Date: 30.08.17 Category: Working For Change
One tenet of risk management over the last two decades has been the assumption of a negative correlation between prices of government bonds and equities. This has become a fundamental part of portfolio construction and lies at the heart of asset allocation and risk-parity strategies. Government bonds have been the perfect hedge for equity allocations: negatively correlated, favoured by regulators, cheap to leverage and with a positive expected return, or ‘carry’. Created through a combination of structural macroeconomic changes, central bank policy and the birth of the modern financial crisis, it has enabled a much more passive approach to risk management. Alas, it is unlikely to last and when it reverts, risk management must too.
“ Central banks are my volatility protection” – Comment in the Financial Times, 05.05.2017.
The 1970s and 80s were an age of disco, questionable fashion and, for a generation of professional investors, a world where higher equity prices meant lower yields on government bonds and vice versa. This positive correlation between bond and equity prices was a persistent feature of the US financial system from 1970 until the summer of 1997, whereupon it changed direction, practically overnight.
In 1997, Alan Greenspan had been the Federal Reserve Chairman for a decade and lauded for his success in ‘price stability’. As the economy and equity markets boomed, inflation did not and the ‘Goldilocks economy’ was in effect. Even temporary tightening of rates in 1994/95 was considered to have been a successful ‘pre-emptive strike’ on inflation, saving the economy from a danger that never even turned up in the data. In short, the Federal Reserve was considered omnipotent, willing and able to manage inflation perfectly.
Analysing inflation and productivity statistics, Greenspan concluded that the reason inflation remained subdued and immune to its traditional influences was that the economy was experiencing a ‘productivity miracle’. However, one could argue that the actual reason was a combination of long-term deflationary factors impacting the global economy, such as computer technology, globalisation, demographics and government policies. In addition, in the preceding years, inflation measurement had been changing with a gradual inclusion of ‘hedonic adjustments’ and focus on ‘core’ inflation figures, further dampening inflation. In any case, Greenspan’s conclusion was that monetary policy could be run looser and hence rates held lower in the future, driving both equity and bond prices higher. Then came the Asian crisis.
“ If something can’t go on forever, it will stop.” – Herbert Stein
The 27th October 1997 marked the first substantial global ‘risk-off ’ event in modern history in which investors across the globe liquidated risk positions in favour of government bonds as a response to a crisis. Over the next 12 months, as the Asian crisis became Russian, the Fed would cut US rates three times in response.
RESPONSE TO FAILURE
Most importantly, in September 1998, as a response to the impending failure of the hedge fund LTCM, the Federal Reserve announced a second, unexpected rate cut during a special meeting. As was the intention, bond and equity prices rallied. The fed had cut rates amidst a booming economy to stabilise equity markets, the now infamous ‘Greenspan put’ was born and the correlation collapsed further, and not just in the US. Other developed markets also experienced this sudden change in correlation in 1997, including Germany (Andersson, Krylova and Vahamaa, 2004) and the UK (Baur and Lucey, 2006). Thus, government bonds could be expected to protect not only against a potential recession but, more importantly, systematic risk in the form of a liquidity driven crisis. By doing so whilst providing a positive carry, systematic risk could largely be reduced in a passive manner. And it worked beautifully. We, humans, tend to anchor expectations of the future in our past, much like old generals are always prepared to fight the last war. The assumption of negative correlation ensures most portfolios are constructed to protect against a sharp rise in market volatility. Or, at least, a systematic event where the pre-programmed response by algorithms and investors alike is to sell risky assets and move to government bonds; in which the implied expectation is that central banks will continue to respond by providing liquidity via even lower rates or through more monetary easing.
Alas, this negative correlation between equities and bond prices can’t continue indefinitely. In the long term, it has been found to be around 0, switching intermittently between positive and negative (Lamont, AON, 2014) as it should, dependent on cycles of growth, rates and inflation. Recent persistent and significant negativity has been propagated by central bank policies which are stretched and transitory, albeit for longer than one could have predicted. A move back to a regime in which there is no or a positive, relationship between equity and bond prices would have a profound impact on two decades of portfolio diversification. Although considered unlikely by many commentators, perhaps this is simply because so much of our understanding of how assets might behave in a crisis is dependent on it continuing. An understanding formed over a period in which global central banks have pushed rates to zero torched and defenestrated the rule book in the name of stability and created trillions of dollars, euros, pounds and yen to buy bonds, ETFs and equities. Consequently, and especially after 2008, this has suppressed risk further across financial markets. This has helped create the perfect environment for passive strategies, ensuring strong returns by combining a portfolio of risk assets with a leveraged position in longer duration government bonds as protection. A hedge which still would have had a positive expected return as rates have continued to fall and yield curves flatten.
The danger is that because government bonds have worked so well as a hedge over the last 20 years, it has invited investors to become complacent and assume that risk can be managed passively. An assumption that the correlation will hold in the future only opens a portfolio to the risk that it might not. By relying on government bonds, investors are only protected if governments again ride to the rescue and central banks continue to create currency to buy assets in the markets. What if they don’t, or can’t?
Risk is multi-faceted and evolving and the suppression of volatility has not eliminated it, merely pushed it into the future, making it more difficult to predict where, or more pertinently, when, it will return. However, return it must. Demographics, high valuations and higher debt levels all point to the potential for a very different risk environment in the future. A more volatile environment where a liquidity-driven crisis like 2008, to which the response is well-rehearsed, is perhaps not the biggest risk to capital. Consider these two scenarios: A future where equity market weakness, a recession, or a crisis isn’t met with lower interest rates, as much as volatile or rising rates driven by policy divergences between governments and central banks, themselves hemmed in by domestic politics or currency considerations. A world where some government bonds are no longer considered as ‘safe’, causing risk to become more transient and local because negative interest rates are fought rather than embraced with no political support for further monetary stimulus.
In other words, markets prone to gap and grind, behaving differently from our recent past but more like our distant past, in which what we believe will protect our portfolio might do the opposite.
“An optimist sees green light everywhere; a pessimist sees only the red stop light. The truly wise is colour blind.” Albert Schweitzer
As prudent stewards of capital faced with an unknowable future, all we can do is to limit our assumptions and be aware of potential risks. To prepare for a future in which the market takes our assumption of safety and diversification and stomps on it. One thing seems certain; we are approaching the end of an era where much of our risk can be managed using a passive hedge with a positive carry. Simply holding government bonds may no longer mitigate risk as much as being risk additive. Thus, investors will need to manage their portfolio diversification more actively and allocate a portion of capital in investment strategies which aim to be uncorrelated to the market whilst profiting from market volatility. Investors with a portfolio populated by passive or ‘benchmark hugging’ funds will either have to be more active in their asset allocation or to allocate to managers or products whose exposure is positioned to benefit from a market that differs from consensus expectations.
One solution that has seen much growth in the last few years is inverse ETFs and traded products based around the VIX. The problem is that their construction makes them primarily useful as trading vehicles over shorter periods of time, necessitating an element of market timing. In addition, as with the VIX itself, they can be both opaque and overtly complex, as well as subject to basis risk and, as anyone who invested in volatility products in 2009 realised, can have significant negative carry. Options and other derivatives can achieve the desired result and whilst the underlying mechanics are not as complex as one may fear, it does require continuous management.
Unfortunately, for many investors, the effort and perceived complexity might leave them susceptible to friendly and ‘competitively priced’ advice from the local investment bank. Another possibility then is to outsource portfolio protection to sensible managers, who have the necessary skills to actively manage a strategy that is uncorrelated and exposed to volatility, whilst offering the possibility of a positive carry. This equally applies to fixed income, where some of the diversification benefits could be maintained in more volatile and idiosyncratic markets by combining security selection with active management of duration, yield curve and geographical exposure. In the end, we must always be ready to adapt. The net result of historical relationships reasserting themselves and central banks losing omnipotence is that in the future, diversification and portfolio construction will become less passive, more challenging, more expensive and more complex. Which is to say, paradoxically, riskier.