mast image

Special Report

Impact investing


SMART portfolio management

Static portfolio management techniques fail because they cannot respond to changing economic conditions. Arun Muralidhar offers an alternative

Last year was a watershed for the investment management industry as dramatic market movements exposed the flaws in the theory and practice of pension fund management. Solvency declined dramatically, globally, as the asset-liability mismatch was exposed (as was the irrelevance of bad LDI); hedge funds did not deliver on the promise of generating alpha (for very high fees); rebalancing policies detracted value because they anchored themselves to a declining portfolio; liquidity dried up; and equity became the only asset investors could sell, causing further problems. However, there had been warning signs, and many analysts had shouted "wolf!" only to be ignored.

The primary source of the problem is financial theory and its standard bearer - the capital asset pricing model (CAPM). Academics who have never managed pension funds ignored the most dominant class of institutional investors - those that delegated decisions to others - and the impact of their behaviour on markets. First, these investors worry about relative risk and relative performance, which has an impact on the choice of investment strategy and the evaluation of risk-adjusted performance and agents.

Second, the whole focus on optimal portfolios (à la Markowitz) made assumptions about asset correlations that masked the bets that investors were making on markets, as well as the factors that drive these bets.

Finally, the CAPM-type results focused on static solutions to portfolio problems, ignoring the time dimension (the inter-temporal CAPM of Merton notwithstanding) in making decisions.

The result was that CIOs had to deal with four major shortcomings:

The fixation of the investment management industry on static prescriptions to manage assets in dynamic markets, especially for long-term strategic asset allocations (SAAs) and naïve rebalancing; Asset managers who offered naïve ‘magic bullet' solutions to sell products rather than solve pension fund problems, most evident in LDI and multi-manager programmes (which, again, are static solutions); Performance measures that did not adjust for risks or skill and, hence, served the asset manager more than the pension plan - most evident in hedge funds, but quickly followed by mainstream asset managers; and Benchmarks recommended to pension funds that are difficult to replicate in the futures market, making it hard for CIOs to be nimble in managing a fund (without taking ‘fake' tracking error), especially as markets zigged and zagged. The value attached to dynamic decision making was miniscule, with many analysts deriding ‘market timing', without realising that every bet in a portfolio, starting with the SAA, is market timing.

We are bound to repeat our mistakes if we do not learn from them, and while there has been considerable introspection on expected return assumptions and volatility, relatively little attention has been paid to the correlation statistic. Brushing this critical factor under the carpet has definitely hurt the interests of the investor. Ignoring the fact that correlations across two assets may be dynamic, investors must focus on a much simpler problem - that of understanding the implied bet in choosing a correlation value. A low correlation between stocks and bonds occurs because they respond differently to economic growth, interest rates, oil, and so on. The same is true for every other correlation statistic between two assets or two managers. Therefore, in setting an SAA or in selecting a portfolio of managers and assuming specific correlations (and expected returns), pension funds are making a bet on these economic relationships, and it is inconsistent to neglect to exploit them in the implementation and management of a portfolio.

Static policies for dynamic markets are undoubtedly flawed and have to be changed with the support of appropriate liquid, transparent and low cost benchmarks; implicit bets (especially in the SAA, rebalancing and liability and currency hedges) need to be made explicit and managed; naïve performance measures have to be improved; and the CAPM needs to be revamped dramatically. But this process can only start with investors taking the time to understand how various market factors influence assets or managers and then develop a set of rules so that, as the factors evolve over time, the optimal portfolio evolves simultaneously.

As Woody Brock of consultancy Strategic Economic Decisions states, the future is less about optimal portfolios than about optimal strategies, and effective CIOs will be those who establish optimal portfolios for specific states of the world and then dynamically adjust their portfolios as the market moves from one state to the next. Developing rules to track market movements, and their impact on all beta and manager decisions, will create a systematic process that generates consistent recommendations that are not easily influenced by emotion (although clearly leaving scope for applying informed judgment in the implementation of these rule recommendations).

SMART (systematic management of assets using a rules-based technique) management of assets and liabilities leads to improved solvency and a lowering of ALM risks. SMART is about introducing good process - namely, only measured and monitored risks can be managed. Hence, the process of developing rules to make explicit the underlying factor relationships alerts overburdened and under-resourced CIOs to make key decisions for better solvency (and not just a return over an investment benchmark), and to make sure that the managers they hire generate, and are compensated for, risk and skill-adjusted performance. SMART is also, therefore, a call to arms to change the way managers are compensated so that the bulk of the fees can be deferred until skill is established.

To summarise, pension funds should incorporate three levels of dynamism in managing the assets and liabilities to improve solvency: 1. Dynamic LDI; 2. View neutral solvency-based beta adjustment; and 3. View-based (or market factor-based) SMART beta and alpha management. Explicit factor analysis and exposition of these rules lend themselves to transparency and good governance, whereas optimised portfolios are derived from black boxes where the investor is not sure whether the decision is driven by the return, correlation, or volatility assumption. SMART presents a new design for pension fund management that allows CIOs to be ‘smart' about managing assets relative to liabilities and at the same time allows them to access alpha flexibly, manage managers dynamically for true alpha, and improve solvency.

It is our hope that we do not follow Mark Twain's advice to "Never put off until tomorrow what you can do the day after tomorrow."

Arun Muralidhar is chairman and CIO at Alpha Engine Global Investment Solutions


Have your say

You must sign in to make a comment


Your first step in manager selection...

IPE Quest is a manager search facility that connects institutional investors and asset managers.

  • QN-2540

    Asset class: All/Large Cap Equities.
    Asset region: UK.
    Size: The fund will be added to our guided fund range.
    Closing date: 2019-05-27.

  • QN-2541

    Asset class: Small/Mid-Cap Equities.
    Asset region: Switzerland.
    Size: CHF 130m.
    Closing date: 2019-06-04.

  • QN-2542

    Asset class: All/Large Cap Equities.
    Asset region: Switzerland.
    Size: CHF 130m.
    Closing date: 2019-06-04.

Begin Your Search Here