Monday, March 22, 2010

Why Did We Keep The House?

(I am just now finishing up this post, which I realize has been sitting half-written for months now.)

In my previous post I tried to at least enumerate the costs of renting the house rather than selling it. The rent checks have to cover not only the mortgage but also property tax, maintenance, and return on equity.

(That last one is often overlooked: if you somehow still have positive equity in your house, then if you don't make at least some profit from renting, you're missing the money you would have made by selling and gaining interest on the cash you'd free up.)

And the short answer is: we are probably slightly behind when renting - that is, the cost of property tax, maintenance, mortgage, and a conservative return on equity* are more than the rent minus property management commission. (With the initial costs of getting the house ready, we're definitely behind but that's perhaps to be expected.)

So why did we keep the house? The short answer is: selling it would have been a lot uglier.

Before going on to the specifics, I have to rant about real-estate agent fees. The transaction costs to sell a house are 6%. I can't come up with words to describe how obscene that is. (Even if I could, I'm told that my mother is linking to this blog for the pet pictures, so I'll keep the language tame.) If you had a stock or bond that had a 6% commission to sell it, you'd throw it straight out the window and never look back.

(Wait - mutual funds did have commissions - these were "loaded" mutual funds, and had something like a 5% fee to buy in. They've gone the way of the dinosaur.)

Unfortunately with housing we're still stuck with high transaction costs - any strategy for managing housing investment has to take them into account. I don't think that FSBO represents a great way to save on transaction costs - if we announce FSBO, we probably have to accept buying agents, and even then we're going to get low-balled because the buyer knows we aren't paying that 3% on our side. For transactions to become more affordable, the cost structure needs to decline across the entire industry.

Besides the high basic transaction fee to sell the house, there was another factor: time on market. At the time of the decision, there were almost a dozen short sales in our neighborhood, many where the house was the same layout as ours. The short sale price (yes, many won't sell at all at that price) was perhaps $20k to $40k below "market", which was already really low.

Would we have been able to sell at market by waiting? Well, having let this blog post sit, we now have a historical data point. Our neighbors have been trying to sell their house. Personally I think they have over-priced it (it's easy to call your neighbor's house over-priced, hard to call your own...) but the operative number is over 200 days on market, or over $10,000 in lost rent.

With the great options of fire-selling (at least $20k loss), leaving the house on the market (at least $10k loss) and paying at least $15k in transaction fees, selling didn't pay.

Executive summary:
  • We "make the mortgage".
  • While this feels good, the net return on the house is probably a loss, when you go account for everything.
  • The specific losses to selling in this environment would be worse.
* How do you know how much "return on equity" you should be getting? One simple way to figure this is to ask what you'd get back if you put the money somewhere else. I would suggest a conservative proxy, like: what would you get in an FDIC-insured savings account. It isn't fair to compare return on equity to a highly volatile investment like stocks. Sure you aren't making 6% on your housing equity, but then it is unlikely to plunge by 50% one year.

Sunday, March 21, 2010

Who Drank the Spicy Soup

So there is one remaining question to answer: who drank the spicy soup? If the effect of securitizing all of this mortgage debt was to create more risk than we could handle, who ate that risk and ended up with a case of indigestion?

To quote top gun, the list is "long and distinguished". (Wikipedia has a chart but I'm not convinced it is accurate.

First: the investment banks.
  • Lehman and Bear Sterns dropped dead. While their losses weren't huge, they didn't have the capacity to sustain much damage either.
  • Merrill Lynch lost the most - almost $30 billion. I'm sure Bank of America feels good about that purchase. ("We want to be just like Citi - flat broke!")
  • Goldman apparently didn't lose anything, although Goldman's financials can be opaque at best. They shorted the market at the last minute to offset losses - on the other side of that bet: AIG
Of course if the Wall Street investment banks are dumb, you know someone has to be dumber.
  • UBS managed to lose $37.7 billion. They were one of the big suckers at the table.
  • Citigroup lost $39.1 billion - their strategy of creating toxic waste, then keeping the toxic part didn't work well.
  • Bank of America lost $7.95 billion. I believe this is the losses they earned the hard way, not the losses they "bought" by buying Merrill before Merrill's losses were fully disclosed. Note that they also own Countrywide - clearly BoA has a talent for aquisition.
  • Washington Mutual and Wachovia both went under and were sold off (to JP Morgan Chase and Wells Fargo, respectively).
Finally we have the insurance companies. Since insurance companies "sit" on your premiums while waiting for something bad to happen, they are natural buyers of securities. Some insurance companies are crazy enough to try to insure financial products. Both MBIA and AMBAC had multi-billion dollar losses, but the big whale, the big sucker at the table was AIG, who have lost over $100 billion dollars since the crisis hit. Don't worry, though, you and I have their backs.

If you want a short list of suckers, basically it's:
  • AIG, who bet wrong on sub-prime for the first half of the housing boom.
  • Citigroup and Merrill Lynch, who were dumb enough to keep their own toxic waste. (Bank of America has managed to buy plenty of sick companies, so Citi and BoA are now the dynamic duo.)
  • UBS, who bought toxic waste late in the game.
Of course, the real sucker is you and me, because it's our tax dollars bailing out AIG, Citi and BoA.

Saturday, March 20, 2010

Synthetic CDOs: I Punch You In the Face

This is where my tortured analogy of spicy soup dies: the synthetic CDO. Let us recall:
  • Tranched bonds (like mortgage backed securities): spicy soup that has been allowed to settle. Some people buy spicier soup for less money.
  • CDOs: let the soup settle again so we can divide it up yet again - a way to try to recover a little more value from the spicy soup.
Now here's where we have to break from reality. In real life if you buy that really spicy bowl of soup and drink it, your mouth is going to be on fire. No one else's. You drink it, you burn. We can't get away from that.

But in the world of finance, pain comes in the form of losing money. This means...pain can be transferred. It's not impossible to set up a deal where you drink the spicy soup and someone else's mouth feels like an inferno.

(Hrm...that might be a really good deal. Remember, the spicier the soup, the cheaper. So you could get cheap soup and someone else hurts like hell...hrm...)

That is the synthetic CDO. You get paid a little bit, and your mouth hurts like hell when someone else drinks the soup. It's as if we have a side agreement that if I drink the soup and it is too spicy, I can punch you in the face.

To get more technical:
  • Normally I get paid interest by taking a little bit of risk. I get more interest for more risk.*
  • Normally the way I take this risk is by loaning someone money. The risk is I don't get paid back.
  • For a CDO, the whole thing has been through a blender. (Hard to spot those Wall Street bankers taking fees with all of that blending!)
  • For a synthetic CDO, you've got my back! I loan some money out, and you take the risk of me not getting paid back. If I don't get paid back, you pay me. I give you a bit of my interest to convince you that this isn't the worst idea ever.
For me, the synthetic CDO lowers my risk (I will get paid back no matter what) and it lowers my return (since some of my interest money has be used to pay you to take this insane gamble).

For you, you get paid a little bit now and you might have to pay me later. If this looks a lot like an insurance policy, that's because it is. You're insuring me against the loss of my loan. In real life, a synthetic CDO might be built out of Credit Default Swaps (CDS), which are just that - insurance on a loan or bond.

