Revathi Asset Management

Comparing P/E and EV/EBITDA

One of the first things that portfolio managers see is the vast difference in valuations that the market is willing to pay across sectors. This later develops into biases regarding which are “cheap” and which are “expensive.” Finally, and more dangerously, these statements morph into the false arguments of “cheap P/E companies/sectors are low quality” or “expensive P/E companies/sectors are high quality.”

Let us begin by comparing the P/E of two stocks, HINDUNILVR and HINDALCO, in two vastly different sectors, fast-moving consumer goods and metals & commodities respectively.

We see that HINDUNILVR, on average, traded at almost 4.5x the P/E of HINDALCO in the last 10 years. Furthermore, it did so with a similar absolute standard deviation, leading to a P/E that is almost 4x as volatile for HINDALCO than it is for HINDUNILVR on an adjusted basis. For example, this is generally where market participants learn that “FMCG is always expensive in India” or that “metals are always cheap in India.”

However, a problem with Price/Earnings is that is calculated using net profit, i.e. net of one-offs. Since these may contribute to significant volatility in the bottom line earnings of a commodity company like HINDALCO, let us compare a more reliable metric that looks only at the core business operations of the firm.

When we look at the average EV/EBITDA that the market has paid for these companies, we see that HINDUNILVR has traded at around 5x that of HINDALCO. However, once we compare the adjusted volatilities in the EV/EBITDAs, we see that the difference is not as pronounced as before; on average, the volatility in HINDALCO’s valuation is only about 1.19x (0.220 vs 0.186) that of HINDUNILVR’s. The reason we also compare the volatility in the metric is to see how the market’s pricing of the companies also changed over time; whereas valuation only tells us the one-off valuation at a given point in time, the std/mean of the valuation also tells us how the pricing of the two companies moves relative to market factors.

Since the adjusted volatility of HINDALCO’s mean is about 0.220 vs HINDUNILVR’s 0.186, one could argue that any volatility that shows up in HINDALCO’s EV/EBITDA for the most part also shows up in HINDUNILVR’s EV/EBITDA. Therefore, from a pure price perspective, this would imply that although HINDUNILVR is much more expensive in P/E terms than HINDALCO, the core valuations of both stocks are similarly affected by market factors and overall sentiment.

Does profit explain valuations?

Since we had established groundwork for the DCF earlier, let us use the same logic to check whether the valuation gap between the two companies can be found in the cashflows of the companies. Assuming that all information eventually flows through to the bottom line, we would expect that, given a 4.5-5x valuation difference, there should be a similarly-significant difference in the net profits of the two companies.

When we compare the actual numbers, we see that HINDALCO closed Mar 2024 with a very similar net profit as did HINDUNILVR. However, the average earnings posted by HINDALCO was around 6050 vs 7100 for HINDUNIVR. More importantly, the volatility in the earnings was much higher for HINDALCO, leading to a std/mean that is over twice as high at 0.74 vs 0.33. We had briefly touched upon this earlier when we had discussed that the markets generally pay a discount for more volatile cash flows. Here, however, we are seeing a 2.25x volatility in similar terminal cash flows vs a 4.5-5x valuation difference. So, net profit alone does not capture the story. Let us once again compare the operating profit to get a more streamlined, core-business-operations point of view.

Here, we see a more muted story for the adjusted volatility in HINDALCO’s operating profit, which is almost the same at 0.37 vs 0.32 for HINDUNILVR. So, while the net profit explains part of the difference in valuations paid for these companies, there is still significant information that we are missing. More importantly, it is not being captured by the cashflows of the core business operations themselves because HINDALCO’s operating profit also seems to be growing at a faster YoY rate. Thus, the other profitability, solvency, and efficiency ratios must be playing a role.

Margins, solvency, and efficiency add colour

Once we compare the net and operating margins of the two companies, we come closer to the true picture. HINDALCO’soperating profit margin is also significantly lower on average (11.31 vs 21.76) and the adjusted volatility in the margins is also higher (0.18 vs 0.134). Additionally, HINDALCO’s net profit margin is both significantly lower and significantly more volatile than HINDUNILVR’s.

This implies that the cashflows from HINDALCO’s core business are more volatile than those of HINDUNILVR’s and that HINDALCO’s one-offs contribute further to even greater volatility.

Solvency ratios explain a further part of the valuation gap between the two companies. We see that HINDALCO also carried much more debt on average at 1.15x equity than HINDUNILVR at 0.01. The interest coverage ratio is not necessary to compare because HINDUNILVR operated at close to 0 debt for the majority of this period. Equity investors are generally happy to see such debt-free companies.

Lastly, comparing an efficiency metric like fixed asset turnover, we again see that HINDUNILVR is a “better” company; it turns over, on average, 4.22x its fixed assets every year vs 0.96x for HINDALCO. This ratio is important because it implicitly tells an investor how much operating leverage is built into the company and whether the company needs much more capital to grow or not.

Relating to the numbers we had seen earlier, even if HINDALCO can continue growing its net and operating profits at a faster pace, it will also need to raise a significant amount of capital, either through debt or equity, to sustain that growth. Thus, HINDALCO’s higher growth rate is not as much of a one-sided positive factor as we might previously have thought.

What about actual stock price performance?

Finally, let’s get to stock price performance because that’s what affects the portfolio. We can do as much analysis as we want but ultimately risk and returns matter just as much, if not more. Just for context, the table below shows the EV/EBITDA for the stocks at the beginning and the end of the period. Looking at valuation metrics alone, HINDUNILVRbecame even more expensive and HINDALCO became even cheaper.

HINDALCO, however, managed to deliver a higher return.

We see that HINDALCO delivered almost 2x the CAGR with lower price volatility (i.e. better Sharpe and Sortino ratios) but with a higher max drawdown (i.e. lower Calmar). In effect, an investor in HINDUNILVR paid up for a “quality” business but received less return and higher relative volatility. He may have had higher peace of mind due to the lower max drawdown, but the “better” company relatively subtracted from his overall portfolio performance.

Let’s take it one step further. Let’s estimate two DCFs for the stocks using the methodology we had established earlier. What we see is that the situation is even more complicated than we had thought so far. Using the latest traded prices (2350 HINDUNILVR, 620 HINDALCO) and dividends (29 HINDUNILVR, 3.5 HINDALCO) at the time of writing, we see that HINDALCO’s dividend requires a 20% CAGR in growth vs 10.5% CAGR in growth for HINDUNILVR’s dividend for the next 10 years, after which we assume that both stocks will flatline to 10% dividend growth at a 12% cost of equity capital. For simplicity’s sake, we assume that margins will be flat for both companies and that conversion from operating profit to net profit to dividend payout will also be the same as historical.

As a result, we can’t say that HINDALCO is an even better buy now than it was before just because it has a lower P/E. Its current stock price requires a certain amount of future growth that is even higher than the historical growth rate (average 16.6% operating profit growth YoY) which was from a much lower base. In that sense, HINDUNILVR appears “cheaper” even though the P/E is much higher because the 10.5% implied growth rate is lower than what was historically achieved by the company (11.9% YoY). As a final point, it is important to remember that these are just implications of the current stock prices; we are not saying that 10.5% growth by HINDUNILVR is more likely to be achieved than 20% growth by HINDALCO. These are just forward estimates.

Valuations only confused us

With this specific example of two well-known sectors having a massive valuation gap in the Indian markets, we have seen that valuation metrics like P/E tell us almost nothing about stock performance or whether a company is “too cheap” or “too expansive” from a portfolio manager’s perspective. At best, valuation metrics offer a basic calculation and at worst, they mislead the investor.

In practice, it is very misleading to say that a stock is too expensive or too cheap based on valuation alone. It is misleading to say that, in this example, HINDUNILVR is a “better” company than HINDALCO because this is factual public knowledge based on the ratios that we compared. Rather, the fact that HINDUNILVR is a “higher quality” company is already priced in and that’s exactly why the valuation gap between the two companies exists. And yet, the “lower quality” company delivered twice the CAGR returns with better risk metrics over 10 years. So which is the “better” company?

Finally, at the same time, one cannot argue the opposite viewpoint that a lower P/E company is inherently better than a higher P/E company. To make qualitative judgements about stock prices and valuations here we need to compare them to estimates of future growth rates. We have just seen that the lower P/E company requires almost twice the growth rate to justify its current stock price, when both are very large, established companies having similar operating profits.

Thus, while it may not be immediately actionable, all we are highlighting is that P/Es and EV/EBITDAs can be misleading, regardless of how you look at them. All we can conclude is that:

1. Valuation metrics can be “too low” or “too high” and end up staying that way or get worse.
2. Core business factors such as profitability, solvency, and efficiency explain much of the differences in valuation metrics.
3. The market already charges accordingly for “high quality” and “low quality” businesses.
4. High P/E doesn’t automatically translate to better stock price performance, but low P/E isn’t automatically a good thing in and of itself either.
5. “High quality” stocks may not necessarily be better portfolio additions.