Why did no one see it coming, asked the Queen in the wake of the financial crisis. This time, she should be forewarned. James Ferguson reads the signs.
Sebastian Junger’s 1997 book, The Perfect Storm, recounts the unhappy confluence on Halloween 1991 of a once-in-a-100-years combination – the collision of two separate cold-weather storm systems in the US, with a warm-weather hurricane from the Caribbean. In financial markets, as with weather systems, the true monsters have not one but many fathers.
We may be looking into the eye of one such Goliath in global bond markets right now. As with storm systems, it’s possible that the vulnerabilities that now crisscross bond markets won’t come to anything. But it’s also possible that one innocuous financial butterfly wing is all it will take to trigger a global perfect storm.
Why did no one see the financial crisis coming, the Queen asked a group of academics at the LSE in November 2008. Cue a lot of stuttering, evasive mumbling and staring at the ground. But academics were the wrong people to ask – certainly, plenty of people in the finance industry knew something was up. The real question raised by Her Majesty’s query is this – even when investors see potential crashes coming, why aren’t they incentivised to take better precautions?
No one, not even the man in the street, was surprised by the dotcom crash. Valuations had become silly, then sillier still. But that’s part of the problem. Those few fund managers brave enough to take a stand and eschew tech stocks ahead of the millennium – such as Phillips & Drew’s Tony Dye in the UK – paid the price with their careers. Dye didn’t get his job back or even an apology when he was proved right.
Two key lessons from the crash
The fact is that institutional investors, those in the best position to recognise the danger signs, have no incentive to act. In July 2007, Charles “Chuck” Prince, then CEO of Citigroup, gave one of the defining quotes of the crisis era to the Financial Times. Discussing the turmoil that had already begun in the US subprime market, he said the party would end at some point – but “as long as the music is playing, you’ve got to get up and dance. We’re still dancing.” A year later, Citi was fighting for survival.
There are two key lessons here. Both the FT and Citi knew things were bad 14 months before Lehman Brothers failed (even while the Federal Reserve was assuring investors there would be no contagion), but everyone chose to ignore the warnings. Secondly, Prince spoke for all bankers when he said “you’ve got to dance”, with “got to” being the critical phrase.
Benny Higgins, now head of Tesco bank, lost his job as head of retail mortgages at HBOS after losing market share to Northern Rock. Two months later, the Rock was bankrupt – but Higgins was not offered his job back.
Fund managers too have limited and sometimes even inappropriate remits. They either have to track or beat a benchmark. The Bank of England’s Andy Haldane noted in a speech a year ago that this makes it very risky career-wise to act counter-cyclically.
Pull out of a rampant bull market, and a manager will immediately underperform, losing funds and perhaps their job. Rational managers would only take the huge personal risk of pulling out of the market if they were convinced a significant and damaging correction were absolutely imminent.
That’s why disaster tends to dawn on almost everyone at once. This makes for what academics call an “asymmetric market” – or what you and I would call a “crash”. So any manager who attempts to reduce risk for his investors is taking huge risks with his own career. Investors’ and managers’ interests are out of sync – never a good sign.
Common sense on risk is wrong
However, risk is an even more insidious foe than that. In his latest letter to shareholders, Warren Buffett bemoans the fact that, “in business schools, volatility [how much an asset moves up and down in price, basically] is almost universally used as a proxy for risk. Though this pedagogic assumption makes for easy teaching, it is dead wrong.”
Wharton business school academic Jeremy Siegel, author of Stocks for the Long Run, looked at more than 200 years of real (after-inflation) returns. He found that while stocks were the most volatile assets over the short term, mean reversion meant they were the most stable (and rewarding) in the long run.
However, the mathematics of volatility is easier to teach and model than the Bayesian statistics of mean reversion. So pension-fund trustees have been pressured by advisers and regulators alike to eschew stable, high-yielding, long-term investments in stocks, in favour of fixed-income instruments (bonds).
The Bank of England’s Procyclicality Working Group found that the weighting of domestic equities in long-term UK institutional portfolios (ie, pension funds and insurance companies) had fallen from 50% in the 1980s-1990s to just 5% now. The fear is that this is due to regulatory insistence on matching fund assets to pension and life insurance liabilities, which is taken to mean selling equities and buying bonds.
This has pushed institutional funds into an extreme over-weighting in fixed-income and debt securities. The flow of such large amounts of money from stocks into bonds has contributed to an almost unprecedented bond bull market. Between 2000 and the end of 2013, UK institutional funds doubled in size to £3trn.
At the same time, their domestic equity holdings fell from about £800bn to just £150bn. That’s £100bn a year being fed into bond markets that institutions would once have funnelled into domestic shares. Multiply that across the developed world and you’re talking trillions of dollars – easily enough to have fuelled the global bond bull.
Bonds outperform stocks
The other problem with risk is that it’s mutable. It may have been a mistake for UK institutions to have held 50% of assets in domestic equities on the eve of the dotcom crash – but as we’ve already seen, managers track the benchmarks they’re given. It is up to asset allocators and owners (you and I, in other words) to change our weightings.
Today, a very risk-averse investor is directed away from shares and into bonds. Why? It has long been recognised that, in the long run, equities beat bonds. In theory, this happens because equities are more volatile – you get the higher return to compensate for the rough ride. So risk-averse investors (who are likely to be thrown off the ride because they can’t cope with the volatility) are directed towards bonds instead, which again, in theory, offer lower returns and so must be lower risk. Trouble is, for the last 30-odd years, bonds have not returned less than shares. Quite the opposite.
Data from both the US and UK suggest that, over 100 and 200 years, the real (after-inflation) return from equities averages about 6%-6.5% a year, and 8%-9% for smaller stocks. Long-term government bonds have managed 2%-2.5%, and only 1% or less from short-dated bonds. But since 1981, things have turned out differently.
Over the last 33 years, stocks returned 7.5% a year – not far off their long-run average. But bonds have shot the lights out. Even before the surge in the last 12-15 months, UK gilts had already almost matched UK shares, while US Treasuries actually beat stocks with an 8.3% annual real return – the first time bonds have returned more than stocks over any 30-year time frame since the American Civil War (1861).
Although this long, sustained bull market has pushed bond volatility very low (volatility is a lagging indicator – it only tends to spike when prices fall), logic dictates that, if bond returns are now equity-like, then bonds must also be every bit as risky as equities. Yet UK institutions have just finished shifting 90% of their domestic share portfolios into bonds precisely because they are supposed to be less risky!
And unfortunately, over the last 12 months or so, what was already a dangerous situation has become very precarious indeed. More than half of the world’s bonds yield less than 1% (bond yields fall when bond prices rise and vice versa). A whopping $5.3trn carry negative yields (a phenomenon so rare that it wasn’t considered even theoretically possible), 60% of which are in Europe. US bond supremo Bill Gross has called German government bonds the short of a lifetime.
As the chart here shows, after a bull market that has already lasted a quarter of a century, bund prices – at least before last week (see page 6) – were accelerating higher. This sort of surge at the end of a multi-year bull market is very similar to what the Nasdaq looked like at the end of the dotcom bubble, or Tokyo in 1989, or oil at $145 a barrel – you get the drift: it’s usually an end-of-an-era thing.
A perfect storm
Perhaps this wouldn’t matter as much if bonds weren’t also demonstrating various other “perfect storm” attributes. Riskier, high-yield (junk) corporate bonds have collapsed in value compared to their safer investment-grade peers on three occasions in the last 15 years alone (2000-2002, 2007-2009 and 2011-2012).
This happens regularly whenever the economy gets a scare. But the hunt for yield over the last three years has pushed all yields ever lower. This allows junk borrowers who might have gone bust to instead re-finance at a cheaper rate, reducing defaults.
In turn, lower defaults make junk bonds seem less risky, which narrows the difference (the spreads) between what junk borrowers have to pay and what investment-grade borrowers are charged. For example, B-rated corporate bonds defaulted about 7% of the time in the 1980s and 1990s (peaking at 17% in 1990). But over the last 12 years, falling yields – the bond bull market – have kept the average default rate at just 1.5%. Junk bonds are so-called for a reason – the next time the system is stressed, defaults will inevitably rise.