With a synthetic CDO, we've separated the party drinking the soup from the party getting burned. What are the implications of this?
  1. I still get some interest, even though you took the risk of not getting repaid. Wha? Well, not getting repaid is not the only risk of a loan. I could make a loan at a low interest rate, then interest rates shoot up. I lose out on the chance to make a loan at a higher rate. This is "interest rate risk" and I still am exposed to it even though you insure me against a total loss. So by selling you this synthetic CDO, I got rid of some of my problems, but not all, and I got rid of some of my return (interest) but not all.
  2. Since I get repaid no matter what, I can now loan money to just about anyone I want. Uncle Joe likes to place bets down at the track, but I don't care. If he can't repay, you've got my back.
  3. I can only loan uncle Joe as much money as I have. But you, you've got a very different situation. If you have my back, you don't have to put up any money right now! You only owe me money if Uncle Joe doesn't make book. So you could insure me against a bunch of loans without actually having the money!** This would be profoundly stupid of you, because I'll have to break your knee caps (and me, because how are you going to make good on the loan if you have no money and no knee caps) but not impossible.
Of course, item 3 is precisely what AIG did - they told Goldman Sachs they'd make good on loans Goldman had made if the debtors went under. AIG didn't have enough money on hand to actually pay off all of the loans. How did this play out?
  1. AIG goes bankrupt.
  2. We (the taxpayers) bail them out, giving them lots of cash.
  3. They use the cash to pay back Goldman, whose original loans have not paid off.
Well, that worked out well, didn't it.

The astounding thing is that AIG wasn't required (and still isn't) required to record in their financials that "in the event of a bad day, we will owe Goldman Sachs a crap-load of money."

If it seems astounding stupid that you could possibly owe someone a gajillion dollars and go around pretending that you don't have a burden on you, well, it is. A good start to financial reform would be to require such obligations to be recorded on balance sheet.

* This is a little bit surprising to normal people because the most basic savings vehicle, the bank savings account, has no credit risk because the FDIC has your back if the bankgoes under. But then that savings account isn't paying much interest is it?

** To be clear: you can sell a credit default swap without having the full amount of money to cover the loss. That's the AIG case. When the insurance is packaged up in the form of a synthetic CDO, in theory the principle of the synthetic CDO should be equal to the amount we might lose. Thus the worst case is losing all of the synthetic CDO. That's why it's called a synthetic CDO - it looks just like a CDO in that it earns a little bit of "interest" and in the event of a bankruptcy loses all of its principle - just like a regular loan.

Friday, March 19, 2010

CDO Soup

We can understanding tranching via soup - it's the uneven distribution of risk and reward in a financial, um, mixture.

So what the hell is a CDO?

Well, recall in our soup example, we let the soup settle and ladle off the "really hot stuff". We can then sell spicy soup as not-so-spicy because we've put a disproportionate amount of the "hot stuff" into just a few bowls.

What do we do with the left-overs - those few bowls of soup so hot that no one can drink them?

Well, if you're Wall Street, this is what you come up with:

We'll just let the spicy soup settle for a day! Then we'll come back and it will have settled into: a really goddamn spicy bowl of soup at the top, and the rest will be drinkable. We can sell that.

That's a CDO*. We cut up and unevenly divide the remnants of what we already cut up and divided.

And if really want to push it to the limit, we'll take that settled soup and we'll let it settle again (no, really!) and try to ladle off some of that as a CDO-squared.

And if we really want to...oh wait, the whole financial system blew up. So much for the CDO-cubed.

* Technically a CDO is a tranched mixture of any pile of underlying stuff, but the ones that have gotten us in trouble are the ones that are a cut up mixture of mortgage backed securities.

Thursday, March 18, 2010

Really Spicy Soup

In my previous post I tried to explain tranching via a soup analogy - you take the spicy soup and ladle it out unevenly so some get more spicy hot paste and some get less. This does leave a few questions unanswered.

Why the hell would I ever drink the bowl with all of the hot sauce?!?

The answer to that is: it's cheaper. If you and I were splitting the cost of take-out, maybe I pay for 2/3 of the soup and you pay for only 1/3rd. That's only fair, because you're going to take all the heat, so to speak.

Tranches are the same way - the tranches with an "unfair" share of the risk also return higher interest. you feel lucky, punk?

Tranching is just dividing up the hot sauce - how does this affect the system?

One of the effects of tranching is to bring players into the market who might not normally buy in.

Let's consider my mother. My mother will not eat spicy food. Her spice tolerance is really, really low. If we have her over for dinner, we're not going to buy the spicy beef soup, even though it is very tasty.

