27 April 2018
Australian investors are bargain hunters at heart. They tend to focus on what looks cheap, with less regard for what they get in return for the price they pay. To most, a stock on a PE multiple of 12x is cheap and one on 24x must be expensive. If it were that simple, the game would be as simple as just spotting a low number. The reality is that it is not so simple.
By definition, the value of a company is calculated as the present day value of all the cash it generates over its lifetime. That is to say, it is the number that spits out at the end of a discounted cash flow or DCF valuation.
However, while a DCF valuation is fundamentally pure in calculating a stock’s intrinsic value, it is, like all valuation techniques, inherently imprecise. A DCF involves making forecasts of future cash flows and estimating an appropriate discount rate, and both steps require subjective judgement. More than other techniques, however, a DCF is a lengthy and quite detailed exercise that takes considerable time to undertake and then maintain. As well, the number the DCF spits out lacks transparency or much meaning on its own, other than for those with their heads in the excel model. Many investors will then choose to use some other simpler valuation technique.
By far the most popular is the PE multiple. It is quickly calculated, intuitive, and most can relate to it quite easily. However, investors should not forget that the friendly PE multiple is ultimately just a crude short-cut measure of value. It comes with a number of blind spots. Investors should be attentive to them, lest they buy into a ‘value trap’ or miss an opportunity hidden behind a high PE multiple that deters most other investors.
The PE multiple ignores the complexities involved in determining the “E”. Earnings are an accounting construct, and they can differ significantly from economic profit. As such, $1 of earnings at one company can be more or less valuable than a $1 of earnings at another.
CSL Limited (CSL) is the dominant player in the global plasma-derived medicines market. Its shares trade on 29x next year’s consensus earnings. Many would consider this too expensive, though we would counter that CSL is a very high quality company that is underestimated for the sustainability and strength of its earnings growth.
Relevantly, its growth is in part the result of heavy investment in research and development, which is aimed at finding new uses for its existing bio-medicines and discovering new ones. Next year CSL will spend approximately US$800 million on R&D. These costs are immediately expensed in the year incurred, notwithstanding that doing so creates a timing mismatch with the revenues generated in the years to come. The effect is to depress the current year’s earnings, by approximately $800 million down to the reported pre-tax earnings of US$2.4 billion. Were this R&D accounted for as a capital cost and expensed over time, earnings would be approximately one-third higher, and the PE multiple of 29x would reduce to a more palatable 22x.
Typically, an industrial company grows by investing in new plant and equipment. This investment usually adds to earnings immediately, but the cost is depreciated over a period of 20 years or even longer. The effect is to enhance earnings and the optics of its PE. In contrast, and as seen in the example of CSL, companies that grow by investing in R&D and other intangible assets are penalised in a PE comparison to the extent of any investment they make in growth. Whether an investment is capitalised on the balance sheet or expensed through the income statement does not affect a company’s economic profit. Nor does it affect a DCF, but the accounting anomaly can change the optics of the PE valuation.
Commonwealth Bank is an example of an Australian bound business that pays Australian tax on virtually all its earnings. The benefit of this is that franking credits attach to its earnings. These franking credits are of value to the company’s Australian resident shareholders, which are passed onto them via semi-annual dividend cheques, and cashed in when they use the credits to reduce tax payable, or in certain circumstances, receive a cash-back rebate in lieu.
A $1 of Commonwealth Bank’s earnings will therefore be more valuable to Australian residents than a $1 of earnings generated by an entity not paying Australian tax. Current examples of the latter include Qantas, Sydney Airports, Scentre and Spark Infrastructure.
The amount by which Commonwealth Bank’s earnings are more valuable is, loosely, the amount by which its earnings gross up for tax paid. Thus, $1 of Commonwealth Bank’s earnings equates in value to approximately $1.42 of Qantas’ earnings. For comparison purposes, the PE multiple must be adjusted accordingly. Interestingly, some of the companies referred to will begin to pay tax soon, including Qantas, which is currently enjoying a tax holiday on account of carry-forward tax losses. When they do, their earnings will reduce to the extent of tax paid, which itself will cause the PE multiple to increase.
