Planning on uncertainty: planning processes could be improved by relying on market indices rather than market forecasts.

"Oh no!" I exclaimed, "here it is again, and I haven't even recovered from the last one." In response to my colleagues' puzzled looks, I explained: "I've been summoned to participate in this year's planning process--another four weeks of intensive meetings that totally waste everyone's time."

"Maybe you think so," said my boss, who had overheard my complaint as he entered the room. "But how would we all function without such a planning process?"

I had to admit that my boss--this particular one, anyway--was right, as usual. To influence him and his superiors, I needed to stop grousing, identify what was wrong with our process and propose some remedies. After some careful thinking, I stopped by his office a week later.

"I think I know how to improve our planning process," I explained. "We spend way too much time haggling about a most probable scenario, which involves making assumptions about all sorts of market conditions that affect our business. Everyone knows that this scenario will be used to create budgets and profit targets, so all the participants have axes to grind. Senior management wants aggressive assumptions that result in high profits and growth, while division managers want assumptions that make their bonuses less difficult to achieve. Given these conflicting interests, it's no wonder that agreeing on a most probable scenario is a long and tiring process."

"You're right," replied my boss, "but I don't see the difficulty of agreement as a terribly serious flaw."

"I agree," I responded hastily. "The real problem is that what actually happens never corresponds to the most probable scenario that we all agreed upon. If market conditions are better than assumed, meeting profit and growth targets is a slam dunk and everyone looks like a hero. But just as often, actual market conditions are worse than we assumed, but everyone is still held to their original profit and growth targets. To meet them, managers naturally tend to take more risk. Sometimes this works, but more often it results in losses that shouldn't have occurred;'

"So what you're saying," my boss interjected, "is that when conditions are better than we assumed, we do less well than we could, and when conditions are worse we take ill-conceived corrective actions that increase our exposure to loss. Can you give me some examples?"

"Sure," I replied. "Take a corporate risk manager. If market conditions impose higher costs than planned in his corporate budget, he has a strong incentive to meet targets by reducing coverage or raising deductibles, regardless of whether this is wise. Similarly, insurance underwriters, in a soft market, have an incentive to reduce prices or write more risky business to meet planned profit and volume targets."

"That's a good diagnosis," my boss said, confirming my favorable view of his capabilities. "But what remedy do you propose?"

"One solution," I replied," lies in taking a clue from the investment world. We don't meet with a money manager to agree on a most probable scenario for the market and then hold him or her to a fixed return target based on that scenario. That's impractical, since we know in advance that what the market actually does won't correspond to the scenario we adopted. Instead, we agree on a relevant measurable index of market performance, and then compare a manager's performance to that index."

"So for the corporate risk manager or the insurance underwriter," I went on, "the planning process would consist of identifying and adopting an appropriate index of insurance prices. At year-end a risk manager or underwriter would be rewarded for his performance relative to this updated customized index. For example, if relevant insurance prices went up, the risk manager's performance would reflect those higher prices, and an insurance underwriter would be held to higher growth and profitability targets."

"Intriguing," said my boss, thoughtfully. "But I know I'll get asked why, if this idea is so good, it hasn't already been adopted."

I was prepared. "Simple. Until recently the necessary market data just wasn't available. Even now it's a bit spotty. And it takes ingenuity to create the necessary indices. But we're in the same situation that investors were in a decade ago, so we can learn from their successes and mistakes."

"OK, let's write a proposal," he replied. "But meanwhile," he said, grinning, "don't forget to attend the planning sessions that start early tomorrow morning!"

William H. Panning, a Best's Review columnist, is executive vice president at Willis Re Inc. He can be reached at [e-mail].