I have espoused this theory at various points, and he agreed with it. But we came to a sticking point: what about the stock market bubble? Debt certainly rose dramatically starting around the same time, but VCs were long money, not superleveraged hedge funds. Households were obviously tapping quite a bit of credit, but that was unsecured credit or mortgages more than margin loans, so the feedback loop is considerably attenuated. And the bubbliest companies weren't using debt, because they didn't have any cash flow.
In the Financial Times (registration required), Frederick Mishkin argues that not all bubbles are created equal:
Are potential asset-price bubbles always dangerous? Asset-price bubbles can be separated into two categories. The first and dangerous category is one I call "a credit boom bubble", in which exuberant expectations about economic prospects or structural changes in financial markets lead to a credit boom. The resulting increased demand for some assets raises their price and, in turn, encourages further lending against these assets, increasing demand, and hence their prices, even more, creating a positive feedback loop. This feedback loop involves increasing leverage, further easing of credit standards, then even higher leverage, and the cycle continues.
Eventually, the bubble bursts and asset prices collapse, leading to a reversal of the feedback loop. Loans go sour, the deleveraging begins, demand for the assets declines further and prices drop even more. The resulting loan losses and declines in asset prices erode the balance sheets at financial institutions, further diminishing credit and investment across a broad range of assets. The resulting deleveraging depresses business and household spending, which weakens economic activity and increases macroeconomic risk in credit markets. Indeed, this is what the recent crisis has been all about.
The second category of bubble, what I call the "pure irrational exuberance bubble", is far less dangerous because it does not involve the cycle of leveraging against higher asset values. Without a credit boom, the bursting of the bubble does not cause the financial system to seize up and so does much less damage. For example, the bubble in technology stocks in the late 1990s was not fuelled by a feedback loop between bank lending and rising equity values; indeed, the bursting of the tech-stock bubble was not accompanied by a marked deterioration in bank balance sheets. This is one of the key reasons that the bursting of the bubble was followed by a relatively mild recession. Similarly, the bubble that burst in the stock market in 1987 did not put the financial system under great stress and the economy fared well in its aftermath.
Sounds convincing . . . but I can't help wondering why credit and stock market prices were rising at the same time.
You are right to wonder. The housing bubble started in the mid to late 90s along with stock bubble, not after the stock bubble burst. And remember, the Dow and the S&P went on to match or exceed the old highs in 2007, at least in nominal terms.
If I were going to interpret Mishkin, I would argue that he and other court appointed economists are laying the groundwork for the new bubbles they are creating. Has anyone else noticed the inverse correlation of the dollar and the stock market lately?
How about this theory:
Strong gain in the stock market improve consumer confidence, as people believe they are either richer from their investments, or will benefit from a stronger economy even if they have no investments.
Improved consumer confidence leads to more debts, as people feel richer and believe they can afford to borrow more.
Perhaps the difference is simply the size of these (excess global savings-induced) bubbles?
What I mean is this: as the global savings surplus increases beyond the size it would be in a free market (ie., if China allowed its currency to float and its consumers to buy what they want), it will naturally be increasingly difficult for said savings glut to find a decent return.
Well, to make a long story short, the size of the Chinese economy by 2007 was about double what it was only back in 1999 or so, and vastly bigger than it was in the late 1980s. So, a (say) 25% savings rate in 2007 translates into a lot more excess cash than it did in the late nineties, and a big increase in debt.
So, the global savings glut simply grew ever vaster, and ultimately this inevitably led to dangerous increases in the leverage positions of borrowers.
The mystery to me isn't so much soaring debt, it's the absence of soaring equities markets.
Yes, it is borrowing on the back of unsustainable rises in the price of an asset that creates the accumulation of risk that becomes a threat to the system. That did not happen in the dot.com bubble because everyone treated the individual dot.com securities as speculative and therefore poor collateral.
The lesson for future financial regulators seems simple and easy to apply. If the prices of an asset class begin to rise outside historic relativities with other prices in the economy, the regulator simply needs to notify the financial agents that this asset class is therefore poorer collateral than it was; and the regulator will treat it as poorer collateral for all purposes of the asessment of the solvency and liqidity of financial institutions.
This measure will not stop asset price rises, many of which will later collapse. However, it will stop most of the damage to the financial system; it should stop the consideration of using economy-wide tight money policy to restrain a relatively minor segment of the economy; and it should finally eliminate the idiotic controversy over whether it is possible to reliably identfy market bubbles in their early stages.
OTOH, DeSoto has shown pretty conclusively that countries with very little leverage are almost by definition the poorest countries.
I'm still not sure I buy the argument the bailouts were necessary to avoid systemic collapse, or could not have been avoided by breaking up these "too big to fail" entities. The 1987 stock market collapse and later dot-com collapse were devastating but involved smaller players with less political clout.
I can understand bailing out depositors, even above FDIC limits. But there seems to be no reason (other than corruption) to bail out the executives, stock holders and bond holders.
Not all bubbles are caused by excess debt, but rather by bad choices. What excess debt does is make bad choices much more dangerous. This is why the tech bubble crashed the economy in a very small, and this crash affected it hugely.
