Home | Business News | Browse by Publication | A | Appraisal Journal

The risk manager of the future: scientist or poet? (Notes and Issues).

Article Excerpt
Quants alone do not represent the future. There is no single deep field of knowledge that is sufficient for competence in the real-world field of risk management. The successful risk managers of today and tomorrow are synthethizers and applied problem-solvers, who focus on loss and capital estimation that is integrated within pricing models and profitability reporting.

In the midst of the industrial revolution, English philosopher John Stuart Mill noted that the technology which had so changed the world did not lighten man's toil: if anything, there appeared to be more work.

Similarly, we should expect the recent advances in risk management thought and technology to make a risk manager's job more hectic, in that both risk and risk management can be judged. Such is the price of moving from record-keeping to affecting the bottom line. The following are thoughts on where risk management is going, and what it's leaving behind.

Why Change?

The current pace of change is driven by data. For a long time data wasn't warehoused in a usable way, making potentially useful theories useless in practice. Additionally, rating agencies didn't publish anything on default rates by grade, so there was little data for the credit risk manager to benchmark against. Banks now have archival, transaction-level data in large relational databases and can apply quantitative techniques to products that previously were only the subject of anecdotes. Moody's and S&P have published historical data on default rates for commercial credits, and Fair, Isaac and others have extensively tested statistical models for evaluating consumer credits.

Think about the following constraints for all previous and many current risk managers:

* A/L management risk was based on a few different 12-month earnings simulations.

* Loan performance data was not archived.

* Loan application information was not linked to account information.

* Risk management could only be judged by top-down charge-off levels.

* Business line performance could only be judged by return on assets.

In such a scenario it was difficult if not impossible to demonstrably improve risk management techniques, since there was no way to compare the alternative risk metrics and strategies. Being a good risk manager was like being a good LAN administrator--hardly worth the effort given the lack of appreciation. The reputation of risk management as a science was such that previous regulatory or business line experience trumped previous risk management experience when qualifying as senior risk managers. This was similar to the way education majors are presumed to know less about education than those who learn about education indirectly through physics or English literature. Luckily for shareholders...