The right price?

THE BIG PICTURE

We apply some simple, but far from fool-proof tools to assess whether investors might be overpaying for stocks, with the majority of our focus being the US.

At the heart of this article is one of Warren Buffett’s famous quotes: “For the investor, a too-high purchase price for the stock of an excellent company can undo the effects of a subsequent decade of favorable business developments.”

Investor returns hinge on three simple factors – the price the business was purchased at (often expressed as a Price/Earnings (P/E) multiple), the sustainable earnings growth or profit generation that the company can produce over time, and the extent to which it pays out cash along that journey.

Future profitability and cash payments are largely unknown, the easiest known starting point is today’s price. We fully appreciate that inflation levels, and thus interest rates, have a major bearing on P/E levels, and comparing P/E levels at different periods can be flawed without accounting for the discount rate (interest rate). That said, interest rates should normalise through a cycle and we also acknowledge that this analysis is rudimentary.

Forward P/E multiples

This first table, to a large extent, eliminates the earnings collapse experienced by companies during COVID-19. It shows the forward P/E multiple that various markets are trading at one year from now (when earnings would largely be back to ‘normal’) and compares this to historical forward P/E’s.

Table 1: Market forward P/E’s versus history

Source: UBS. Date: May 2021

With the exception of the UK and Japan, markets are trading at elevated levels relative to their historical averages. Although this says nothing about specific companies, it tells us a lot about aggregate valuations in markets.

Bull markets – do they begin when P/E multiples are high?

A different approach is to gauge whether bull markets are created off high P/E ratings. The challenge once again with this analysis, is that P/E multiples in their simplest form can give false negatives. In other words, at times P/E multiples look very high – a sign of a very expensive market, but are being distorted by artificially low earnings or a depressed “E” as is typical in recessions. In table 2 below, this occurred in 2003, 2009 and over the past year. As a launchpad, the 2020 P/E multiple started off a higher base than the 1960 and 1974 bull markets.

Another insight is that all the bull market periods since 1960 have ended at lower P/E multiples than the present rating of the S&P. Currently, Bulls would argue that the “E” is depressed and we need to look through this slump, which we agree with. Even so, the 1 year forward P/E multiple would still be higher than the bull market exit P/E multiples in 1968 and 2007.

Table 2: S&P 500 bull markets since 1960

Source: Shiller website, Northstar Asset Management. Date: 31 May 2021

Cheaper markets are better predictors of higher future returns

Our final piece of analysis simply assesses whether an investor is better off buying a market on a lower or higher P/E multiple. Again, the data is noisy, but conclusive: lower P/E multiples provide a margin of safety so that future returns are simply better and more predictable.

Chart 1: Probability of positive returns & average historic subsequent return from a P/E of 24

Source: Shiller website, Northstar Asset Management. Date: 31 May 2021

Nominal GDP growth is an important driver of sustainable corporate profitability

Future returns also depend on sustainable profit growth. In Chart 2, we show nominal GDP growth in the US economy over 3 decades and we compare it to the profit growth of the S&P. The following are worth noting:

  • A strong relationship exists between nominal GDP growth and S&P earnings growth over time.
  • Since the Global Financial Crisis (GFC) in 2008, it has been a challenge to grow nominal GDP, despite all the stimuli. In fact, growth was lower during the two preceding decades.
  • Concurrently, earnings growth has been muted since 2009.
  • After each crisis, GDP and earnings growth surged back. However, after the 2020 COVID-19 crisis, the extent of the rebound has exceeded anything seen before – a function of uber stimulus.
  • Nominal GDP growth is expected (market consensus) to return to benign levels post 2022, similar to post GFC growth numbers.

Chart 2: S&P earnings vs United States of America nominal GDP

Source: S&P Capital IQ, Shiller website, Northstar Asset Management. Date: 31 May 2021
Chart LHS scale cuts off 793% outlier peak in 2010.

If both GDP and earnings growth have been muted over the past decade, why has the US market done so well? The answer lies in historically low inflation and interest rates. Without persistently low interest rates in the years ahead and assuming that future GDP growth rates (consensus) will be suboptimal, it is likely that corporate profits after this initial rebound will be muted. This poses risks to overvalued components of the equity market.

Navigating equity valuation risks

Right now, central bankers and governments have created an environment conducive to risk taking. Consequently, investors have driven and been rewarded by elevated market levels. Going forward, threats to this paradigm exist, which require mitigating actions. Our approach on behalf of our clients relies on the following interventions:

  • Utilise investment products that allow flexibility to reduce risk as appropriate.
  • Reduce overall equity exposure within our Northstar Global Flexible strategy when our bottom-up valuation work clearly indicates that stocks are not offering value on our Buy List.
  • Produce analysis that deeply appreciates the drivers that sustain earnings and ensure that these profits are fairly accounted for by the market.
  • Remain proactive and nimble to position portfolios in securities adhering to our ‘quality at a reasonable price’ philosophy.

Northstar’s offshore funds have generated market leading returns for many years, we anticipate a tricky investing landscape after the current euphoria subsidies, but believe that our investing framework is well placed to account for this.