NGDPLT will not eliminate recessions

Noah Smith responded to my recent post:

Update 6: Scott Sumner comments. He disagrees with my observation that the freshwater/saltwater conflict has waned. His answer to the question of “what causes recessions” is “aggregate demand shocks”, but this doesn’t really answer the question of what is causing those shocks (of course, Scott Sumner thinks NGDP targeting can perfectly cancel out any aggregate demand shocks, so maybe he doesn’t even care where the shocks come from). He agrees that modern macro methods (DSGE models) have given us no useful technology of any kind. And he endorses replacing the current mainstream macro profession with “Market Monetarists” like himself and David Beckworth.

Yikes!  Just to clarify:

1.  I’ve done many posts answering the question of “what is causing those shocks.”  Demand shocks are caused by changes in either the supply or demand for the medium of account—and they occur for a reason.  Or one can think of them being caused by bad monetary policy, if you think (as I do) that it’s the central bank’s job to deliver a stable growth path for expected future NGDP.  But outside factors obviously impact the demand for base money.  In late 2007 and early 2008 a slowdown in base growth was the main culprit.

2.  I do not think all recessions are caused by demand shocks, nor did I say so.

3.  I do not think NGDPLT can perfectly stabilize demand, nor did I say so.

4.  The suggestion that Beckworth replace Bernanke was sort of half-joking, in response to a point by Noah Smith.  I know that won’t happen.  The serious point is that I’d like to see market monetarism replace the current standard model in macro.  I don’t seriously expect a handful of market monetarists at small schools to replace the entire establishment, if one is thinking in terms of bodies, not ideas.

I would add that if NGDPLT is adopted, there will come a time when it will be perceived to have failed, because it can’t perfectly stabilize AD, and there are also real shocks.  So there will still be occasional business cycles, although I’d expect them to be milder.  When they occur, people will say that NGDPLT “failed,” just as when China finally has a recession, they’ll claim the mythical “China bubble” has collapsed.

However it’s important to realize that even mild recessions like 1991 or 2001 did not seem mild at the time.  I recall articles in both years about how “this one is different” perhaps “heralding the end of the American dream, blah blah, blah.”  People always tend to overreact to current events.  Remember after 9/11 how all the experts assured us that we were in a new world, a “war on terror.”  (And then the other side never showed up for the war.)  So perhaps the adoption of NGDPLT would not be the end of history that I’ve claimed it would be, but I do think it would help to gradually unite macro and micro, leaving only RBC analysis on the macro side.

And I do think NGDPLT can prevent severe slumps like 1921, 1930, 1938, 1982, 2009, etc.

PS.  Merry Christmas!

PPS.  Because of the holidays I won’t have time for much blogging, but may post a few from my queue.



42 Responses to “NGDPLT will not eliminate recessions”

  1. Gravatar of Saturos Saturos
    25. December 2012 at 07:16

    A quick reply –

    I was under the impression that Scott had argued that NGDPLT could essentially eliminate the business cycle. I can’t remember where he said it more explicitly, although I am pretty sure he did so, but I do remember these posts: and the one after.
    I think that is where Noah is getting his opinion from as well:

    I have a distinct sense (perhaps mistaken) with this post that Scott is walking back from earlier claims that he could more or less solve “demand” problems completely with NGDPLT, offhand as well as in the body of posts, though I can’t find the links. Can anyone help me?

    Anyway, Merry Christmas to all!

  2. Gravatar of marcus nunes marcus nunes
    25. December 2012 at 07:17

    NGDP-LT will certainly make recessions milder. But as usual, in that sort of world ANY recession will fell “like the end of the world”. But at least we would not be mourning, like John Taylor, the 5 year anniversary of the end of the Great Moderation!

  3. Gravatar of ssumner ssumner
    25. December 2012 at 07:33

    Saturos, It would end large demand side business cycles. I’d be very surprised if someone could find a post where I said it would eliminate supply side business cycles, such as 1974–and there are many, many, many posts where I say it would not end supply side business cycles.

    As for the claim that it would work “perfectly”—does anyone really believe I’m that stupid?

  4. Gravatar of ssumner ssumner
    25. December 2012 at 07:43

    Saturos, The post you cite does suggest that demand shocks play a role in virtually all US recessions, but supply shocks do exist, and NGDPLT can’t eliminate them. The post you linked to mentions that Japanese real GDP did fall briefly after the tsunami. However I believe the US business cycle would be far milder without demand shocks.

