There are inefficiencies in most listed real estate portfolio structures, resulting in significant loss of alpha opportunities, writes Witold Witkiewicz
As institutions increase their allocations to real estate, listed securities can offer efficient exposure to the underlying asset class. However, investors opting for active management are presented with poor portfolio construction practices that systematically restrict alpha generation opportunities. This is a structural problem and is the opposite of what active management should provide.
As investors look to eliminate inefficiencies from their portfolios, active management of listed real estate securities is likely to be impacted. At the same time, innovative solutions can provide investors with new ways to build more efficient programmes in the asset class, both passive and active.
It is well known that the threshold for active management as measured by a portfolio’s active share is at 50% and that below this level managers hold positions that cannot possibly be expected to outperform a given benchmark. Therefore, below an active share of 50% by definition a portfolio includes negative alpha exposures and is directly contrary to what investors should expect from active management.
Although active share has become a widely used headline measure it is the composition of active share itself that can give investors far greater insights into portfolio construction practices and a manager’s investment process. It provides a detailed profile of how a manager is pursuing outperformace, which parts of the opportunity set (investable universe or benchmark) he/she focuses on in terms of stock picking and the degree of conviction that is being exercised across the opportunity set to create unique portfolios and to generate alpha for clients. In our view, understanding the composition of active share is as important as the measure itself in evaluating active management.
A few years ago Kania Advisors conducted an industry wide study focused on active share and it’s composition in actively managed global real estate securities funds. The study evaluated the magnitude of active share across large-, mid- and small-cap segments, each defined as one-third of the relevant benchmark, and a fourth segment composed of out-of-benchmark holdings. The hypothesis is that in a portfolio with an active share of 50% and no out-of-benchmark holdings, if the manager pursues alpha generation equally throughout the opportunity set, each segment should contribute an equal proportion to the overall measure, that is, one-third of 50% or about 17%.
However, the results revealed that in the first half of 2012 actively managed funds across the sector had low active shares ranging between approximately low to high 30%’s, ie, meaningfully lower than the 50% threshold for active management. Further, the composition of active share was in some respects counterintuitive. It showed that in relation to capacity (market capitalisation of a stock), the large-cap segment accounted for only a nominal mid-to-high-single digit contribution to active share, less than half the expected level.
It is reasonable to assume that managers can be expected to have detailed knowledge of large-cap stocks in their investable universe, should have strong knowledge and views on these companies and should use the higher capacity to construct unique portfolios to generate alpha.
This is certainly what investors expect from active managers. Contrary to this, for all practical purposes managers were systematically ‘closet-indexing’ in the large-cap segment. Instead, the bulk of active share was generated in the small-cap segment with additional contribution from out-of-benchmark holdings. Mid-caps had roughly the expected contribution.
Well, aside from direct implications of low overall active share in the 30%’s, the most important issue is that managers were systematically under-utilizing a major part of the opportunity set. They were mainly providing index-like exposure to a large part of their opportunity set, large-caps, arguably failing to demonstrate their expertise and doing so at active cost to investors. Another issue was that portfolios were generally tilted to small-caps in aggregate with out-of-benchmark holdings, which implies that the overall exposure could potentially be less liquid compared to benchmark characteristics.
One potential explanation for these suboptimal portfolio construction practices could be that there is a structurally better opportunity to generate alpha in small-caps compared to large-caps. In that case it would certainly make sense for managers to focus their underwriting efforts on identifying those opportunities. We looked at the alpha opportunity within each segment by comparing best- vs worst-performing clusters of stocks within each segment and found no evidence of this.
The magnitude of alpha opportunity is similar across the three segments, ie, large-caps offer as much potential to generate alpha as mid-caps and small-caps. Perhaps then it is the case that large-caps are well analysed by the market making these stocks relatively more information efficient and managers do not feel that they have sufficient ‘information edge’ to take meaningful active positions without risking potential downside as a consequence. Instead, they might feel that their efforts are better spent focusing on less analysed and less information efficient small-caps in order to find value that others might have missed. Perhaps.
But that was then and this is now
Fast forward to today. After much industry coverage of active share and active management in general, are these structural issues now resolved? The short answer is, partly.
For listed real estate securities funds with global focus, active share has increased across the board compared to a few years ago to just above 50% currently so the sector is now actively managed. Also, the number of funds with active share below 50% has dropped during the same period. Therefore, on headline figures both are important improvements. However, that’s where the good news for investors ends. The not so good news are several.
Firstly, the composition of active share has not changed. Active managers are still not meaningfully active in large-caps where they have ample opportunity to demonstrate to investors their deep knowledge of these companies and clear value-add of their investment process and expertise. Active share generated in the large-cap segment is still at approximately only half of the theoretical level and opportunity, it has not changed. For investors this means that although large-caps represent a significant part of the opportunity set for alpha generation, it continues to be a largely index-replicating portfolio exposure. It remains a structurally underutilised opportunity by active managers by systematically limiting alpha generation.
Secondly, on average portfolios have migrated further into small-caps and out-of-benchmark exposures, which together now account for nearly 40% of portfolio holdings, reducing the proportion of large-caps and mid-caps and tilting portfolios further into potentially less liquid exposures.
Thirdly, as managers respond to previous criticism and construct more concentrated portfolios, the fact that portfolios are migrating into small-caps and out-of-benchmark holdings implies that these potentially less liquid holdings are also becoming more concentrated, presenting other types of risks. While the average size of holdings in large-caps has fallen in the past few years, the average size of holdings in small-caps has increased to ca 4.5x the relevant stock’s benchmark weight, an increase of approximately 15%.
All this means that investors are systematically and increasingly exposed to small-cap tilts potentially taking on higher event risks as these less liquid holdings are now becoming more concentrated. At the same time the large-cap segment continues to be structurally under-utilized. Active managers are not pursuing truly active management in a large portion of their coverage universe. And investors are taking notice as should consultants and fiduciary managers.
Real estate securities have recently been given their own GICS classification and will become more visible to investors who are likely to revisit how exposures and mandates are structured today and how they want them to be structured going forward.
Today, investors are also increasingly offered a wider range of products with which to structure allocations more precisely and more efficiently, for example in listed markets through indexing and ETFs including specific factors or characteristics within those.
If investors take a view that active managers fail to provide consistency in their investment process, they are likely to revisit how allocations are best implemented. In the case of listed real estate securities it might become valid for investors to ask themselves if active managers only should be given mid- and small-cap focused mandates and large-cap exposures being implemented through more cost efficient eg, market-cap weighted solutions.
We estimate that the cumulative threshold of reaching ca 75% of theoretical active share compared to a market-cap benchmark ranges between 30-50% of the benchmark market capitalization. Clearly, this is not a sustainable proposition and active managers should take notice and respond accordingly or risk placing up to half of their AUM in the firing line of more cost efficient product development.
So, can investors get truly active listed real estate securities allocations throughout the entire opportunity set of the investable universe including in large-caps? And can investors get that exposure without the tail risks of small-cap tilt, out-of-benchmark exposures, corresponding illiquidity and increased concentration risks?
Yes, now they can. To solve these, what we consider, suboptimal portfolio construction practices Kania Advisors has developed an alternative beta index designed to provide investors with new tools to construct more efficient listed real estate securities allocations. The index is based on Kania Advisors proprietary methodology of Consistent Active Indexing (CAI), Kania Global Real Estate CAI Index, and has important structural characteristics.
Firstly, while there already exist alternative beta indices for the sector, we believe those are largely based on general equity market factors instead of capturing what is really specific to real estate markets. There is plenty of evidence that listed real estate securities over time correlate more with real estate markets rather than equity markets, hence in our view applying general equity market factors on listed real estate securities markets may be inconsistent with the asset class. The index developed by Kania Advisors is based on factors specifically relevant to real estate with insights from over two decades of quantitative analysis of global real estate and real estate securities markets, and nothing else.
Secondly, the CAI methodology is designed to use large, mid and small-cap segments in equal proportion. This is important. As each segment of the index has the same weight as that segment’s weight in the benchmark there is no small-cap tilt (as is the case for alternative beta in general and in actively managed listed real estate securities funds as discussed above). Also, there are no out-of-benchmark holdings. The result is a more liquid overall index.
Thirdly, the index is structured to provide a consistent separation between beta and alpha with each segment contributing equally to the overall active share of 50%, or put differently each segment itself has a 50% active share. This means that it provides a solution to current portfolio construction practices of active management, the index provides equal and full alpha generation potential across the entire opportunity set, no segment is underutilised. This is ensured at each semi-annual rebalancing of the index.
Also, the methodology can be applied in a transparent way to a client’s specific benchmark or a customised universe, eg, by applying an ESG screen, to generate unique exposures. Lastly, other than the direct benefits above, the methodology provides consistency for structuring overall allocations between passive and active options. There is no need for investors to spend time or resources to either evaluate or monitor the active profile or indeed any ongoing migration of the active profile and potential effects on other exposures.
Witold Witkiewicz is founder of Kania Advisors