According to a new report by ETFGI, smart beta equity ETFs/ETPs listed globally, gathered USD53.7billion in the first 10 months of 2015. There were 764 smart beta equity ETFs/ETPs, with 1,336 listings and assets of USD399billion from 106 providers listed on 31 exchanges in 27 countries.
"If you look at some market studies," says Konrad Sippel, Global Head of Business Development at STOXX, "people are forecasting that by 2020, around 30 per cent of investor assets will be in smart beta products. Currently that figure is around 5 per cent or less so there is a huge growth potential for the industry as investors become more familiar with the concept and more funds and indices gain three-year track records or longer.
"Right now, we have EUR6billion of assets in smart beta ETFs, which makes up about 25 per cent of all European smart beta assets and 10 per cent of total assets linked to STOXX market indices (7,000 in total). If that 30 per cent forecast is realised, it offers plenty of room to further expand our business."
In recent times, smart beta products have caught the attention of investors allocating to active managers. Although many products would be regarded as passive, the ability to re-weight indices and build customised products based on, for example, a dividend tilt or a growth or momentum tilt, has allowed smart beta to fit the needs of active investors, who might previously have allocated to systematic managers.
"Investors realise they can take some of the skills and sophistication of active managers and use them in a systematic way. That has resonated with investors for two reasons: they get more control over their investments, and they are able to chase lower cost solutions. This is important because not all investments into active strategies have necessarily delivered what many expect of active returns; they've been too closely correlated to traditional market cap passive strategies," comments Howie Li, Co-Head of Canvas at ETF Securities, one of Europe's leading ETF providers.
At Mellon Capital, custom beta, where the team tailors exposure to specific factors based on client requirements, plays an important role in its investment universe.
It currently has three multi-factor smart beta strategies in its roster that invest in i) the US, ii) Global Developed Markets; and iii) Global Emerging Markets.
"We also offer a number of high-dividend strategies that may be viewed as smart beta. Most smart beta strategies follow a clear rules-based approach to portfolio construction, and provide simple exposure to well-understood market anomalies (Value, Small Cap, Quality, etc.).
"Although we still believe that fully active strategies can produce stronger risk-adjusted results over time, an increasing number of investors prefer to invest in a cheaper alternative. These investors are willing to sacrifice potential alpha by investing in smart beta strategies rather than fully active ones," says Mellon Capital.
Sippel says that STOXX regards smart beta as anything that takes a strategic approach to tilting traditional market cap weightings towards a specific investment objective.
"We have a lot of indices that focus on dividend investing, generating high dividend yields, we have products in the low risk space to reduce volatility as well as factor-based minimum variance products. More recently, we've developed increased combinations of the high dividend, low volatility theme and also themes such as quality, strong balance sheets, ESG, which all fall under the smart beta wrapper, to some extent," explains Sippel.
Single versus multi-factor investing
By far the widest selection of smart beta products resides in equities. This is understandable given the depth of liquidity in global equities, the sheer breadth of market indices, and the fact that equities lend themselves to rules-based investing.
When smart beta first emerged, it started with single factor equity products. The primary driver for this was demand from institutional investors for tailored indices and an ability to generate returns based on isolating specific risks in the market to harvest something a little `smarter' than pure market beta.
"Someone who is less sophisticated and less attuned to equity investing is better off starting with a multi-factor product if they are moving away from trusting an active manager to smart beta. As they get more confident, that's when they may consider switching in and out of different single factor strategies at the appropriate time.
"People need to realise that single factor is not necessarily an all-weather approach. You've got to know when to use a single factor and why you are using it. Single factor equity products have been the first major wave of smart beta, but they require investor education," says Li.
When looking at smart beta, the uninitiated investor is advised to take baby steps. Rather than invest in single factor products, of which there are countless choices in the market, they should start by taking a multi-factor approach. This will allow them to understand how different factors respond in specific market conditions.
"It gives them time to understand how value, or momentum, or size or low volatility, works within a multi-factor strategy. The next step could then be to integrate certain single factor products into their portfolio when the understanding of how each factor behaves increases," adds Li.
Low volatility & European quantitative easing
One firm that prides itself on leading the ETF revolution, and which offers a wide range of single factor smart beta ETFs, is Invesco PowerShares. "Having single factors as building blocks is really important," says Dan Draper, Managing Director of Global ETFs at Invesco PowerShares.
With divergent global monetary policy, the US ending quantitative easing just as Europe enters phase one, there has been a trend among investors to shift away from US equity ETFs towards non-US equity ETFs.
As a result, in May this year, the firm launched PowerShares Europe Currency Hedged Low Volatility (Ticker: FXEU), which is hedged back to the US dollar.
"In 2011, we launched PowerShares S&P 500 Low Volatility ETF (SPLV), and with approximately USD5billion in assets it has become one of our flagship products. What was interesting to us was that by taking broad-based US equity exposure, and adding a low volatility tilt, during the period when the US was going through QE, led to a period of noticeable outperformance; the combination of low volatility in the face of QE really worked well.
"Our thinking with FXEU was to introduce a product, whilst Europe is still in the early stages of QE, and apply the same low volatility factor as we had done for SPLV, hedging it back to the US dollar. We've already raised USD170million. The fund is up approximately 3 per cent YTD. Many similar products in that category, which do not use a low volatility factor, are in negative territory over that same time period," explains Draper.
Multi-factor products
Multi-factor products allow the investor to rely on a strategy that takes into account different economic cycles; these are products that know when to put more emphasis on value companies, or when it is more beneficial for momentum investing etc.
As mentioned, not all investors have the sophistication to know which single factors to pick out, such as low volatility in FXEU, which Draper confirms is aimed at sophisticated institutional investors who want to implement single factor strategies.
Fund sponsors are therefore building multi-factor products to help investors and give them more of a `ready-made' strategy.
"There are fewer multi-factor products in the market but they are growing," says Li. "The interesting thing about multi-factor is that the dynamic model, the engine behind it, matters much more in terms of how the product provider differentiates themselves compared to a single factor product; there are only so many ways you can apply value or momentum to a single factor FTSE 100 product. There's probably a bit more room for differentiation with respect to the dynamic allocation in a multi-factor approach."
Multi-factor products, with their exposure to a number of different and complementary factors, will generally give a more consistent set of risk-adjusted returns over time.
"Single-factor products, however, may also have an application in investors' overall asset allocation process. For example, a single-factor smart beta product may be used as part of a completion strategy in order to lend more exposure to lower beta stocks to an equity portfolio with a higher risk profile," explains Mellon Capital.
Dual factors
More institutional clients want single factor ETFs to use as building blocks, once they have familiarised themselves with the benefits of smart beta. According to Draper, institutional investors typically adopt a core and satellite approach, using multi-factor in a core portfolio and overweight or underweight single factors in a satellite portfolio.
"We've also looked at single factor strategies that have worked well, such as high dividend, and combined them with SPLV (low volatility) to create a product (Ticker: SPHD) that uses just two factors to minimise the value trap potential; something that one can experience in high dividend ETFs in certain times in a market cycle. That's what's really interesting about smart beta. Once you've got a wide range of single factors, you can start thinking about optimising combinations to create new solutions for investors," confirms Draper.
Fixed income smart beta on the rise
Although there are far fewer smart beta products in the fixed income space, innovative providers such as London-based ETF sponsor, Source Ltd., are steadily building a range of fund options for investors.
When looking at fixed income, the main risk factors at play tend to be duration risk and credit risk.
"Generally, if an investor wants to invest in sovereign bonds of the main developed economies (US, UK, Germany, Japan) they would have exposure to mainly duration risk. If they want to diversify away from duration into credit a first step would be to allocate into a corporate investment grade bond product. If they want to overweight credit risk further, they could allocate to a high yield product.
