Launching a New Program

–You need to know about variations in your program's class of business. Let's say you're going to do a social-service program. In the social-service industry, you have halfway houses, Meals-on-Wheels programs, etc. The list goes on to about 100 subclasses. You need to figure out which subclasses you're going to be in.

–You need to know how the program's members make money. For example, we have a program for the senior-living industry, which includes assisted-living facilities. One of their goals is to operate at full capacity. The more residents they have, the more money they make. To remain at full capacity, however, they may keep people too long. Eventually, residents become too old or too sick for an assisted-living facility to care for. At some point, skilled care–i.e., a nursing home–becomes a better option. But if an assisted-living facility is too focused on the bottom line, it may keep people beyond the point where they can care for them properly. That creates a liability issue. You need to be aware of such issues as you go into a new program.

–You need to know the trends. I'll give you another example from the senior-living market. We are seeing an increase in what are called "congregate care" facilities or "retirement communities." These essentially are facilities that offer independent living, assisted living and skilled care on one campus. Currently, these facilities represent about 15% of the industry's beds. But in terms of new construction, they represent 50% of the beds. That's the new wave. Whatever your niche, you need to know what's coming down the pike.

–You need to know the industry associations. Associations can be extremely valuable sources of information. When we decided to start a senior-living program, we sponsored a cocktail party for all the state association presidents, at a convention of the national association. They gave us a lot of information and ended up inviting us to a presentation the next day on the industry's demographics. Associations also often have access to databases that you otherwise wouldn't know about.

–You need to know the competition. We're big believers in identifying the one or two competitors that are really the "players" in a program's niche. If you can't compete with them, then you're unlikely to achieve much penetration into the market. Assess their strengths and weaknesses. By understanding how they do business, you might determine how you can do business a little differently, and thus find a place for yourself in the market.

–You need to know the coverage requirements. If you are in too much of a hurry to get your program off the ground, you may overlook vital coverage requirements. For instance, you initially may tell your carrier that you can do the program on a nonadmitted basis and only later discover that admitted paper is crucial to the program's viability. Maybe claims-made coverage won't do, or perhaps an umbrella is crucial to penetrating the market. You have to know all these things up front.

Perhaps the best way to find out what you will need is to approach producers and get copies of policies they're selling in your startup's niche. In some cases, you also can obtain copies of competitors' policies from state insurance department filings, although that approach can be hit-or-miss. If an insurer hasn't made a filing lately, your chances of finding whatever they originally filed may not be good.

Underwriting considerations

Assuming you know enough about your program's market, you can prepare proposed underwriting guidelines. Among other things, credible underwriting guidelines will help you demonstrate to carriers that you have expertise in your program. In fact, without underwriting guidelines, your proposal is going to be dead on arrival.

You must be able to provide solid reasons why you want to exclude certain accounts and write others. A well-reasoned ineligible-accounts list is vital to your credibility. You also need to have an idea of where the pricing needs to be. Other underwriting considerations include the following:

–Referral accounts: Determining which accounts will have to be referred to a company's underwriter usually comes toward the end of negotiations with the carrier. Sometimes the insurer will say they're not crazy about this or that exposure. By the time they enumerate all their concerns, they may have narrowed your underwriting authority so much that you won't have a viable program.

One option in such cases is to reply that you will refer all accounts of concern to the carrier's underwriting staff for approval. That can work well if the insurer doesn't narrow the field too much. If the insurer is unreasonably tight, you can just load up their underwriters with submissions–which I confess we have done. When the company underwriters get out from under the avalanche of submissions, the insurer may "get real" and give you a wider underwriting box.

–Account selection: The above notwithstanding, you must demonstrate to the insurer that you have a plan to give the carrier a preferred class of the business, or at least a subclass within the segment that is going to produce an underwriting profit.

Marketing considerations

Even if a carrier loves your program, you will have to explain how you intend to sell it and produce a reasonable volume. So you must have a distribution strategy that makes sense–whether you're selling direct to insureds, through a limited number of producers who have expertise in the field or through a broad producer network. Other considerations include the following:

–Advertising: You need to have a good idea of what will work best. Direct mail? Print advertising? E-mail blast? Perhaps you can turn to vendors within the industry who have a vested interest in seeing that practitioners are well-insured. We've used vendors quite successfully. We have a program in which the manufacturers of equipment our prospects use have been desperate to see them get proper coverage. We've actually used the manufacturers' newsletters to advertise our product. They've been happy to let us do so.

–The marketing rollout: However much you may think you know about a startup program and its market, you're still on "training wheels." After you launch the program, you will continue to learn how to underwrite and handle the business. What we like to do–and I think this helps reassure carriers–is initially promote the product in a small territory. Alternatively, we may approach a small number of producers to launch the program, rather than advertise it broadly. That gives us a chance to get our feet wet, to get a good fix on what submissions are going to look like, and what we're going to want to write and not write. By starting small, you don't find yourself unexpectedly inundated with submissions and then behind the eight ball in regard to service. Also, you may discover an underwriting issue you previously overlooked and have an opportunity to address it before really rolling out your program.

Other considerations

Since carriers take a big gamble on startup programs, they want to know what kind of resources a program administrator has. Do they have good underwriters? Do they have systems for issuing policies appropriately? What is their track record on their existing programs?

We're big on quality control. We have a full-time auditing department that audits our underwriters monthly. The underwriters are "graded" on such things as whether they've calculated five-year loss ratios and factored them into pricing or declinations, and whether they've obtained MVRs and have underwritten them properly. We post underwriters' scores in the break room and send them to insurers, so they can see how each underwriter performs monthly.

We also have an underwriting-assistant audit, which is basically a compliance audit. It determines whether underwriting assistants are issuing policies correctly, with the right forms and endorsements, the right effective dates, the right limits, etc. Their scores are posted too. These internal controls don't replace a carrier's audits, but they help reassure an insurer that we run a tight shop.

One thing you must determine from the outset is whether the market is going to be big enough to justify your efforts and satisfy the insurer. That can be tricky. One of the first things most program administrators do is check the number of businesses specified in the applicable SIC code(s). Sometimes, however, SIC codes aren't that well-managed. There may be classes within an SIC code that don't fit what you're doing. For new or evolving classes of business, there might not even be an SIC code. As an alternative, industry associations can be good sources to check; often they have done this sort of research themselves.

Sometimes you can find information sources on the Internet, but you have to be careful. You must validate the information, because anyone can put anything on the Internet. You can't assume the data is correct–or isn't outdated. You don't want to be well into negotiations with a carrier and have it discover that your figures are 10 years old. You must take the time to validate your data with multiple sources. I don't think we've ever tried to launch a program using population estimates based on just one source of information. The more sources you have, the better.

After estimating the population of an industry segment, you must estimate how many will be eligible for your program. If your underwriting guidelines are so tight that only 10% of the accounts will fit them, you need to know that in advance.

Once you've estimated the population of your industry segment, the number of eligible accounts and the average premium, you can come up with a rough estimate of how much premium you theoretically could write. Then you multiply that figure by a factor to reflect what your market penetration realistically will be. Unless you are in a hard market, it's extremely unlikely that your penetration will exceed 5% in the first year or 15% in the third year.

Defining your competitive advantage

For any program, the big question is, What's your competitive advantage? There are a number of possible answers to that question, including the following:

–An attractive price: Maybe you can offer low rates, based on superior risk selection. Perhaps you've found a subclass within a niche that is better than the overall niche. In negotiations with your carrier, you might be able to justify aggressive pricing for that subclass, knowing that you can still make a good underwriting profit–if you have valid loss information backing you up. Reducing coverage is another way to obtain a lower rate. Maybe the members of your niche are not that big on some coverages typically included in packages sold to them. If you can remove those coverages, you can come up with a more competitive rate.

–Value-added services: A great example, of course, is risk management, particularly for programs geared to larger accounts; i.e., those paying more than $50,000 in annual premium. If you have a great risk management department, you can help your large insureds run their businesses better and reduce their losses. Ultimately, that will lower their premiums.

–Crucial coverages: For example, in the late 1980s, we had one of the first programs to offer sexual-abuse coverage to daycare centers. The coverage was claims-made, offered as a sublimit and was quite restricted. But it was coverage, and that gave us a competitive advantage.

–Outstanding service or ease of doing business: This can be a compelling advantage when your program deals with "transactional business"; that is, accounts with annual premiums less than $5,000. If you can get out quotes quickly while not requiring excessive amounts of information, if you can issue policies rapidly, etc., producers–and even insureds–are likely to look on your program favorably.

Case studies

I'm going to give you four examples of startups we've done to give you an idea of how the process can work under different scenarios. My first example is a program we targeted to a subclass of a much larger niche. It consisted of quite small accounts. In fact, when we first came out with this program, the annual premium for the lowest available limit was just $195.

Essentially, our subclass consisted of in-home businesses engaged in more or less the same work as that performed by more traditional businesses in the niche–i.e., those with typical workplaces to which the employees drove, a sizable volume of business, etc. We discovered that the competition was treating our subclass as if it were populated by typical businesses. For instance, they were offering them commercial general liability policies with $1 million/$2 million limits.

We created a stripped-down liability product that offered claims-made coverage and low limits. It contained many exclusions aimed at exposures of more customary businesses. The insureds had to sign off on these exclusions on a separate document that was attached to the policy. While we offered a limited product, its price was appealing to the people in our subclass. Most were working out of their homes because they needed to supplement their incomes. They did not have a lot of cash lying around with which to buy commercial insurance. To accommodate their budgets, the lowest liability limit we offered was $25,000/$50,000. So we approached this subclass in a completely different way from the rest of the market. Today, we're the dominant player in it.

My next example is a startup program that complements an existing program and is for a similar class of business. In discussions with our existing clients, we determined that while we knew a certain amount about this class, we didn't know enough to design and start a program. So we agreed with the carrier to split the cost of hiring a consulting firm with expertise in this particular industry. (The fee was about $35,000.) The consultant introduced us to their clients (our prospects). We spent time in their boardrooms, learning a great deal about their businesses and the carrier from which they bought insurance. We obtained copies of their policies. The consultant also provided much information about the business models followed in this industry.

As a result of the formal study that came out of all this, we picked a couple of subclasses within the industry that we wanted to pursue, although we didn't have hard claims data or loss runs on individual accounts. Still, we had underwriting expertise with a class that had similar exposures. That and the in-depth research we had conducted on the new segment persuaded the carrier to give the startup a try.

My next example is a program we recently launched that serves one of the few niches for which there is still a hard market. This was a hard sell, because many carriers had pulled out of this class in the preceding two or three years because of adverse loss experience. When a producer approached us with the idea of doing a program for this niche, we thought he'd lost his mind. He had some premium and loss data from the accounts he had written but not nearly enough for an actuarial study. The great opportunity was that the market was fragmented and there wasn't much competition. But the thought also crossed our minds that perhaps this class wasn't even insurable.

The agent had an amazing amount of knowledge about this industry, for which the use of a certain piece of equipment is fundamental. The agent had gotten to know the manufacturers of this equipment and had learned what was safe and unsafe, based on his own experience. In fact, he had become so knowledgeable that the manufacturers actually would ask for his opinion about their designs for new equipment. And he would advise them–without compensation. He simply volunteered his services, which gives you an idea of how much passion he had for this industry.

The first couple of times we met with this agent, we really didn't take him seriously. But he finally convinced me to go with him to a convention, where equipment was on display. He took me by one piece after another, explaining why each piece was safe or unsafe. Then he started testing me. He said, "Now I want you to look at that one and tell me whether you think it's safe or not." In short, this guy was a fanatic–and you need a fanatic to start up a program for that sort of industry.

After searching the Internet, we came up with some injury data on this industry. Then we turned to an actuarial service, which used it to develop a rating model for us. Then we talked with the actuaries for a carrier with which we had a relationship for a totally unrelated program. Ultimately, the company's actuaries decided they could accept the model if we'd raise the rates by 10%. We did so and got the program off the ground.

Our competitive advantage was that we could select accounts based on the expertise of this independent agent. Not only had he figured what equipment was safe and unsafe, but he also had determined that there were three types of potential insureds in this industry and had concluded that only one type used the equipment in a way he deemed safe. That was his–and our–market.

This agent had a fair share of the market in one state. We agreed to insure his book. For the rest of the country, he agreed to be our underwriting consultant. We pay him an override on that business. We will not even quote on an account of decent size unless the agent first talks to the prospect. He really is an underwriter masquerading as an agent, and he was indispensable to getting this program launched.

My last example of a startup is a workers compensation program we recently launched to complement an existing program. The startup was for a workers comp class that scares the daylights out of most carriers. However, we had obtained excellent underwriting results from the package product we were selling to this class. There was some overlapping experience between the package and the comp, because the primary exposure in the package was auto. Whenever you had an auto-accident claim involving employees, you also usually had a workers comp claim.

The National Council on Compensation Insurance had a class code for this niche. They also had a number of subclass codes that really didn't seem to fit what our insureds were doing. That left us wondering whether our subclass was one of the better ones or one of the worst. If it was at or near the top, it could be a good class to write.

The NCCI publishes workers comp information for 70% to 80% of the states. We found that a couple of the other states had developed much more detailed class codes than the NCCI had. These states had a subclass for our kind of business (among others that the NCCI didn't list), and the rates for it were much lower than for the subclasses that the NCCI addressed.

In essence, it appeared that the rates charged in our subclass (in the NCCI states) were subsidizing those of the other subclasses. That immediately told us–and our carrier–that we might have opportunity. While the carrier still wanted us to stay out of certain states where they felt rates weren't adequate, eventually it agreed to a program that operates in more than 20 states. Another factor that helped us sell the carrier was that we had an in-house loss-control specialist for our package program whose expertise extended to workers compensation.

Starting a new program is never easy, as the aforementioned case studies illustrate. The process can be slow and frustrating. You end up going down a lot of dead-end streets, so be prepared. Winston Churchill once said, "Success consists of going from one failure to another without loss of enthusiasm." That's how you have to pursue a startup program. You will run into one wall after another. Whether you succeed will depend on how determined you are, and how hard you're willing to work.