyclical.”

    Yet if, noticing the fall in Y, we change plans and lower G to $160 in the second year, then this is pro-cyclical… anti-stimulus… austerity.

    Of course, that all assumes that we actually agree on what is pro- versus anti-cyclical policy. But I hope you catch my meaning.

  19. Gravatar of Morgan Warstler Morgan Warstler
    9. February 2012 at 16:50

    nunes, two things:

    1. do you agree that a level target NGDP (Scott will accept 45), would quickly make either/or choices apparent on where the growth gets to come from (public vs. private)?

    2. The real numbers are that F,S,L employees in held to 1998 compensation + inflation would be at about $12.5, instead they are at about $1.75T.

    That’s $500B too much this year, it has grown steadily above inflation since 1998…. combined the waste plus interest – comes in over $7T.

    I want to convince Scott, that IF he believes the fiscal multiplier is zero (maybe less), then there has been a serious negative effect on RGDP and a positive effect on inflation (though NGDP trend was +5%), we have grown public wages rather than demanding they give us 3-5% productivity gains every year (firing Bob to pay for Tom and Steve’s raises).

    I can’t begin to model this and would be grateful for the lesson in how to try.

    My grand plan is to show conservatives that inflation targeting grows govt. faster than NGDP targeting.

  20. Gravatar of Morgan Warstler Morgan Warstler
    9. February 2012 at 16:50

    $1.25T

  21. Gravatar of marcus nunes marcus nunes
    9. February 2012 at 16:58

    Morgan
    BTW call me Marcus.
    Scott wrote above that your “obsession” with PS employees is becoming “unhealthy”. Maybe so, but to each his preffered obsession!
    One “transmission channel” from IT to more G/Y is that there´s always the risk (like in 2008) that with the ZLB being reached MP loses “power” and you have to do Fiscal Stimulus (more G/Y). Under NGDPT level targeting that “channel” is “closed”.

  22. Gravatar of ssumner ssumner
    10. February 2012 at 08:24

    DR, You said;

    “You don’t get it both ways.”

    Fine, but the LEVEL of the stimulus in 2011 was still huge, the world’s third largest deficit. So the more you emphasize the decline from 2010, the more you emphasize just how huge the 2010 stimulus must have been. You can’t have it both ways.

    Richard, I’ll defer judgement on the interest income question, I’m simply not qualified to judge. But but do they take into account that when you buy a thirty year bond at a time when short term rates are at the zero bound, the annual interest payment on the bond may greatly exceed the total expected return? Borrowing short and lending long isn’t a magic money machine. But again, I’ll defer any final judgment until I learn more about that issue.

    Keynesian tend to focus on government spending, and Marcus Nunes’s post shows that the rise in G was unusually large for Britain.

    Major Freedom. You said;

    “The increase in slopes of NGDP are all greater than the increase in slopes of RGDP.”

    I think your eyes deceive you. Slopes are different from gaps.

    DR, Do you really think that in future recessions, when the UK debt is much higher than in 2007, it will be possible to do much bigger stimulus packages than the very Keynesian Gordon Brown did? I’m not saying there isn’t SOME level of stimulus that would work (think WWII as an obvious case) but that level is simply not feasible in a modern welfare state starting with a big public debt.

  23. Gravatar of Aggregate Supply Really does Matter « Economics Info Aggregate Supply Really does Matter « Economics Info
    10. February 2012 at 11:01

    […] Source […]

  24. Gravatar of D R D R
    13. February 2012 at 15:57

    Scott,

    I don’t accept your framing at all. Who cares? Ours was bigger, and Greece’s was smaller. There were 14 countries with net lending as a higher share of GDP than the UK, and neither Greece (#20) nor Latvia (#57) were among them.

    I don’t care if the deficits were large. Unlike some, I do not conflate deficits and “stimulus.” I don’t consider it stimulus to let revenues fall endogenously with output, and I don’t consider it stimulus to maintain exogenous government expenditures when output falls. Each of these I consider not throwing gasoline on the fire. It’s a nice start, but the least one could do.

    From what I can gather, Brown’s announced stimulus was for about 20 billion pounds over 18 months, or 3.3 billion pounds per quarter and left an output gap of about ten times that amount. At 35 percent of GDP, this implies a revenue loss of about three times the amount of the stimulus– on top of the tax cuts.

    If you have better numbers let me know, but I don’t think it was the stimulus what done that deficit.

  25. Gravatar of Britmouse Britmouse
    14. February 2012 at 02:54

    I was so annoyed by Portes’ revisionism, I started a blog so I could attempt a hatchet job:

    http://uneconomical.wordpress.com/2012/02/10/apocalypse-then/

  26. Gravatar of ssumner ssumner
    14. February 2012 at 17:35

    DR, I just think that when a country has one of the three biggest deficits in the world the argument that they are sliding into recession because the deficit isn’t even bigger has no credibility. Sure it was partly due to the recession, but the rest of Europe saw similar declines in GDP during 2009, Britain wasn’t unique.

    And what was 35% of GDP? Surely not the output gap?

    Britmouse, Excellent post.

Leave a Reply