Risk-weighted Capital: Whose Model is it anyway? | Hacker News Reader

Risk-weighted Capital: Whose Model is it anyway?

economist.com

23 pointsakg_6710 comments

This article basically just outlines regulatory capital and (I think) avoids economic capital. In very very shorthand - regulatory capital is the capital needed to comply with local regulations (with guidance from BASEL and BIS), while economic capital is the more nuanced calculation used to calculate the risk of capital using the bank's own deduction and modelling. A great metaphor for the actual difference is "Regulatory Capital is to Economic Capital as Road Test Is to Driving" (http://www.americanbanker.com/bankthink/regulatory-capital-r...). Good banks use both, alongside other metrics to evaluate deals.

Economic capital takes into account the particulars of individual companies, while regulatory capital takes a broad brush approach. In reality, using reg capital leads to set ratings without taking into account the actual risk of the asset to the company. I would argue that the best approach would be economic capital as the main capital calculation with review and monitoring by regulators of the model and its assumptions.

I wish the article made a clearer distinction between the capital ratio and the equity ratio. In places it makes it sound as if only shareholder equity counts towards capital requirements.

But today that's essentially true. The other tiers of capital are much thiner and easier to manage. To get more equity the only way is more profits, or a right issue.

In the enthusiasm to raise the capital requirements there is a fundamental schizophrenia. We can't ask banks to lend more, in particular to borrowers with lower credit quality, and at the same time force banks to deleverage. We can't ask banks to hoard capital to strengthen their capital ratios and at the same time nationalise all their profits by imposing gigantic fines that are scaled more to be headline grabbing that to the actual scale of the misbehaviour.

>In the enthusiasm to raise the capital requirements there is a fundamental schizophrenia. We can't ask banks to lend more

There is no schizophrenia. We don't want more lending. The economy is already over leveraged as it is.

The banks want more lending, of course.

Some governments want more lending:

https://www.gov.uk/government/news/new-40-million-investment...

http://www.washingtonpost.com/business/economy/obama-adminis...

“The idea, in Europe and beyond, is to limit how far banks can adapt their spreadsheets to make their assets look less risky.”

They should be required to publish live spreadsheets on their websites showing their RWAs with daily marks. SpreadServe [1] would be perfect for this job.

[1] http://spreadserve.com

The data is not that readily available. Most Banking reporting is very far from realtime, it's "eventually consistent". Not because the Banks are (necessarily) hiding anything but due to the complexity of the systems.

Regulators in first world countries already require accurate and timely reporting. It's really not as simple as just publishing a data feed. That's why things like Basel II(+) exist.

The group aggregated RWA calcs for a global tier 1 bank run substantially in arrears - perhaps as much as 2 months, and IIRC are submitted to the regulators quarterly.

The author is also using the word "spreadsheet" as shorthand to mean "the entire modelling infrastructure". For the trading book, this can be non-trivial.

"Banks are optimally opaque institutions. They produce debt for use as a transaction medium (bank money), which requires that information about the backing assets – loans – not be revealed, so that bank money does not fluctuate in value, reducing the efficiency of trade."

-Banks As Secret Keepers: https://www.aeaweb.org/aea/2015conference/program/retrieve.p...

There's a recent but very influential paper that shows that banks need to keep their assets secret to function well.

Reply on news.ycombinator.com