Using the “I told you so” argument

Paul Krugman frequently notes that he has been consistently right in claiming that our fiscal and monetary policies would not lead to high inflation and high interest rates.  I can’t say I blame him (which is not surprising, given that I have made identical predictions.)  At the same time, I feel a bit uneasy about judging the merits of an economic theory based on the forecasting skill of its leading proponents.  I believe markets are relatively efficient, and thus that it is hard for even a brilliant economist to forecast shocks that the markets don’t see.  I am far more impressed by the fact that Irving Fisher’s Phillips curve model explains the Great Contraction of 1929-33, then I am worried by the fact that he failed to predict it before it happened.  But I’m the exception.

I was reminded of this issue when I read a great post by Ryan Avent entitled “What Happened over the Summer?”  Ryan dismisses several theories of the economic slowdown, before settling on the following explanation:

So let me tell you a story. In late April, fears of a serious European debt crisis began to emerge. These fears sparked a mild panic and a renewal in the flight to safety. This flight manifested itself, in part, as a rush to buy American government debt. Treasury yields had been rising in the months prior to the crisis, but plunged from April through the summer. The dollar shot up; the trade-weighted dollar rose nearly 5% from late April to early June. In response to the pressure within markets, the Fed reopened currency swap lines it had used in previous stages of the crisis. It did not, however, take steps to offset the impact of the financial hiccup on growth expectations.

Markets reacted. The Dow fell over 13% from late April to early July, and was still 10% off its April peak in late August. From January to April, 10-year inflation expectations were stable at around 2%. These began falling sharply, and were down to around 1.5% by the end of the summer. Every signal available began flashing a decline in economic expectations starting in late April. But the Fed didn’t act. Not until late August did Ben Bernanke hint at a course change, and matters improved almost immediately. The Dow has risen by nearly 13% since then. Inflation expectations leveled off in October. And the pace of private hiring has returned to early spring levels.

You may not buy this story. We obviously can’t be sure one way or another. It strikes me as fairly compelling, however. And if we do accept it, the story implies that the Fed, by waiting until August to signal a policy change, cost the economy between 100,000 and 200,000 jobs a month for four months.

Of course people could argue that “hindsight is 20-20” and that the Fed had no way of knowing last spring that the euro crisis would end up hurting the US more than Germany.  Unless, of course, they read TheMoneyIllusion.  The following essay was published by the online Economist in August, but note the embedded quotation that was published in my blog back in May:

Back in May and June there was a lot of talk about the bleak outlook for the euro zone. Recall that the problems in Greece, and more broadly all the so-called “PIIGS”, had created doubts about the soundness of banks in France, Germany, and the Netherlands. In late May I made this observation in my blog:

“So stocks in the heart of the eurozone, the area with many banks that are highly exposed to Greek and Spanish debts, are actually down a bit less (on average) than the US. Perhaps the strong dollar is part of the reason. Perhaps monetary policy has become tighter in the US than Europe.”

The loss of confidence in the euro led to a rush for safety, and the demand for dollars rose sharply in the spring of the year. Because the interest rate in America is stuck at 0.25%, and the Fed is reluctant to use unconventional policy tools, there was no policy action taken to offset the increase in the demand for dollars. Monetary policy became effectively tighter.

The results were predictable. Whereas the euro had traded in the range of 1.35 to 1.45 to the dollar in the first four months of 2010, the exchange rate has dropped to the 1.20 to 1.32 range since the beginning of May. Because Germany has an export-based economy, this contributed to a fast rise in output. Just the opposite happened in the US, where a recovery that looked on track in the first quarter of 2010, suddenly stalled in May and June. Some have argued that the winding down of fiscal stimulus caused the recovery to weaken in the US. But spending rose briskly in the second quarter; the problem was a widening of the trade deficit.

Many economists overestimate the importance of real shocks in the business cycle of large diversified economies, and underestimate the importance of monetary shocks.

Let me first clarify one point.  I am not trying to claim to be some sort of Nostradamus.  I simply try to infer market expectations from indicators such as stock, forex and TIPS prices.  Some indicators, such as a strong dollar, can be ambiguous.  It might indicate a strong real economy, or it might indicate a tight money policy (including an increase in money demand.)  In those cases I look at other markets for confirmation.  In the long run I’ll predict no better than markets.  Where there are unexpected shocks (like bubbles bursting) I will fail to predict them.  But consider how much I was able to predict, just buy using market indicators:

1.  In late 2008 I warned that all sorts of market indicators were predicting a severe fall in AD, and that trying to fix banking was merely treating the symptom.  I said much easier money was needed.  We now know that NGDP began plunging two months before the banking crisis of September.

