Dynamic Portfolio Management and Tools for Building Risk Adaptive Portfolios

The Traditional (Beta-Driven) Approach and Current Markets

In the current environment of low yields, pricey equity markets and lower expectations for portfolio returns, institutional investors may find it challenging to meet their return objectives over the next several years. Traditionally positioned portfolios with a high dependence on equity returns are likely to be even more hard pressed. The traditional approach to institutional investing begins with setting a long-term strategic asset allocation ("SAA"), hiring managers and periodically rebalancing back any portfolio drift to the SAA weights.  This approach could be called a "beta driven" approach as the bulk of the risk of these portfolios is derived from the beta of the portfolio.  In an upward trending market, like that experienced over the last nine years, dynamic risk management and innovative portfolio construction, that attempt to add meaningful risk-adjusted alpha, are not always appreciated.

"Risk-focused" portfolios, by contrast to beta-driven programs, seek to find the best reward per unit of risk from both pure beta exposures as well as active managers and strategies, and hence require reallocation dynamically to the best opportunity.  When equity markets are surging, and diversification keeps well-structured risk-focused portfolios from attaining the same level of returns as some of their beta driven peers, many thoughtful CIOs are asked by stakeholders, "Wouldn’t our plan do just as well by indexing the portfolio and saving all the hassle?" It is an important question that every CIO of a risk-focused plan knows that they will be asked. In such environments, risk-focused portfolios are not appreciated for having higher Sharpe ratios if they earn lower returns and certainly, their risk-adjusted ratios cannot be used to pay pensions or fund a university endowment’s obligations.

Many investment programs utilize a beta driven portfolio approach with a keen focus on tracking error, mean variance optimization, benchmarks and cost minimization. Reputational risk from running a portfolio with this type of structure is limited as when markets go up, the plan participates well and when markets go down losses can be attributed to these market shifts.  Over time the upward bias of the market provides these investment programs with a combination of risk premiums over cash which provide the basis for the growth of their investments. This somewhat passive or static approach to investing in contrast to more dynamic approaches is therefore often appealing to allocators.

The appeal of this approach to investing can be especially compelling toward the end of bull markets, when volatility has been low and equity multiples have been rising for a prolonged period.  During those times risk management and more dynamic approaches to portfolio management are often unappreciated.  Investors fear missing out on the next leg up in the market and tend to ignore old concepts like cash flow analysis, active management, value and diversification.  Late in major market cycles, those activities seem counterproductive as market leadership narrows to a small group of stocks like RCA and General Electric in 1929, Coca Cola and Avon in 1972, Cisco and Microsoft in 1999, and FAANG – the group of stocks made up of Facebook, Apple, Amazon, Netflix and Google's parent company Alphabet – in this cycle.  Market participants are rewarded for more passively investing and ignoring risk management, diversification, rebalancing and cash.

However, when volatility begins spiking and a new bear market arrives, a risk-focused portfolio not only helps protect capital but also enables the dynamic investor to seize opportunities.  Having a diversified, cash flow focused portfolio that is less impacted by market volatility can provide an investor with the confidence and portfolio liquidity necessary to take advantage of market dislocations and more attractive long-term investment opportunities when they occur.

Wisdom of the Past (or Dare to be Different)

Some of the greatest insights on how to approach investing dynamically can be gained from reading the works of Benjamin Graham, who was regarded by many as one of the greatest investment thinkers of all time.  He was a teacher, employer and friend to Warren Buffett and transformed the investment industry into a modern profession back in 1934 when he and co-author David Dodd published the textbook Security Analysis.  Graham also wrote The Intelligent Investor, which is an invaluable guidebook regarding the emotional mindset necessary to succeed over time in investing. The Intelligent Investor articulated the philosophy that the returns an investor should aim for or expect were not tied to the amount of risk that the investor was willing to take but rather the amount of intelligent effort the investor was willing and able to bring to bear on the task of investing and portfolio management. Graham believed that the minimum returns over time would go to the passive investor who wants safety and freedom from concern and that greater returns could be realized by the alert and enterprising investor who exercises maximum intelligence and skill.

When Benjamin Graham referred to intelligence and skill in the context of investing, he was addressing a disciplined analytical approach to the understanding of the market and individual companies and how they should be valued based on their growth prospects, price multiples, capital structures and cash flows as represented by their dividends.  He believed that investing was a discipline focused on having a longer time horizon and creating diversified portfolios with a margin of safety, while speculation was centered on attempting to profit from shorter term price movements.  He generally viewed short-term trading strategies as speculative but believed that longer term strategies focused on buying great growth companies or finding companies trading at significant discounts to their long term intrinsic value could be effectively employed by enterprising investors.

For enterprising investors, Graham also advocated the contrarian investment philosophy, commonly ascribed to Baron Rothschild, of buying markets when they are cheap and selling them when they are dear.  Dynamic rebalancing focuses on this principal of dynamically adjusting portfolio allocations based on relative value.  This disciplined approach to asset allocation provides a risk posture that offers alert and enterprising investors a greater foundation to take advantage of market dislocations and other opportunities dynamically when they are presented. This approach has been exemplified by many of the investments made by Graham’s prodigy, Warren Buffet, over time.

While Buffet is probably best known for investing in great companies over the long term, he has also proven himself as an alert and enterprising investor who is able to take advantage of market opportunities when they present themselves.  Over the years, he has prudently diversified his holdings with investments in insurance and reinsurance, which do not correlate to the equity markets, and has also taken great advantage of market dislocations.  During market downturns, he has provided debt and equity capital to companies in crisis when few others had cash available or the wherewithal to evaluate risk in dislocated markets.  When banks needed to offload market risk, he made attractive deals selling long dated market puts to them. Warren Buffet exemplifies how long-term capital can and should be allocated dynamically over time.  Also supporting this concept, Samuel Kunz, Managing Director, Asset Allocation & Investment Strategy at the University of California and Arun Muralidhar, Founder and Client Chief Investment Officer of AlphaEngine Global Investment Solutions, demonstrate how investors can commit to their investment beliefs and "dare to be different" around three major areas: 1) investing style (formulaic or discretionary); 2) optimization method (for setting the SAA and tactical asset allocation ("TAA")); and 3) rebalancing process (mechanical or informed).¹ Similarly, there are other examples of pensions and endowments that have shown great capacity for exercising intelligence and skill as enterprising investors.

Dynamic Portfolio Management

The bull market that began in March 2009 was one of the longest on record and would have been nine years old on March 9, 2018.  The S&P and Dow possibly topped prior to this anniversary, but the NASDAQ exceeded it.  However, despite this sustained bull run in equities, many pension funds are still underfunded and hence require an unreasonably high future rate of return on their assets.  Long bull markets make it easier for pension plans and other institutional investors to meet their immediate target return objectives, but the higher priced assets also lead to lower expected returns in the future.  Many institutional investors recognize this fact and have lowered their future actuarial targets over the last few years, reflecting the concern that beta may under-deliver in the future, while the US stock market continued its ascent to historic levels.  This trend suggests a concern in the industry that equity dependent beta-driven portfolios may struggle in the future.  The chart below illustrates modest annualized returns over the last decade for most liquid assets other than US equities.  These lower returns for most liquid assets have caused many well diversified portfolios to have difficulty in achieving their needed rates of return.  With a low overall level of interest rates and historically high equity prices, it is unlikely that beta-driven programs will find it any easier over the next several years to meet their return objectives.

Return expectations of thoughtful allocators are driven by their understanding that price change, income and cost are the three components that make up future returns for financial assets like stocks.  When the income component of returns is at historically low levels and future price appreciation is limited due to high relative prices for most assets, future return expectations should come down.  Equity multiples and credit spreads at levels that have historically been precursors to market downturns, remind institutional investors that it may be very challenging to achieve their required rate of return over the next several years. Not all programs, however, are equally challenged.  Some institutional investors may weather future market volatility and a low return environment better than many of their peers.  These exceptional investment programs generally have a set of values and investment beliefs that differentiate them from the herd.  To be exceptional, by its very nature, generally means that these investors and their programs don’t follow the herd or blame the market for losses.²  Their leaders do not manage portfolios the way they do because it is easier, they often do it despite the challenges and greater personal risk because of a belief that it is a better way to serve the entities that they represent. To manage a portfolio that is better rewarded for risk than a typical beta-driven program, the portfolio manager must have a greater focus on market risk and valuations and have a disciplined approach for dynamically adjusting their portfolio to adapt to the market environment.  It is important to try to protect capital when valuations are extreme by historic standards or relative to expectations and it is equally important to seize opportunities when the investor is being overcompensated for risk.  This process of varying risk exposures to adapt to the market environment is a discipline that is sometimes referred to as Dynamic Portfolio Management ("DPM"). With a greater focus on capital protection and opportunistic investing, DPM can offer some tools to enhance the probability of realizing more attractive returns over time and increasing the possibility of achieving investment return objectives. The foundation of DPM is not a new concept or methodology for managing portfolios. It is rather an "old school" informed decision-making process for actively managing portfolio asset allocations and sub-asset allocations.  It generally embraces Modern Portfolio Theory ("MPT") and mean variance optimization ("MVO") but attempts to overcome some of the known shortcomings of these useful modeling tools.  Through fundamental and technical analysis, dynamic portfolio managers seek to optimize their portfolios’ structures and allocations to generate better risk-adjusted returns. This approach to portfolio management may be of even greater importance as market volatility begins to spike and opportunities to buy assets at attractive price levels reappear.

Dynamic Portfolio Management as a Complement to Modern Portfolio Theory

For most allocators, MPT provides the mathematical framework for optimizing the expected return of their portfolio for a given level of risk.  That optimization process is challenging due to its requirement for anticipated asset class returns and correlations as the critical data inputs for MVO calculations and analysis.  Often, correlation inputs used for MVO models are derived from past market experience and are extrapolated forward.  The extrapolation of the recent past into the future can dangerously mislead allocators at major market turning points.

Asset class correlations vary significantly over time and depending on the assets may range from strongly positive to strongly negative levels.  In general, MVO is most beneficial when sets of assets are available with stable low correlations to each other.  The weight of any asset class in a portfolio is then determined by how it impacts the overall performance of the portfolio rather than just by its individual risk-reward profile.  In the current market environment investors face a world in which they are not only dealing with richly priced assets that may offer lower future returns, but also assets that have higher correlations.  These heightened yet unreliable asset correlations lead to reduced benefits from the portfolio construction tools of MPT.

As the chart above illustrates, the correlation benefits of major asset classes have varied dramatically in even the last two market cycles which, until recently, were dominated by a secular bull market in bonds.  Diversification benefits were clearly greater during the dot-com crash than during the more recent global financial crisis.  Going into the global financial crisis, many portfolios were optimized based on MVO models utilizing correlations of the prior market cycle.  Those lower correlations failed to be realized during the market downturns resulting in higher overall portfolio volatility and larger portfolio drawdowns than anticipated. Markets remain in a period of elevated correlations due to the global flood of liquidity that has been unleashed by central banks around the world.  This liquidity has pushed capital into almost all asset classes available to institutional investors.  Future correlations will depend upon many factors including synchronization of various market cycles, central banks, cash flows and valuations.

DPM may overcome some of the correlation challenges of MVO modeling.  Typically, MVO modeling seeks to optimize returns for a specific level of risk.  DPM by contrast seeks to allocate capital, on an absolute or relative basis, to assets where risk appears to be overcompensated based on historical fundamental and technical relationships.  While MPT and MVO seek to take advantage of market efficiencies, DPM seeks to take advantage of market inefficiencies. DPM is the foundation of active management, contrarian investing and the adage of "buy low and sell high."  A portfolio constructed utilizing a DPM methodology should generally be expected to reduce risk exposure as asset values increase and increase risk exposure as asset values decline.  Ironically, while gross exposure to a portfolio's array of asset classes typically varies more significantly in a portfolio utilizing DPM than MVO, actual volatility is often dramatically lower for a DPM portfolio than for an MVO focused portfolio with a similar SAA.

While DPM is a traditional investment approach, it has several key portfolio construction differences from typical MVO portfolios regarding risk, assets and cash levels.  Most DPM advocates adhere to the belief that an asset’s price level is a better determinant of true risk or capital loss than its volatility.  At its heart, DPM relies on the traditional fundamental analysis of estimating an asset's future cash flows and discounting them at a rate sufficient to compensate the investor for the range of probable outcomes regarding the investment.  A more volatile asset trading at a price lower than its fully discounted price may offer less risk of capital destruction than a less volatile asset trading above its discounted price.

A focus on discounting cash flows may lead to a higher percentage of high cash flow investments in a portfolio due to the investors appreciation of the more predictable return stream and the lower volatility of the cash flow aspect of the investment than the price change component.  DPM driven portfolios often also tolerate higher levels of cash than typically seen in MVO focused portfolios. This tolerance of higher cash levels is derived from an acceptance of the "real option value" of cash.  Typical MVO portfolios consider cash as a drag on portfolio returns.

While cash is considered potentially useful for reducing volatility, under MVO it is generally considered lacking in the ability to drive returns.  By contrast in a dynamically driven portfolio, cash is considered to have the return profile of not just cash, but the weighted average return of cash and the future return of whatever asset you might buy with that cash after a market dislocation.  In other words, cash has option value and cash expected returns should be enhanced by the real-option value to buy distressed assets.  Basically, as asset prices become more overvalued and the probability of a dislocation increases, the real option value of cash increases and the potential exposure to it under DPM is likely to increase as well.

Dynamic Portfolio Management Principles and Guidelines for Portfolio Construction

There are several basic principles and general guidelines that provide investors with a foundation for utilizing DPM and the concept of building risk adaptive portfolios in conjunction with a plan’s SAA. These include:

• Allocations should be actively managed: Enhanced risk-adjusted performance may be derived from active management of allocations throughout a portfolio at the asset class, sub-asset class and security levels through informed tactical deviations from a SAA benchmark;
• Allocation tilts should be driven by expert information: Information to inform allocation tilts is driven by price and market behavior indications and should be driven by expert informational sources focused on allocation decisions at that level. Asset class tilts may best be driven by a blend of broad market valuation metrics, relative value between asset classes, global macro factors and technical indicators. Sub-asset class and security selection weights may best be driven by extensive fundamental and valuation work at the manager level;
• Price and valuation matter: The long-term expected return for holding any asset is highly dependent on the price you paid for it. If you pay too much for a great asset, your return may be worse than if you buy a less attractive asset at a better price. Ideally, you want to buy great assets when everyone else is selling;
• Cash flows matters: An informed investor should not be indifferent to the source of an investment’s returns and adheres to the discipline that understanding and utilizing cash flows of investments is a fundamental aspect of risk management. Normally, the cash flow and cost aspects of an investment are the most stable components of return, while price change is less reliable and more volatile as multiples change significantly over time;
• Leverage is neither good nor bad: It is a tool which amplifies both positive and negative performance. In a positive market, it may significantly enhance the returns of an investment while in a bear market environment or with a weaker company, it may wipe out an investment that would have otherwise survived. The type of leverage utilized is often as important as the decision to use leverage and stable term financing will often prove much more valuable over a market cycle than lower cost options such as total return swaps;
• Default position should be driven by the SAA: If there is not an informed view regarding a variation of any allocation aspect, the default position should be driven by the SAA and all deviations from target are active decisions which should be monitored and measured through attribution analysis for impact on portfolio risk and performance.

Dynamic Portfolio Construction Tools

In general, DPM finds value in active management, but recognizes that the additional incremental return derived in some of the most efficient markets may generate limited additional returns relative to the additional cost.  In those circumstances, the dynamic investor may find a greater ability to generate additive returns by focusing on dynamic portfolio construction tools such as a portable alpha program that would allow the investor to focus on more attractive levers for generating incremental return.  With a portable alpha program, the investor may drive returns through: 1) dynamic rebalancing; 2) dynamic implementation; and 3) alpha pool construction.

Dynamic Rebalancing

The macro aspect of DPM is otherwise known as dynamic rebalancing, whereby the investor varies the allocation to the major asset classes within a target range relative to the SAA based upon fundamental and technical factors.  An example of a very successful dynamic rebalancing program would be the "Informed Rebalancing" program at the San Bernardino County Employees Retirement Association ("SBCERA").³   Their program has been in operation for over a decade and has added between 50 and 100 basis points on an annualized basis at the total fund level for the five and ten year periods while reducing portfolio volatility.  The program has served as possibly the single biggest contributor to total alpha and risk management for the portfolio at a very low cost. It was built on the Mcube Investment Technology’s AlphaEngine® Platform using peer reviewed journal articles to create trading rules to alter the weights between pairs of assets. It was predicated on the idea that portfolio ranges in the SAA were either a source of unmanaged portfolio drift or could be managed intelligently – thereby improving governance – for alpha and risk management by the investment team using a Benjamin Graham / Warren Buffet style approach to reallocating to the most attractive asset class.  The original model also focused on a low turnover approach which reduced portfolio turnover relative to range bound rebalancing models.

Dynamic Implementation

Dynamic implementation refers to the process of varying the synthetic instruments and structures used in creating a beta overlay.  The most basic type of overlay or beta replication is a simple futures contract used to duplicate the desired beta on a one-for-one basis.  This may be referred to as a "delta one" contract, where the valuation change of the contract, if overlaying cash, is approximately the same as the valuation change if the index was owned.  As volatility changes over the market cycle and fear and greed vary, the dynamic investor may take advantage of derivative price volatility through dynamic implementation of their synthetic beta exposures. At various times in the market, the investor may be able to structure overlays with superior risk-adjusted characteristics than a traditional delta one futures position.

Beta Overlay Programs

The construction and management of the collateral pool used in a beta overlay program is another opportunity for DPM at the asset and sub-asset level. The overall collateral pool or "alpha pool" used in a beta overlay program is typically a blend of low volatility hedge funds paired with a significant allocation to cash. It is critical that the collateral pool for the derivative overlay strategy be managed with a strong focus on maintaining ample liquidity for any needed derivative transactions that would be required to maintain targeted asset allocations even in a severe market downturn or crisis.  That said, the allocation and blend to the hedge fund portion of the collateral portfolio may vary to some extent depending on outlook, market environment, portfolio risk exposure and interaction of the beta overlay.  In most environments, strategies that exhibit limited beta, low volatility and greater liquidity are optimal components for constructing the beta overlay collateral pool. Market neutral equity strategies and relative value long-short credit strategies are types of strategies which are often suitable building blocks for constructing relatively stable collateral portfolios.

Dynamic Rebalancing at Sub-Asset Level through Managed Custody Accounts

Strategic partnerships with managers and the ability to tap into their market insight can provide extremely valuable information for managing portfolio sub-asset allocation and security selection decisions.  The Managed Custody Account ("MCA") is a powerful tool to help manage that information flow and enhance alignment of interest between the asset manager and the dynamic institutional investor.  The MCA can also help institutional investors with one of the most common problems inherent in institutional portfolios, over-diversification.  Many institutional investors hire multiple managers in an asset class to limit risk to the manager’s specific style or some unforeseen risk.  Unfortunately, this process also often leads to an expensive index type return profile.  By working with a smaller number of managers, with MCAs in place and a strong alignment of interest, the institutional investor may concentrate their portfolio in that sub-asset class into the best ideas of their managers, while maintaining a well-diversified portfolio overall.

Dynamic Strategies Potential Impact on Portfolio Returns

The potential impact of the various dynamic portfolio management tools on returns and volatility varies significantly.  In building out these programs, a focus on margin of safety and prudent diversification levels should be paramount.  The extent to which the strategies will be utilized, and the amount of time and energy expended on them are likely to have a significant impact on their overall effectiveness.  Each of the strategies should be considered a major structural development program with dedicated personnel and resources.  Developing and building out all of these programs can entail a substantial time commitment and originally at SBCERA took almost a decade. With two senior developers of these strategies, who had already implemented these programs elsewhere, the Texas Tech University System ("TTUS") was able to build out these programs in just a couple of years.

