What CRE bubble?

I am not a believer in bubbles.  That doesn’t mean I don’t think bubbles exist, rather I don’t know of any coherent definition that is useful.  At the same time I gladly admit that (ex post) the bubble theories do seem to match the price movements in housing circa 2004-06, or tech stocks around 1999-2000.  Krugman argued here that bubble theory also explains commercial real estate (CRE) price movements over the past decade.  I do not agree.

First a bit of perspective.  CRE has always been much more volatile than residential housing.  I recall many examples of sharp declines in the prices of big city office buildings.  I would lump commercial real estate in with commodities and equities as three assets that are very pro-cyclical.  This is especially true if the cycle is caused by a demand shock, a sudden and unexpected fall in NGDP.  Indeed I view these asset price movements as a much more important part of the monetary transmission mechanism than the fed funds rate.

When you first glance at the graph in Krugman’s post, the two lines seem very closely correlated.  But on closer inspection there is a crucial difference.  Housing peaked by mid-2006, and declined thereafter.  That is why it is often viewed as a bubble.  High housing prices could not even be sustained in a relative strong economy during 2007.  In contrast, CRE prices peaked in early 2008, and fell only because the economy moved into recession.  There was a modest decline from 1.9 to 1.7 during the mild phase of the recession, and then prices fell sharply as soon as NGDP began declining in the second half of 2008.  This pattern is exactly what you would expect to occur during a period of tight money that caused the sharpest fall in NGDP since 1938.  Of course almost everyone else seems to think money wasn’t tight in 2008.  So what’s left when you have no other explanation for events?  A bubble.

The term ‘bubble’ is really just shorthand for saying; “I don’t know the cause.” 

I don’t know the cause of the housing cycle.  Hence I can call it a bubble.  I do know the cause of the CRE cycle.  Hence I cannot call it a bubble.

Update:  I just noticed that Arnold Kling had a similar post.


Tags:

 
 
 

36 Responses to “What CRE bubble?”

  1. Gravatar of q q
    8. January 2010 at 08:32

    > Krugman argued here that bubble theory also explains commercial real estate (CRE) price movements over the past decade.

    I think you are overstating what Krugman claims in his blog post. Is there a technical and commonly agreed definition of the word ‘bubble’?

    Off the top of my head, I’d think that the changes in housing finance (exotic mortgages, home equity loans, etc) would make the residential mortgage market more like the commercial real estate market.

  2. Gravatar of OGT OGT
    8. January 2010 at 08:39

    As Calculated Risk often points out it is typical for commercial real estate to peak five quarters after residential. Commercial real estate is a concurrent or lagging indicator, while residential is usually a leading indicator.

    Both the development cycle (they rarely build the retail or office space before housing) and lending cycles are slower than residential real estate, however, they are very highly correlated. There is almost never a boom or bust in one without a similar movement in the other, with residential usually in the lead.

    http://www.calculatedriskblog.com/2007/08/lending-and-cre.html

  3. Gravatar of OGT OGT
    8. January 2010 at 08:41

    Personally, I use the Minsky-ite definition of a bubble, any time asset prices move into “Ponzi” finance levels. By which I mean priced over the ability to produce revenue able to finance the debt payments. (Even with NGDP growing at it’s historic averages.)

  4. Gravatar of Mike@pvl Mike@pvl
    8. January 2010 at 09:04

    Q, in fairness to Scott, Krugman has presented this CRE “bubble” a number of times as proof that government played no role in the housing bubble.

  5. Gravatar of rob rob
    8. January 2010 at 09:28

    My definition of a bubble is pretty simple: when The Economist magazine calls it a bubble, it is a bubble. They confidently called the tech bubble a bubble and the housing bubble a bubble. They also wrote that when housing bubbles crash it usually hurts the economy a lot. I don’t see why Roubini gets so much credit.

    In 2007, with nearly every stock market and asset class in the world going up — doesn’t it seem natural that when they fell, they would likely fall together? (Has there been a time in history when stock markets were that correlated?)

