Very thoughtful post. Everyone who wants "more" banking regulations should look more towards the capital reserve and other leverage measures instead of trying to nit-pick things like banning X fee, Y technique or Z instrument. Markets work when risk is priced at its real level - as opposed to the governments bail-out moral hazard and bankers' group-think.

His conclusion that there are still problems with how risk is weighted is certainly true as well - though I'd say that a lot of this could be solved using super-transparency.

Let require releasing a programmatic model for every instrument that will determine pricing for the underlying assets (i.e. all the mortgages in a MBS would be individually modeled with real data inputs). Then we can have a real idea of the risk from the sum of the underlying assets.

The new requirements on loan level data are fairly extensive and will certainly facilitate micro-analysis of portfolios of asset backed securities. The problem is that this will result in even more financial engineering. The government basically doubled the amount of informaiton that loan underwriters have to take when writing loans. It even requires information such as how far the recipient lives from their primary work place (in the case of home loans).

That kind of information is not easy to come by and it's certainly not easy to incorporate into an analysis. It raises the cost to the loan underwriter, to the loan issuer, to the securitization underwriter, which ultimately goes full circle back to the end consumer (credit card rates are already at 15 year highs due to the Dodd bill). There is also another side effect, this legislation, pariticularly as it pertains to securitizations which fund the majority of home loans and consumer debt, will facilitate further complex financial engineering (the more data you have, the more you're likely to use it). It will also give certain funds the opportunity to discern demographic information, which is currently not "technically" possible but will be a lot easier to do once the flood of data comes in from issuers. What i'm hinting at is that institutional speculators and investors will be able to discriminate against pools by analying income, location of workplace, years of employment (another factor), income history etc.

I may be ranting so I'll cut short here. Long story short, the costs of banking are about to sky high for everyone. I am not sure the American public would support this legislation if they realized just how much it would effect their lifestyle, access to capital, cost of capital, and consumption.

"What i'm hinting at is that institutional speculators and investors will be able to discriminate against pools by analying income, location of workplace, years of employment (another factor), income history etc."

Isn't this a good thing? Aren't these the factors that a bank should consider when deciding to lend money? Wasn't it the lack of such analysis that led to over-valuing mortgage securities that turned out to be worthless?

I'm confused.

You would think so but the reality is that these factors, when enforced, place certain demographic groups at distinct disadvantages. That disadvantage can be augmented when a bank can no longer hide (for all intents and purposes) who they are lending money to.

For example, I could analyze a pool of loans by a Michigan based auto lender. Under current guidelines, i basically get id numbers, fico score, stats on seasoning, outstandnig balance, etc. Who knows where these people live. With new data i could start guessing with reasonable accuracy where these people live (never on an individual basis, but you can get an idea of what counties or towns). I could discern that Bank B is lending to a "riskier" demographic than Bank A. While Bank B cant legally discriminate in their lending policies, the capital markets can certainly discriminate on which notes they want to buy. This would raise the cost of capital to banks lending to "disadvantaged" groups in society (minorities, immigrants, etc).Now a quick fix is for people to analyze the data and make sure that there is no humanly way that analysts like myself can figure narrow down where people in a pool live (which allows us to make assumptions on makeup), but from what i've seen that isnt going to be the case.

Now you're in a predicament. African Americans are over twice as likely to default on a home loan than their white counterparts. Who stands to lose here?

Look its just a thought. Finance is so complex that it's virtually impossible to figure out the second and third layers of unintended "consequences" of monumental legislation like the Dodd bill. Things are still playing out and there are armies of lawyers trying to predict how it will affect capital markets. So much is still uncertain (which explains why the markets are behaving the way they are). One thing isnt, the middle and lower classes are going to get hit hard. Not because the rich arent being taxed enough, but because these regulations are going to make the cost of funding consumption noticeably higher across the board. My group is predicting that within 2 years we will see an end to "free" credit cards that come with rates less than 9%. With risk-based pricing out the window, issuers need to find a way to imbed reserves into their loans or recoup higher potential for loss.

"My group is predicting that within 2 years we will see an end to "free" credit cards that come with rates less than 9%."

I still see this as a good thing. More people cutting up their credit cards is one of the best things that can happen for long term financial stability, in my opinion.

I must admit that I have no interest in the internals of how banks run themselves - I suspect that bankers can "innovate" their way round any potential regulatory restrictions anyway.

However, what I do care about is the moral hazard aspect - if a bank implodes and requires a bail using taxpayers money I sure as hell want individuals who gained from this situation to be held personally responsible. I would require directors of all banks over a "too big to fail" threshold to guarantee that their personal assets are on the line if there is a state bail out.

I think that might concentrate minds a bit so that they work out what is in their own self interest.

More transparency would definitely help, but I am not sure how solvable the problem is at a fundamental level. As Steve Waldman points out (http://www.interfluidity.com/v2/716.html), there is a large confidence interval around any measure of asset worth, and hence of capital levels.

I am not sure what the solution is, but increased transparency, simpler measures such as the leverage ratio, and potentially the separation of complex securities businesses from simpler deposit-and-loan businesses all come to mind.

You could probably say that confidence is a a rule of thumb that uses crowdsourcing (or group think) the assessment of risk - but transparency can serve to improve this even further. Confidence in people/system is derivative (and has feedback from) hard numbers.

I have a banking background, and the articles greatest argument is that 2015 is too far away. Although I agree it's far away, banks rely on so much leverage that unwinding assets can take some time. Mainly mortgage and other long-term loans make 2015 relatively short notice.

Mildly off topic, but still looming large is the bonus system within banking. I know that across the board bonuses are centered around short-term goals. From the CEO all the way down to the lowest teller, the majority of employee and management incentives have to be aligned with the long-term viability of the company. Until this is fixed, the financial sector will continue the boom and bust cycle.

When I read the Black Swan by Taleb the crucial point was to realize how weak we are when we rely on financial experts. Be careful, bad regulations can be opening the door for new creative thinking. For example in Spain the new accounting system gives more freedom to imagination when giving value to assets.