Tag Archive | "Actuary"

Considerations When Designing New Products: An Update

Working in the F&I industry stays interesting because of the great products that our industry continues to develop to meet the needs of the public. In 2011, we wrote a short article on developing new products. Since that time, the amount and complexity of new products in the F&I space has only increased.

In developing a new product, there are, of course, many issues regarding such concerns as policy language, pricing, marketing, and systems. We want to focus on some general considerations when developing a product from a pricing and accounting standpoint. Let’s tackle the subject in FAQ format.

Do I Have Any Claims Experience?

Obviously, if you have related claims experience, it can be of tremendous help in determining the potential exposure for your product. Industry experience can be helpful. Census data, crime statistics, and data designed for the auto industry can help solidify assumptions.

How Is the Product Marketed?

Similar products can have vastly different claims experiences depending on how they are marketed. Will the product be marketed to new buyers or to current owners? Is the product marketed at the dealership or direct marketed? What is your expected pricing?

The amount of money that a consumer pays can impact claims consciousness as well. A low-price product will have lower awareness.

Embedded products, which are part of the sale of every applicable item and not charged separately, are increasingly popular. These products typically will have a much lower claims rate. This may be due to consumer’s forgetting about the benefit since they didn’t explicitly pay for it.

A great example of this is that many credit cards include protection for items bought within a certain period. Have you ever made a claim for an item bought on a credit card? Do you think you could have?

How Am I Going to Earn the Premium?

This is a critical assumption for evaluating early experience. Claims occur on F&I products at vastly different rates depending on the product. Some products (even profitable ones) have an initial surge of claims as the buyer may seek to repair some preexisting issues that cannot be fully excluded. Never assume that exclusions can prevent all claims.

To evaluate the experience correctly, you must earn the premium in the same ratio as the claims flow. You may need to earn premium differently for accounting and refunding than you do to evaluate the program. While earnings are necessarily subjective, be careful to use your best estimate.

Some programs, such as lease wear-and-tear, will have virtually no claims experience until the contract is at, or near, the end. You may want to make more careful assumptions about these products because once you see the results you may have a large amount of unearned exposure.

How Will I React to the Results?

Everyone who develops a product expects it to succeed. Don’t let your prior assumptions blind you to the results in your book!

Most F&I products are relatively high-frequency/low-severity products that reach credibility in a short amount of time. If you are earning your exposures correctly (see above), your results could be actionable in a few months. Plan on frequent monitoring after the product launch to ensure that the program is performing to your expectations.

Divide the experience into months or quarters by policy inception date and examine the experience of the most mature contracts carefully. For example, if you expected a “claims surge” for the first few months after the sale of a product, are you seeing the claims from your oldest contracts dropping? If not, you may need to reevaluate your assumptions.

If an Insured Makes a Claim, Will They Be in a Better Financial Position?

Most insurance products operate on the premise that the insured will not be better off financially after a claim. For example, auto insurance will pay the actual cash value of your car. In theory, you could go out and buy the same model car with similar mileage. You cannot buy a brand-new car.

Some products, such as GAP insurance, improve the customer’s financial situation. If you have a GAP claim, the negative equity in your vehicle is erased. A similar example occurs in homeowner’s insurance, which typically offers “replacement value” for your house and contents. If a fire destroys your house, you will receive new furniture and clothes — not the old stuff you had.

These insurance products work because the vast majority of people do not wish to have a fire or a car wreck, even if it improves their personal balance sheet slightly. However, you can expect that these types of products will have higher claims because there are always a few people who will use the product to their advantage.

Will the Claims Be Correlated With the Economy?

For most products, this is not true. Unexpected repairs and collisions are examples of random events. While certain people and certain vehicles may be more likely to have a claim, in general, the claims among similar risks are random.

However, this is not true for all products. Some products will show a higher propensity of claims due to the economic environment. This type of risk can occur when insuring the underlying value of an asset, such as with GAP or residual value insurance. If the used car market shows a big drop in prices, the likelihood and severity of claims will increase across the board. If your product is correlated with the economic environment, be prepared for a wide range of results depending on the conditions of our economy.

New products are what make the F&I business one of the most exciting places in insurance for product development. While the considerations listed above may give you pause, we encourage our industry to continue to develop products which meet the needs of all participants, by providing peace of mind to the buyer and an adequate return to the underwriter and business partners.

Posted in ActuaryComments (0)

VSCs in 2016: New Terms, New Costs

For the 2016 model year, there are some major changes in the warranty market which will impact service contracts.

First, of course, it will increase the costs for the same service contract. Some claims that were previously covered by the manufacturer’s warranty will now be covered under the service contract. Second, the claims would be projected to occur, on average, earlier in the contract period. Therefore, the revenue should be recognized — or premium earned — more quickly than in the past.

What is the impact?

For an administrator, the reduced claims covered by the manufacturers are unwelcome news. On the positive side, the decrease in warranty terms increases the “value perception” of a service contract. Unfortunately, there are no easy answers in addressing the impact of the additional coverage without a detailed analysis of actual claims. The decrease in powertrain coverage will not increase claims equally by make, since some makes will have proportionally more claims in the powertrain portion than other makes.

The impact on service contract costs is a complicated question and there are a number of factors to consider:

  • Make/model of the car: There will be variation in the powertrain claims by vehicle.
  • Starting mileage: Since the time limit remains unchanged, used vehicles with remaining warranty will be impacted less. For example, a used car with 36,000 miles that is three years old will have minimal cost increase since the warranty would be active for two more years in both cases.
  • Driving patterns: Notice that the time remains the same and the new term offers 12,000 miles per year while the old term averaged 20,000 miles per year

In order to calculate the impact of this change, we made assumptions which are general in nature and may not be appropriate for a specific book of business. The assumptions are that powertrain costs are 50% of total service contract costs.

In addition, we assumed claims would increase by 10% per year as the car ages. Finally, we assumed the contract holder would drive on average 15,000 miles per year, with some driving as few as 8,400 miles per year and others 21,000 miles per year at most.


Note that the increase in costs is only a rough estimate due to the decrease in warranty terms; it would not include any increase due to new technologies or more expensive repairs.

Why the increase in costs? Note that the previous warranty of five years/100,000 miles effectively eliminated the powertrain portion except for the last year. Relatively few drivers “miled out” of the previous warranty. Effectively, the new terms penalize the high-mileage drivers, because a normal driving pattern would not have exceeded 60,000 miles in five years by a great margin.

Earnings are also included under a “Reverse Rule of 78s” method. This method is often used for earning new car vehicles. It uses a sum-of-the-digits method in which the earnings are in proportion to the month. For example, in Month Three of a 12-month contract, the earnings would be (1+2+3)/ (1+2+3+4+5+6+7+8+9+10+11+12) or 6/73 or 8.2%. So in Month Three, a total of 8.3% would be earned.

While this method is easy to implement, it only does a fair job of approximating earnings. It tends to earn too fast early in the contract when there is very little exposure due the manufacturer’s warranty.

More interesting are the hypothetical earned experience curves. While the examples above are hypothetical, it does show that earnings will speed up to some degree under a decreasing manufacturer’s warranties.

Are you still using triangles?

Actuaries typically use triangles when analyzing service contracts. They are typically organized by purchase date, term and type of car. They are easy to produce but past trends can be problematic. In this case, the patterns in the past will show too little development at the end of the contract. If you don’t adjust for this type of exposure, you will not only be facing increased costs but may not realize for a number of years. We prefer “triangle-free” approaches using miles outside the warranty. We will discuss this more in a future article.


Of course, a detailed analysis of the specific factors in your book would be necessary to quantify the impact of a change in the manufacturer’s warranty. Extended eligibility is another concern for a couple of reasons. First, these vehicles will show significantly different earnings patterns since the expiration of the manufacturer’s warranty will occur sooner. Also, the manufacturer may extend the warranty on these vehicles.

It is important for administrators to know both the underlying cost and the correct earnings rate of their book of business. Administrators need to be prepared to understand the impact of these changes on their service contract offerings.