    Demand shocks cannot be totally eliminated, even with NGDPLT, indeed even with futures targeting. So there will still be a business cycle, but much milder than what we currently experience. I probably would not even have responded to Smith if he had not used the term “perfectly”. When people use terms like that I know they aren’t engaging in a serious discussion, but rather just trying to score debating points.

    In many post I emphasize the great improvement we’d get from NGDPLT, which probably leads people to think I view it as a sort of magic wand.

  5. Gravatar of W. Peden W. Peden
    25. December 2012 at 08:31

    Great post.

    Similarly, NGDPLT won’t prevent a little (temporary) inflation from an oil shock or a world war, but it will prevent inflations as occured in the 1970s.

    I was reading the London Times and I noticed that one argument tentatively put by David Smith (who seemed basically sympathetic to NGDP targeting, which is not surprising since he seems to be fairly monetarist in orientation) against a switch to NGDP targeting was that the kind of “flexible” inflation targeting that we’ve had in the UK is roughly the same as NGDP targeting.

    However, I do think there is a very important difference: inflation-targeting today is roughly in the same condition as monetary aggregate targeting circa about 1987. Policymakers tend to pick definitions of inflation that fits what they want to do, just like latter-day monetarism. Base drift and missing targets have become commonplace. All this means that when AD picks up again and starts to become excessive, inflation-targeters will have plenty of ways to avoid tightening, e.g. “core inflation is still low”, “we expect inflation to fall over the next years”, “inflation statistics are not necessarily reliable”, “inflation is one of many factors, including the exchange rate, interest rates, asset prices, and unemployment, which we look at” etc.

    The best thing about NGDPLT is not that it is an answer to the current crisis, but that it is also an answer to the next crisis.

  6. Gravatar of W. Peden W. Peden
    25. December 2012 at 08:33

    Also, while NGDPLT doesn’t avoid supply-side recessions, it does imply the most rational response to them, which is to maintain the growth of spending at a steady rate (avoiding permanent inflation, as in the 1970s) and allowing the market pricing system to do its work e.g. providing incentives for energy-producers to invest and energy-consumers to cut use during an energy-crisis.

  7. Gravatar of Jon Jon
    25. December 2012 at 09:14

    Yes, Noah seems to be affirming the consequent. You say if ngdp instability then recession. Noah heard if recession then NGDP instability.

    Incidentally, although AS shocks will cause declines in real output–aka a recession, they won’t necessarily cause unemployment. When they do cause unemployment that has more to do with automatic COLA in wage contracts or the knock on effects of inflation targeting.

  8. Gravatar of ChargerCarl ChargerCarl
    25. December 2012 at 10:36

    have we had any supply side recessions post WWII? I remember you did a post on this a long time ago and you argued that we’ve had only one, and it was iffy at best.

  9. Gravatar of Bill Woolsey Bill Woolsey
    25. December 2012 at 11:01

    I don’t accept the definition of recession as reductions in real output.

    A asteroid destroys half of the U.S. We calculate real GDP. It has fallen by 40%. Oh, there was another Great Depression?

    In my view recessions are situations where real GDP is below potential GDP. Potential GDP is not the trend of real GDP, but rather what real GDP would be if prices and wages were fully flexible. (That is only rough and ready.)

    So, if potential output grows more slowly, or even falls, and real output stays with it, there is no recession.

    NGDPLT will result in recessions when spending alls below the target growth path. There will be powerful forces that tend to prevent this from happening and reverse such devations allowing for rapid recovery. But it will happen.

    Sumner (and I) doubt that it will be stuck below target and there seems to be nothing that can be done to get it back.

    Second, it will sometimes go above target, and the powerful forces getting it back down will involve slower real output growth and perhaps an absolute decrease. For the most part, this will not be a recession, however, but rather a return of real GDP to potential. There will be powerful incentives to take care in an overheating economy. But errors might be made.

    While shifts in potential output (and changes in the price level due to shifts in the supplies of particular goods) will sometimes result in mild recessions.

    Of example, if the supply of a good with inelastic demand increases, the price level rises more than in proportion to the decrease in real income. For total spending to remain on target, spending on other goods must be dampened. It could go the other way.

    If all decreases (or slowdowns) in real GDP count as recessions, then we have slower growth in potential, errors causing nominal GDP to grow more slowly or shink, and then a return of nominal GDP to the target path after an error has caused it to rise above target.

  10. Gravatar of Suvy Suvy
    25. December 2012 at 11:11

    I actually think that targeting nominal growth may increase the amount of recessions, but it would decrease their severity. I think it would to this because it prevents the economy from getting too out of control as it is a better tool to prevent an overheating economy than inflation targeting is.

    I would also like to add that I think a permanent low inflation low unemployment equilibrium is impossible in an economy. Stability itself is destabilizing and for a long stretch of stability; you will get a long stretch of instability. This is why central banks should not target inflation and unemployment. They should target nominal growth and debt/income ratios.

  11. Gravatar of Scott of course “Recessions are always and everywhere a monetary phenomena” « The Market Monetarist Scott of course “Recessions are always and everywhere a monetary phenomena” « The Market Monetarist
    25. December 2012 at 11:31

    […] Sumner has a new post in which he claims that “I do not think all recessions are caused by demand shocks”. […]

  12. Gravatar of jknarr jknarr
    25. December 2012 at 11:31

    Please add a pinch of Minsky spice to this Christmas pudding.

    The stability engendered by NGDPLT programming will be “hackable” via leverage. The stability of NGDPLT policies will likely engender more leverage. Hence, it will still create overleveraged whipsaws for the real economy.

    Patient capital may even be able to hack the NGDP response mechanism quite profitably and pro-cyclically. That is, NGDPLT will utimately be discredited unless accompanied by removing debt discretion power from the Fed, ie reserves.

    So can we call it NGDPLT-RE? (reserve elimination) Again, a central bank can target and stabilize the economy or debt, but not both.

    Or possibly a 50% NGDP 50% debt targeting regime, but oy veh…

    Debt destabilizes, even the pure world of NGDPLT. Moving toward NGDP economic targets must mean weaning the banking system from the money authority teat. This would keep debt to systemically unimportant levels, and stabilize the economy.

    I’m not certain why there is no discussion of leverage. Leverage will”price in” NGDPLT stability and destabilize the economy, simply with a different reaction function.

    This board is in cloud cookoo land if we fail to address debt and reserves under NGDPLT.

  13. Gravatar of Don Geddis Don Geddis
    25. December 2012 at 11:34

    Suvy: you make assertions, but don’t back them up. Why (do you think) NGDPLT would “increase the amount of recessions”? Why (do you think) that “a long stretch of stability” necessarily causes “a long stretch of instability”?

    Neither of those assertions are commonly accepted in macro, so you’ll need to do a better job of defending your ideas.

  14. Gravatar of Suvy Suvy
    25. December 2012 at 12:07


    I’ve explained these a lot on this blog. I think that targeting NGDP would be a good way to prevent an economy from overheating. For example, if you were to have 6% growth in the US with a 15% increase in debt/income ratios, you have an economy that has major issues as the boom is unsustainable and will lead to a crash. With NGDP targeting, you would be cooling down the boom and creating inflation when there is a crash. So in a situation of sustained deleveraging; you’re helping the economy delever by creating inflation. You would be reducing the period of deleveraging–thus reducing the severity of the recession. The reason I think you may have more recessions is because I think you would be preventing the economy from overheating more often. I would like to add that this particular assertation could easily be wrong as NGDPLT has never really been done before. It’s simply a prediction.

    The other part is straight from Minsky. It starts with the fact that the future is unpredictable; so expectations actually shift over time. If you have a period where a crisis just occurred, both lenders and borrowers are very safe. As the memory of the crisis fades, lenders and borrowers become more reckless as expectations of returns (whether income related earning off assets or speculative gains off assets) start to rise due to the change in expectations. This causes debt/income ratios to rise over time. As they rise, this adds to demand. Basically, there ends up being a positive feedback loop between rising leverage and asset prices. The rise in debt is not a zero-sum game because the creation of credit adds to demand. Eventually, debt/income ratios get to a point where they cannot rise any more; that’s when a financial crisis occurs. Then, as debts fall, you get stuck in a debt deflation. Therefore, the central bank comes in to stop the collapse of credit and creates inflation/monetizes the debt in order to reduce debt/income ratios. You get a period of stagflation as the debt/income ratios correct and then the cycle starts over. I don’t agree with the view that is commonly accepted in macro that the economy is in equilibrium except for exogenous shocks. You have to include the fact that there is a financial sector whose job is to increase the capital stock. However, the financial sector also finances rising leverage and that can trigger a massive rise in asset prices.

    My view is like this, the economy is a dynamic, nonlinear system that is constantly changing states. Equilibrium may exist and is important; however, equilibrium itself is a transitory state. Not only that, but the existence of equilibrium (or multiple equilibria) does not say anything about the stability of the system itself. For example, the equilibria itself could be unstable. Or the equilibria could be locally stable, but globally unstable. This is very common for the behavior of nonlinear dynamic systems. Not only that, but the economy (due to the fact that it is nonlinear and dynamic) will be highly sensitive to the initial conditions. We could also have an issue where the economy heads to limit cycles as equilibrium behavior. In other words, the economy is path dependent. A very simple example of the kind of behavior that I am talking about is displayed in very simple nonlinear dynamic systems–like the Lorenz attractor. Now, when we add human behavior and the shifts of expecations/risk profiles over time; we get very, very complex behavior.

    As you can see, I am not a fan of the rational expectations theory and I think the EMH needs to be changes. I think the EMH needs to start incorporating the fact that there are correlations in the data, that price movements and errors are correlated(look at the Hurst exponent, it clearly varies over time as there exist periods of strong trend structure). As for rational expectations, it doesn’t account for herd behavior. It doesn’t account for the fact that expectations shift over time.

    I think a better alternative to the EMH is the Fractal Market Hypothesis put forth by Edward Peters while a better alternative to the rational expectations theory is the Financial Instability Hypothesis.

  15. Gravatar of Philo Philo
    25. December 2012 at 16:13

    All I wanted for Christmas was a TheMoneyIllusion post, and I got it! Thanks, Scott, and Merry Christmas.

  16. Gravatar of Dan Dan
    25. December 2012 at 18:09

    Do Macro-Economists ever give a thought to how an economy actually works? Every post I read of Sumner leaves me with the impression that all that matters to him is for the national economic numbers to “look good”. Whether people have productive jobs, whether they enjoy freedom & liberty, whether they enjoy a high quality of life seem to be inconsequential to his thesis.

    Yet the only reason we should care about national economic policy is so for the citizens of that nation to enjoy freedom, peace & prosperity.

    It is not just that macro-Economists, including Sumner and Krugman, have the cart before the horse. They present economic models that disregard the need for horse. They then argue amongst themselves about how to move the cart faster. All the while the realists keep asking, “Where is the horse?”

  17. Gravatar of flow5 flow5
    25. December 2012 at 18:13

    “In late 2007 and early 2008 a slowdown in base growth was the main culprit”

    The payment of interest on excess reserve balances prevented the correction of subpar “base” growth. And the IOR policy will prevent correcting it in the future.

    Preventing the 1974 contraction would have been no different than adjusting other business cycles since 1942. It’s the same. It’s extraordinarily easy.

    SEE Article: Member Bank Reserve Requirements — Analysis of Committee Proposal; published — Feb, 5, 1938 laid it out:

    “In 1931 this committee recommended a radical change in the method of computing reserve requirements, the most important features of which were…”
    (1) “Uniform percentage requirements against the volume of deposits of both types and in all classes of cities”
    (2) “Requirements against debits to deposits”
    (3)”the establishment of a low reserve against time deposits in 1914 has facilitated the growth of bank credit without a corresponding growth in reserves”
    (4)”the committee recommends that all member banks and all deposits be treated alike for reserve purposes”
    (5)”the committee proposed that reserve requirements be based upon the turnover of deposits”

    Strangely, this research paper was DECLASSIFIED on March 23, 1983.


  18. Gravatar of flow5 flow5
    25. December 2012 at 18:20

    “In late 2007 and early 2008 a slowdown in base growth was the main culprit”

    Unfortunately the Federal Reserve doesn’t gauge the volume & timing of its open market operations in terms of the amount & desired rate of increase of member commercial banks costless legal reserves, but rather in terms of the levels of interest rates.

    By using the wrong criteria (interest rates, rather than member bank reserves) in formulating & executing monetary policy, the Federal Reserve became the business cycle’s nemesis.