  6. Gravatar of Philo Philo
    8. January 2010 at 09:51

    You write: “I don’t know of any coherent definition [of 'bubble'] that is useful.” But then you give a definition: “The term ‘bubble’ is really just shorthand for saying: ‘I don’t know the cause’.” That’s not a useful definition *for economic science*, since it contains an indexical reference to the speaker; but I think it’s quite useful as a cynical observation on much economic punditry.

  7. Gravatar of Tom Dougherty Tom Dougherty
    8. January 2010 at 10:22

    Different schools of thought probably have different ideas about what a bubble is. Some think bubbles come about due to speculation or greed, but the Austrians idea on a bubble is when the Federal Reserve expands credit more than what actual savings allows. The greater the expansion the greater the bubble.

    One way to look at this might be to look at what the Taylor rule says the Federal Funds rate should be and compare it to the actual Federal Funds rate (although, this is probably not ideal for or endorsed by Austrian economists). If the actual rate is below the the Taylor rule then you have the makings of a bubble in the economy. The longer the rate stays below what the Taylor rule says then the bigger the bubble should be.

    There is no way to know in advance where the bubble will appear in the economy. That depends on the where most of the excess credit is going. So how about the definition of a bubble is when the Fed Funds rate is held excessively low (relative to the Taylor rule) for a prolonged period.

  8. Gravatar of David David
    8. January 2010 at 11:21

    I don’t think commercial real estate was a bubble either, because you didn’t see the same kind of speculation or financial shenanigans. Nobody quit their landscaping job to “flip” office buildings. No “buy strip malls with no money down” seminars. It may not be “coherent” but I maintain it’s useful to find bubbles in the zeitgeist instead of in macro statistics.

  9. Gravatar of Mike Sandifer Mike Sandifer
    8. January 2010 at 13:19

    I look at bubbles in a very different way than any economist I’m aware of. Perhaps sans the distractions of sophisticated macro paradigms(and maybe even dogma), I make use of some simple observations as an investor.

    First, asset prices can only grow so quickly. Vertical increases aren’t possible in a finite universe.

    Second, the higher the growth rate, the shorter the period of sustainability.

    Third, the gap between the price near the vertical and that near trend, holding supply fundamentals constant, tells you much about the severity of the burst.

    I bet the reasons for the above are pretty obvious, but I’ll list mine if requested.

    So, contrary to the assumptions of everyone I’m familiar with, bubbles can often be spotted, the period of busts can be predicted with precision(increasingly vertical, contracting temporal sustainability), and the magnitude of the corrections can also be predicted with useful precision.

  10. Gravatar of Mike Sandifer Mike Sandifer
    8. January 2010 at 13:21

    I should mention that external shocks can always much things up, to cover the obvious.

  11. Gravatar of Simon K Simon K
    8. January 2010 at 14:24

    OGT – the trouble with Minsky’s ponzi finance definition of a bubble is that it equally well describes what happens when asset prices rise and never crash back down again. Its not only reasonable, but a good idea to buy an asset at a price not justified by its current returns provided you can justify it based on expected future returns, which you might well expect to realize through appreciation in the asset price once everyone else catches up with you.

    Its similarly perfectly reasonable, indeed profitable, for lenders to finance such purchases even when the asset doesn’t generate the cash flow for the debt payments, by postponing the principal and even some of the interest until the borrower’s expected gains are realized.

    The thing that makes ponzi finance ponzi is that the expected gains never come. But this is just saying you can only really recognize bubbles in restrospect – its not helpful.

    I think the only way you can identify a bubble is by looking at strategies and expectations – if everyone buying into a market is expecting to pay for it through asset price appreciation, and justifying it from historical price data rather than expected changes in real returns, that’s a bubble. But of course, the strategies and expectations of market participants in a market like housing are incredibly hard to measure – even returns are hard to measure for owner-occupied housing. Even when you have better measures (eg. in the stock market you can look at futures and options, debt secured against the stock, company’s bond prices) the EMH still seems to hold – even when the stock is in a bubble the other related prices are usually consistent with it.

  12. Gravatar of David David
    8. January 2010 at 15:32

    When the marginal Homebuyer says, “Sure, I’ll take the teaser rate–or interest only, that sounds good too. Yeah, I can swing 40% PITI and 5% down. By the time it resets the house will be worth twice as much, just like last year.” And the Bank says, “put your e-signature here, sir, no documentation needed.” And Wall Street says, “Looks good. I’ll take two million just like it.” And Moody’s says, “Oooh what a cute little tranch you made! Aaa!” And the World says, “High grade debt at these premiums? It’s too good to not be true!” it’s a bubble.

  13. Gravatar of Doc Merlin Doc Merlin
    8. January 2010 at 15:40

    What I use when looking at charts:
    Prices increasing faster than exponential, with a log periodic power series.

    What I use when investing:
    Prices that are far above growth and profits that can be sustained for more than a generation.

    Anyway, Scott, I agree about CRE’s they are a lagging indicator and not a bubble.

    Anyway, look at population movement over the past decade. People have been fleeing NY and Cali and coming to Texas, Alabama, etc. Yet housing prices in Cali and NY rose faster than in Texas and Alabama. Also, the states that faired the best in this recession were low tax, low benefits states with high agri output, like Texas, Nebraska, etc. This is despite the food price crash that happened.

    It seems to me what we are seeing is the effect from high-regulation, high-tax, high-benefits states towards low-regulation, low-tax, low-benefits states. My thoughts are confirmed when I look and see what states and cities are almost ready to default.

  14. Gravatar of ssumner ssumner
    8. January 2010 at 19:46

    I am out of time tonight, but if any of you starting reading my blog after March 2009, you might be interested in this post. I still think it is my best post on bubbles, and explains my odd view that bubbles might exist, but that that fact is of no use to us.

    http://www.themoneyillusion.com/?p=695

    I will try to answer the comments tomorrow.

  15. Gravatar of scott sumner scott sumner
    9. January 2010 at 06:50

    q, How am I overstating Krugman? Didn’t he claim CRE was also a bubble?

    OGT, You said;

    “As Calculated Risk often points out it is typical for commercial real estate to peak five quarters after residential. Commercial real estate is a concurrent or lagging indicator, while residential is usually a leading indicator.”

    This strongly supports my view. The recent bubble in residential RE was the first nationwide bubble in RE in my lifetime. Since these are often correlated, it is probably simply a cyclical move in CRE, not a bubble.

    OGT#2, That definition doesn’t work in a country with high secular inflation.

    Thanks Mike@pvl.

    rob, Roubini’s reputation was not for calling the bubble, it was for predicting that the financial system would crash.

    I know of only one investor who predicted that (I forgot his name) and he made billions of dollars in 2007-08 selling financial assets short.) If everyone knew it was going to happen, everyone could have done the same.

    Philo, I was sort of half-joking. I was saying “when the speaker says ‘bubble,’ I assume he means he doesn’t know the cause.”

    Tom, I think you are confusing easy money with easy credit. Money was much easier in the 1960s and 1970s but we didn’t have bubbles (until the unimportant gold bubble in 1980.) In the 1920s, 1980s, 1990s and the 2000s money was much tighter, and we had big bubbles. There is no correlation between easy money and bubbles.

    David, Good point.

    Mike, I wish I could predict when bubble will pop. I’d be rich.

    Simon K, Yes, I agree. Unfortunately I don’t follow individual stocks, but haven’t there been stocks in tech and biotech that exhibited bubble-like rises, but stayed high? (Microsoft? IBM?, Amgen? Genentech?) Of course they have ups and downs, but aren’t there stocks that soared and stayed at a relatively high levels? Maybe someone has better examples.

    David, You need one more. Housing supply restrictions. Many areas of central USA did not experience bubbles because the supply of land for building is almost unlimited. But that is a good list.

    Still, it begs the questions of why people thought prices would keep rising.

    Doc Merlin, I agree with your analysis of the regional problems. I am not a believer in chart watching. (See my answer to Simon k.)

  16. Gravatar of OGT OGT
    9. January 2010 at 07:36

    I’d say quite the opposite. Most local housing bubbles like Texas in the eighties and LA in the early nineties were also accompanied by CRE bubbles. It’s generally more accurate, in fact, to speak of real estate bubbles.

    I read your linked post on bubbles. I do think its necessary to separate the bubble response of investors and policy makers. The former should be ever vigilant to bubbles, otherwise how are markets supposed to work. As for policy makers, I generally go with the DeLong/Gagnon line of policy makers focusing on credit expansion rather than asset prices (though misaligned asset prices is a warning sign)

    http://www.econbrowser.com/archives/2010/01/monetary_policy_2.html

    I’d also push back on the EMH thing, as I have an opposite reaction to the seriousness of a writer when I see it waved around casually. It’s certainly not enough to tout EMH if one doesn’t know the specific incentive structure of a particular market or the distribution of information. Mike Konczal as a good post on the limits of arbitrage at Baseline Scenario:

    http://baselinescenario.com/2009/08/21/the-limits-of-arbitrage/

  17. Gravatar of OGT OGT
    9. January 2010 at 08:10

    Sumner- You said: OGT#2, That definition doesn’t work in a country with high secular inflation.

    Well, then, of course, the authorities have other issues to deal with. But, the asset spike would still pose a problem assuming the authorities had any intention of returning historic NGDP growth rates, especially if leverage rates are high.

  18. Gravatar of rob rob
    9. January 2010 at 08:53

    Scott, you said:

    “I know of only one investor who predicted that –(the housing bubble)–(I forgot his name) and he made billions of dollars in 2007-08 selling financial assets short.) If everyone knew it was going to happen, everyone could have done the same.”

    I’m not saying everyone knew it, but an informed minority did. A lot of people such as myself put off buying a house because of the reported bubble, mainly expressed in the historically low ratio of monthly income/mortgage payment. Shorting it required a financial sophistication I don’t have — and it requires getting the timing right — the real hard part. In 2005-2006 I believed strongly that prices would be lower in the next 5 years, but I didn’t know when. I just took a strategy of renting until they did…

    I don’t know: maybe Wall Street professionals think they are too sophisticated and smart to take their cues from headline stories in The Economist. (although I suspect The Economist is taking its cues from professionals).

  19. Gravatar of Doc Merlin Doc Merlin
    9. January 2010 at 09:50

    A lot of people predicted the bubble, only very few had the positioning to take advantage of it on the short end. It was very, very hard to short MBSs. Heck, even Cramer was talking about a looming housing crisis in late 2006/ early 2007.

  20. Gravatar of rob rob
    9. January 2010 at 10:08

    Just a few links to The Economist confidently calling the housing bubble:

    1-12-2006:
    http://www.economist.com/businessfinance/displaystory.cfm?story_id=E1_VPRTJVJ

    “The Economist’s long-running quarrel with Mr Greenspan is that he chose not to restrain the stockmarket bubble in the late 1990s or to curb today’s housing bubble.”

    10-27-05
    http://www.economist.com/opinion/displaystory.cfm?story_id=E1_VDRQGVV

    “If confirmed, Mr Bernanke will take charge of the Fed at a particularly tricky time, when the American economy looks unusually unbalanced, with a record high current-account deficit, the biggest housing bubble in history and rising inflation.”

    6-16-05
    http://www.economist.com/businessfinance/displaystory.cfm?story_id=E1_QDSJDNS

    “Not only does this dwarf any previous house-price boom, it is larger than the global stockmarket bubble in the late 1990s (an increase over five years of 80% of GDP) or America’s stockmarket bubble in the late 1920s (55% of GDP). In other words, it looks like the biggest bubble in history.”

    5-26-2005
    http://www.economist.com/world/unitedstates/displaystory.cfm?story_id=E1_QDTDQVG

    “More evidence that the housing market has lost touch with reality”

  21. Gravatar of ssumner ssumner
    9. January 2010 at 11:35

    OGT, I said there were no national bubbles before 2006. The local bubbles you mention probably weren’t bubbles, just mistakes by the market. There is a difference BTW, something that many people overlook. Coastal California prices are still very high, so I don’t see much evidence that the run-up in the 1980s was all that irrational. And of course Texas is easily explained by the oil bust. Prices never got all that high, and naturally fell when the oil market tanked.

    I don’t get the baseline scenario argument at all. LTCM was a bunch of arrogant smart guys who thought the EMH was wrong and thought that they could beat the market. They failed miserably, and this is evidence against the EMH? To me all it shows is that the market is smarter than LTCM. They were picking up nickels in front of a steamroller based on there being no Black Swans out there. Not a smart strategy.

    rob, OK, but a lot of smart guys on Wall Street do have the sophistication, and only one did it. So it can’t have been that obvious.

    BTW, some commeneters will say “they knew it would crash, but just not when.” But that info would be useless. All markets crash on occasion, so saying it will at some point is meaningless.

    Doc Merlin, Just short the big banks.

    rob, Thanks, I recall The Economist making those predictions. But they did not draw the correct policy implications. They did not and do not understand the need to maintain stable NGDP growth. Didn’t they oppose some of the Fed’s rate cuts in 2008?

  22. Gravatar of David Pearson David Pearson
    9. January 2010 at 11:45

    Scott,

    I seems to me that you define a “bubble” as “irrational valuation” or “overshooting in price”. I agree that this is not a very useful construct, but there is a better one.

    The true definition of a bubble is a positive feedback loop whereby asset prices affect the real economy and the real economy affects asset prices. The feedback loop proceeds until its own dynamic undermines it. Valuation is a part of this definition (as it helps us understand how far the feedback dynamic has progressed), but only a part.

    A concrete example: asymmetric monetary policy led to an skewed expected return distribution for real estate assets, which in turn caused credit to flow to those sectors (the real economy response). The more the return distributions were seen as skewed (“real estate prices don’t fall, and the Fed won’t let them), the more prices rose, the more credit flowed, the more demand, and the more prices rose, which in turn skewed the expected return distribution still further. The self-destructive dynamic, of course, was that eventually prices would rise above the ability to service the debt, and demand would evaporate. This is, in fact, what happened.

    So how could one predict the timing of such a bubble’s demise? It was, in fact, relatively easy. Once one could ascertain that the marginal buyer could not qualify (or self-qualify) for the marginal loan, the bubble would begin to unravel. In the case of housing, the marginal loan was an option arm (literally the lowest-payment loan that could possibly exist), and a useful tool for timing the demise of the bubble was a flattening-out in the penetration of option arm loans into the mortgage market. This occurred in the summer of 2006, which, in fact, was the peak of the market in Southern California.

    So I suggest that bubbles can be well understood, and in my fortunate experience, even timed. The nature of true bubbles is that the feedback loop can be studied, understood, and predicted with some degree of certainty.

  23. Gravatar of ssumner ssumner
    9. January 2010 at 11:59

    David, It sounds like you did well in this case. But I have a question about your definition. First of all, there is no such thing as a “true definition,” rather there are useful definitions, and I think that is what you mean. Lots of people talk about the silver bubble of 1980, although silver doesn’t affect the macroeconomy. So definitions are partly reflective of how terms are used in the real world. I think you are saying the only useful definition of a bubble is when there is feedback to the economy.

    But what if the Fed had targeted NGDP? Would that have prevented feedback from the housing bubble? In that case it seems that almost by definition a bubble invloves bad monetary policy.

  24. Gravatar of rob rob
    9. January 2010 at 12:25

    Scott, you said:
    “I recall The Economist making those predictions. But they did not draw the correct policy implications. They did not and do not understand the need to maintain stable NGDP growth. Didn’t they oppose some of the Fed’s rate cuts in 2008?”

