Mark-to-Market vs. Mark-to-Model: What Ever Happened to Real Value?

Banks Get New Leeway in Valuing Their Assets:

The change seems likely to allow banks to report higher profits by assuming that the securities are worth more than anyone is now willing to pay for them. But critics objected that the change could further damage the credibility of financial institutions by enabling them to avoid recognizing losses from bad loans they have made.

Here are two good summaries of why it is worthwhile to keep marking-to-market: from CAP and from Baseline Scenario. Mark-to-market versus mark-to-model is an incredibly esoteric and dorky thing to talk about, even for accountants and actuaries, but the underlining idea of the debate is easy to think about in other contexts - let me give you some in a less serious context for Friday.

The Baseball Card Bull Market of the Early 90s

So in the early 1990s, when I was in Junior High, there was a craze about collecting and trading baseball cards. Our classroom would have a corner during lunch where we’d all compare, with our binders and those 3×3 plastic containers for cards, who had what, and we’d trade back and forth accordingly. And of course we had our model, The Model, in fact, the Black-Scholes of our trading:

So while other kids were mowing lawns or delivering newspapers, I decided I was going to make my profit by arbitraging the volatile Frank Thomas rookie card market. I took a highly leveraged position in the Upper Deck Frank Thomas rookie card - I borrowed against future allowances, and bought several cards for $7 each from a kid who wanted to get out of collecting baseball cards in order to try hanging out with girls (loser!).

Upper Deck is like the AAA of baseball cards. The guide said that these cards were worth $9. Buy at $7, sell at $9, instant money. My dad took me to the convention center, and I was all ready to make some cash money, when I found out that all the tables were only buying them for $5. Sensing my frustration (and also perhaps worried, since, like the FDIC or those with a savings account, he was providing all the leverage for me), he asked me, “wait, what are those cards worth again?”

I answered that they are worth $9. That’s what the guide, my model, says they are worth. No doubt those guide values are created by the most brilliant minds available. My dad, not in business or finance, was very clear in trying to explain to me “no son, they are only worth what someone is willing to pay you for them.” I responded that this card convention center was completely wrong in how they were valuing my baseball cards. I was but a little financial engineer back then; now I would have known to say “Dad, clearly the Sox are having a bad season, and/or this isn’t the time in the year-long sporting cycle when demand is reasonable for baseball cards. I don’t feel I should get punished for the normal ups-and-downs of the baseball cycle.”

I may have also noted that the flood of crap Fleer-brand baseball cards, the Mortgage Backed Security of its day, was destroying liquidity in the market, but that I had the trust of then Treasury Secretary Brady to start buying up those crappy baseball cards and get them transferred onto the government’s balance sheet.

So who was right? Me, with the model of the Baseball Collectors Guide and the excuse of the business cycle, at $9? Or my dad, who says they are worth whatever someone is willing to pay you in an open market, at $5?

How you answer that question should color how you are disposed to marking assets to the model, versus marking them to the market. Mind you, this was not just an academic exercise - at $5, my dad realizes I’ve made some terrible calls, and is probably going to make me start mowing lawns until I can pay him back. If I could convince him they were worth $9, I would not have to mow lawns (which I sincerely did not want to do) but instead could probably borrow some more…

Dorm Room Debates

I was chatting with a distinguished older businessman I was introduced to; educated with an MBA, brilliant business mind. He was complaining about Mark To Market.