  19. Gravatar of flow5 flow5
    25. December 2012 at 18:30

    At the height of the stock market bubble, Greenspan initiated a “tight” monetary policy (for 31 out of 34 months). A “tight” money policy is defined as one where the rate-of-change in monetary flows (our means-of-payment money times its transactions rate of turnover) is no greater than 2-3% above the rate-of-change in the real output of goods & services.

    Greenspan then wildly reversed his “tight” money policy (at that point Greenspan was well behind the employment curve), & reverted to a very “easy” monetary policy — for 41 consecutive months (i.e., despite 17 raises in the FFR, -every single rate increase was “behind the curve”). I.e., Greenspan NEVER tightened monetary policy.

    Then, as soon as Bernanke was appointed to the Chairman of the Federal Reserve, he initiated a “tight” money policy (ending the housing bubble in Feb 2006), for 29 consecutive months, or at first, sufficient to wring inflation out of the economy, but persisting until the economy plunged into a depression).

    The FOMC continued to drain liquidity despite its 7 reductions in the FFR (which began on 9/18/07). I.e., despite Bear Sterns two hedge funds that collapsed on July 16, 2007, & immediately thereafter filed for bankruptcy protection on July 31, 2007 — as they had lost nearly all of their value), the FED maintained its “tight” money policy (i.e., credit easing, not quantitative easing).

    I.e., Bernanke didn’t initiate an “easy” money policy until Lehman Brothers later filed for bankruptcy protection (& it was one the Federal Reserve Bank of New York’s primary dealers in the Treasury Market), on September 15, 2008.

    And Greenspan didn’t start “easing” on January 3, 2000, when the FFR was first lowered by 1/2, to 6%. Greenspan didn’t change from a “tight” monetary policy, to an “easier” monetary policy, until after 11 reductions in the FFR, ending just before the reduction on November 6, 2002 @ 1 & 1/4% (approximately coinciding with the bottom in equity prices).

    I.e., Greenspan was responsible for both high employment (June 2003, @ 6.3%), & high inflation (rampant real-estate speculation, followed by widespread commodity speculation).

    Bernanke relentlessly drove the economy into the ground, creating a protracted un-employment, & under-employment rate, nightmare.

  20. Gravatar of Dan Dan
    25. December 2012 at 18:31

    A dose of Austrian to explain what all the NGDPLT in the world cannot explain:

    “Capitalism creates a critical frame of mind which, after having destroyed the moral authority of so many other institutions, in the end turns against its own; the bourgeois finds to his amazement that the rationalist attitude does not stop at the credentials of kings and popes but goes on to attack private property and the whole scheme of bourgeois values.” – Joseph Schumpeter

  21. Gravatar of Saturos Saturos
    26. December 2012 at 00:47

    Scott, so knowledge of market monetarism will still be useful after NGDPLT is implemented? Good to know. I assume you still think that fiscal stimulus would be useless, though.

    Philo, ok that’s sad… I got chocolate for Christmas… lots of it… people know what I like.

    And please don’t say you’re one of those economists who think we should all give cash for Christmas, or the ones that think Christmas is the day to celebrate Scrooge. I’ll tell you where to to start improving the profession, ban those articles, along with any talking about how commercialized the holiday is (poor misers who didn’t get any presents themselves).

  22. Gravatar of Saturos Saturos
    26. December 2012 at 00:51

    Dan, Scott wants people to have jobs at least as good as they would have if monetary disequilibrium (such as Hayek wrote about later on) didn’t take them away, instead of the situations like we have now, where the mass suffering makes people more skeptical of capitalism and liberty and leads to worse and worse policies. But generally maximizing liberty and prosperity is a separate challenge, that’s not where Scott’s comparative advantage is. Scott has said this before. Read through the archives; there’s actually a lot more to this blog. (You might learn a fair bit about movies, for instance.)

  23. Gravatar of W. Peden W. Peden
    26. December 2012 at 05:51


    “no greater than 2-3% above”

    A minimum cannot be a range.

  24. Gravatar of W. Peden W. Peden
    26. December 2012 at 05:52

    * And a maximum cannot be a range either.

  25. Gravatar of Negation of Ideology Negation of Ideology
    26. December 2012 at 06:08

    jknarr –

    “This board is in cloud cookoo land if we fail to address debt and reserves under NGDPLT.”