"We try to launch products that have a clear investment thesis behind them. They have to be products we believe in and which we think investors should be considering; that's why we've only launched eight fixed income products since 2011," explains Fabrizio Palmucci, Fixed Income Specialist at Source.
Palmucci confirms that assets have noticeably accelerated since 2013. Across the eight fixed income products, assets currently total USD6billion, averaging USD700million per fund and underscoring the success that Source has had with each launch.
"The PIMCO EUR Short Maturity ETF has close to EUR3billion and is now the single biggest actively managed ETF in Europe. The fund is actively managed by PIMCO and invests primarily in short-term investment grade debt. The average portfolio duration will vary based on PIMCO's economic forecast and active investment process, and will not normally exceed one year."
Short Term High Yield has also been a success story at Source. The premise it offers to investors is that if they are investing in the short-term high yield segment, Source can offer a similar yield on a portfolio with a 1- to 5-year duration profile to that with a 1- to 15-year duration.
"The way we structure the product is innovative and efficient. We have just launched a GBP-hedged share class of the Short-Term High Yield Corporate Bond Index Source UCITS ETF," confirms Palmucci.
Two of Source's most recent product innovations, both of which launched at the end of 2014, invest incorporates, with a short duration bias – PIMCO Low Duration Euro Corporate Bond Source UCITS ETF and PIMCO Low Duration US Corporate Bond Source UCITS ETF.
Both are active strategies. The `smart' component is that the strategy focuses on low duration bonds, giving a better risk/return profile given where interest rates currently are.
"We believe rates are likely to rise soon in the US and even though that's not necessarily the case in Europe for short-term rates. We don't see a lot of upside in longer duration, which has historically been correlated with US rates.
"In our view, what is smart about these products is that, first of all, the majority of assets in Fixed Income ETFs are in broad duration; approximately EUR17billion in total assets. When investors think about exposure they take the broad benchmark, which includes all maturities. Lower maturity and duration typically mean lower yield.
"However, because our ETF is active, we supplement this lower yield through credit selection. The overall effect is to generate a yield that is similar to the broader index, but with less duration risk. This is smart positioning in the context of where rates are today," explains Palmucci.
Minimum variance methodology
For the last three years, STOXX has been running a series of Minimum Variance Indices, giving investors exposure to 24 different geographies. There are two versions: one constrained, the other unconstrained.
The constrained index version is based on a composition that is similar to the underlying index, but possesses a lower risk profile. This index version enables market participants to closely follow a certain benchmark index and still benefit from a risk optimised portfolio. The unconstrained index version might differ more strongly from the underlying index, while offering the potential benefit of an even better risk profile.
"Whichever region, or index version you look at, you will see a typical pattern, which is that returns fall more or less in line with the standard benchmark but with a significant reduction in volatility. As a result, all of our Minimum Variance Indices have a higher Sharpe ratio; in other words, providing a similar level of return at a significantly lower level of risk," says Sippel.
The objective of Minimum Variance is to provide access to the respective markets by varying the weights of the stocks of the underlying broad indices in such a way that the overall portfolio of the new index has the lowest possible volatility.
In order to achieve robust results, a covariance matrix that is estimated using a fundamental factor model developed by risk software specialist, Axioma, is used.
"With a Minimum Variance approach, what you are trying to do is predict the optimal portfolio. This relates to Markowitz's Modern Portfolio Theory where if you take a set of securities and you simulate every possible portfolio scenario, you end up with a line, called the Efficient Frontier. Any portfolio across that line has the most efficient risk/return structure," says Sippel.
Of course, nobody has a crystal ball. This requires using historical data to predict where those portfolios could lie in the future. Academically, it has been shown that by using historical data it is easier to predict future volatility, as well as the correlation among individual stocks.
Minimum Variance uses a covariance matrix to measure correlations of every pair of securities within an index. To reduce the complexity of the calculation, the returns of securities are broken down into various factors to come up with an optimised portfolio composition. This is done using a fundamental factor model developed by risk software specialist, Axioma
"To do that you need a good factor model and that's why we partner with Axioma. They've got one of the leading factor models and optimisers in the industry. The factor we are looking to optimise against, in this case, is minimum variance (low volatility)," adds Sippel.
Maximum variety methodology
Since 2005, Pierre Filippi, CEO of Paris-based Fideas Capital, has been refining and applying a unique smart beta methodology to equity and multi-asset portfolios. Referred to as "Maximum Variety", it aims to minimise volatility by focusing on individual risks within assets that are weakly correlated.
Fideas believes this approach optimises the risk reward properties of the portfolio by effectively reducing the global risk of the portfolio without having to miss out on the high return potential of volatile assets.
The strategy, which exists in two portfolios – Betamax Europe, focusing on European equities and Betamax Global, a multi-asset growth fund – has been catching the eye of institutional allocators of late. France's incubation fund, Emergence, and its investment adviser, NewAlpha Asset Management, recently announced that it was investing EUR35million in Betamax Europe. Over the last two years, assets at Fideas have doubled to more than EUR400million.
What makes Maximum Variety interesting is that it works in a very different fashion to Minimum Variance.
Both smart beta strategies use the concept of diversification to reduce risk, but whereas Minimum Variance does this by tilting the index towards low volatility stocks, Maximum Variety weights risk in such a way that it still gives exposure to higher risk stocks, whilst still limiting volatility.
The formula on which the Betamax algorithm runs is to maximise the weighted average volatility of assets divided by the volatility (ex ante) of the portfolio. Assets are measured by their risk rather than by their quantity i.e. if an asset is more volatile there is less of it in the portfolio and vice-versa.
"If one compares this methodology to a dominant risk-based smart beta methodology such as minimum variance, there is a major difference. With minimum variance, you are only taking half of our formula, which is to maximise the asset's weighted average volatility divided by ex ante portfolio volatility.
"By doing this, there is an inherent preference for low volatility assets. As a consequence, the outcome of the portfolio is mainly based on a low volatility anomaly i.e. that low volatility assets are capable of producing a good return compared to high volatility assets.
"We do not base our formula on this negation of the classical risk-reward equation of the market. Our formula means that we still seek to buy risk. Instead of looking mostly at low volatility assets, we balance assets by factoring all the potential risks that you can buy in the market. We might look at growth versus value, small-cap versus large-cap, high beta assets versus low beta assets. We try to capture everything in the market," explains Filippi.
Contrary to multi-factor strategies, that try to balance all five risk factors into a single portfolio, the Maximum Variety strategy runs on pre-determined performance factors. Different risk factors are active in the market at certain times, and silent at other times.
"The Betamax model is attracted to those risk factors that are active in the market now. Not over a period of time. Currently, two types of factors are very active in the market. One is energy, the other is small-cap stocks. These two factors are well represented in our portfolio," confirms Filippi, who confirms that around 60 per cent of the global portfolio is held in equity and commodity assets.
One important aspect of Maximum Variety is that allocations are made at the sub-index level, unlike most quantitative smart beta strategies that work on a stock-by-stock basis. This approach, says Filippi, avoids the need to identify and manage stock-specific risk. "By using sub-indices, it is easier for investors to understand the biases used (i.e. Norway would be indicative of an energy bias). This is very hard to determine in a smart beta portfolio that looks at individual stocks within a single index," adds Filippi.
Fideas has just announced a merger with credit specialist, Rivage Gestion. Over the coming years, Fillipi hopes that the Maximum Variety methodology will be successfully applied to credit markets, thereby widening the suite of Betamax products.
"We see further growth in the secular smart beta ETF industry. We estimate that these products will grow at around 25 to 30 per cent over the next three to five years," concludes Draper.
