Current Actuarial Modeling Practice and Related Issues and Questions

The future of the actuarial profession is a matter of modeling and putting models into practice. One change taking place in the profession is the realization that all actuarial techniques are founded on models and that it is the models, and not the techniques, that are important for the future of the profession. In the U.K., we recently discussed a paper, ‘‘The Future of the Profession’’ (Nowell 1996), that pinpointed, in particular, weaknesses in financial economics and stochastic methods. What this really meant was that actuaries have focused too closely on their techniques — what I call computational tools — and have not kept their eyes above the horizon where progress has, by and large, been made by others. The U.K. has been a little slower than the U.S. and Canada in putting this to rights, but that is now happening. And modeling is the key.

In the first part of this paper I set out a ‘‘model of models.’’ It does not attempt to explain everything about models in any philosophical way; it is meant only to give an intuitive guide, so we can take an actuarial tool and recognize where it fits into a modeling framework. Then I discuss at more length two modern actuarial tools—profit tests and model offices—and ask, Are these models at all? And, because I am very interested in model offices and I believe these are just at the beginning of their development, I discuss some of the more down-to-earth practical problems about creating and using them.