Worthwhile Canadian Initiative

As people in the humanities would say, this post’s title is an “homage” to Nick Rowe.  JimP sent me this interesting Bloomberg article:

Montreal undergraduates may help reshape the Bank of Canada’s monetary policy and give Federal Reserve Chairman Ben S. Bernanke and Bank of Japan Governor Masaaki Shirakawa clues about how to ward off deflation.

About 240 students so far have spent two hours in a 25th- floor computer lab near McGill University, earning an average of C$30 ($29.88) by viewing combinations of economic data, including unemployment and gross domestic product, and then predicting what would happen to inflation. Central bank researchers are taking part in the project to see whether people can make such forecasts more easily if policy makers target specific levels in the consumer price index instead of the inflation rate — which might help households and companies make better decisions about spending and investing.

The more accurate the test subjects are, the more they earn. One did so well, “we tried to track this person to see if we could hire her,” said Jean Boivin, 38, a Bank of Canada deputy governor who is helping with the research and has also co-written paperswith Bernanke.

The experiments will help Canada decide if it should switch from inflation targeting to price-level targeting in 2012 and may help the bank better communicate its policies to the public, Boivin said. The test results also might benefit Fed policy makers, who discussed price-level targets on Oct. 15 and voted Nov. 3 to inject another $600 billion of reserves into the banking system to avoid deflation — a widespread drop in prices that has plagued Japan for more than a decade.

Broader Agenda

“Central banks need to know more about how expectations are formed, and so we see that as part of a much broader agenda,” Boivin said in an interview at the Bank of Canada’s Ottawa headquarters in the room where he, Governor Mark Carney and four other policy makers decide on interest rates, including a decision tomorrow that’s scheduled for 9 a.m. New York time.

A few weeks back a Bentley student named David Norrish pointed to a flaw in my NGDP targeting idea.  Why [he asked] should we expect the largest economy in the world to experiment with a risky monetary regime that had never been tried out anywhere else?  I seem to recall that ideas like inflation targeting were pioneered by smaller economies such as New Zealand.  So I’m glad to see that the Canadians are considering serving as guinea pigs for price level targeting, before the policy is deployed in the much more important American economy.

BTW, the critieria for success should not be defined solely in terms of accurate inflation forecasts.  Short term forecasts may be relatively accurate under inflation targeting, even if the price level follows a random walk.  It’s more important to have accurate inflation expectations over the life of nominal contracts (wage and debt contracts.)  That’s where the advantages of level targeting are strongest.

PS.  Canadian readers:  I was just kidding—playing the ugly American for cheap laughs.



48 Responses to “Worthwhile Canadian Initiative”

  1. Gravatar of W. Peden W. Peden
    6. December 2010 at 10:51

    Prof. Sumner,

    In your opinion, what would the range of sensible targets in a price level targeting regime be? For example, with inflation targeting, it seems to be generally agreed that something like 0.1% would have certain avoidable problems and 10% would have certain avoidable problems.

    So, for instance, what might happen if a price level targeting regime targeted a constant price level over 5 years?

  2. Gravatar of jj jj
    6. December 2010 at 11:18

    As a Canadian I happily volunteer as a guinea pig. But only now that it might actually happen have I thought of what adjustments I’ll need to make. Here’s one: What will happen to my mortgage rate? My mortgage rate is based on the bank prime, and I wonder what will happen to that after it’s uncoupled from the BOC overnight rate target.

    Right now my rate is variable, calculated as [my bank’s prime] – 0.5%, where [my bank’s prime] has been stable for months at [BOC Overnight Rate target] + 2.0%. In practice my rate changes every 6 months or so.

    Once the BOC abandons the overnight rate target, what will my bank use to set its own prime rate? They could use actual overnight rates, but those probably won’t yield nice round numbers like 3.0. And what will the frequency and magnitude look like? Right now all the banks change their primes on the same day as the BOC changes its target. It’s also easy to see that they use the same x% when setting prime = overnight + x%.

  3. Gravatar of Tomasz Wegrzanowski Tomasz Wegrzanowski
    6. December 2010 at 11:43

    Not terribly related, but are there any prediction markets for GDP and unemployment in various countries? Even intrade disappoints here.

  4. Gravatar of JimP JimP
    6. December 2010 at 13:13

    Scott is quoted here – at the end.


  5. Gravatar of Nick Rowe Nick Rowe
    6. December 2010 at 16:48

    Thanks Scott!

