There are many ways to cannibalize a business.

For example, there’s simple cannibalism where a company introduces a new product that eats the old. The hope here is that the new product will grow fast enough and carry high enough profit margins to make up for the demise of all revenue from the old line.

And there’s delayed cannibalism where a company introduces a new product that still eats the old. But the hope here is that the new product will replace enough of the revenue from the cannibalized line that the company can make a transition to a totally different product.

No company faces cannibalism happily. It’s eat your old business or die.

And that’s exactly the situation facing AT&T (T) and Verizon (VZ) right now.

AT&T has opted for simple cannibalism by linking its future to Apple’s (AAPL) iPhone.  (For more on the future of the iPhone see the August 6 post on my JubakPicks.com blog “Apple and Google go head to head in China’s mobile phone market” https://jubakpicks.com/2009/08/06/apple-and-google-go-head-to-head-in-chinas-mobile-phone-market/.)

Verizon has opted for delayed cannibalism by hoping that its cell phone business can be the bridge to a future built on fiber to the home homes and businesses.

What the two companies share, though, is a present in which their land-line phone business is not so gracefully aging.

Take a look at the results that Verizon just reported for its second quarter. Phone lines to homes and business fell almost 10% from the second quarter of 2008 to 3.3 million. Some of that is due to the current recession.  Companies that are laying off workers aren’t exactly rushing to order new phone lines. But the drop continues a trend that began before the recession and has just accelerated now that times are tough and companies and families are trying to save money.  A significant number of phone customers are dropping their landlines and going completely wireless. More than 20% of U.S. households had only wireless service, according to a National Health Interview Survey conducted in the second half of 2008.

Growth in the number of wireless customers looked like the solution for a while—but only as long as the U.S. market was under-penetrated.  Now, however, there are nine mobile devices in circulation for every 10 people in the United States, according to the CTIA, the wireless industry trade association.

You can see growth slowing as market penetration increases.  At the end of 2008, when the percentage of the U.S. population with wireless phones hit 87%, according to the CTIA, there were 270 million wireless subscribers in the United States. That’s more than a doubling from 2000 when there was 110 million wireless subscribers in the United States and market penetration was at 38%.

That slowdown has led to intense competition among wireless service providers for more customers and better customers. In the last quarter, when No. 1 (By number of subscribers) Verizon gained 1.1 million customers and No. 2 AT&T gained 1.2 million customers.  No. 3 Sprint Nextel (S) lost a net 214,000 wireless customers.

But Sprint’s net loss in the number of customers gets much worse when you look at what kind of customers it lost. The company lost 991,000 contract customers—you know, the people who sign 2-year deals and pay a bill every month– in the quarter. Sprint made up most of that loss—remember, it’s net loss was just 214,000—by adding 770,000 customers without contracts that came in through programs like its Boost Mobile unlimited prepaid partnership with Virgin Mobile USA (VM).

There are just two problems with replacing contract costumers with prepaid customers. First, prepaid customers pay a smaller monthly bill, $34 on average to the $56 average paid by contract customers. And second, prepaid, contract-less customers are more likely to change service. What’s called “churn” in the industry, the rate at which customers don’t renew, was just 2.1% at Sprint for contract customers in the second quarter but 6.4% for Boost Mobile prepaid customers.

Since a customer lost means a customer that has to be replaced after expensive advertising and marketing campaigns, the three times higher churn for prepaid customers results in a huge hit to Sprint Nextel’s bottom line.

AT&T’s solution to declining land line numbers and to slowing growth rates in the wireless market is, you guessed it, Apple’s iPhone. In the second quarter, when the company added 1.2 million wireless contact customers, AT&T activated 2.4 million iPhones.

That growth comes at a price. AT&T is subsidizing the new 3G iPhones introduced this quarter to the tune of $325 a phone, according to estimates from Oppenheimer. (AT&T pays Apple an additional $100 per phone for every iPhone sold in an Apple store. The industry subsidy for the average smart phone is $200. ) Those subsidies drove operating profit margins down to 38.3% in the quarter. Without the subsidy margins would have been a little more than 40%.

But the subsidies are worth it–in the long run. On average iPhone users spend about 60% more a month than the average wireless user. And they’re more loyal too. Churn among contract customers fell to 1.09% for the quarter.

Verizon is practicing a delayed cannibalism that uses growth in its wireless business—albeit growth at a slower rate than at AT&T—to offset the decline in its landline business. And uses the time bought by that strategy to build a new fiber to the customer business called FIOS.

In the second quarter Verizon added 303,000 net FIOS Internet customers and 300,000 FIOS Television customers. As with AT&T’s iPhone customers, these are heavy data users who rack up bigger bills than the company’s average land line or wireless customer.  On average FIOS customers pay a monthly bill of $135, about twice the bill for the average Verizon customer.

If AT&T’s cost for being an iPhone cannibal is the huge $325 a phone subsidy to Apple, Verizon’s is the bill for building out the optical fiber network that will enable the company to get FIOS into homes and offices. The company is spending $23 billion this year and next to install fiber-optic cable. (One beneficiary of that build out is Jubak Picks 50 stock Corning (GLW) See my July 29 update.)

How does an investor pick between these two cannibals?

On current results, there’s not much that separates the two strategies. Wireless revenue at both companies (once you adjust for Verizon’s acquisition of Alltel) was about 9%.

In the intermediate run—say in an economy rebounding from the Great Recession—I’d give the edge to AT&T. The company’s network offers the best integration of land line and wireless services for business customers and I think AT&T is therefore more leveraged to a rebound in corporate business.

In that same intermediate period, however, much will depend on what happens when AT&T’s deal to be the exclusive seller of Apple’s iPhone ends in 2010. Will the two companies renew? What will be the terms? Will Apple add Verizon to make it a sell-a-trois?

In the longer term, however, I give a nod to Verizon’s strategy. The company is building a whole new heavy data, fiber to the user, Internet and video network on the bones of its current landline business.

In the long-run, in my opinion, the race goes to the bravest cannibal.