Guest Post: A Bubble So Big We Can't Even See It

Tyler Durden's picture

Authored by Steve Keen via Real-World Economics Review,

Before the current turmoil began, Federal Reserve Chairman Ben Bernanke's hope was that rising asset prices would lead to a "wealth effect" that would encourage the American consumer to start spending again, and thus help the American economy finally leave the "Great Recession" behind.

His predecessor Alan Greenspan argued in February that this would work because:

"...the stock market is the really key player in the game of economic growth... The data shows that stock prices are not only a leading indicator of economic activity, they are a major cause of it. The statistics indicate that 6 percent of the change in GDP results from changes in market value of stocks and homes." (Greenspan 2013)

This is the so-called "wealth effect": an empirical relationship between change in the value of assets and the level of consumer spending which implies that an increase in wealth will cause an increase in consumption.

Greenspan's sage status is somewhat tarnished post-2007, so I don't think anyone should be surprised that his definitive statement involves a sleight of mouth. The "6 cents extra spending for every dollar increase in wealth" found in the research he alluded to was for the relationship between changes in the value of housing wealth and consumption, not stocks. In fact, the authors argued that the wealth effect from stocks was "statistically insignificant and economically small":

"Consistent and strong evidence is found for large but sluggish housing wealth effects... the MPC [marginal propensity to consume] out of a one dollar change in two-year lagged housing wealth is about 6 cents...

Furthermore, a statistically insignificant and economically small stock wealth effect is found ...

Additionally, there is evidence that the housing wealth effect is significantly larger than the stock wealth effect... these results suggest that it is necessary to take into consideration the potentially substantial difference between consumers' respective reactions to fluctuations in the housing markets and stock markets." (Carroll and Zhou 2010, p. 18. Emphasis added)

So the empirical data does not support Greenspan's notion that the stock market drives the economy (though the housing sector might). But equally the economy isn't booming sufficiently to make the reverse case that the economy drives the stock market. So what is causing the markets to boom right now?

Let's start by taking a closer look at the data than Alan did. There are a number of surprises when one does - even for me. Frankly, I did not expect to see some of the results I show here: as I used to frequently tell my students before the financial crisis began, I wouldn't dare make up the numbers I found in the actual data. That theme continues with margin debt for the USA, which I've only just located (I expected it to be in the Federal Reserve Flow of Funds, and it wasn't - instead it's recorded by the New York Stock Exchange).

The first surprise came when comparing the S&P500 to the Consumer Price Index over the last century - since what really tells you whether the stock market is "performing well" is not just whether it's rising, but whether it's rising faster than consumer prices. Figure 1 shows the S&P500 and the US CPI from the same common date-1890—until today.

In contrast to house prices, there are good reasons to expect stock prices to rise faster than consumer prices (two of which are the reinvestment of retained earnings, and the existence of firms like Microsoft and Berkshire Hathaway that don't pay dividends at all). I therefore expected to see a sustained divergence over time, with of course periods of booms and crashes in stock prices.

Figure 1: The S&P 500 and the CPI from 1890 till today


That wasn't what the data revealed at all. Instead, there was a period from 1890 till 1950 where there was no sustained divergence, while almost all of the growth of share prices relative to consumer prices appeared to have occurred since 1980. Figure 2 illustrates this by showing the ratio of the S&P500 to the CPI - starting from 1890 when the ratio is set to 1. The result shocked me - even though I'm a dyed in the wool cynic about the stock market. The divergence between stock prices and consumer prices, which virtually everyone (me included) has come to regard as the normal state of affairs, began in earnest only in 1982.

Until then, apart from a couple of little bubbles in stock prices in 1929 (yes I'm being somewhat ironic, but take a look at the chart!) and 1966, there had been precious little real divergence between stock prices and consumer prices.

Figure 2: Ratio of stock prices to consumer prices from 1890 till todays


And then, boom! What must certainly be the biggest bubble in stock prices in human history took off—and it went hyper-exponential in 1995.

In 1982, the ratio of stock prices to consumer prices was only 1.8 times what it was in 1915. By 1990, the ratio was substantial at 4 times - well above the level of 1929 (2.6:1) but below the peak reached back in 1966 (4.1:1). Then it just exploded to 12.5 times by the peak of the DotCom bubble in 2000.

Since then, it's been doing the Jitterbug. The current rally has erased the crash of 2008 in nominal terms, but at a ratio of just over 10:1 today, it still stands shy of the two previous peaks of 12.5:1 in 2000 and 10.5 in 2008.

So are stocks in a bubble? On this view, yes - and they have been in it since 1982. It has grown so big that - without a long term perspective - it isn't even visible to us. It has almost burst on two occasions - in 2000 and 2008 - but even these declines, as precipitous as they felt at the time, reached apogees that exceeded the previous perigees in1929 and 1968.

But this of itself doesn't truly establish that there is a bubble however, since as noted, even I expected to see a trend in the ratio of stock prices to consumer prices over time. Perhaps 1890-1950 was the abnormal and this is now a restoration of it?

So is there any other series that looks anything like this? Oh, let's try one at random - say, the ratio of margin debt (on the New York Stock Exchange) to GDP (see Figure 3).

Figure 3: NYSE Margin debt as percentage of GDP


OK, I had my tongue in my cheek, but again this data had even me gob smacked when I first plotted it. I had not expected this correlation: my analysis actually runs from change in margin debt, rather than its level. So this outright match blew me away - particularly when I put the two series on the same chart (see Figure 4 - and yes Alan, feel free to use this one on the ABC News!).

My causal argument commences from my definition of aggregate demand as being the sum of GDP plus the change in debt—a concept that at present only heretics like myself, Michael Hudson, Dirk Bezemer and Richard Werner assert, but which I hope will become mainstream one day. Matched to this is a redefinition of supply to include not only goods and services but also turnover on asset markets.

This implies a causal link between the rate of change of debt and the level of asset prices, and therefore between the acceleration of debt and the rate of change of asset prices—but not one between the level of debt and the level of asset prices. Nonetheless there is one in the US data, and it's a doozy: the correlation between the level of margin debt and the level of the Dow Jones is 0.945. 

Figure 4: Margin debt compared to the DJIA—correlation 0.945


Of course there are elements of spurious correlation here: they were both generally rising over 1955-2013. But one can also make a causal argument that increasing levels of debt levered up the gap between asset and consumer prices. This assertion of course directly contradicts a famous proposition in academic finance—the "Modigliani-Miller theorem" that the level of debt has no impact on the level of asset prices—which is another good reason to take it seriously.

In devising my "aggregate demand is income plus change in debt; aggregate supply is goods and services plus net turnover on asset markets" relation, I was never sure whether the measure of asset market turnover should be based on the level of asset prices, or their rate of change: this was something that only empirical research could clarify. And on this point, the US data is again exceptional: both the rate of change of margin debt (relative to GDP) and the rate of acceleration of margin debt correlate strongly with change in the Dow over the past six decades.

Figure 5: Change in margin debt & change in the Dow--correlation 0.59


The correlation of the change in debt with change in the Dow is stronger than the correlation of acceleration - 0.59 versus 0.4 - but both are pretty strong for correlations over more than half a century, especially since conventional wisdom asserts they should both be zero.

Figure 6: Margin debt acceleration & change in the Dow--correlation 0.4


The correlations have risen too as the level of debt has risen - both aggregate private debt and, in the USA's case, margin debt which is specifically used to buy shares.

Figure 7: Change in margin debt & the Dow in recent years—correlation 0.69


Figure 8: Margin debt acceleration & change in the Dow - correlation 0.6


Now comes the complex question: which causes which? Does rising/accelerating margin debt cause the stock market to rise, or does a rising stock market entice more people into margin debt? Obviously there will be some cumulative causation here: both statements are going to be true to some degree. But this also implies a positive feedback loop, which is part of the explanation for why stock prices are so volatile.

Regardless of that complex causal loop, this data scotches Greenspan and his causal argument that a rising stock market causes a rising GDP. The market - and recently the economy - has risen not because of "the wealth effect", but because of "the leverage effect". Leverage has returned to the stock market, driving up stock prices and aggregate demand in the process.

How far can it go? Margin debt is still shy of its all-time high as a percentage of GDP, so there is certainly some headroom for further rises. But at the same time, the market is still in territory that was uncharted before the Loony Zeros (my "Roaring Twenties" candidate for how we should describe the last decade and a half) drove it higher than it has ever been before. Fragility, rather than sustainability is the message I would take from this data.

I'm reassured in this prognosis by the fact that Greenspan made precisely the opposite point in that interview, when he stated that "the price-earnings ratio is at a level at which it cannot basically go down very much." As some other commentators have observed, Greenspan  expressing confidence in the stock market is a reliable contrary indicator.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.
duo's picture

I guess buying gold miners on margin was a bad idea.

ZerOhead's picture

Just a whacko idea from a self confessed nutjob here... but what are the chances that the intentional UNDER-REPORTING of inflation through changes in how the CPI is calculated might have something to do with this...

knukles's picture

Probably a lot... seems intuitive... good on ya'

Other one would be effect of survivor-ship bias in the DJIA or other equity index composition calculations.  Bias toward (significant) overstating of equity returns as the bankrupt, out of favor, obsolete, etc., fall out of favor/composition over time.

SO essentially CPI is grossly understated while equity returns are grossly overstated meaning.... 
Lemme see here...

My old equity geek bosses would have really blown a gasket at this one!
If the truth hurts, ignore it....

kaiserhoff's picture

But there are so many barriers to entry,

and it's an obsolete business model no one wants to be in anyway...

say wut?

spine001's picture

Dear Steve Keen and Zero hedgers: the associasion shown is causal but not in the way described. If you look  arefully there is a time delay between the curves. If you do a causality analysis you will see that the level of the DJI is what driv3s the level of margin debt and not the opposite. The dynamics are complex and I dont disagree that the re is an associati9n between the availability of margin debt and the level of the DJI since one feeds on the other one. But the behavioral model is a lot more complex. With respect to The Modigliani Miller theorem. The author clearly stat3s the assumptions f9r it to be valid in his paper and since those assumtions are never valid in real life "well functioning rational markets" among others, the main use of the theorem is in detrmining from a violation of its assumptions what conditions make the structure of capital relevant to valuation.

Stoploss's picture

Steve, overlay periods of US war on the Margin Debt chart, fills in the gaps of the connection between high margin debt and wars and or acts of war, I.E. 2001 for instance.

All Risk No Reward's picture

Steve, nice analysis of the chair defects on the Debt Money Titanic.

Now how about addressing the fundamental fraud of loaning a nation its money supply such that the lenders' wealth is, BY DEFINITION, society's INEXTINGUISHABLE DEBT?

Nobody seems to "go there."

I get its scary, but isn't living life hiding the "devil's toolbox" a bit scary, too?

Does every establishment and anti-establishment establishment wonk want that on their tombstone's resume?

Debt Money Tyranny

It is all fine and dandy calling Greenspan and Bernanke deceptive, but let's get to the brass tacks...  they are criminals...  serially breaking Section 2A of the Federal Reserve Act:

Weapons of Mass Debt

It wasn't a stock market bubble, it was a debt money bubble released into stocks.

It wasn't a housing bubble, it was a debt money bubble released into houses.

It isn't a stock market bubble, it is a debt money bubble released iinto stocks again.

The root cause is the Federal Reserve mega banking cartel's Trojan Horse's criminal debt bubble blowing operation!

Someone in the establishment has to man up and call these foul criminals foul criminals.

There is no doubt this is what they are - read Section 2A of the Federal Reserve Act and then look at the exponential debt growth they created contrary to that law.

spine001's picture

You are correct and the authors article is flawed in several technical levels.see my other post if interested.

dadichris's picture

Debt Money Bubbles - Made in U.S.A

RockyRacoon's picture

I'm no chartist (either in belief with this manipulated market, nor in practice as a useful tool), but what's that on the right side in Figure 1?   Is it the fabled head-and-shoulders pattern?

kaiserhoff's picture

I think that's the broke back mountain pattern, but what do I know?

(Knucks nails the jello to the wall.)

blackbeardz's picture

my chartleaves say we're at the top of the head stage, pleasurable but sensitive area.

CheapBastard's picture

The consumers ain't gonna be spending a lot any time soon, fellow. I read their unemployment benefits are getting cut from $289 by a whopping $43  bucks.


What are they going to buy on $1,000 a month (besides pop-eyes, cigarettes and Boon's Farm)?

And, oh yeah, all those rental  houses as an investment in a society where 50,000,000 are on food unemployment is "robust."


Good luck!

asteroids's picture

The stock market "wealth effect"  as a driver is another myth! The average american has no holdings in the stock market. The odds are he is living month to month. He's unemployed and feeding his family with food stamps. How the hell is this guy going to find the cash to invest in GS? Why the hell should he? The average guy understand "moral hazard", why should he put his money with a FED and Obozo that doesn't. The average guy does not believe in the reality distortion field that the FED and Obozo project.