Posted in ActuaryComments (0)

CNA National Welcomes New Chief Actuary

Scottsdale, Ariz.—Effective June 13, Karen Queen has been appointed as the chief actuary for CNA National Warranty Corporation.

“We are very happy to have Karen as part of the CNA National team,” says Joe Becker, president and CEO. “Her expertise will help us to continue providing accurate rates for our expanding product line. Having spent the last nine years with our parent company, CNA, she is an ideal candidate to be part of the executive team that is moving CNA National to the next level.”

Queen will be responsible for the direction and management of CNA National’s pricing strategies and actuarial forecasts to support the company’s objectives and growth.

During her tenure with CNA, Queen was most recently the assistant vice president for professional liability pricing and previously the assistant vice president for excess and surplus pricing. In addition, she played key roles in the actuary departments for Berkley Underwriting Partners, Encompass Insurance, Safeco Insurance and Allstate.

“I’m excited to be in Scottsdale with CNA National,” says Queen. “I look forward to the opportunities this new role will present.”

Queen holds a Bachelor of Science in Mathematics from Purdue University.


Posted in P&A NewsComments (0)

Staying Ahead of the Earnings Curve: Alternative Methods of Examining VSC Experience

One of the most common metrics utilized in evaluating the profitably of a book of vehicle service contract (VSC) business is the earned loss ratio, calculated by dividing the current cumulative losses paid or incurred by the earned premium-to-date. Earnings curves help a company appropriately match revenue with expected VSC liabilities by providing the percentage of premium that should be recognized as revenue at each point throughout the life of a VSC.

In our previous article, we examined the impact of cancellations and refunds on earned premium and we recommended an alternative approach to developing earnings curves which incorporate the impact of cancellations in an effort to produce stable earned loss ratios throughout the terms of the VSCs. However, as we noted, there is likely constant shifting in the mix of business underlying a given earnings curve, i.e., the distribution of loss and refund dependent variables within the applicable segment is changing from one year to the next. Because of this, maintenance of the earnings curves and monitoring of the loss and refund emergence patterns by the company’s actuary is required.

In this article, we will review a method of testing the appropriateness of the utilized earnings curves. First, however, we will examine various methods to review the experience which are entirely independent of the earnings methodology.

Written Loss Ratio Development

Developing the optimal set of earnings curves is no easy task. Inappropriate earnings curves distort earned loss ratios and can mask problem areas of the VSC book. In addition, earned loss ratios for the most recent contract years can be highly leveraged resulting from few expected claims and a relatively smaller percentage of the premium earned. But what if we could examine the experience while removing the reliance on the earnings curves all together? One way we can do this is to observe the development of the written loss ratio, that is, the ratio of cumulative losses to net written premium as the book matures, as displayed in the following graph:

*Note the amounts provided in the graph above were developed for illustrative purposes only and do not represent actual results.

The above graph provides a comparison of cumulative written loss ratio development for each contract posted year as of each calendar month-end. As expected, each curve varies in length and increases from inception of the contract year until all contracts expire. For example, the 2011 curve extends to 48 months of age (12/31/2014), while the 2014 curve extends to only 12 months of age. Assuming each contract year performed identically, the curves would be on top of one another. However, we can see that each year is quite different. For example, the 2012 year (green line) has grown to a 61.3% loss ratio as of 36 months whereas the 2011 year (red line) had a 53.4% loss ratio as of the same age. This may be an indication that the 2012 year will ultimately have a loss ratio significantly higher than the 2011 year. In addition, the 2013 (orange line) and 2014 (blue line) years have significantly higher loss ratios than previous years at the same age, but why?

When building and examining this type of graph, it is desirable to achieve an appropriate balance of credibility (data volume) and homogeneity (similar characteristics across the years). At some point, as we filter the data down to a smaller segment of the overall book, homogeneity may increase, but we will have lost too much credibility to rely on the results. So we must try to find the delicate balance between the two.

