A Review of the 2008 Financial Crisis
Bridgewater is focused on understanding the historical cause-effect relationships of complex economic situations. As we look back at the events of 2008, we invite you to explore some of the crisis-era research and decision making that enabled our investment team to understand what drove the market and helped us weather the greatest financial storm of a generation.
Identifying the Signals of a Crisis
Long before the events of 2008, Bridgewater's investment team set in place the tools needed to recognize and understand signals of economic turmoil. Hear firsthand from key members of the investment team about how this framework led to insights and critical decision making:
How Bridgewater Navigated the 2008 Financial Crisis
The story of Bridgewater's journey through the financial downturn, as told by the people who led the way.
A Conversation with Ray Dalio
Bridgewater founder Ray Dalio and senior portfolio strategist Jim Haskel flesh out some of our key research around debt crises, and apply it to what’s happening in markets today.
Download Our Book: Big Debt Crises
Principles for Navigating Big Debt Crises provides a unique template for how debt crises work and the principles for dealing with them well.
The Crisis, As It Happened
Revisit the events of the 2008 financial crisis through the research and recollections of the Bridgewater investment team:
Setting the Stage
Zooming out, the last big debt crisis in the US was in the 1930s. At that time:
DebtsDebts had grown much faster than incomes and debt burdens were high.
Interest RatesInterest rates hit zero in the downturn, and could not be lowered sufficiently to prevent a depression.
Money PrintingTo end the depression, policymakers abandoned the gold standard and printed money.
In 2008, the US economy once again faced a similar set of challenges:
DebtsFrom the close of World War II all the way to 2008, debts grew faster than income, setting up what became a debt squeeze.
Interest RatesThe latter stages of that leveraging occurred alongside a secular decline in interest rates, beginning in the 1980s. In 2008, interest rates hit zero and could no longer be lowered enough to relieve the burden of now very high levels of debt.
Money PrintingThis required that one very important part of the ultimate resolution of this situation was the printing of money through quantitative easing.
Managing money through the 2008 financial crisis felt in many ways like trying to navigate our way through a blizzard—a flurry of information would hit us day after day that we had to make sense of in order to be able to react well to the evolving conditions. Living through that experience reinforced the value of having a template to help us understand what was going on. While most people have not lived through many big debt crises, these crises have happened many times throughout history. By studying them we were able to understand their commonalities, allowing us to process the dynamics unfolding in the economy and markets in the context of how big debt crises typically play out. While there were many moments along the way where we did not know what would happen next or were wrong in our assessments, our template was like a map that helped us figure out which path we were on. This approach helped us enormously, enabling us to manage money well and keep our clients’ money as protected as we could, despite the extreme market environment.
In the timeline below we’ll revisit our experience living through the financial crisis through the lens of the research that we were carrying out at the time. This will give you a window in to the process we followed to improve on our mental map for how a crisis of this sort typically plays out—a map that we began building decades earlier, as we managed money through the emerging market debt crises of the 1980s and Japan’s bursting bubble and depression of the 1990s. While this is by no means a comprehensive account of the 2008 crisis (and a more in-depth treatment is available in Section 2 of the book), it will give you a sense of how we processed the arc of events as it unfolded.
Getting High on Stimulants
The risk of an economic crisis in the US was on our minds in the early 2000s. In the years after the dot-com bubble burst, there were many signs that the US had been close to entering a depression. The Federal Reserve was ultimately able to successfully engineer an economic recovery by lowering interest rates to stimulate borrowing and spending. But interest rates had come dangerously close to zero, and the economy had been very slow to respond. Moreover, since 1980, each recession had required progressively lower interest rates than the last to stimulate a recovery. Those ever-lower interest rates had encouraged borrowing, causing debts to rise higher and higher relative to incomes and making the economy more sensitive to tightenings and less responsive to easings. There was a real risk that interest rates would hit zero in the next downturn and standard monetary policy would become ineffective. We had seen this happen in Japan, where policy makers were still struggling with the aftermath a decade after its bubble had burst. We explored these risks in 2002:
Most people, especially Americans, are hooked on credit. Make it cheap and readily available, and they’ll consume it abusively. So, history suggests that reversing a recession is easy (i.e., has always happened), provided interest rates can be lowered (i.e., aren’t close to 0% before easing begins)…
Because a) the great economic risk increases as we approach a 0% interest rate (because debt service costs can’t be reduced and new consumption can’t be stimulated), b) the next cyclical low in interest rates will probably be below the last cyclical low and c) the last cyclical low in the Fed funds rate was 1 ¾%, the risk of hitting 0% and having a really bad economy will be much higher during the next recession than during the last one (i.e., the one that just ended).
The Risk of ‘Depression’
Since the very beginning of Bridgewater, we have believed that the most effective way to learn and compound understanding is through a systematic process. That is to say, when we discover insights about how the world works, we explicitly specify our understanding into a set of rules that can be debated, stress tested, and programmed into our investment systems. We have found this approach very helpful, even when we are dealing with things that are inherently hard to measure. As we improved our understanding of depressions, we followed the same process of systemizing our learnings. In one of our Daily Observations from September 2002 (“The Risk of ‘Depression’”), we wrote:
Depressions occur when interest rates can’t fall (e.g., because they can’t go below 0%), so debt service burdens can’t be relieved, so asset sales have to occur to raise cash… Since bubbles consist of essentially the opposite dynamic (rapid debt growth causing asset values and GDP to be exceptionally strong), it is after a bubble, when debt burdens are high and short-term interest rates are close to 0%, that concerns are most warranted.
Because the relationships that drive the economy and markets in depressions are so radically different than we are used to at other times, we have developed a depression risk gauge to alert us to this risk. While not precise, this proprietary measure signifies that the risk of a deflationary depression is relatively high (above about 25%)… This gauge is now signaling caution.
Monetary Policy in the Great Depression
Big debt crises and depressions have happened many times before. In order to be prepared to navigate one, we revisited past cases (such as the Great Depression and the ongoing depression in Japan) to deepen our understanding of how they typically work. One of the areas we wanted to understand was the effectiveness of monetary policy. We published some thoughts on this topic in November 2002:
The key is that a depression, in our view, is not the same as a long and large recession, but is rather a different phenomenon entirely. Normal economic linkages are maintained in recessions, but break down during depressions… In both [the US in the 1930s and Japan’s current deflationary depression] monetary policy (falling interest rates) failed to stimulate economic growth or financial markets, leading to an unmanaged economic contraction in which central bank control is very limited.
Dangers of the Present Value Effect and Easy Leverage
The big rate cuts after the tech bubble crash had stimulated borrowing and spending, but they also set the stage for the subsequent bubble, which grew most rapidly between 2004 and 2006.
During this period, US economic conditions looked excellent by most measures. But not enough attention was being paid to the role leverage was playing in driving strong economic conditions and rising asset prices. Rising housing prices had improved household net worths, and with credit easily available, households borrowed extensively and debt rose rapidly relative to incomes. But much of the “wealth” that was being created was likely an illusion, setting the economy up for problems down the road. We elaborated on this point in our research:
People… are comfortable borrowing and spending more based on their high net-worth, since they feel that they’re worth more. However, most of what we are worth doesn’t yet have much value – it is the discounted present value of the expectations of what will be earned in the future… treating the present value of future expectations the same way as current wealth encourages one to borrow and consume wealth that has yet to be created. Said differently, we are essentially spending tomorrow’s earnings today… the “present-value effect” and the availability of leverage together have created a much riskier environment and a lot of current consumption at the expense of future consumption.
It is interesting to us that the amount of debt in the economy is so large… at the same time as the credit spreads are so low… We think that the longer-term risks (i.e., those in the 2 to 5 year forward period) are very large.
Growing Financial Risks
In early 2007, the bubble approached its top. At this point, hardly anyone was concerned because both the markets and the economy were still doing great. In our Daily Observations of February 5, 2007 (“Growing Financial Risks”), we wrote:
At this time the markets are discounting the lowest risks in decades, yet we believe that the imbedded risks in the system are quite large… It seems to us that money is now being thrown at financial instruments like it is being thrown at the art, jewelry and high-priced real-estate markets. Prices of risky assets, particularly those with positive carry, are being driven up, and yields/carries are being driven down, making expected future returns low. Simultaneously volatility has shrunk; as a result, low volatility is being assumed to continue and reaching for yields has caused increased leverage to be employed in order to try to squeeze more return out of the puny spreads/carry trades.
…Though we haven’t done enough research on this to say for sure what impact such a crisis would have on other financial institutions and corporations, it appears to us that the financial system is similarly not overly exposed to a moderately bad market/economic environment, but is seriously exposed to a really bad environment. The size of derivative exposures outstanding is enormous. Specifically, the total amount of outstanding OTC derivatives as of the end of last year was $262 trillion (yes, that’s with a t), which is about 5 times what it was in 2000… Our gut level view is that these positions would produce problems if there was a big expansion in credit spreads and liquidity premiums.
Is This the Big One?
By June 2007, tightening pressures had flowed through to the first broad sign of financial distress: rising foreclosures and delinquencies started to translate into meaningful losses for bigger banks. Two hedge funds run out of the investment bank Bear Stearns that invested in subprime mortgage-backed securities faced growing losses and a wave of investor redemptions; the funds would eventually collapse in July. Economic growth still looked good, as the debt and tightening conditions hadn’t yet passed through to the economy, and US equity markets hit new highs in mid-July. But the bubble had already begun to unravel. On July 26 we wrote:
[We] believe that interest rates will rise until there is a cracking of the financial system, at which time everything will reverse (i.e. there will be a move to focusing on fear from focusing on greed, volatilities will increase, and carry and credit spreads will blow out). We had (and now have) no idea exactly when this will occur and if what’s happening now is the big one. We just know that 1) we want to avoid or fade this lunacy and 2) no one knows how this financial market contagion will play out.
…A few months ago we undertook an extensive study to see which market players held what positions, especially via the derivatives markets. So we read all studies by government overseers and financial intermediaries, we gathered and examined all data we could obtain, and we delved into 10-K reports of financial intermediaries. And we concluded that no one has a clue. That is because one can only vaguely examine these exposures one level deep… But we do know that these exposures have grown rapidly (about four times as large as five years ago) and are huge.
A number of hedge funds struggle and some are forced to liquidate as housing market weakens
The Really Big Picture: Not Just a Normal Recession
At the start of 2008, cracks began to appear in the economy and markets, as weak economic data led the Federal Reserve to cut interest rates. Many at the time thought that the economy was in recession. Our research at the time described how we saw a different process at work, one interest rate cuts might be unable to reverse:
We believe that we are approaching the end of an era – i.e. the end of the post-1981 period that was characterized by falling interest rates, falling inflation rates and strong growth… The “R”[ecession] word has been used a lot to describe the possible contraction in economic activity because all contractions are now called recessions. However, to use that term to describe what’s happening would be misleading… A “D”[epression] is an economic contraction that results from a financial deleveraging that leads assets (e.g. stocks and real estate) to be sold, causing asset prices to decline, causing equity levels to decline, causing more forced selling of assets, causing a contraction in credit and a contraction in economic activity, which worsens cash flows and increases asset sales in a self-reinforcing cycle… Equity levels fall relative to debt levels despite interest rates declining while credit spreads widen until risk free interest rates fall to 0%, and monetary policy ceases to work.
