The WorldCom fiasco is shaping up as the kind of story where anyone can find evidence for his favorite business lesson. For many, it reinforces the view that bogus accounting and growth at any cost lie behind the telecom industry's accelerating slide.

But why did WorldCom and other challengers to the industry's status quo fall off the growth curve to begin with? The answer is tied to a fundamental difference between the cost structures of newer challengers and the big incumbents.

The old telecom giants typically find their costs evenly split between fixed items like depreciation and variable items like access or interconnection charges. Given that balance, revenue growth tends to buy margin improvement.

Newcomers like WorldCom, on the other hand, tip more steeply toward the variable cost side—spending roughly two-thirds of their outlays there. The networks owned by upstarts are incomplete, so they must lease capacity. Here, undisciplined growth can lead to a sea of red ink.

Consider the cases of America's Verizon and Equant. With its fixed-cost model, Verizon posted an increase of 7.4% in earnings before interest, taxes, depreciation and amortization from 2000 to 2001 as revenues grew 4.1% over the same period. No such correlation exists for upstarts. Equant saw Ebitda fall 24% from 2000 to 2001 despite a 62% increase in revenues. Revenue growth does not guarantee stronger earnings.

The new kids on the block are admittedly playing against some tough odds. As an industry, telecom companies are flailing under about $2 trillion in debt. World-wide, the sector has lost more than $2 trillion in market value—a 60% drop in market cap—in the past two years. WorldCom's demise pours gasoline on the fire.

In a dismal business climate, however, young companies are most vulnerable. According to our analysis, telecoms don't start to get the full benefit of their fixed assets until they reach very large scale—with total assets approaching $50 billion.

Established companies generally have higher, more uniform levels of network utilization. And the WorldComs of the world feel the squeeze sooner when prices collapse. With prices falling 20% to 30%, or even 50% annually for capital-intensive services, companies have to add new capacity just to earn the same revenue. That plays havoc with the relationship between assets and revenue. Sure, it hurts all players, but it's less of an issue for incumbents, whose top line comes largely from local service, where prices haven't dropped as sharply.

The disadvantages for companies like WorldCom don't end there. The average ratio of capital expenditure to revenue has remained stubbornly high for the challengers as a group. Our research indicates that industry upstarts can't match the capital expenditure ratios of incumbents until they have been in business for at least a decade.

Considering all that, there's a compelling argument for the challengers to try a different strategy against industry behemoths. Once they face up to the fact that their cost structures are different, they'll be halfway to the solution. Optus in Australia managed to offset its disadvantages, turning from a net loss in 1999 to a net after-tax profit of $238.5 million in 2001. This performance led Singapore Telecommunications into acquiring Optus last August. Despite a rough transition, revenues grew 7%, double the overall growth rate for the industry, and SingTel Optus has continued to outperform its counterparts around the world.

WorldCom's meltdown creates tougher operating conditions for upstarts and for the established players who are likely to start buying distressed assets. Survival now depends on how quickly they can adapt their business models to reflect the prevalence of variable costs. Rather than assume that any growth is good, they need to treat growth selectively: What's the real cost of the next dollar of revenue? What's the right price?

For too long, telecoms have run their businesses by measuring Ebitda, as analysts do. But focusing on Ebitda can be dangerous because it ignores the telecoms' scarcest resource—capital. Even before WorldCom's meltdown, investors had started to focus on more reliable performance measures. But it's rare to find reporting systems that emphasize return on capital, and rarer still to find top managers whose incentives are driven accordingly.

In a changed industry, young telecoms cannot assume that growth will automatically drive margin improvement. They must be realistic about the cash-flow outlook for their businesses, and they shouldn't benchmark their performance against the financial ratios of industry behemoths. The only way out now is by playing to their strengths.

Mr. Harrop is a vice president in the Sydney office of Bain & Co. Mr. Altman codirects Bain's Telecom and Technology Practice.