Stock analysts and market commentators are a very imaginative lot. They can always be counted on to find ways to justify some predetermined conclusions or narratives. This is partly why we see “new” valuation metrics all the time when the old ones do not work to fit the story.
Case in point: We have previously explained how the price-earnings (PE) methodology was used to justify higher and higher prices for glove stocks at the height of the Covid-19 pandemic — multiplying short-term supernormal profits by the historical average PE ratio, even though doing so is quite clearly absurd. When this no longer worked — when share prices started falling with vaccination rollouts — the valuation methodology was “changed” to discounted cash flow (DCF) or some modified combination of DCF-PE-cash flow. To the layperson, it all sounds very scientific. But, in fact, it is just a mirage.
We have said this many times before — we cannot decide to change the methodology whenever it suits us because there is ever only one way to value companies, and that is the DCF methodology. All other valuation metrics, such as the PE ratio, are just shortforms (simplified versions) of DCF.
So, unless analysts are completely incompetent — we would hope not, although it is more than likely some in fact are — the only explanation must be that they do it for other self-serving reasons. It is not the first time, nor will it be the last.
During the dotcom mania, sky-high valuations for internet companies were derived based on measures such as customer clicks, eyeballs and time spent on a website. Analysts touted these “new” valuation metrics with all the confidence that they were justified by the “new” economy and “new” business models. In reality, many of the internet start-ups do not even have a viable business model or strategy on how to turn a profit.
Thankfully, the most outlandish of these have since fallen by the wayside, in the aftermath of the dotcom bubble burst. But some persisted. One valuation metric that has been gaining serious traction in recent years is the price-to-sales (PS) ratio. In formula terms, PS = Market capitalisation ÷ Sales.
This valuation metric comes in particularly handy for high-growth and start-up companies yet to make a profit. So, there are no earnings (the E) to compute the PE ratio. Some also argue that PS is a new and better methodology — partly because, increasingly, earnings are being distorted by accounting standards — but it is not. (Admittedly, we are also increasingly frustrated by some of the recent changes to accounting standards, with accountants pretending to be economists, actuarial scientists and experts of other disciplines that they are not.) It is yet another short-form for DCF — and one that builds in even more assumptions than the PE ratio.
The use of PE ratios, especially when comparing across industries, comes with a whole lot of assumptions, including on growth, business risks and company-specific risks. That is why we say a PE of 9 or 229 times tells us nothing about relative valuations.
We have proven, mathematically, how fair PE valuations can differ greatly, depending on the expected long-term growth rates even assuming exact same risk profiles, which of course is not realistic. All companies are not created equal (for a more detailed explanation, please see “A bull or bear market in 2021?” [The Edge Singapore, Issue 971, Feb 15, 2021]).
Comparing stocks using the PS ratio makes a huge assumption on margins. Companies have different cost structures: fixed costs, marginal costs and cost of borrowings (interest expenses). Even companies within the same sector — say, technology — can vary greatly in terms of their cost structures. So, comparing a PS of 2 and 20 times is just as meaningless as comparing a PE of 9 and 229 times.
To demonstrate this, we did DCF valuations based on 10 years’ cash flows at different discount rates for Companies A, B and C (see Table).
Company A is in a high-growth business where sales are growing faster than costs. Examples of this type of companies would be platform- and software-based tech companies such as Amazon.com, ServiceNow and Adobe, where the marginal cost will decrease, as sales are spread over high upfront costs.
Company B is also in a high-growth business, but one in which costs rise in lockstep with sales. Real-life examples include tech stocks that are more focused on hardware sales such as Apple — though the company is shifting towards a more services-driven (subscription) business model that has decreasing marginal costs — and fast-growing, global coffee chain Starbucks Corp.
Company C is in a low-growth business, where the profit margins of its products are relatively constant. Carmakers such as Toyota Motor Corp and Volkswagen Group are good examples as is computer hardware manufacturer Dell Technologies.
As we have shown previously, the fair PE valuations of high-growth companies would be higher than those of low-growth ones, all else being equal. And as the discount rate — which reflects the risks specific to each company (risk premium) plus the risk-free rate (typically government bond yields) — rises, the fair PE valuations will drop, and the fall will be steeper for high-growth companies because earnings further in the future will be worth far less today. This explains the current selloff in growth stocks as US yields rebound from historic lows.
The fair PS valuations of low-growth and relatively fixed-margin companies, such as Company C, will be lower than those of high-growth companies with relatively fixed margins such as Company B. High-growth companies with expanding margins (decreasing marginal costs), such as Company A, will command the highest fair PS valuations, all else being equal.
That is to say, Company A trading at 3.8 times PS is no more expensive than Company B trading at 1.4 times PS, and Company C trading at 0.9 times PS is no cheaper than Companies A or B (at the discount rate of 10%, assuming they all have the same risk profiles, which in itself is unrealistic).