Businessman ((BM)): The problem with Mark to Market is that the market can’t get the true value of an asset all the time.
Me: But isn’t the true value what the last person is willing to pay?
BM: Yes but what about assets that are longer than the business cycle, that are getting hammered over the short term? And isn’t it a backwards way of viewing things, from a risk-management perspective?
Me: I worry too that Mark-to-Market is pro-cyclical instead of anti-cyclical - it encourages risk-taking when things are good when solid risk-management requires leaning against the business cycle. I’m more than open to improvements there, but isn’t the general idea of marking to market the right basis?
BM: Not necessarily. What if the true value of the asset is higher than what the market wants to pay for it though?
Me: You keep saying true value, and I don’t know what that means in the context of assets outside of markets. I’m not actually sure if I know what it means for anything… [preparing to drop the Rorty bomb!]
BM: It means what it is actually worth, really worth, independently of markets and cycles.
Me: Wait, what? Is this like a Neo-Kantian thing?
BM: Wait, Huh? No, it is like any asset has a value, an essential value that it really holds outside of people, who may be confused or constrained or otherwise unable to acknowledge the value of the asset, bidding on it. The people holding that can know that better than the market sometimes.
Me: But isn’t the whole point of a market economy that commodities find their essential value within the bidding and trading mechanisms? Isn’t the informational aspect of markets like, the whole reason we have capitalism? Even if I grant your point, insider agents working with secretive ‘models’ shouldn’t be considered the best information agents here. Stocks and bonds valuations don’t, or shouldn’t, exist in an a priori plane; the fact that they go through the business cycle like everything else is factored into their valuation by markets.
BM: But what if markets are wrong, because they are blinded by short term interests or their own illiquidity, to see the essential real value of the assets? What if they can’t understand the additional information, or ignore it, or don’t recognize it as information?

At this point, I almost expected him to say “the problem is that all we can see is the shadow of the assets projected on a wall, not the real Form of a bond comprised of $500k loans to junkies. If only we could see them outside the limited cognition of our cave, see these mortgage-backed securities in the light of the actual Sun….”

Want to know how I know your markets are f**ked? Because distinguished men of business and finance sound like hipsters trying to make their way through a seminar on epistemology.

Comments

[Lilguy:]

This same discussion occurs all the time about every asset. The one we most commonly see is that the stock market is "under valued" or "over valued." Gold & other commodities are similarly characterized.

Well, folks, they're not either. Assets in a generally competitive capitalist system (where the market has priced in all information) are the value a willing buyer & seller agree to. Their value is not determined by analysis or models or astrology or tea leaves.

Now one may believe these assets may appreciate or depreciate in the future and invest accordingly, but at the moment, they are what they are. And sometimes one's assessment of the future value of the asset is right, sometimes its wrong. That's where risk enters the picture--and it can take many forms.

Apr 18 10:08 AM

[Mike Razar:]

Guess what, children. It makes no difference what accounting rules you use. A balance sheet consists of lots of data. An accounting number (like an assigned asset value) is an arbitrary calculation made by, umm, an accountant. Why not act like capitalists and let the market decide? Clearly MTM is ambiguous in illiquid markets. But it is usually possible to estimate various mtm values. The problem is that ALL valuation models are flawed. They rely on a collection of unverifiable parameters. The oft cited "discounted cash flow" model is a good example of this. You must choose a discount rate. Do you think you know the correct one? Is there even a correct one? At best there is a probability distribution of various discount rate values. Then there is the probability distribution explaining future defaults. Do you know how to calculate those? I sure don't. The way to trade profitably is to disagree with someone else's probabilities and be right, or lucky.

So let a thousand flowers bloom. If you believe in transparency, report a variety of "values" for an assset, each based on a clear set of assumptions. Use a cash flow model, but identify it as such. The same with a mtm value. The same with Barney Frank's opinion of its value. Maybe then we can end this stupid argument once and for all.

Apr 18 10:12 AM

[dcb:]

alas, I think it will never be resolved. Opinions on this matter will not change that is for sure. I do like how in every "crisis" the fault isn't with the people who made the mistakes, it is in some rule or almost anything besides their own mistakes. After all the masters of the universe don't make mistakes do they. when ever markets behave irregularly (which they do on a regular basis) it is the fault of the market. Even the idea makes me laugh, and the level of denial makes me laugh more.


