Click here to download a .pdf version of this newsletter.

Return to Main Page


In My Opinion
Actuarial Credibility in Question
Paul Lacko  

Did you respond to the Opinion Writers Survey conducted by the Joint Task Force for Enhancing the Reputation of Casualty Actuaries? One part of the survey asks you to evaluate some proposals to improve "actuarial credibility." I think you're supposed to assume that "actuarial credibility" is inherently and indisputably a good thing, that more of it is better than less, that it can be improved, that actuaries can improve it, and that the proposed improvements will, in fact, improve things. The survey doesn't say what "actuarial credibility" is, why actuaries should be responsible for its production, or whose names and addresses we should print on the shipping labels.   

After you finish and submit the survey, the survey software locks the door behind you and won't let you return. You can't go back and review, print, or revise your responses. You can't even get in to see a blank survey form. In other words, you're expressly forbidden to reveal how your opinions might have changed as a result of the survey experience itself. What are the real problems? Were the questions on the mark? What other questions do you wish you could have answered, had they only asked? If every respondent were to answer exactly as you did, would that increase the odds that the world might someday be a better place?   

Hotshot quantitative analysts—read "actuaries"—should certainly be prepared to answer that last question, don't you think? Or, as some of those who question our credibility might suggest, is this survey just another confirmation that "actuary" may be defined as "a highly educated and well-trained person who has achieved the qualifications deemed necessary by an 'actuarial society' to stand in a dark room and throws darts at the far wall, trying to score a bull's eye on a black dartboard"?   


The first thing we need to explain to our critics is that this definition is too simplistic. The room isn't dark, and the dartboard isn't black. There are a lot of open windows, and gusts of wind shoot through without warning. There are many objects scattered around the room that obscure our view of the target and deflect our darts mid-air. Some of the objects are stationary, and we can work around them. Others move like pinballs, whose shifts in velocity are random and radical. And the dartboard shifts position every time a moving object bumps it or the wall near it. And we don't find out if our last throw came close until after several more throws. Oh, and our darts are all different sizes and shapes. We hardly ever get to throw identical darts twice in a row.   

We are doing the best that can possibly be done. We have developed a broad collection of actuarial models, tools, and techniques that are useful and effective in certain situations. We are well aware of their limitations. And we keep trying. We keep trying. And we keep trying to improve.   

The problem seems to be that we can't give the critics what they want, and they don't want what we can give them. What they want are exact answers and guarantees. At the very least, they want to know about potential problems that could be reduced or even prevented, given sufficient warning. They would appreciate something like what meteorologists do when storm systems form over the south Atlantic—forecasts of storm path, timing, and possible landfall.

We can't give exact answers or guarantees. We can only give ranges and approximations. When we select a single number, be it a rate per hundred dollars or a "reasonable" reserve level, we are no longer acting as Actuaries, with an upper-case "A," but as company managers or as regulators. (Vernon Rice, ACAS, described the distinction between lower-case "actuary" and upper-case "Actuary" more than twenty years ago to us members of his pricing staff. Don't blame Vern for what I say here in a much different context.)
The problem seems to be that we can't give the critics what they want, and they don't want what we can give them.

A company can't operate without informed management, of course. Federal, state, and local governments will not permit the companies to operate regulation-free. I enjoy serving the public in the corporate decision-making process. Many actuaries participate in setting up and enforcing the rules of acceptable insurance industry behavior. I would argue that we have an Actuarial role that is separate from any actuarial function we have in company management or regulation. To the extent that we confuse these roles ourselves, we confuse the people who depend on us. We thereby create our critics, both CEOs and Insurance Commissioners.   


I don't think Actuaries can accept responsibility for the carried reserve numbers in published financial reports, GAAP or Statutory. Selecting specific values may be part of an actuary's job, but it is outside the range of an Actuary's formal training and expertise. I do think an independent Actuarial Statement of Opinion is a good thing—by an Actuary who is not a member of the company's management team—but the best it can do is affirm or deny that the carried reserves are at least in the right ballpark, probably, more or less. Consequently, I think the proposals in the Opinion Writers Survey are more likely to make things worse instead of better.   

I do not mean to suggest that outside actuaries should prepare a complete, bottom-up reserve analysis for every insurer. The Statement of Reserve Opinion should be done in-house to supply regulators with the specific information they want disclosed. There's no point in doing over what the company actuary has already done, especially without the company actuary's detailed knowledge of the company's operations. Senior management should bring in independent Actuaries to audit the in-house Actuarial function. Don't validate a company's answers to regulatory questions. Analyze the strengths and weaknesses of the in-house reserve analysis. Recommend improvements. Verify that effective in-house actuaries are effective Actuaries. Help management and regulators to always have the best information that the Actuarial profession can provide and the best advice that in-house actuaries can offer.   

Credible actuarial audits would improve "actuarial credibility" substantially. They might also lead to meaningful standards of what "fully credible actuarial analysis" means. Granted, this wouldn't be a cure-all. But it would accomplish much more than any proposal in the Opinion Writers Survey.   


Click here to write a Letter to the Editors

Copyright © 2014 Casualty Actuarial Society. All Rights Reserved.