Financial Armageddon for Dummies

There seems to be a lot of confusion about the nature of the crisis that is consuming our financial markets. I thought to start this thread to clear some of this up.

What are we dealing with is something called a “credit crunch” or a “liquidity crisis.” There is a lot of talk about “bad mortgages” or “bad debt” or “bad CDOs.” How could we have written this “bad” debt?

The first fundamental bit of confusion usually comes over the meaning of the term “bad.” Chances are that when you hear a debt is “bad” you interpret this to think that it’s in default, or not paying or not likely to pay.

While some of these securities and mortgages are “bad” in this sense, this is not what is generally meant when people are talking about the billions of dollars of bad debt choking the financial market.

These mortgages and debts have all been securitized into bonds and bond-like instruments, and they trade in the financial markets. A bond is said to be “bad” if it becomes illiquid, that is, you can’t trade it.

The vast majority of the debt that we are talking about is still paying interest, is not in default, and is not expected to be in default. It is simply illiquid.

To understand this one needs to know that bonds typically trade much differently than stocks. Bonds tend to be held for longer periods of time and don’t necessarily trade every day. Imagine if GE stock traded only once every couple of months, and you get the picture.

The way traders trade bonds is by doing something called discounting, usually off a treasury. Treasuries tend to trade all day every day, so if you can figure out how a given bond compares to treasury you can simply watch the treasury price and interpret that price movement towards your bond. Another way to do it is to find a bond that traded recently that is similar to the bond you own. If that bond traded up, chances are your bond is worth more, too. This is why ratings are important. If your bond is rated AA, you can watch how other bonds with this rating trade and figure out what a fair price for your bond is.

One might say, for example, that AA rated bonds are trading 200 basis points off their treasury upon observing a 30 year corporate bond compared to a thirty year treasury. A basis point is simply 1/100th of one percent. So, if the thirty year treasury is paying 4.36% and a AA rated bond is paying 6.36% you know that the AA rated bond is trading 200 basis points off the treasury.

Now, if you have a AA rated bond that is a ten year bond, you can use this information to price your bond. You might see that the 10 year treasury is paying 3.74% and therefore interpolate that your bond shoud be priced to yield 5.74% Not all AA rated bonds are equal. Some have features in them, calls, sinking funds, and other things that may make them more or less valuable then other AA rated bonds. You adjust your price based on this. You make comparisons to other bonds which have traded to confirm or adjust this price. This is why ratings are important.

That’s how bonds are priced and that’s how they trade. If you have an investment portfolio that contains bonds you may see your price change month to month even though not a single bond of that issue has traded hands. There are services that due this for traders and firms and issue guesses as to what the price of a bond would be if it traded. The largest of these is JJKenney.

For the most part, this is what traders and the financial community are talking about when they say a bond is “good” or “bad.” They are not talking about it’s ability to pay which is its “credit” but its ability to trade which is its “liquidity.”

A “good” bond is a liquid bond. This means it trades reasonably often and it trades reasonably predictably and with strong adherence to its benchmarks. You can simply look at a treasury and know what your bond is worth by comparison. More importantly, everybody else knows it too. Because of this you will be able to find a good and predictable price for your bond and sell it at that price pretty easily.

A “Bad” AA bond might be the scum of the AA universe. It might have weird call features (when it can be redeemed.) It might lack collateral or seniority. It might not trade that often and when it does it might trade 200 basis points off its treasury one time and 250 basis points off the next time.

All things being equal you would prefer AA rated bonds that are good from a liquid sense.

To give you a good analogy, say you want to spend 25k on transportation. That 25k could buy you a Prius of which there are millions on the road, or a Delorean of which there are few. Pretending for the moment that these cars are equal in terms of their transportation value, which would you want?

Probably, you would want the Prius. If you chose the Delorean, don’t trade bonds. It’s easy to buy and sell a Prius. Buying and selling a Delorean isn’t. It’s a specialty item that doesn’t trade all that often. If you buy a 25k Prius today and you decide to sell it in six weeks, you shouldn’t have too much trouble getting your 25k back. You might have a tougher time finding someone to buy your Delorean.

From a liquidity standpoint Prius is good, Delorean is bad.

Bonds work the same way.

For a long long time mortgage backed securities were very very good. A securities dealer with an inventory of CDOs (a packaged mortgage backed bond) was like a car dealership full of Prius’.

Very suddenly it became like holding a dealership full of Deloreans.

Now remember, for purposes of our analogy there is no difference in the transportation value of a Prius versus a Delorean. They are both the same.

That’s liquidity.

The bad debt is bad in the sense that nobody wants to own it, nobody knows how to price it, and there is not a market for it.

Until they get rid of it, they can’t do business with each other. If they can’t do business with each other that means money get’s “tight.” Credit is simply not available. Businesses can’t borrow money. Individuals can’t borrow money. Liquidity seizes up, transactions grow scarce, inventories can’t get replaced, people get fired, and the whole economy grinds to a halt.

What you have then is a depression.

The “bailout” is simply a proposal to buy some of this delorean type debt. The government will hold it, collect interest and when liquidity resumes they can sell it back to the market.

In the past, like the S&L bailout, the government made a profit doing this.

So, the bailout is not the government spending money the way it does if it buys a tank. It will receive interest and will likely be able to sell this debt back to the market when liquidity resumes, likely at a profit.

That’s the mechanics of the crisis and the proposed solution. Questions?

No questions, just a comment,

Thanks for posting that.


While you’re calling this debt bad because it is illiquid, it is nonetheless the case that housing prices are sharply down, mortgage default and foreclosure rates are sharply up, etc. Some non-trivial portion of this illiquid debt is also bad in the sense that it’s not ever going to be repaid. In fact, is it not the case that it’s mostly illiquid because no one has any clue how much of it is bad in the latter sense?

How, then, can you or anyone else predict that the government might be able to sell the debt back at a profit?

If there’s money to be made in buying up this debt, why aren’t private investors snapping it up? And isn’t that an indication that the book value of these loans is fictitiously high? Seems like the only justification for involving the government would be to buy debt that nobody else wants, or at a price that nobody else is willing to pay, so doesn’t that amount to an outright subsidy to the current holders of the debt, who after all bought it knowing the risks? Doesn’t backstopping such risk-taking only encourage more of it in the future? And if the real value of these debts is truly much lower than their book value, doesn’t a bailout simply prolong the inevitable, as it did in Japan in the 90s? Also, several hundred billion dollars have already been injected through the FM bailouts, discount window operations, etc., yet the credit market has gotten tighter, not looser. How do we know that won’t happen with the bailout money, too?

Can’t we sell the bonds to the Chinese and get on with the spending?

That should solve the problem.


No. That is not a fact. Bill Gross addressed this issue this week in Barron’s. These hybridized securities contain components of mortgage pools with built in assumptions and contingencies for default and prepayment rates. Some of these are great, some not so great. One can plug the cusip numbers for these components into a model and examine the creditworthiness of the underlying assets based on various models for default rates and price them accordingly.

Having so modelled them, one creates a series of PSA assumptions for the prepayments, default rates and credit-worthiness of the guarrantors (if any.) One can then do a present value calculation for the sum of the anticipated cash flows. If this is done correctly and the issues are purchased at or below this present value than the government cannot lose money by simply holding the asset through maturity. There is no guarranty that the market value will exceed that value before final maturity. If it does, the government can sell it for a profit. If it doesn’t they hold it.

What caused this very sudden change?

You’re losing me here.

  1. If they can be priced in such a fashion, why are they illiquid? You said in the OP, “nobody wants to own it, nobody knows how to price it, and there is not a market for it.” Now I’m all confused. :slight_smile:

  2. How can you guarantee you won’t lose money by holding them to maturity when your value calculations rely on assumptions? Assumptions pretty much by definition could turn out to be false, else they’d be knowledge rather than assumptions.

Wheres the market? How do you know you can sell them and there will be a buyer.

There are some that are. There’s so much though that it’s a drop in the bucket.

Actually no. The book values are fictitiously low. There is an accounting requirement called “marking to the market.” This means that certain institutions are required to carry these instruments on their books at the price that they can be sold at. This price determines the value of their assets which may be counted as collateral. They may only borrow so many times their collateral. If the value of their collateral drops they must either reduce their borrowing or put up more collateral, or sell assets, i.e. “raise capitol.” When they sell assets that further drops the market price of those assets. Other firms that hold those assets than must “mark to the market” and lower the values of their collateral. Than they must raise capitol. It starts to snowball and then their is a rush or a collapse of liquidity in the asset which means that it simply cannot be sold at any price. Once this happens it can’t even be marked to to the market. These writedowns continue and reach a point where the prices bear no resemblance to the value of the asset in question. Unlike many other assets the value of a bond can be tangibly known as the sum of the present value of its cash flow.

Marking to the market works great when markets are functioning normally. It doesn’t work at all when they lose efficiency.

IMO, yes. Absolutely right. The bailout introduces “moral hazard,” into the equation which means that entities are more likely to take unreasonable risks if they expect somebody to bail them out.

Yes it does. One needs to make a “lesser of two evils” sort of judgement based on the risks of creating moral hazard versus letting this thing to do its damage in order to form an opinion as to whether it’s a good idea or not.

No. For any given set of PSA assumptions (default, prepayment, etc) there is a definable present value for any given security. For this to work, either your assumptions have to be accurate or the price has to be so low as to fall below even the most pessimistic assumptions. The prices now on the majority of these securities are below even worst nightmare type assumptions as they’ve been driven to that extreme by forced liquidations and mark to market action.

We don’t, for sure. For it to work the companies that have their securities bought need to reinject that liquidity into the market with new lending or purchasing of debt liquid debt instruments. This is their business and they really want to do business, so the expectation is that they will do so. They must also do so successfully and not simply do something that gets them “stuck” from a liquidity standpoint again. At the outset some of this freed up liquidity will immediately get stuck again.

To make up an analogy think of freeing up a a clogged drain. It may get clogged again a little bit lower or the pieces you free up may contribute to another clog, but you work your through and keep unclogging because… what else can you do? That’s how you fix a clog.

Whether or not this plan has merit is not what I’m arguing. There are many who are against it and think it’s a bad idea. Some think we should do nothing. Others recommend other courses of action. What I’m trying to do here is present the mechanics so people can understand what the actual problem is, what is being proposed to fix it, and decide if they like it or not on their own.

That’s not definitively answerable. A combination of factors most likely with market saturation being the predominant one.

This one’s my fault. Sorry. For our purposes every bond has three prices. The first is it’s liquidation or “fire sale” price (if you had to sell it today, what could you get for it.) The second is its hold to maturity price (If you hold onto it and collect interest and get money back at maturity, and discount the sum of those future cash flows to their present value than what number is that) The third is the market price (If you examine current market conditions, and take your time, sell it for a reasonable price what should that number be.)

Typically when you mark to the market you are using market price. You do this the same way you might price your house. You use comps. If you’re only comps are fire sales than there really is no market price. If the comps are wildly disparate what do you use? You’re not allowed to use the hold to maturity price, so you’re forced to use fire sale price. If nobody is buying at all (and we went months in the CDO market with essentially NO transactions) than there is not even a fire sale price. It becomes impossible to price them.

A trading firm has no use for the hold to maturity price. They are in the business of providing liquidity, so they can’t use that price. Nobody knows how to price them in this context.

You need somebody to price them in a different context, a third party not under the same constraints. Enter the government. They price them not from the context of a trading firm creating liquidity but from a discount to the hold to maturity price (they can afford to look at it from this perspective. A trading firm cannot.) Once they start buying at this level suddenly there is a market price, the firms can mark to the market and price efficiently from their liquidity context.

You can’t be “sure” sure. You can be “pretty damn sure” sure though. Right now, the liquidation price of these securities is mostly at "everybody defaults, even the government backed portions get defaulted on, the credit default swaps are worthless, and inflation goes to 25% a week tomorrow (I hyperbolize, but not all that much.) The liquidation price puts the assumptions at post-apocolyptic-we’re-in-a-Mad Max-movie levels. If you don’t think that’s where we are currently at than the assumptions are too low and the bonds are well below hold to maturity values.

Okay, that made a lot more sense. Thanks for this thread - clearer than anything I’ve read in the media.

Anywhere there’s a buyer and a seller.

You don’t for sure. For the markets to work efficiently there has to be confidence that they will. You can’t make that for sure. If the government buys below the hold to maturity value (present value of the sum of future cash flows) than they still make money in the unlikely event that no market ever develops.

In the really real world though, we can be pretty sure a market develops and there will be buyers as a consequence of homes being bought and sold from time time and those sales requiring mortgages and those mortgages requiring credit markets for liquidity.

It’s all interconnected and it all has to work or none of it does. Either that or we’ve trashed the whole system and gone to communism or something else.

My pleasure. This is pretty esoteric stuff. I’ve been posting for 10 years and my area of expertise has never been particularly germaine to most discussions.

Imagine if you were a professor of 19th Century French Literature, and everybody’s eyes glazed over any time you talked about it.

Suddenly there’s a worldwide Baudelaire crisis that threatens everybody, but nobody understands “Les Fleurs Du Mal.”

That’s me. I’m glad to be useful.


Do you have any estimates or references for how much (meaning, what %) of the stock will be bought with the $700 billion? In theory, enough needs to be purchased in order to accurately price the things in a reasonably liquid market, correct? So the $700 billion is some % defined as $700B/[stock]. After that people should be able to correctly value them, buy, sell, borrow against, etc. Correct?

So how is the price for all these debt papers to be determined. ? If they know we are going to buy a trillion dollars worth, wont they stick it to us. ?

No. I’m kind of taking Paulson and Bernanke at their word that that’s what’s needed. It’s not stock though that we’re talking about though it’s debt instruments (though there may be some equities tied up with some of it.)

The 700 billion is what they figure is needed to unwind the leverage and allow the holding institutions to meet their collateral requirements without undue hardship or collapse. After that, ideally these instruments price themselves at some as yet undetermined but efficient (or efficientish) market price.

That market price will likely be significantly lower than the discounted hold to maturity price the government has paid. The liquidity than helps the credit markets recover. Housing and mortgage markets stabilize and the doomsday assumptions built into this efficient-ish market price improve. Everything gets better and then the market rallies so that this price increases to or past what the government paid and then the whole market, including the government becomes liquid, and everybody’s happy again.

I know that sounds fishy, but that’s the way it works most of the time. These credit crunches are relatively rare.

That was my question too.

So, someone must have some idea of what this maturity price is/could be?

Scylla, I’ll chime in with another “thank you”! Some of it is still too much for this dumb engineer but it helps a lot.

Again, Bill Gross wrote a great article in Barron’s last week about this. The hold to maturity value is something that shouldn’t be a big problem to calculate. The government simply offers a discount to this level to protect itself.

Realistically, we are talking about a lot of transactions here. Is the government going to get screwed on some of them? Yeah, it’s pretty likely. Are some of the institutions going to screw each other? Yeah, it’s pretty likely. Will there be abuses? Yeah, it’s pretty likely.

It has to be done right, and those abuses have to be minimized for this thing to work.

Again, I’m trying to be mechanistic in my answers rather than trying to sell the plan. It could get screwed up pretty easily. Competance is necessary.