Eight reasons why programs fail to fly

Armed with reams of data, we started looking for common problems in failed programs of the past-so they could be avoided entirely or dealt with in an effective manner. We identified eight major factors, upon which I'll elaborate in this article.

1) Problems with profitability. These problems include the following:

• Failure to understand profitability issues. We found that many program administrators (and insurers, for that matter) didn't grasp how a program's profitability is affected by loss development, how a loss in today's dollars is usually a multiple of that number three to five years from now. They failed to account for the impact of such trends as social inflation, legal inflation or medical inflation. Nor did they grasp the effect that increased limits factors can have on the long-term profitability of an excess casualty book.

• Failure to subject the portfolio to actuarial analysis. In the last few years, it has been harder for program administrators to place programs with carriers–and for carriers to evaluate the mountain of program proposals they receive. One simple solution, but a vital one for both parties, is to obtain an actuarial study of the program. Our experience suggests that program administrators who have their programs analyzed by an actuary, whether an employee or an independent contractor, markedly increase the odds their programs will be accepted. Even though we subject our entire book to this kind of analysis, we think it's healthy for program administrators to do the same. We love nothing more than to have their actuaries and ours together look over the data and resolve–or at least identify–any discrepancies in their opinions.

• Failure to act on actuarial findings. This factor jumped off the page at us. Sometimes program administrators go to the time, trouble and expense of having actuaries look at their programs–and then just file away the reports. We saw this in several instances. You have to do something about any problems an actuary uncovers.

• Failure to monitor changes in rates and exposures. Changes in rates–up or down–will affect profitability, of course. But don't overlook exposures, either. The general economy has been growing for the last 24 to 30 months. You need to factor changes in insureds' revenues, payrolls, property values, etc., into the actuarial analysis of your program's profitability. Analyses need to be updated to reflect the results of premium audits and underlying changes in the book itself. For example, payroll shifting into and out of various SIC codes in a program for contractors can have a material impact on exposures in the book. If these changes are not recorded, tracked and acted upon, you may have an inaccurate picture of where the book is going.

• Falling into the "too little, too late" syndrome. Essentially, this means taking only incremental steps to restore a program's viability when more drastic measures are needed. It's like the noise in your car: It usually doesn't go away by itself, but rather needs diagnosis and treatment.

2) Problems with claims handling. These include the following:

• Conflicts of interest.We have several programs in which a program administrator owns the third-party administrator handling a program's claims. There is an inherent conflict of interest in such relationships, however, and smart program administrators and their carrier partners understand that conflict and actively work to minimize it. Unless well-documented claims handling agreements are carefully constructed, with responsibilities of both the underwriting side and the claims side spelled out contractually, conflicts can arise that can lead to the failure of the program.

• Failure to spot claims trends. In many failed programs, we noticed that an external factor of one sort or another began to affect a program's profitability but was not identified until it resulted in a crisis.

• Failure to derive book-of-business findings from claims analysis. Many insurers have learned the hard way that such claims analyses should not be overlooked. To prevent this from happening to us, our loss-control people prepare type-of-loss and cause-of-loss analyses on a quarterly basis for each of our programs, both in terms of claims counts and incurred loss dollars, as well as loss adjustment expense. When an insurer provides this information to its program administrators, they can use it to adjust their underwriting, claims management, risk management and loss-control procedures.

• Failure to manage litigation. Litigation expenses have been the death of some programs. They will skyrocket, if permitted to do so; and it is incumbent upon both the program administrator and the carrier to see that they don't. One way to rein in these costs is to try mediation or arbitration before concluding it is necessary to go to trial. Another is forcing your defense counsel to show you a budget–before commencing litigation-and then requiring the attorney to stick to it. It's also helpful to have defense counsel provide you with quarterly or even monthly status reports on all claims in litigation. Even if a program administrator is not participating in the underwriting risk, we believe it is imperative that he or she know what's happening in the claim files that are in litigation.

3.) Problems with underwriting. These problems include the following:

• Failure to develop and adhere to risk selection and pricing criteria. Many insurers are amazed and a little embarrassed to admit that some of their programs have this elementary problem. But some programs have underwriting guidelines that fit on a single piece of paper. A successful program, on the other hand, will have documented underwriting guidelines that are updated annually.