The smoothness (or lack thereof) of the lines will give a quick indication of the volume of data behind the graph. Regarding homogeneity, one must be mindful of the underlying mix of business across the years. For example, if all term lengths are aggregated and we know that the more recent contract years have a significantly higher percentage of shorter term business, this type of graph may show the more recent contract years performing worse than older years at the same relative points in time. However, this conclusion may be the result of a misinterpretation.

Since the average term of the more recent years is shorter, the loss ratios will develop to their ultimate values faster and will initially appear less profitable than prior years. To help avoid misinterpretation of the results due to shifts in the underlying mix of business, we must compare year-over-year results by filtering to a level of sufficient homogeneity. However, if we were not aware of shifts in the VSC book toward shorter term contracts, this graph may help uncover the shift and we can now drill deeper into the data and review each term separately.

Reviewing at this more granular level may then uncover another trend across the years and a mix shift of another variable. We may never uncover all underlying mix shifts, possibly due to the capture of insufficient variables or lack of credibility at a sufficiently granular level, but understanding the shifts and results at a granular level will help us explain trends at a more aggregated level.

Building the Graph

As discussed earlier, the graph of cumulative written loss ratio development is built by using purely raw data and is not dependent on any earnings curves. We can break it down into its components of loss development (numerator) and net written premium development (denominator).

*Note the amounts provided in the graph above were developed for illustrative purposes only and do not represent actual results.

The above graph shows how the cumulative net written premium for each contract posted year has developed as of each calendar month-end. Each year’s curve rapidly increases through 12 months of age as additional new business is written. After the peak at 12 months, the negative development is due to premium refunds on canceled contracts.

*Note the amounts provided in the graph above were developed for illustrative purposes only and do not represent actual results.

The above graph shows how the cumulative paid losses for each contract posted year have developed as of each calendar month-end.

The first step to building premium and loss development graphs is capturing the data. At a bare minimum, we need the appropriate dates (contract posted date, cancelation posted date and the claim payment date) and the appropriate measures (original written premium, premium refunded and claim amount paid). Contract and cancelation posted dates are typically used to be more consistent with the accounting aspect of the contract and will freeze each data point into the future. The contract purchase/effective date, cancellation reported date and claim reported date could be used, but the points of each line — especially the end points — may change when the graph is refreshed in the next period due to posting and payment lags.

You may be asking, “Do we need a data warehouse that captures a snapshot of our entire database at each calendar month-end?” The answer is ”No.” These graphs can be built with your current inception-to-date database. Without going into the details, all it takes is some creative arranging of your data, some fairly sophisticated database programming and data visualization software. (Tableau is used here.) This can be done utilizing Excel, but data visualization software will make these graphs much easier to build and refresh. Additionally, data visualization software can help you quickly dive into the data to identify unusual data points and understand the origin of trends. This type of analysis can be built one time and requires little maintenance on a regular basis.

Development of Other Measures

Since loss ratios are dependent on premium levels, varying premium levels may cause differences in written loss ratio development curves even if the mix of business and loss levels were identical across all years. However, we can remove the impact of changing premium levels by building development graphs on other measures that are independent of premium such as claim frequency (claim count per contract) and loss costs (losses per contract). Similarly, we can evaluate cancellation activity by graphing cancellation frequency percentage of contracts cancelled per original contract) and refund ratio (percentage of premium refunded per original premium).

Earned Loss Ratio Distortion

So how distorted are the earned loss ratios? Unfortunately, we can really only test this in hindsight since no contract year’s business is created equal. However, as long as mix shifting is gradual, we can get a reasonable idea of the appropriateness of the current earnings curves by observing the earned loss ratio development as displayed in the following graph:

*Note the amounts provided in the graph above were developed for illustrative purposes only and do not represent actual results. Amounts in this graph are not consistent with amounts in previous graphs.

The above graph shows how the cumulative earned loss ratio for each contract year has grown as of each calendar month-end. In a perfect world, the earned loss ratio for a given contract year would be the same at 12 months of age as at 120 months of age. However, even if perfect earnings curves were used for all segments, any difference in profitability between shorter and longer term contracts will result in an increasing or decreasing earned loss ratio as the contract year matures. So once again, homogeneity and the mix of business must be considered not only when comparing one contract year to the next, but also when examining the development of a given contract year. In general, to appropriately examine the earned loss ratio development of a given contract year we must filter the data to a level that we believe will not lead to misinterpretation due to differences in profitability among the rating variables — especially term length.

Utilizing graphs of earned loss ratio development can assist in further refining the earnings curves, but building these graphs takes sophisticated database programming and substantial computing power and space. Do you need a data warehouse that captures a snapshot of the entire database at each calendar month-end to build earned loss ratio development graphs? The answer is still ”No.” The current inception-to-date database can be used. However, since we need the earned premium for each month-end, we need to build a table with a separate record for each month in the life of each contract. Said differently, a 36-month contract occupying a single record in your database would need to be expanded into 36 records in this table. For a database with hundreds of thousands of contracts, populating a table with tens of millions of records to build the above graph could easily require tens of gigabytes of space.


The graphs presented in this article can help us identify problem areas early, compare contract years without the distortion of earnings patterns, identify shifts in the mix of business, break down the experience into its components and help us make sense of the earned loss ratios.

Are these perfect tools? No. All methods of evaluating experience have strengths and weaknesses, but there are other useful tools to monitor experience without the dependence on or distortion of earnings curves. There are many variables driving loss and premium development, only some of which can be captured in the data and possibly used as rating variables.

Being able to explain every difference between point X and point Y is impossible. However, in our experience, these development graphs are helpful in showing that something has changed and provide some indication of the significance of the change. After a difference has been found, the next step is to figure out what has changed and why the results were impacted.

In our next article, we will examine various ways to visualize data in an effort to investigate and quantify the impact and cause of mix shifts and unprofitable outlier segments.

Posted in ActuaryComments (0)

Actuarial Approaches to Maximize Profits

I recently had the pleasure to serve as moderator of a panel presentation titled “Actuarial Insights into Today’s F&I Products” at the P&A Leadership Summit. The panel was made up of three independent actuaries: George Belokas of GPW & Associates, Lee Bowron of Kerper Bowron & JH Briscoe, and Michael J. Covert of Perr & Knight. For those of you who were unable to attend this session, here are some interesting vehicle service contract (VSC) highlights discussed at the session.


The panel started at a high level to review methods used to analyze a VSC portfolio. The traditional approach involves loss triangles organized by original purchase date, where loss emergence is tracked over the life of the contracts. The end result is a triangle where the oldest contracts are fully earned and the newest contracts have only a couple of data points. The triangles allow a comparison of one year of experience to another to see how losses are trending. A loss emergence pattern can then be selected for a given block of business and applied to forecast ultimate losses for each policy year. This approach is fairly easy and works well for rate filings and statements of loss reserve opinions, but can be problematic if the underlying business is changing.

For example, consider the changes that were made by many manufacturers back in 2006 and 2007 to increase the term of their underlying warranties. A loss pattern based on 2005 data and used to project 2007 ultimate losses would be inaccurate because the exposure on the 2005 contracts started at an earlier point and generated higher losses during the first two years. One technique to combat this problem is to segregate the triangles by variables such as underlying warranty, starting odometer miles and term, but this can generate a lot of triangles for review. If a program had just five mileage bands, five underlying warranty variations and five terms, there would be 125 different loss triangles to analyze (assuming they were all large enough to be credible).

An alternative approach is to remove the triangle concept and consider the underlying unit of coverage; which is the number of miles driven per VSC outside the manufacturer’s warranty. This method requires an analysis that takes every contract and strips out the manufacturer warranty and reviews driving patterns to calculate a base cost per mile driven. Driving patterns can be developed by analyzing cancel and claim data to see how fast customers are expiring their contracts. This type of modeling is more complicated and not ideal for a rate filing, but it allows for more detailed analysis over a number of variables such as starting mileage, coverage, deductible, class, age of contract, and whether it’s reinsured. This model could even be used to develop an earnings pattern specific to that contract.

Impact of Starting Odometer

The panelists noted that there has been a large increase in the starting odometer on used VSCs. The typical used car mileage bands have been in the 40-60K mileage range, but many providers are allowing higher mileage vehicles into their programs; some even up to 150K miles or more. As one would expect, these older vehicles tend to have more losses, which increases the claim frequency. In addition, the average claim cost is higher because more expensive items are breaking. What the actuaries see from providers is that some of these older vehicles are not priced adequately to reflect the exposure and the segment is typically unprofitable. A mix of business shifting towards a higher mix of these vehicles may be particularly alarming.

New Car Profitability

New car profitability continues to increase year over year. In other words, if all business were combined for a single term/mile segment, a VSC written on a new 2009 model year vehicle is more profitable than a comparable 2006 vehicle. Possible explanations include increased underlying manufacturer warranties and continuing improvements in vehicle quality. Another possible explanation is a migration toward Asian vehicles from domestics. This trend is supported by provider portfolios as well as public data from industry sources such as NADA. Asian vehicles traditionally have had lower loss costs and the shift in business could explain an overall increase in profitability of new cars. The panelists agreed that it is too soon to tell what the impact of more sophisticated technology will be on new car loss experience.

Distribution Channel Impact

The distribution channel impact on profitability is largely a pricing issue. Many providers use the same reserves (i.e. “rate charts”) for a VSC sold by a dealer as they do for a VSC sold to a customer shopping on the internet, despite clear evidence of adverse selection on the latter. From a profitability standpoint, the best time to sell a VSC is at the dealership or financial institution when the car is new and the car’s future performance is unknown. As the car ages, defects become known and the opportunity to shop in anticipation of a claim increases.

Mix of Business

A typical VSC rate chart tends to be priced by component coverage, new/used, term/miles, and initial odometer mileage. The result is that one can end up with very lengthy rating manuals that are difficult to analyze and price. The panel suggested that not enough time is spent pricing each individual segment. For example, as a general rule, most classes are priced incorrectly. Asian vehicles are typically priced too high, domestics are about breakeven and Europeans are generally underpriced. But huge rate charts make it very difficult to discern these differences. If one slices the data by vehicle make, these differences would become obvious since the Asian loss ratio would likely be too low and the European loss ratio too high. The same analysis should be conducted on each rating variable, including term and initial mileage. Shorter terms are easier to analyze since losses emerge quickly, but one must pay close attention with longer term business or there can be an unpleasant surprise several years down the road after it’s too late to turn the ship around. Without digging in at the outset, it’s hard to understand what is happening and why.

Reimbursement vs. Default Coverage

Most VSC business is insured via a reimbursement type insurance policy where claims are paid when the vehicle breaks down. There is a significant amount of historical data to analyze when pricing that coverage, which increases an actuary’s confidence in the rate adequacy. There is a lesser used type of insurance policy that is only triggered when there is a default by the contract obligor. The reason for default is often related to financial, economic, or other market related reasons that are much harder to predict. Thus, the question of profitability under one type of coverage versus the other can be tricky. Default coverage has a very low frequency, especially when the economic environment is good, and thus a lower insurance fee. But if a loss does occur, the severity is significant. By definition, any individual default policy will either be priced too high or too low and one just hopes that in aggregate, an insurer’s premium and surplus is adequate to cover any potential default. There is also the question of when to earn an insurer’s premium with default coverage. Some choose to wait until the contract expires, earning 100% at contract termination.

Why Do Some VSC Providers Lose Money?

Why do some well-intentioned providers lose money on an auto VSC portfolio when there is so much data to analyze? The answer often comes down to poor rate development and inaccurate earning curves. Rate chart development is typically performed by 1.) finding a competitor and copying theirs or 2.) performing an actuarial analysis to determine rates. The risk with copying a competitor is that plans are often lengthy, difficult to understand, and not priced correctly. And over years of rate changes and tweaking, the underlying relativities that once existed among the rating variables (such as class plan, new vs. used, mileage) have become blurred. So if you copy a competitor plan you could very well be replicating their pricing inadequacies. The more appropriate approach is to take an existing rate chart and try to back out a set of base rates with relativity factors. A new rate chart could then be generated by multiplying those out to reestablish those important rating factors among variables. One must also consider a competitor’s underlying forms in conjunction with a rate comparison to make sure that any coverage differences among programs are addressed. For example, the competitor might have different car components covered in their plans that don’t cleanly map to the plan being reviewed.

