The return of beta
Something unexpected is happening: the renaissance of beta. As question marks over alpha are growing, good old market beta stages a comeback. But there is more to come: the discovery, or invention, of "alternative beta". Some institutional investors already feel the need to review their investment thinking - again!
In simple terms, by alternative beta people mean sources of risk-adjusted return that have traditionally not been accessible to investment managers. This area is so fresh that you find all sorts of concepts and labels like "exotic", "clever", "dynamic" or "hedge fund beta", "structured alpha"(as opposed to "pure alpha"), or even "hidden assets".
The terms beta and alpha were originally developed by Modern Portfolio Theory. Beta captures the exposure of a portfolio to market risk, ie investment risk that cannot be diversified away. In today's discussions, people refer to alpha as the ability to outperform the average market return.
Beta risk had become rather discredited in the pensions world in the early 2000s when the stock markets collapsed, and the bonds held proved inadequate to cover pension liabilities. In the new wave of investment, the focus turned on (market and liability) risk management. However, active fund managers reacted with the promise to spice up the more pedestrian return prospects with extra returns from various "alternative" or "alpha strategies".
Now, right at the time when most pension funds have finally succumbed to "alpha is in, beta is out", the tide may be starting to turn again. What happened? As so often, a combination of things - changes in the investment environment and changes in people's heads:
❑ Equity markets have strongly recovered from the lows of 2002/03, offering investors attractive beta returns by simply being in the market;
❑ The delivery of new alpha strategies, in whatever form packaged, has frequently not been up to the promise;
❑ In many cases, they have also shown surprising correlation characteristics with the main markets, visible even by naked eye;
❑ Pension funds can get beta returns almost for free these days through passive funds or index products, while the search for the new, but still uncertain, alpha comes dear.
As time goes by, more data from new wave investing is available, and academics have started to take a closer look at the nature of alternative returns. Hedge funds in particular have claimed that, unlike traditional funds, their returns are primarily due to fund manager skill while hedging unwanted market exposure out. In fact, the research presented by the alpha camp has for some time shown a positive alpha not only for many individual strategies but even for the hedge fund industry averages. The preferred reference model for return analysis thereby is the Capital Asset Pricing Model (CAPM), which conceptually breaks returns up in two components: beta returns (ie by passive exposure to the market) and alpha, the residual, which is being claimed as "active fund manager skill".
However, there is now a growing community of people who believe that the CAPM analysis overstates alpha, and thereby the contribution of the human skill factor. The beta camp argues that there are other contributors to return at work that are not captured by the CAPM market beta.
In the tradition of the Asset Pricing Theory (APT), they use factor models of return explanation, where most of the original alpha return is effectively sliced into a number of new factor betas. The remaining pure alpha is tiny, if at all positive (and post fees, quite likely to be negative).
What are all the additional factors? Much research had been devoted from the 1970s to identify such systematic risk factors. Typical examples include the size of the company (ie excess returns for small caps), credit (eg spreads paid for default risk of debentures), volatility and liquidity (ie excess returns for more volatile and less liquid assets).
Now, this sounds hardly new. In fact, most investors have over the years tried to capture such (fundamental or technical) risk premia through portfolio tilts and specialist investments. Classic examples are small cap stocks, value style portfolios, corporate and emerging bonds, private equity, real estate, and so on.
Despite the long experience with these specialist investments, the controversies have never really ebbed, neither in theory nor in practice: Do these risk premia really exist? Are they persistent? Can investors exploit them?
The critics insist that market beta is the only beta that deserves a risk premium while these factor premia are either:
❑ Non-existent, ie, the result of mis-specified research models;
❑ Too small to be worth chasing, taking all costs into account; or
❑ Only temporary (excess returns may be there for behavioural or institutional reasons, but once detected, demand increases and they evaporate soon).
Unfortunately, neither alpha nor beta can be directly observed, as their measurement depends on the model you choose for their analysis. But investors cannot wait for judgement day, and in practice they have taken a much wider definition of "the market" than just their local equity and bond index.
So what has really changed? There is a renewed alpha versus beta debate that has emerged from the analysis of hedge fund returns. Some researches think they have found new alternative betas, and a number of providers think they can make them newly available to investors.
What is meant by this? It is probably best to define alternative betas as systematic risk factors (ie, uncorrelated with global markets and with positive expected return) that are not available through the passive exposure in asset markets. In other words, they are factors additional to the market beta and the traditional (fundamental and technical) factor betas.Alternative beta covers common risk factors that drive returns in trading strategies, such as
❑ Volatility and correlation trading;
❑ Other typical hedge fund strategies such as merger, convertible, fixed-income "arbitrage";
❑ Insurance against downside risk or extreme events;
❑ Provision of liquidity, eg, to hedgers.
In this definition, the distinction to traditional beta is through the involvement of trading techniques and hedge fund practices (eg, derivative instruments, shorting, leverage, rather than asset class. Beware of the misnomer "alternative asset class": a premium for buying and holding a less liquid asset class, eg, real estate, still falls into the traditional category."
But with trading strategies and timing decisions involved, what really distinguishes alternative beta from alpha? One answer is: when alpha skills become repeatable processes, more widely known, they become alternative beta. How to filter out such alternative betas from alpha is, and will remain, controversial.
The discovery, or invention, of alternative beta has far-reaching implications. After all, chasing alpha (unlike beta) is only a zero-sum game. It requires losers on the other side. Also, there is a scarcity issue: many trustees are sceptical about the recent mushrooming of "exceptional investment talent". If the alternative beta camp were right, additional sources of return would become available to investors in much purer, simpler and cheaper form than through hedge funds.
In fact, the investment industry is already reacting. One way to play alternative beta are so-called strategy portfolios, eg by gaining more direct exposure to volatility, correlation, yield curve, credit and other trading strategies. Such purer alternative beta products could well be interesting for pension funds with an advanced asset allocation and risk management process. For example, they could start diversifying more confidently in the alternative spectrum, and avoid the costs and complications of fund of funds.
Institutional investors always found it difficult to deal with the uncertainty and imponderability of the human skill element. Cynics look at alternative beta as an attempt to institutionalise in hedge fund investing via a manufacturing approach that aims at higher capacity, consistency, transparency and lower fees. Pension plans will be asking a lot of questions in this process, among them:
❑ Are we really dealing with undiversifiable risk that deserves a premium? What is the economic and financial rationale behind?
❑ The alternative data history is short. Do the measured betas have any predictive value for the future?
❑ What exactly is the alternative beta element in asset classes like currencies or commodities?
❑ Volatility per se does not necessarily pay a risk premium (see the volatile results of gambling). Similarly, for illiquidity or complexity. What is the specific systematic source of return?
Pension trustees can these days see the point of enlarging the investment universe. Particularly the new generation of investment officers would be open to the idea of accessing hidden sources of return so far unavailable to them directly. However, the mechanism of separating the passive from the active element in trading strategies will need to be thoroughly scrutinised.
Georg Inderst is an independent consultant based in London