• Failure to understand the underwriting performance of the book. What's happening from one year to the next? What's changed? Geographically, where have you been successful in writing business? Where have you lost accounts? If you are writing all your business in one of the bottom 10 states in the U.S. Chamber of Commerce's litigation study, you could have a problem in your book that will show up in two or three years when losses start developing.

• Failure to keep up with the external environment affecting the portfolio. Underwriting should change, for example, to reflect changes in state or federal law. Such changes can create problems–but also opportunities to add a product line, enter a new territory, etc.

4) Problems with segregation of duties. This issue came up more than we expected. Examples include the following:

• Failure to segregate management from all other staff. It's critically important that a program administrator's senior management (and the insurer's, too, for that matter), be actively involved in the business. But the people who report to senior management must have the authority, experience and training necessary to do their jobs-and then they should be held accountable.

• Failure to segregate sales and marketing from underwriting. In some failed programs, salespeople have the authority to bind business-sometimes before underwriting has a chance to assess the risk and make an informed decision. Quite frankly, we still see some of that going on today-although not in any of our programs-and we find it surprising and troubling.

• Failure to segregate underwriting from claims. I previously mentioned the inherent conflict of interest between program administrators and TPAs. A common feature of failed programs is that someone other than a claims professional was deciding whether or not to pay claims.

• Failure to separate finance from all other operations. The accounting, comptroller's and treasury functions of a program administrator require the same segregation of duties, job descriptions, experienced personnel and periodic audits of performance that any large or public company does.

5) Financial problems. Here are examples of problems we identified.

• Failure of the program administrator to operate as the insurer's fiduciary. The program administrator has an integral fiduciary responsibility to the carrier, and their organization must be geared to fulfilling it without fail in all transactions.

• Failure to understand and address credit risk. Most programs run on a credit basis, using an extended bordereau, account current, etc. Typically, there is a long interval between the effective date of coverage and the date the carrier receives the premium. Whether we like to admit it or not, that's a credit risk. Those of us who have been in this business awhile understand that the rating agencies take a dim view of carriers that do not aggressively manage the credit risk in their portfolios. This is one reason why, generally speaking, program carriers are thought poorly of by the financial community–because of all of this enterprise risk, which doesn't exist in a traditional brokerage environment.

• Failure to reconcile premiums and cash. One problem that amazed us was that some program administrators were not reconciling the cash coming in off the account current with the premiums being reported on the bordereau or other kind of policy register. A wise insurer will put in place mechanisms that track both sides of the process and look for the anomalies, e.g., a premium reported under a policy number, but with no corresponding cash on the account current, or vice versa. This task requires a considerable investment in people, systems and technology. Lexington has a team of employees who reconcile these numbers all day long and do nothing else.

• Failure of senior management to have and test financial management controls.

6) Problem with risk management such as the following:

• Failure to understand the business being written. We interviewed some program administrators we used to do business with, and a common theme was, "The business changed, and we didn't catch it."

• Failure to identify and address changing risk exposures. For example, California recently enacted legislation for homebuilders that gives them a chance to fix homebuyers' problems before they can file construction-defects lawsuits against them. It remains to be seen whether the change will be for the better or worse-but it is a change. Program administrators and carriers need to be aware of such changes and what they might mean to clients and to the portfolio.

• Failure to upgrade the portfolio. Program administrators should continually look for the "bottom 2%" of accounts, from either a loss-ratio or exposure standpoint, and either make them better through underwriting, risk management and loss control or make them go away. They also should continually look for opportunities to use risk management to improve the book.

7) Problems with technology. Just a few of the main points here:

• Failure to automate processes. It has become nearly impossible to run a program in a manual environment anymore. The sheer volume of work, for all intents and purposes, requires the use of automation.

• Failure to test for data anomalies. One common scenario is the failure to use technology to look for "outliers," unusual data values that may signal errors. For instance, if a policy with a $500,000 limit turned up in an excess casualty written through a program administrator where the insurer never issues an umbrella policy for less that $1 million, the insurer should be able to identify that policy as having been coded incorrectly.

• Failure to upgrade technology. We all wish we could write a check and get a technology platform that serves our needs forever. Technology, however, must be continually upgraded, and in failed programs, it often isn't.

• Failure to adopt disaster recovery plans. In light of last year's hurricane season and similar disasters, we have spent a lot of time talking to our program administrators about their plans for resuming business after a natural–or, God forbid, man-made–disaster. We're also spending a lot more time upfront with prospective program administrators, asking tough questions about their systems technology and how they're positioned to respond, should something happen.

• Failure to own the source code. We added this point because of a specific case in which a program administrator and his IT manager had a falling out. When the IT manager departed, the program administrator learned he did not own the source code for the software that ran his proprietary rating engine and policy-issuance system. It took him months to rebuild the software and get back to issuing policies in an automated environment. The IT manager was an independent contractor, and there had been no formal agreement on who owned his work product.

8) The interests of the parties are not aligned. Aspects of this problem include the following:

• A failure to communicate.Management at both the program administrator and the insurer must talk regularly. Senior executives on both sides must know each other well and know what issues are important to both parties. Together, they need to identify and address program deficiencies and changes in the external environment.

• Developing a "too big to fail" mentality. On the carrier side, management can fall into a trap where, even when the program is going poorly, management believes it has to be retained at all costs. On the other side, the program administrator might think, "I'm just too big a part of their portfolio. They won't get rid of me." In fact, such thinking often leads to the biggest failures.

Overcoming the problems

The solutions to all of the problems discussed in this article are contained within the problems themselves. Here's what we think needs to be done on a clear, consistent basis.

• Have a clear strategic vision. That is, both parties must have a complete understanding of what they are trying to accomplish. The goal for us is to articulate to the brokerage community why we are in the program business, what we're looking to get out of it, and what we can accomplish together.

• Have definitive operational tactics. From the articulated vision flows the specific steps that must be taken daily to achieve desired results. Progress toward the objectives must be measured, and the findings can't just be talked about and filed away until the next quarterly meeting or annual review.

• Pay attention to people and infrastructure. It has taken an enormous investment of capital on our part to create a business unit that can effectively deal with program administrators and with program business. The same kind of investment is required by program administrators who want a business model that can weather peaks and valleys and be sustainable over the long term.

• Stress partnership and communication. Those are easy words to throw around but much harder to put into practice. They require a lot of work and attention to detail, but are essential to long-term success. We've had a number of programs for years, but that is only because we and the programs administrators have put in the necessary hard work. When we get together to discuss strategy and objectives, we both know we're going to roll up our sleeves and not quit until we've agreed on a plan.

• Audit, audit, audit. Program administrators who work with us know we live, eat and breathe that mantra. Continuous auditing of all parts of a program relationship has been the singular most important reason for the success we've had in the last five years. We no longer can afford just to assume that something is going to get done. The programs we have known the most about–because we have audited them every step of the way–have been the long-term survivors.

I'm not referring just to underwriting audits. Our financial auditing team tracks the money all the way through the program administrator's processes until it gets to our bank account. It has been an eye-opening experience for some program administrators to see the weaknesses we found in their financial controls.

We also do risk management audits. We turn our loss control people loose and say, "What does this book of business tell you? What's happening in the external environment? What's happening within the portfolio? And what can we do with that information?"

We also do claims audits. We have several third-party administrators under contract to do claims adjusting work for our programs. It's vital to know whether they are doing the job they were contracted and paid to do for our program administrators and us.

We also do technology audits to see what kinds of systems program administrators have and whether they have been stress-tested for system failure or natural disasters. Backup measures must be in place to ensure the program administrators' viability.

• Last but not least, know when to walk away. There comes a time–even if all the necessary steps have been carried out–when there just is not a fit anymore. When and if that time comes, the program administrator and carrier must be prepared to part company. Usually, it's not the end of the world. Of all the programs we lost over the last 11 years, I know of only two that no longer are in business. I would say 90% of them found a new home elsewhere.

Our analysis of what went wrong in the past has served us well in the present. In 2004, we launched seven new programs and lost two, which we think is a pretty good ratio. In 2005, we remain optimistic about program-business opportunities.