The causes of the last crisis are more likely to be baked into market prices today, according to Orbis Investments’ Ashley Lynn, who said lightning seldom strikes the same spot twice.
Going through a crash is scarring, said Lynn, adding: “You feel the pain when it happens, and afterwards you spend lots of time trying to understand exactly what caused the crash and what protected against it.
“While you’re licking your wounds and retracing the steps in your head, other investors are doing exactly the same thing.”
She added: “Meanwhile, the real risks are the things few people are worried about, because if fewer people are worried about them, they are less likely to be reflected in market prices.”
When most regular people think of the word “risk,” they mean something simple, said Lynn, such as the likelihood of losing money.
Yet, when many economists and financial professionals talk about risk, they mean something quite different.
“Economic risk metrics largely consist of backward-looking measures like volatility and beta,” she said. “But as we know, the past is not always a good predictor of the future.
“In fact, a steady, low volatility climb upwards—while it will look great on these risk metrics—can mean that an asset is now priced too high, and very risky indeed.”
In late 2006, she said, if you looked back over the previous eight years, there was an asset which had delivered 13.7 per cent per annum returns, with only 6.5 per cent annualised volatility. Better returns than the stock market, with bond-like volatility.
“On any backward-looking statistical measure, it looked like a slam dunk investment. Unfortunately, that asset was high-yield commercial mortgage-backed securities, which went on to lose 64% of their value, wiping out all the returns from those great 8 years,” she said. “The lesson? Backward-looking risk measures do not predict the future.”
Backward-looking risk metrics are easy to measure, she noted. When you look forward, everything is hazy and uncertain.
“When you look backward, all the numbers are there, just waiting to be calculated to as many decimal points as you’d like. It feels precise, it feels certain—it feels reassuring,” she said. “For the most part, financial analysts don’t like uncertainty—we want to be able to put a number on everything.”
But what do these measures actually tell us?
“They do have their uses, and if assets were always rationally priced, they would be our best shot at predicting future price moves,” she said. “But if assets can, in reality, be priced ‘too high’, and therefore be risky, or priced ‘too low,’ and therefore contain a margin of safety, then those facts will be much more important than backward-looking price metrics in assessing the risk of loss going forward.”
The reality is that assets are not always rationally priced, she pointed out, adding that investor emotions like fear and greed affect the prices of assets, lifting them up past reasonable levels when they enjoy “golden child” status, and pushing them down below intrinsic value when they become unpopular or carry obvious risks.
“The past is more precisely knowable than the future, but the future is what’s actually going to affect your pocketbook.”
Lynn (pictured) provided some historical illustrations of the perils of looking backwards while hunting for returns in the future.
The first example is about the bond market. Over the past 30 years, government bonds have offered good protection in nearly every major stock market crash. Bonds rose as stocks fell, cushioning the blow for multi-asset portfolios. This has trained investors to buy government bonds for crash protection, but one crucial thing is different today.
In stock market crashes over the past 30 years, government bond returns started with significantly higher yields; so yields had room to come down, and prices had room to go up. Today bond yields are at rock-bottom levels.
Over $12trn of bonds globally trade at negative yields. This provides a natural ceiling on how much further the prices of these instruments can go up: at some point, people will not be willing to pay governments for the pleasure of lending them money.
“Long-dated government bonds cannot provide the same degree of crash protection that they did previously for a simple reason,” said the Orbis fixed income head. “Their prices are already about as high as they can get.”
The next example shows that low-volatility stocks appear to be in a similar boat. As central banks have pushed down bond yields, some investors have been forced out of bonds and into equities in the hopes of earning an acceptable return.
Naturally, these reluctant equity investors have gone for the most bond-like stocks they could find –– low-volatility stocks, and preferably ones paying a steady dividend. This popularity has pushed up the prices of stable shares.
At a high-enough price, however, any asset can be risky—even one with a historically stable price. When you buy a stock, you are buying a stake in a business, so you should always do your homework on the company fundamentals.
“Today, some low-volatility stocks have very high valuations and very stretched fundamentals,” she said. “In our view, that’s a dangerous combination.”
The important point is that the riskiness of any asset changes with the asset’s price. At some point, if widespread investor fear drives the prices of “low-risk” assets far past their true value, they can no longer be considered safe.
“The most risky thing an investor can do is buy something for more than it’s worth, simply because ‘everyone knows’ how safe it is, or because it has appeared safe in the past,” she said. “This isn’t risk reduction—it’s participation in the madness of crowds.”
Twelve years ago, Lynn said, giving another example, we heard about another “completely safe” asset whose price was uncorrelated to equity and only went up. Pretty soon we had the subprime mortgage meltdown and a full-blown global financial crisis on our hands.
The portfolio manager believes investors can avoid this rear-view-mirror driving by asking the right questions.
Instead of asking “what was low risk in the last crisis,” investors should drill down to fundamentals in order to find out which assets are priced fairly, and which are likely to be low risk today.
Whether a potential investor is analysing a stock or a bond, the questions should be very similar: “what am I paying for, exactly? Do the assets, cash flows, and potential upside justify the risk of loss?” If the answer is no, then the asset does not deserve a place in client portfolios—regardless of how popular or “low risk” it may be perceived to be.
“At Orbis, when we think about risk, we are thinking about the chance that the fund loses money. And in our view, the best way to lose money is to pay more for something than it’s actually worth.”
She explained that when her team judges the riskiness of an investment, they aren’t only concerned with how its price has behaved in the past. Instead, they focus on whether that price looks too high today, and hence might go down permanently in the future.
Lynn is a portfolio manager on the $43.4m Orbis Global Balanced fund, which is headed by FE Alpha Manager Alec Cutler and seeks to balance investment returns capital growth and risk of loss through investing in portfolio of global equities and government bonds.
Performance of fund vs sector since launch
Source: FE Analytics
Since inception of the strategy it has returned 85.42 per cent against 60.98 per cent gain for the IA Mixed Investment 40-85% Shares sector. It does not levy a management charge but has a performance fee of 50 per cent of the outperformance by the fund of its composite benchmark.