On Apr 18 10:12 AM Mike Razar wrote:

> Guess what, children. It makes no difference what accounting rules
> you use. A balance sheet consists of lots of data. An accounting
> number (like an assigned asset value) is an arbitrary calculation
> made by, umm, an accountant. Why not act like capitalists and let
> the market decide? Clearly MTM is ambiguous in illiquid markets.
> But it is usually possible to estimate various mtm values. The problem
> is that ALL valuation models are flawed. They rely on a collection
> of unverifiable parameters. The oft cited "discounted cash flow"
> model is a good example of this. You must choose a discount rate.
> Do you think you know the correct one? Is there even a correct one?
> At best there is a probability distribution of various discount rate
> values. Then there is the probability distribution explaining future
> defaults. Do you know how to calculate those? I sure don't. The
> way to trade profitably is to disagree with someone else's probabilities
> and be right, or lucky.
>
> So let a thousand flowers bloom. If you believe in transparency,
> report a variety of "values" for an assset, each based on a clear
> set of assumptions. Use a cash flow model, but identify it as such.
> The same with a mtm value. The same with Barney Frank's opinion
> of its value. Maybe then we can end this stupid argument once and
> for all.

Apr 18 10:36 AM

[dcb:]

Maybe I just don't get it. You want to sell me a bond. Based upon a model I use I will pay a certain amount because the cash flows are worth a certain amount. Look if these bonds were such a great deal, people would be buying them. the fact that they aren't says all that needs to be said.


On Apr 17 09:45 PM mathgeek wrote:

> Article adds nothing to the debate.
>
> I think one nice way to test your conviction about mark to market
> is this - Since house prices were transacting in an open market,
> do you agree that house prices were totally appropritate and correct
> to use as a basis for lending at the peak of the so-called bubble?
> (And I say so-called, because if the most recent market price is
> really the only indication of value, there cannot be such a thing
> as a bubble)
>
> There are a myriad of other issues with mark to market - for example,
> m2m instantly amortizes expected changes to a future cash flow to
> present earnings... kind of the opposite of the time matching that
> accounting is supposed to acheive...
>
> And that leads to the ultimate question here... what exactly are
> you trying to acheive? If you think market to market makes it easier
> to evaluate a bank's real financial situation, you have either never
> done it or are a fool... and have no business trying to understand
> a bank balance sheet. On the other hand, if you are a trader or
> a hedgie who is trying to make a buck exploiting the extra volatility
> created by mark to market, it makes perfect sense.
>

Apr 18 10:41 AM

[henarl:]

There is a big difference between an asset that just sits there earning no income (like a baseball card), and one that that pays an income while holding it (like a mortgage). If a mortgage defaults and stops earning that income, THEN mark it to it's instant market value, but not before. As long as it has income earning power it should be valued based upon that and it's likelyhood of continuing to earn said income.
MTM is only valid for non earning assets.

Apr 18 12:53 PM

[Mike Razar:]

dcb, you are basically correct in your cynicism. If shareholders and Boards did their job, none of this could have happened. (www.americanthinker.co...)
If you re- read my first post, I think you will see that I agree that masters of the universe do make mistakes. I learned on Sesame Street that everybody makes mistakes, including me. So let each potential investor decide how much value to give to "ambiguous" assets. Why let FASB or Tim Gethner or Barney Frank decide? They know less than a potential buyer who does his own due diligence. I am a dedicated free-market fanatic. Markets can only fail if there is hidden information or if an outside force imposes unreasonable rules or limitations on who can participate and at what cost. (Think of fixed commissions on exchanges circa 1970.)


On Apr 18 10:36 AM dcb wrote:

> alas, I think it will never be resolved. Opinions on this matter
> will not change that is for sure. I do like how in every "crisis"
> the fault isn't with the people who made the mistakes, it is in some
> rule or almost anything besides their own mistakes. After all the
> masters of the universe don't make mistakes do they. when ever markets
> behave irregularly (which they do on a regular basis) it is the fault
> of the market. Even the idea makes me laugh, and the level of denial
> makes me laugh more.

Apr 18 01:24 PM

[William Cowie:]

Like dcb said. that's as far as true market value is concerned.

As for accounting, that's a different issue. There is no perfect system of asset calculation for accounting purposes. Never has been, never will be. Let's just get that out of the way. Hundreds of years of accounting policies, tried and true, accepted universally, are based on "historical cost" asset valuation. Granted, it's not perfect, but its imperfections are well understood by millions and it delivers that elusive goal: consistency across industries and year to year. Once an asset is on the books, it never changes value. The sole exception is where a very unusual and significant event occurs, in which case any adjustment is only down, never up.

As I said, it's not perfect; no single accounting system is perfect. But historical cost is like the gold standard: everyone understands it, and has learned to work around its imperfections, because everyone knows where they're starting from. Mark to anything is just too slippery, just too inconsistent, too imprecise. Nothing else will ever work any better than old faithful, which has served us well for hundreds of years, in boom and bust.

There's no dishonor, only wisdom, in bringing back something everyone understands (warts and all) and has served so many industries so well, in so many countries, and over so many years.

Apr 18 02:40 PM

[Mike Razar:]

Bill, Would you put up your own money to loan the "historical value or close to it for an asset whose current value seems lower (by whatever criteria you like).? Of course not. The thing about mtm is that, while imperfect, it has the virtue of passing the "your own money" vs. "other people's money" test. (YOM vs. OPM). Why doesn't anyone react to my idea to provide a range of values according to all the various valuation methods available and let the investor choose which to use.?


On Apr 18 02:40 PM William Cowie wrote:

> Like dcb said. that's as far as true market value is concerned.
>
>
> As for accounting, that's a different issue. There is no perfect
> system of asset calculation for accounting purposes. Never has been,
> never will be. Let's just get that out of the way. Hundreds of years
> of accounting policies, tried and true, accepted universally, are
> based on "historical cost" asset valuation. Granted, it's not perfect,
> but its imperfections are well understood by millions and it delivers
> that elusive goal: consistency across industries and year to year.
> Once an asset is on the books, it never changes value. The sole exception
> is where a very unusual and significant event occurs, in which case
> any adjustment is only down, never up.
>
> As I said, it's not perfect; no single accounting system is perfect.
> But historical cost is like the gold standard: everyone understands
> it, and has learned to work around its imperfections, because everyone
> knows where they're starting from. Mark to anything is just too slippery,
> just too inconsistent, too imprecise. Nothing else will ever work
> any better than old faithful, which has served us well for hundreds
> of years, in boom and bust.
>
> There's no dishonor, only wisdom, in bringing back something everyone
> understands (warts and all) and has served so many industries so
> well, in so many countries, and over so many years.

Apr 18 03:16 PM

[oldtimertoo:]

If you can't grasp the concept that "price is what you pay - value is what you get", then you have no business investing in the stock or bond market. Maybe you should stick to trading baseball cards.

Banks and hedge funds don't want to buy mortgage securities because they are subject to unpredictable marks. They don't want to sell the ones they own because they perceive way more value than the occasional, opportunistic bids.

Apr 18 04:59 PM

[ed233:]

I don't see how the market is all that difficult. And when it comes to valuation of the banks toxic assets mark to market valuation helps nobody including the banks other than perhaps to enable them to buy some time hoping things will turn around.Toxic assets are intrinsically flawed and will continue to be flawed unless and until real estate values start to recover and there is a turn around in the employment situation. Unfortunately at this present time the two ingredients making up the price for toxic assets are going south with no signs of abating. If you buy these toxic assets there is a good chance that your valuation is depleting and who knows for how long. That being the case buyers will offer prices only at substantial discounting. That is the price in the open market. And banks who are forced to sell could find their financial position as being insolvent without further government assistence. The market finds the true value by the interaction of buyers and sellers arriving at a mutually agreed upon price. The buyers are there but the sellers can't afford to sell at the present market offering. Maybe the government could wave a magic wand and make the price something other than what the market is demanding or let on they're baseball cards.

Apr 18 05:33 PM

[CaptainJJack:]

This example is a little distorted.

The rule I understood them to have followed before the mark to market rule was a mark to model, BUT with an important qualifier:

The model used had to be the same model you normally use to determine the price to BUY the security if it were offered for sale, AND it has to be validated against past security purchases, as well as current market prices, if available.

Apr 18 06:05 PM

[whisperonth...:]

I don't know who is right, or who is wrong, but I do know that if you can't manage to write the article without using what I consider profanity, you aren't saying anything worth reading. Part of education is in being able to get your point across using words equal to the knowledge you wish to impart. The level of your knowledge sinks to the level of your language. Clean it up. Then maybe I'll be interested in reading it.

And I personally don't care if "everyone says it." That still doesn't make it acceptable. It just makes it garbage.

Apr 18 10:42 PM

[whisperonth...:]

I don't know who is right, or who is wrong, but I do know that if you can't manage to write the article without using what I consider profanity, you aren't saying anything worth reading. Part of education is in being able to get your point across using words equal to the knowledge you wish to impart. The level of your knowledge sinks to the level of your language. Clean it up. Then maybe I'll be interested in reading it.

And I personally don't care if "everyone says it." That still doesn't make it acceptable. It just makes it garbage.

Apr 18 10:42 PM

[AndOrNot:]

Let me see if I have this, very complex, M2M stuff straight.

Borrowers purchased homes at three-four times fair market value, using “sub-prime instruments” and “stated income” while padding the pockets of Realtors, Appraisers and Lenders along the way, who only winked at each other in rapacity, while shirking all fiduciary and professional responsibilities.

And NOW the Lenders want us to honor those phony, self-serving price hikes, so they now can “State” their-own income, as they again, feloniously, inject into that premise, that, WE are just “Too Big To Fail.”

Like we cannot live without giant liars, frauds and cheaters among us?

Wake Up World! The euphemism “Too Big to Fail” just might (seemingly) mean “Too Evil to Resist.”

Apr 18 10:47 PM

[imnotadoofus:]

M2M is based on the efficient market hypothesis. That theory is no longer assumed to be valid, especially so after the past few years. Behavioral Finance would suggest that emotional reactions may have more impact on valuations than a discounted cash flow model. So modification of M2M is a realistic adjustment and will not result in erroneous wasting of capital. Remember, there are three levels, and level I and II will not be affected by phoney assumptions as there are market transactions that confirm value. It is the level III that must be adjusted and not be confined to ABX or CMBX derivative indices.

Apr 19 12:31 AM

[Friar Hilarius:]

If I understand it correctly America built too many dwellings for too few buyers and found them hard to sell

So the surplus was sold to people who couldn't afford to buy them at the previously "inflated" prices when supply and demand were relatively "in balance"

When the indolent, and unhealthy, and unemployed couldn't keep up with their payments (which they could never have afforded anyway) the dwellings were confiscated

The banks then moaned and groaned that they had a surplus of dwellings which again no one could afford

So they asked Big Daddy in Washington for some cash to tide them over in making bonus payments to their world beating executives

Why is this a better model than Christian Communism which is well described in the Acts of the Apostles :-

All the believers were one in heart and mind. No one claimed that any of his possessions was his own, but they shared everything they had. With great power the apostles continued to testify to the resurrection of the Lord Jesus, and much grace was upon them all.

There were no needy persons among them. For from time to time those who owned lands or houses sold them, brought the money from the sales and put it at the apostles' feet, and it was distributed to anyone as he had need. [Acts 4 32-35]

Have I missed the point of where capitalism has failed?

Friar Hilarius

Apr 19 01:12 AM

[Kinabalu:]

A center-left radical with a sense of humor. Normally I wouldn't like it but in this case I can't resist. Huzzah!

Apr 19 08:53 PM

[Ravi Nagarajan:]

The fundamental problem of mark to model is the inherently subjective nature of defining the model. Someone earlier wrote about Buffett and the long dated derviatives. These derivatives have in fact been totally misunderstood (I wrote about it here: www.rationalwalk.com/?...), and Buffett DID in fact say that the Black-Scholes computation made no sense, but nonetheless, he supports the concept of Mark to market in his latest letter.

Of course, Black-Scholes itself is a model. But it is a known and accepted "yard stick" so to speak. When I read a financial statement, I don't want to have to read different models to figure out just what is on the balance sheet. I'd like a yard to be a yard, an inch to be an inch, etc. It is important to have a common language, set of assumptions, etc in accounting.

This does NOT mean that these standards are always right. It does NOT mean that market prices are always right. But the reader of a financial statement can make those decisions using the statements as a baseline knowing that a common set of assumptions were used for all similar companies rather than subjective models invented by each firm. From an investment perspective, it is the job of the analyst to figure out when market values do not reflect actual intrinsic value and to act accordingly.

Now of course, mark to market is causing all sorts of problems related to the levels of capital required for a financial firm to be considered solvent. I think you deal with this by modifying the manner in which regulators examine financial institutions. Wouldn't it make more sense to have regulators loosen the standards (assuming they believe that mark to market does not reflect true capital position) rather than keeping the regulatory standards static and messing with the accounting such that investors and others cannot compare the statements of similar firms?

Apr 19 09:07 PM

[petyaczar:]

mike your comment "It makes no difference what accounting rules you use." has to be the most ignorant and ridiculous comment I've ever heard regarding financial reporting. Perhaps you misspoke.
M2M = the change to the accounting system= has resulted in the loss of $11TRILLION DOLLARS in the market value of holdings by the American people AND ALSO the establishment of Socialism over the free market system. A government Coup deTat.

Fascinating!!! I will assume you misspoke, miswrote, and/or were suffering an ischemic event of some sort.

ROFLMAO....

PS If the accounting system didn't matter, then why use one at all, why not abolish it (The system) and allo each individual entity to use their own individual accounting system.

Your comment was IMO ridiculous in the extreme. -if that's possible.


On Apr 18 10:12 AM Mike Razar wrote:

> Guess what, children. It makes no difference what accounting rules
> you use. A balance sheet consists of lots of data. An accounting
> number (like an assigned asset value) is an arbitrary calculation
> made by, umm, an accountant. Why not act like capitalists and let
> the market decide? Clearly MTM is ambiguous in illiquid markets.
> But it is usually possible to estimate various mtm values. The problem
> is that ALL valuation models are flawed. They rely on a collection
> of unverifiable parameters. The oft cited "discounted cash flow"
> model is a good example of this. You must choose a discount rate.
> Do you think you know the correct one? Is there even a correct one?
> At best there is a probability distribution of various discount rate
> values. Then there is the probability distribution explaining future
> defaults. Do you know how to calculate those? I sure don't. The way
> to trade profitably is to disagree with someone else's probabilities
> and be right, or lucky.
>
> So let a thousand flowers bloom. If you believe in transparency,
> report a variety of "values" for an assset, each based on a clear
> set of assumptions. Use a cash flow model, but identify it as such.
> The same with a mtm value. The same with Barney Frank's opinion of
> its value. Maybe then we can end this stupid argument once and for
> all.

Apr 19 10:26 PM

[Whippet:]

Ravi- interesting article at the other end of your link. Seems to me the only variable that Buffett has a problem with in the Black-Scholes equation is volatility, and he thinks the predicitive premium values of the BS equation are "ridiculous." Seems to me by his quote that the world's greatest investor is a little touchy on this subject. Fact is, even if you use a similar volatility window of 10-20 years, the number is HIGH. If he thinks he has found mispriced future volatility, wonderful- make the trade. The only responsible accounting model one can use when valuing sold puts 10-20 years in the future is a complete and utter loss, as if the markets went to zero, plus the premiums collected. Write it off the books now, and show a windfall gain if indeed they expire worthless then. If this is unacceptable, hedge. He has no business whining about the Black-Scholes model- it works. And he can laugh all the way to the bank when they expire worthless... I know if I could sell puts on SPX or QQQQ for 2020 (and I can't), I would have to put up rolling collateral based on the current SPX or QQQQ mark. He should be even more conservative, as the CEO of a publicly traded company.

Apr 20 01:58 PM