Saturday, April 17, 2010

Rational Actors: What We Knew

I have blogged in the past on our decision to rent our house out (rather than sell it) and just a little bit on the ideological food fight over human decision making going into the crisis. Before the crash, did lenders and home buyers act rationally (according to perverse incentives) or did they simply lose their minds? I can only provide insight into one tiny transaction within the housing boom, the one I was involved in.

Why did we buy the house? What were we thinking? In hindsight it is clear that the correct decision would have been to defer buying the house to avoid the 25% decline in asset valuation. But when we bought in 2006* (near the top of the market), what did we know?
  • We knew we wanted a house. We were old enough, and had rented for long enough that we wanted to try it the other way.
  • We knew that we would have to live in the house for at least 4 or 5 years just to break even on high transaction costs related to housing purchases. We had no intention to sell the house on a shorter time frame. (In hindsight we ended up leaving after 4 years - a bit earlier than expected.)
  • We knew the house was unlikely to appreciate in value. We were aware that the rent-to-mortgage ratio was pretty far out of whack, and we believed that house prices couldn't really go up any more. In other words, we had no speculative interest.
  • We "knew" that high housing prices were sustainable. Before Washington DC, we had lived in Boston, and I had worked in San Francisco, two markets with sustained unusually high housing costs. So it didn't seem implausible to us that in the suburbs of Washington DC, housing prices could simply remain high.
  • We knew that we could get decent financing with a manageable fixed long term interest rate, and that this was desirable. We put down a very large down-payment. We had no expectation of strategic default at all.
  • We knew that there had been rapid turn-over and price appreciation (a "boom") in the housing market over the last few years, with crazy things going on (bidding wars, no inspections, etc).
  • We knew that there were tax incentives to pay a mortgage rather than rent, and we could do a monthly-cost analysis to show that we'd get a reasonably good deal (in terms of monthly payment) despite rent-to-mortgage ratios.
We did not know that underwriting criteria had so totally fallen apart; stories of NINJA loans and liar loans were not mainstream and the Giant Pool of Money was two years away. And we had no idea why Wall Street would make loans that would not be repaid (nor did we realize that this they were doing such a thing).

We also did not think that Washington DC was likely to suffer a real-estate bust they way Texas did in the 80s; it seemed that the government was hiring lots of people and would keep on doing so and that the housing build up was reasonably permanent, similar to other high-priced east coast cities. The notion of a nation-wide, rapid, steep housing crash wasn't on our radar; we didn't recognize that the volatility in housing prices we were seeing was, in fact, real volatility.

So putting it all together: our purchase of the house was not speculative. We expected flat home values and tolerable expenses, and we were willing to accept that to not rent. We had no sense of the price volatility that a house might display.

Better Late Than Never

I started reading finance books after we bought the house; before the house purchase, my strategy was "save money for the house." It was only after the house down-payment was paid for that I started to ask "what else to save for."

So one of the questions I have to ask is: had I read all of the books first, would we have avoided buying the house? Upon writing up our assumptions, I think the answer is "no". In particular, simply making a case that houses are tied to interest rates and interest rates can be volatile would not have scared us off. Houses are illiquid and expensive to sell, so we would have expected a rate spike (even a serious one) to only slow the market down, not murder it. The missing piece of information was the huge exposure to ARMs and deteriorating lending quality. It's not that this information wasn't know-able back then, it's that we didn't know about it.

If you want to categorize this transaction, the best label is probably "information asymmetry".

* This is not meant to be a "cry-me-a-river" post regarding the house; we are very, very lucky that we were able to absorb the loss without having to put our career and life plans on hold. Many others are not so lucky.

Wednesday, April 14, 2010

Too Big To Fail? I'd Be a Fool Not To Invest!

The question of what caused the housing and banking crisis we are now in* is more than just an academic question - the root causes have profound implications on what we might change to avoid doing this all over again in 2020. There are a number of schools of thought on what made bankers do what they do: were the banks the dumb money, did they knowingly dig a hole, or were they pushed in the direction they were pushed by the structure of the system?

