Primary navigation:

QFINANCE Quick Links
QFINANCE Reference

Home > Blogs > QFINANCE Editor > Paving the Way for Dodd-Frank and Basel III - an interview with Michael Carhill

Paving the Way for Dodd-Frank and Basel III - an interview with Michael Carhill

This interview has been contributed by marcus evans events.

Anticipation is mounting as the global banking industry awaits the implementation of new game-changing financial regulations. Although implementation of the changes brought about by the Dodd-Frank Act and Basel III will occur over several years, banks must start modifying their risk management, capital adequacy and stress testing programs now in order to ensure stability and facilitate compliance in the long term.

Michael Carhill, Director of the Enterprise Risk Analysis Division (ERAD) of the Office of the Comptroller of the Currency, US Treasury Dept, gives his thoughts on how new regulations will affect capital management practices in the banking industry. Carhill will speak at the marcus evans 5th Annual Capital Allocation and Stress Testing Conference taking place in New York City, NY, January 31 - February 1, 2011.

How helpful and accurate is stress testing as a means for banks to manage their regulatory capital position?

MC: Referring to scenario analysis stress testing - for instance in banking book interest rate risk, scenario analysis has been used for many years. Scenario analysis is a primary tool by which banks manage their interest rate risk. In other areas, in particular portfolio credit risk, this concept is relatively new. We believe it's helpful for banks to analyze the stress test, but the question of how accurate the models can be has yet to be determined. Banks have been doing stress testing for credit portfolios since about 2005, and at this point there are only one or two banks that have even finished their models.

Even before finishing the stress-testing credit models, banks are finding them useful for planning for likely macroeconomic scenarios and for helping to establish their loan-loss reserves. These models do not necessarily need to be highly accurate to be useful to banking executives. That said, it remains an open question of how accurate the credit portfolio stress tests are. When you talk about a stress test using banking book interest rate risk, a common stress test is to impose a parallel, instantaneous shock to interest rates. The shocks differ depending on how close you are to the line of business. For the investment portfolio manager, that shock might be ten basis points, or for executive management that shock might be 100 basis points. Those shocks never actually materialize in exactly the way the model assumes. For something like investments securities the models are accurate - we've had a situation where we're close enough to see the assumed shock the models predict very well. But in credit when you're talking about something like, say, unemployment increasing three percent over the next year, it's hard to get chances to test those models. Unemployment just increased three percent two years in a row and we know what the credit losses were then, and that's being used to build the models. But whether they will predict well the next time unemployment goes up three percent remains to be seen. But at any rate, it at least gives you a ballpark estimate, which is new, and we think that will be very useful information for executive management.

What are the common challenges in stress testing?

MC: There are a lot of quantitative, technical issues that are very difficult to solve. They fall into two categories - one is the response of the model to assumptions. How will the variable of interest, in this case profit and loss, respond to the assumptions? The main tool used to analyze that is econometrics and there are all kinds of issues that come up in econometrics. What are the factors driving gains and losses? As you do your econometric work are you sure that you do not have a spurious relationship? And what is the shape of the function - are losses linear in the underlying driving factors or are they non-linear? Those are the challenges that banks face, so as a result of the technical difficulties, it's turning out that the institution of portfolio credit stress testing is taking banks a good deal longer than we had hoped. Very few banks have completed their models and they're in the process of building them.

What is the relationship between the Internal Capital Adequacy Assessment Process (ICAAP) and the bank's other lines of business?

MC: Aside from the resources that go into developing enterprise-wide stress tests, the lines of business aren't always involved. It's an overlay. It is the executive management that uses ICAAP as a tool.
Banks have different ways of doing this. One way is you build a team of quantitative modelers who develop their models using historical data that has been provided by the lines of business. Once the quantitative modelers have their models built and have their enterprise-wide stress tests developed, they give it to executive management. If executive management finds the risk unacceptable, they will tell a given line of business to rein in the risk. For example, even though this enterprise-wide stress testing is still under development, we have seen one bank that has pulled in its banking book interest rate. It was reduced quite substantially as the result of an enterprise-wide stress test because executive management was uncomfortable with the losses that resulted. At the other extreme of the enterprise-wide stress test, a macro-economic scenario is given to each line of business and then the line of business forecasts its own losses. Even then it is not a problem integrating with the line of business's main work, but it does take resources, labor and time in order for the bank to build its enterprise-wide stress test.

Can any efforts be made from the banking sector to help jumpstart the US economy, do you believe?

MC: As always, banks play a crucial role in the economy by connecting savers with investors. They must do so safely and soundly and with adequate capitalization to ensure the banks' viability as a going concern. In our view, that's a constant.

The marcus evans 5th Annual Capital Allocation and Stress Testing Conference will take place in New York City, NY, January 31 - February 1, 2011. QFINANCE is a media partner on this event.

Contact information can be found on QFINANCE's Upcoming Events page.

Tags: banking , banking reform , Basel III , Dodd-Frank Act , dodd-frank bill , regulation
  • Bookmark and Share
  • Mail to a friend


or register to post your comments.

Back to QFINANCE Blogs

Share this page

  • Facebook
  • Twitter
  • LinkedIn
  • RSS
  • Bookmark and Share

Blog Contributors