ideas-business-openBookFinancialCorporations

one of my theories is that the root of the current financial problem is that companies aren't modeling their credit risk well, especially the credit risk of credit insurance.

consider the following two news snippits:

" "At the 2007 conference he noted that his company worked with a “global swath” of top-notch entities that included “banks and investment banks, pension funds, endowments, foundations, insurance companies, hedge funds, money managers, high-net-worth individuals, municipalities and sovereigns and supranationals.”

Of course, as this intricate skein expanded over the years, it meant that the participants were linked to one another by contracts that existed for the most part inside the financial world’s version of a black box. "

-- http://www.nytimes.com/2008/09/28/business/28melt.html

"AIG is mostly a safe, well-run insurer. But its financial-products division, which accounted for just a fraction of its revenues, wrote enough derivatives contracts to destroy the firm and shake the world. It helps explain one of the mysteries of recent years: who was taking on the risk that banks and investors were shedding? Now we know." -- http://www.economist.com/opinion/displaystory.cfm?story_id=12263158

this makes me think that one problem is that, when trouble hits, financial firms don't have sufficient information about who the counterparties are in each other's derivatives portfolios and so, to be conservative, they consider all other financial firms to be a potentially large credit risk. i've heard other comments to that effect in the news in the past few months.

which makes me think: perhaps this is the perfect time/industry for the application of transparency to corporations. if financial firm X was totally transparent as to all its obligations, then when trouble strikes, anyone holding contracts against that firm would be able to accurately assess whether or not X might be in trouble.

this would make X's derivatives contracts (and all other contracts) more valuable. Consider a non-transparent competitor of X called Y. In case of a financial storm, if you hold a contract to which Y is the counterparty, an accurate model of the contract's worth would now assess a large uncertainty penalty reflecting the possibility that Y is vulnerable to the current storm and will go bankrupt. However, if you hold the same contract with X as the counterparty, after the storm hits you can inspect X's books and make a more accurate judgement as to the possibility of their going bankrupt.

As a counterparty of X, this extra knowledge increases predictibility in the worth of your portfolio in the case of a storm. This increased predictibility should be worth something.

Of course, when you initially buy the contract from X, there is only a small chance that there will be a storm during the duration of the contract. So extra value of predictibility in the case of a storm must be discounted to reflect the small chance of a storm occurring at all.

Still, all in all, X should be able to sell all of its products for a little bit more money than its competitor Y can. In addition, this sort of competitive advantage would be very hard for competitors to match, because the business strategy of an investment bank with open books will probably be very different from the business strategy of an investment bank which relies on secrecy.

In the long term, the advantage would probably become even larger. The regulators would certainly rest easier if people were forced to buy their credit insurance from someone with open books, rather than someone like A.I.G.. Therefore, if open-book investment banks became a fixture, the regulators would probably make rules that explicitly encourage corporations to deal with them, much like some entities are required to hold debt with a high rating. This would entrench the competitive advantage of an open-book investment bank.

Notes: (1) currently, one expects auditors to perform this function; they inspect the open books and certify them. but the current financial crisis shows that, when an unexpected storm hits, people disregard previous auditing results, on the (i think, correct) assumption that the auditors wouldn't have raised a red flag solely on the basis of vulnerability to this particular type of storm, because no one expected this particular type of storm to occur. therefore, there is still additional value in opening one's books

(2) of course, the "man on the street" doesn't have time to comb through your books and identify all of the suddenly-risky trades when a storm hits. only other large financial corporations will have the manpower to make use of this directly. but perhaps some rating agencies would sell reports about you after the storm hit, which would allow smaller investors to benefit also,

(3) of course, this scheme works better the more standardized and machine-readable your books are

(4) i've heard that even today, large financial corporations open their books to their counterparties, under an NDA. However, apparently the counterparties often find it difficult to parse the disclosures, which are not written in any standardized format. So, basically, the amount of human effort required for the counterparty to understand the disclosure grows with the size of the disclosing corporation. So the total amount of effort required for n companies to exchange this information, where each company engages in m separate lines of business, is O(n^2 m), which is large (derivation: each of the n companies must understand each of the m lines of business in each of the other n companies). If, instead, corporations disclosed the information to the public, then a third party could parse the disclosure once and for all, lowering the processing effort to O(n m) (because, for each company and each line of business, one entity must parse the disclosure). And if this system led to the development of machine-readable standard formats for the disclosures, the effort would fall to O(n) (because now there are only m formats to understand in the entire industry, rather than n*m; all the parsing entity has to do now is run their parsing program once for each company; although in this situation, you don't really even need a third-party parser, because the books are already released in a manageable form; however, you would still probably have third-party analyst/rating agencies, which would crunch the numbers in the disclosures for everyone else).

(5) another reason that public disclosure is better than NDA-protected disclosure is that, with public disclosure, you can do n-th order credit risk analyses; i.e. so X is owed money from Y; but what's Y's credit risk? If only direct counterparties get disclosure and you are a direct counterparty only of X and not Y, then you can't well-estimate the credit risk you get from your indirect exposure to Y. But if there is a whole ecosystem of transparent finance, then you get calculate that X is owed money from Y, and Y from Z, etc, and at each step you can do your own assessment of the credit risk introduced.

(6) the same way that open books would allow your counterparties to evaluate your credit risk, you could open all of your email in order to allow others to evaluate your operational risk. this is a more radical change however.