Monday, October 6, 2008

A Scared Accountant is a Dangerous Thing

Alright, ladies ‘n germs, you might as well settle in, because this is going to be a long one. If you’re a pro in the financial world, you might want to skip to the end, where I describe the hellfire wrought by mark to market accounting. For others, the good news is that by the end of this post, you will understand, or at least know about, most of the salient aspects of the recent financial crisis. More important, you will understand why this particular financial dislocation has been so severe.

Here’s the part you probably already know: this financial crisis has been created, in large part, by a bubble in asset values and the underlying financing for those assets. In this case, the assets concerned were real estate assets, and the financing were residential mortgages, for both first homes and investment properties.

Now, economies suffer bubbles in asset values all the time. We had the stock market bubble in 2000. We had a similar real estate bubble in the 1980’s. There was the quintessential tulip bubble in Holland in the 1600’s. Often, these bubbles were accompanied by “easy money”, which led to the proliferation of unsustainable financial contracts like we have seen in this most recent bubble.

However, most of the bubbles in history did not cause the extinction of the investment banking business. This bubble has been markedly different in the severity of the liquidity crisis which has unfolded, and its effect on financial institutions.

What caused this crisis in the first place? Well, there are a lot of people who have been keen to point the finger at one particular thing or another. It was either the “predatory lenders” or the easy money of Alan Greenspan which caused it. Barney Frank is my favorite bogeyman.

In reality, there were a number of trends which, together, created this mess. Then, there was one thing which made it much, much worse than any other bubble we’ve seen, possibly since the Great Depression.

Very quickly, let me explain the conclusion, and then I will give you details later in the post.

The Things Which, Together, Contributed To The Bubble:

Relaxing of mortgage underwriting standards
Very efficient, but complex, capital markets
Low interest rates


The Thing Which, in Turn, Caused the Wholesale Destruction of the Financial World:

Mark to market accounting

Without this last item, mark to market accounting, in my opinion, we would be facing issues in our financial system not much more severe than that which we suffered during the Russian debt crisis or any number of other financial dislocations over the course of the last century.

Instead, this time, Merrill Lynch is gone. Goldman Sachs has converted itself into a commercial bank. Lehman Brothers has been liquidated. After two hundred years of being in business, Wall Street is simply gone.

If you’d like to know why, read on.

First, the bubble itself.

The Things Which, Together, Contributed To The Bubble:

Relaxing of mortgage underwriting standards: For most of recorded history, the mortgage business has been quite mundane. Banks had rules of thumb, developed over decades, which determined how large a mortgage a person could secure, given his particular financial situation. They used a ratio of mortgage payment to total monthly income, a ratio of downpayment to total value, and various other measures, and were generally quite inflexible. At some point in the 1990’s the federal government got involved in a crusade to increase home ownership, particularly among the poorer, which often meant minority, populations. In the course of doing so, the government applied pressure to financial institutions, often in the form of regulations and pressure to meet requirements of the Community Reinvestment Act of 1977, but also by simply urging financial institutions to relax their underwriting standards for minority populations. A very hot smoking gun can be found here (http://www.bos.frb.org/commdev/commaff/closingt.pdf), a 1998 manual printed by the Federal Reserve of Boston specifying the ways that banks might do such a thing, including overlooking those old standards I mentioned. The result was that banks began to relax those standards, and people who previously could not qualify for a mortgage became able to do so. In addition, truly creditworthy people started being able to get more aggressive mortgages, including no downpayment, interest-only, etc. This trend was magnified by the actions of Freddie Mac and Fannie Mae, which could (and did) purchase or guarantee the payments on certain qualifying mortgages and resell them to investment banks or other institutions.

Very efficient, but complex, capital markets: This one is little more difficult for those not in the financial industry to understand, but the development of the collateralized debt obligation (“CDO”) and collateralized mortgage obligation (“CMO”) massively increased the scale of the financial difficulties associated with the real estate bubble. In developing a CMO, an investment banking firm would purchase a great number, say 5,000, mortgages, and pool them together, creating a massive set of cash flows from the payments on the combined mortgages. Then, they would take these cash flows and divvy them up into slices, some of which were senior to the others, and thus more secure, and others of which were junior, and thus more risky, but which would pay higher interest rates. All of the slices would be rated by the rating agencies and sold to institutional investors of all stripes. The development of these financial instruments is a wonderful idea, and they will be permanent parts of our financial industry for years to come, but they added several elements of risk to the financial system which helped to create the current mess:
Complex and opaque financial instruments: CMO’s are inherently complex. Any analysis of the cash flows underlying such securities is time consuming and difficult, since the securities are an aggregate of hundreds and sometimes thousands of other financial contracts, each of which has its own particular complexities. While it is possible to estimate certain risks associated with such securities, such analysis is, at best, a guess, especially when unexpected events occur (like economic cycles or, as in this case, a housing bubble).
Distance between the originators of mortgages and the ultimate owners of mortgages (the “agency” problem): With the advent of the CMO, the entity which ended up holding a mortgage was very rarely associated with the entity which originated the mortgage. In fact, mortgage bankers, operating by themselves instead of on behalf of banks, started gaining market share in the mortgage marketplace. These mortgage bankers were incentivized solely by the number and amount of mortgages that they originated, rather than the quality of those mortgages, since they generally sold them very quickly to investment bankers eager to pool them into CMOs. The problem with this scenario is that it tended to magnify the shortcuts that mortgage originators took in the due diligence they performed on their borrowers, since they would not be the ones to suffer if the mortgage became impaired at some point in the future.
Low interest rates: With the advent of the CMOs, the increasing aggressiveness of mortgage bankers, and the ubiquity of financial information on the internet, it became increasingly easy to obtain an aggressive mortgage starting in the late 1990’s, a trend which increased into the early 2000’s. In addition to these trends, the United States (and the world) saw a very low interest rate environment early in the 2000’s as the federal reserve attempted to jump start the US economy after the recession of 2000-2001 associated with the collapse of the dotcom equity bubble. With low interest rates, borrowers started to be able to obtain even more aggressive mortgages, since there is an inverse relationship between interest rates and the amount that someone could borrow, given their income. “Mortgage Calculators” became ubiquitous on the internet, wherein a person could input various financial figures, and then, based on the prevailing (low) mortgage rates, figure out how much he could borrow to purchase a home. As people were given more and more money to purchase a home, not surprisingly, home prices continued to rocket skyward.

The situation was a bit like the old snowball rolling down the hill. Sometime in the 1990’s a bit of snow got kicked of the top of a mountain and started rolling down the hill. The federal government added a bit more to the snowball in the 90’s and into the 2000’s as it encouraged the loosening of credit standards for mortgages. It continued to get bigger and bigger as Wall Street developed fancy new ways of packaging mortgages and selling them to unsuspecting institutional investors. Finally, as interest rates achieved new lows in the 2000’s, and consequently home prices began to surge, the snowball was by now huge, and flying down the hill, approaching the bottom. People and events started to get in the way of the hurtling snowball.

Sometime in 2006, people started to notice that the default rates on the most aggressive mortgages (mostly what they call “sub-prime”) started to creep up. There are reasons why the default rates were higher than in days past, but we shall not go into them here, because it is not particularly germane to the discussion. But business continued as usual. In fact, people were so pleased with the surging housing markets, that normal Joe Sixpacks were buying up real estate on speculation, hoping that they would make a killing when they resold the property six or nine months later.

By August of 2007, however, the snowball really started bashing into things at the bottom of the hill. Default rates on mortgage portfolios had started to climb to almost unprecedented rates. Because of the opacity and complexity of the securities which actually held most of the aggressive mortgages, people in the financial world had a hard time even figuring out which securities would be affected, and by how much. In that month, trading of most types of these securities ceased altogether for a period, as people backed off to try to understand what was happening.

Investment banks, which had put together billions of dollars of these types of securities in hopes of selling them to long-term institutional buyers, were caught with illiquid securities on their balance sheets. Nobody would buy them, because nobody knew what effect the surging default rates would have on the complex cash flows of these securities. Banks could neither originate, nor package new mortgages into new CMO’s.

At the same time, the freezing up of the mortgage origination marketplace started to affect home prices. For the first time in ten or fifteen years, home prices were starting to drop, furiously in some places. The snowball was now massive, and crashing into the entire housing marketplace with a fury.

This was the situation in the winter of 2008. No bank or financial institution had yet failed, but people were getting increasingly concerned about the amount of collateralized debt obligations sitting on the balance sheets of financial institutions. In the quarter ended September 30, 2007 and December 31, 2007, banks, investment banks, insurance companies and other financial institutions started trying to estimate their financial exposure to the worsening housing crisis, and started taking charges, aggregating into the hundreds of billions of dollars by early 2008.

Trading in the most exotic of collateralized securities had virtually ceased. Since nobody had the time to figure out what they were really worth, nobody would bid for them. Rumors started to fly to the effect that certain institutions were more exposed to losses than others, and people and institutions started pulling their money out of the accounts of certain institutions. There was a run on Bear Stearns, one of the most aggressive banks, in March, and it was swallowed up by JP Morgan over a tense weekend.

Then, it just kept getting worse. By September, the federal government had to step in and save both Fannie Mae and Freddie Mac, the entire investment banking industry was gone, and Congress was forced to provide up to $700 billion for a bailout of the world’s financial institutions.

How did we go from having an asset bubble, which may have been no more significant than others in history, to the extinction of the investment banking business and governments around the world having to commit hundreds of billions of dollars to prop up failing credit markets?

In my opinion, there is one overriding reason, and that is mark to market accounting.

The Thing Which, in Turn, Caused the Wholesale Destruction of the Financial World:

Mark to market accounting:

First, let me do a little bit of back of the envelope mathematics for you. Currently, in October 2008, the aggregate default rate on mortgages in the United States is about 6%. Most people, even though they may even have a home which is worth less than the amount outstanding on their mortgage, are still paying their mortgages, because they don’t want to default! Now, 6% is a very high default rate given historical averages. It’s scary high, but let me continue, because it’s not as bad as you may think. Let’s make some obscenely aggressive assumptions. Let’s say that 100% of the defaulting mortgages are foreclosed (it would never happen…). Further, let’s assume that the original loan to value ratio of all of those mortgages is 100% (again, absurd). Finally, let’s assume that the actual market value of the homes underlying those mortgages is only 70% of the mortgaged amount (likewise, absurd). The lenders on those homes will loose 30% of their principal, ignoring transaction costs. OK, so now let’s do the math. The total value destruction so far should be the total mortgage market times 6% times 30%, or 1.8% of the total mortgage market. The total mortgage market in the United States is about $10 trillion. 1.8% of that number is $180 billion. A big number, to be sure.

But, as of today, the financial industry has taken something like $800 billion worth of write-downs, and still the carnage continues. When it’s all said and done, people are estimating that the problem might cost well in excess of $1 trillion in aggregate, and possibly many trillions.

What gives? How can $180 billion of possible losses translate into a $1 trillion of actual losses?

Well, by an accounting adjustment.

Remember those opaque CMOs that nobody could figure out? Well, those securities are holding all of the defaulted mortgages. But since they’re so complicated, and since the housing market is so dynamic at the moment, nobody can really figure out which CMOs are going to be affected in what amounts. So nobody will buy them. They sit, illiquid, on the balance sheet of the institution which happened to own them when the music stopped.

Excluding Fannie Mae and Freddie Mac, financial institutions might have several trillion dollars worth of these securities sitting on their balance sheets, and, as we have calculated before, there may be something like $200 billion of real losses hiding in there so far.

But how do these financial institutions value these complex securities which sit on their balance sheets? Do they make a bunch of assumptions about the economy and housing markets, and analyze the resulting cash flows and pick a discount rate and assign a value to them?

As of the last several years, the answer is no. Because of increasingly aggressive accounting policies which finally resulted in the adoption of something called SFAS 157, financial institutions have been required to value those securities at the price which others in the marketplace would be willing to bid for them. This is called “mark to market” accounting. When markets are liquid and functioning properly, and when institutions hold securities primarily to sell them or trade them, mark to market accounting makes a lot of sense. It is the most appropriate measure of value of a security which is held for sale.

But, didn’t we just conclude earlier that the market for these securities has become almost illiquid? At what price would people be willing to purchase complicated securities with unknown values in an illiquid marketplace? At the very least, someone would have to be given a very substantial discount to a security’s true value in order to induce them into a trade under such circumstances.

In fact, this has been the case. Whatever trading has occurred in the marketplace has been at severely discounted prices, for all of the reasons that we have discussed. The most talked about example has been the $30.6 billion worth of CDOs which Merrill Lynch sold to a private equity investor called Lone Star Financial for approximately 22% of the stated face value of those securities.

If we had to value all of the CDOs sitting on the books of financial firms at those kinds of discounts, there would be paper losses at those firms amounting to many hundreds of billions of dollars, maybe even trillions of dollars.

In fact, this is exactly what has happened, notwithstanding the likely fact that the real losses which will be incurred because of the housing crisis will be much, much less. All of this would be fine if financial institutions did not have capital requirements (regulatory, in the case of banks, and simply as a matter of creditworthiness for everyone else). Mark to market accounting has created its own, self-fulfilling prophecy. Because accountants have marked these securities down to a value far below their true, intrinsic value, credit markets have seized up as nervous institutions have stopped lending to other institutions with suspect balance sheets.

And, since institutions are not being able to borrow, based on paper losses, they are forced into liquidity crises. Often, those liquidity crises force the institutions to actually sell their suspect securities at firesale prices, well below their true intrinsic values, thereby turning paper losses into permanent reductions of capital.

This is how Bear Stears, Lehman Brothers and the others disappeared on various Sunday nights. A simple accounting treatment has outrageously caused a crisis of confidence which has resulted in an historic credit meltdown and the extinction of the investment banking business.

And it did not have to happen.

When I catch my breath, on another post I will describe how all of this could have been avoided.

PS: Understanding all this, if you want another villain, how about the trial bar? Here’s how it goes: Enron -> trial lawyers -> collapse of Arthur Anderson -> Sarbanes-Oxley -> proliferation of outrageously, ridiculously conservative accounting principles -> SFAS 157 -> extinction of the investment banking industry.

I am only half-kidding.

No comments: