The Next Big Threat

Now that the TARP is dead, there's another threat looming on the horizon. One that lawmakers on both sides of the aisle are attempting to perpetuate on the American people. And the American people are ignorant of the facts. So here they are, free for the taking. Read this - it's in plain English - then write your Members of Congress, and tell them in no uncertain terms how you feel.

What they're proposing is to eliminate the "mark-to-market" accounting rule, at least temporarily. First I'll explain the rule and how and why it came about. Then I'll explain what they want to do, and I'll de-bunk the myths they're floating out there to sell this bad idea. Finally, I'll explain the consequences if they get it done.

What is Mark-to-Market Accounting?
Mark-to-market accounting was ushered in with Statement of Financial Accounting Standards (SFAS) No. 115, back in 1993 (I wrote an industry white paper on it at the time.) SFAS 115 required financial institutions - banks, S&Ls, credit unions, insurance companies, etc. - to account for their investments as follows.

First, they had to classify their investments into three categories, based on the following criteria, and with the following implications for financial statement reporting:

Held-to-Maturity (HTM). For investments classified in this category, the institution has to show "the positive intent and ability" to hold the investments on their books until they mature. They can record these on the books at historical cost, without taking any mark-to-market adjustments through either the income statement or the capital account on their balance sheet.

However, if an institution sells one security out of the held-to-maturity account, the entire portfolio is considered to be "tainted," and it must then mark its entire portfolio to market. So again, it has to have the "positive intent and ability" to hold the investment until it matures.

Available-for Sale (AFS). The institution can sell these securities freely, but for frequent, short-term trading activity, the next category applies. Securities in this category must be marked to the current market value, and the difference between historical cost and market value is recorded as an adjustment to capital on the balance sheet, but it doesn't flow through the income statement so it doesn't affect earnings.

Note that the adjustments go both ways. If the value of the security rises, the result is an increase in book capital, or net worth. If the value of the security falls, capital declines.

Note also - and this is very important - that an institution can transfer securities from AFS to HTM at the current market value, thereby avoiding future mark-to-market adjustments on those securities. But it must then hold them to maturity, or risk tainting the portfolio. We'll come back to this point later.

Trading (T). If the institution employs frequent, short-term trading in and out of a portion of its portfolio (or all of it), those investments get classified as "Trading." The accounting treatment here is similar to that for the AFS category, except that it's the full Monty: the mark-to-market adjustment flows through both the income statement and the capital account on the balance sheet. So gains and losses affect both current earnings and book capital.

Now let's turn our attention to some practical implications.

Early Treatment and Current Practice
Initially, many institutions - especially smaller and more conservative ones, like credit unions, S&Ls, and community banks - classified everything HTM to avoid having to make mark-to-market adjustments. These institutions live and die by their monthly budgets, so they sought to avoid potential losses. They also tended not to sell investments. They invested in short-term government bonds, typically maturing in less than five years, and held them until they matured.

SFAS 115 took effect beginning in 1994. That year, two things were happening in financial institution-land. First, interest rates were rising as the economy grew out of the 1990-91 recession and the subsequent two years of job market stagnation. Bond prices fall when interest rates rise, and vice versa. So institutions were even more reluctant to classify their securities AFS and have to take the mark-to-market losses. (We should note here that these are "paper" losses only, or "unrealized" losses. The loss is "realized" when we actually sell at a loss. Until then, any mark-to-market gain or loss is "unrealized."

The other dynamic in 1994 was that people started saving less now that the recession and the job market crunch were over, and they started borrowing more. So banks, S&Ls and credit unions had less deposits coming in, but more loan demand. To free up liquid funds to make loans, they needed to sell some investments - but they'd classified their portfolios as HTM, and they didn't want to taint the whole portfolio.

So, in either 1994 or 1995 - I forget which, and I'm too lazy to research it, besides which it isn't really important for our purposes - the Financial Accounting Standards Board (FASB), which creates the accounting rules, opened a window of time during which institutions could make a one-time transfer of part or all of their HTM portfolios into AFS, so that they could then sell the securities to free up cash to make loans.

Today, most institutions have seen the light, and carry everything or almost everything in AFS. They recognize that the ability to generate cash to make loans when deposits are scarce overrides the mark-to-market adjustments they may have to make, which they've learned to manage - for the most part.

Dirty Little Secrets
Fast-forward to today. Those smaller institutions are still fine, for the most part, as the vast majority of them avoided the subprime and Alt-A contagion.

But the big boys didn't. They sought the juicy yields offered by bonds backed by those loans. But now, with defaults soaring on the underlying loans, those bonds' values have been decimated. And, having to mark their values to market, big institutions are taking huge quarterly write-downs, affecting at a minimum, their capital, and in many cases, their earnings.

The big banks and Wall Street firms, now aided and abetted by Congress, are calling for the elimination of mark-to-market accounting, at least temporarily. Dave Ramsey - normally a voice of reason - is apparently drinking the kool-aid, and he sums up the argument nicely:

"However, (mark-to-market accounting is) part of what's caused this in the news now. Merrill Lynch was sitting with $30 billion tied up in sub-prime loans with houses. Stupid! They get what they deserve for doing that, and I'm with you on that. Those houses didn't become worthless all of a sudden because (Merrill) couldn't sell their bonds. Since they couldn't sell them, they basically gave them away for 22 cents on the dollar. Now do you think all those houses lost 80% of their value underneath that deal? No, they didn't, so they gave them away for 22 cents on the dollar (about $6 billion total) because there was no market for them. Nobody wants to buy sub-prime bonds because they suck. They're junk bonds. But at 22 cents on the dollar, it's a bargain because even if you foreclosed on every one of the houses in there, you'd probably get $20 billion back out of $30 billion, and so the company that bought those for $6 billion got a deal! But there's no market for them. That's where these companies are stuck. They can't sell this stuff, but accounting-wise, they've had to mark it down to market and it's frozen the marketplace.

Economist (Brian) Wesbury is saying that if we change that one rule and don't force them to mark down to market value and just let them hold on to all the stuff, and say just on sub-primes for this period of time you can change that rule - a temporary change - that'll free the market up. It's seized right now; it's frozen. This will thaw it out and get it going again. He says that'll solve 60% of the problem ... and I think he's right.

That one accounting rule is what made Merrill Lynch sell out. That one accounting rule is what's driving other ones into the dirt. Would you rather let them change their accounting rule or loan them $700 billion for us to buyout their bad paper?"

Myth-Busting
I like Dave Ramsey. He understands personal financial management, and he's fiscally conservative, like me. I respect that. But he doesn't understand how the fixed-income market works, he doesn't understand mark-to-market accounting, and he doesn't understand the institutional investment world. Allow me to de-bunk these myths.

First and foremost, mark-to-market accounting did not cause market values to fall. Mark-to-market accounting adjustments, by definition, come after the market values have fallen. Saying the rules even played a role in the current mess is like saying the diagnosis made the patient sick.

Second, let's examine closely what happened with the Merrill sale. Merrill did not sell its bonds because of "that one accounting rule." Merrill would have been better off continuing to hold the bonds under "that one accounting rule." Here's why.

Merrill - and so many other financials - have been marking their bonds to market, all right. But the market values they've been using have been artificially high. They don't want to take the full write-downs. That's why we get so many ugly surprises after the auditors come in. I know of one large institution that initially reported $7 million in earnings last year. Then, their auditors arrived. They fought the auditors until July (normally, the audit is done in March), and when it was all said and done, they were forced to admit that their initial valuations had been too high, and that they actually lost about $50 million.

Here's one of the dirty little secrets the industry and Congress would have us believe: that financial institutions are having to mark their bonds too low. Ben Bernanke told Congress they're being forced to write investments down to "fire-sale" prices.

If that were true, then when they subsequently sold them, they'd sell them for more than where they were carrying them on the books, and they'd record gains. That has yet to happen in any instance.

Oh, they argue, but they're not selling. Nobody's buying. That's why the prices are so low.

I've said it before and I'll say it again: I see this stuff trade every day. I see bids, and I see offers. And the Merrill sale proves that sales are taking place. The fact of the matter is that institutions aren't marking their investments low enough. That's why they want to change the rule. Next time the auditors come in, they'll take bigger hits. And it will be revealed that they're actually insolvent. But more on that later.

So back to Merrill: why did they sell, if not for the accounting rule? They sold because, quite simply, they were running out of money. Nobody wanted to lend to them anymore, because they were destined to be the next Bear Stearns or Lehman. So they had to sell something to raise money. And the 22 cents on the dollar they got belies the fact that they had to lend much of the money to the eventual buyer. Factoring in the loan terms, they actually sold for about six cents on the dollar.

As for Ramsey's assertions regarding the recovery potential on the homes whose mortgages underlie these bonds, we need to understand how these bonds are structured, and what the actual collateral is.

Financial "Innovation"
Take a bunch of mortgages, bundle them together, and you have a mortgage pool. Now, let's say those mortgages are all subprime loans, and all the houses are in LA, or Vegas, or Phoenix - three of the hardest-hit markets in the housing bust. What do you think the chances are that such a combination of concentration and credit risk could earn that pool a triple-A rating? If you answered, "slim to none," you get a gold star.

So, we have to slice up the cash flows. We create a "senior" bond, which has the greatest protection from the collateral's risk, and it gets the AAA rating. How do we do that? We create some crappy, risky bonds called support bonds that get low ratings, and they take the first hit when the defaults start rolling in. On top of those, we create some less risky "mezzanine" bonds that take the secondary hit if the supports all go away through defaults. Those get an intermediate rating.

The problem is that the ratings agencies - like most of the financial world - missed the boat when it came to foreseeing the inevitable correction to the housing bubble. They didn't anticipate the high level of defaults. They didn't anticipate the complete erosion of the supports, which has happened, and the mezzanines, which is happening even as we speak - and rapidly. Thus, they didn't anticipate the threat of real losses to the senior bonds.

Ramsey's assertion that when you sell the homes you'll get $20 billion back out of $30 billion - a two-thirds recovery rate - is consistent with home price trends. Home prices in Vegas are down 33% from their peak at last count; in LA, they're down 30%. But for one thing, they're expected to fall further. And for another, many foreclosed homes fall into disrepair or get vandalized - often by the former owners on their way out the door. Their neglected swimming pools become pestilent breeding grounds for mosquitoes. There are significant clean-up and repair costs to be borne by the lenders before they can be sold, which reduces the recovery rate, as do real estate commissions.

But the real disconnect between Ramsey's two-thirds recovery on the eventual real estate sale, and the value of the bonds themselves, has to do with the bonds' structure. Let's understand that better.

If Merrill owned the support bonds, which erode first, they probably are worth only six cents on the dollar. Since they take the first hit from defaults, their value could fall all the way to zero. By the time there's any recovery from selling the foreclosed homes, the supports - and the mezzanines, in many instances - are long gone, and the only recovery will accrue to the senior bonds.

But the senior bonds are being rapidly downgraded, due to the erosion of the supports and the mezzanines. That has contributed to their valuations. Investors initially paid up to get a AAA rating on a bond that's now junk-grade. In essence, they overpaid because the ratings agencies got it wrong.

Also, some of the bonds had further "credit enhancements" to get up to the AAA level. That often came in the form of bond insurance. The biggest bond insurers are MBIA, Ambac, and FGIC. You may recall seeing those names in the news. The reason is that they, too, have been downgraded due to bad bonds they themselves own. Thus their ability to make good on the mounting claims made by holders of bonds they insure has been impaired. And that has resulted in further write-downs of the bonds they insure. Again, investors paid up for the insurance "wrap" that gave the bonds they bought a triple-A rating. Thus the value has been decimated.

Finally, in some instances the collateral isn't first mortgages, it's home equity lines, or HELOCs. The problem there is that if there's less than a 100% recovery rate on the sale of the home, the first mortgage holder gets all that is recovered, and the HELOC holder gets nada. I know of one HELOC-backed bond that has FGIC insurance, and has been downgraded from AAA to junk status. The issuer? IndyMac, the failed California lender. The price is currently being quoted at 26 cents on the dollar.

Why is this last bond an important example? Because thus far, not a single payment to the bondholders has been missed. That's why investors argue the bonds' market values are excessively pessimistic: "But the bonds are performing! They're paying as scheduled!"

Well, that's not how bonds trade, as I've noted in a previous post. Bonds trade based on expected future cash flows. And the future cash flows on a California HELOC-backed bond issued by a failed lender and insured by a downgraded and failing insurer, whose support bonds are gone and whose mezzanine protection is evaporating rapidly, don't look promising. If anything, a price of 26 may be optimistic.

Remember when I said Merrill may have benefited from "that one accounting rule" by holding the bonds? That's because they could have marked the bonds pretty much wherever they wanted, at least for a while. Once they needed cash so desperately they were forced to sell, they had to take the real market value, which is whatever the next guy will pay for the bonds. Having hinted at that distinction, let's examine pricing sources.

"Mommy, Where Do Prices Come From?"
Pricing relatively illiquid and structurally complex fixed-income securities is nearly as much a fairy tale process as the stork or the cabbage patch.

Theoretically speaking, you project the cash flows into the future, discount each cash flow at the discount rate associated with when it's projected to occur, and sum it all up in a net present value calculation (if you don't understand that, take it on faith).

Where you run into difficulty with this process is in projecting cash flows that can prepay. So you make assumptions - based on a model built using past prepayment behavior - as to when those cash flows will occur, factoring in prepayments.

Prepayments can occur when a mortgage borrower refinances. By the same token, if he defaults, the mortgage may never get paid off, at least in full.

Prepayment models don't always do a great job of factoring in when those things will happen based on past experience, because sometimes we encounter new territory in terms of the economic conditions that influence prepayments. To wit, until a few years ago, in the history of mortgage securitization, we'd never experienced a one percent Fed funds rate, which resulted in mortgage rates so low for so long that refinancings reached previously unfathomable levels. The models just weren't calibrated for that scenario.

Likewise, we never anticipated the housing slump we're in today (okay, some of us did, but I'm no longer in the business of building prepayment models).

The prepayment models are used to determine when the cash flows will occur. The timing of the cash flows determines their net present value, which is the "modeled price" of the bonds. Different investment firms have different prepayment models they use, thus their "modeled prices" may differ. Obviously, when they're faced with two different prices from two different models, they choose the price that results in the smallest unrealized loss.

The problem with using modeled prices is that models don't buy bonds (except maybe Gisele, but she'll want those denominated in euros, thank you very much). The actual market value is what the next guy will pay you for the bond. Absent a live bid, though, you've just got to go with the modeled pricing.

That's how the financials overstate the value of their bonds. They use the most optimistic assumptions possible to calculate the highest possible price. Until their auditors show up.

The Threat
Okay, we now understand mark-to-market accounting. Simply put, it's designed to introduce price transparency. Why are the financials opposed to that?

Because they don't want the investing public, their auditors, or the ratings agencies and regulators to know how little their bonds are really worth.

At a time when most market participants are clamoring for greater price transparency in markets for things like credit default swaps (a whole 'nother topic), the financials and Congress and even Dave Ramsey now want to take an eraser to it. Why is that a bad idea?

Look at a graph of WaMu's or Wachovia's or Lehman's or Fannie Mae's stock price over the last year or two. It will look much like a ski slope, if we smooth it out, beginning to decline in July 2007, when the Bear Stearns hedge fund melt-down started the credit crisis, eventually approaching zero when it failed or was acquired.

Why the gradual decline? Simple. With each quarterly earnings release, we learned of further write-downs. We had transparency about what the company was really worth. So we could factor that into their stock price. The ratings agencies had an early-warning mechanism to be able to downgrade those firms' own debt ratings, so that unsuspecting investors didn't buck up to buy their bonds, thinking they were safe. The regulators also had an early warning system, so they could step in before any depositors lost money.

Absent mark-to-market accounting, those firms would still eventually run out of capital. They'd become insolvent. They'd fail. Their poor investment decisions caused that to happen, exacerbated by a now-difficult lending, bond-selling and underwriting environment (which they brought upon themselves) that is crimping revenues.

So what would their price graphs look like? Well, without transparency, instead of a ski slope, you'd get a cliff. The price would perk along merrily at a high level until it dropped immediately to zero.

No early warning for the ratings agencies. No early warning for the regulators.

No early warning for investors.

And that last point is the most ominous one. Absent price transparency on the financials' bond holdings - and thus the financials themselves - no intelligent investor is going to buy that debt, or provide capital or credit to the financials.

You think credit markets are frozen now? How about a scenario where no one knows what anything is worth, from subprime bonds to bank stocks? Hey, if the regulators can't assess what my bank's worth, I'll be darned if I'm going to keep any money deposited in it. And the bonds that are trading now - the very bonds that the financials say aren't trading - will truly stop trading. And then the waters will get really murky.

Transparency is a good thing. Early warning mechanisms always are. I'm thankful for NOAA hurricane tracking reports and tornado sirens.

In summary, think about just the period from last Friday to yesterday - basically a four-day weekend. WaMu, Wachovia, Fortis, Dexia, Glitnir, and Bradford & Bingley - all financial institutions, from the US, the UK, Benelux and Iceland - effectively failed. They were either taken over or bailed out by their respective governments, or bought by another bank just before they would have imploded.

We knew it was coming. We'd watched their respective stock price declines. We'd seen one credit counterparty after another snub them. We'd seen them scramble for suitors.

They all failed, either Friday in the case of WaMu, or over the weekend in the case of the others. What was the market reaction to this unprecedented weekend of global bank failures?

The Dow opened down Friday from Thursday's close, but rallied to finish up about 160 points on the day. Monday morning, it opened down about 150, and fell another 150-200 points during the morning. But that had more to do with concerns the bailout wouldn't pass than the failed financials, which the market largely took in stride.

The Dow Jones Euro Stoxx Index behaved similarly, as did the UK's FTSE 100, as all eyes were on the US bailout vote, which would cover foreign banks as well.

Why were the markets calm? Because the gradual price declines of those firms' shares were already priced into the market. In short, we expected their demise, because price transparency had revealed their long, slow death march.

Now, imagine the blood-bath that would ensue if all of those names failed in a span of four days - only two of them being trading days - if the declines in those firms' values had not been anticipated. If they perked merrily along at a price of say, $45 a share in the case of WaMu, then suddenly went to zero. Overnight. All six of them at once.

There would be panic in the markets. Yesterday's 800-point sell-off would look like a rally. There would be a global run on banks, causing a financial collapse, as the insurance funds couldn't contain the carnage.

But see, investors would know this in advance. They like price transparency. So they'd simply stop investing, thus freezing the credit markets. Depositors with a lick of sense would go ahead and withdraw their funds anyway. So we'd have a true credit market freeze, not just the imagined one that the financials are whining about now.

Is credit near-frozen, though? To be sure, it is. But that's not a bad thing. It means you probably can't get a loan today unless you have squeaky-clean credit and not too much existing debt.

That's as it should be. That's as it should have been all along. We wouldn't be in this pickle had that been the case three to five years ago.

The fact is, if you're reading this from somewhere in the US - or even the UK, or Europe, or Australia - you probably fall into one of two categories.

Either you already have more debt than you should, and you need another loan like an overdose patient needs a fix, or you live so well within your means and are so content with what you have that you have no need to borrow money.