Radical Transparency

In a March 2009 Wired article, Daniel Roth calls for radical transparency in financial reporting as the path to recovery and a more secure financial system.  He argues that the reporting requirements today allow companies to obscure what’s going on and that the way to fix things is as follows.   Embrace a markup language with which bite-sized chunks of standardly defined pieces of financial data are thrown out to the world so that users can crowdsource the true picture of a company’s financial health.

Given my previous post, you’d probably suspect that I think this is a good idea.  And at first blush it is.  But the article misses two important points which render the plan fairly impotent.  The first is that since public disclosure only applies to public companies, the radical transparency plan erodes further the incentive for companies to go (or remain) public,* hence driving the creative accounting underground.  And that’s assuming that such disclosure has the potential of making public companies more financially transparent.**  Let’s assume for the moment that it does have such potential, the next question is, will it?  I think not.

The issue is that it’s not data that produces transparency, but rather data plus analysis.  Yes, you have to start with accurate data for the analysis to be effective, but good data alone is not sufficient to produce the kind of price-correcting (aka naive-investor-protecting) transparency that Roth believes will result.  Or more to the point, good data by itself will enable sophisticated traders and quantitative arbitrageurs to make more of a profit, but that excess profit will be at the expense of the average investor.  In theory — as Roth points out with his LendingClub example — good data levels the analysis playing field, but in practice the field will never be level as long as there is profit to be made by deeper analysis.  The only way true financial transparency will be achieved is if there is incentive for the best possible analysis to be open-sourced.***

Maybe someone here will come up with a clever mechanism (like this one) that aligns the market incentives with the policy goals.  If so, that would be great.

But I think we need an even more radical form of transparency.  I read about the concept a while ago in John Allen Paulos’ book A Mathematician Plays the Stock Market.  Paulos suggests that the concept of insider information should be abolished and that we should allow all employees and stakeholders to buy and sell shares of their own companies without penalty.  A corollary is that there should also be no penalty for giving “inside” information to outsiders (except possibly in cases of trade secret or non-disclosure violation).

When I first thought about Paulos’ proposal, I dismissed it as unworkable and an idea with many bad unintended consequences.  But the more I think about it and challenge my assumptions about what’s important for individual companies to thrive and for the markets to be more bubble-free, the more it makes sense to me.  Thoughts?

* Let’s face it, the only real reason for a company to go public is to provide an exit for early stage investors and liquidity (read giant payday) for founders and employees.  This is a great economic driver and not something I advocate doing away with.  But now there are so many private and pseudo-public ways for stake-holders to get their paydays without the burden of public scrutiny and company accountability that it doesn’t make sense to undermine the incentive to IPO.  If anything, we should be providing extra incentive for companies to be public and to take on the concomitant burdens.

** If it doesn’t, then what’s the point?

*** Note that the real transparency benefits claimed from the LendingClub example came when LendingClub itself decided to post not only the data but also good algorithms for analyzing that data.

  • Steve

    I think data is the right idea, though I think the financial transparency is an example where there would be least benefit.

    The popular example today is people want ‘transparency’ into how the bank bailout money was spent. Assume you have detailed transaction data on all bank transactions before and after the bailout money was given. Now, assume the loans given by the bank increased by 1M and the salaries paid by the bank decreased by 1M. Looks good, bailout money used to fund loans and not to pay bank employees. However, perhaps without the bailout they would have still loaned 1M and instead would have cut salaries by 2M. So, the bailout may have been used to keep on ’employees who got them into this mess’ (see quote from AIG bonus scandal today). btw, I have a similar objection to insider trading - what if you were going to sell stock, but got an insider tip that earnings would be great, you never get caught on this because your insider tip avoided a transaction.

    I can give more examples outside the bailout fund - spent more of my life than I care to admit working with multinational companies and their accounting. I don’t think accounting nor the disclosures we have for public companies is perfect, but I think there is lots of data available. The difference ‘radical transparency’ would make in this area I think is relatively small.

    Now, if you want to use the same argument about improving healthcare - particularly causes for disease, this is where the linked data idea would have the largest difference.

  • kevindick

    I think you’re right but for a slightly wrong reason. Data vs data + analysis is not really the issue. The issue is whether prices incorporate all of the available information.

    Preventing insider trading prevents prices from incorporating the information available to insiders (which less face it is the most up to date and contextualized information). Just like you don’t want to stop short selling because it prevents the market from incorporating information, you don’t want to prevent insider trading.

    Now, you don’t want insiders to be able to perpetrate fraud by purposefully providing public information that is in the opposite direction of their private information. But the solution to this is simple, all insider trades should be immediately made public. So if an insider says good things publicly, you can watch his account to see if he’s buying.

  • Rafe Furst

    I think you nailed it. In my mind the keys (which you explain either explicitly or implicitly) are:

    1) Value has the potential to be created whenever there is an asymmetry; in economic terms this manifests as the “law of comparative advantage”
    2) Value is created via coordinated activity (i.e. cooperation)
    3) Money emerges when mutual knowledge becomes common knowledge about personal value relationships to goods and services

    Here are some related musings of mine: