Our friend Doug McIntyre at 24/7 Wall St. is great at spotting industry trends that affect stock prices. He discusses one such trend (broadband strategies at telcos) in his post AT&T: New High Every Week:
AT&T trades at 2.24 times sales. BellSouth (BLS) trades at 3.64 times. AT&T (T) must have paid a premium. But, Verizon (VZ) trades at 1.22 times. And, Qwest (Q) at 1.2 times. Qwest, being the smallest of the lot does not have the balance sheet or revenue to compete with cable, WiMax, and VoIP as well as the rest. Or, at least that is the conventional wisdom about why it carries no premium….
Verizon says it will spend $20 billion building its fiber system according to the company’s public statements. It also says that the service is now available to six million homes, about 20% of their customers. Of course, that does not mean that the customers are using it.
AT&T and BellSouth have not announced intentions anywhere near as aggressive for upping the ante to get faster wires into their customer’s homes. And, that may be why Verizon trades at a discount.
However, Verizon’s significant discount only applies on the basis of price/sales. On a trailing P/E basis it is actually cheaper, and on the basis of price/book it is just a smidgeon more expensive than AT&T. As it turns out, accounting arcana are largely responsible for the price/sales differential. Specifically, the method each company uses to account for its majority (60% for AT&T and 55% for Verizon) stake in its wireless operations. Let’s have a look.
On page 24 of their 2005 annual report (.pdf), Verizon says:
Our Domestic Wireless segment provides wireless voice and data services and equipment sales across the United States. This segment primarily represents the operations of the Verizon Wireless joint venture with Vodafone (VOD). Verizon owns a 55% interest in the joint venture and Vodafone owns the remaining 45%. All financial results included in the tables below reflect the consolidated results of Verizon Wireless.
The consolidated wireless results (100%) were revenue (in $millions) of $32,301. Of that, Vodafone’s 45% share amounts to $14,535. Were this excluded from Verizon’s total revenue of $75,112 the adjusted revenue for Verizon would have been $60,577. Instead of 1.37x sales (using 2005 revenues and the current share price), Verizon’s market cap would represent 1.70x sales.
Turning to AT&T things get a bit murky. To start with, AT&T (the long distance company) is only included in the merged company’s 2005 results for 45 days. Fortunately they do provide pro-forma data as though it had been included for the full year. A second issue is their method for reporting their wireless operations. We present their comments on their wireless division (p. 28 of their annual report):
We account for our 60% economic interest in Cingular under the equity method of accounting in our consolidated financial statements since we share control equally (i.e., 50/50) with our 40% economic partner BellSouth in the joint venture. We have equal voting rights and representation on the Board of Directors that controls Cingular. This means hat our consolidated results include Cingular’s results in the “Equity in net income of affiliates” line.
In layman’s terms, this means that they don’t record any of Cingular’s sales (but they do record 60% of its net income.) Here’s how this is described in Financial Statement Analysis: A Global Perspective (pp. 561-562):
There are three approaches to accounting for (investments over which the company has significant influence): the equity method, proportionate consolidation, and consolidation.
Each of these three methods yields the same amount of net income for the period. Likewise, each yields the same amount of net assets (or equity) for the period. However, the methods yield different cash flows for the period. Additionally, the details may be different. Consequently, key financial ratios differ under the various methods.
Specifically, under the equity method AT&T doesn’t record any revenue, and records its share of Cingular’s profits as a line (Equity in net income from affiliate) on the income statement. Meanwhile, Verizon records all of Verizon Wireless revenue and profits, and deducts Vodafone’s share as a line (Minority interest) on its income statement.
So what adjustments need to be made for AT&T? First we need to start with the pro-forma income statement, which treats the long-distance business as though it had been included for the full year. This gives us a revenue line of $66,061 rather than the $43,862 reported in 2005. This adjustment alone reduces AT&T’s price/sales (apples to apples for the way we calculated Verizon’s) to 1.86x.
Now we need to add in the proportionate (60%) share of Cingular’s revenue of $34,433, or $20,660. Now our total proportionate revenue for AT&T is $86,721. After this adjustment, AT&T’s price/sales is 1.41x, lower than the apples-to-apples 1.70x for Verizon.
Since both methods result in the same net income and equity, neither the P/E multiple nor price/book multiples are affected by the accounting choice. So we can accept at face value (at least with regard to their accounting for the wireless divisions) the Yahoo! Finance TTM P/E of 18.81x and P/B of 2.22x for AT&T, and the 14.98x P/E and 2.30x P/B for Verizon. Again the multiples end up being fairly close when the accounting treatment is apples to apples. Verizon is more expensive on price/sales and price/book but cheaper on P/E.
This is one of the cases where the market has largely picked up on the accounting difference. This doesn’t mean that investors understand the accounting intricacies, but more reflects the fact that investors more frequently use P/B or P/E rather than P/S. In fact, if Doug is correct in his assessment that Verizon has chosen the wrong strategy, this could be a great investment opportunity since Verizon is trading at about par with its more conservative peer.
So you can see why we like to dig into the numbers - sometimes you dig up a nugget that is pure gold.
T 1-yr comparison chart with PE ratio:
VZ 1-yr comparison chart with PE ratio: