Is this the End of Risk-Factor Investing? Or Only the Beginning?
To view this article in digital format, click here (http://www.ai-ciodigital.com/ai-cio/june_2013#pg78).
Latin was born in Europe a couple thousand years ago. Its pure form was used by only the best-educated in society. It was a way of communicating between the clergy, royalty, and administrative classes, and was the subject of huge scholarly projects that are still on display today.
The general public had no access to learning this tongue, and the snippets they did catch were irrelevant without the grammar and theoretical background essential to its comprehension. Latin's benefit to them was minimal, save the efficient running of the states in which they lived.
Compare that history to this one: Risk-factor investing was born in Europe a couple of decades ago. Its pure form was debated by the best-educated in society. Its pioneers were academics who had limited access to real investment but thought there must be a more efficient way of running money. Libraries of academic papers have been written on the subject, but the general public-and most institutional investors-have little grasp of its meaning or why it might be relevant to them.
With me so far? Good.
What happened next to Latin is key to understanding the current evolution in risk-factor investing.
Through various channels-the church, town administration, general need for efficiency-Latin began to spread across Europe. In towns and villages from Brindisi to Londinium, marketplaces buzzed with nominative nouns and ablative clauses that had once been the reserve of their leaders.
However, after barely a couple of decades, someone from Brindisi would have struggled to be understood in Londinium. This is because the language of each populace had adapted to its own needs. It had evolved. The ruling classes soon cottoned on and Latin in its purest form faded into the background.
Risk-factor investing is going through a similar process. What started as a theory debated in lecture theatres and symposiums has been adapted by those who have looked at it. Some stuck close to the original theory but have attempted to bring it up to date; some have taken the basic thesis and broadened it out; and others have chosen to loosely base their portfolio construction on the idea, as they think simply looking at numbers does not give a full view of the world.
Just as the new European languages had the same root, the original idea for these risk-factor evolutionists is the same: Identifying your exposure is the most important part of the job. Without it, you are blindly allocating assets. This can quickly get you in trouble-as it has done in the past.
Essentially, this group-now spread across the continent-has turned accepted theory on its head.
Jaap van Dam, managing director for strategy at the [euro]140 billion Dutch pension management company PGGM, was one of the first practitioners to realise there was a problem in the late 1990s.
"If you look at the world through an investment category lens, you think you have to invest in a range of asset classes, as they all have different names. But if you look below the surface there is more commonality. There is something more fundamental to notice; there are common factors," he explains. "Just looking through an asset-class lens, you might think you have a well-diversified portfolio, but when you consider what is driving performance, this is not the case. We thought about considering risk and agreed that it should be a bigger part of our solution."
The acknowledgement that assets are instruments allowing investors to access risks-and avoid them-is fundamental to this way of thinking.
"All asset classes can be viewed as structured products," says Pieter van Foreest, head of client risk management at APG in the Netherlands. "You have to look at what it does in your portfolio for interest rates, credit spreads, currency, etc. Credit is a single asset class, for example, but it has several risk components."
It is these components that the academics identified back in the 1960s but by way of models and technical languages that would have cut no ice with mainly lay investors. A lack of urgency had a part to play in the general ignorance of the new ideas, too-pensions were almost universally well-funded.
"In the 1980s, a lot of pension funds were invested in government bonds, so they didn't really have a diversifier, such as equities," says Jose Manuel Suarez, a director and adviser to several Dutch pension funds. "Coming from the 1980s, when we had bond yields of 5% to 7%, it was easy to match liabilities and long-term interest rates, so when the need increased to get outperformance in the 1990s, the introduction of the diversified portfolio arrived and institutional pension funds came to a different view of investing."
But the risk-factor model was still in an inaccessible form for most. Van Dam, van Foreest and others shuffled through the various academic papers for inspiration.
"You could do it in a purely scientific way," says Suarez Menendez, "but then you have to get into the market and explain to the world that this factor needs to be calculated this way and looked at that way. For trustees, it's very difficult if they cannot physically see the risk factors. We needed something that was perceivable by the market."
In the most sophisticated pension investment offices up and down the European continent, these pioneers attempted to bridge the gap between theory, in which they saw tremendous value, and practical implementation they could take to their trustees.
"If we were being purely scientifically-led, then it would never land," says Suarez Menendez, "as it would be an academic discussion between academics on how risk factors are modelled."
There are some who stick closely to the original thesis-visit Vatican City and you'll find signs written in Latin-but they may have a tough time convincing their trustee board to follow their lead.
"True believers say it is the only way to look at the world," says van Dam, "but at some point they are going to have to recognise that there's more than 'factors' and 'risk' only, and you can't bring risk down to one or two numbers. Risk is a multi-dimensional thing and it should be looked at in a multi-dimensional way."
The investor sticking closest to the original scholastic method, it could be argued, is the Danish state pension provider ATP. The "risk buckets" it introduced in 2006 created a stir in Europe-and are still too complex for many to try to unpick and recreate.
For Anders HjA|lmso Svennesen, co-CIO at the fund, however, its approach is a mixture of art and science rather than strict number crunching. "We have to identify which factors exist and make sense of them. Each asset class has different risk premia, so we look at what we want-what we think we will be rewarded for holding-and hedge out the rest," he says.
He agrees that ATP's approach has very direct links with the original theories formed in the 1960s and 1970s but argues that it has developed and adapted the approach to its needs. "The big difference is the huge expansion of the derivatives market; there are many ways to hedge out risks that just weren't available 30 years ago," he notes. Swaps, hedges, shorting facilities, futures, options, puts-the list of instruments a sophisticated investor can apply seems to grow every day, but there remain gaps for the Danes.
"To invest in the way that we do, you need to have the right instruments," he says, "and we don't have them all yet, but they are developing. What is also essential is having a board that understands the process and allows you to use them; if they don't, it is 'Mission Impossible'."
And not all boards do-or have had the opportunity to do-which is why risk-factor investing is evolving differently across the continent.
ATP has left the stock-picking of companies (for bonds, equities, and other investment instruments) largely to external managers. These third parties-who are hired on a total return basis-analyse the characteristics of a firm and feed it back to the pension's investment team, which draws out the relevant risks of the asset and company-specific characteristics.
PGGM, on the other hand, is moving in the opposite direction.
"We are starting to think that the financial sector has moved away from the real world in the last two decades and all the serious theory that now shapes the paradigm really doesn't connect the two," says van Dam. "We feel that's not good enough."
Politics, human behaviour, and other uncontrollable events all affect investment outcomes, so PGGM is "coming back to earth" to get closer to what is really happening.
"Ten years ago, we would have said: 'We don't care; we're buying a stream of beta and that's it.' I think we'll go back a bit toward a fundamental view of the world," concludes van Dam.
There is one thing these investors do have in common, however: None are trying to shoot the lights out with performance. "It's not about making super-stellar returns," says Suarez Menendez, "it's about getting your outcome more in line with expectations you've expressed beforehand."
Monitoring third-party managers remains important, however, as, if they are underperforming the market, you are paying them to lose your money-risk-factor based or not.
It is in this community that the next stage of the evolution will have to occur, according to van Foreest. "You may have asset managers who are very good at finding instruments, but they don't know what exposure you want. This happens a lot. They want to sell you a product, promise high returns-and it will probably be actively managed, as the fee is higher-but what does it add from a portfolio or balance sheet perspective? Board members have to ask about whether risks are rewarded or not, and I think we are going to see a lot of changes."
In the UK, which has a larger number of smaller pension funds, progress has not been as rapid. "I extoll the theoretical virtues of risk-factor investing," says Chetan Ghosh, CIO of the [euro]8 billion Centrica Combined Common Investment Fund. "But it is almost impossible for us to implement as some of the larger funds do in Europe-it pervades their very investment philosophy and for them it makes sense." According to Ghosh, size is a stumbling block for smaller funds without tremendous internal investment capabilities. This doesn't mean he has rejected the idea entirely. "Instead, we use risk factors to provide another lens to appraise risk and identify any overexposure."
So risk-factor investing is not dead-nor is Latin, for that matter. They are both simply evolving to their users' needs. What is going to be interesting is seeing how each protagonist pushes out his new version and which stands the test of time.
"There will never be just one model that works-in fact, all models are wrong as we live in the real world," says van Foreest. "We have to adapt to what is the greatest priority and what is not."
These adaptations are not the end of the theory heralded as risk-factor investing; they are just the beginning of the next chapter. Or Ormutantur omnia nos et mutamur in illis, as they say in the Vatican.