Anniversary

1981

Education

University of Michigan
B.S. Mathematics

Credentials

FSA, MAAA, CERA

Extra Curricula

Former Chair,
SoA Technology Section

Former Member,
American Academy
of Actuaries
LFRIC Committee

 

 

 

• Browse Team Biographies

You started your career at PolySystems in 1981. What services did PolySystems offer then?

PolySystems was a timesharing company in 1981. This was before PCs came in to use. Mainframe computers were very expensive, so companies paid for the time they used on mainframes. PolySystems offered software to calculate statutory reserves, cash values, GAAP reserves and asset shares. Tax reserves had not been invented yet. We made it easy to transfer the reserves to company administrative and valuation systems via magnetic tape.

The business model was to let the mainframe computer generate revenue. The employees were around to keep the machine running and to find new customers.

There are more than 85 people working for PolySystems now. How many worked at PolySystems then?

PolySystems employed 14 people when I started, and a few of those 14 are still with PolySystems today.

What were some of the events that began the transformation of PolySystems from what is was then to what it is today?

In 1985, companies had to compute separate tax reserves for the first time. A new wrinkle was that tax interest and mortality assumptions might change with every year of issue. That was a huge problem for a lot of company valuation systems. We created a solution which brought together people, software and hardware to get the job done. The combination of the three was more valuable than the sum of the parts.

In the late 1980s, PCs became fast enough for us to consider doing enterprise work on them. We demonstrated how it was faster to calculate values as needed rather than depend on pre-computed factor libraries. There were numerous obstacles, but there was tremendous cost advantage in using them.

During the 1990s, products and regulations made a first principles operation even more compelling. Many lines of business demanded this approach. Late in the decade, attention was on Y2K. The practice of using two digits to hold a year seemed safe in the 1960s, but in the late 1990s, doom and destruction were predicted. Many companies replaced aging systems.

In the current millennium, we have seen dramatic growth in risk management, we await new techniques for valuing liabilities, and we watch products change to be attractive in the challenging environments.

What changes do you expect to see over the next several years?

The demands of discretionary practices such as we see in risk management and required reports proposed by PBA, Solvency II and the new IFRS will be intense. There will be pressure to depend on compressed models, because the quantity of calculations are immense and granular models are needed for accuracy and auditability. These are uneasy, trade-offs.

Actuarial systems will undergo significant development. We need high quality data from other systems. We need to manage the many layers and sets of assumptions efficiently. We need to spread the calculation workload intelligently over a large number of computing resources. We need to make use of affordable computing facilities.

One interesting new technology is cloud computing. The idea is that we can run our peak jobs on someone else's equipment and just pay for the time and resources that we use. I am not sure how this will work out, but it sounds a lot like timesharing which is where PolySystems began life.