    I’m certainly not going to argue with you over them drawing the wrong policy implications. My point is that were spot on in calling the housing bubble (and the tech bubble). So while people can talk about how “difficult” it is to spot bubbles, The Economist seems to have a mastery of it. Could it be because they aggregate a lot of studies and then take a common sense approach to what seems reasonable?

  25. Gravatar of David Pearson David Pearson
    9. January 2010 at 12:38

    You are right — I should have said useful rather than true when referring to the definition. With regard to different definitions of bubbles, I meant to create a distinction between pure valuation anomalies (silver, tulips) with real-economy feedback loops (housing, internet stocks, railroads).

    I’m not sure all feedback loops like the housing one involve monetary policy, but since it is both artificial (let’s say, man-made) and impactful, monetary policy tends to play a starring role.

    The key question for NGDP and bubbles is whether NGDP level targeting is truly symmetric. By definition, it is, right? In other words, there is no inherent asymmetry where target levels are RAISED when NGDP grows faster, but kept when it grows slower or shrinks. At least, that’s the theory as I understand it, and therefore it should actually eliminate bubbles caused by monetary policy.

    There are two problems with the theory. The first is that the Fed is given towards asymmetry, and therefore will have a tendency to relax as targets are exceeded. This would be a big mistake. Second, if the achievable trajectory of the real economy falls below that inherent in the NGDP level target, the Fed will end up lifting long term inflation expectations as the market will expect asymmetry. That is, they will expect the Fed to not tighten when nominal NGDP rises but RGDP treads water. What matters is not whether the Fed will do this, but whether the markets think the Fed will do this. You need credibility for symmetry, and after decades of a Greenspan/Bernanke/Mishkin/Kohn Fed pushing asymmetry as dogma, this is a tall order, in my view.

  26. Gravatar of David Pearson David Pearson
    9. January 2010 at 13:11

    BTW, Scott, I would ask whether the feedback loop bubble definition violates the EMH? Markets seem to be blind to the somewhat predictable terminal point of the loop. How can this be? What I witnessed in this particular bubble is that markets “invent” modes of analysis that are used to predict away the terminal point. One could say that, therefore, EMH still applies.

    To be specific, during the housing bubble underwriters used “sophisticated” econometric models to predict defaults. These models resulted in strange and counter-intuitive underwriting policies. For instance, there was no correlation found between “liar loans” and defaults, because obviously, the more liar loans, the more housing demand, the higher housing prices, the easier to refinance, the less defaults. So Wall Street took this and said, “let’s make liar loans with abandon — they will never go bad”. And the credit market said, “we’ve analyzed that too, we agree, and please sell us more liar loans”. Is this market efficiency or inefficiency?

  27. Gravatar of OGT OGT
    9. January 2010 at 15:00

    OK, I’ll bite what is the difference, in your definition, between a market mistake and bubble.

    I think the Baseline article is relevant for two reasons, it stresses the importance of understanding the structure of a particular market in understanding its dynamics and its susceptibility to large scale over valuations. One would never advocate the same betting strategy in Texas Hold ‘Em and Blackjack. Talking generically about ‘markets’ is in many ways not useful.

    Second it demonstrates a concrete inefficiency, and it’s this inefficiency along with the feedback loops, or reflexivity, and the high expense of information that David wrote about above that make feedback loops possible.

    You seem to be arguing a very weak form of EMH, that markets are hard to predict, but OK, so? Policy makers should still probably review credit conditions and capital requirements in markets moving significantly away from fair value given the asymmetric risks.

  28. Gravatar of rob rob
    9. January 2010 at 15:02

    Scott, you said:

    “OK, but a lot of smart guys on Wall Street do have the sophistication, and only one did it.”

    If only ONE shorted it, then doesn’t that seem like a pretty loud indication that a herd-like mentality was present among Wall Street professionals? Perhaps today’s financial professionals have less common sense about finance than those outside of it. Perhaps they are merely sophisticated lemmings. Isn’t being a sophisticated lemming nearly a prerequisite for getting in the door at investment banks in the first place?

    Amazing all those Economist articles and only ONE GUY decides to short it…

  29. Gravatar of Mike Sandifer Mike Sandifer
    9. January 2010 at 15:40

    Dr. Sumner,

    If you haven’t anticipated this, if you look at some graphs of possible bubble phonomena, perhaps it’ll help illustrate my points.

    For example, for the recent housing price run up and crash:

    http://mysite.verizon.net/vzeqrguz/housingbubble/index.html

    Or the Tech bubble as compared to some other bubbles (the author’s concerns not withstanding):

    http://bigpicture.typepad.com/comments/2005/03/nasdaq_and_othe.html

    And yet another comparison, now including the stock “bubble” of 1929:

    http://www.vanenschot.com/finance/stockmarketcrash.jpg

    I quickly grabbed these off google, so my apology for any sloppy examples, but to my memory these seem to represent the data upon which my original conclusions shared above rested.

    From my perspective, spotting great short opportunities during bubbles is not always all that hard.

  30. Gravatar of Doc Merlin Doc Merlin
    9. January 2010 at 18:18

    The problem is predicting when the bubble will actually pop. Its pretty obvious when you are in a bubble, but most attempts to short can push up prices all at once (the short squeeze is one mechanism that does this).

    In my experience its much harder (and expensive) to short than long, even when the bubble is obvious. :-(

  31. Gravatar of Doc Merlin Doc Merlin
    9. January 2010 at 18:19

    Oh forgot to say, looking at the amplitude of the log periodic behavior of the bubble can tell you roughly when it will pop, although now that this is well know established, I don’t know if that behavior will remain.

  32. Gravatar of Mike Sandifer Mike Sandifer
    9. January 2010 at 19:30

    Doc, it’s interesting you should say that.

    I first tried this approach near the end of the housing bubble and it “worked”,but I was too timid to take full advantage of it. I’d been burned before by what I thought I understood.

    Then my lack of macro knowledge cost me dearly in late ’08, but I’ve already learned enough here and elesewhere(especialy here) to help avoid those sorts of mistakes in the future, or so I hope.

  33. Gravatar of scott sumner scott sumner
    10. January 2010 at 18:48

    rob, More likely they simple cry “bubble” whenever some asset class has a huge rise.

    I am confident that if the Fed decided to do NGDP level targeting, they could be symmetrical. The problem is they have no desire to do this, so we can’t judge the feasibility based on how they act now.

    As I said, there are lots of things that look inefficient, and thus seem to violate the EMH. But somehow the anti-EMH theory fails to be useful. I don’t know why.

    OGT, You said;

    “You seem to be arguing a very weak form of EMH, that markets are hard to predict, but OK, so? Policy makers should still probably review credit conditions and capital requirements in markets moving significantly away from fair value given the asymmetric risks.”

    I think it is a huge mistake to do this. We will never be able to predict markets, or know when they are under or overvalued. But we do know how to prevent severe recessions–keep NGDP growth fairly stable. If we try to outguess markets it will be a distraction from what we should focus on. Bubbles are not public policy concerns if we have sound monetary policy and get rid of TBTF. Investors may lose some money on occasion, but that’s their problem. The Fed turned a regional housing bubble into a national one by letting NGDP fall sharply. The “correct” value of housing is very closely linked to NGDP growth, and that’s the Fed’s domain.

    An example of a mistake is when oil prices were high right before the 1991 Gulf War, and then suddenly plunged when the war looked like a rout. An example of an inefficiency is when stock prices dropped 40% around August to October 1987. The first was an honest mistake, as markets overestimated how much the war would effect oil shipments out of the Persian Gulf. The second was simple irrationality, as there was no important new information that would justify such a sharp price change.

    Rob, you said;

    “If only ONE shorted it, then doesn’t that seem like a pretty loud indication that a herd-like mentality was present among Wall Street professionals? Perhaps today’s financial professionals have less common sense about finance than those outside of it. Perhaps they are merely sophisticated lemmings. Isn’t being a sophisticated lemming nearly a prerequisite for getting in the door at investment banks in the first place?

    Amazing all those Economist articles and only ONE GUY decides to short it…”

    That’s like saying “I could play better pass defense than the Packer secondary.”

    Come to think of it I could, so maybe you are right.

    Mike, I can’t quite follow those graphs. They seem to simply show stocks going up a lot, and then down a lot. (I don’t even believe 1929 was a bubble.) How do they show predictable bubbles? Any volatile time series will have big ups and downs.

    Doc Merlin. You said;

    “The problem is predicting when the bubble will actually pop. Its pretty obvious when you are in a bubble, but most attempts to short can push up prices all at once (the short squeeze is one mechanism that does this).

    In my experience it’s much harder (and expensive) to short than long, even when the bubble is obvious.”

    If your experience tells you that, if you lost money, then it wasn’t a bubble. I keep telling you guys that these are mostly optical illusions. Graphs make time series look very predictable—until you try to do it.

    Example; My mutual funds are up 80% since March. Do I sell? Are you sure? If it is a bubble then it is 100% true that I should sell, even if the rally has farther to go. But I have no idea if I should sell. It isn’t just because I don’t know if there is another 20% up, I don’t know whether stocks will be higher or lower 1, 2 or 3 years out. And if this 80% run-up really was a bubble, I should be pretty sure.

    In the mid-2000s the Fidelity Latin America fund rose 5-fold, that’s 400%!!! Was that a bubble? Was it a good time to sell? No, it promptly doubled again, for a 10-fold increase.

  34. Gravatar of Mike Sandifer Mike Sandifer
    11. January 2010 at 16:35

    Dr. Sumner,

    You replied: “Mike, I can’t quite follow those graphs. They seem to simply show stocks going up a lot, and then down a lot. (I don’t even believe 1929 was a bubble.) How do they show predictable bubbles? Any volatile time series will have big ups and downs.”

    The point is that in each of those comparisons of “bubbles,” the rise was near vertical before the sharp decline in each case. And the decline can’t be far behind, given that verticality is obviously an insurmountable limit to such rises.

    So, my only point is that the larger the rise versus trend, assuming fundamentals remain near trend(tricky to read these at times, admittedly), and the more vertical the rise, the sooner and more severe the downturn that will occur. Timing usually isn’t at all difficult with near vertical rises, as the rate of change over time is obviously extreme, leaving a small window for the bubble to burst.

    Sure, this doesn’t slways hold up, but it does often enough to trade on, and it needn’t only describe what one might call bubbles. If you look at graph after graph, obviously most rises are like this in general.

    These are the only points I have to make.

  35. Gravatar of ssumner ssumner
    12. January 2010 at 13:51

    Mike, You said;

    “The point is that in each of those comparisons of “bubbles,” the rise was near vertical before the sharp decline in each case. And the decline can’t be far behind, given that verticality is obviously an insurmountable limit to such rises.”

    Actually it can be far behind. The Fidelity Latin America fund was just over 5 in August 2002. By early 2005 it was nearly 20. That’s a bubble right? No, it kept rising for several more years up to 60, and it still at 45 today.

    http://quote.morningstar.com/fund/f.aspx?t=FLATX

    Anyone who sold in early 2005 at 20 was an idiot. I realize it doesn’t look vertical right before early 2005, but that’s because it was a log graph. If it was a regular graph it would have looked pretty vertical.

    As far as your broader point that it usually works, I can’t say, you might well be right. All I know is that it is very hard to get rich, so it can’t be as easy as many people make it seem.

  36. Gravatar of TheMoneyIllusion » Which state had the most bank failures during 2008-10? TheMoneyIllusion » Which state had the most bank failures during 2008-10?
    3. September 2010 at 09:36

    [...] earlier posts I argued that the commercial real estate market does not appear to have been a bubble.  It held up [...]

Leave a Reply