    This experiment is intended to help resolve one of the questions that came up at the Bank of Canada workshop a few weeks back:

    One of the advantages of price level path targeting over inflation targeting is that it should create “stabilising expectations”. If there’s a negative shock, and the price level falls below target, then, if people understand price level path targeting, they should expect higher than normal inflation in the near future, which, for a given nominal rate, reduces the real rate, increases demand. And this increased demand acts as an automatic stabiliser to partially offset the negative shock.

    But if people can’t figure out the difference between price level and inflation targeting, we won’t get these stabilising expectations.

  6. Gravatar of Mark A. Sadowski Mark A. Sadowski
    6. December 2010 at 17:26

    Thanks to its research program the Bank of Canada is of course a good source of literature reviews on price level targeting. For example:


  7. Gravatar of Morgan Warstler Morgan Warstler
    6. December 2010 at 17:57

    As always, Level Targeting is a big YES, as long as we use it to piss on booms.

    Once we run 4% inflation for two years, that means no matter what happens to unemployment we dial down hard on the screws.

    The issue is what happens when unemployment is 8% and inflation is 4%? It’s only when we BELIEVE the Fed has that internal strength, will anyone believe this is something they can bank on.

  8. Gravatar of JimP JimP
    6. December 2010 at 17:59

    The whiff of deflation.


  9. Gravatar of scott sumner scott sumner
    6. December 2010 at 18:02

    W. Peden, I hate to say “it depends” but it does. With a futures targeting regime a stable price level would be ok. With our current monetary regime, which seems unable to avoid liquidity traps, 3% inflation would be better.

    jj, I think they’d still use interest rates as short term targets, they’d just change their long run target from inflation to the price level.

    Tomasz, That angers me too. I don’t recall seeing any.

    Thanks JimP.

    Nick, I’m not to worried about people figuring things out. The people you really need to figure things out are the financial markets–and they’ll have no trouble. If the Fed had been doing level targeting in late 2008 then asset prices wouldn’t have crashed as far, which means the recession would have been much less deep.

    First you guys create a superior banking system, and now superior monetary policy.

    Mark, Thanks for the info. I made up my mind 20 years ago when I read something by Barro explaining the idea. It just seems to make a lot of sense to me, for all sorts of reasons. It’s very connected to how I visualize the monetary policy transmission mechanism.

  10. Gravatar of scott sumner scott sumner
    6. December 2010 at 18:06

    Morgan, Yes, but NGDP level targeting is even better. But either is better than what we have.

    JimP, Thanks, it’s odd looking at their graph. Bernanke reacts in horror to the thought of above 2% inflation, but they were cutting rates in 2007 when inflation was above 2%.

  11. Gravatar of marcus nunes marcus nunes
    6. December 2010 at 18:51

    This paragraph from the SF Fed Letter is “great”:
    The main difference between slowing prices in the early 1980s and current disinflation is that the large deceleration during and after the 1981 recession was due to a severe tightening of monetary policy. This tight policy stance significantly reduced inflation in 1982 and is sometimes called the Volcker deflation after then-Federal Reserve Chairman Paul Volcker (Blanchard 1984). By contrast, the most recent recession was caused by the bursting of the housing bubble and the ensuing financial crisis. Despite exceptionally low interest rates, the breadth of disinflation has remained quite large since the beginning of the last recession. In October 2010, half of consumer spending was on items whose inflation rate had declined 1.2% or more over the past year.

    The current recession/disinflation cannot be due to MP (which isn´t tight because i is extremely low)!

  12. Gravatar of Morgan Warstler Morgan Warstler
    7. December 2010 at 03:47

    “Thanks, it’s odd looking at their graph. Bernanke reacts in horror to the thought of above 2% inflation, but they were cutting rates in 2007 when inflation was above 2%.”

    Scott, perhaps you should do a post on all the recent times when the Fed should NOT have been cutting rates, according to level targeting.

    Like Fisher, your stuff is much more interesting if it is first being used as a wet blanket… if we don’t have the fortitude, we can’t use your stuff.

  13. Gravatar of Doc Merlin Doc Merlin
    7. December 2010 at 09:16

    ‘It’s more important to have accurate inflation expectations over the life of nominal contracts (wage and debt contracts.) That’s where the advantages of level targeting are strongest’

    Damn right!

  14. Gravatar of Cameron Cameron
    7. December 2010 at 11:05

    More evidence the fiscal multiplier is 0?


    Obviously the question is whether or not the Fed thinks like this, but I it’s pretty likely the Fed would have done more in the absence of fiscal stimulus in 2009 (Bernanke recommended fiscal stimulus after all – although now he claims the Fed has the tools it needs). The stimulus bill may very well have hurt the economy if the impact of fiscal stimulus is overestimated.

  15. Gravatar of Rod Everson Rod Everson
    7. December 2010 at 15:49

    Bad tax policy, and I include over-regulation and government corruption in the definition of “bad tax policy” is, and always has been, capable of driving any economy to its knees.

    And all the monetary and fiscal stimulus in the world won’t fix that problem. Excessive monetary stimulus will just add inflation to the misery mix (recall the Carter-era “misery index”) and excessive fiscal stimulus will just ensure that not only will the current generation suffer the ill effects of bad tax policy, but so will future ones.

    Trying to address the needs of the U.S. economy right now with monetary policy brings to mind the old saying “When you’ve got a hammer, everything looks like a nail.” Well, the Fed’s got a hammer all right, but the problem isn’t a nail, it’s a screw.

    And frankly, I’ve had it with being screwed. But I’m not deluding myself into thinking Bernanke has a chance in hell of changing things. On the other hand, the current House of Representatives just might have some ideas about bad tax policy (including over-regulation and government corruption.) We’ll see, but don’t hold your breath for relief via monetary policy.

  16. Gravatar of Mark A. Sadowski Mark A. Sadowski
    7. December 2010 at 16:55

    As long as IOER is in effect the hammer the Fed seems to be using so furiously is made of rubber (or more likely feathers).

    With respect to bad tax policy I can give you an earful. It’s been my main research interest hitherto.

    But lately monetary policy excites me more. And I’m considering a number of research projects in that vein.

  17. Gravatar of scott sumner scott sumner
    7. December 2010 at 18:43

    Marcus, And didn’t Janet Yellen used to be president of the SF Fed? I’d guess it’s a rather Keynesian Federal Reserve Bank.

    Morgan, If the Fed is serious about 2% inflation and not one basis point more, then monetary policy has been too expansionary for the vast majority of the past 20 years.

    Cameron, The market reaction seemed to be slightly higher inflation expectations, so I’ll accept that there was a small positive multiplier from this announcement, but you are right, the likely monetary reaction makes any multiplier estimate highly suspect.

    Rod, I have earlier posts pointing out that just because someone has a pneumonia, doesn’t mean they don’t also have a knife wound. The US economy has multiple problems. Inadequate NGDP is one of them. There is no single problem, or single solution.

  18. Gravatar of Rod Everson Rod Everson
    7. December 2010 at 22:48


    Yes, we have inadequate NGDP, but if the main cause is due to lousy governance, rather than tight money, as I believe to be the case, then trying to inflate our way out of the problem just adds to the misery index.

    Just because someone is sick, doesn’t mean they have either pneumonia or a knife wound. The point is that a proper diagnosis has to be conducted before a treatment can be prescribed. I’m extremely skeptical of both the diagnosis of monetary tightness and the prescription of QE, QE2, QE3, etc.

    Bernanke is running Fed policy exactly as he feels he must at the zero bound. Unfortunately, he’s wrong about what can be done at the zero bound so he’s choosing the ridiculous QE route and trying to influence economic activity in an extremely inefficient way.

    It’s ironic that Bernanke is on record saying he could raise interest rates in 15 minutes if he had to, but doesn’t realize that that is exactly what he must do to regain control of the money supply. He’s bound by his own theory and we’re screwed until he changes it, or until he’s replaced by someone with a better theory.

  19. Gravatar of Rod Everson Rod Everson
    7. December 2010 at 23:06


    I’d agree with you that IOER is an issue at higher interest rates. At near-zero, under present Fed procedures, removing the IOER would have the immediate effect of dropping fed funds to zero percent and excess reserves still wouldn’t circulate due to the absence of an opportunity cost.

    IOER is a mistake, and letting fed funds trade at either the IOER rate or very close to zero percent is a mistake. Either action takes the Fed out of the money supply management game and we end up with actions like Quantitative Easing, actions never proven to work because we’ve never tried them before. But what the hell, it’s just a multi-trillion dollar economy populated by 300 million people that they’re experimenting on. Why should anyone object?

