Investors are increasingly ‘chasing beta' as a proxy for taking less risk. But is this actually achievable? Neil Cable warns against peddling an unhelpful myth
Something strange seems to be happening in the real estate industry: some investors appear to be following an investment strategy that has almost no chance of delivering on its promise. Rather than chase the high alpha of the heady early-to-mid 2000s, there now seems to be a growing desire for ‘delivering the market', or beta. In one sense, this is perfectly understandable. People naturally react to extreme situations by becoming increasingly risk averse. However, there is a danger for the real estate industry - both advisers and fund managers themselves - in perpetuating a belief that beta can actually be delivered.
Fund managers should not shirk their responsibility to manage their properties professionally - all funds, from core to opportunistic, need to manage tenants, maintain buildings, protect income, control non-recoverable costs and so on - and advisers should not be drawn in to believing that beta can be delivered either by simply being large, or by following a strategy of - for example - buying ‘prime' assets, focusing only on long leases, or focusing only on high-quality tenants. All of these things will deliver different risk profiles for clients, but history suggests that none of them will stand much of a chance of achieving beta - or ‘index returns'.
Providing beta in other asset classes relies on an ability to replicate an index, either via derivatives or via a tracker fund or exchange-traded fund, which automatically buys and sells stocks based on an index. This is, quite simply, not achievable in an asset class like property, where every deal and every lease agreement is the result of a private negotiation.
Investing in more than one fund will certainly diversify manager risk, and investing in one or more large funds will certainly achieve a reduction in specific risk within the underlying portfolio. But neither can effectively be designed to achieve beta, especially given time lags in investing and disinvesting (that is, subscription and redemption processes), fund structures (closed ended, open ended, etc) and an inability to gather data on - and therefore to replicate the performance of - each individual asset.
Derivatives are the trump card for the beta lover - they have come a long way in property markets and are a very useful tool for investors, especially if they simply want to ‘buy or sell the index'. However, depth of the markets and a lack of true market-making makes this a reliable and efficient option for the few, not the many.
As recently as the early 1990s, many property investors around the world invested principally in their own domestic markets. The terms ‘core' or ‘core-plus' hardly existed and the terms ‘alpha' or ‘beta' were not common parlance among property fund managers. Investment Property Databank (IPD) was a nascent force, but property fund managers assumed they were supposed to outperform their competitors and actively manage their portfolios. Few in property thought about tracking an index.
As the 1990s drew to a close, the first wave of international funds were launched; IPD was expanding rapidly, but the European Association for Investors in Non-listed Real Estate Vehicles (INREV) was yet to be properly established. Pension fund consultants had started recommending property again for the first time in years and, with all the new data available, increasing sophistication in both performance measurement and attribution analysis started to become commonplace. Along with cheap credit and new market entrants came a raft of ‘opportunity funds'. The focus, and expectation of clients, moved from returns being driven principally by income to capital growth. For a while it seemed that large numbers of funds could indeed produce double-digit returns from property, and many assumed this was due to the property industry honing its skills and property generally becoming a mature asset class.
Then late-2007 arrived and much of the ‘alpha' that had been embraced by so many investors disappeared in a puff of debt.
Last year, a new report by the Urban Land Institute (ULI), ‘Have Property Funds Performed?', revealed that the alpha that many funds ‘generated' (albeit during a short six-year period of analysis) was indeed nothing of the sort and the risks many property funds took with clients' money would shock the most aggressive equity fund managers.
A measure of this is to look at tracking errors. Put simply, any tracking error, positive or negative, indicates a failure to track the index - or deliver beta. While opportunity funds would have been deliberately trying to produce a high alpha - and therefore a relatively high tracking error - the core funds should have had a relatively low tracking error.
If we compare these numbers to equity funds over the same time period (using an internal Fidelity analysis of over 200 equity funds), we can see even the core real estate funds delivered tracking errors which would be comparable to ‘high alpha' equity funds (that is, funds typically looking to outperform their benchmarks by an aggressive 5-8% per annum). The ULI study concluded that the core European real estate funds had a higher-than-expected level of market risk.
At the other end of the spectrum, ULI suggests the tracking errors of some of the ‘high alpha' property funds would be exceptionally high - perhaps unacceptably high - if observed even in high alpha equity funds.
The ULI data is stark, but has some acknowledged drawbacks, not least the relatively short time period covered by the data. It is also worth acknowledging that beta can be delivered in other asset classes even with relatively high tracking errors, although this relies on a number of financial tools unavailable in the property market. But the comparison between the asset classes throws into sharp relief the high tracking errors experienced by property funds.
To try to ascertain whether the ULI conclusions are supported by longer term data, or a wider sample of funds, data was obtained from IPD covering the UK market - where we have reliable data over much longer time periods. Figure 2 provide conclusive evidence that over a 20-year period - that is, a period covering different cycles and different market conditions - not one single fund in the databank delivered a tracking error of less than 50bps. Nor was time period a factor: the same was true over 15, 10 or even five years. In fact, only two funds out of 88 measured over the full 20 years delivered a tracking error of less than 200bps. Or, to put it another way, an equities analyst interpreting the IPD data, would conclude that 98% of funds were either ‘low alpha' or ‘high alpha' funds - and none at all delivering anything close to beta.
‘Beta chasers' - an illusive dream
There is a seductive appeal to targeting beta as a strategy, but it seems to be fuelled by the same illusive ether that intoxicated those perpetually chasing the high alpha of the early 2000s. The supposedly safe option of beta may turn out to be just as illusive as the hunt for high alpha ultimately proved during the boom.
Indeed, it is worth noting that despite the pioneering work and global success of IPD over the past quarter of a century, property indices are still, in some respects, in their infancy. Taking the UK as an example, IPD estimates the total market size of the commercial property market as £238bn (€286.8bn), yet its index (£135bn in size) only covers 57% (source: IPD UK Annual Digest 2010). The AREF/IPD UK pooled funds index - used by many for benchmarking - has a total net asset value of just £28.5bn, or just over 10% of the market. One includes the impact of fund management fees and leverage, the other does not. One provides much more detailed data than the other, enabling better attribution analysis.
To deliver beta, one needs to be able to credibly replicate an index and two things are clear looking at the property market: first, it is simply not possible to replicate property indices; and second, it is questionable whether replicating available property indices would be providing beta in any case.
So what is the solution? If we are to learn anything from the last boom/bust cycle, it is that risk is as important as return. As the ULI paper showed, the risk that investors thought they were taking was very different to the risk they were actually taking - and of course, they did not get the returns either. Not only does the industry need to improve measuring and to control and explain the risks it is taking, it has a wider fiduciary duty not to mislead. It is possible that some in the industry talk of beta because they want to appeal to clients' current risk aversion, and what they really mean is low alpha. If so, they should say so. Investors tend to react badly when fund managers say one thing and do another, and if the reality is low alpha rather than beta then investors need a clear explanation of how and why that level of performance and risk is achieved, rather than a misleading perception of lock-and-leave index returns.
Writing last year in his article ‘How much later allocator?' Andrew Thornton, chief executive at Internos Real Investors asked why many pension funds were not allocating new capital to the sector, given its attractive risk-return characteristics. Perhaps some capital is going into ‘beta plays' or perhaps some investors are still smarting from the empty promises from the last boom. Either way, as we emerge from a damaging credit crunch, if we are to encourage pension funds to make renewed allocations to real estate, now is surely the time to be brutally frank about the level of performance - and risk - they should expect.
This leads to three main conclusions:
• Real estate is the ultimate alpha asset class: Fund managers, consultants and pension funds should not hide behind a mythical cloak of beta. Certainly, some funds will take more risk than others, and some will have higher outperformance targets than others, but we should not pretend that delivering an index return is an achievable goal. Alpha - even low alpha - must not be dressed up as beta;
• It is about risk just as much as return: One of the ironies of the headlong dash from alpha to beta is that the real estate industry still seems to be obsessed with quantifying return, but not yet quantifying risk. INREV's new definitions are a welcome step in the right direction, although they probably have some way still to go. Fund managers need to be honest about the level of risk they are taking in order to generate a return, and be able to show their clients how and why they are controlling that risk;
• More transparency, not less: Investors have made it very clear that they require more transparency, better governance, better communication and more sophisticated analysis of risks than in the past.
If the real estate industry is to claim its rightful share of capital from investors, it needs to address all of these issues head on. Promising beta is not a promising way to start.
Neil Cable is head of European real estate at Fidelity International