Why does a barbell portfolio make sense?

This is the exact question asked by a client referring to one of our charts shown below. As can be seen, although there has been a clear shift in our portfolios towards cyclicals over the last few years reflecting the value emerging there, we continue to hold exposure in tech/internet: hence what can be termed a barbell (see charts below).

This split serves a purpose given the growth versus value debate of late, but we prefer to say that we have broad exposure to Asia and emerging markets or multiple themes with a bias towards the most undervalued at any given time.

I’d make two observations about our Invesco Asia ex-Japan Equity strategy before specifically addressing the question:

1. yes, we have a value bias currently, and we would expect to have a slight value bias through the business cycle – that’s because we tend to have more conviction in our ability to spot undervalued companies which have temporary setbacks yet priced as permanent – versus putting a lot of faith in the long term growth trajectory of a business. On balance we think the risk/reward is more favourable when there is little optimism built in
2. a barbell approach is another way of saying that we are unconstrained: we can ride our winners for as long as fundamentals allow and tend to be contrarian at purchase – this can offer good differentiation against peers in addition to good returns

So, to reverse the question slightly, why should we constrain ourselves to one style?

The same can be said about active investing. We should not have to rely on a set of conditions or factors to outperform. Indeed, themes and factors go in and out of fashion over a business cycle and labelling companies as single factor bets is overly simplistic.

Every company has its own cycle, its growth profile, and setbacks, which is why even at the height of the valuation discrepancy between growth and value in 2020 we found idiosyncratic reasons to retain a well balanced portfolio, although with a perhaps controversial overweight in cyclicals given the contrarian opportunities there.

As a general rule, we think it’s in the best interest of our clients to hold a variety of ideas at different stages in their transition from contrarian to popular, while ensuring that they continue to meet our total return threshold (3-year CAGR >10%). Some of our holdings are secular growth stories; others benefit when the cycle turns; some have re-rating potential without implying aggressive assumptions; and we own high yielders.

This means that we tend to hold a mix of growth and value or tech/internet and cyclicals, as illustrated above, and our track record demonstrates that this broad approach works. Over the last 20 years or so, the outcome for clients on average has been >10% p.a. and around 3% alpha p.a. net-of-fees, as intended.

In a recovery, cyclicals have shown to offer better earnings growth as well as re-rating potential – and our process is to trim/sell when the components of total returns (business growth; re-rating potential; dividends) no longer add up to 10%.

It is notable that some of the fastest growing companies and best performers so far this year were originally considered value and rated ‘sell’ by consensus analysts when we initiated the positions.

Pacific Basin (+112% YTD) and Asustek (+53% YTD) come to mind. Both stocks could still be considered value trading on a forward P/E of 8x with dividend yields of 7%, but the earnings upgrades over the last 3 months (+100% and +40% respectively) suggests that consensus expectations were considerably off the mark and are only now catching up.

Are these now value or growth stocks? Historically, our largest positive contributors began as value ideas and then graduated into growth stocks. It’s obviously not a pre-condition for outperformance but perhaps indicates that it not the style that matters but the journey and we wouldn’t contemplate disassociating business growth from valuations.

Two additional points on why some value exposure makes sense: firstly when the investment case doesn’t go as expected, the downside risk is usually limited as expectations are already low; secondly, cyclicals offer some diversification today in the context of many Asian and emerging market portfolios seemingly overly exposed to growth stocks.

Also, as the market grapples with the implications of inflation on valuations, this could be more challenging for long duration assets, particularly as the scarcity of growth premium is less obvious in the context of a rising tide lifting all boats.

We continue to hold GARP stocks and remain selective. We have been reducing the wider tech/internet exposure over the last 12 months but still have a slight overweight in tech hardware and media & entertainment, where new opportunities have emerged.

Buying structural growth stocks can be rewarding if future growth is not already in the price. However, even when growth prospects look strong buying stocks on unreasonable valuations is too risky and not a sustainable way to make money.

It is equivalent to giving away a lot of the value to the seller of the shares with the hope that the company will manage to sustain its high growth further into the future. Whilst conceivable in a small number of cases we tend to steer away from making optimistic predictions about long term trends and recognise that competition and regulation can play a part in levelling the playing field. More importantly, if the rosy scenario does not pan out, the high valuation multiple is a vulnerability.

Fortunately, we can avoid frothy parts of the market and find good growth at a reasonable price. Some of the 2020 winners including unprofitable internet and ESG-related stocks have lost between 1/2 and 1/3 of their value so far this year.

The dominant Chinese internet companies have also not been immune to the recent turn in sentiment caused by anti-monopoly regulation. Some of these are back on our radar as we contemplate good entry points at more compelling valuations. A too strict interpretation of value can lead to missed opportunities.

Conclusion

As long as there are mispriced opportunities across the wider market, we would expect to have multiple themes in the portfolio. Today our portfolios appear to be well balanced with a cyclical tilt (or value bias) but each stock is held on its own merits.

As the market becomes less binary, with cyclicals recovering from their lows and tech/internet retrenching from their highs, we have the flexibility to uncover any type of opportunity. We very much adhere to the motto Price is what you pay, value is what you get regardless of style which is why we spend most of our time working out fair value, and as such new ideas tend to come from unloved parts of the market.

Our DNA is to lean against consensus when too extreme, and to act with conviction if we can back it up fundamentally. We have a total return approach, agnostic to factors and styles, but being contrarian at the point of purchase can lead to differentiated portfolios.