2.  In late 2008 and early 2009 I said fiscal stimulus would not even come close to solving the AD problem, and that monetary stimulus was the key.

3.  On the day after the March 2009 QE I said it would provide a slight boost, but was nowhere near enough to solve the problem.  The economy did do a little bit better after March.

4.  Like Krugman, I said the right-wingers predicting high inflation (because of QE and ultra-low rates) were totally off base, and that inflation would stay very low.

5.  In May 2010 I suggested that the euro crisis might well hurt the US more than Europe.  That was before we had any second quarter GDP data showing the sharp slowdown in the US and the robust growth in Germany.  Ryan Avent is right, the Fed had the market signals it needed to act in the spring and cost hundreds of thousands of jobs by waiting until November (or late August for the verbal easing.)

6.  Back in early March 2009 I suggested that the Fed needed to do three things; QE, lower IOR, and level targeting and/or NGDP targeting.  Recently they began discussing all three options, and adopted QE.  The talk of QE and the other options clearly boosted stocks and foreign exchange prices in September-October, which confirmed my prediction that monetary policy is not impotent at the zero bound (as had been widely assumed in late 2008 and early 2009.)

7.  My prediction that the slowdown that began around May was due to an increase in the demand for dollars implied that monetary easing should reverse the slowdown.  It’s too soon to say for sure, but Avent’s right that the early indications are that the economy picked up a bit in October.  Certainly QE has already moved asset prices in a direction that should (ceteris paribus) lead to more NGDP growth.

This does not mean prosperity is just around the corner.  Markets seem to be indicating that inflation will be a little bit higher, and that the odds of a double dip recession have receded.  But the recovery is still likely to be slow, albeit a bit less slow than forecast in mid-August.

I don’t have any spectacular predictions that other missed.  I’m no Roubini.  But at the same time I think it’s fair to say that if monetary policymakers had paid more attention to market signals after mid-2008, we would have had a more expansionary policy.   And even those who are skeptical of the current QE policy can hardly deny that more AD would have been beneficial in late 2008.

Rather than look for spectacular predictions (something we can never expect from policymakers anyway) it makes more sense to look back at what models were telling us, and whether in retrospect the advice now seems wise.  In 1930 no one knew for sure that easier money was needed.  And in 1966-68 no one knew for sure that tighter money was needed.  Now almost everyone agrees about what went wrong in those two episodes.  I’m impressed by models that would have told us that at the time.  I’m claiming that the target the forecast approach that relies on market forecasts of NGDP indicators would have led to better monetary policy over the past 3 years, indeed the last 100 years.

Paul Krugman occasionally predicts things markets don’t see coming.  When he’s right, he racks up successes that I miss.  Often we have the same prediction.  But like the tortoise in Aesop’s fable, I believe that methodically inferring market forecasts of the effect of monetary policy will win out in the long run.  Indeed I think that markets have recently exposed some weaknesses in his view of the relative potency of fiscal and monetary policy at the zero bound.

There’s still a long way to go, and markets will continue to make mistakes on occasions.  But if we’ve learned anything over the past three years it’s that monetary policymakers ignore market signals at their peril.

PS.  I just noticed a German blog called Kantoos Economics, which seems to also favor stable NGDP growth.  I can’t read German, but the graph is worth 1000 words.

PPS.  I just noticed The Atlantic picked up Avent’s post, perhaps the press will discover my blog someday.  🙂


Tags:

 
 
 

41 Responses to “Using the “I told you so” argument”

  1. Gravatar of Morgan Warstler Morgan Warstler
    9. November 2010 at 15:07

    Why are we told that Japan gets exactly the deflation they want… but we now assume the fed wants something different than what we have.

    “And if we do accept it, the story implies that the Fed, by waiting until August to signal a policy change, cost the economy between 100,000 and 200,000 jobs a month for four months.”

    Ahem, why would we risk taking over the Congress?

    The Fed neutralizes stimulus. It INSTRUCTS the government to start cutting the deficit. Why in the world, would it make it easier for DeKrugman/Obama to do more Fiscal?

    Those 4 months could have changes the ENTIRE narrative, the entire story arc.

    And how long have I been predicting that?

  2. Gravatar of woupiestek woupiestek
    9. November 2010 at 16:18

    I can read German. The entry is a rant criticizing the ECB (EZB in German) for letting the economy fall of a stable growth path.

    The first sentence says: “this entry is inspired by Scott Sumner, who since the beginning of 2009 writes: the Fed is guilty.” Can’t parse “Aber der Reihe nach.”, it’s probably German idiom. The next two paragraphs state that central banks should target expected NGDP.

