AT&T (T 0.76%) has been the most aggressive of the three major wireless carriers with device promotions this year. The release of Apple's iPhone 12 prompted it to offer its existing customers the same promotions as new customers, something T-Mobile (TMUS 0.62%) and Verizon (VZ 3.19%) haven't done.

But if you ask AT&T's management whether it was a mistake to make such an aggressive offer while the competition may have been more conservative than expected, executives will say absolutely not. Instead, they see it as an opportunity to lock in existing high-value customers long-term.

Let's examine that logic and what AT&T isn't saying.

A line of people all staring at their smartphones.

Image source: Getty Images.

Keeping churn low and improving lifetime value

The biggest reason subscribers leave AT&T for a competitor is "because they could go get a new device somewhere else," CEO John Stankey said at an investor conference last month. "It wasn't because they didn't feel like they could go get good service."

As such, investors should expect the AT&T offer to produce lower postpaid phone subscriber churn than Verizon or T-Mobile when it reports fourth-quarter results. That will be the first measure of whether the strategy works.

CFO John Stephens pointed out several other benefits of the offer at another investor conference earlier this month. "When you think about the fact that customers trade in a phone, they sign -- they stay on for another 30 months. These are customers that we know their history of payments, their history of churn," he said. In other words, AT&T can mitigate the cost of the new devices with the trade-in requirement while generating better-than-average expected customer lifetime value from the customers it approves for the offer.

"On top of that, we get the opportunity to sell them up on whether it's adding HBO Max, whether it's going through the Unlimited Elite programs," Stephens added. If AT&T can upsell its services to existing customers, it will see an even faster payback period. Bringing them into its (virtual) store is an excellent opportunity to do that.

So the second factor investors should be looking at is average revenue per subscriber. Numbers like HBO Max activations may be muddled, as most legacy subscribers still haven't switched, but looking at total HBO subscribers in conjunction with wireless average revenue per user (ARPU) could be a better indicator to see if AT&T's successfully upselling its high-end bundled wireless offering.

It looks more like desperation than sound strategy

T-Mobile CEO Mike Sievert wasn't afraid to call out AT&T's desperation at an investor conference last month. "If I was AT&T and thought I was about to lose all my customers, I'd do something extraordinary, too," he said. Indeed, making an offer to keep subscriber churn low in a period of extremely low industry churn to begin with is far from a sign of strength. 

What's more, it may indicate concern from AT&T that it won't be able to keep up in developing its 5G network. AT&T's spectrum position leaves much to be desired, particularly in mid-band spectrum.

T-Mobile's building its ultra-capacity 5G network on the back of its 2.5 GHz spectrum acquired in its merger with Sprint. The network offers speeds up to 300 Mbps, or roughly seven times faster than its LTE network. As of the end of 2020, it already covered markets with a combined population of 106 million.

The FCC is currently auctioning off similar spectrum, but AT&T's facing fierce bidding from rivals like Verizon, which are capable of outspending it. With bidding now surpassing $80 billion total, it's likely Verizon's securing more of the highly valued A block licenses, which it will be able to use this year. AT&T's likely settling for BC block licenses in most markets, which won't clear until 2023.

All this is to say, AT&T's probably desperate to lock in customers today to buy time until it can show progress on its 5G network. And the timeline for that is looking longer and longer. It's really a short-term fix for a much larger problem it faces.