An alternative way to undertake the comparison is to calculate the PE multiple of companies like Commonwealth Bank by using its pre-tax earnings as the denominator. This way, Australian investors can compare the value to them of Commonwealth Bank with Clydesdale, Wells Fargo, or indeed any other offshore bank.
The PE multiple takes its cue from just one year’s earnings - typically a forecast of next year’s earnings - and ignores what happens to earnings in subsequent years. Interestingly, one year’s earnings represent a relatively very small part of the value equation. Mathematically, if a stock trades on a PE multiple of 10x or more - which means most stocks - its earnings next year will ordinarily account for less than 10% of its total valuation. Over 90% of its value then derives from earnings in the years thereafter. What happens in these outer years is therefore of prime importance.
These outer years may entail material growth that belittles the earnings of the reference year. Looked at this way, a stock on a high PE multiple can actually be undervalued - in fact materially so - owing to strong compound growth.
CSL provides a classic example. For decades, its shares have traded on PE multiples many have considered too expensive, as is the case now. However, regardless of when its shares were bought over this time, its shares have gone on to deliver outsized shareholder returns. One concludes from this that the shares have been undervalued throughout. That includes when the shares traded on a PE multiple of almost 90x in late 2001, when investors became bullish on an industry consolidation story. Since hitting its peak PE, investors have achieved a compound annual return of approximately 14%, significantly outperforming the market and any reasonable estimate of investors’ required rate of return.
CSL floated in June 1994 at an IPO price of 77 cents (on a split-adjusted basis) and has since risen to around $160 per share. In simple terms, the rise in the share price over that time can be accounted for in two parts: firstly, the forecast PE multiple has risen by about 70% from 17x to 29x; and secondly, CSL’s EPS has risen from 4.5cps to A$5.38 per share, about 120x or 12,000% higher. With the benefit of hindsight, it is clear that the starting PE multiple was far less important than the company’s subsequent growth. This will naturally be the case the further one looks out for the purpose of assessing value. This is not to dismiss the importance of the PE multiple. It’s just the PE multiple must be considered in the context of what the “E” will look like over time.
In our view, the Australian market is overly short-term focused. The best evidence of this is its fixation of PEs and a corresponding focus on earnings just one year out. This provides other investors with an opportunity. Looking ahead of the purview of the market, to earnings subsequent to next year, will allow you to see what the market will soon enough be focused on. This time arbitrage manifests itself most valuably in stocks with long duration growth. With CSL as an example, the Australian market systemically underappreciates the value of compound growth.
Cyclical companies experience profitability that rises and falls with the economic cycle. A PE multiple, however, will be based off earnings for a given year, and will therefore ignore this cyclicality.
The mining companies provide a good example of the trap this can present. Seasoned investors will tell you the best way to play resources is to buy on a high PE and sell on a low one. This might seem perverse to most but the strategy has logic.
The miners’ PE multiples typically peak near the bottom of the earnings cycle. This is when commodities prices are at their lows, investment in the sector has dried up, and the supply-demand equation is starting to shift up to support higher prices. It is also usually the point when the miners’ share prices bottom, which in turn is the best time to buy them. Conversely, when commodity prices are at their highs and miners’ earnings hit a peak, their PE multiples tend toward their lows and share prices hit their highs. Here, miners are increasingly investing in new production, and the demand-supply equation is weakening.
As the graph below shows, and typical of most miners, the share price and PE multiple of BHP have been inversely correlated. Investors have been best off buying at high PEs and selling when they fall to their lows.
By focusing only on earnings, the PE multiple ignores the strength (or weakness) of the balance sheet. The balance sheet can be a source of material value that is undervalued, or goes unrecognised, in a PE analysis.
Using a PE multiple as one’s valuation guide actually penalises cash-rich companies. This is because a PE multiple effectively only gives value to earnings, and cash earns minimal interest income.
Flight Centre, the travel retailer, provides a good example. Its shares trade on 19x next year’s consensus earnings, although it must be said that this is after its shares have almost doubled over the last year.
As at its last balance date, Flight Centre held just over $1 billion in cash, another $203 million in investments - mostly liquid short-term fixed interest securities - and had just $91 million of debt. In effect, it had net cash of $1.122 billion (after allowing for its debt).
Now, $745 million of this amount was held on behalf of customers, who pay Flight Centre in advance of their trip. Such is the power of Flight Centre’s business model that it operates with negative working capital, meaning it collects payment from customers before being required to pay the airlines, hotels and other suppliers. Importantly, however, Flight Centre gets to keep the interest earned on this ‘free float’ of customer funds. This float is a valuable asset, which is currently undervalued on a PE analysis as it only captures the interest earned, which right now is very little.
Even allowing for customers’ funds, the company had a net cash position of $377 million (again, netted off for debt). Again, the PE multiple will only capture the value of the interest earned, which is approximately $8 million, or less than $6 million on an after-tax basis. Applying the current PE multiple of 19x to this interest income implies this $377 million cash pile is only worth $106 million. Or looked at another way, the correct PE multiple for the interest earnings is currently 67x.
Conceivably, Flight Centre could return excess cash to shareholders, via a dividend or buyback. It could even add to this by taking on a moderate amount of debt. Either way, there is unappreciated option value in the prospect of monetising its strong balance sheet.
As it is, the strong balance sheet affords Flight Centre the capacity to tough out difficult trading conditions and to be opportunistic in respect of any investment opportunities that arise. This provides defensiveness and optionality that contributes in an assessment of its quality, and which is often not reflected in a PE multiple.
Conversely, an over-geared company presents risks that might not be reflected even in a very low PE rating. Quite possibly, it may need to raise cash at inopportune prices that dilutes existing shareholder value, and turns what was previously a low PE multiple into a high one.
A PE multiple also ignores the value of non-earning assets. The classic example is of excess property that is valued on the balance sheet at its original cost but which could be sold for a lot more.
Sometimes, potentially valuable assets may even be loss making. In this case, their losses depress earnings, which in turn causes the PE multiple to appear higher. Relying just on the PE multiple will result in undervaluing the stock.
A good example is provided by Seek Limited, whose stock trades on a PE multiple of 28x next year’s earnings.
Seek is best known for its online employment classifieds business in Australia, Asia and elsewhere. But the company also has a division called Early Stage Ventures which invests in venture capital-like opportunities, all broadly in related sectors such as human resources management. These early stage ventures incur quite significant losses.
Currently, the Early Stage Ventures division is running losses at about $30 million per annum. These losses act to depress Seek’s group earnings, which for context, are approximate $350 million annually. This, in turn, has the effect of penalising Seek’s PE multiple, and hiding the potential upside value of these early stage ventures.
If these investments were made by a venture capital firm, they would show up on the balance sheet as an investment. If one were to apply Seek’s current PE multiple of 28x to this division it would generate a valuation of -$600 million. That’s a liability, not an asset.
The advice that follows is to be careful in taking a PE multiple at face value. It is useful as a crude measure of value, particularly for companies with a steady, predictable earnings trajectory. However, given the valuation analysis is so straightforward, it would be rare that such companies trade at out-of-whack valuations.
Increasingly, investors are being required to undertake more complex analysis to uncover value. After all, value is more than just a low PE. In fact, a low PE multiple can very often foretell earnings troubles ahead, the so-called ‘value trap’. And conversely, value can be found in high PE stocks. An investor who tries to cut short the valuation analysis by just focusing on the high PE multiple might be thrown off the scent and miss material pieces of value.
As Warren Buffett would say, price is what you pay and value is what you get. In the end, the PE multiple is just the price you pay per unit of a year’s earnings. The value you get depends on looking out farther, wider and deeper.
We have discussed above some stock examples demonstrating the blind spots in relying on the PE multiple. Fortunately, there are other valuation techniques that appropriately account for these blind spots. For example:
Ultimately, there is more to the friendly PE multiple than meets the eye. And therein lies the risk, and the opportunity.
Download a copy of the article here.
If you enjoyed this article, please share to help others find it. If you’d like the next article delivered to your inbox, select 'subscribe'.