Larger plans are likely to benefit most by developing these programs internally but may find external service providers helpful in their program’s development. Smaller plans where staffing or staff retention may be constrained, may benefit by outsourcing these programs to an external service provider. In most cases, pursuing an incremental and iterative approach to building out these programs is beneficial in developing needed skills and infrastructure. Starting with smaller alpha overlay and MCA program sizes while refining insight into the operations of those programs can be prudent. Tighter initial rebalancing ranges can help validate model effectiveness. Over time as programs mature, sizing and rebalancing ranges may be expanded with the increased knowledge and skills of the teams implementing them. The chart below illustrates some potential starting goals and longer-term targets that may be possible from fully developed and well run dynamic programs.

 

Importance of Effective Reporting

Despite the many benefits of Dynamic Portfolio Management, it does add layers of complexity to a portfolio.  This added complexity creates the need for enhanced portfolio transparency, analysis, monitoring and reporting. The transparency is needed to drive greater insight into the portfolio and understand what facets of the program are working well and which are not.  In a dynamic portfolio, it is not enough to be performing well relative to peers, it is important to understand the risks that the portfolio is exposed to and how well positioned the plan will be to take advantage of opportunities as they arise.  Unlike some programs where information and transparency are not always requested, because of an inability to process that additional information, in Dynamic Portfolio Management it is critical to not only obtain transparency, but process that data into actionable information.

"Managing the TTUS program dynamically, utilizing a portable alpha construct and substantial MCA programs has proven to be a better risk-adjusted approach for the university system," said Tim Barrett, Chief Investment Officer of TTUS.  "With that said, it comes with complexities, such as tying the beta and alpha returns together for reporting, independent derivative pricing, as well as separate account administration for the numerous MCAs.  With the right partner, any pension or endowment can overcome these hurdles. In our case, Maples Fund Services has been our partner to address these administrative issues."

In the years that DPM has been in place at SBCERA, it has helped generate significant additional compounded cash returns over its own policy as well as traditional 60/40 program benchmarks with dramatically less risk.  Similarly, at TTUS, since Tim Barrett’s arrival five years ago, DPM has enabled persistently strong risk-adjusted performance.  In a strongly rising market, the total return for these plans is typically in line with their peer group with significantly less volatility and downside risk.  In flat to down markets they often strongly outperform their peers through their greater focus on cash flow and significantly reduced drawdown risk.  With their typically higher levels of cash reserves, lower equity allocations, and greater access and ability to dynamically hedge their portfolios, these dynamic portfolios are likely to significantly outperform peers during the next market downturn.


 

Conclusion

DPM empowers CIOs and their teams with a set of tools that may be used to complement their existing portfolio structure and enhance their risk-adjusted performance.  These tools are based on time tested investment principals of diversification, valuation and margin of safety.  This dynamic process can be enhanced with the right set of strategic partners to help drive the dynamic aspects of a portfolio's asset allocation and to create market intelligence and insight from a portfolio’s data.  While this risk-focused process is somewhat more complex than a beta-driven process, it is achievable – even with small but enterprising teams as well as high quality strategic partners to provide operational and administrative support – and offers the potential of allowing plans to meet their long-term investment objectives in the lower return environment that we may all soon be facing

¹ Kunz, Samuel and Muralidhar, Arun, Dare to Be Different: How to Commit to Investment Beliefs Through Knowledge Management (March 14, 2018). 2
Available at SSRN: https://ssrn.com/abstract=3140928

² Kunz and Muralidhar, 2018.

³ Barrett, Timothy and Pierce, Donald and Perry, James and Muralidhar, Arun, Dynamic Beta: Getting Paid to Manage Risks (December 1, 2011).
Journal of Investment Consulting, Vol. 12, No. 2, pp. 67-78, 2011. Available at SSRN: https://ssrn.com/abstract=2004667