Update No. 205 – 06/03/15

I mentioned last week that I would go over my two favourite investment metrics this week.

It is one of the biggest mistakes inexperienced investors make to overestimate the usefulness simple metrics. There is no single, simple all-purpose metric that can be applied. Each application of a metric must be countered by critical thinking about what in the metric may be flawed.

Price to Earnings (P/E) is the most commonly used metric. To be sure, it is popular because it is both simple and useful. There have been plenty of studies that set out how buying a ‘basket’ of low P/E stocks has pretty consistently outperformed the broader market over time.  The famous “What has worked in Investing” (.pdf) paper by Tweedy Browne sets out (in table 15 on page 16) a pretty clear advantage to low P/E stocks.

There are flaws in using P/E (or many other metrics) in isolation. In the case of P/E, the first key danger is that the numbers that makes up the ‘E’ could easily be flawed. There could be ‘one-off’ gains (or losses) that make the ‘E’ figure unreliable, leading to artificially high (or low) numbers. So while buying a basket may work with diversification spreading risk, using P/E as a determinant for the purchase of an individual stock must be done with considerable caution. The ‘one-year of unreliable earnings’ risk can be mitigated somewhat by using the CAPE, which uses a 10-year rolling average earnings. Ben Graham also set out the importance of looking at earnings over a full cycle.

Another danger in using P/E is that it ignores the capital structure. Imagine two companies with $100m market capitalisations each. Each earns $10m post-tax. They each therefore have a P/E of 10. Imagine however, that Company A has $50m of net-cash on their balance sheet, but Company B has $50m of net-debt. Without examining this fact, an inexperienced investor may wrongly be indifferent between the two as investments. Assuming their future earnings prospects are similar, Company A is a meaningfully better option.

This is why I prefer to use what I refer to as Price to Enterprise Value (or the ‘ungeared’ P/E ratio). It’s really enterprise value to earnings, which I will express hereon as EV/E. In this situation, Company A above would have an EV/E multiple of 5x after subtracting the cash and Company B would have an EV/E of 15x after adding the debt into EV. It tells a very different story of two $100m capitalisation companies that are earning $10m, but it is much more accurate. For cash and debt are genuine elements of the capital base that must be properly considered. Unencumbered cash is available to be returned at any time to shareholders or to be used to grow earnings per share somehow (buybacks acquisitions etc.). Debt usually has a known due date, but it is generally best to behave as though it could be called at any time. The EV/E simplifies this (probably oversimplifies it should be noted as debt can be prudently used in some circumstances to augment returns) by considering equity and debt interchangeably.

The trick then is what multiple to apply to a given company as a ‘fair value’. As with P/E, it will depend very much on the industry and the competitive environments that the company operates in. Some businesses that are very capital light, with negative working capital and favourable industry dynamics could easily command a pretty hefty EV/E multiple. Some businesses that need to make huge investments in working capital, and PPE should correctly command a fairly low multiple.

We will now think about the ASX200 in terms of EV/E. As I quill this missive, the ASX200 trades at about 5880 points. After the completion of earnings season, I understand the trailing 12 months earnings of the index is about $361, which give the market a P/E of about 16.3x at present. I should point out that for the first time in a few years I will declare is definitely too high (I have said roughly fully valued in the past) based on historic levels and the current condition of the Australian economy (where the majority of the index’ profits are earned), but in light of 2% central bank rates, it is not entirely unexpected that it trade above historic levels. Purchasing the ASX200 at this level is likelier than not to leads to below historic returns over the next 5 to 10 years in my view. Better than average stock selection is a must.

Working out the gearing level is trickier. This BRW article back in 2012 pinpointed the ASX200 levels of gearing as having fallen to nearly 20%:

Excluding banks and real estate investment trusts, the top 200 stocks covered by Goldman have gearing ratios approaching 20 per cent – the lowest level in 30 years. Heading into the financial crisis, average gearing levels were close to 50 per cent.

I think you really have to include banks & REIT’s as they genuinely comprise part of the ASX200’s composition if you own the index. In any case, in the absence of better research on the subject, I would suggest the gearing including these elements would not presently be materially different from perhaps 25%. Again, with the ASX200 at 5880 points, 25% leads us to estimate the ASX200 debt load as $1,470. The EV/E under this scenario is about 20.4x, using the earnings of $361 and the enterprise value of 7350.

Your job as an investor now is to find companies that have growth prospects at least as good as or better than the ASX200, with an EV/E that is lower. Do that and over time, your investments will perform better than the index…With index levels where they are, you will need it.

I also mentioned Price to Free Cashflow (P/FCF) last week. The post has gotten long so I will be brief – it is another favourite of mine because it helps us to eliminate or normalise for capital intensive businesses. The trick with FCF is determining what elements of the Capex need to be deducted from operating cashflow. It is among my great bugbears as a stock analyst that most businesses do not at least make some effort to clearly set out in their reporting what of their Capex is ‘maintenance’ related and what is ‘growth’ Capex.

Ordinarily FCF purists just deduct all Capex, but there are times when this will cause you to overlook a highly cash-generative business. Vocus (VOC) for example, which EGP doesn’t/hasn’t owned (unfortunately) has a both a depreciation schedule for their fibre assets which serves to disguise/moderate the NPAT profits (they depreciate the fibre at least 3x faster than the assets likely useful life) and a heavy Capex budget that serves to make FCF appear to be negligible (or negative…). Fortunately for potential investors in VOC, they have a CEO who sets out the Capex maintenance/growth story better than most. For the majority of industries, the depreciation charge serves as a pretty good proxy for what the maintenance Capex is likely to be, but the alert analyst needs to be watchful for industries/situations such as set out above where you may be able to finagle an edge – Tony Hansen 06/03/2015

EGP Fund No. 1 Pty Ltd. Up by 4.26%, trailing the benchmark by 8.56% since July 1^st 2014. Since inception, EGP Fund No. 1 Pty Ltd is Up by 77.62%, leading the benchmark by 32.00% all-time (April 1^st 2011).

^{*1 after 31May 2013 dividend of 2.333 cents per share plus 1.000 cent per share Franking Credit & 31 May 2014 Dividend of 7.000 cents per share plus 3.000 cent per share Franking Credit}

^{*2 calculated based on dividends reinvested}