Liquidity Liquidity Everywhere

[mib]

[i]how do you get put back 15 points on a loan with a 2 duration?[/i]



may be it happens because nobody sane enough to be trusted with any money would bid more than 85 for that loan?

[rowdyroddypiper]

"may be it happens because nobody sane enough to be trusted with any money would bid more than 85 for that loan?"



Except it's not distressed, not even close.  Plus if they are writing new business that looks substantially similar in quality, I'd argue they are bidding par.

[mib]

so, are there any real bids above 85 for these particular loans? if not, the rest is irrelevant: we already knew that life is tough, people are stupid etc. if there are - the marks are crazy and in most banks the problem would be solved by escalating the issue to management

[rowdyroddypiper]

There are no real bids for almost anything at this point but in the absence of liquidity it's worrisome that bankers are just picking a number out of thin air to mark you back.  Especially when their business is getting done at par for comparable product at similar spreads.  My point is not woe is me, it's not even that people are stupid, it's that despite the idea that risk management is there to actually manage risks in practice they often just take the most reactionary and draconian view of things.  This is worrisome to me.  When you have people that don't know the business attaching marks that are inconsistent with what the institution is doing on their own book that is worrisome to me.  Plus it's not as if it's beneath banks to take back collateral in a dislocation, sit on it for a few months and sell it at an enormous profit.

[KangaXX]

85 with a 2 duration and you thinks its not crap? Show me.

[kr]

I'm not sure the 'theoretical' comments here are worth much. The trading desk's job is to manage liquidity, which sometimes means not trying to find / force a market clearing price. The job of market risk people is different, though sometimes they construe their job as requiring through demonstration that liquidity could be found at a price for any jar on the shelf at any time of day. The overriding principle is that the entity desires to profit so long as its regulatory requirements are met, and nothing more. So unless a total freeze-up is imminent, there is - and should be - massive latitude for flexibility when it comes to illiquid assets in bad markets. As a side effect, numbers sometimes show on screens which are completely disconnected with reality - i.e. nobody traded there, and no market would clear there (except the one in the textbooks).

[mib]

yes, there are some people whose job is to ensure that a number on a screen equals a price at which one could trade. this concept is called "mark-to-market" and I fail to see how it contradicts entity's purpose. There can be no harm if a number in column entitled "market price" is the real current price and not what the trader thinks could be the price had the world been fairer. What to do with the "riskless" position once it is marked down 20% is an entirely separate issue.

[doctorwes]

The term "price" encompasses several operationally distinct concepts. So does the term "market clearing", I think. For a couple of years, until a few months ago, we had reached the point where the practical difference between these concepts had become small, and this made it easy to confuse them. But that ability to oversimplify and blur distinctions was a function of market conditions at the time. Things have clearly changed.

[kr]

the problem is a bit like black-scholes - conditions for an mtm to exist in the theoretical 'frictionless' sense don't hold absolutely or in a static way. Being asked to report a single and unqualified number leads others to make decisions based on that same dangerously incomplete information - what other purpose does reporting serve? So particularly for me, where my starting point is always far from the theoretical ideal, I see serious risks that have cost others a lot of money (i.e. bad 9-figure P/L result).



At the end of the day, this situation is a modernised agency theory challenge... the agent, who has the biggest information set, can't be trusted by the principals to extract the maximum reward at minimum risk. The contracting arrangement, to no suprise, is a bit in favor of the agent. The MTM metric, especially for the purposes of quantifying risk, enters because it appears precise, but I wouldn't suggest this is the only path.

[Energetic]

I know: it ain't over until it's over. Is it over or not?