  20. Gravatar of jj jj
    8. December 2010 at 08:37

    How are you defining “inadequate GDP”? I think the most useful definition is that GDP is lower than the market was expecting when wages and contracts were set. Under that definition it doesn’t matter at all what the main cause is — monetary policy, lousy governance, ‘recalculation’, or an asteroid destroying the east coast. Deviating from expected NGDP causes purely unnecessary and unproductive pain, above and beyond the real pain caused by lousy governance or the asteroid.

  21. Gravatar of jj jj
    8. December 2010 at 08:38

    Scott, what do you think of what I wrote above? What’s the ideal monetary policy to handle immense natural disasters?

  22. Gravatar of Rod Everson Rod Everson
    8. December 2010 at 12:28


    All I meant by “inadequate NGDP” is that obviously the economy is running well below capacity now. 10% unemployment tells you that. I’m not trying to create a new statistical measure here. It’s just common sense that GDP could, and should, be higher today than it is.

    If that were due to an external shock like an asteroid, it would be acceptable, but it’s much more likely that the people we’ve chosen to govern us have, over time, driven the economy into the tank. That’s unfortunate and should be unacceptable. But most of them are still there running the show.

    And to blame all of it on poor monetary policy, or worse, to claim that any variant of monetary policy is capable of correcting the problem is just grasping at straws.

    As for deviating from “expected NGDP” it always has and always will. I’d prefer deviations upward, caused by real growth rather than inflation, rather than the other options. Right now we primarily have a deviation downward caused by bad governance, i.e., crappy economic policy out of DC.

  23. Gravatar of Jj Jj
    8. December 2010 at 20:24

    Agreed on the common sense definition of inadequate GDP. But I disagree that monetary policy can’t fix it. Monetary policy should bring back full employment, and GDP back to trend. That won’t make us artificially richer, either — it will just allow the economy to run at full capacity, whatever that might be. The quality of government is one factor that determines what we actually get at full capacity.
    In other words,
    good govt + good monetary = 3 BMW
    Bad govt + good monetary = 3 Lada
    Good govt + bad monetary = 2 BMW
    Bad govt + bad monetary = 2 Lada

  24. Gravatar of scott sumner scott sumner
    9. December 2010 at 07:12

    Rod, Lousy governance reduces RGDP. Only tight money can reduce NGDP. Zimbabwe had lousy governance; low RGDP and high NGDP.

    jj, I prefer NGDP targeting even if there is a large supply shock. In theory, wage targeting is slightly better, so if the shock was large enough I suppose NGDP would be less effective than wage targeting.

  25. Gravatar of Rod Everson Rod Everson
    9. December 2010 at 09:25


    Take the hypothetical case where monetary policy is run perfectly and the price level remains absolutely stable.

    Then the economy receives a shock from a stupid tax policy and over a year real GDP drops significantly.

    Why on earth would you then look to monetary policy to return real GDP to normal?

    All I’m maintaining is that it’s possible that the current situation, given the idiots in charge in DC today.

    (Note as evidence that the House is actually willing to raise taxes on everyone in spite of their governance being soundly rejected on Nov 2nd, and in spite of the fact that President Obama seems to finally have learned some economics, or maybe just who to listen to next…)

    It’s also possible that this recession is due entirely to poor monetary policy, but not, in my opinion, very likely. Elections have consequences and in 2008 we voted to reduce our standard of living in the U.S. Not that the Republicans in charge before that would have done much better, I’ll concede.

  26. Gravatar of Rod Everson Rod Everson
    9. December 2010 at 09:30


    I really don’t understand your statement that only tight money can reduce Nominal GDP.

    Price level at 100 throughout. Real GDP at 1,000. Nominal GDP at 1,000.

    Economy suffers huge shock, internal or external and real GDP drops to 900, but Fed maintains stable prices so price level stays at 100. Nominal and real GDP both fall to 900.

    Please explain where I’m wrong. Maybe NGDP doesn’t mean Nominal GDP? I’ve been assuming that it does.

  27. Gravatar of Doc Merlin Doc Merlin
    9. December 2010 at 10:15

    @Rod Everson:

    Scott DEFINES “tight money” as “reduction in NGDP” then he uses that circularly to try to combat anyone that says that money is/was too loose.

  28. Gravatar of Rod Everson Rod Everson
    9. December 2010 at 13:12

    Is Doc Wilson correct that you define tight money as a reduction in NGDP, Scott, or is he misstating your actual position? It does seem consistent with what you wrote and would explain things.