Lets just say the roof on my house is leaking so I climb into the attic and notice that the beams in one corner are damaged and sagging which is causing the leak. So I go into the opposite corner of the attic and remove sturdy beams from a solid portion of the roof and use them to repair the damaged section and the leak is fixed from a short term point of view. Each time the roof leaks I go up and swap beams around to fix it and each time it works. I get more and more confident that I have discovered the solution for keeping my roof in good condition in perpetuity as I have more and more evidence that beam swapping fixes leaks. Of course any outside observer would quickly realize that someday the amount of beams that need replacing will exceed the number of good beams left that can be moved and that this approach is doomed to fail. A second observation is that each individual leak is likely to be bigger than the last as the overall long term health of the roof deteriorates. The lesson is simple and clear, if you view each roof repair as a separate and independent event and draw conclusions from those independent events you will eventually have an outcome which is completely opposite, and fully not understandable, from your working theory.
If you view each bubble separately and assume each fix that is undergone has restored the economy to full health because your short term measurements (GDP growth) tell you so you will eventually be extremely shocked to find out that previous fixes suddenly don't work despite a consistent history of having them work. Not only that but each bubble will appear larger and more damaging than the previous one- that is until the 'roof' so to speak is replaced.
I won't ask you to take it on faith either. Lowering the federal funds and discount rate targets have been the primary tools for fighting recessions in the US since Volcker pushed those rates to historical highs in the early 1980s. During the past three decades the general trend has been to lower rates to stimulate the economy during periods of slow growth or recession and to raise rates after growth has returned. The catch is that with one very minor exception* every time the rate has been raised it has been raised LESS than the previous reduction. Each recovery that has been encouraged in this way has taken the federal reserve one step closer to the infamous liquidity trap, and along the way the bubbles have become larger and more ingrained in the real economy.
If you primarily measure the health of the economy by changes in GDP (or UE or any measure that emphasizes short term changes) then you would have totally missed that each recession (or threatened recession) has cost the federal reserve ammunition which makes each successive recession harder to fight (especially when measured as "ammunition required" as a % of "ammunition" left.
* the federal funds rate had a short term peak in 1995 of 6, was dropped to a low of 4.75 in 1998 and was raised to 6.5 in 2000.
Pretty Bubbles...What kind is this ?
Financial Times
Mother of all carry trades faces an inevitable bust:
Rally in risky assets.
http://www.ft.com/cms/s/0/9a5b3216-c70b-11de-bb6f-00144feab49a.html
You think all the people buying internet stocks at inflated prices were doing it with real money? It was all on margin. It didn't nuke the banking system because it wasn't banks buying these stocks. Make no mistake: the internet bubble was as debt-fueled as the housing bubble. Just look at the money supply stats leading up to the crash.
The leverage in a retail brokerage margin account doesn't seem to be anywhere near what you get in a mortgage loan though.
I think what made this crisis the perfect storm was that leveraging up came from three distinct groups.
First, the home buyer. In their case, it wasn't just a matter of taking out a loan to buy a home, but the insane products out there that I suppose, in a certain sense, is liking buying stocks on margin, e.g., interest only loans, option ARMs, 120% financing, etc. So home buyers weren't just leveraged up, but were massively leveraged up (or perhaps another way of looking at it is that there's the prudent way to leverage up (slap 20% down) and then the imprudent way (e.g., go NegAm big time).
Second, the financial institutions that invested in MBSs and CDOs, often at 35-to-1 leverage ratios or greater.
Third, the banks that are over concentrated in commercial real estate (residential real estate was less of a problem for banks since much more of those mortgages were sold on the secondary market and became a problem for the second group above).
If any one of these groups develops balance sheet problems, you might have a recession. But when all three groups are massively leveraged up, and all that leverage is tied to the same asset (real estate), and the asset collapses so that all three groups develop balance sheet issues, it seems you get our current predicament (the Great Recession, or will it be remembered more as the Great Stagnation if we have a lost decade type situation?).
Megan,
Stock can only be bought on 50% margin and mortgages could be mortgaged 90% plus. That is the main answer to your query at the post's end. It's the degree of leverage and concentration of risk that cause the problem.
The real issue is whether there are large institutions that are holders of bubbled assets that are themselves highly leveraged, because that begins the process of transmitting the problem from the asset class itself to the rest of the real economy through a cascade of defaults in the financial intermediary system.
If you have thens of thousands of individuals holding NASDAQ stocks in January 2000 even on 50% margin, the collapse is diffused throughout the economy and over time as they will not sell all at the same time. But if you have three institutions holding a trillion dollars of bubbled assets with only $100 billion of equity supporting it, you have the potential for $900 billion in defaults in a very fast time frame and that is where the problem comes in.
You are all missing the major form of leverage employed during the stock market bubble.
Inflated internet stocks were being used to purchase real world assets, including other companies and equipment.
The valuation of these stocks were based on extremely small floats, so the leverage employed was from a valuation of the entire company based on a small amount of available equity. These companies would then use the remainder of the stock as a way to purchase some real world goods. Effective leverage was 20 to 1 or better in many cases.
This of course was much smaller than the total amount of money available in real estate, but it was large enough to dramatically impact the real economy when the spending went away.
I should mention that I am not surprised that Mishkin missed the form of leverage being employed in the stock market bubble. He has proven to be less than insightful on a variety of topics.