FlipFlop's picture

M-M theorem refers to company balance sheet? In presence of tax, it does matter.

Now, does it matter if investor levers up on behalf of company?

I am suspicious of anything prof Keen writes.

spine001's picture

You are correct. The article is technically incorrect. I wonder why was it written? M&M is only useful for determining under what conditi9ns debt structure is relevant since its assumptions are never met in real life.

franzpick's picture

The biggest thing in the world to ever hide in plain sight.

AssFire's picture

I don't know...have you seen my wife's ass under her hoop skirt?

Boop's picture

Thank God, no, I haven't.

ebworthen's picture

"More cushion for the pushin'" as the saying goes.

franzpick's picture

She'll be in the running when she pleads for and starts wearing a TARP.

kaiserhoff's picture

You're married to Moochelle, too?

Oh wait.  That would be ugliest.  My bad;)

dwayne elizando's picture

Obviously Mr. Keen isn't familiar with this chart by Didier Sornette. Equity bubble shmequity bubble.

dwayne elizando's picture

Obviously Mr. Keen isn't familiar with this chart by Didier Sornette. Equity bubble shmequity bubble.

All Risk No Reward's picture

Societal asset stripping debt based money is the answer - it is the biggest fraud ever and hidden in plain sight.

Debt Money Tyranny Exposed

Al Huxley's picture

I'm betting cheap money encourages margin debt which results in the rising markets, while rising interest rates trigger deleveraging and corresponding selloffs.  Hence the observation that there's never been a serious bear market in a low interest rate environment (so much for fundamentals).  Equity bulls might want to pay a little more attention to the bond market right now....

NoDebt's picture

You beat me to it.  Would seem reasonable leverage would be used more when it's cost drops- as it has been doing with few interruptions for 30 years. 

The last month may have been the turn off the bottom, though.  Roughly 100 bps in a month should be enough to grab just about anyone's attention.


JJ McApe's picture

the titanic is unsinkable...

knukles's picture

Thank God for some truths in these uncertain times

Bangin7GramRocks's picture

401k created the big boom. If Dubya could've gotten privatized Social Security the market would have tripled again. No fundamentals, just an increase in play money to play with.

SunRise's picture

"the titanic is unsinkable! . . . once she bottoms

falak pema's picture

Has the CPI measurement changed over time, since 1890?

I would think that it has, energy for instance.

JustObserving's picture

If inflation was measured now as it was in 1980, it would be about 7% higher than the official number.

If the US govt. has been understating inflation by an average of 3% since 1980, US GDP now is $6 trillion instead of the official $16 trillion figure.  Incidentally, Dagong, the Chinese rating agency, claims that US GDP is about $6 trillion now.

falak pema's picture

you can obviously see the implication of this on the S&P/CPI ratio, in the POST 1980 period; as if the old CPI were used rather than the new CPI the ratio would be much smaller; as the denominator would be much larger.

JustObserving's picture

Exactly.  As someone said - "We are an empire now.   We create our own reality."  And we create our own fake CPI and GDP and the biggest stock bubble in history.  But is difficult to argue with 700 military bases in 130 countries.  Or the NSA collecting 97 billion pieces of intelligence every month. Or France, Spain, Portugal and Italy closing their airspace 3 hours into the flight of Evo Morales putting his life at risk.  

OneTinSoldier66's picture

I would say that's a pretty Observation of the Situation, JustObserving.

August's picture

If you're not cop, you're little people.

knukles's picture

Hah ha ha ha ha
A Chineese entity telling anybody else about fudged numbers.
Sorta like warmongers getting a peace prizez

Or sumptin', Lucy

JustObserving's picture

You mean to say you would believe Goldman Sachs over Dagong on American economic statistics?

And the most prolific warmonger in the world did get the highest peace prize.  It is a Bizarro world.

MaxThrust's picture

Sorta like war mongrels getting a peace prize.

Yes they do, Obama.

So you can see, strange things do happen

socalbeach's picture

According to John Williams' ShadowStats, the CPI has been monkeyed with since 1983, which coincidentally or not is when this article says there's been a divergence between share price and CPI (2nd graph from top - Ratio of S&P500 to CPI).

Renfield's picture

I have often wondered what would have happened had Volker not raised interest rates so high in the early '80s. I may be oversimplifying, but I've thought of this huge decades-long 'wealth effect' bubble as being the only (and predictable) result of going off the gold standard.

Today, the 'wealth effect bubble' has become such a leviathan that I don't think any interest rates can save the fiat system now, even if TPTB were in a position to raise rates. Today, central banks and governments are, I think, too insolvent to pay any real interest at all, and they can't rely on the knock-on effects of some other market's prosperity, because they are ALL now in the same boat.

If interest reflects risk, then in a purely fiat system, there can be no nominal ROI risk and so interest rates are irrelevant except as a measure of confidence rather than risk. A low-rate environment leaves a bank very little room to manoeuvre - zero being the floor, since negative returns are obviously unsustainable even in a short term - and so if low rates last a long time, and there's no apparent way for them to move up, then complete loss of confidence is close. Since the government borrows from the people, then if the people (creditor) find out that the government (debtor) is broke, they will not lend their (labour/assets) to the government any more and govvy fiat becomes useless. We are reaching the point now where people are beginning to understand that their debtor, the government, is broke and they will never be repaid no matter what nominal 'interest rates' may be.

Am I over-simplifying, or is there nothing new in this big picture, except for the 'global' size of it?

All this to say, thank you for putting some detail into the big picture of what happens when any bank/government moves to a purely fiat currency, and specifically what's happening this time.

Dick Buttkiss's picture

"I may be oversimplifying, but I've thought of this huge decades-long 'wealth effect' bubble as being the only (and predictable) result of going off the gold standard."

Yes, but it was only temporary, so don't worry:

The third indispensable element in building the new prosperity is closely related to creating new jobs and halting inflation. We must protect the position of the American dollar as a pillar of monetary stability around the world. ...

I have directed Secretary Connally to suspend temporarily the convertibility of the dollar into gold or other reserve assets, except in amounts and conditions determined to be in the interest of monetary stability and in the best interests of the United States. — President Richard M. Nixon, August 13, 1971

bonin006's picture

Even more temporary than the income tax, so far.

andrewp111's picture

The next currency system will be a unitary global all-electronic money system. This will make it possible to impose large negative interest rates and put a "color" on money because there is no way to withdraw cash and no foreign currency. The Mark Of The Beast financial system will be the only possible way to maintain the existing world order in desperate times. The ultimate in can-kicking.

Morla's picture

Capital controls do have some limitations.. As the regime becomes more oppressive, things like government cheese and victory gin become black market money. Try and ration THOSE down and you open a whole different can of worms.

If the real goal, like that of the drug war, is simply to criminalize vast swaths of the population leaving them subject to the arbitrary power of the police, then yes capital controls will do that job nicely.

ebworthen's picture

Margin debt is used to bid up stock prices.

Increasingly, millions of Americans are trapped in equities for IRA's and 401K's.

Banks and Wall Street hedge funds use margin debt (with help from the FED) to artificially value stocks higher.

Working Americans have a perception of wealth that is illusory because it is on paper - and most can't get the money out.

This trapped money (and from Pension funds) is used as the base for Wall Street profiteering.

Ramp up equities bought at a low level using margin debt, then sell them and make scads of money.  Rotate in and out of sectors and asset classes as the ramp up is occurring to maximize the amount of cash you make.

The average American is trapped in this casino, and are the ones who lose when the bubbles pop.

To add insult to injury and to try and entice them back into the equity casino and chase yield the FED pays next to nothing or zero on savings, while loaning banks all the cash they can handle at 0.5%.

Need a loan?  That will cost you 4.5%.  Have money you want to save?  We'll give you 0.25%.

The American citizen getting bent over and bled to death by the banker/crony capitalism/government cabal.

It is as plain as day if you open your eyes.

desirdavenir's picture

Totally agree...just I didn't see you message when I started typing...

Dr Benway's picture

You are absolutely correct. There are two things holding up the share market ponzi, debt and granny investor pension money. Paper prices could rise as long as these inflows were increased. But when rates have been lowered to zero, and granny funds stubbornly refuse to get fooled again, what to do.

The wealth effect always struck me as borderline insane though. So if we manipulate asset prices above their fundamentals, then the resultant economic boon from spending will be enough to pay for it all, with improved fundamentals to justify the higher prices effected through manipulation. But if the return is greater than the expenditure in this scheme it would be a perpetual motion machine.