    This board has discussed debt and reserves. If there’s a problem with debt, it can be addressed through improved regulation, financial education and changes to bankruptcy law. I don’t see how a central bank can target debt – maybe you could give an example of how that would work.

    I do see how NGDPLT could work. And it could work with 1% bank reserves or 100% reserves. It could work in a credit card culture or in a culture where everyone saved and paid cash for everything. The monetary base would be a different size (relative to GDP) in each case, but the central bank could hit the target (or at least come close and make up deviations as Scott descibes).

  26. Gravatar of Cthorm Cthorm
    26. December 2012 at 08:01

    So a reasonable summary of your entirely reasonable view is:
    “NGDPLT will minimize the occurrence and severity of recessions, in so far as monetary policy as the power to do so.”

    That doesn’t mean NGDPLT is the cure-all, there are lots of policy improvements that can help ensure consistent growth.

  27. Gravatar of flow5 flow5
    26. December 2012 at 08:14

    “A minimum cannot be a range”

    Then delimited by a range? The point is that no money supply figure standing alone is adequate as a “guide post” for monetary policy. And Vt is the only valid velocity figure. Vi is contrived. Money by itself is as useless (a monetary transmission mechanism) as interest rates. And the effect of using interest rates is indirect, varies widely over time, & in magnitude.

    I had a Ph.D. in statistics (in 79) validate the debit series against gDp & AAA corporates. He said there was no correlation unless there was a lag. When I established the lag for real-gDp & inflation the data was highly correlated. When the G.6 was discontinued I used the next best metric. It subsequently worked better for real-gDp than inflation, but still works. There’s nothing better than the 2 numbers combined.

    These metrics aren’t coincident or lagging indicators. They are ex-ante because of their periodicity. Forecasts are virtually infallible. None of this is in the economic literature.

    Greenspan & Bernanke (preeminent scholars) still don’t understand how they failed. If the average citizen knew that Bernanke was responsible for taking their check, their food, & the roof over their head, Bernanke wouldn’t last long.

  28. Gravatar of Philo Philo
    26. December 2012 at 08:28

    @ Saturos
    The sad truth is that chocolate is not good for you. I prefer to eat a piece of fruit instead, while exulting in my dietary virtue. (But even better is reading good online commentary on economic policy.)

  29. Gravatar of Negation of Ideology Negation of Ideology
    26. December 2012 at 08:45

    Cthorm – Great summary. Monetary Policy can’t solve everything.

    Dan – I think the problem you’re having is that monetary policy can only deal with monetary problems. Deciding whether the currency should be tied to NGDPLT, CPI, gold, silver, discretion, or whatever is incredibly important, but not the only important thing.

    There is no Utopia on this side of the grave. If someone said the federal highway system should use a different kind of pavement would you argue “You road experts think building better roads will solve all our problems!”? Or would you evaluate the proposal on its own merits?

    A national government has to make a decision on what it will use for money – for taxes, fines, government spending, civil lawsuits, etc. The question is simply what system of public money will be most beneficial (or least harmful) to the citizens of that country. I believe it’s NGDPLT.

    That doesn’t mean that taxes, regulation, education, infrastructure and everything else isn’t important.

  30. Gravatar of Saturos Saturos
    26. December 2012 at 09:30

    Philo – fruit? Were you naughty this year Philo? Are you in Santa’s bad books? ;(

  31. Gravatar of Saturos Saturos
    26. December 2012 at 09:34

    But then perhaps you’d agree with Dogbert…

  32. Gravatar of Brian Donohue Brian Donohue
    26. December 2012 at 09:40

    heh. ‘this one is different’ has become go-to material in the event of any economic slowdown, along with ‘jobless recovery’.

  33. Gravatar of ssumner ssumner
    26. December 2012 at 10:25

    W. Peden, You said;

    “The best thing about NGDPLT is not that it is an answer to the current crisis, but that it is also an answer to the next crisis.”

    Good point.

    Jon, Good point.

    Bill, Suppose VAT increased sharply–that could cause a recession under NGDPLT.

    jknarr, The real problem is that we are in a new world of permanently lower real interest rates. That’s what will cause all the leverage.

    Dan, Have you ever considered reading my posts advocating deregulation, free emarket reforms, privatization, vouchers, eliminating income taxes, etc.?

    Saturos, Yes, fiscal stimulus will still be worthless. I think part of the confusion yesterday came from people drawing too broad a conclusion from various isolated comments I made at very different times.

