Indices & Benchmarks - January 15, 2014

Let’s stop saying anything and everything about smart beta - an interview with Noël Amenc

In this month's interview, we talk to Noël Amenc, Professor of Finance at EDHEC Business School, Director of EDHEC-Risk Institute and CEO of ERI Scientific Beta, about the reasons behind EDHEC-Risk Institute's creation of the Scientific Beta platform, the concept of smart beta and the current state of the market for index provision.

Noël Amenc

In April 2013, EDHEC-Risk Institute launched an important initiative in the area of smart beta with the creation of the Scientific Beta platform. What is your assessment of this operation at the beginning of 2014?

Noël Amenc: Even though the Scientific Beta platform was only set up recently, we can consider the launch to be a success because the platform has been very warmly welcomed both by asset owners and asset managers.

Today there are more than 3,700 users of the platform, who employ Scientific Beta indices either to benchmark their investment management or smart beta investment, or directly by replicating the indices. There have been more than 24,000 visits to the www.scientificbeta.com website since it opened.

The philosophy of the Scientific Beta platform is quite appropriate for mandates because it involves allowing investors, as part of the Smart Beta 2.0 approach, to make their own choices of risk factors and diversification methods to construct a benchmark which corresponds not to embedded risk solutions imposed by the index provider, but instead to a benchmark that is representative of the investor or manager’s risk allocation decisions. Nonetheless, our smart beta investment approach based on transparency and a very competitive economic model has given rise to interest from ETF providers and the year 2014 will see the first ETFs replicating Scientific Beta indices listed both in Europe and in the US.

But do you not fear that EDHEC Risk Institute, through this initiative, might lose its neutrality and therefore its ability to influence and the credibility drawn from its academic status?

Noël Amenc: This question is fundamental and in fact is part of the DNA of EDHEC-Risk Institute.

We are convinced that management research in general, and financial research in particular, is only of interest if it influences practices. Management research was established as a reaction to excessively theoretical approaches to economics. There is a risk today that researchers in finance, driven by the legitimacy that they think derives from increasingly formal and quantitative research, will forget the very purpose of research in finance, namely its usefulness and the application of the results.

The dictatorship of the “publish or perish” model, which is the key factor in academic careers today, is in our opinion detrimental to the relevance of the research. It does not give sufficient value to the efforts that researchers make to popularise their research outside the circle of their peers and academic journals that are not widely read by investment professionals. It ignores application of the research. Speaking of applied research in business schools or university is still dimly viewed. Even though the reality of the Nobel Prize is that it often highlights the originality and the importance for society of the researcher’s work without seeking to elect the person with the largest number of publications, it appears that the scholasticism of the rankings dominates academia.

Naturally, this de facto withdrawal of academic research from real life applications leaves room for what is termed “practitioner” research, but the approach to this research is often much too commercial. For example, in the area of smart beta, it should be observed that the articles published are written almost exclusively by authors who have a commercial interest in publishing and in supporting a thesis that is favourable to their business or to that of their employer, and who sometimes do not hesitate to use all of their skill and scientific capabilities to justify practices or messages that are less scientific.

EDHEC-Risk Institute was set up as a response to this two-sided nature of management research. For the 95 people that work within the institute, it involves implementing a collective research project, the goal of which is to serve the investment industry with the highest level of scientific rigour. We aim for both academic qualification and business relevance.

Unlike the populist opinions that, notably in Europe, dominate the thinking of political leaders, we think that the development of finance and its technical sophistication is a good thing for the economy and for society as a whole. All serious academic research shows that restrictions on financial activity contribute to an increase in the cost of capital that is detrimental to economic growth and employment. The challenge is not therefore to reduce the role of finance but to improve its practices. That is what guides EDHEC-Risk Institute’s actions. For more than 12 years we have trained thousands of investors and managers in our executive seminars, hosted tens of thousands of investment professionals at our conferences and distributed our research to millions of readers with the sole aim of producing an improvement in investment industry practices. And it is for that same reason that we set up ERI Scientific Beta last year.

