Smart beta: The battle for the middle
Smart beta has quickly established itself in today’s investment jargon. The financial crisis demonstrated the shortcomings of market capitalisation-weighted indices while investors became ever more focused on cost-efficiency and the optimisation of their risk budgets. The result has been a sudden increase in the number of indices, strategies and products designed to offer exposures that seek to correct or even exploit the flaws in traditional benchmark indices.
Although a broad definition of smart beta is becoming harder as the range of strategies and products increases, there are common characteristics. They all begin with systematic, rules-based strategies to re-weight portfolio constituents in some other way than by market capitalisation. But while some products – particularly indices, of course – look like they are supposed to be considered an alternative way to make a passive investment, others look more like straightforward quantitative active-management strategies.
Indeed, the growth of smart beta, at least in product terms, has been so great that many now question whether all of its iterations can still be grouped under a single banner – and definition is important, not just for measurement and controlling risk, but because of its impact on investor expectations.
The most basic question is probably whether an investor regards smart beta as part of her universe of passive solutions or simply another part of her universe of systematic active strategies. Money coming into the space seems to have come from both traditional, cap-weighted passive allocation and traditional, cap-weighted-benchmarked active allocations.
Some would argue that many smart-beta indices have elements of passive products because they are systematic, rules-based and transparent; some would go further and argue that they even correct biases in the cap-weighted indices. Others would insist that even a systematic smart-beta index must be defined as active because it requires a degree of trading to maintain the underlying index and avoid style drift as market prices move, above and beyond what is required for a cap-weighted index.
At a glance
• Smart beta has proliferated since the financial crisis.
• It is so varied as to defy definition.
• Is it a response to a desire for new passive indexing?
• Or for better benchmarks for active management?
• If it is about the latter, why do investors so often maintain cap-weighted benchmarks for smart beta-based strategies?
Furthermore, some argue that much of a smart beta index’s outperformance is generated from this trading of index constituents, rather than their intrinsic value. According to INTECH research, which has based its investment strategy on this belief, systematic rebalancing back to the original smart-beta portfolio consistently buys low and sells high, exploiting the short-term price movement of stocks, which is due to natural volatility rather than fundamental data or events. For example, Intech’s research found the buy-low, sell-high trading of a strategy looking to exploit the size effect by investing in smaller CAP names, even if rebalanced only infrequently, explained all the long-term outperformance of small-cap versus large-cap indices.
“The whole idea of ‘smart’ has connotations in today’s jargon of something that does what it is meant to do intuitively, without the need for skill,” Schofield says. “However, under the bonnet there is something else at work. The main driver for the vast majority of products’ returns comes from the rebalancing necessary to maintain their weightings. That makes them inherently active because they have to be traded to be maintained. Many providers don’t realise it is this rebalancing, rather than the inherent value of the stocks in the index, that generates returns.”
“Being passive is like standing still in the road when the traffic lights have changed regardless of what is hurtling towards you. It is not a low-risk strategy”
There is also a spectrum of passivity within the range of smart beta strategies on offer. While some simply offer alternative weighting methods across an entire universe of stocks, others define a sub-universe with a particular risk or factor-based tilt. Some are delivered in highly transparent and replicable indices through very cost efficient vehicles, but others look decidedly different.
The point at which smart beta looks more like active management comes where an index is less transparent and not easily replicable. The cost will be higher as a result.
“Where it is not clear how an index is being calculated because the manager is using a proprietary black-box is where we move towards active quantitative strategies, especially where a manager has the discretion to change the strategy over time,” argues Phil Tindall, senior investment consultant at Towers Watson.
Perhaps this is the real point of smart-beta concepts: to provide a better risk-management foundation for active strategies?
That makes intuitive sense. After all, in many cases, smart-beta strategies and products are nothing new. Many, such as enhanced indexation or factor tilts, have long existed in active management and some products have simply been rebranded as smart beta.
Interest in smart beta has arguably not been driven by the concept of a new form of indexation at all. Instead, investors have often been responding to two issues brought starkly into light by the financial crisis, namely: the shortcomings of the cap-weighted benchmarks against which their active managers’ performance is judged, and the excessive fees they have been paying those active managers.
Many of those managers faced criticism that they charged alpha-level fees for beta risk and return in the wake of 2008. Competition has subsequently driven down the cost of purely passive cap-weighted exposure, active managers are being forced to focus much more on genuine alpha, and a cost-efficient lucrative middle ground has emerged for strategies that lay their foundations in cost-efficient, systematic, rules-based alternative-weighting strategies, to enhance purely passive exposures.
“When we first started looking at things like fundamental indexation, investors were still paying 30-40 basis points versus around 10 basis points for cap-weighted exposure,” says Tindall. “Now the difference is only a handful of basis points.”
The smart-beta question – indeed, the beta question as a whole – is critical, given the increasingly large passive exposures institutional investors globally have built in recent years. Figures from ETF provider iShares show cumulative net new flows into equity and fixed-income index funds have totalled nearly €1.5trn since the onset of the financial crisis in 2007, while active equity and active fixed-income fund net new flows have been effectively zero, driven by significant outflows in 2008 and 2011.
And yet, particularly since the financial crisis, it has become almost reflexive to believe that cap-weighted indices are not optimal. They are inherently impacted by herding; they are over-exposed to over-valued constituents and they offer no downside protection.
“Many organisations look at cap-weighted benchmarks as a hugely inefficient way to harvest beta,” says Tim Gardener, global head of consultant relations at AXA Investment Managers. “Being passive is like standing still in the road when the traffic lights have changed regardless of what is hurtling towards you. It is not a low-risk strategy. Behaviour distorts market capitalisations. It tends to reward those that have done well historically, but to harvest beta means allocating to those that are more likely to do well in the future.”
“Calling a group of strategies ‘smart’ implies they will always outperform a cap-weighted benchmark. Our research has proven this is not the case”
Smart beta is about eliminating risk where it is not getting rewarded, agrees Beltran Lastra Aritio, a portfolio manager at JP Morgan Asset Management.
“At the peak of the tech bubble, 60% of the risk in the S&P 500 was attributable to tech stocks,” he observes. “Financials are currently 45% of the MSCI World. If an investor wants that, fine, but it should be an explicit view by the investor. That decision about whether to take such a long position shouldn’t be outsourced to the market.”
In this way, smart beta allows investors to remain in an essentially rules-based world while gaining more control of their underlying market exposures. And if smart beta is really about better benchmarks for active management rather than creating better market indices, that greater level of control is not only natural but desirable in the quest for more active risk-taking in active-management mandates.
But, as that suggests, with control must come ownership and, in turn, career risk. This can present a conundrum in terms of appropriate measurement of smart-beta strategies as well as how they are sold and managed. The conundrum arises from a well-entrenched tendency to compare smart beta performance against cap-weighted benchmarks – even though one might assume that active strategies that take as their starting point some kind of smart-beta concept should be benchmarked against the relevant systematic smart-beta index.
The practice is not as counter-intuitive as it sounds, suggests Alain Dubois, managing director and head of new business and product development for index at MSCI. “Investors should compare the performance of smart-beta strategies to market-cap indices as they are a natural starting point, represent the global opportunity set and are macro consistent,” he explains.
But doing this can pose significant problems for both investor and manager. When smart beta strategies are outperforming the cap-weighted index, all is well. In falling markets, or where a cap-weighted benchmark has a strong, unrewarded long position, the question of tracking error becomes almost irrelevant.
“The risk for investors is not tracking error, but loss of wealth,” as Gardener puts it. “If the cap-weighted benchmark falls 30%, it is not good to have zero tracking error.”
However, there will inevitably be potentially lengthy periods when smart-beta solutions underperform – usually in times of market strength.
“Many providers don’t realise it is the rebalancing, rather than the inherent value of the stocks in the index, that generates returns”
“Calling a group of strategies ‘smart’ implies they will always outperform a cap-weighted benchmark,” suggests Ursula Marchioni, head of iShares EMEA Equity Strategy. “Our research has proven this is not the case.”
Different strategies suit different market outcomes. Each tilt offered by a smart beta index is likely to generate outperformance in certain scenarios and underperform in others. In a period of underperformance, ignoring risk controls versus cap-weighted indices is more difficult, for both investor and manager.
“Ultimately, somewhere in the investing organisation someone is very likely to be judging smart-beta performance relative to cap-weighted indices,” says David Schofield, president of Intech’s international division. “It would be foolhardy to ignore the risk controls relative to that index. For many providers of smart beta, the steps taken to control those risks are either non-existent or very limited, which is dangerous if investors care about the performance of the portfolio relative to cap-weighted indices.”
Schofield argues that providers should consciously and deliberately manage the risk relative to cap-weighted indices and limit the length and magnitude of underperformance.
Comparison to cap-weighted benchmarks is potentially a problem for providers not just because it could mean assets are less ‘sticky’, but also because, so far, few smart beta strategies go out of their way to control their tracking-error risk relative to their cap-weighted benchmarks. It is an issue that is becoming more important as equity benchmarks, in particular, continue to do well.
Smart-beta strategies have the potential to underperform the cap-weighted benchmark for longer than trustees’ tolerance lasts. Although they are long-term investors, justifying a three-year period where an investment is underperforming is increasingly difficult. As a semi-passive strategy, the underperformance can feel even more acute given the lower cost of cap-weighted indices.
“When we first started looking at things like fundamental indexation, investors were still paying 30-40 basis points versus around 10 basis points for cap-weighted exposure. Now the difference is only a handful of basis points”
“Over a market cycle, if a strategy can’t beat the market-cap index, then it has failed because tracking the market-cap index remains the cheapest option,” as Axa’s Gardener puts it.
Ultimately, smart beta has probably been as successful as it has precisely because it can be adapted to solve both of the problems identified in this article.
As Phil Edwards, European director of strategic research at Mercer, says: “Smart beta is interesting for investors wanting to reduce costs in their equity portfolio, but retain some of the biases they might otherwise access from active managers; or for smaller investors wanting to improve their passive exposure without significantly increasing complexity and costs.”
What really matters, though, is whether smart beta indices perform in line with investor expectations, which not only depend on which of these two things they think smart beta is doing for them, but which are still closely tied to how cap-weighted benchmarks perform. Tolerance of underperformance will be limited in an increasingly cost-conscious world.