    “Komplett versagt” means completely failed. He tell us that if the economy strays more then ten percent of the target growth path, the central bank completely fails. And then he show the graph.

    The last three paragraph in summary:
    “The ECB has not done everything it can. Such a great fall in NGDP is a catastrophe that a central bank should never allow.

    The Fed is no better, but can learn. I’m skeptical about the ECB that according to Germany is doing so well. What do you think?”

    I like the careful build up to the graph; the rest of the entry is of lesser quality.

  3. Gravatar of “Eu vi o que vocês fizeram no verão passado” « Historinhas “Eu vi o que vocês fizeram no verão passado” « Historinhas
    9. November 2010 at 17:35

    […] Scott Sumner (aqui) tem um ótimo post elaborando a questão. (Como participei desse processo desde o início, posso […]

  4. Gravatar of JTapp JTapp
    9. November 2010 at 17:36

    You can use Google Translate to read the website you linked to. Above the graph: “What has the ECB done? Completely fails:”

  5. Gravatar of Benjamin Cole Benjamin Cole
    9. November 2010 at 17:39

    The world is noticing Sumner. Just keep honking, keep using language that packs a punch, and stay on point.

    You may also wish to post shorter, and only one a day.

    Not for me; I deeply enjoy your posts, that I consider the foremost economic commentary in America today.

    I am thinking of a broader audience.

  6. Gravatar of JimP JimP
    9. November 2010 at 18:07

    Its interesting what has been going on in the stock market the last three days. First a 200 point up day as the QE2 news is digested. And now two pretty weak days. I think the rise of the deflationista rants against Bernanke have played a roll. Larry Kudlow and various others are promoting the idea of political pressure to force the Fed to stop the program. And Rand Paul, of course, wants to close the Fed and go back to gold. And he will apparently be the head of Fed supervision in the congress. The crazed deflationistas are really hard at work.

    I don’t think Bernanke has the backbone or the ability to resist this pressure all alone. Now really is the time for the President to speak. I have said this before. We need Obama to do a Roosevelt. If Obama is willing to debate the Republicans on the issue of NGDP growth vs deflation I think he can win and perhaps get himself re-elected. If he doesn’t, both he and we are in trouble. Clearly there is a segment of opinion here that actively desires the Japanese outcome. Will we get it?

  7. Gravatar of Niklas Blanchard Niklas Blanchard
    9. November 2010 at 18:08

    Sometimes I think that I should stop blogging and just link to your posts, as I often share the same opinion about the subject you happen to be writing about.

    I’m often very frustrated that the yards stick in which people judge science is its predictive power. Similarly, many people dismiss economics as not a “real science” because of its presumed lack of specific predictive power. But in reality, science doesn’t predict anything, it just described a plausible process such that if you find the conditions, YOU can reliably predict the result.

    That is very useful, as running around being mystified (or mortified) by natural phenomena constantly is not a useful way to spend your time.

  8. Gravatar of Morgan Warstler Morgan Warstler
    9. November 2010 at 18:20

    “I don’t think Bernanke has the backbone or the ability to resist this pressure all alone.”

    JimP, I’ve been saying this, I’ve been warning this…

    The only way Ben / Scott get to do a formal QE program is if they go out and do Full Bore, Double Barrel, Show ’em Their Panties Austerity Cheers…

    Show Kudlow, show Palin, show Rush, show everyone that they are to be trusted…. the message is: Obama screwed up! And when this Austerity Program (plus QE) takes off – we all know who delivered.

    “the enemy of my enemy is my friend.”

    Right now, they aren’t flying colors, they need to hoist that red flag high.

  9. Gravatar of scott sumner scott sumner
    9. November 2010 at 18:52

    Morgan, You said;

    “Why are we told that Japan gets exactly the deflation they want… but we now assume the fed wants something different than what we have.”

    Right now both countries have a bit less inflation than they’d like; over the past 20 years both countries got roughly the inflation they wanted. Where is the inconsistency?

    Thanks Woupiestek and JTapp.

    Benjamin, Yes, I keep telling myself to do shorter posts.

    JimP. You might be right. BTW, I vaguely recall Obama defending the Fed overseas. I guess that’s more politically acceptable, as if there’s one thing American’s agree on, it’s that we don’t want foreigners telling us how to run our monetary policy. 🙂

    Niklas, Maybe I should stop blogging and turn it over to you and Beckworth and Woolsey. I am always short of time these days.

    Morgan, Too bad you don’t agree with me. Together we could write posts with a lot of punch–we’d blow “DeKrugman” out of the water.

