Auction rate securities, a primer

Auction rate securities are a scam. There, I said it. It's a way to procure longterm financing with short term rates with the note holders assuming all of the auction failure risk. I've looked through auction rate contracts and they are very complicated and give little or no recourse to the note holder when an auction fails.

But let's step back a second and figure out how these beasts work. A company needs to raise capital. They decide to borrow money for a long term project but they want to borrow at the low one month rate rather than the expensive 10 year rate. One month rates can be 2% less than ten year rates, so who wouldn't want to do that. But there is a catch. If they borrow at the one month rate, they have to pay the money back, or refinance the debt in one month. Enter the auction rate note.

The auction rate note allows the borrower to pay a little more than the the one month rate but gets to keep the money for a longer term. A quick digression on why borrower pay more to borrow for a longer term. It's pretty simple. First suppose a co-worker comes to you and asks to borrow $20 until tomorrow ("I forgot my wallet"). If you have the money and you know the person, no problem. You'll see him tomorrow, and the next day so you can be pretty confident that you'll get your money back. If the same guy asks to borrow that same $20 but promises to pay you back in five years, you might have misgivings. Are you going to be able to find him in five years? where will you be then? And a whole lot of similar questions.

The same thing happens for corporations. Lending for a longer period is riskier. There is a good chance the company's financial health could be different, better or worse. Also, the longer you have until a bond matures, the more the price might move. Look at two extreme examples. In bond math, the duration of a bond is considered to be roughly speaking the time it takes to get your money back. For a short term bond the duration is really short, a long term bond the duration is much longer.



So now back to auction rate notes. The borrower and underwriter (investment bank) start the process like any long term note. Remember, that's where the risk is. Then they finance the transaction by issuing auction rate notes. The difference between these notes and other short term paper is that the borrower is not obligated to pay back the loan when the notes are "auctioned" each month.

When most people hear the word auction, we think of someone with a gavel selling Picassos. If the seller's minimum price is not met, then the piece isn't sold. When the financial community hears auction, it's in the context of highly liquid securities like US treasuries where everything always gets sold. Auction rate securities are like art auctions. People bring there bonds to the auction, the auctioneer (trader) sends around a marketing memo around and interested parties bid on the notes. No bids, then the owners go home with their bonds and try again next month.

But what happens now, the auction failed and the owner still wants to sell the bonds. The auction rate desk that runs the periodic auction says it's not their job so the security goes to the corporate desk who evaluates it like a longterm note, because that's what it is. And they will put a price in line with what comparable 10 year bonds are priced at.

In fixed income, prices work inversely to coupon. Suppose a company issues two bonds that are similar in every way except coupon. Bond one, H has a high coupon. Bond two L has a low coupon. The price of the bond H will be higher than bond L. So now the corporate trader will price the auction rate security (with a coupon 2% less than long term bonds) against a higher coupon long term bond and the price will be low. But the owner of the auction rate note was expecting par, and now suddenly they're getting 80-90% of par, a huge loss for a "riskless" investment.

So how did it go wrong? the problem is that the banks sold these securities with the assumption that market for these securities would remain liquid. That there would always be buyers and anytime someone wanted to sell, they could. The documents stipulate that when there are no buyers, the sellers keep the bonds and try again. There is no obligation from the borrower or the underwriter to either pay back the bonds or make a market in these securities. Also, most of these securities have a cap in the rate the borrower pays. This cap was set when times were sunnier and credit was easy.

And when these securities do trade, they trade as longer term rather than short term securities.

Leaving the owners holding the bag. A

Citibank's $12 billion dollar loans

It's easy to see this headline and think subprime loans. Don't. These are not subprime loans. They are not based on subprime loans. In fact they have nothing to do with mortgages. They are corporate loans.

And what is a corporate loan? it's a loan to a corporation. It's a short to medium term loan that is secured by the corporation's assets and comes with a set of stipulations that the corporation has to follow. Recently it has been a key part of the LBO (leveraged buy out) market that took companies of all types and sizes private. It differs from high yield bonds (junk bonds) that fueled the Drexel junk bond LBO craze of yesteryear because they are senior secured instruments.

So why was Citibank stuck with so many? Citibank, and every other investment bank, was involved in the loan market because it was instrumental to the buyout market. To participate in a large multi-billion dollar buyout, a bank or a syndicate of banks bids for the right to loan the acquiring company the money to buy it. So for example, private equity fund PEF is buying company XYZ for $30 billion dollars. They go to their bankers and say "give me your best deal!". The banks figure that out, and loan them the money and the transaction goes through.

What do the investment banks do now? well they don't want to hold the loans, they want to sell them. And usually they do, to hedge funds, bond funds and CLOs. And there was a healthy demand for the paper, so the banks could be aggressive on what they charged the private equity firms. They were waving in deals and printing paper. But there's a lag between when a bank commits it's resources and when a deal closes and in that time the credit markets ground to a halt.

So the banks were suddenly left with billions of dollars of loans clogging up their balance sheets. This is not what investment banks are supposed to do. They are supposed to buy and sell, or loan and sell. The sell part of the equation had disappeared. the buyers had disappeared. And to get buyers, the banks had to lower the price.

So now $12 billion dollars of loans on the books was work 80 to 90 cents on the dollar, or a $1.2 to $2.4 billion loss.

So the banks are losing money on good assets. Not to be confused with the subprime crisis, where the banks are losing money on bad assets. How can you lose money on good assets? buy at the wrong time.

Plenty of people bought tech stock before the bubble burst. Some bought brain freeze stocks like pets.com, but lots of people bought good solid companies for more than what they should have. They were supposed to keep going up, they went down and money was lost. The companies were still good and are still in business now (look at Oracle for example).

So Citibank is just cleaning house. This frees up some capital, allows them to take a loss and move on.

And the brilliant part of it is that the private equity firms, the very people that borrowed the money, are buying the debt. Not all of the debt, but a good deal of it. They're doing this because of three main reasons. The first, is that it's a good deal. They like the company, believe it's going to get bigger and better and will make money in the long run. They already own the equity (an LBO buys the equity of a company), so why not the debt too. And at a discount. And as the company improves, and the credit market improves, the debt will become more desirable and will be traded.

The second reason is more pragmatic. Private equity firms need the investment banks to raise money. It might not be working now, but it will in the future. People have long memories.

The third reason is a pure CFO play. The loans were made in a very aggressive market and are very low. If they were refinanced, it would be at a much higher rate. By buying the loans, they are helping out the acquired company's balance sheet.

So where did the credit problem start?

This is the question I get asked the most. Akin to "who's fault is it?". It's also a pretty hard question to answer. Well, not hard, but it's not a short answer. It started in the mortgage market and it started a couple of years ago. First let's time travel to when things were still good. That would be 3 or 4 years ago and people were buying and selling real estate, mortgages were easy to come by and values only went up.

1. Mortgage Backed Securities

Mortgages are not the simple transactions they used to be. Back in the stone age, you'd go to a bank, fill out an application and hopefully get a loan to help buy the home. The bank would take that loan and either keep it on it's books or sell it to one of the pseudo government agencies (FNMA, FHLMC or GNMA).

And then there a couple of innovations happened that radically changed the playing field. First off, banks started selling their loans to investors, other banks and special purpose vehicles (SPVs). This got rid of the loans, with the associated risk, off of their balance sheets and allowed them to make more loans. From here investment banks started selling pools of loans, then selling structured pools of loans (more about that later) and then all sorts of structures that appealed to every investor.

Now savings and loans and commercial banks (the originators) didn't have to worry about the borrower paying back the loan in 30 years. All they cared about was getting through the next six months and it was someone else's problem. And the investment banks (who securitized the loans) felt comfortable with their risk exposure since they just packaged together the loans and sold them as new asset backed securities (ABS securities). And the buyers of these securities felt secure with their investment because they had the backing of the rating agencies and thought that were being adequately compensated for the risk that they were exposed to.

So everyone was happy. And everyone was making money, so more people got into the business. With more people in the business of making loans, the amount of money made on loan went down, so volume had to go up. The housing market cooperated, prices kept going up, so people used the equity built up in their houses to fix them up or move to a bigger and more expensive houses.

And then lenders started making more aggressive loans to people with worse credit. Since the market was great and defaults were low, there was not much to differentiate good risk from bad risk. So banks could lend money at a much higher rate, but without the same increase in risk. Remember, this is when housing prices just went up. When a borrower got in trouble, more likely than not they could sell their house for more than they owed. The default risk was being masked by a fired up real estate market.

And all these loans got funneled through the system. Investment banks couldn't get deals done with good loans because there wasn't an "arbitrage", or a way to make money, with loans that had low interest rates. So the investment banks started adding riskier (with higher interest rate) loans into the pools.

And then the cracks began to appear. At first it was loans that went bad before they could be securitized. The investment banks had the right to send them back to the originators, who would buy them at cost. The originators always kept a reserve available for a certain percentage of loans going bad, but suddenly that reserve was blown away and the returned loans kept coming. This is when the investment banks started buying up all of the subprime issuers, look at MortgageIT sale to Deutsche Bank for example.

Meanwhile at the investment banks, there was a push to create indexes based on the
ABS securities. These indexes followed a set of 20 ABS securities, picked by committee, that gave a representation of the whole marketplace. Every six months the reference securities were changed so the index referred to current deals rather than old deals. Investors could buy ("long") or sell("short") the index. If you thought that the market was going well, you'd go long. If you thought the market was heading into the sewer, you'd go short.

Lots of people went short. Goldman Sachs made a huge amount of money by going short.

Everyone wants to short the index, hardly anyone wants to go long. Since the index is based on actual bonds, then the implied price of those bonds is also tanking. Now one more step, if you owned a bond that looked just like a bond in the index, then the "implied" price of that bond was also tanking.

So then end of month prices come out. And they are horrible. And the collateral reports on the loans come out, and they are horrible and the rating agency monitoring reports come out and they're horrible too.

And that's the short story.

Introduction to the blog

So here it goes. I've worked in the credit markets (what used to be called fixed income) for more years than I care to admit, most recently originating CDOs. Everyone I know asks me what the heck is going on. I try to explain but it's a complicated and long story that tends to both bore and confuse the listener. There's no one bad guy. It's a complicated story that starts years ago and is just exploding now. It involves mortgage backed securities, credit derivatives, complex structures, rating agency models, mortgage bucket shops and a host of other large and small players. Oh and lax oversight, that too.

So I thought that maybe if I tried to write it down, organize it a bit, the story will come together and I won't spent fifteen minutes boring people trying to explain what's happening.