Why use multiples?
There is no doubt that discounted cash flow technique is the
fundamental valuation measure. But multiples are a useful tool in an investor’s
tool kit. They help you:
-
cross check your DCF forecast by comparing it to
market’s forecast (e.g. by backing out implicit growth rate forecast by the
market based on the multiple at which the company currently trades);
-
compare between peers – multiples are a great
way to compare peers and get a feel which companies the market believes have
better prospects & which ones the market things are likely to suffer); and,
-
are a great way to search for bargains. I use
multiples as a useful starting point for bargain hunting. However, it is very
important to recognise that a high multiple does not necessarily mean
overvalued stock; and a low multiple does not mean there is a definite bargain
to be had. There can be real economic reasons for why a stock trades at a low
or a high multiple, and searching for those reasons are what lead to investment
decisions.
Which ones?
There are three widely used multiples for investors:
-
Price to Earnings (P/E)
-
Enterprise Value / Earnings before interest,
tax, depreciation and amortisation (EV / EBITDA)
-
Enterprise Value / Earnings before interest and
tax (EV / EBIT)
P/E is a widely used measure in stock valuation and there
are many research reports supporting low P/E as a good measure of value. On
average, low P/E stocks have returned more in the long term than high P/E
stocks. However, I don’t like to make buy decisions simply based on low P/E; I
only use low P/E as a good starting point for further investigating a stock. P/E
has its drawbacks, the main ones being:
-
Firstly, P/E is distorted by GAAP &
non-operating items such as one off gains & losses, write-offs,
amortisations etc. This can cause significant distortions and be difficult to
wrap your head around if you want to make the adjustments to get to a
comparable P/E;
-
Secondly, P/E distorts the picture by mixing the
capital structure with the operating performance. For example, two similar
business will show very different P/E ratios if one is highly geared and the
other has nil or negligible debt. It’s better to use a multiple which is a good
gauge of operating performance and is not distorted by the capital structure.
In spite of P/E’s drawbacks, I would suggest using it in
screening for value stocks, but don’t buy simply based on low P/Es as it is
likely that low P/E is for a reason, and that reason could be a blow up coming
soon.
EV/EBITDA is a widely used multiple by practitioners. In my
experience this multiple is used commonly in the private equity and M&A
world – for example, it is common to see new releases by buying & selling
companies where they talk about the EV/EBITA multiple at which a business is being
bought or sold. It is also common to see
a sector or industry EV/EBITDA multiple. However, a major drawback of this
measure is that it ignores depreciation – which is the accounting equivalent of
capex required in the future to sustain the business. If a true measure of
value is the free cash flow, it is dangerous to ignore depreciation as it will
come out of cash flows at some point. Followers of Buffett will note that he
hates this measure for the same reason. However, this can still be a useful
measure in industries with very little maintenance capex requirements (e.g. some
new technology or social media companies).
EV/EBIT solves for the depreciation issue inherent in
EV/EBITDA by accounting for accounting depreciation. However, EV/EBIT too has
its drawbacks:
-
Firstly, EBIT accounts for amortisation expense
which tends to be a accounting only measure – e.g. if a company has made
acquisitions at a premium and recognised intangibles to account for the
premium, it would amortise the intangible. This will distort its multiple
compared to a similar business which has grown organically (organically generated
intangibles don’t need to be recognised or amortised for accounting purposes).
-
Secondly, two companies may have very different
depreciation based on the market value they paid for their assets being
depreciated.
-
Thirdly, there may be a big difference between
accounting depreciation and the actual cash required for maintenance capex.
In summary, the best multiple to use depends on the specific
circumstances of the company being investigated. I would personally start off
with EBIT and make necessary adjustments as follows:
-
Compare accounting depreciation to real cash
required for maintenance capex and use the latter if it significantly differs
to the accounting depreciation – i.e., EBIT + accounting depreciation –
maintenance capex. The cash flow statement is usually a good place to check for
capex; try to estimate the average capex to sales ratio over the years and use
this to predict capex needs based on your forecast sales.
-
Secondly, I would also add back amortisation expense
to EBIT for my multiple measure.