  10. Gravatar of Greg Ransom Greg Ransom
    9. November 2010 at 20:33

    I call B.S.

    You’ve not felt uneasy in the least doing this with Hayek’s macroeconomics — even when reminded that Hayek in the 20s and 30s was working with much, much worse data than exists today, and spent almost zero time studying even that data after 1930 …

    Scott wrote,

    “I feel a bit uneasy about judging the merits of an economic theory based on the forecasting skill of its leading proponents.”

  11. Gravatar of Mark A. Sadowski Mark A. Sadowski
    9. November 2010 at 21:04

    I’m not distressed by Irving Fisher’s failure to predict. The job of economists is not to predict but to propose. Public policy is the key. No one can predict with absolute accuracy.

  12. Gravatar of Full Employment Hawk Full Employment Hawk
    9. November 2010 at 22:49

    “In late 2008 and early 2009 I said fiscal stimulus would not even come close to solving the AD problem …”

    Once again, you and Krugman agree, although for different reasons. Krugman argued that the stimulus was too small to do the job.

  13. Gravatar of Full Employment Hawk Full Employment Hawk
    9. November 2010 at 23:04

    “Many economists overestimate the importance of real shocks in the business cycle of large diversified economies, and underestimate the importance of monetary shocks.”

    Right On! Real Business Cycle theory’s impact on Macroeconomics has been extremely negative. And what has been happening since 2008 is refuting it.

    With the exception of genuine supply shocks, like the petroleum shocks in the 1970s, which are relatively rare and easily recognized, the PROXIMATE cause of business cycles is fluctuations in aggregate demand. And these are primarily caused by shocks from the financial system. But the financial shocks do not only include pure monetary shocks, but also such shocks as collapses of financial bubbles and financial crises.

  14. Gravatar of Joe Joe
    10. November 2010 at 00:17

    Use google translate to read it. It works quite well.

  15. Gravatar of Doc Merlin Doc Merlin
    10. November 2010 at 03:34

    @Full Employment Hawks
    ‘Right On! Real Business Cycle theory’s impact on Macroeconomics has been extremely negative. And what has been happening since 2008 is refuting it.

    With the exception of genuine supply shocks, like the petroleum shocks in the 1970s, which are relatively rare and easily recognized’

    You seem to be missing all the real adverse supply shocks that have been happening the past two years.
    1. Lower than expected yields on almost every crop,
    2. massive fires in Russia destroying their wheat harvest,
    3. Russia temporarily shutting down the has pipline to europe,
    4. China closing off its rare earth mineral supply
    5. Increases in tariffs and protectionism in the US
    6. Increased riots and work stoppages in europe.
    7. Wheat rust spreading
    8. ~50% increase in minimum wage in the US
    9. The EPA shutting off the water to farms in California
    10. The Oil spill in the gulf
    11. The federal government’s drilling moratorium in the gulf following the spill
    12. Russian drought has resulted in russia shutting down all exports of grain
    13. Cotton crop is so low that
    ‘Mike Stevens, a veteran Louisiana cotton broker, said: “The report is the most bullish in our lifetime.”’ Note, bullish means rising prices and that can be very bad in commodities. Also, note that this comes at a time with low aggregate demand, so it should underscore the severity.
    http://www.ft.com/cms/s/0/249211fc-ec1d-11df-9e11-00144feab49a.html

    14. The US federal government increased reporting requirements for businesses (added costs).
    15. Venezuela is destroying their infrastructure.

    I could keep doing this all, day, as supply shocks have been very severe and adverse this past two years.

  16. Gravatar of W le B W le B
    10. November 2010 at 04:39

    History shows that when something becomes a target it ceases to function as a reliable indicator. Have you begun to think what would need to happen when once the forecast becomes the target?
    You say that market signals mid 2008 were showing amber to red, but most of the cutest business people I know were actually responding to conditions they detected in mid 2007 or even before. Perhaps their profit taking, around that time, which expressed itself as deleveraging, is an even earlier warning sign.
    I realise this doesn’t fit in with your EMHism but perhaps in less liquid markets it is possible to spot the early movers and movements.

  17. Gravatar of Full Employment Hawk Full Employment Hawk
    10. November 2010 at 06:23

    “by waiting until August to signal a policy change, cost the economy between 100,000 and 200,000 jobs a month for four months”

    This also cost the Democrats a significant number of seats in this election. With the high unemployment rate, the Democrats would still have lost a lot of seats, but if the Obama administration had had the Recovery Summer it had counted on, the losses would have been less severe.

    The Obama administration in leaving two vacancies on the BOG vacant for an extended period of time instead of proptly filling them with full employment hawks made a major blunder.

  18. Gravatar of Full Employment Hawk Full Employment Hawk
    10. November 2010 at 06:30

    “You seem to be missing all the real adverse supply shocks that have been happening the past two years.”

    These are minor supply shocks that have only a very small effect on the economy. None of these have reversed the disinflation that the U.S. economy is undergoing. Only severe supply shocks, like, FOR EXAMPLE, the petroleum shocks of the 1970s are a significant cause of business cycles.

  19. Gravatar of Full Employment Hawk Full Employment Hawk
    10. November 2010 at 06:33

    “11. The federal government’s drilling moratorium in the gulf following the spill”

    The moratorium was only on deep well drilling, the rest of the drilling continued. While this may have had some effect on the local economy, the effect on the U.S. economy was highly insignificant.

  20. Gravatar of Full Employment Hawk Full Employment Hawk
    10. November 2010 at 06:36

    “I’m not distressed by Irving Fisher’s failure to predict.”

    Irving Fisher’s insistence that the ecomomy was in good shape and that the high stock prices were justified to a large degree discredited him at the time.

  21. Gravatar of Full Employment Hawk Full Employment Hawk
    10. November 2010 at 06:43

    “We need Obama to do a Roosevelt. If Obama is willing to debate the Republicans on the issue of NGDP growth vs deflation I think he can win and perhaps get himself re-elected.”

    If the unemployment rate has not come down significantly by 2012 Obama does not have a snowball’s chance in hell of being reelected.

    If I could tell Obama just one thing, I would tell him

    “It’s the jobs, Mr. President.

    Quantitative easing has also been getting a lot of criticism from the left, for example, the Huffington Post. Now that Sarah Palin has come out against it, this criticism will be largely muted.

  22. Gravatar of Morgan Warstler Morgan Warstler
    10. November 2010 at 07:18

    Biggest shocks: Obamacare, stimulus, efforts at Cap ‘N Trade, Wall Street reform

  23. Gravatar of JimP JimP
    10. November 2010 at 07:29

    Martin Wolf defends the Fed – but is critical because they should have done more.

    http://www.ft.com/cms/s/0/93c4e11e-ec39-11df-9e11-00144feab49a.html#axzz14y4TfwBh

  24. Gravatar of MRPO MRPO
    10. November 2010 at 09:35

    “Like Krugman, I said the right-wingers predicting high inflation (because of QE and ultra-low rates) were totally off base, and that inflation would stay very low.”

    That’s nice, except regular old conservatives, Republicans and folk from elsewhere around the political spectrum have also been very concerned about the risk of inflation. Hardly just “right wingers.” Little things like this always cause me to fear there’s an a least periodically blinkered ideologue hiding behind a mask of elegant writing.

  25. Gravatar of Jeff Jeff
    10. November 2010 at 11:49

    @JimP

    You said:

    I don’t think Bernanke has the backbone or the ability to resist this pressure all alone. Now really is the time for the President to speak. I have said this before. We need Obama to do a Roosevelt. If Obama is willing to debate the Republicans on the issue of NGDP growth vs deflation I think he can win and perhaps get himself re-elected.

    If Obama comes out strongly for QE2, he is in effect admitting that it should have been done long ago. Why wasn’t it? If people start asking themselves that, they’ll quickly notice that

    1. Obama nominated Bernanke for reappointment all the way back in August of last year, and thus has to take responsibility for whatever Bernanke did or didn’t do since then, and

    2. Obama failed to nominate anyone to the open seats on the Board of Governors until very late in the game. And even then, he didn’t push their nominations very hard, evidently because he wanted to focus on ObamaCare.

    Furthermore, an admission that expansionary monetary policy is the right medicine also makes Obama’s stimulus bill look like a huge waste of resources.

    Finally, I think it should be noted that Bernanke’s bailout of Goldman Sachs, er, AIG, and his endorsement of the “let’s scare the hell out of everyone so we can get TARP passed” strategy has greatly damaged the Fed’s reputation among the Tea Party people. I know at least one spouse of a Fed employee who told me that she used to proud that her husband worked at the Fed, but now she never mentions it to anyone for fear she’ll be called on to defend the indefensible.

  26. Gravatar of Joe Joe
    10. November 2010 at 15:03

    “I am far more impressed by the fact that Irving Fisher’s Phillips curve model explains the Great Contraction of 1929-33, then I am worried by the fact that he failed to predict it before it happened. But I’m the exception.”

    As long as you only feel “somewhat” uneasy, I’m with you. After all, crises may be tipped off by very odd things, and no model should be thrown overboard because its proponents didn’t pay attention or fully understand something – that’s nothing more than inputting imperfect information, after all.

    However, “I was right” is and should be very important, especially on (relatively) simple things like “will we have significant inflation?”

    There are, to a rough approximation, infinity potential models of the economy. And economics is very politically fractured. Therefore, you can almost always create a plausible explanation of what happened in hindsight – at least in the eyes of your partisans.

    You can only get rid of that bias by asking whether your model works, prospectively, in the real world. (If your inputs don’t involve significant judgment calls, you can retrodict to a point after the model was invented and get the same benefits, so long as you don’t cherry pick events – this could be good for data purposes).

  27. Gravatar of Full Employment Hawk Full Employment Hawk
    10. November 2010 at 17:09

    “Furthermore, an admission that expansionary monetary policy is the right medicine also makes Obama’s stimulus bill look like a huge waste of resources.”

    Not so: A very strong case can be made that with the economy in the worst shape since the Great Depression, both monetary and fiscal policy are needed to get the economy out of the ditch. That, for example, is my position: For your run of the mill recession monetary policy along with the automatic stablizers are sufficient, but this is a lot worse and you need to go forward with both engines at full throttle.

    Therefore supporting expansionary monetary policy is in no way an admission that the stimulus was a huge waste of resources.

  28. Gravatar of Jeff Jeff
    10. November 2010 at 19:39

    @FullEmploymentHawk,

    That argument only makes sense if you have some empirical evidence that fiscal policy does, in fact, work. But there has never been any evidence that running bigger budget deficits creates productive employment. The failure of the stimulus bill to keep unemployment below the promised 8 percent is just one more flop in a long string of fiscal policy fiascos.

    We know for sure, however, that stimulus bills ARE enormously costly. At best, you can say that Obama tried something he knew would be extremely expensive, and he did so having no good reason to think that it would work. Do you really expect him to admit now that a much cheaper and more effective monetary policy alternative was available?

    Even people who think it is possible to write a stimulus bill that can work don’t really defend the one passed last year. Obama himself admits that there were no “shovel-ready” projects in there, and the only jobs the administration can claim were saved by it were state and local government jobs. When most citizens think the government is already too big, that’s not much of a defense.

    There’s no way around it: the economic policies of both this administration and the one before it have been incredibly bad. I almost said “stupid”, but the people making these decisions are anything but. Rather it’s the people who think the politicians have their best interests at heart who are in need of remedial education.

  29. Gravatar of Full Employment Hawk Full Employment Hawk
    10. November 2010 at 20:50

    “The failure of the stimulus bill to keep unemployment below the promised 8 percent is just one more flop in a long string of fiscal policy fiascos.”

    This argument is pure sophistry. It ignores the fact that initially the Obama adminstration failed to realize how bad the economy was when they took over and this forecast was based on that underestimate of the size of the problem.

    By the time the stimulus passed, the unemployment rate was already above 8%.

    Keynesian advocates of the stimulus, like Paul Krugman, argued that the stimulus was too small to do the job before it even passed, and the failure of the unemployment rate to come down can just as easily be interpreted as a vindication of their position as an indication of the position that it cannot work.

    When the Obama administration took over, output was falling at a rate of 6%. By the middle of the year, the economy had stabilized and started growing. This can be interpreted as showing that the stimulus, while too small to complete the job did give output a strong upward boost. And the research of reputable economic forecasting companies, like IHS Global Advisors and Moody’s Economy shows that it did.

    There is a solid body of theory and empirical evidence that when an economy is depressed and the central bank is not trying to offset it, expansionary fiscal policy does work.
    But strong ideological blinders can keep people committed to laissez-faire from considering it.

    The most serious problem with macroeconomics is that ideologically based dogma plays much too much of a role in theoretical modeling, interpretation of empirical results, and especially policy making.

    I agree with Scott Sumner on monetary policy, which is why I am on this site, but I agree with Paul Krugman about fiscal policy.

  30. Gravatar of scott sumner scott sumner
    11. November 2010 at 05:48

    Greg, I never once criticized Hayek based on his forecasting skill. Not once. Your argument seemed to be that he didn’t know NGDP had fallen sharply. I found that implausible, as one didn’t need detailed research to understand that world prices and output were falling sharply in the early 1930s. It was headline news.

    Mark, Yes, I agree about the role of economists.

    Full Employment hawk, I agree about Krugman, but I also think the economy underperformed the Keynesian model’s prediction, even given the inadequate stimulus. Especially in 2010 Q2 and Q3.

    We agree about the RBC view.

    Joe, Thanks.

    Doc Merlin, Those sorts of shocks occur all the time. I do agree that the minimum wage has modestly increased youth unemployment–the others don’t have significant impact. Remember, supply shocks are inflationary.

    W le B.

    1. I don’t worry about that issue, because it is NGDP forecasts that I am interested in stabilizing. That will help stabilize the labor market. In a sense, I care more about NGDP forecasts than I do NGDP itself. It is forecasts that influence wage contracts, investment decisions, etc.

    2. The Lucas critique predicts that some relationships may break down with adoption of a new target, but the EMH says forecast targeting should not produce this problem. And recall that Lucas is a ratex guy, so he could hardly argue his critique applied to forecast targeting.

    FEH, You said:

    “The Obama administration in leaving two vacancies on the BOG vacant for an extended period of time instead of promptly filling them with full employment hawks made a major blunder.”

    Yes, and I am pretty sure I pointed this out before his supporters like Yglesias and DeLong started mentioning the issue.

    Yes, Irving Fisher was discredited at the time, but for the wrong reasons. Indeed his idea of raising the price of gold was arguably the only clearly effective macro policy during the entire 1930s. Had his policy been adopted in 1929, the Depression and WWII never would have happened.

    Morgan, Again, not big enough factors to explain high unemployment.

    JimP, That’s a very good article.

    MRPO, You said;

    “That’s nice, except regular old conservatives, Republicans and folk from elsewhere around the political spectrum have also been very concerned about the risk of inflation. Hardly just “right wingers.” Little things like this always cause me to fear there’s an a least periodically blinkered ideologue hiding behind a mask of elegant writing.”

    You wrote this in response to a post where I said both the left and the right were wrong? If I am partisan, what about Krugman or Rush Limbaugh?

    Joe, Yes, they should recognize current problems,like inflation or deflation. But there were no current problems with the economy at the time Fisher made his infamous prediction. Everything looked great.

    FEH, Your point about unemployment rising above 8% by the time stimulus was passed is valid, but I think some stimulus proponents overlook an implication of that argument. It implies that the earlier predictions that unemployment would peak at 8% or 9% were faulty, DESPITE THE FACT THAT THEY WERE MADE WELL AFTER THE MAJOR FINANCIAL SHOCKS. This tells me that models aren’t good at predicting macro trends (which of course we all knew already.) In that case, Obama has no right talking about jobs saved vs. some hypothetical counterfactual growth path, as we really don’t know what that path would have been–especially as the Fed might have been more aggressive w/o fiscal stimulus.

    I’m not saying you are wrong–fiscal stimulus might have helped, but it’s awful hard to find solid evidence.

  31. Gravatar of Doc Merlin Doc Merlin
    11. November 2010 at 11:52

    @Full Employment Hawk

    “These are minor supply shocks that have only a very small effect on the economy. None of these have reversed the disinflation that the U.S. economy is undergoing. Only severe supply shocks, like, FOR EXAMPLE, the petroleum shocks of the 1970s are a significant cause of business cycles.”

    The food shocks are actually very severe. While each individual food type may have a high elasticity, food as a whole is very inelastic elastic. This makes the effect much worse than you would expect.

    Second of all, this recession is very similar in its starting to the 70’s. We also had what amounts to an oil supply shock to the western world. (Yes I know globally it wasn’t an oil supply shock, but the a positive demand shock in china of a good that we don’t produce but demand very highly looks a lot like an adverse supply shock to us.)

  32. Gravatar of Doc Merlin Doc Merlin
    11. November 2010 at 11:58

    ‘Doc Merlin, Those sorts of shocks occur all the time. I do agree that the minimum wage has modestly increased youth unemployment-the others don’t have significant impact. Remember, supply shocks are inflationary.’

    They are only inflationary if the fed isn’t targeting inflation. If it is targeting inflation (which ours was) then they cause an NGDP drop.

  33. Gravatar of Jeff Jeff
    11. November 2010 at 16:07

    @FEH,
    The

    research of reputable economic forecasting companies, like IHS Global Advisors and Moody’s Economy

    is like the the Mark Zandi crap Scott has criticized here before: it consists of simulating models that have Keynesian assumptions built in and then pretending that the simulations say something important about the validity of those assumptions. It’s pure nonsense.

    I’ve posted on this before, but it bears repeating. The right way to test a macro theory is to figure out what restrictions it implies on the parameters of a VAR (Vector Auto Regression, an atheoretic statistical model), estimate the VAR with and without those restrictions, and do likelihood ratio tests to see if those restrictions are rejected or not. If they are, you’re done, the theory is wrong.

    However, even if you don’t reject the restrictions implied by your theory, you should go further and estimate something like a Litterman BVAR (Bayesian VAR) that is both atheoretic and known to be a pretty good forecasting technique. If your theory is actually informative about how the real world works, imposing the VAR restrictions it implies on the BVAR should improve it’s out-of-sample forecasts. If it doesn’t, how can you say it really tells you anything worth knowing?

    Of course, people have tried this stuff many times. What they usually find is that some of the Keynesian-style restrictions are not rejected, but imposing them does not improve forecasts. And if you are familiar with how the Keynesian models have been developed over the past 50 years, you’ll recognize that much of that development was data-driven. For example, people noticed that real wages tend to be procyclical, while the older models implied the opposite. So the modelers dropped sticky-wage assumptions and went with sticky-price assumptions instead. That helps them get procylical real wages in their simulations. But should we then pretend that this is evidence their model is correct? I don’t think so. Adapting models to fit stylized facts about the economy doesn’t make them correct, it just makes them less obviously wrong.

  34. Gravatar of scott sumner scott sumner
    11. November 2010 at 18:55

    Doc Merlin, But there is no scenario where they cause a drop in inflation, is there? In any case, they weren’t large enough to cause the huge NGDP drop in late 2008.

  35. Gravatar of Doc Merlin Doc Merlin
    11. November 2010 at 21:54

    @Scott Sumner:
    ‘Doc Merlin, But there is no scenario where they cause a drop in inflation, is there? In any case, they weren’t large enough to cause the huge NGDP drop in late 2008.’

    Supply shocks can cause a fall in NGDP if the shock spurs structural changes or innovation that lower the elasticity of the good. They can also cause NGDP drops if highly leveraged companies.

    In our case, the problem wasn’t that supply shocks caused inflation, it was the supply shocks caused huge PPI inflation, which made an inflation targeting fed to raise rates which caused the housing bubble to bust, which then tightened money supply immensely.

    However, I agree they weren’t *directly* for the NGDP drop, inflation targeting was. They were responsible for raising the PPI substantially, and fearing cost-push inflation the fed tighten up, which dropped NGDP and kept CPI low.

  36. Gravatar of Doc Merlin Doc Merlin
    11. November 2010 at 21:55

    THEN after that, the demand is so crushed from the NGDP drops that you get demand driven deflation. But it all starts with centrally planned monetary policy combined with supply shocks.

  37. Gravatar of Mark A. Sadowski Mark A. Sadowski
    11. November 2010 at 22:08

    Doc Merlin,
    I’ve examined the situation using Tabarrok and Cowen’s AS/AD diagram.

    If you place the LRAS curve at 2.7% real growth and move the SRAS and AD curves so that they all intersect at an inflation rate of 3.3% (deflator) you will get the situation that existed in 2006. To get to 2007 (2.1% real growth and 2.9% inflation) the only way possible is by moving the AD curve. Ditto for 2008 (0.4% growth and 2.1% inflation). 2009 (-2.4% real growth and 0.8% inflation) is more complicated as the Tabarrok/Cowen diagram implies that there is combination of a negative AD shock and a positive SRAS shock.

    In short using their diagram and annual data I find absolutely no evidence of a negative supply side shock. Rather, there appears to be a positive supply side shock in 2009. Interestingly labor productivity also soared (and unit labor costs plummeted) in 2009.

    How does anyone who considers themselves rational find evidence of a negaive supply side shock out of all this?

  38. Gravatar of scott sumner scott sumner
    13. November 2010 at 10:09

    Doc Merlin and Mark, There is one limited case where Doc’s argument makes some sense. The high oil prices of mid-2008 did raise headline inflation in the US. I don’t think they directly affect growth very much, but they may have scared the Fed away from cutting rates in June-September, and thus indirectly helped cause the adverse AD shock. So I’ll grant you that–but fundamentally it was an AD problem.

  39. Gravatar of Doc Merlin Doc Merlin
    15. November 2010 at 12:53

    @Scott:

    It wasn’t just oil. Just about every metal was behaving the same way.

  40. Gravatar of ssumner ssumner
    16. November 2010 at 06:03

    Doc, Yes, and again it was increased demand more than low supply–increased demand for commodities doesn’t cause worldwide recessions, it reflects strong worldwide growth.

  41. Gravatar of Let’s see who’s right! | Kajulew Let’s see who’s right! | Kajulew
    18. November 2010 at 19:18

    […] For the Monetarists: Scott Sumner who says that the Fed’s QE2 will lead to slow growth with modest inflation, and that monetary policy is not impotent at the zero bound when looking at market prices.SS […]

Leave a Reply