Stock valuation of Telecom companies through PBR regression analysis
Value investors have traditionally chosen stocks with low PBR. It is said that Benjamin Graham, author of “The Intelligent Investor” considered stocks with PBR of less than 2/3 as one of the investment criteria.
One of the ways to use PBR is to compare a target company’s PBR to the average PBR of the peer group when determining whether the target company’s stock is undervalued or overvalued. However, it has a problem. The criteria for selecting comparable companies may be subjective. And even companies in the same industry may have different business composition, risk, and growth rate. In that case, you can evaluate the stocks after adjusting for differences through regression analysis.
Case: Telecom companies listed on the US and UK stock markets
The six companies in the screen above are large telecom companies listed in the US and UK. The market cap ranges $12 billion to $250 billion with the PBR of 0.55 to 3.91. When considering PBR alone, Vodafone is most undervalued, and Verizon is most overvalued. However, when you consider fundamental factors together, it is different.
In this case, it assumes that two fundamental factors affect the PBR of the telecom sector, ROE and growth rate. The growth rate refers to Thomson Reuters’s estimation over the next five years.
Considering the two fundamental factors, you can approach each company’s PBR more analytically.
It seems that Vodafone’s PBR is low since its ROE is negative, and Verizon’s PBR is high because its ROE is high. In the case of T-Mobile, ROE is relatively low, 5.25%, but the growth rate is high. That is why the PBR seems to have increased. Although the number of samples above is small, it exercised a regression analysis based on them by placing ROE and growth rate as independent variables. Then, the following equation came out.
PBR = 0.12 + 9.73 ROE + 6.85 g
(R2 = 82%)
The R2 of 82% means that ROE and growth rate account for 82% of the PBR. It is relatively high, so the result appears reliable to some degree. It calculated the theoretical PBR by the regression equation above.
The gap between the expected PBR and the actual PBR is residuals. A negative residual means that the actual PBR is lower, which means the stock is undervalued.
Based on that, BT was most undervalued at -0.95, and Vodafone and Verizon were slightly overvalued at 0.50 and 0.44, respectively. However, in the case of Vodafone, the actual PBR will not decrease to 0.05, like the result of the regression equation. Also, since the above regression does not include all factors affecting the PBR, you’d better not give a big meaning to the small difference between the expected and the actual PBR.
Nevertheless, in the above case, the residual of BT is quite large, -0.95, so it is worth analysing more to see whether it is a signal to buy. The implications of the residuals will become even more critical as R2 increases by adding other independent variables that you believe to affect the PBR as well as ROE and growth rate, or by increasing the number of samples.
- (Excel Financial Model) PBR valuation via a regression analysis of ROE and growth variables
- (Blog) PBR and Evaluating the Stock Price of Tesla and Toyota