AQR To Investors: Get Intentional About Your Unintentional Risks
Institutional investors with allocations to hedge funds and private equity funds could be exposing themselves to unintended portfolio risks, according to quantitative investment manager AQR’s latest research piece, Was That Intentional? Ways to Improve Your Active Risk.
Active management and tactical asset allocations are two methods that investors take to try and outperform their strategic asset allocation benchmarks – i.e. intended risks - but they should be paying as much, if not more, attention to unintended risks, according to the paper.
Authored by the firms’ Portfolio Solutions Group, AQR’s paper examines the effect that correlated bets from different third-party investment managers has on the likelihood of beating a benchmark. Investors tend to allocate capital to multiple third-party investment managers to diversify away idiosyncratic manager risk and to give them access to different sources of returns, like bottom-up stock picking and momentum in equities, two very different investment and trading styles, for example. However, AQR’s paper suggests that in some asset classes active management is not particularly diversifying because of the correlation of managers’ returns to each other. AQR’s paper says that hedge funds have one of the highest correlations of returns across managers, at just under 0.5. The paper says that the correlation of hedge fund returns to equities generally is even higher, at about 0.6. In comparison, traditional long-only equity correlations across managers is less than 0.2 for U.S. large cap equity and emerging markets equity and less than 0.1 for international equity.
A solution to manage this risk is to run correlation and regression analyses to help evaluate active manager performance and inform changes at the portfolio level. Additionally, “From the active risk perspective, investors should consider benchmarks that match ‘passive’ exposures as well as possible to reduce the impact such exposures could contribute to the active risk budget,” AQR’s paper says.
Illiquid assets like private equity have become increasingly prevalent in institutional investors’ portfolios in the past two decades, and AQR points to the implementation lag between an investor’s desired long term target PE allocation and the time it takes to achieve said allocation as another unintended risk of which investors must be aware. It offers an example of an investor which wants to increase its PE allocation from 5% to 10% (sacrificing public equity exposure from 55% to 50%, maintaining the 40% bonds allocation in the standard 60/40 model) which will take them 5 years to achieve. To reflect this asset allocation change, they can either change the benchmark weighting of PE immediately to 10%, or they can stagger the 5 percentage points increase over the 5 years at 1 percentage point per year.
The former option creates a larger tracking error: 0.8% in year 1, versus the latter option which produces on 0.2% tracking error. Similarly, the likelihood of underperforming the investor’s strategic asset allocation is significantly lower over the 5-year period if they stagger their benchmark change versus changing it all in one go: AQR’s paper shows the odds of underperforming the benchmark by 50 basis points with the staggered approach is just 0.1% over the 5 year period versus 13% using the immediate change scenario.
Benchmarking private equity allocations can also be challenging because of the differences in the reported monthly, quarterly, and annual returns of private equity funds when compared to public equities. AQR’s research looks at the median private equity returns for U.S.-based public pension plans and finds a 10.7% tracking error when benchmarked to the S&P500 but only a 2.7% tracking error when benchmarked to the Cambridge Associates U.S. Private Equity Index. Hedge funds don’t fare much better: again, using the median hedge fund returns for U.S.-based public pension plans, the tracking error to cash is 5.3% and only 1.7% to the HFRI fund-weighted composite index. The data shows the challenges that investors have of whether to pick a public proxy for benchmarking purposes, or a peer group proxy, for their illiquid allocations.
“Customizing existing public benchmarks (e.g., including lags to “smooth” benchmark returns), using peer benchmarks, or removing these difficult-to-benchmark asset classes from portfolio-level active risk budgeting altogether,” are a few solutions to this challenge, AQR says.
AQR’s paper analyses other ‘unintended’ risks, such as cash buffers kept by investors, rebalancing schedules and currency risks. It stresses that while these risks can diversify each other, they can also be additive, leading to more unintended risk. Furthermore, the unintended risks can outweigh the intentional risks.
“Outperforming an SAA benchmark is hard enough – making sure your portfolio’s active risk is as intentional as possible can make the journey a bit easier,” is AQR’s bottom line message to investors.
© The Sortino Group Ltd
All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or scanning or otherwise, except under the terms of the Copyright, Designs and Patents Act 1988 or under the terms of a licence issued by the Copyright Licensing Agency or other Reprographic Rights Organisation, without the written permission of the publisher. For more information about reprints from AlphaWeek, click here.