Ahead of the Pack

Estrella Seguros uses SAS® Customer Intelligence to improve profitability in a market with squeezing margins

Estrella Seguros is part of the Generali Group, a company that was originally founded in Trieste (then a part of the Austro-Hungarian Empire) in 1831 and has been doing business in Spain since 1834. In 2002 Generali had 175 companies on its balance sheet, of which 108 are insurance companies with consolidated premiums of €46.9 billion (US$56.6 billion), making it the fourth largest European insurance company (and the largest in life insurance).

Two of Generali's insurers are based in Spain: Vitalico Seguros and Estrella Seguros, a highly customer-oriented organization with 59 branch offices in Spain, 397 agency offices and 2,161 computerized agents. Estrella Seguros generated €1.25 billion in total revenues in 2002, of which €245 million was in motor insurance premiums, accounting for the largest share of non-life business.

Profits under pressure
"Auto insurance is a very tough and complex business," says Alberto-José Macián Villanueva, manager of the Motor Insurance Department at Estrella Seguros. "It is a compulsory product – you cannot drive a vehicle without insurance – and there are a large number of insurers competing for business, so the market is highly mature allowing for only a tiny growth in the volume of business. Profit margins, if positive, are very thin."

Consequently, effective technical and management control is absolutely essential to keep Estrella Seguros' motor insurance business profitable. In the early 1990s, under the pressure of increasing domestic competition and the entry of foreign insurers into the market, Spanish motor insurance companies became increasingly strategic in their approach and started the process of decentralizing decision making to lower levels of the organization. Financial and management control systems took on a greater importance in supporting this process, enabling senior managers to monitor results by company, by branch and by unit, and to identify deviations from the budget or plan. Insurance companies increasingly used budgetary instruments, such as assigning incentives linked to evaluation of performance, and took corrective actions to improve internal procedures and competitive positioning.

Predicting risks for a premium advantage
"The key to such management control is the definition of indicators that guide the behavior of each department and the formulation of a model to predict performance," says Villanueva.

In the specific context of motor insurance, management control and predictive models are particularly important to ensure that customer-facing decision makers can decide what risks should be covered, what conditions and covers should be offered, and what level of premiums should be charged. These are the basic underwriting rules or "law" in the insurance business.

However, the real complicating factor is that motor insurance is a constantly moving target. Beneath the "basic law," there are constantly and rapidly shifting quantitative technical parameters, such as the level of premiums, the frequency of claims and the average cost of the claims. "This makes it extremely difficult to forecast changes in the market," says Villanueva. "Insurers that are unable to stay ahead of the game and adapt rapidly simply collapse and disappear," he adds.

With competition squeezing margins to the limit, accurate premium setting, or rating, is fundamental to profitability, and you need to be able to process increasingly large volumes of data to arrive at the right premium for each individual motorist: claims history, geographic and demographic information, social status, and information on the vehicle. This data is related to the technical parameters to determine the average number of claims for the motorist's profile, and the average value of the claim, to arrive at the correct risk premium.

"There are typically hundreds of variables relating to a motorist and vehicle, so we rely on statistical analysis to identify what are the most significant variables in rate setting," says Villanueva. "It is obviously essential that any insurer should not only be able to set a competitive yet profitable rate, but also should be able to provide a quote to the customer instantly and automatically."

Like most motor insurers, Estrella Seguros uses SAS to perform the tariff segmentation and data analysis required to deliver accurate rating. "If you do not use SAS, you are dead," says Villanueva. "SAS has become the de facto standard analytical tool for rate setting in motor insurance. Anyone who is not already using SAS will probably be out of business in the very near future. It's as simple as that."

In this situation, the way in which SAS is applied as a technology solution, as opposed to SAS itself, can be a key success factor. Villanueva's philosophy is to give the analyst as much scope as possible to "do it on your own" with minimum dependence on the IT department. "We built a simplified data warehouse, specific to the requirements of the Motor Insurance Department and put the emphasis on the correct modeling of the data, as opposed to query tools," he says.

Estrella Seguros then used SAS rapid applications development software to build two applications, the Actuarial Laboratory for rating, and the Multidimensional Actuarial Laboratory, for control of the key business variables through free navigation. Users at Estrella Seguros can use these two sophisticated applications flexibly and easily, without in-depth knowledge of SAS software.

Fine-tuning automotive insurance
"As a result of these applications we can decentralize decision making while maintaining a high level of technical control," says Villanueva. For example, on a regular basis Estrella Seguros monitors sales for underwriting quality and analyzes the reasons for non-renewals, and it can monitor results against budget (in terms of claims frequency, average cost of claims and the evolution of claims reserves). "Two or three years ago we were in a very bad phase of the economic cycle in the motor insurance industry and experienced some poor results," says Villanueva.

Many companies tried to get through the crisis by dropping prices to increase market share, and were severely punished. "But thanks largely to the knowledge delivered to us through SAS, at Estrella Seguros we were able to fine tune our strategy more effectively, allowing us to emerge from the downturn ahead of the pack. SAS enabled us to anticipate the next phase in the cycle and to reposition and re-price our offering accordingly," says Villanueva.

Estrella Seguros has now been using SAS for actuarial analysis and for technical control over a period of five years. Based on its good experiences the company has recently implemented, together with its sister company Vitalico Seguros, a corporate knowledge base to support marketing, financial control and general management of all areas of business. Villanueva believes this will be key to the future performance of the companies in an industry that will become even more competitive.

"In the motor insurance industry, it is very easy to go out of business if you lack adequate knowledge of customers and insight into market trends," Villanueva says. "Without a doubt, SAS provides the best technological solution, not only to deal with vast amounts of information, but also to perform the sophisticated analyses that are necessary to survive."