Are we living in a time of asset bubbles?

Here is one typical complaint about bubbles, from Jesse Eisinger, excerpt:

We are four years into the One Percent’s recovery. Now, we are in Round 3 of quantitative easing, the formal term for the Fed injecting hundreds of billions of dollars into the economy by purchasing longer-term assets like Treasury bonds and Fannie Mae and Freddie Mac paper. What’s that giving us? Overvalued stocks. Private equity firms racing to buy up Arizona real estate. Junk bond yields at record lows. Ratings shopping on structured financial products.

These are dangerous signs of prebubble activity.

Here is a Krugman rebuttal.  I will offer a few points on a series of debates which in general I have stayed away from.

1. I don’t find most predictive discussions of bubbles interesting, while admitting that such claims often will prove in a manner correct ex post.  “OK, the price fell, but was it a bubble?  I mean was there froth, like on your Frappucino?”  Or to quote Eisinger, it might also have been “dangerous signs of prebubble activity” (what happens between the “prebubble” and the “bubble”?  The “nascent bubble”?  The “midbubble”?  The “midnonbubble”?)

2. Good news and improving conditions may well bring more bubbles or greater likelihood of bubbles, but that is hardly reason to dislike good news and improving conditions.

3. Relative to measured real interest rates, stocks look cheap right now.  That doesn’t mean they are, but reread #1.

4. No one understands the term structure of interest rates, no matter what they tell you.  Reread #1.

5. I don’t see why anything particular about the current state of affairs, at least in the United States, needs to be “unwound.”  I sometimes draw a distinction between those of us who have been thinking about interest on reserves since S. Tsiang, Fischer Black, and the Reserve Bank of New Zealand, and those of us who have not.

6. One coherent definition of bubble is that of a hot potato, traded in a world of heterogeneous expectations, but which must ultimately pop, because eventually the price of that asset will consume all of gdp, a bit like those old Tokyo parking spots.  Fair enough, but I don’t see that in many asset markets today if any (Bitcoin for a while?).

7. Another coherent definition of a bubble has less to do with a dynamic price path and ongoing resale for gain, but rather there may be a (temporary) segmentation across classes of asset market buyers.  The obvious candidate here is that  many people and institutions have been frightened into Treasuries and away from almost everything else.  That could mean we have a real interest rate bubble, but it also could mean that lots of other assets are undervalued, at least if the liquidity effect defeats the higher real interest rate effect of moving out of Treasuries.  (It would be odd to think that a shift of funds out of Treasuries and into stocks would cause stock prices to fall, but perhaps some people fear this.)

I don’t agree with this view, but I do feel I understand it.  The most likely “bubble” is then in real interest rates, due to a (temporary?) skewing of the risk premium.  That all said, I do not think this should be called a bubble.  Changes in the risk premium and “bubbles” have traditionally been considered alternative explanations for asset prices.  Reread #1, and reread #4 while you’re at it.

8. Ruchir Sharma made some interesting points yesterday:

Far from fighting off a deluge of foreign capital, leaders from India to South Africa are struggling to attract a greater share of global capital flows in order to fund widening current account deficits. Over the past decade, the foreign exchange reserves of the developing world grew at an average annual rate of 25 per cent, swelling from $570bn in 2000 to $7tn in 2011. But over the past year, the average rate slowed to a crawl of barely 5 per cent.

The idea that money is still flooding emerging markets misses the big picture, which is that global cross-border capital flows are down 60 per cent from their 2008 peak. The largest shares of cross-border capital flows are in bank loans, trade and foreign direct investment, which are slowing worldwide.

9. I expect the real economy over the next twenty years to be more volatile than it was say in the 1990s.  In that sense, many current asset market prices may be revised and quite dramatically.  Still, I don’t find the bubble category to be so useful in this regard.  We really don’t know what is going to happen and that is why the current prices are wrong, not because of a “bubble.”

10. I am probably done blogging about bubbles for a while.  Satisfying you was not the goal of this post, but that is in the nature of the subject area, not out of any desire for spite.


:Reread #1, and reread #4 while you’re at it."

I re-read 1 to 10. Very well-written , I think. When all we are seeing are Black swans , perhaps it may be time to remove the colored glasses.

Except you can't see Black Swans, that's the point of the Black Swan.

A lot of bubbles are misnamed herd effects. The herd will overvalue assets from time to time. A bubble is when the herd can continue rushing in with no wider economic effects because it is being artificially propped up. For example, if everyone got into real estate, other investment assets should drop as money is siphoned away. Or consumption should fall.

Some bubbles, like the Internet, have real value. Had you bought tech stocks in 1996 or 1997 you came out ahead even if you held through the crash.

However, in almost every case: tech stocks in 1999, housing in the 2000s, you can see the Fed pursuing an easy money policy. Today, there is no easy money policy because the entire economy is experiencing credit deflation. The Fed and fedgov deficit spending is propping up the entire economy. Thus, if there is a bubble it is in GDP. Otherwise, there is not a bubble.

I am not very astute about these things, but I find it hard to understand how the recent acquisitions of Instagram and Tumblr are not bubbles. Is there a good argument why these companies are worth the price they were bought at?

IT seems to have an impending bubble.

That one's easy. "A billion dollars" means something very different to you, I, and Yahoo. (They aren't quite saying "here kid, have a nickle ...)

Yahoo is paying $1.1B for Tumblr. It's quite difficult to value companies based on a steep forecasted growth curve, but one estimate for Tumblr was to go from $13M in revenue for 2012 to $100M for 2013. Yahoo is essentially attempting to buy growth (and excitement) they cannot generate internally. Execution, however, is critical, and Yahoo's history over the last decade does not lead to optimism. Think of GeoCities, HotJobs, Altavista....

So it's not so much that Tumblr was overpriced, but whether or not Yahoo was the correct buyer. We'll see about that.

The Instagram deal looks better. (Remember that part of the purchase price was in overvalued Facebook stock.) But the real difference is that Facebook appears to not be screwing Instagram up -- for example, Instagram customer base doubled since the Facebook deal, as opposed to Facebook's slower growth.

So you might say both Facebook and Yahoo bought growth; Facebook clearly did not screw up its acquisition, and it remains to be seen whether Yahoo will.

What can you possibly mean by this?

@enoriverbend @john personna

I don't know much about investing, but is the current situation much different from right before the last dotcom bubble? Aren't Price-to-earnings ratios through the roof again (not sure what the 1999 numbers were)?

Many of these current favorites don't even have a P/E ratio since they've either made a loss or even worse, don't even have a revenue stream in the first place! Will the rosy expectations work out this time around?

Yahoo is for the Nth time paying hundreds of millions if not billions of dollars for assets that cost tens of millions to create.

The last time Yahoo did that, dozens of people were spending tens of millions to create hundred million to billion dollar NASDAQ assets to dump, creating the NASDAQ bubble in the 90s.

My favorite example of this is A billion dollars of vapor created and sold in a year.

@Rahul, P/E ratios aren't in the same ballpark as 1999. I think they're around 15-20 right now, somewhat higher than historical averages.

But the reciprocal, the "earnings yield" is a more useful number (currently, 5%-6%).

The earnings yield can be compared to fixed income yields (maybe 4% for long duration [inflation risk!] with some credit risk right now.) In a low-yield environment, you would expect stocks to support a higher P/E ratio (lower earnings yield.)

Shiller thinks you should use 10-year averages for these numbers, but I think you lose information this way, and right now those numbers are depressed by lousy earnings years in 2007-09. The 10-year average numbers that a lot of people use are scarier.

What concerns stock investors is that current profits (historically high as a % of GDP) are not sustainable. I'm not so sure. Companies have shown an ability to make money in a tough economy over the past four years. Part of me feels like high corporate profits and high unemployment are not a coincidence. In other words, adding marginal employees may hurt profits. In this case, though, as we are told here all the time, there will be countervailing positive macro effects.

The network effect that instagram and tumblr have created among us consumers is valuable to advertisers... as soon as someone could figure out how to use it.

I have no idea if the prices paid for instagram and tumblr are worth it, but someone thinks that they can figure it out.

I think it is important to distinguish between those who have built a network and those who have merely captured attention. The later is more fleeting. Even knowing that though, you might pay a billion for fleeting attention, if billions mean less to you. (As an aside, I find this week's headline that "kids abandon Facebook for Twitter" amusing. Twitter with crude old tech keeps converting people from more advanced forms. The attention they've captured is amazing.)

Yes, but if that "someone" is systemically wrong, it becomes a bubble, doesn't it? And I see a trend. Lots of Internet startups are selling for huge amounts with very little to offer by way of revenue-streams. Neither do they come with much physical-assets, nor large employee-capital (didn't Instagram have less than 50 employees?) nor strong Patent-portfolios.

I'm skeptical if a billion dollar investment will pay off on network effects alone. OTOH, a part of that "billion" is accounting-wizardry I guess since a lot of that payment is in parent company stocks I suppose?

'Twitter with crude old tech'

Smart phones aren't quite crude old tech - and twitter is pretty much about data entry as it works best with a camera equipped smart phone.

SMS is pretty old crude tech - whether one considers Twitter a simple update from a one to one method to a one to many model is another discussion.

felix salmon observed that "For Yahoo, having $4.4 billion in cash plus Tumblr is clearly going to be better in terms of the future than having $5.5 billion in cash." Having $5.5 billion in cash makes the purchase a "soft billion" for Yahoo.

"The network effect that instagram and tumblr have created among us consumers is valuable to advertisers… as soon as someone could figure out how to use it."

But if some other teenager comes up with a cuter name and spends a year hacking instead of living and has parents who let him drop out of high school, tumblr oe instagram could quickly go the way of the Beta to Facebook's VHS. (Murdock screwed up Myspace big time.)

In the last 14 months, prices in the Case Shiller 20 Cities grew at rates that differ from 28% in Phoenix to 0.5% in New York. House price increases are driven by lower mortgage rates resulting from the Fed’s purchase of long term bonds. This analysis explains the city differences that correlate at 85% to three factors: months of inventory, income growth, and the percent decline from the 2006 peak In about half of the 20 cities price growth driven by investors is at bubble levels.

The regions outside the Case Shiller 20 called “Heartland” represent the bulk of the U.S. housing market. This market accounted for 75% of housing sales in the last year but it got less than 5% of the attention. Why? It’s boring. House prices are one third that of the Case Shiller average and one fifth of the California average. Inventories are high. They did not experience a big gain in the housing bubble and the house prices average 15% below the inflation adjusted price at the start of the bubble in 1997, while Case Shiller cities are 30% above. Prices in these regions are slowly but steadily recovering.

Re point #9, I’ll suggest that pushing investors into stocks and high yield bonds (via the Fed’s portfolio balance channel) raises the risk that these asset classes will be found out – ex post – to have been priced to deliver poor returns. (It can have the unintended consequence of encouraging investors to follow their natural instincts – anchoring, availability heuristic, recency bias, etc. – and chase returns instead of weighing prices against potential default rates, earnings growth, etc.)

If so, then asset prices will be more volatile than they would otherwise be, contributing to the economic volatility that you and many expect.

I’m not disagreeing that the term "bubble" is nebulous and often unhelpful – there are many excellent points in this post – but this is how I interpret Jesse Eisinger’s concerns about overvaluation and “dangerous signs of prebubble activity.” It’s not a reason to “dislike good news and improving conditions,” as you say, but it is recognition that the good news may be at least partially linked to a greater likelihood of bad news in the future.


I don't think that we can say the S&P500 or HY bonds are in a bubble right now. But they certainly seem priced to deliver poor capital gains/poor returns...

As an individual investor, nearly all asset classes seem priced to deliver poor returns over the long term.

That's a little different than in a classic bubble in which certain asset classes (tulips, internet startups in the 1990s, residential housing in the 2000s) seem to promise high returns, particularly in the short term.

Yes, I would agree with you.

Working in the financial industry, I can tell you it's always a tough time to 'jump in' the markets. It's almost never easy to make the first investment... You're always afraid of things going wrong.

But, right now, it feels even tougher than usual. Apart from truly market-neutral L/S and systematic HFs (i.e. strategies supposedly de-correlated from markets' direction), it's hard to buy anything right now. And shorting? My God, that seems a receipt for disaster right now. I might consider buying CDS HY protection, though. At 340 (for the US) and 380 (for EU Crossover), it's hard to imagine it going lower. At worst, it ought to stay flat-ish. Or buy volatility. VIX below 13 seems a steal. I mean, yes, it went lower in 2005-2006. But THAT was a bubble...

Compared to what? Stocks may be moderately overpriced based on historical standards, but IMO, investors are still risk-averse (i.e. bearing risk is lucrative, as it has been for the past four years for the most part.) There's a lot of money still in cash, but, as the market has continued to climb, the dumb money is starting to come back. Happens every time. Is this a bubble, or a cycle?

@Brian: "Investors are still risk-adverse".

Not according to the VIX or CDS HY indexes.

But I am not saying we're in a bubble. Indeed, I think I've said we weren't. Things can be over-priced or risks look badly priced (at least to me) without it being automatically a bubble...

This is also true for Tyler's 'real interest rates bubble'. The ultra-low IR are the results of policy actions, not markets. At most, as someone else was saying, low interest rates are enablers of bubbles.

I have an off-topic, technical question brought on by the sentence:

... because eventually the price of that asset will consume all of gdp, ...

As I understand it, GDP is an _annual_ figure, i.e. its unit is technically "$/year" rather than "$". (And that GDP growth would then be in "1/year^2") Thus pedants should translate the above to:

... because eventually the price of that asset will consume one whole year gdp, ...

Both his and TC's original sound reasonable. But imagine if the conventional unit was "$/attosecond" then it would be ridiculous to worry that " eventually the price of that asset will consume one whole attosecond of gdp, ...".

So what is so special about the year?

Correction: GDP growth has units of 1/[time], not 1/[time^2]. Assuming we are talking about fractional growth rather than absolute dollars.

All this talk of a bubble,

On a site

That believes in Efficient Markets.

Where are those folks who talked about the need to Buy Gold?

Was that a bubble?

I was scanning for someone who would name gold. I've lived long enough to see 3 cycles of gold boom and bust. It seems very much an on and off market. It is a fear play, until it is not. I'd certainly call each replay a bubble.

Silver is even a better example, considering the Hunt brothers, and how well their attempt to corner the market played out in the high inflation of the late 1970s.

"Was that a bubble?"


I suspect your "nope" is part of a semantic dodge. ;-) If you undefine bubble, nothing is.

We definitely have just seen another housing bubble burst this weekend:

"I am probably done blogging about bubbles for a while"

Wait, what...did I miss it?

Let's try another definition for an economic bubble:

An economic bubble exists whenever the price of an asset that may be freely exchanged in a well-established market first soars then plummets over a sustained period of time at rates that are decoupled from the rate of growth of the income that might be realized from owning or holding the asset.

Examples provided at the link above. Applying that definition, and looking at today's situation, there is no bubble at present in the stock market, nor has there been since the end of QE 2.0.

There does however appear to be a new bubble forming in the U.S. housing market, which looks to have begun in July 2012. Here is a longer term view of that data - the present rate at which median new home prices are escalating is consistent only with those seen during the first U.S. housing bubble.

A lot of people, especially those with an interest in seeing housing prices rise, are trying to sell the story that what is going on in the housing market is really a recovery. But to believe that, one would also have to believe that the housing market was never in a bubble previously. Those are not people to be taken seriously.

Interesting graphs indeed but how do the presently very low real interest rates affect the calculations?

Very low real interest rates do not affect the calculations (and keep in mind, they weren't a factor at all for the Dot-Com stock market bubble).

They are instead one of three enabling factors that must exist in markets where assets are purchased primarily through debt-financing in order for a bubble to form and be sustained in them, where "falling interest rates" in addition to "low interest rates" are a contributing factor. (Here are links to Part 2 and Part 3 of that series for the history of the first U.S. housing bubble.)

I think a discussion about real estate without considering real interest rates and people capacity to borrow is more or less incomplete by definition... And you kind of contradict yourself when you then refer to it as an 'enabler' - a very good descriptor by the way.

But if real IR are low and have been low for a long time, you can see why housing might benefit, all else being equal. In general, I think the point goes to Barkley Rosser, though, about supply elasticity.

London real estate prices stabilised a bit during the crisis and that was it. No drop, let alone a meaningful correction. That's to do with S/D.

Housing is probably not in a new bubble, although a handful of markets might be. A widely used definition of a bubble involves price going substantially above some fundamental, although what constitutes fundamental in any given market is often up for debate and in some virtually unmeasurable (gold is an example). However, for housing there is widespread agreement that the price/rent ratio is an excellent measure, although the fundamental value of that should vary with the real mortgage rate. In any case, rental markets generally reflect immediate real supply/demand relations in housing markets.

So, some markets have heated up in the past year, notably Phoenix, Las Vegas, and Northern California, particularly San Francisco. Glaeser has long pointed out that places vary in their supply elastiticities of housing, with Texas very high so that neither rents nor prices rose much during the national housing bubble. Boston and Northern CA, OTOH, have very low supply elasticities, so any increase in demand will show up as both rising rents and prices, although the price/rent ratio certainly rose during 1998-2006 in the Bay area.

I do not know what is going on with rents in Phoenix or Las Vegas, but in San Francisco rents are rising with a real demand driven by the high tech industry and particularly the decision by Google to locate their HQ in the city itself. The pricea/rent ratio is probably not going up there. No bubble, and in most of the US housing prices are simply not rising all that much, in many places still held back by a backlog of foreclosure sales. So, maybe there are a few markets getting a bit bubbly, but most of the US housing market does not look bubbly, or even pre-bubbly at all.

More broadly, I agree with most of what Tyler says here, with a few minor exceptions I shall not dwell on, although I have been publishing on the topic of bubbles for decades and can claim to be one of those who called the last housing bubble early on and warned that its collapse would have major repercussions througout the financial system much earlier than most. As it is, I see very few markets now that can be seriously described as being bubbles at the moment, with indeed real interest rates one of the few possibilities, as Tyler notes.

"I don’t find most predictive discussions of bubbles interesting"

If you can't use your knowledge of the past to predict the future and change actions accordingly of what use is it (besides entertaining you).

Actions need to be decided pre- not post-.

Does Tyler believe bubbles can happen at all? What do we call 2003-2007?

Something about credit snobs.

That was Alex, wasn't it?

'What do we call 2003-2007?'

The Dean Baker era - Prof. Cowen, being typical of his profession, does not like to discuss this time period, however.

Exactly. Economics is a subset of archaeology.

I consider bubble primarily a psychological phenomenon.
Sometimes the price of something can be too high, but it doesn't really become a "bubble" unless people become convinced in some way that it's just going to keep rising forever, thus leading to a positive feedback loop. Like "peak oil" a couple years back led people to believe that the price of oil was just going to go up permanently. Or how everyone came to think that buying a house was a perfectly safe investment. Or the DOW 50,000 predictions of the mid-90s (the stock market is just going to keep going up FOREVER!). Any time popular wisdom decides that the laws of economics have been broken, or someone has dicovered a formula for perpetual growth, that's pretty much automatic bubble territory, just be virtue of the fact that believing that down is not a possibility will cause a positive feedback loop.

My definition is similar to this. It is when people like a thing, but get caught up in too strong of a "beauty contest" feedback loop. Everyone, thinking everyone else will continue to love a thing, doubles down too many times. You like gold, or houses, buy some, see the feedback loop as other people buy some, recommit yourself that more and more people will buy some, until you go overboard. It crashes when prices reach a level which exclude new speculators. The flows, and beauty contest effect, reverse.

And no discussion of the logic of choice in production goods across time.


No talk of the changing liquidity of assets and the money substitute effect.

Very interesting.

"An economic bubble exists whenever the price of an asset that may be freely exchanged in a well-established market first soars then plummets over a sustained period of time at rates that are decoupled from the rate of growth of the income that might be realized from owning or holding the asset."

It happens.

Economists seem incapable of using the "toolbox of economics" to understand the phenomena.

Fascinating and very interesting.

Picking up your key points:

"And no discussion of the logic of choice in production goods across time."

"No talk of the changing liquidity of assets and the money substitute effect."

The data suggests that these factors don't come into play until *after* a bubble has formed and has been inflating for some time. Keep in mind that bubbles form quickly after being sparked - the feedback from the price signals take time to alter the inertia in the economy with respect to production decisions. Likewise for money substitute effects.

It would be nice is someone went through some of the data.

Of course, the knowledge problem blocks much ever everything working in the causal mechanism from being known to anyone but God.

Are we living in a time of asset bubbles?

Yes we are. When we depend on central bank credit expansion to keep everything up the answer is clearly yes.

You need to add some jargon if you're going to convince this crowd.

[Insert asset] have reached what looks like a permanently high plateau.

Wrong crowd, but I never miss a chance to mention Robert Brenner's What is Good for Goldman. Asset price Keynesianism etc.

I am always up for at least skimming things like this. But the author lost me on page 34, when he claimed that the NASDAQ fell "400 per cent" between 2000 and 2002.

OK, wrong denominator ; perhaps he meant 75% . But is that alone sufficient to dismiss the paper? Or are there many other things wrong with it?

prebubble activity

Bah, I'll start worrying when we get to mezzanine bubble activity.

The price action of this buying climax is reminiscent of 1987 which resulted in a single day drop of over 20%. Here is an overlay of this year's market compared to the ramp up to the '87 crash:

gold buble right now raising into sky and than crash landing
so if something was too high or too low, it will reverse someday with an amazing power

My definition of a bubble is:

It is a bubble if most buyers are buying because the asset price has been going up and without regard to fundamentals. I see not bubbles right now. Stocks are going up because their returns look better than alternatives.

Good thoughts, thanks for sharing.

The idea that money is still flooding emerging markets misses the big picture, which is that global cross-border capital flows are down 60 per cent from their 2008 peak. The largest shares of cross-border capital flows are in bank loans, trade and foreign direct investment, which are slowing worldwide.

Yep, money is tight. The Fed should not have started worshipping at the altar of the inflation target. It was a good policy for the 1980s and most of the 1990s, but it doesn't hold up well to big nominal shocks. Like Newtonian physics before relativity, let us refine.

What's up, for all time i used to check web site posts here early in the morning, because i love to gain knowledge of more and more.

Comments for this post are closed