The Mckinsey book on Valuation has a very good chapter on how
to use multiples to triangulate results. It recommends using a EV/EBITA measure
which his similar to the above.
What does it mean?
It is important to never forget the fact that the value of
an asset is equal to the present value of its future cash flows. A multiple
based valuation measure is only of use if it links in with the fundamental
valuation measure based on cash flows. In order to make sense of multiples, I
love the explanation in the Mckinsey book on valuation and I have had good
success in being able to use this method in practice to make sense of multiples
between peers which is rare for most theoretical measures.
If you accept that free cash flow is the fundamental driver
of value, then value can be expressed by the following formula:
NOPLAT is basically your estimate of free cash
flow (e.g. EBITA). Using your estimated EBITA, the fundamental value driver
formula translates to:
Note that T is the effective tax rate, which most companies
will report. The (g/ROIC) determines the amount the company needs to keep
re-investing in order to grow at the projected rate. ROIC can be estimated based
on growth in sales over growth in capital employed, alternatively you could
simply use the return on capital employed achieved by the business; as for
growth rate, this should be the same as the forecast you have used for your
DCF. WACC too should be your estimate for the cost of capital – cost of debt is
easy to calculate, whereas cost of equity will depend on your assumption for
the premium you demand as an investor.Now the next step takes you to the key formula which relates
the multiple to the free cash flow formula. Simply divide both sides by EBITA:
The above is what a multiple tells you about value. In
particular, the multiple is a function of the following variables – tax rate,
growth rate, Return on Invested Capital achieved by the business, and its cost
of capital. Always remember this when using multiples.
Variables
|
Impact on multiple
|
Tax rate
|
Higher (lower) the tax rate, lower (higher) the multiple
|
Growth rate
|
Higher (lower) the growth rate, higher (lower) the multiple
|
ROIC
|
Higher (lower) the ROIC, higher (lower) the multiple. However, is
very important to note that if ROIC is < WACC then higher growth actually
wipes out value. ROIC always must be greater than WACC for growth to add
value
|
WACC
|
Higher (lower) the WACC, lower (higher) the multiple
|
As you will know, the above variables are key in calculating
a DCF value. And by knowing how they link in with a multiple based value, it is
possible to test your DCF assumptions with what the market is saying, and also
a useful measure of comparing between peers.
Other important points to bear in mind when using multiples
Enterprise Value
When using Enterprise Value based multiples, it is important
to calculate the Enterprise Value correctly. Note that Enterprise Value is the
price a buyer of the company will need to pay when taking over the whole thing.
Therefore, it not only includes the value of equity (market capitalisation),
but also the debt, minority interest, preferred shares, pensions liabilities, and
any other fixed obligations (e.g. capitalised leases), and from this you will need
to deduct cash & cash equivalents and the value of any non-operating assets
(e.g. investments not used in the operations) held by the company. Although
most online tools give you a number for the Enterprise Value, I would highly
recommend that you work through the company’s balance sheet and detailed notes
in arriving at this number. There can be a lot of useful details on fixed
obligations hidden in the notes which some of the freely available tools
ignore. Broadly, use the following as a guide to calculate Enterprise Value:
Enterprise Value
= Market Cap + Debt (both short-term & long term) + minority interest +
preferred shares + lease obligations + any pensions deficit + other fixed
obligations (e.g. litigation expenditures) – cash & cash equivalents – non-operating
assets or other long-term investments
Using the right measure
When using multiples for valuation ensure that you are using
the denominator consistently across comparables – e.g. if you are using EBIT
and adjusting it for amortisation and maintenance capex, ensure you do the same
across comparables.
Using the right peers
If using multiples as a relative value measure, it is important
to use the right peers (i.e., companies within the same sector and with similar
operational characteristics).
In conclusion - If you use multiples correctly, you will be able to see the
differences between peers, reasons for them, and be able to compare your
assumptions with those of the market. It is a great way to identify mispricing
and form a thesis. The other opportunity is where you have a conglomerate with discrete
businesses where you can compare the component parts with listed peers and
identify value add if/where that component part is being spun-off by the conglomerate.
