A deep dive into EV/EBITDA and P/E multiples

A deep dive into EV/EBITDA and P/E multiples

Author: Joris Kersten, MSc

Owner Kersten CF: M&A advisory + Valuations at The Netherlands. www.kerstencf.nl

Valuation training: 5 day valuation/ modelling training at Amsterdam (4th – 8th November 2024). www.joriskersten.nl

Source used: Morgan Stanley Investment Management, Counterpoint Global Insights, “Valuation Multiples: What they miss, Why they differ and the link to fundamentals”, April 23rd 2024, Michael J. Mauboussin & Dan Callahan.

?

Introduction

The basis for value creation lies in the return on invested capital (ROIC), growth and risk.

Companies create value when their investments earn a return that is higher than the opportunity cost of capital (e.g. weighted average cost of capital (WACC)).

Higher growth generates higher value for firms that earn a return (ROIC) above the WACC!

So higher growth leads to a lower value when the “spread” is negative!

ROIC/ WACC spread = ROIC -/- WACC.

And growth has no impact on value for a company that earns (ROIC) the WACC.

With multiples, they provide NO direct impact into the magnitude of a firm’s investments, or whether the assets make enough returns.

?

Multiples & Intangible assets

Another issue with multiples is that the major investments a company makes is on “intangible assets”.

For example “customer acquisition costs” and “branding”.

But companies often “expense” these investments in the income statement when they occur.

Accountants record these investments as “selling, general and administrative” (SG&A) expenses.

Or as “research & development” (R&D) expenses.

Both reduce the current earnings.

?

Matching principle

But in accounting there is the “matching principle”.

This means that a company should match its expenses to revenues.

The three main expenses subtracted from revenue, to arrive at EBIT, are:

1.????? COGS (cost of goods sold);

2.????? SG&A;

3.????? R&D.

Within this perspective, research is done (see source used for this blog) among listed firms, listed from the 1960s to 2010s.

It was found (again, see source used for this blog) that COGS match revenues effectively for all companies.

But while SG&A matches revenues well for the cohort listed from 1960 – 1990, it practically “un-matched to revenues” for firms listed from 1990 – 2010.

The issue is “expensing investments”, instead of “capitalising”.

Multiples are supposed to reflect the magnitude, and return, of an investment.

But the shift to “intangible assets”, and how companies record them in the financial statements (“expensing”), has made ROIC calculations problematic!!

The rise of intangible assets mean that:

·???????? Earnings are understated (investments in P&L);

·???????? Invested capital is understated (intangible assets not on balance sheet).

This weakens the signal that earnings & multiples provide, compared back to the old days when companies merely had capitalised tangible assets.

Partial conclusion:

Ideally accountants should match expenses to revenues well!

When this is not done for intangible assets, the CF analyst needs to clean NOPAT and invested capital.

In a later blog I will show how this is done.

?

P/E and EV/EBITDA multiples

P/E and EV/EBITDA multiples are the most popular multiples equity investors use.

But what do you do when two companies:

·???????? have a similar P/E but a different EV/EBITDA multiple;

·???????? have a dissimilar P/E but same EV/EBITDA multiple.

It is logical that the correlation between P/E and EV/EBITDA multiples is high.

But there are cases where two companies are close on one metric, and different on the other metric.

The P/E captures the relationship between the company’s “equity market capitalization”, and its earnings (or net income).

P and E are often expressed “per share”.

The P/E is a “leveraged ratio” because it is measured after interest.

The EV/EBITDA multiple compares the firm’s EV (enterprise value) to EBITDA.

EV = equity market cap + debt and other liabilities -/- cash.

And this multiple is “unlevered”.

P/E is almost always higher than EV/EBITDA for a profitable company.

Reason is as follows:

·???????? Imagine a company with no debt and no excess cash;

·???????? Here P equals EV;

·???????? And E will be lower than EBITDA;

·???????? Because of tax, and non cash items are not added back;

·???????? This results in higher P/E than EV/EBITDA.

This does not hold for a company with negative income.

?

Differences between P/E and EV/EBITDA

The differences between P/E and EV/EBITDA multiples come from:

1.????? The mix of tangible and intangible investments;

2.????? Differences in the capital structure;

3.????? Non operating expenses;

4.????? Tax rates.

First, as we have seen, COGS tend to match revenues for most companies.

COGS are the majority of total expenses for “tangible asset intensive companies”.

But SG&A is much more prominent for “intangible asset intensive companies”.

So actually, in order to get a consistent picture, intangible assets should be capitalised & amortised.

The result:

·???????? Earnings and EBITDA will increase very little for “tangible asset intensive companies”;

·???????? Earnings and EBITDA will increase a lot for “intangible asset intensive companies”.

So multiples have lost relevance because of the widening gap between earnings!

But it is good to note here that:

·???????? Capitalisation of intangible investments does not affect “free cash flow”.

Partial conclusion:

Keep using “free cash flows” for valuation alongside multiples. So do not forget DCF valuation + LBO analysis since they both use some sort of “free cash flow”.

?

Capital structure

The second reason why P/E and EV/EBITDA multiples can differ is “capital structure”.

The impact of changes in the capital structure on multiples is complex.

Imagine a company that is purely financed with equity and has no excess cash.

Now introduce debt.

