Which earnings valuation tool works best?

Harry is a wily member of Intelligent Investor and is excited by the value emerging in blue chip companies. He recently enjoyed reading our special blue chip edition, which featured valuations of property developer Lend Lease and fund manager Perpetual.
Yet Harry scoffed at our valuations. He’s mistrustful of reported earnings, but particularly earnings before interest and tax (EBIT) and its loose cousin, EBIT before depreciation and amortisation (EBITDA). ‘Why can’t you just stick to using net profit and the more commonly used price to earnings ratio (PER)?’ If you aren’t familiar with PERs, we highly recommend reading Getting to grips with PERs.
It’s the first two letters of any ‘EB’ acronym – Earnings Before – that give Harry heart palpitations. He sides with Warren Buffett, who considers EBITDA a tool of fraudsters and crooks and once quipped, ‘does management think the tooth fairy pays for capital expenditures?’
Any time you see EBITDA used as a metric, treat it with skepticism. Even in the pages of Intelligent Investor Share Advisor. The only time you’ll ever see us use it is when it’s all we have (as was the case in our recent divisional analysis of Lend Lease titled What's Lend Lease worth?) or where we think our own estimates of depreciation and amortisation are better than the accountants’ (see Treasure MAp). The more letters that follow ‘B’ in terms like EBIT, EBITDA or EBITDAX (for resource companies), the further away from net profit they are.
On the other hand, PERs aren’t infallible, either. Apart from all the usual accounting shortcomings, they ignore debt. That’s alright if you’re trying to value fund manager Platinum Asset Management, for example, which is debt free. But if you’re analysing media business APN News & Media, which owed its creditors $936m at 30 June 2007, you might also consider using a valuation multiple that accounts for lenders claims on the business.

A useful solution is the enterprise value (EV) to EBIT multiple. To demonstrate, we’ll refer to the table as necessary. Please note that we’re using APN’s 2007 financials (the 2008 numbers were too messy to be of use).

Enterprise value

EV is the sum of a company’s market capitalisation (number of shares on issue multiplied by the share price) and its debt (or interest bearing liabilities), less any cash that’s surplus to requirements. EV represents the price you would have to pay to own the business outright, after paying off banks or other lenders.

We discussed this term at length in The low-down on enterprise value. Inasmuch as the EV relates to the ‘P’ in a PER, it incorporates the way a business’s assets are financed into the price (which the traditional measure of ‘P’, being market capitalisation, does not do).

Now, having adjusted the term equivalent to ‘P’, we also need to adjust the number that approximates ‘E’. So, having added debt into the numerator (EV), we also add interest and tax back to the denominator (EBIT). This allows us to compare apples with apples using an EV/EBIT ratio.
Where a PER compares the current market capitalisation of a company with the earnings available to the shareholders, the EV/EBIT multiple compares the total implied value of the business, including debt, to the total earnings available to all stakeholders in the business, including the tax office.
Now, having considered the difference between the two, let’s turn to the task of evaluating an EV/EBIT multiple.

If you’re interested in detail on pinpointing an appropriate multiple, our Value blog has an interesting conversation on the subject. But in this Investor’s College we simply want to provide a flavour for the topic. Essentially, we’re trying to determine a comparable EV/EBIT multiple for any given PER.

Lower your expectations

First, and most obviously, you need to use a significantly lower multiple for EV/EBIT multiple than you would for a PER. For a business with no debt or surplus cash, a PER of 10 is equivalent to an EV/EBIT multiple of 7 (the 30% difference compensates for the current corporate tax rate).
Platinum’s figures in the accompanying table illustrate this. Its EV/EBIT multiple of around two thirds of its PER accounts for the group’s tax bill and large cash surplus. The equivalent EV/EBIT ratio should always be lower than a PER to account for the tax office’s claim on the business. But for businesses with debt, the relationship is a little more complicated.

That’s because interest on the debt is usually tax deductible, meaning a company with some debt in its capital structure will pay less tax than a business funded purely by equity. Academics will tell you that this increases the overall valuation of a business with debt compared to one without, because there is less ‘leakage’ to the tax office.
For Australian investors owning Australian businesses, though, much of that advantage is clawed back through the franking credit system. That said, for a business with a prudent amount of debt (and there’s no hard rule on what that is), it’s possible to justify a slightly higher EV/EBIT ratio than would otherwise be the case.
This is all fiddling at the edges, though. The important thing to understand is that an appropriate EV/EBIT multiple should be approximately 30% lower than the equivalent PER you would pay for the same business.  

Chalk and cheese

Armed with this background information, let’s consider the two companies in the table. Platinum, as we’ve already seen, doesn’t have any debt. So it doesn’t make much difference whether you use a PER or an equivalent EV/EBIT ratio (which is 30% lower). For APN, though, the two ratios paint a vastly different picture.
Based on 2007 earnings per share of 34 cents, APN’s PER is 3.4. That looks attractive on the surface. But, due to the debt involved, the EV/EBIT ratio tells a truer story. It’s a substantially higher 5.1 times (the equivalent of a PER of 7.3).

Different perspective

From a different perspective, imagine if APN replaced all of its debt with new equity. That’s the adjustment we’ve made in column two of the table (and a less extreme variation on that theme may be forced on APN by its bankers at some stage). In this case, the EV/EBIT multiple doesn’t change. But let’s now consider the PER.

As interest is no longer being paid to bankers, net profit has risen by 16%. But despite this increase in ‘E’, the PER hasn’t fallen. In fact, it’s doubled due to a much higher ‘P’. Before replacing the debt with equity, the ‘P’ only revealed part of the price story - that part funded by shareholders. The new result produces a more realistic - and less appealing - valuation.
So, after all of this, what can we say to Harry? The answer is that regardless of whether we use the EV/EBIT multiple, PER, book value, or any other valuation method you can think of, choosing the best tool for the job is what counts. Often this will require a combination of methods. As is often the case in the financial world, a blend of practical experience, a study of financial history and a dollop of common sense should steer you in the right direction.