Jeffrey Friedman refutes the notion that bankers acted irresponsibly because they knew they were too big to fail. This is probably true. (I would argue that they acted irresponsibly because they'd make huge amounts of money in the very short term for doing so.) But the entire issue is a bit of a red herring; the problem with Too Big To Fail isn't that it makes bankers do stupid things; the problem is that it makes investors do stupid things, and it forces us to pick up the bill.

There are two fundamental problems with really, really big banks:
  1. They are so big that we don't have a practical way to disassemble them if they go bust. The FDIC does a great job of taking apart small banks. But this process won't scale up to an institution as large as Citi or BoA. See Simon Johnson's comments.

    So the first problem with TBTF is that should a TBTF bank go bust for any reason, we're going to end up picking up the bill with a giant bail-out. It doesn't matter whether the banks sink the bank intentionally; unless you believe that bankers are infalliable and will never screw up, you have to recognize that TBTF is hazardous to the well-being of the American taxpayer.

  2. Even if the bankers aren't counting on a bail-out, the bond-holders may be. If there is a lesson to be learned from this crisis, perhaps it is this: BoA and Citi's debt is safer than that of smaller institutions because BoA and Citi can't be put into bankruptcy or liquidation due to their enormous size. Therefore if they do fail, a bail-out is a more likely solution than liquidation.

    The problem with this is that this is going to make their debt less risky to bond-holders; private investment will be directed toward these large institutions because of their "structural guarantees" (that is, they are so huge that their disposal in the event of disaster is going to be in the form of a bail-out, not a liquidation). Thus they will have cheaper funding (since their debt is less risky) than smaller banks, and thus they will be more profitable, crowding out otherwise more fit competition. (And you thought there was no economy of scale in banking?)

In summary, my problem with TBTF is not that it caused the crash; my problem with TBTF is that it caused me to get stuck with the bill, and it's going to make the next crash more likely.

The Democrats talk about regulation, and the Republicans talk about ending bail-outs, but I am waiting for either of them to do what would really make a difference (but go directly against a huge stream of lobbying dollars): break up the biggest banks into smaller pieces. Too Big To Fail is just Too Big.

* Some insist that it is strictly a banking crisis - those of us who have lost significant value on the houses we own, or are in some other way connected to reality, think otherwise.

Tuesday, April 13, 2010

Rent This House - Pay No Tax

In past posts I have tried tried to describe the math behind keeping the house as a rental. Main points:
  • Consider all expenses when calculating whether rental pays. Property tax is a big one!
  • Consider return on equity; if you have equity, you need to earn more than you would by selling and putting the money you get in a bank. (Of course, the banks will give you 0.0001% now, so that may be moot.)
  • Consider high transaction costs - it's just not cheap to buy and sell real-estate. In our case, we considered the market dislocation as driving up transaction costs (in that we'd have to have the house on the market for a while).
Well, it's April and the tax man is upon us.* Here is what I have learend:
  1. Losses on the rental property (there are a lot of up-front expenses; since we rented in October we're definitely in the red against those up-front costs for 2009) do not lower taxes on your regular work-type income. So we just "carry the loss forward" - that is, we can use our loss to pay less taxes next year if/when we make some money on the house.
  2. We have to depreciate the house - that is, we claim for tax purposes that it loses about $3000 of its value as an asset every year. This produces another "loss" for tax purposes.
If we have a good year with the house, we will still make less than $3000 in profit, so it turns out that due to depreciation, we will probably never have to pay taxes on the house.

To explain this second point in more detail: think of your rental house as a business; the business "buys" a house and thus has an asset. The IRS requires you to treat that house as losing value over 30 years. (This is silly because you will almost certainly be able to sell your house for more than $0 in 30 years, but hey, I only work here.) Your business recognizes a "loss" every year of a little bit, and thus you are less profitable to the IRS than you are in real life. (Clearly you didn't lose any real money to depreciation.)

There is a flip side: when we go to sell the house, the price we paid will be lowered by all of that depreciation. We paid $400,000 for the house (plus closing costs), so the odds of us making money on selling the house are approximately zilch. But for tax purposes, the price we bought at will appear lower for every year of depreciation.

(The IRS calls this "recapture" - the idea is that we really did think the house was trending toward $0, so when we actually sold it for more than $0, we were surprised that we made that unexpected money. Again, this strikes me as very silly indeed, but I'm not in charge.)

The final piece of the puzzle: if we have accumulated "losses" with the IRS every year (it's possible, because depreciation will be larger than our profit margins) we do get to use those losses against selling. So when it finally comes time to sell the house, the taxes on any gains (should they, due to some freak accident, exist) will be made larger by depreciation but smaller by losses carried over.

Does this affect the net calculation of whether it pays to rent a house out? I don't think it really matters substantially. It does change all of the numbers a little bit, but the way I see it, there are enough unpredictable factors (how long will you be renting, what will future interest rates be, what will the rental market bear) that the unknowns dwarf the imprecision of calculations done without correct tax treatment.

* You should really not treat anything I say on this blog as professional advice. The best thing to assume is that I am a computer programmer who is woefully under-informed in financial matters, because, well, I am a computer programmer who is woefully under-informed in tax matters.

Come On Man...Everyone's Doing It

This American Life has some of the best long-form reporting on the financial crisis. In particular, normal people like us have a prayer of understanding what happened.

Their latest story coveres the Magnetar Trade. Depending on who you believe, Magnetar either bet both for and against the housing market (that would be Magnetar's view - they were just doing their jobs and "hedging") or they created particularly crappy CDOs just so they could bet against them (that would be ProPublica's view).

You can understand the Magnetar trade like this:
  • You put 5% down on a house.
  • You come to me and ask me to put in 95%. This is an equity sharing arrangement, not a loan; that is, I will own 95% of the house, and you will own 5% of the house. I am nervous; to entice me, you say that you'll eat the first losses if the house value goes down. (In other words, you are enticing me by offering up your 5% investment as some kind of "price protection".)
  • You take out fire insurance on the entire house.
  • The house burns down. You lose your 5% up front but collect the entire insurance payment. I'm out my 95%.
The claim against Magnetar is that they requested the house be built out of matchsticks - they requested that the CDOs be built as badly as possible. This implies that they expected return on their insurance payment, not their initial small investment.

To look at this another way: in order to take insurance out on the house, Magnetar needs the house to be built. They put up their 5% money to lure others in. (Come on man, come invest, everyone's doing it, we have our money in.)

There are two separate sets of losers from the Magnetar trade:
  • Whomever bought the end CDOs that lost value (burned down) lost their investment.
  • Whomever sold the insurance to Magnetar lost money when they had to pay out.
For the end CDO buyers, ProPublica lists some of them, and I can't imagine they won't sue. In particular, I would expect them to sue the investment banks that built the CDO with Magnetar's money. The problem is that the banks building the CDOs may simply have been stupid.

(It is a certain special kind of stupidity of course, one that earns the banks lots of up-front fees. The problem that the bank can take your money, lose it all, earn a fee in the process, and not be considered criminals is a topic that will have to wait for another blog post.)

I don't know who sold insurance to Magnetar, but it may have been a similar group. Wall street banks can turn insurance into synthetic bonds. (This post by Felix Salmon may help explain how this works.)

The real problem here is: I don't see how we'll ever really know what happened; since Magnetar is a hedge fund, they're not required to disclose who they traded with, and they're not offering up a list of trades to prove their claim that they were not hoping to burn the house down. Since Magnetar is quite successful, there won't be an autopsy. (Compare to LTCM, which failed in 1998 and is now described in detail in dozens of financial history books.)

I'm Only Happy When It Rains...

On On the Media, Adam Davidson describes the difficulty of explaining anything financial on the Planet Money podcast. And this has to be a concern to all of us for a simple reason: we (the tax payers) got totally looted this time around, and if we can't even understand the arguments about how to fix the problem, Congress isn't going to "fix" the systemic problems in our financial system in a manner that works out well for us. Simply put, if we can't understand it, we're going to get looted again.

So what do we do when we find explanations like this? Friedman is arguing that people buying homes at of market were acting rationally, the bankers weren't acting irresponsibly due to their paychecks, and that regulation, not deregulation caused the crisis. His arguments are more nuanced than I am describing; I am phrasing them a bit more bluntly to draw out how ludicrous they are. Mike Konczal takes apart point three here.

But can we even understand what they are arguing about? What are these capital minima, and why do we care?*

Rainy Day Funds

As part of our personal finances, we maintain "rainy day" money - that is, money that just sits in a boring old FDIC insured bank account for the sole purpose of having money on hand if something bad happens, like I lose my job or the car gets hit by a meteorite.

Now we have a friend, let's call him Joe. Joe has a good job that pays a lot of money, and Joe has expensive taste. He pays for his car, mansion, caviar, champaign, and Armani suits using credit cards, mortgages, car payments, etc. In other words, Joe has a ton of debt, but he also has strong cash flow coming in to pay that debt. He swears he has the situation "under control".

Joe has a rainy day fund too! But...his rainy day fund is only $5000. We keep telling Joe "that's too low". What if something happens?

Unfortunately Joe's high paying job was in banking, and he was laid off last year. The $5000 of rainy day money wasn't even close to being able to absorb the avalanche of debt he was faced with. In fact, it got a lot worse; since his house went down in value, he can't sell his house to make back the money he owes on the mortgage. Clearly $5000 wasn't enough.

Rainy Day Funds for Banks

You can think of reserve capital as rainy day funds for banks. Every time a bank makes a loan, they are required to put a little bit more money into the rainy day fund, just in case the loan goes bad. The Friedman and Konczal are arguing about what effect the government minimums on the rainy day fund had.

Konczal's argument is an important one to understand, and it is basically this:
  • If banks are going to have the minimum rainy day funds the law will allow, then banks are trying to live on the edge, and the law has to assume that bankers are crazy.
  • If banks are going to have larger rainy day funds than the law will allow, then you can't blame the law for being "too lenient" because the bankers are choosing their own (larger) rainy day funds and thus it is the banks own decision that is right or wrong.
In other words, you can't claim that banks are grown-ups who can manage their own finances and then blame the law when banks fail.

This begs a question: why wouldn't banks set up larger rainy day funds to avoid failing? The answer is that banks profitability is measured relative to their rainy day fund - that is, relative to their capital base. Thus the closer to the edge a bank runs, the more relatively profitable it is as a business.

I would argue that there is also assymetric risk (something that deserves its own post). Basically if the bank is more profitable, the bankers might get higher pay, their stock options might be worth more, they might get a bigger bonus pool. But if the bank fails completely, once the bank is bankrupt, life doesn't get worse for the bankers if it is more bankrupt. If the bank fails gently (is insolvent by $1) the result is the same for the bankers as if the bank completely implodes and needs a huge taxpayer bailout.**

One more thought: while I think Konczal is right to point out the flaws in Friedman's logic regarding regulation, I consider the capital requirements (rainy day arguments) a little bit silly; the rainy day requirements that would have been necessary to keep a bank functional in the face of the crisis at hand now would have been so high that anyone suggesting them would have laughed the idea out of congress or the boardroom. The requirements would have been too large for banks to function at all.

The heart of the problem isn't banks having inadequate reserves for a crisis, the heart of the problem is the banks making an astounding number of loans that will never be paid off. It's the equivalent of Joe buying the state of Florida on his bankers salary; no amount of rainy day savings is going to protect against that.

* Regarding whether bankers who made money by doing deals regardless of whether they failed may have had incentive to do stupid things, and the mentality of home buyers in 2006, I'll leave that to you to judge - human nature is what it is.

** That result under Bush and Obama appears to be that the bankers get to keep their jobs, which is astounding.

Monday, April 12, 2010

So...How Did We Get Here?

Housing wise, it's been a good four years. Let's review:
  • Housing prices are down about 25% where we used to live, more so in less fortunate parts of the country.
  • The economy is on its ass, with high unemployment and poor growth.
  • Interest rates for consumers are very low - if you have money in the bank, you're not getting much for it.
  • Credit for businesses is hard to get, see also poor growth.
  • The Federal Bailout Agency^H^H^H^H^H^H^H^H^H^HFederal Government has spent all kinds of money on the crisis* so when we come out of all of this, our national total national debt is going to look at lot more like Italy's than it used to.
You might ask yourself: how did I get here?

Let's trace backward. The problems we have today come from two sources:
  1. The results of a housing bubble. The housing bubble represents a massive mis-allocation of resources. Simply put, we spent the Bush years (I don't want to call them the "oughts") building houses we didn't need; this hasn't set us up for future growth, rather it was squandered investment.
  2. The bursting of a credit bubble. There's a lot less credit than their used to be, which makes it hard to invest in new growth.
These bubbles are actually interlinked; since most people finance their houses, house prices are tied to housing credit (that is, cheap mortgages). So really we had a mortgage bubble that is now hitting us twice: it hurt our real economy by misdirecting investment into houses, and it hurt our banking system by creating a lot of bad debt.

How did we end up with a mortgage bubble? The answer to that is: financial alchemy.

Simply put: Wall Street banks discovered that they could turn lead (sub-prime mortgages and other low quality debt) into gold (AAA-rated CDOs) because the credit rating process wasn't very good. Banks could stuff their securities with "thin file" borrowers (borrowers with a good credit score but short credit history) and low-score borrowers and the average was good enough for the pool to be acceptable.

Wall Street made money on these transactions from the difference in the cost of raw materials (crappy mortgages) and the price of the finished product (high-grade AAA debt). This explains the strong pressures for lower lending standards: the worse your mortgage, the less the bank has to pay to buy it; if they can still manage to turn it into a CDO, they make more money by lowering their cost of supplies.

You can think of this as painting lead bars in gold paint because the gold inspector only looks at the bars, rather than testing them more carefully. Get the cheapest lead you can to make the most money when you resell your fools-gold.

The banks building these CDOs would have done quite well by this strategy except for one problem: they kept their own defective product. In trying to answer the question: were they malicious or were they incompetent, this pushes me to suspect gross incompetence; if you know you are making a truly defective product, the last thing you want to do is keep it. If nothing else, I think the CDO losses being eaten by the major banks indicate that they didn't realize the scope of the mess they were making.**

So to summarize: Wall Street discovered they could make money by financial alchemy, and the raw material was housing debt; this demand for housing debt as raw material drove down the cost of financing and thus drove up the price of houses, which caused a building boom. (You gotta love how commodities markets react to prices.) When the whole thing fell apart, we're left with too many houses, too much bad debt, and not a whole lot of money. Not a whole lot of money unless, of course, you received bonus payments for creating CDOs.

* I am very critical of the Fed and Treasury's handling of the crisis, which I think has created moral hazard and burned fiscal and monetary resources without either fixing real problems for "main street" or addressing any kind of long term problems. This approach has been consistent under both Bush and Obama.

** This doesn't make their actions even remotely acceptable; but I think the issue is important when we consider how to prevent this from happening next time. Regulations to stop "bad guys" aren't going to help; we need regulations that stop the amount of damage that can be done by "dumb guys who have our money".