But it’s ever less likely that anyone will see this coming. That’s because the quality of covenant protection (contractual obligations that serve as early warnings for bondholders, basically) has almost halved since 2002 (according to the OECD think tank), while 60%-70% of bonds issued in Europe each year are exempt from requiring a prospectus.
The scariest thing about bonds
Yet the scariest feature of bond markets today may not be inadequate protections, nor that institutions are holding too many, nor even negative yields. Most worrying of all may be the lack of liquidity. Most bonds effectively cease trading about a month after issue anyway. But what matters most in a panicky market is finding a buyer to establish a floor under prices. When this doesn’t happen – as in the subprime securities crash of 2008, or the credit markets in early 2009 (when euro junk yields, which are 3.6% today, hit 27%) – prices, for a while at least, have no safety net and the market fails to clear.
The banking crisis has led to new rules which have made securities much more expensive for investment banks to hold on their trading books. As a result, bond inventories at US primary dealers are today just a quarter of what they were pre-crisis. Investment-grade bond turnover has fallen to a lower level than even at the worst point of the 2008 crisis.
What this means, in practice, is that if a big enough shock hit the global bond market, chances are that no one would be in a position to bid for large institutional blocks. For bond funds which let their retail investors access their money on demand, this could result in something akin to a bank run – with angry investors queuing to redeem holdings that the fund can’t find a buyer for.
Even once you’ve created the conditions for the perfect storm, the winds could change – clear skies may yet prevail. And it’s true that the world faces more of a deflationary threat than an inflationary one just now, which is one factor driving the bond bull market.
However, once a market is widely seen as offering little or no value, the specific trigger that starts the selling is far less important than the distance prices have to fall before they attract a large enough group of bargain-hunters. For example, I still don’t know exactly what triggered the 1987 crash. Or precisely why the dotcom rollercoaster went over the top.
What will the trigger be?
Maybe the trigger for a bond rout will be emerging-market borrowers struggling with the strength of the dollar. Or fears from speculative bond buyers that the European Central Bank will taper quantitative easing (QE). Or that QE will work too well, and revive inflation. Or maybe Greece exits the euro. Or China turns south. Or companies default en masse. Or something else.
Whatever the trigger, at that point, the liquidity constraints across the system will be revealed and the fracture lines become all too clear. Worse still, bonds tend to be the preserve of the most cautious of investors – those least able to tolerate a loss. So it makes sense to weatherproof your portfolio in advance (see below for more on how to do it).
Obviously, financial forecasts, like weather forecasts, can only ever offer a probability of being correct. So I won’t claim to know exactly what’s coming. I can only present the damning evidence. But at least you can tell the Queen, if she asks, that this time, you saw it coming.
• James Ferguson is a founding partner of the MacroStrategy Partnership LLP.
So should I sell all my bonds?
There’s a diversification argument for holding bonds as part of a portfolio, writes John Stepek. Typically, bonds are not tightly correlated with equities – they frequently move in different directions. And if we remain in a QE–driven, deflationary environment, bonds could get even more expensive than they are now. So we’re not saying “dump every bond you hold”.
However, we would suggest you consider your asset allocation. Bonds are unquestionably overvalued relative to history, so your allocation to them should be lower than it would otherwise be. And understand what it is you own – shorter-dated bonds (those with less time to go until they mature) are less sensitive to changes in interest rates and the economic environment, so they wouldn’t suffer as badly in any sort of “perfect storm”.
A bonds crash would be bad for equities too. Cheap money has encouraged companies to borrow to buy back their own shares, driving up share prices. Anything that pushes up interest rates will hurt, and a bond-market crash would mean a panic. But timing it is impossible. So we’d invest in markets that offer decent value – the US is very expensive compared to history, but various eurozone markets and Japan look more reasonable.
“Safe-haven” assets would do well in a crash. These include gold – it has no counterparty, so benefits when people are worried about solvency. So own some. The US dollar would likely do well too – it’s still the dominant currency, and remains an asset people tend to run to in a crisis.