    1. Comments that people overrate the importance on non-monetary shocks. They do, but real shocks still matter.

    2. Comments that nominal shocks are especially important in big diversified economies like the US.

    3. Comments that the Fed could almost perfectly peg NGDP expectations (but forgeting that actual NGDP can change.)

    4. People overlooking my many comments that NGDPLT would have made the 2008-09 recession MILDER. I must have said ‘milder’ 100 times in the blog. That’s well short of “perfection.”

    5. I frequently call myself a “moderate supply-sider” Does that sound like someone who thinks real shocks have zero effect?

    My hunch (it’s just a hunch), is that under NGDPLT recessions would be at worst like 2001-03, which maxed out at 6.3% unemployment, and would occur every 10 to 20 years (i.e. less frequently.)

  34. Gravatar of ssumner ssumner
    26. December 2012 at 10:28

    Saturos, I care about Christmas because I care about other people (well, at least a few others) and other people care about Christmas.

  35. Gravatar of Doug M Doug M
    26. December 2012 at 11:22

    I don’t see the distinction between a supply shock and a demand shock.

    Resources may become more scarce than thought. The weather could be bad. We may over-invest in new technologies based on optimistic assumptions. Companies sometimes lose money. The future is hard to predict. Shocks happen.

    When a shock happens, inefficient markets are slow to react to new information. Prices and wages are no-longer market clearing, resulting in unemployment.

    So, the question: Would NGDPLT make these markets more effiecint? Would prices and wages to react more qickly to changing informaton?

  36. Gravatar of Federico S Federico S
    26. December 2012 at 11:32

    Hi Scott,

    One small question — I’ve always interpreted your view of AD as being NGDP. So implementing NGDPLT should, by definition, stabilize AD. So when you say that NGDPLT won’t always stabilize AD, what is it that you mean by AD? Or are you just referring to corner cases where measurement errors distort NGDP?

    Thanks and happy holidays!

  37. Gravatar of anon anon
    26. December 2012 at 12:38

    Scott, so your claim is that we can still have demand-caused recessions, because the path of actual NGDP will be noisy, even though NGDP expectations are pegged, and they’re always realized on average? This is a very interesting point, which may deserve more attention. Here’s a relevant post by Vaidas Urba.

    BTW, is this why you support automatic fiscal stabilizers – because of their short-term effect on the NGDP path?

  38. Gravatar of Bill Woolsey Bill Woolsey
    26. December 2012 at 12:49

    Forget NGDPLT targeting.

    NGDP minus taxes on production and inport taxes targeting.

    (I am not sure about the imports taxes.)

  39. Gravatar of Scott Sumner Scott Sumner
    26. December 2012 at 14:32

    Federico, Policy lags–nothing works perfectly.

    anon, No, I support automatic stablizers because they minimize the deadweight loss of taxes and inefficient fluctuations in production of government output. But if they also help overcome the lag in monetary policy, that’s great (I don’t have an opinion either way on that question.)

    Bill, Yes, those tweaks would help make recessions even milder. But there are other issues such as payroll taxes, etc, that could also cause problems.

  40. Gravatar of Major_Freedom Major_Freedom
    26. December 2012 at 21:46

    NGDP targeting brought about by inflation CAUSES recessions, because it distorts the price system and misleads investors who work in a division of labor, where investments are made independently of each other and people are partially ignorant of other people’s preferences.

  41. Gravatar of Greg Ransom Greg Ransom
    27. December 2012 at 00:32

    It would be poetic justice if Noah continued to ascribe these views to Scott which he doesn’t hold for another 3 years.

    If Scott feels free to invent bogus positions and false ascribe them willy-nilly to other economists, what right would Scott have to complain.

  42. Gravatar of TallDave TallDave
    28. December 2012 at 06:45

    Excellent post, one quibble.

    (And then the other side never showed up for the war.)

    Well, yes and no. AQ has largely failed to execute here (some were caught, like Richard Reid) but that’s because they’ve been kept busy elsewhere — they’ve killed thousands of U.S. troops and tens of thousands of Iraqis/Afghanis since 2001, not to mention the first U.S. ambassador since 1979 just recently. Sometimes these things look less serious in hindsight precisely because serious action was taken then — like, say, the 1987 crash.

Leave a Reply