  29. Gravatar of jj jj
    9. December 2010 at 13:36

    You asked: “Why on earth would you then look to monetary policy to return real GDP to normal?”
    I don’t: I do not think monetary policy is capable of returning real GDP to normal, if the cause of the drop in GDP is not monetary. I don’t think the fed can conjure up real GDP growth out of thin air. What they CAN do is conjure up nominal GDP growth out of thin air.

    The government can and does affect real GDP, for better or usually worse. However, allowing this to also drag down Nominal GDP is just biting your nose off to spite your face — it amplifies the original problem.
    If the Fed were to force NGDP to follow a certain trend, it would not paper over any real problems in the economy — it would just prevent real problems in one part of the economy from dragging unrelated other parts of the economy down through monetary effects.
    Falling NGDP hurts a lot of sectors of the economy, far beyond the real negative effects that the government is causing. Scott has estimated that of the 8% drop in NGDP in 2008, 1/4 (2%) was due to real effects, and 6% due to bad monetary policy. Under an NGDP level-targeting regime, NGDP would have continued to grow at trend, but real GDP would have fallen 2% (how I’m not sure — extra inflation?). Anyways the idea is that you can’t avoid the real effects but you don’t need nominal effects creating further real pain.

  30. Gravatar of Rod Everson Rod Everson
    9. December 2010 at 19:47

    jj, Nominal GDP did not fall 8% in 2008. From the 2nd quarter high in 2008 to the 1st quarter low in 2009 it fell about $350bn or a little over 2 percent.

    Real GDP fell almost 4% over approximately that same time period as the housing market contracted significantly.

    I haven’t got a clue what you’re talking about, but the last thing I want the Fed to do when the economy is suffering a real GDP contraction unrelated to Fed policy is to have them then try to maintain Nominal GDP. Why? Because then we’d have both falling real GDP AND inflation. That’s how you drive the misery index higher. Carter presided over such a time and he’s not exactly considered the best President we ever had.

  31. Gravatar of Mark A. Sadowski Mark A. Sadowski
    9. December 2010 at 23:09

    Rod wrote:
    “Because then we’d have both falling real GDP AND inflation. That’s how you drive the misery index higher.”

    Except we’ve had falling real GDP and barely any inflation. In that past two years we’ve seen RGDP decline and price levels barely rise. What is your point?

    This represents a major NGDP/AD crisis. Estimate the SRAS slope and tell me otherwise.

    If you can still make a coherent argument in favor of stagflation that I commend you. If not prepare to be taken apart limb from limb.

  32. Gravatar of Mark A. Sadowski Mark A. Sadowski
    9. December 2010 at 23:12

    “that” should read “than” obviously

  33. Gravatar of jj jj
    10. December 2010 at 06:34

    Sorry about the 8% figure, it was off the top of my head. I should have said “8% off trend”. And I don’t remember which quarter-to-quarter period Scott came up with for that. If I can find the data I’ll post it here.

    Regardless — falling GDP and higher inflation is not the worst thing in the world. A depression, or great recession, is worse. I’d gladly take a 2% drop in real GDP, and 2% higher inflation, if it could prevent the 8% drop in real GDP and 10% unemployment we’re getting now.

    The fed under Carter wasn’t practicing NGDP targeting, so it’s not fair to assume we’ll get the same result today.

    Do you think this entire recession can be explained by real causes? Like, everybody who is now unemployed was in an unproductive job?

  34. Gravatar of Rod Everson Rod Everson
    10. December 2010 at 08:05


    I get frustrated with all the moving targets in here. I wasn’t talking about the current situation. If you’ll notice, I preceded the statement that you’re reacting to with this qualifying remark:

    the last thing I want the Fed to do when the economy is suffering a real GDP contraction unrelated to Fed policy is to have them then try to maintain Nominal GDP.

    I’m frankly astounded that anyone can hold the position that the only way NGDP can be reduced is via Fed actions. That’s ridiculous and I was just pointing out one of the logical ramifications of holding that position.

    The present situation might, or might not, be such a case; I wasn’t addressing the present situation, however.