In summary, the PS ratio can be useful as a valuation metric for companies that are making only marginal or no profit — in which case, the PE multiple would be very high or infinite — but have strong growth prospects, high upfront costs and very low variable costs. For these companies, marginal cost will decline with rising sales and PE would drop rapidly in the future. Conversely, it makes little sense in using PS to value stocks where margins are likely to be fairly stable, even when sales are growing quickly.
In addition, investors must always bear in mind the risks of looking at PS ratio in isolation. For instance, a stock with high sales growth may appear to be a bargain in terms of lower and lower PS. But the company may never turn a profit and could even be on the verge of bankruptcy. After all, it is easy to grow sales if one keeps selling below cost. Eventually, the business will run out of cash (see “PS is useful for valuing loss-making companies, but it comes with great risks”).
We have written a fair bit on valuations and methodologies because we believe they are educational — too many commentaries and analyses simply do not make sense. For sure, ratios such as PE and PS are popular and widely used because they can be calculated quickly — rather than forecasting multi-years of cash flow based on many variables — and are easy to communicate and understand. They can be useful. Just remember that, when the experts start marketing new valuation methodologies, they are merely putting a new spin on an old concept. There is no such thing.
PS is useful for valuing loss-making companies, but it comes with great risks
Every investor dreams of finding the next Amazon.com or Google, the trillion-dollar companies that rose to global dominance with the rapid growth and adoption of the internet.
Internet stocks generated great excitement through the 1990s as many predicted, correctly, that the internet would change the world as we knew it back then. We see similar exuberance today, in stocks related to what are seen to be major secular trends such as those in the electric vehicle, new energy, fintech and cloud computing sectors.
Almost all start-up companies will incur losses at the beginning, before their businesses achieve sufficient scale. As such, they are particularly difficult to value, to separate the wheat from the chaff.
As mentioned in the main article, at the height of the dotcom mania, the market was valuing internet start-ups using metrics that were completely divorced from cash flows. We all know how that ended.
History does not repeat itself, but it often rhymes. Many of today’s start-ups in the hottest sectors are similarly loss-making. In the absence of earnings, one of the most popular valuation metrics is to compare their price-to-sales.
Following up from our earlier examples in the main article, Company D is currently loss-making but sales are expected to grow at a much faster pace (20% annually) than costs (10%) over the next 10 years. Assuming a discount rate of 10%, its fair PS valuation would be 3 times — higher than that of Companies B and C, which are profitable but whose sales and costs are expanding at the same rate.
This explains, in part, why many lossmaking start-ups such as electric vehicle manufacturers Nio and Nikola Corp are trading at significantly higher forward PS multiples than profitable carmakers such as Toyota Motor Corp, the world’s biggest carmaker by revenue and volume sales — because their sales are expected to expand at a far faster clip from low base.
Clearly, the PS ratio can be useful, but it also carries great risks, something that investors must keep in mind — particularly when expectations become irrational. For starters, assuming the same discount rate when comparing companies is not realistic because of different risk profiles.
Critically, sales and costs become increasingly less predictable the further we go into the future, especially when the industry landscape is also evolving rapidly. In our example, if everything goes to plan, Company D will turn profitable in Year 4. But if, say, costs turn out to rise faster than expected, for instance at 15% instead of 10%, Company D will break even only in Year 6. In this case, its fair PS valuation today should, in fact, be only 0.4 times instead of the original 3 times, which would result in a huge valuation contraction (correction) — and big losses for investors.
The Global Portfolio closed 4.4% lower for the week ended March 24, slightly worse than the MSCI World Net Return index’s 2.1% loss. Last week’s losses pared total portfolio returns to 50.6% since inception. Nevertheless, this portfolio is still outperforming the benchmark index, which is up 39.4% over the same period. The biggest drag on the portfolio was Geely Automobile Holdings, which fell 19% last week. China’s biggest carmaker reported lower sales and profits last year, attributed primarily to the Covid-19 pandemic. Other notable losers were Builders FirstSource (-10.8%) and Taiwan Semiconductor Manufacturing Co (-8.3%). There were only three gainers for the week: Home Depot (+4.5%), Johnson & Johnson (+0.7%) and Adobe (+0.1%).
Disclaimer: This is a personal portfolio for information purposes only and does not constitute a recommendation or solicitation or expression of views to influence readers to buy/sell stocks, including the particular stocks mentioned herein. It does not take into account an individual investor’s particular financial situation, investment objectives, investment horizon, risk profile and/or risk preference. Our shareholders, directors and employees may have positions in or may be materially interested in any of the stocks. We may also have or have had dealings with or may provide or have provided content services to the companies mentioned in the reports.