Bear Stearns acquired in fire sale by JPMorgan Chase
The Loan Losses Are Still to Come
Conditions continued to worsen in early 2008. In March, Bear Stearns, a systemically important investment bank, was sold to JPMorgan Chase at fire sale prices. In the wake of this near-collapse of a major financial institution, one of the big questions was how much deeper the losses might be across the financial system. We wrote:
Financial institutions have gradually come clean about the losses they sustained on derivatives and securities of many kinds… the losses from the old way (bad loans) are just about to come to a head… Loans don’t go sour until the cash flows go bad, while markets go sour in anticipation of problems… Given how easy credit conditions were, how strong the economy had been, and how quickly both those turned, the problems in the loan books will likely be vast.
Entering the Second Stage of the Deleveraging
By mid-August 2008, markets appeared to be stabilizing, helped by steps from US policy makers to shore up confidence in the financial sector. These measures, however, were insufficient to address the underlying credit problems that were becoming only more pressing. As we described:
While the Fed did a great job of providing liquidity where it reasonably could, the accounting adjustments (e.g., allowing losses to be written down over several years) weren't made, so we are approaching a solvency crisis that we think is about to result in an avalanche of asset sales. So now the question is whether they will create a safety net in time to catch these assets so that they don't crash and bring down the financial system and the economy with it. Frankly, we think that this will be a race to the wire… Without a safety net, the knock-on effects will extend way beyond the holders of these assets. They will extend to virtually everyone as all forms of equity decline in value, capital availability contracts and the real economy weakens. So I think that we are approaching an inflection point - i.e., that either the policy makers will move or we will have a worsening of the crisis.
Where We Are Now
On September 15, Lehman Brothers declared bankruptcy. As Lehman was teetering, policy makers were struggling with whether and how to provide a bailout, but did not yet have programs in place to handle a default. The lack of clarity around counterparty and credit risks, coupled with the first-order impacts of the bankruptcy, dramatically accelerated the process of forced asset sales and triggered seizures of the financial system. We warned at the time:
With interest rates heading toward 0%, financial intermediaries broken, and the deleveraging well under way, it appears that we are headed into a new domain in which the classic monetary tools won’t work and the Japan in the 1990s and U.S. in the 1930’s dynamic will drive things.
Congress passes bill creating Troubled Asset Relief Program (TARP) to handle toxic assets, after a failure to pass a bailout bill on September 29 sparked a major equity selloff.
US policy makers launch a huge variety of programs in the wake of Lehman’s collapse, including the AMLF, CPFF, MMIFF, and many others, to address the variety of issues in the crisis without achieving full resolution.
A Template For Understanding What’s Going On
Less than two months after the Lehman bankruptcy, amid continued extreme market volatility and ongoing attempts by policy makers at the Federal Reserve and the Treasury to control the crisis, we took a step back to provide perspective on the economic forces we saw at play. We described our goal as follows:
Rather than just examining the D-process, in this report we would like to convey our big picture template for understanding economic movements (of which the D-process is a part). This template consists of three big forces – 1) productivity growth, 2) the “long wave” cycle and 3) the business/market cycle. We believe that if you understand these forces and how they interact, it will go a long way towards helping put what has happened and what is likely to happen into perspective. We also hope that it might help your strategy for the future.
Why We Expect More Shock & Awe and What It Will Mean
On November 25, the Federal Reserve and the Treasury announced $800 billion in lending and asset purchases aimed at pushing down mortgage rates (to help the housing market). This was their first quantitative easing (QE) program. This was a classic and critical step in managing a deleveraging. Central bankers in the midst of crises are forced to choose between 1) “printing” more money (beyond what’s needed for bank liquidity) to replace the decline in private credit, and 2) allowing a big tightening as credit collapses. They inevitably choose to print, which is when things change dramatically. They began to do so now. We wrote at the time:
Today’s Fed’s announced moves are just the latest steps down the path of continuing to broaden the securities bought and increase the amounts spent to bring down credit spreads and add liquidity to the system. We expect more because we expect they will do “whatever it takes” that they can get away with. So we expect similar big steps in the future, including both the obvious ones (buying long term bonds and perhaps buying stocks) and the less obvious ones (getting dollars in the hands of entities that are unimaginable, probably indirectly – e.g., countries that aren’t politically acceptable via the IMF – and dropping dollar bills out of airplanes).
Federal Reserve begins purchases of Treasuries and expands MBS purchases
Can Increased Money Make Up For Falling Credit?
By the spring of 2009, policy makers had begun to produce a massive easing of fiscal and monetary policy, which was needed to stabilize the financial system and stimulate the economy. March 2009, in particular, brought aggressive easing programs, designed to provide the money needed to make up for contracting credit. First the Federal Reserve significantly expanded their quantitative easing program, increasing their MBS purchases and adding purchases of Treasury bonds. A few days later, the Treasury Department announced expanded purchases of troubled assets from banks. We were cautiously optimistic about the new policies, writing:
What the Fed (and other central banks) is doing this time that is different than what has been done before in that it is acting faster and more aggressively in increasing money to offset decreasing credit. So, the outstanding questions are 1) to what extent can money creation replace credit creation (e.g., what will happen if debt growth doesn’t pick up yet money growth increases by more than enough to offset it; or, how well can we get along without credit creation) and 2) will this money growth lead to credit growth or will it lead to the inflation of inflation hedge assets…
These shock and awe counter-attacks contributed to big rallies in stocks and produced increased optimism. It should be noted, however, that in other depressions big policy moves from other governments’ counter-attacks produced similar big rallies and waves of optimism that were repeatedly disappointed by continued deterioration in the real economies. So, the lesson to be learned is: don’t get optimistic because of stock rallies until they are confirmed by improved actual credit and economic activity.
Following this wave of monetary and fiscal easing, the economy showed signs of stabilizing, and markets rebounded sharply in April 2009. One key question for the strength of the recovery was the health of the banks. Despite recent drips of good news, there wasn’t broad transparency on whether the banks were still encumbered by toxic assets or a big need for capital. In early May, the Federal Reserve released the results of its banking stress test. We wrote:
For the first time in the last two years we are confident that the regulators really do understand the scale of the banking problem! Our only concern now re the banks (and it’s not a big one) is that they will focus too much on the amounts of capital the banks will have to raise ($75B in common stock) to meet the regulators’ requirement rather than look at the amount they have to raise to support lending in the amounts that the economy needs to be healthy.
A Beautiful Deleveraging
While policy makers’ actions had successfully produced an economic recovery, in the aftermath of the crisis debt burdens were still extremely high. For the economy to heal, debt levels would have to be brought down through time. In 2012, we laid out our thoughts on how such a deleveraging process could be managed so as to be “beautiful” and took stock of how policy makers had managed this process in the US:
The deleveraging process reduces debt/income ratios. When debt burdens become too large, as is now the case in debtor developed countries, deleveragings must happen. These deleveragings can be well managed or badly managed. Some have been very ugly (causing great economic pain, social upheaval, and sometimes wars, while failing to bring down the debt/income ratio), while others have been quite beautiful (causing orderly adjustment to healthy production-consumption balances in debt/income ratios)…
Like the US deleveraging in the 1930s, the lead-up [to the current deleveraging] consisted of a debt-driven boom, and the deleveraging has transpired in two stages: a contraction in incomes followed by reflation and growth. However, because of a swift policy response from the Fed, which was prompt in guaranteeing debt and aggressively printing money, the contractionary period only lasted six months (versus over three years in the 1930s), and since then there has been reflation and debt reduction through a mix of rising nominal incomes, default, and debt repayment.
Following the reflation that began in March 2009, incomes recovered, debt burdens fell below their initial starting level to around 335% and stocks recovered all of their losses. At this time, the credit markets are largely healed and private sector credit growth is improving. Thus far, this deleveraging would win our award of the most beautiful deleveraging on record. The key going forward will be for policy makers to maintain balance so that the debt/income ratio keeps declining in an orderly way.