It is because we judged that the industry’s practices in the area were not satisfactory and because, in spite of our publications, we observed that the market’s smart beta offering was both poorly designed and sold in an unsatisfactory way, that we intervened.

The limitations of smart beta therefore justify the existence of ERI Scientific Beta. Is that not contradictory with your desire to make ERI Scientific Beta a leader in the provision of smart beta indices?

Noël Amenc: Not at all. We think that the truth always triumphs in the long term and brings value not only to the market but also to those who seek to speak the truth.

In the area of smart beta, people should stop saying and doing anything and everything.

The idea that underlies the smart beta approach consists in recognising that even though cap-weighted indices remain the reference in measuring the performance of the market, they are not necessarily the right choice for every investor. It is therefore highly regrettable that the practical application of the approach in numerous investment management offerings or smart beta indices does not live up to the conceptual and practical challenge, which is to respond effectively and honestly to the limitations of cap-weighted indices.

In concrete terms, what is your principal reservation in the area of smart beta offers or messaging?

Noël Amenc: Our main reservation is that providers have organised themselves in such a way that there is no real criticism. The debate on transparency today clearly opposes EDHEC-Risk Institute and the vast majority of index providers, who consider, in spite of the recent scandals in the area of market references, that the question of the transparency of index methodologies and track records is a non-issue and, above all, that investors are not asking for that transparency.

We will soon be publishing two studies1 that show that both in North America and in Europe, an overwhelming majority of investors are asking for this transparency and are dissatisfied with the information (or lack of information) supplied by their providers. Index providers are not in favour of this transparency and consider it to be pointless. Of course, if it involves convincing a significant business prospect, the index provider is prepared to organise carefully defined transparency by giving the prospect access to data that is subject to a strict confidentiality clause, but it is out of the question for a researcher or a competitor to have access to the details on the methodology, or even to the historical compositions of the indices so as to be able to carry out independent research or offer criticism that would be useful to the investor community.

While most index providers use their researchers to seduce investors and publish in scientific journals to make their track records and methodologies credible, these same firms refuse to provide free access to the data that underlies their research. Ultimately, the promoters of smart beta want to be seen to be providing a response to the inefficiency of the securities market, but refuse that an efficient market for indices, which would necessarily be based on transparency, be created.

As part of our discussions with our business prospects, how can we say that cap-weighted indices are too concentrated and poorly diversified if we do not provide the composition of our own smart indices, which would allow the effective number of stocks2, or the GLR concentration measure3, to be calculated with complete independence when we are presenting them as a solution to poor diversification?

More generally, how can we publish simulated track records for indices that have been set up recently and are based on fairly complex methodologies without letting the market access information that enables them to challenge these track records and to independently analyse the risks that underlie the published performances or outperformances?

The term “smart beta” has become very popular, to such an extent that some consider it to be more of a marketing slogan that is often a source of confusion for investors. Do you share this opinion?

Noël Amenc: We do indeed have considerable reservations about the confusion that is perpetuated on the very concept of smart beta.

There are many sources of confusion and a detailed description of these would fill a book rather than an interview. I would however like to draw the reader’s attention to three of them. Firstly, it should be recalled that the concept of diversification is the basis for the very existence of smart beta. It is because cap-weighted indices are reputed to be poorly diversified that the demand for new better-diversified benchmarks, commonly known as smart beta benchmarks, has emerged in the first place. In that respect, not all alternative forms of weighting can be qualified as smart. The smart nature of a weighting is not self-proclaimed, but comes from the concrete fact that the weighting, through its objective and methodology, explicitly targets diversification, or at the very least, deconcentration. As a result, speaking of fundamentally-weighted indices as smart indices is a confusion in meaning. The weighting of stocks through fundamental characteristics does not in any way take into account the relationship between the stocks and the objective is not therefore to diversify the benchmark but to represent the market in a different way (alternative size measure index) or to offer a less volatile proxy for the Value factor. Ultimately, this alternative weighting scheme, ex-ante, targets neither deconcentration nor diversification of the benchmark. It is therefore not surprising to observe, ex-post, that these indices are fairly concentrated and highly exposed to sector risks that investors do not really choose when they invest in a fundamental way.

The second confusion perpetuated on smart betas is that they are all the same and that their performance comes from the same drivers, namely a bias towards Value and, above all, Size factor exposures, as well as the rebalancing effect. In the end, according to such misleading claims, the difference between the various weighting schemes is supposed to be their relative ease of implementation, i.e. their turnover and capacity or liquidity effect. Notwithstanding the fact that the promoters of these statements “forget” to integrate the turnover control methodologies for indices other than their own in their comparison of the simulated track records of the different forms of indices, it is totally confusing to make investors believe that all indices are exposed to the same risks on the pretext that any non-cap-weighted index is necessarily less exposed to Large-Cap and Growth stocks than its cap-weighted equivalent. The Smart Beta 2.04 approach precisely allows the choice of risks to which one wishes to be exposed to be distinguished from the choice of weighting scheme that will enable the benchmark to be diversified. It is therefore entirely possible to create a smart beta index on a selection of very Large-Cap or even Growth stocks and as a result end up with a Large-Cap-tilted and/or Growth-tilted portfolios, in the event that such tilts happen to be attractive for a particular investor.

Besides, reducing the risks of a benchmark to the sole micro-economic Fama-French or Carhart-type factor exposures means forgetting that sector risks, which serve as proxies for underlying macro-economic risks, can have a strong influence on the return and the volatility of the portfolio in the short or medium term. For example, during the financial crisis, the difference in relative drawdown with respect to cap-weighted between a maximum deconcentration Value index with or without sector neutrality is noteworthy. For the Euro zone, the Scientific Beta Value index experienced a maximum relative drawdown of 20.5% over the last 10 years, compared to only 10.5% for the sector-neutral version of the same index. The sector-neutral Value index avoided the strong overweighting of financial stocks that was present in the maximum deconcentration Value index without sector neutrality constraints. Clearly, despite being exposed to the same Value factor, the two indices had very different risk exposures.

The third and final confusion, which in our opinion will lead to major disappointment for some of the most sophisticated investors in the market, is the one that is perpetuated between the concept of factor model and smart beta investing.

Investing in a smart beta index is not the same as investing in a portfolio seeking to replicate the performance of non-investable long/short factors, whose loadings are observed ex-post. Here too, not all factor indices are smart – in our view only well-diversified factor proxies can be called smart. The idea of maximising the exposure of an index to a given risk factor is in fact equivalent to constructing highly concentrated, and therefore very poorly diversified, portfolios, which, as such, produce unattractive risk-adjusted returns over the long term, and, in the short term, can be significantly impacted by exposures to specific or non-rewarded risks. In the end, the risks of underperformance compared to market indices are such that the promoters of these factor indices have often seen fit to add tracking error constraints with respect to broad cap-weighted indices, despite the fact that this is completely inconsistent with the positioning and messaging on the search for factor purity. In fact, the true challenge in factor investing is the choice of well-rewarded, i.e. well-diversified, investable proxies and the allocation between these proxies. On the same subject, another reservation that we have with regard to current smart beta practices is related to the lack of consistency of the approaches proposed. For example, the same index provider who initially promotes the usefulness of investing in the low volatility anomaly, will suggest doing so by using the minimum volatility index, even though a well-diversified benchmark constructed using a selection of low-volatility stocks would provide more efficient and more direct access to the benefits of this perceived anomaly. In the same way, if the Size effect is what is being sought, why not build a smart factor on the basis of the selection and diversification of smaller cap stocks for a given universe rather than counting on the implicit but uncontrolled exposure to the small-cap risk of a benchmark that contains all the stocks in this universe, whether large-cap or smaller cap, in the same proportions.

Do you not think that your direct presence in the smart beta market makes the reservations that you are formulating less audible?