  35. Gravatar of Rod Everson Rod Everson
    10. December 2010 at 08:37


    Yes, I think we have 10% unemployment (almost) and 15% underemployed mostly because of real causes, i.e., bad governance. FNMA and Freddie Mac were the direct result of bad governance. Tax policy that encouraged massive speculation in housing was bad governance. Electing a President and Congress who think banana republics are models of effective governance was bad governance (by the electorate.) Passing on free trade deals ready to be inked was bad governance. Overturning centuries of law and stiffing GM’s bondholders was bad governance. Needless to say, I could go on.

    If you honestly think the Fed is responsible for 10% unemployment, or could do anything about it beyond making the unemployed even more miserable by raising the price of everything they buy, then you have to explain to me how the sort of governance I cite above is supposed to result in economic growth. Good luck with that.

    As for your last comment: “Like, everybody who is now unemployed was in an unproductive job?”, why must they have been in “unproductive” jobs? For example if you tax something and you usually get less of it. If you tax a productive enterprise too highly, it might no longer be productive. So why must everyone unemployed today have been in unproductive jobs? Makes no sense.

    And Mark, I’m not saying the Fed had absolutely nothing to do with this recession. I’m just pointing out that if they did, they had a bit part to play, not the leading role. Now, if we’re still in it two years from now, I suspect they’ll very much have the leading role in causing that.

  36. Gravatar of jj jj
    10. December 2010 at 14:02

    Rod, first reply:
    Nor was I necessarily describing the current situation. I think that in a hypothetical situation, a 2% fall in real GDP unrelated to fed policy will turn into a much larger drop in real GDP if the fed doesn’t respond appropriately.

    It’s almost a philosophical distinction to ask if only the Fed can reduce NGDP.

    NGDP = RGDP * [monetary policy]

    Obviously either the real economy or the fed can act on NGDP, or counteract the movement of the other. This blog suggests that the most efficient policy, to enable fastest RGDP growth, is for the fed to exactly counteract RGDP such that NGDP grows along a particular path.
    Certainly under today’s regime, factors other than the Fed could reduce NGDP. But the whole argument is that they should not; that nothing is gained by allowing NGDP to fall; and that the Fed is capable of preventing this.

  37. Gravatar of jj jj
    10. December 2010 at 14:13

    Rod, second reply:
    I agree with you on everything the government did wrong. That should reduce real wealth, and make everybody less productive. But would we not be better off if, under the exact same government, everybody was working? Before I waste a lot of words, would you agree up front that even with our bad government there might be a way for everybody who is employed today to remain just as productively employed, while also creating jobs for the unemployed?
    Right now the unemployed are doing nothing. Yet they aren’t starving to death, so they’re consuming resources. As a bare minimum they could do something (like my laundry), no matter what the taxes on them are, and the economy would be better off.
    This isn’t a great or even good reply but I have to rush off; I’ll pop back in later and hope to continue this discussion.

  38. Gravatar of Rod Everson Rod Everson
    11. December 2010 at 08:14


    Taking your second reply first: Be very careful when you talk about “creating jobs.” Who is going to do the creating? We have hundreds of thousands of “green jobs” right now that only exist because our wise legislators decided to create them with subsidies and favorable tax policies. What if, after all is said and done, those green jobs actually burn more resources than they generate? Are we collectively really better off?

    So, no, I don’t agree with you if you are saying that government should set about specifically creating jobs. Instead, government should pass sensible tax policies and stick to what doing what individuals on their own can’t do (on their own or collectively.) Build roads, maintain public utilities, defend the country, etc. What you will get with a philosophy of “putting virtually everyone to work” is an even more collectivist government.

    In “The Road to Serfdom” Hayek stated that Great Britain was too wealthy a country (even by then, in the early 19040’s) to allow its less fortunate citizens to go hungry, without shelter or sufficient clothing. Yet he wrote the most anti-collectivist book I’ve ever read (besides Orwell’s “1984”.) His statement on providing for the poor focused the matter for me better than anything I’ve ever read. We are rich enough to provide for the less fortunate among us, no matter how conservative our leanings.

    The trick is to set the limits of such government support prudently so that productivity doesn’t go into the ash can. 99 weeks of unemployment (and rising soon) is probably too much, and many of the other limits we’ve set are probably unnecessarily generous also.

  39. Gravatar of Rod Everson Rod Everson
    11. December 2010 at 08:33

    jj, regarding your first reply on NGDP:

    With respect, your statement “It’s almost a philosophical distinction to ask if only the Fed can reduce NGDP,” is sheer nonsense. There’s no a “philosophical distinction” involved. It’s a simple question. Can only the Fed reduce NGDP? And the answer is obviously no. If California fell into the sea tomorrow, both Nominal GDP and Real GDP would plunge regardless of the Fed’s actions.

    The real “philosophical distinction” arises because some people here, Scott and possibly you included, seem to think the Fed should always maintain a targeted NGDP, without acknowledging the absolute fact that this is simply saying that the Fed should add inflation to our woes whenever bad governance or an external shock cause a fall in Real GDP.

    Now, I understand the logic of the Fed maintaining a targeted NGDP. Essentially, it’s aimed at stopping the Fed from screwing up and generating a recession on its own, as has happened more times that they would ever admit, incidentally. And I think that would work most of the time, especially since I believe that most of the recessions of the last half of the 20th century were caused by tight money.

    But what if the external shock or governance were positive? What if, by some miracle, our legislators and our good fortunes combined to create a decade of 6% real GDP growth? If the Fed were following a NGDP formula, would they then slow money growth to stay on target, thereby lowering the price level? And would that slowing end up functioning like it always has in the past and also slow the economy? Would Reagan’s economy have sputtered in the late 80’s, diminishing the power of his ideas on taxation and governance? As I said elsewhere, operating outside of a policy of maintaining stable pricing gives the Fed the power to determine the fate of presidents, and presidential candidates too, for that matter. No single unelected government agent, i.e., no Bernanke, should have that power. Seriously.

    This is why I prefer a Fed policy of maintaining a steady price level. I really don’t want either inflation or deflation to be foisted on me by the Fed, and I firmly believe that no Fed chairman should have the latitude that has been granted to the current one.

  40. Gravatar of Doc Merlin Doc Merlin
    11. December 2010 at 10:37

    “Do you think this entire recession can be explained by real causes? Like, everybody who is now unemployed was in an unproductive job?”

    Thats not what “real causes” means. A massive natural disaster can be a real cause. A change in government policy can be a real cause.

  41. Gravatar of scott sumner scott sumner
    11. December 2010 at 15:11

    Rod, I define “tight money” as a fall in NGDP.

    A fall in M*V is tight money in my book, because the Fed should offset any fall in V with more M.

    Regarding the fall in NGDP, I recall it was 3% from 2008:2 to 2009:2. Or maybe 2008:3 to 2009:3. That’s more than 8% below trend.

  42. Gravatar of Rod Everson Rod Everson
    11. December 2010 at 15:52


    Well, if that’s your definition it doesn’t seem to allow for money to be “tight” when Nominal GDP is rising. Guess I would have a problem with that. A Fed-induced, tight money, recession can start and end with nominal GDP never falling, for that matter, but possibly you’ll amend your definition to be “falling below trend?”

    Also, I can easily perceive of a situation where nominal GDP is falling due to an external event (along with real GDP, of course) where the Fed was actually being quite easy, just not easy enough to hold nominal GDP steady (or I guess growing steadily in your formulation.) The only way they could restore nominal GDP to trend in that case would be to cause a generalized inflation.

    I don’t know why you want that, but I certainly don’t. As I said before, all you’d be doing is increasing the “misery index.”

    Tight money cannot be rationally defined as “a fall in NGDP.” It makes no sense. Your definition essentially puts the blame for every recession (and some non-recessionary stalls) on the Fed. Certainly you don’t believe that? Or maybe you don’t believe the Fed can determine whether money is “tight” or “easy”?

    Try stepping back from your formula and ask what really makes sense, because if your formula is telling you “tight money”=”fall in NGDP” it’s deceiving you.

  43. Gravatar of Rod Everson Rod Everson
    11. December 2010 at 16:00

    @Doc Merlin

    Guess you were right Doc. Thanks. I don’t get it, but you were obviously right. Got a feel for how this discussion will end? I’m guessing you’ve probably seen the ending before.

    @ Rod Everson

    Scott DEFINES “tight money” as “reduction in NGDP”

  44. Gravatar of ssumner ssumner
    11. December 2010 at 19:20

    Rod, To be more specific, I meant money is tight whenever future NGDP is expected to be below target. Thus there is no absolute “tight” or “easy” money, only easy or tight relative to some policy objective. You’ll have to forgive me because I often cut corners in responses, as it’s hard to fully explain all the nuances in quick replies.

    My definition certainly does not put the blame for all recessions on monetary policy. I wouldn’t blame monetary policy for the 1974 US recession, or the Zimbabwe depression, or the Indonesian depression of 1998, or the Russian depression of the early 1990s, or many other depressions.

    A recession is a fall in RGDP, not NGDP. It can be caused by real or nominal factors. Often it is both. Indeed I think most US recessions are partly monetary and partly real.

  45. Gravatar of Rod Everson Rod Everson
    12. December 2010 at 09:44


    Expected by whom? What target? You’re absolutely right when you say (by your definition) there is no tight or easy money. As I implied before, the “moving target” aspect of your site can be difficult to engage.

    With respect, I’d suggest you no longer take the particular shortcut where you say “tight money = a fall in NGDP” both since it’s incorrect and since you don’t mean it anyway.

    Incidentally, by my reckoning (and I actually have seen data to back up what I say) most U.S. recessions were initiated, and ended, by monetary policy. By the time Congress gets around to addressing a 6 to 9 month contraction caused by tight money (my definition, not yours), we’re usually already coming out of it due to easy money. Typically, we don’t even know we’re in a recession for the first 4 or 5 months after one starts.

    This one is different, however. No economy can stand the combined assault of a housing depression and a radical economic agenda hitting at the same time. The Fed could ease all they want, but they can’t stop what they didn’t cause (mostly anyway–I’d concede that they were too tight prior to the housing crash and in that sense triggered it, though a housing crash was inevitably going to happen anyway.)

    Oh, and my definition of tight money is a slowing of the rate of growth of the money supply. Where I get into trouble with mine is defining what the money supply is, since changes in banking regulations affect how money is held, e.g., the introduction of NOW accounts.

  46. Gravatar of Jj Jj
    13. December 2010 at 16:21

    I maintain that ultimately only the fed can reduce ngdp. No matter what the real economy does, the fed can print up enough money to make ngdp whatever it wants. If California slides into the sea that mint mean trillions of dollars per month, for years, but ngdp could be kept at the same level. It seems like you’d call that a terrible policy, but it would work in nominal terms if not real, nonetheless.
    I called it philosophical distinction because the fed does in fact allow ngdp to fall. This is like asking, does God cause suffering or just allow it to happen? When it comes to fiat money the fed is God– there is no natural amount of money, only what the fed creates.

  47. Gravatar of Jj Jj
    13. December 2010 at 17:01

    Also, I for one acknowledge the fact that targeting ngdp when rgdp falls DOES mean adding the misery of inflation. So why would anyone be in favor? Well, the idea is at you are also SUBTRACTING the misery of inadequate AD, and in the balance you come out ahead.
    A lot of productive jobs are lost due to inadequate AD, and it is just unproductive and unnecessary pain. The inflation created by ngdp targeting would be productive and necessary pain.

  48. Gravatar of scott sumner scott sumner
    13. December 2010 at 18:08

    Rod, You said:

    “With respect, I’d suggest you no longer take the particular shortcut where you say “tight money = a fall in NGDP” both since it’s incorrect and since you don’t mean it anyway.”

    That’s a bit strong. 99% of the time when actual NGDP falls significantly, expected future NGDP also falls below target. But I’ll try to be more careful in the future.

    You said;

    “This one is different, however. No economy can stand the combined assault of a housing depression and a radical economic agenda hitting at the same time. The Fed could ease all they want, but they can’t stop what they didn’t cause (mostly anyway-I’d concede that they were too tight prior to the housing crash and in that sense triggered it, though a housing crash was inevitably going to happen anyway.)”

    The economy survived the housing depression (which occurred between mid-2006 and mid-2008) just fine. The sharp rise in unemployment after mid-2008 was caused by falling NGDP, the housing crash was 80% over by mid-2008.

    I don’t agree that the economy was hit by a radical political agenda. Obama’s health care bill was not that big a deal, and the economy boomed when LBJ was much more activist.

    But let’s say you are right, you have provided an explanation for falling RGDP, not falling NGDP. I’m trying to explain falling NGDP.

    The money supply is a highly unreliable indicator of the stance of money policy. Changes in both the supply and demand for money matter a lot. And the Fed can and does influencing both.

    In my view, the most expansionary monetary policy in US history occurred after March 1933—and yet the money supply was flat. The dollar was sharply devalued.

    jj, I’d just add that the problems usually assumed to be caused by high inflation are actually caused by high NGDP growth.

Leave a Reply