Module 3: Overview of Asset Allocation

What is investment governance?

Investment governance ensures that appropriate individuals or groups make informed investment decisions and conduct oversight activities on behalf of investors. The objective of effective governance is to match the investor’s objectives with their constraints while ensuring that investment decisions comply with relevant laws and regulations. Investment governance also seeks to improve investment performance by aligning asset allocation with implementation.

Describe elements of effective investment governance and investment governance considerations in asset allocation.

Effective investment governance models do the following:

• Establish long-term and short-term investment objectives.

• Allocate rights and responsibilities within the governance structure.

• Specify processes for creating an investment policy statement.

• Specify processes for creating a strategic asset allocation.

• Apply a reporting framework to monitor the investment program’s stated goals and objectives.

• Periodically perform a governance audit.

Strategic Asset Allocation (SAA)

The investment committee, which is the highest level of the governance structure, will typically approve the SAA decision. A proposed asset allocation will be developed after (1) the IPS is constructed, (2) investment results are simulated over the appropriate time horizon(s), and (3) the risk and return attributes of all possible asset allocation strategies are considered. In addition to approving the asset allocation, good governance should also specify rebalancing decisions and responsibilities.

Reporting Framework

A reporting framework allows stakeholders to evaluate performance, investment guideline compliance, and the investment program’s progress toward achieving its stated goals and objectives

Economic balance sheet

an economic balance sheet contains an organization’s financial assets and liabilities, as well as any nonfinancial assets and liabilities that are applicable to the asset allocation decision. These nonfinancial assets and liabilities are referred to as extended portfolio assets and liabilities because they are not included on traditional balance sheets.

Compare the investment objectives of asset-only, liability-relative, and goals-based asset allocation approaches.

There are three types of asset allocation approaches: (1) asset only, (2) liability relative, and (3) goals based. These asset allocation approaches attempt to match investors’ goals with their optimal level of risk.

Asset-only approaches

Asset-only approaches make asset allocation decisions based solely on the investor’s assets. An example of an asset-only approach is mean-variance optimization (MVO), which incorporates the expected returns, volatility, and correlations of asset classes. The investment objective for this approach is to maximize the expected return per unit of risk (e.g., maximize the Sharpe ratio). The chosen investments should consider investor constraints (stated in the IPS) as well as investor risk tolerance.

Liability-relative approaches

Liability-relative approaches involve asset allocation decisions based on funding liabilities, with the objective of paying liabilities when they come due. An example of a liability-relative approach is surplus optimization, which is based on principles from mean-variance asset allocation. The surplus is computed as the investor assets value minus the present value of investor liabilities. Modeling liabilities may be achieved by shorting an amount of bonds that matches the duration and present value of liabilities. Liabilities may also be modeled by creating a portfolio designed to hedge the liabilities. Asset allocation focused on funding liabilities is also known as liability-driven investing (LDI).

Goals-based approaches

Goals-based approaches are geared toward asset allocations for subportfolios, which help individuals or families achieve lifestyle and aspirational financial objectives. For example, goals could involve maintaining a current lifestyle or donating money to a university at some point in the future. To achieve the stated goals, it is necessary to specify the type of cash flows needed (e.g., even, uneven, or bullet payment), the time horizon(s), and the level of risk tolerance in terms of the probability of attaining a certain goal. Each sub portfolio will have a unique asset allocation designed to meet the stated goals. Summing these asset allocations will produce the investor’s overall portfolio SAA. Asset allocation focused on the investor’s goals is also known as goals-based investing (GBI).

Diffe

Both liability-relative and goals-based asset allocation approaches are based on meeting liabilities. The main difference is that liability-relative approaches focus on liabilities of institutional investors, while goals-based approaches focus on liabilities of individual investors. Institutional investors have legal obligations or debts, whereas individual investors wish to meet specific lifestyle goals. This suggests that penalties for not meeting liabilities are much higher for liability-relative approaches.

Contrast concepts of risk relevant to asset-only, liability-relative, and goals-based asset allocation approaches.

Risk concepts associated with asset-only approaches focus on asset class risk as well as constructing effective asset class combinations. The relevant risk measure for MVO, the most popular asset-only approach, is the standard deviation of portfolio returns, which incorporates asset class volatilities and asset class return correlations. Other risk sensitivities, such as relative risk and downside risk, can also be measured with a mean-variance framework. For example, the risk relative to a benchmark can be modeled with tracking error, and downside risk can be modeled with value at risk (VaR), semivariance, or maximum drawdown.

Monte Carlo simulation is a statistical modeling tool often used to complement MVO. For example, a manager could begin by selecting several optimal portfolios using MVO that have acceptable risk and return for the client. The manager could then use Monte Carlo simulation to generate multiple simulated paths, which display how these portfolios would perform over time. This action would provide useful information on downside risk when portfolios encounter a market stress scenario. These results can then be used to refine asset allocation decisions.

Contrast concepts of risk relevant to asset-only, liability-relative, and goals-based asset allocation approaches.

Risk concepts associated with liability-relative approaches focus on not having enough assets to pay liabilities when they come due. The volatility of contributions used for funding liabilities is also a risk. The standard deviation of the surplus may be used as the relevant risk measure. In general, the differences between asset and liability characteristics (e.g., size, sensitivity to interest rate changes) are the main drivers of risk for liability-relative asset allocation approaches.

Contrast concepts of risk relevant to asset-only, liability-relative, and goals-based asset allocation approaches.

Risk concepts associated with goals-based approaches focus on the risk of not being able to achieve the stated financial goals. If an investor has multiple goals, then the risks will encompass multiple future time periods. Thus, portfolio risk under a goals-based approach is the weighted sum of the risk that is attached to each goal.

Describe the use of the global market portfolio as a baseline portfolio in asset allocation.

Financial theory proposes that the first asset allocation to consider should be the global market portfolio. This portfolio contains all available risky assets (e.g., global equity, global fixed income, real estate, etc.) in proportion to their total market values. It is also the portfolio that minimizes diversifiable risk because it is the most diversified portfolio possible. The market portfolio is found on the efficient frontier by drawing a line from the risk-free asset that is tangent to the efficient frontier (known as the capital market line). The point of tangency is the location of the global market portfolio.

Tactical asset allocation (TAA)

Tactical asset allocation (TAA) is an active management strategy that deviates from the SAA to take advantage of perceived short-term opportunities in the market. TAA introduces additional risk, seeking incremental return, often called alpha. These deviations from the SAA weightings by asset class should be restricted by risk budgets or rebalancing ranges that control the amount of deviation. The deviations may be based on forecasted asset class valuation, business cycle state, or stock price momentum. A multiperiod view of the investment horizon is sometimes referred to as dynamic asset allocation (DAA). DAA recognizes that asset (and liability) performance in one period affects the required rate of return and acceptable level of risk for subsequent periods. Changes to the SAA may be limited to simply adjusting the mix between stocks, bonds, and cash. Conversely, global TAA may involve a broader and more complex multiasset approach.

With TAA, there is a tradeoff between potential outperformance and tracking error. A key limitation of this approach is the additional trading and monitoring costs as well as possible capital gains taxes. Thus, the decision to implement a TAA should be evaluated under a cost-benefit approach.

Passive and active management choices

Under passive management, investor insights or expectations do not impact the composition of the portfolio. Examples of a passive approach include indexing or holding bonds to maturity. With indexing, the portfolio may add or drop positions based on the index holdings, but it would not react to the changing expectations of investors regarding valuations. In contrast, under active management, the portfolio composition changes when investor insights or expectations change. The goal of active management is to earn risk-adjusted returns that exceed an associated passive benchmark.

The decision of where to invest along the active/passive spectrum depends on the following factors:

• Availability of appropriate investments (e.g., a relevant index)

• Active management scalability in terms of value added from each active decision

• Investor constraints, such as social and environmental concerns, when using a passive approach

• An investor belief in efficient markets, which would discourage the use of active management

• The cost-benefit tradeoff where additional transaction costs are needed for achieving excess returns

• The tax status of investors, which differs between taxable and tax-exempt investors

Risk Budgeting

Risk budgets can be established in either relative or absolute terms and stated in either money or percentage terms (e.g., 10% return volatility). These risk budgets specify how risk should be distributed among portfolio assets without regard to asset expected returns.

Rebalancing Approaches

Calendar rebalancing is rebalancing the portfolio to its strategic allocation on a predetermined, regular basis (e.g., monthly or quarterly). Generally, the frequency of rebalancing depends on the volatility of the portfolio, but sometimes rebalancing is scheduled to coincide with review dates.

The primary benefit to calendar rebalancing is that it provides discipline without the requirement for constant monitoring. The drawback is that the portfolio could stray considerably between rebalancing dates and return to its strategic allocation ranges on the rebalancing date. In other words, rebalancing is related to the passage of time rather than the value of the portfolio.

Rebalancing Approaches

With percentage-range rebalancing, rebalancing is triggered by changes in value rather than calendar dates. The manager sets what are called tolerance bands or corridors that are considered optimal for each asset class. For example, a corridor of 50% ± 5% would indicate that the related asset class must stay within a band of 45%–55%. If the asset class wanders outside that corridor, which would no doubt mean other classes have violated their corridors, the portfolio is rebalanced.

By not waiting for specified rebalancing dates, range-based rebalancing provides the benefit of minimizing the degree to which asset classes can violate their allocation corridors. Cost is increased by the time and expense of constantly monitoring the portfolio (as compared to only checking valuations on the specified calendar dates) and by potentially more frequent trading.

Strategic Considerations for the optimal width of a corridor for an asset class in a portfolio

• Transaction costs. Obviously, the more expensive it is to trade, the less frequently you should trade. If an asset is particularly illiquid, for example, trading can be quite expensive. In that case, the corridor for the class should be wide. In general, the more illiquid the asset, the wider the corridor.

• Risk tolerance. More (less) risk-averse investors will have tighter (wider) rebalancing corridors.

• Correlations. The more highly correlated the assets (allocations) in a portfolio, the less frequently the portfolio will require balancing. If all assets tend to move together, their values will tend to stay within acceptable ranges.

• Momentum. If investors believe that current trends will continue, an argument can be made for using wider rebalancing corridors. Conversely, if investors anticipate mean reversion, tighter rebalancing corridors should be applied.

• Liquidity. Illiquid investments, such as private equity and real estate, are typically associated with larger trading costs. These liquidity costs encourage the use of wider rebalancing corridors.

• Derivatives. Rather than selling underlying assets, a derivatives overlay strategy can be used to synthetically rebalance a portfolio. This approach results in lower transaction costs, lower taxes, and can be executed quicker and more easily compared to rebalancing with only the underlying stock and bonds. The tradeoff is that derivatives require additional risk management when used as a rebalancing tool.

• Taxes. When making rebalancing decisions, taxes must be considered because realized capital gains and losses will impact investor taxes. Therefore, taxable portfolios will typically have wider rebalancing corridors than tax-exempt portfolios. The corridors may also be asymmetric due to tax savings (i.e., loss harvesting). This suggests that the range may be less below a certain target weight than above (e.g., the tolerance band may go from 48% to 55% for a 50% target weight).

Asset class volatility

Asset class volatility also has an impact on optimal corridor width. In the simplest situations, higher volatility most likely calls for narrower corridors to control risk. But the impact of volatility can be more complex than that. Higher volatility (in the absence of high positive correlation between classes) will lead asset class weights to shift more quickly.

common tilts when implementing the global market portfolio is to overweight

when implementing the global market portfolio, common tilts (biases) include overweighting the home-country market, value, size (small cap), and emerging markets