All too often, dividend investors base their investment decisions on yield alone, with little regard for the price they paid for the share.

As long as the dividend keeps rolling in, the thinking goes, I don’t have to worry about the share price.

Ignoring the capital is nice in theory, but difficult in practice. It’s rather challenging, for instance, to stay focused on a share’s 4% yield when the price has fallen 25%. And if you end up selling the share for a loss, that’s less capital that you can then reinvest elsewhere to generate the income you desire.

It’s a given that share prices are volatile, but I believe we can reduce the probability of permanent losses of capital by paying more attention to valuation when making buying and selling decisions.

Ideally, we want to purchase shares for less than their ‘intrinsic value’.

Put another way, we want to pick up £1 coins for less than £1. Investors might see that £1 coin as being worth 90p one day or £1.10 the next, but at the end of the day, it’s still worth £1. The market might make mistakes in the short-run, but in the longer-run it is a highly-efficient machine that more accurately prices a company’s true worth.

Valuation is an extremely broad topic and we won’t be able to cover it all here, but we will address the two major schools of valuation – relative and absolute valuation – and provide some methods you can put to practice straight away.

Relative valuation

By far the most common valuation method is relative valuation. With relative valuation, the investor compares the subject company’s multiples – price-to-earnings, price-to-book, et cetera – with those of comparable companies.

The underlying principle is that the market should price similar companies in the same industry the same way. If the subject company is trading with lower multiples than its peers, for instance, then it is said to be relatively undervalued.

To illustrate, an investor interested in grocery shares might put together a table like this:

Tesco Morrison Sainsbury Marks & Spencer
P/E 10.1 9.3 10.2 11.6
P/B 1.6 1.1 1.1 2.2
P/S 0.4 0.4 0.3 0.6
Yield 4.2% 4.4% 4.7% 4.6%

Note: The data in this table is for illustrative purposes only.

From such a comparison, we might deduce that the market is undervaluing Tesco1 and Wm. Morrison relative to Sainsbury and Marks & Spencer on a price-to-earnings basis.

You can probably see why most equity research reports and investment decisions are based on relative valuation methods – it’s quick, simple, and market-based.

But whilst relative valuation is helpful, it does have some drawbacks.

First, we need to understand what goes into each multiple. For example, the price-to-earnings ratio is a function of earnings growth and the required rate of return. All else being equal, a share with a lower price-to-earnings ratio might have lower growth expectations and/or higher risk than its comparables, and this could make its lower multiple justified.

Metric Is a function of… A higher metric a result of…
P/E Earnings growth, required rate of return Higher earnings growth expectations and/or lower risk
P/B Difference of return on equity and required rate of return Wider spread between ROE and required rate of return
P/S Profit margin, required rate of return Higher profit margins and/or lower risk
Yield Earnings growth, required rate of return Lower earnings growth expectations and/or higher risk

Note: Ceteris paribus!

Second, we need to identify the differences between the businesses we’re comparing. For example, Tesco has a bank division whilst Marks & Spencer does not and that could account for the different multiples.

Finally, it’s possible that the market is wrong and the group could be over- or under-valued.

To illustrate, in the late 1990s, a dot-com share trading with a price-to-sales ratio of 100 might have been relatively undervalued versus other dot-com shares, but that doesn’t mean it was actually undervalued.

Once the dot-com bubble burst, it didn’t matter much that the share was relatively undervalued, as all of the shares in the industry collapsed.

Absolute valuation

The alternative method of valuing shares is absolute valuation, which as its name suggests seeks to determine the absolute – or ‘intrinsic’ – value of a given share, regardless of how it might stack up relative to its peers at a given point.

The underlying principle with absolute valuation is that, if you’re correct in your outlook (a big ‘if’), the market will eventually correct itself and the share price will move closer to your fair value estimate.

There are a number of absolute valuation methods, the most popular of which are the dividend discount model (DDM) and the free cash flow to firm (FCFF) or to equity (FCFE) models.

In each case, the investor estimates future cash flows, and discounts the expected cash flows to the present using an appropriate rate to arrive at a fair value estimate.

The simplest absolute valuation model is the Gordon Growth Model:

Fair value = Next year’s dividend per share / (required rate of return on equity – dividend growth rate)

To illustrate, let’s say we’re looking at a share that analysts expect will pay a 100p per share dividend next year. Let’s also assume that the required rate of return is 10% and that we expect the dividend to grow at 2% in perpetuity.

Fair value = 100p / (10% – 2%)

Fair value = 1,250p

If this particular share is trading for less than 1,250p in the market, then, we might conclude that it is undervalued and worth buying.

As with relative valuation, there are a few drawbacks to absolute valuation to keep in mind.

First and foremost, you’re making forward-looking estimates — often very far into the future — and those estimates may not reflect what actually happens. If you over-estimate the company’s prospects, for instance, you’ll over-value the share and may end up overpaying. As such, it’s critical to assume a number of possible scenarios when using absolute valuation.

Second, the absolute valuation method you choose needs to reflect the type of business you’re analysing. A company that doesn’t pay a dividend (and doesn’t expect to), probably shouldn’t be valued using a dividend-based valuation model. (Though, in theory, it can be done).

Finally, absolute valuation methods can require substantial spreadsheet work and thus have a steep learning curve and typically require more time than relative valuation methods.

Which valuation method is best?

Most research reports that you’ll read will use either relative or absolute valuation, but the best approach may be to use both. The two methods have their benefits and drawbacks, but when used together they can help identify miscalculations and establish a better range of valuation possibilities.

To illustrate, you might run a dividend-discount model only to find that your fair value estimate doesn’t make sense on a realistic price-to-earnings basis. Similarly, you may find a share to be relatively undervalued on a price-to-earnings basis, but when you run a FCFF model you find that the share is actually overvalued.

Whichever methods you employ, it’s important that dividend investors consider valuation and not simply purchase a share based on its yield. In the end, you still want to purchase a share for less than it’s worth and reduce your risk of losing capital.

For further reading

Realising that valuation is a massive topic, I’ve provided links to three resources that I consider to be quite valuable and helpful if you decide to learn more about valuation.

  • Damodaran Online — The homepage of NYU finance professor, and arguably the world’s foremost valuation expert, Aswath Damodaran. Here you’ll find helpful data sets and valuation spreadsheet templates, as well as his finance class materials. Definitely one to bookmark!

Have a great 2013!

You can bookmark all The Analyst’s articles on dividend investing. The archive will be updated as new dividend articles are posted.

  1. Note: The Analyst owns shares in Tesco.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes:

<a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>