BI’s Article search uses Boolean search capabilities. If you are not familiar with these principles, here are some quick tips.

To search specifically for more than one word, put the search term in quotation marks. For example, “workers compensation”. This will limit your search to that combination of words.

To search for a combination of terms, use quotations and the & symbol. For example, “hurricane” & “loss”.

Login Register Subscribe



Terry Van Gilder's insurance career began in 1971 as an underwriter for Chubb's Department of Financial Institutions. Mr. Van Gilder eventually rose to the position of executive vp of Chubb & Son Inc. and chief underwriting officer of Chubb Group in Warren, N.J.

He left Chubb in 1995 to become chairman of Menlo Park, Calif.-based Risk Management Solutions Inc. RMS is widely known for its computer catastrophe models that are now used throughout the insurance industry and have, in some cases, replaced traditional actuarial methods.

Before leaving Chubb for RMS, Mr. Van Gilder helped spur a research and development relationship between Chubb and RMS. Chubb became an investor in RMS through Morgan Stanley's Merchant Banking Group. Mr. Van Gilder also encouraged Chubb to become a shareholder in Bermuda-based Cat Ltd. The idea to form Cat Ltd. was born at RMS, in part as a vehicle for testing the underwriting capabilities of RMS' software.

Mr. Van Gilder, 53, recently purchased a historic restaurant and inn that dates back to 1755 and is located on the eastern shore of Maryland. He is overseeing its restoration and business development. He also is a board member for RMS, which continues developing new systems that could shape insurer and risk management operations.

Mr. Van Gilder recently spoke with Associate Editor Roberto Ceniceros about catastrophe modeling and RMS' role in the evolving relationship between insurers and reinsurers.

What drew you to catastrophe modeling?

As senior underwriting officer and then as chief underwriting officer at Chubb, one of the issues I was trying to deal with -- and the whole industry was trying to deal with, although it was just in its very early stages in the late 1980s -- was this whole issue of managing catastrophe exposures.

Most insurance companies, even today, have not equipped themselves very well to deal with catastrophe exposures. They have a very difficult time understanding their accumulations.

I just felt the tools available to us and the techniques that we were using, although tried and true from a traditional perspective, were not adequate for the task.

So how did you get involved with RMS?

I was looking for people or companies who could help me manage catastrophe risk better. It was a concerted effort. Then we bumped into some RMS people at an industry conference.

By comparison to what RMS does today, it was a real little program then. At that time, they were not even sure if the insurance industry would be a significant customer base. They were mostly technical and engineer oriented.

What I did that was beneficial to Chubb and RMS is I entered into a research and development relationship. They were helping me figure out how to deal with the set of issues that I had and I was helping them figure out how to deal with insurance companies, how the product should be developed, how they should explain it and why it would be useful.

How many clients does RMS now have?

We have about 370 clients. Virtually every significant insurer and reinsurer in the world is a client of RMS.

So has RMS data played a role in the evolving relationship between insurers and reinsurers?

Oh yes. In many cases, reinsurers are much more knowledgeable today and involved in the affairs of the primary companies they insure. That is usually on a cooperative basis, not on the basis of imposing themselves.

And they are using the software that we create for them and the methodology around it to not just understand what primary companies are doing, which was the original purpose of most reinsurers.

They also manage their capital more directly and we have a whole set of tools that pertain to that.

How long has that been happening?

I would say since the early '90s.

Have reinsurers been assuming even greater influence in the business of primary insurers?

The greater involvement and the insistence on the part of reinsurers that they understand what the primary companies are doing -- and by understanding it, then seeking to alter their behavior in some respects -- grew out of a couple of issues, including the explosion of tort liability. That caused people to view casualty reinsurance very differently, and caused capacity constraints and differences in terms and conditions, many of which were imposed by reinsurers.

How has your involvement with RMS shaped your view of the industry?

That is how I came to believe -- and I still believe -- that the future of the industry very much lies in becoming more sophisticated in the use of automation. Companies have traditionally equipped themselves with systems that are really oriented toward reducing the cost of transactions and keeping statistical records.

Still, it is rare to see on an underwriter's desk a very sophisticated piece of equipment that pulls in information from a variety of sources and allows better decisions to be made that relate to the rest of the organization, such as its entire financial situation. RMS today is working with a number of clients on building workstations that do that.

In many respects, the insurance industry as a whole has been very conservative and slow in utilizing more sophisticated means to understand its business.

It has to adopt better analytical tools that help determine which way to take a company. They're either going to find ways to utilize methods and tools appropriate for the industry or they are going to suffer rather badly in competition or in comparison to other financial institutions. If they don't get a better grip, they are either going to fall by the wayside or become part of something else.

What are the shortcomings of the traditional underwriting methods, such as evaluating a risk's probable maximum loss?

The problem I see with much of the industry is that what they are doing is very static. It is the dynamic part of the book of business that is interesting.

That is what is wrong with looking at individual risk PMLs. They look at a static deterministic event, at one thing. But the range of possible events, the probability associated with them, and the magnitudes of all those things have a lot to do with it. That risk is part of a book of business. It doesn't exist by itself and you can make thousands or hundreds of thousands of decisions in an accumulated book of business and it will kill you because you are not managing the dynamic quality of it. You are not managing the total of it. You are managing individual risk decisions.

The concept I tried to get infused into RMS is optimization. You are not just randomly presented with risk and selecting risk. If you are alert, you are purposely going about business in a particular way, so as to attract certain kinds of business and take it into your company in a particular way at a particular price level and under certain terms and conditions because it makes sense in the total context of what you are managing.

There have been optimization theories applied to portfolio management in other kinds of financial institutions for years. This is not new, but it is real new in the insurance industry. The idea is that you conscientiously go about constructing a portfolio of a particular type.

Where does that lead an insurer?

Strategic capital allocation. It's a concept I was working on at Chubb several years back and some companies have a sense of it, some do not. It's basically about a pool of capital you are trying to employ and produce a return on.

There are an infinite number of ways you can deploy that capital: You can underwrite risk, you can invest in bonds, you can invest in other things. It really gets into the question of the possibilities you have available to you. Are you employing your capital in the most logical way to produce the best return?

Most insurance companies until very recently didn't.

So what we are trying to do today is work with insurers and reinsurers more on the concept of strategic capital allocation. You have this list of things you can do.

Do they produce an ultimate book of business that you look at and you say, "This makes sense. This is who we want to be. This is who we are. This is what we have the expertise to do and this produces a meaningful return in the context of whatever we say to our shareholders we are going to achieve."

It's very difficult for companies to do that.

There has not been a lot of expertise focused on that issue in the industry overall, although you will hear a lot of lip service. It's about developing a model of the business which takes into account, much like we do in the catastrophe area, various contingencies that are likely to occur, how likely are they to occur, what trends look like.

Some of it is based on traditional actuarial methods and some of it is based on our own expertise in terms of analyzing a series of events and the probabilities associated with them and things that can randomly skew results, much like hurricanes do.

We are applying those same skills of catastrophe modeling to the issue of managing an overall portfolio of disparate things and then trying to pull that all together into some body of material that can be communicated effectively, not just to management but throughout the company.

This is what we do in the catastrophe arena every day for companies.

So, you are helping insurers lay out goals and over time change their risk mix?

Sure. And become in the process much more successful entities that can survive in a hostile environment and have a much better grip on what they are doing. Those are really the issues. That is a very unique new thought process, the ability to lay out a sophisticated model.

Are you doing anything similar for risk managers?

For corporate clients -- who are working on a broader sense of risk management and working on managing the overall risk of a corporation, but have traditionally looked at the downside of risk -- we are trying to apply the analytical skills that we have to the issues of quantifying, mitigating and managing risks, whether or not it has anything to do with natural hazards.

We are taking RMS people who have backgrounds in software development, engineering, statistics and actuarial areas and applying those skills to broad corporate issues of managing risks in a very different and more intelligent way.

The idea is there are upsides and downsides to risks. There is also the question of maximizing the upside, and can you figure out over time what things are more likely to lead to a positive result and lessen your focus on cutting off the downside risk?

The question is, can you figure out which things are more likely to be more productive or lucrative for the company to do? It's very much in the research and development stage.