The P of P/E will be unaffected since it reflects the value of equity only.

But E will go down because of interest.

So same P, and lower E, will increase the multiple, all else equal.

Concerning EV, adding debt increases EV, since it is equity + debt.

But is has no effect on EBITDA, since EBITDA lies above interest and taxes.

So here also EV/EBITDA will go up, all else equal.

Now our CF theory suggests that an optimal capital structure is reached, when the tax shield is maximised, taking the costs of financial distress into account.

And the pecking order theory of capital structure says on financing growth:

1.????? Start with cash generated internally;

2.????? Then go to debt;

3.????? And finally go to equity.

Companies with high ROIC often have sufficient cash flow from operations to finance their growth.

Companies with low ROIC often have insufficient cash flow to fund the business, and they need to use debt or equity.

As a result:

·???????? Industries with high ROIC tend to have lower debt to capital ratios;

·???????? Industries with low ROIC tend to have higher debt to capital ratios.

In addition:

·???????? Companies with low ROIC tend to have D&A (depreciation & amortisation) that is a higher percentage of EBITDA, than companies with high ROIC.

This implies that the gap between P/E and EV/ EBITDA multiples is generally higher for low ROIC companies with lots of debt.

And this implies that the gap between P/E and EV/EBITDA multiples is generally lower for high ROIC companies with little debt.

?

Non operating expenses & tax rate

P/E and EV/EBITDA multiples can also differ as a result of non operating expenses from:

·???????? Restructuring programs;

·???????? Asset write offs;

·???????? Reorganisations;

·???????? Unrealised capital gains/ losses.

And at last, they can differ due to the tax rate.

?

EBITDA and the Depreciation Factor (DF)

When pricing equity with EV/EBITDA, you should know how much the D&A from EBITDA is.

D&A is an estimate for maintenance CAPEX.

So for two firms with the same EBITDA, the one with the higher EBIT will have more cash flow to distribute to the shareholders, and has a higher value.

The ratio of EBITDA over EBIT is called the EBITDA “depreciation factor” (DF).

For example, the EBITDA DF is as follows:

·???????? Sector consumer discretionary = 1.3 (100% EBITDA/ 75% EBIT);

·???????? Sector utilities = 1.8 (100% EBITDA/ 56% EBIT);

·???????? Overall = 1.4 (100% EBITDA/ 70% EBIT).

And the amount of D&A is defined by:

·???????? Capital intensity;

·???????? Assumed asset lives;

·???????? Level of acquisitiveness.

EBITDA DFs provide insight on the ROIC/ WACC spread:

·???????? Companies with low EBITDA DFs commonly have positive ROIC/ WACC spreads (e.g. EBITDA DF under 2.0);

·???????? Companies with high EBITDA DFs commonly have negative ROIC/ WACC spreads (e.g. EBITDA DF over 2.0).

This means that companies with low EBITDA DFs will have higher multiples, than companies with high EBITDA DFs, for a given rate of growth.

?

Amortisation

The amortisation of (acquired) intangibles has gone up from about 2% to 20% from D&A in the last 40 years.

Amortisation is (largely) the result of a company buying another company at a premium to the (tangible) book value.

This because the buyer revises the balance sheet from the seller, to reflect its tangible assets and intangible assets.

This is called purchase price allocation (PPA).

Intangible assets that arise from contractual rights, or other legal rights, that can be separated from a company must be amortised over its useful life.

This results in the amortisation of intangibles.

Then the rest of the “intangibles” that do not meet the right criteria are labelled as “goodwill” on the balance sheet.

Companies do NOT amortise goodwill, this must only be checked periodically to see whether the goodwill needs to be “impaired”.

?

Amortisation from 2002

Since 2001 there is a change in accounting rules for business combinations.

Until 2001 companies could use one of two accounting methods:

·???????? The pooling of interest method;

·???????? The purchase method.

The pooling of interest method basically allowed companies to combine their balance sheets, so there was no need to record intangible assets.

The purchase method required companies to show the goodwill on the balance sheet, and to amortise it over a period up to 40 years.

But those options changed from 2002, and the FASB (financial accounting standards board) changed the rules so that the pooling of interest method could not be used anymore.

Also goodwill did not need to be amortised anymore.

From 2007 FASB introduced another change, which clarified what companies could see as “intangible assets” and what as “goodwill” (purchase price allocation).

As a result, intangible assets increased relatively to goodwill on balance sheets.

?

Conclusion

Summarised, it is great to look at multiples like P/E and EV/EBITDA.

But do not get lazy ! And keep focussing on the fundamentals as well !

The fundamentals are:

·???????? Cleaned NOPAT;

·???????? Cleaned invested capital;

·???????? Cleaned ROIC;

·???????? WACC/ risk;

·???????? Growth;

·???????? Free cash flow.

And realise that multiples (can) fail to consider:

·???????? Investment needs;

·???????? Correct matching of costs & revenues (expensing instead of capitalising intangibles);

·???????? Gaps between P/E and EBITDA multiples.

Hope this was useful.

See you next week again with a new article,

Best regards, Joris????

?

Source used: Morgan Stanley Investment Management, Counterpoint Global Insights, “Valuation Multiples: What they miss, Why they differ and the link to fundamentals”, April 23rd 2024, Michael J. Mauboussin & Dan Callahan.