But wait -- if we just tranche it, we can buy the beef soup! All Lori and I need to do is not stir it up and serve the first bowl to ourselves, taking all of the hot paste to ourselves. The remaining soup is quite mild, and my mother, who normally would not be able to drink it (if it were evenly mixed) will be able to drink the remaining tasty broth. She'd probably like the beef ribs too.

If you are the Korean restaurant, this is a hell of a discovery - you can sell a lot more soup now, because you can sell soup to everyone, not just customers with a taste for spicy food.

There's just one hitch: you have to convince a few people to drink that first bowl, the hot bowl, the one with most of the spicy paste. I have had that bowl, and let me tell you, it's an experience!

Motivation to Lie

Now before I continue, I must be clear: the Westborough Korean restaurant has never lied or misrepresented the spiceyness of their food. They have never suggested that we not mix the soup, and they have never suggested that drinking the first bowl of unmixed soup would be a smart thing to do. Our discovery that the soup will "tranche" itself happened by accident, and this whole post is just a (badly stretched) analogy.

So if I can sell more soup by letting it settle (so the less spicy part is palitable to a wider audience), I might find that the limiting factor is finding people to drink the first bowl. Sure the first bowl might be cheaper, but how else can I sell it?
  • I could simply misrepresent how spicy it is. When someone asks me: "will I roll over and scream in pain if I chug the first bowl", I'll just answer "oh no - it's spicy but it's not that bad. You'll be fine." This isn't a very good idea - my customers might get angry at me for setting them up for pain, but that'll happen tomorrow, and I want to sell soup now.
  • I could agree to drink the first bowl myself if no one else wants it. This is a damn stupid thing to do because I might have to drink the soup, but again, I want to sell soup now.
That is pretty much what Wall Street did: they got the ratings agencies to represent the more dangerous tranches as "not that bad". In some cases, they agreed to buy the tranches back, or they kept them for themselves, a move that has caused them to lose a lot of money.

Those stupid could they be so dumb? Oh wait - they made bonus money when they sold the tranches. And...wait - we (the tax payers) are bailing the firms out now that they're in pain. Maybe we're the dumb ones.

Wednesday, March 17, 2010

Subprime Soup

I'm a huge Michael Lewis fan, and his interview on Fresh Air is a good one (as is his 60 minutes piece).

The Planet Money team recently bought their very own toxic asset. They look surprisingly cute in the animation for something that has completely destabilized our economy.

One barrier to getting real reform is that most people's eyes glaze over as soon as you say the word tranche - and yet tranching is fairly close to the heart of the shell game that was built on top of some really poor mortgages.*

If you want to understand most of structured finance in one delicious meal, just buy the spicy beef soup from the Westborough Korean Restaurant, our new go-to take-out place.

The soup is a delicious beef broth with noodles, beef ribs, bean sprouts, and this really spicy red paste - in small quantities the paste is delicious. But if you pick up takeout, then by the time you get the soup home, the paste has settled to the top. The first bowl of soup is going to have more of the red paste than its fair share, and is going to be really, really hot. You'd have to be a little crazy to drink it.

That is tranching, in a nut-shell. Given a pool of "stuff" with a little bit of danger mixed in, tranching just means the Wall Street firms who are dividing the soup up put more of the spicy sauce in some bowls and less in others.

This is generally good for those getting the less spicy bowls - it allows them to make a pool of "stuff" less dangerous.

What do you call those pools that have more than their fair share of hot sauce? Those pools are called 'toxic waste', and that's where our toxic assets come from.

* I believe the true heart of the crisis is not any one financial technique - it is asymmetric risk - that is, the ability of Wall Street bankers to make money when a bet wins and have someone else pay when a bet loses. But in order to get anyone to play a game like that with them, they need to make the game complicated, so that the absurdity of the rules is less obvious. If you slice and dice your securities into enough tranches, it becomes a lot harder to see what you actually own.

Of course, when you actually look, it's an eye opener!