Incorrect earning patterns also cause some providers to lose money because a flawed earnings pattern can disguise poor loss results. For example, a pro rata earnings curve might be an acceptable curve for used VSC’s initially but over time as historical data is built, customized earnings curves should be created to more accurately predict ultimate loss ratios. A correct earnings pattern captures the emergence of losses so that the loss ratio remains constant over time. Said another way, the loss ratio evaluated at month one should be the same as in month 60. VSC providers can use many different customized earnings curves within their portfolio to capture important differences by segment. Any of the three actuaries in attendance at the P&A Leadership Summit would be happy to assist you with this project.

Posted in IndustryComments (0)

Big Data – What is it?

Big data is the new buzzword. Business publications are full of stories about big data. A Google search will yield a lot of information about big data — and a lot of vendors trying to sell you on Big Data. But what is Big Data? It depends on whom you ask. But a general rule is probably that it is more data than you know what do with, or how to handle.

There are several problems with the Big Data paradigm that is being reported in the media and marketed by vendors.

1. It is largely driven by consulting and computer companies, who are always eager to jump to on the next bandwagon. In many ways, they are overselling the potential benefit, which inevitably leads to disappointments. Already, one recent media article declared that the era of Big Data is dead.

2. For service contract providers, we’ve always had lots of data. The amount and type of data is not really changing. Most of the explosion in data in other industries is from automated sources, such as radio frequency transmitters, monitoring stations and smartphones. Industries that use these technologies are experiencing the surge in data.

For example, Progressive Insurance has a product that monitors the driving habits of the insured. This would record the speed, braking, length of the drive and the time spent driving (along with other factors). This information is transmitted to the company and calculates a new rate. This is significantly more data than normal insurance company transactions.

3. Another issue is that even if you have an explosion of data, you might need only to take a sample to analyze it. After all, you don’t need to ask every voter to compute a poll. We often use a technique known as “bootstrapping” where we only analyze a sample of the data, but do this process multiple times. The difference between the predictions of different models is often indicative of how predictive the underlying data is. However, since we are only analyzing a sample, the amount of data is usually not an issue and the calculations are much quicker.

4. Most importantly, it is really not about the data, but what you do with it. Data should drive business intelligence, which should then drive actions. The data is only the first step.

Big Data (or maybe just data) – What can you do with it?
One of the most helpful things we like to do with service contract data is to adjust the exposures (an exposure being one month of service contract coverage). First we adjust the exposures for the underlying manufacturer’s warranty. If the manufacturer’s warranty is covering the claims, we want to eliminate these months.

Second, we adjust for average driving patterns. At the end of the contract, some drivers will have “driven out” out of their coverage — we want to eliminate these. Finally, we want to adjust for the age of the contract. Cars may have more repairs as the contract ages, though this is offset some by sales, total losses and the “forget factor.”

Using this type of analysis allows the administrator to build a powerful model, which can predict losses and is not dependent on segregating data into small buckets based on the type of coverage and term. In this way, you can calculate the current exposures for a contract and the future exposures. A base cost is calculated from historical data and adjusted for deductible, coverage, purchase mileage, class and any other factors.

What else can a VSC writer do with their data? Here are some ideas for projects:
• Analyze the characteristics of dealers who terminate their dealerships. Build an early warning system that predicts the probability that a dealer will terminate their business in the next 12 months.
• Build a profitability index for dealers based on volumes and underlying business.
• For a dealer, build a model which predicts the probability of a sale using historical close data. This could vary by salesperson, customer characteristics, vehicle or even time of purchase!
• For an administrator, build a model which predicts whether an inspection will be cost effective.

The Big Data revolution may be oversold — but that doesn’t mean that your data is not the key to increasing customer penetration and profitability.

Posted in ActuaryComments (0)

Page 1 of 3123