Noël Amenc: We define ourselves as an “activist” academic research centre. As such, if we think that the market is not capable of setting up the right offerings by itself, or, worse, is moving in the wrong direction, we intervene and assume our responsibilities. We do not only want to be market observers or chroniclers, but providers of academically-robust and implementable solutions. That is what we are doing with the smart beta index platform, which aims to offer a choice of diversification strategies, and methodologies for implementing this diversification, that are well documented in the academic literature and rigorously tested. The idea is not to position ourselves as the promoter of a particular weighting scheme or index but to provide investors with the means to make informed choices on a variety of smart beta strategies.

But the scope of EDHEC-Risk Institute’s intervention is not limited to smart beta. We are also very active in other areas of asset management research. We have made freely available to the industry not only our research but also benchmark construction methods as part of an approach combining dynamic risk budgeting and life-cycle investing. This involved taking into account short-term risk budget constraints, which have increased considerably in recent years with the prudential regulations that weigh upon long-term investors, while providing access to optimal allocations integrating the investment horizon. This initiative was a response to the inadequacies of deterministic glide paths used in some current allocation offerings and notably first-generation target-date funds, which do not integrate satisfactorily either the mean reversion of equities, which itself relates to the recognition of an investment horizon, or the loss aversion of the investor, who cannot be satisfied merely to avail of an efficient long-term solution, but must also be able to respect maximum absolute or relative loss thresholds. The conviction that things needed to change in the area of asset allocation also led us to set up and make freely available Solvency II benchmarks that are aimed at reconciling an optimal long-term allocation strategy with the short-term solvency constraints introduced by the new European regulation Solvency II.

I could also cite the number of interventions that we have made in the area of regulation when we consider that the projects presented by the legislative bodies are based on a lack of knowledge of the academic realities and of the indisputable empirical results that support these realities, whether involving the negative effects of the short selling ban, the dangerous pointlessness of a financial transaction tax, or the danger of designating scapegoats such as hedge funds instead of addressing the real problems of both the sub-prime crisis and the sovereign debt crisis.

Whether it involves regulation, investment solutions, portfolio construction, risk analysis or risk management, EDHEC-Risk Institute is not content to merely publish, but takes action in order to accelerate the adoption of its research results. This is a constant and ERI Scientific Beta is one of the most comprehensive forms of our commitment to better finance.

  1. “Call for Reaction to the EDHEC-Risk North American Index Survey” and “Index Transparency: a European Perspective,” forthcoming.
  2. The effective number of stocks is a measure of portfolio diversification given by the reciprocal of the Herfindahl measure, which is itself a commonly used measure of portfolio concentration that is computed as the sum of the squared portfolio weights.
  3. Goetzmann, W. N., L. Li, and K. G. Rouwenhorst (GLR, 2005, “Long-Term Global Market Correlations,” Journal of Business 78 (1): 1-38) use the ratio of the variance of the portfolio returns to the weighted average variance of individual stock returns to take into account not only the distribution of weights in the portfolio but also the correlation properties.
  4. Amenc N. and F. Goltz, Smart Beta 2.0, Journal of Index Investing, Winter 2013

About Noël Amenc

Noël Amenc, PhD, is professor of finance at EDHEC Business School, where he heads EDHEC-Risk Institute, and CEO of ERI Scientific Beta. He has a master’s degree in economics and a PhD in finance and has conducted active research in the fields of quantitative equity management, portfolio performance analysis, and active asset allocation, resulting in numerous academic and practitioner articles and books.

He is a member of the editorial board of the Journal of Portfolio Management, associate editor of the Journal of Alternative Investments, and member of the advisory board of the Journal of Index Investing.

He is also a member of the scientific board of the French financial market authority (AMF), the Monetary Authority of Singapore Finance Research Council and the Consultative Working Group of the European Securities and Markets Authority (ESMA) Financial Innovation Standing Committee.

URL for this document:

Hyperlinks in this document: