Financial Education

Asset/liability management approaches seek to match assets with future liabilities. The strategy is subject to three major risk factors:

1. Interest rate risk – assets decline in value as interest rates rise. If assets must be sold in order to meet one liability, a shortfall could arise with respect to future liabilities.
2. Contingent claim risk – contingent claims could halt interest payments or result in an early return of principal which would have to be reinvested at lower prevailing interest rates.
3. Cap risk – caps on floating rate payments prevent assets from performing in line with interest rates.

In a pure indexing strategy, the goal is to produce a portfolio that perfectly matches the benchmark. All bonds would be owned in the same proportion as their weight in the index.

Full replication is more commonly practiced in equity portfolios, as equities tend to be far more liquid than bonds. Bonds’ illiquidity makes the strategy difficult to implement. The difficulty, inefficiency and high costs of implementation result in this strategy being attempted only rarely.

Mean-variance optimization is subject to significant levels of error due to the need to estimate return, risk and correlation for each asset class included. As a result, no single optimization can lend much confidence in the output.

Many investors will take several simulations to generate efficient frontiers under a variety of assumptions regarding return, risk and covariance. These simulations can be integrated into a single resampled efficient frontier.

A resampled efficient frontier tends to result in a more diversified portfolio and an asset allocation that remains more stable over time.

Strategic allocation sets the investor’s long-term exposures to systematic risk. Tactical asset allocation (TAA) involves short-term adjustments to asset weights based on short-term predictions of relative performance.

While strategic allocations are revisited only periodically or when there is a change in the investor’s circumstances, TAA is an active, ongoing discipline. In effect, it is active management at the asset class level. It often takes place as an interim step between the strategic asset allocation and the actual asset class management decisions. Alternatively, the asset management can be set and a derivative overlay strategy used to alter the asset class weights tactically.

Tactical asset allocation is based on three principles:

1. Market prices explicitly describe the returns available (either cash yield or a yield + growth formula)
2. Relative expected returns reflect relative perceptions of risk
3. Markets are rational and mean reverting

Tactical asset allocation frequently seeks asset classes where risk premia are well above normal levels in anticipation of mean reversion. This requires judgment, as the mean reversion process can take many years.

If fiscal and monetary policy are both tight (or loose), economic growth is likely to slow (or accelerate.) When fiscal and monetary policy are at odds with each other, the outcome is less certain. The slope of the yield curve can help investors interpret the relative policy influences.

Loose monetary and fiscal policies encourage growth at the risk of inflation. As a result, the yield curve tends to be steeply up.

A loose monetary policy combined with tight fiscal policy has less inflation risk, and the yield curve slopes up more modestly.

Tight monetary policy combined with loose fiscal policy causes short term rates to be higher than normal, and the yield curve to be flat.

Tight monetary and tight fiscal policies should slow the economy and curb inflation, often resulting in an inverted yield curve.

The process of setting capital market expectations follows a general framework:

1. Specifying the final set of expectations that will be needed (which asset classes, risk, return, etc.) including the time horizon to which they apply
2. Research the historical record to determine past average performance as well as performance during periods similar to the current
3. Specify the method(s) and/or model(s) that will be used and the information that will be required in order to use them
4. Determine which source or sources can best provide the needed information
5. Interpret the current environment using the selected data and methods, applying experience and judgment
6. Provide the set of expectations that are needed, documenting all conclusions
7. Monitor actual outcomes and compare them to expectations, providing feedback to improve the expectations-setting process

Once the expectations for each asset class have been set, a strategic asset allocation strategy can be developed, maintained and modified as necessary.

At the CFA Institute annual conference in April/May 2007, Wharton professor Jeremy Siegel discussed the “age wave” in the U.S. and other developed nations.

The declining number of workers relative to retirees are increasingly less able to produce the needed goods and buy the assets of retirees, which in turn leaves retirees unable to fund their retirement. Possible solutions include raising the retirement age or allowing younger workers in developing nations to buy the retiree’s assets (which would transfer a massive amount of capital to developing nations.)

Siegel concludes that globalization is vital for the well-being of both developing and developed countries.

Endowments are permanent funds established in order to provide long-term benefits without depleting the initial principal value. The nature of endowments yields specific investment objectives and constraints.

Risk objective

Risks taken by endowments should be consistent with the goal of providing stable, real income over time (intergenerational equity.) Higher risk can be tolerated if the institution’s spending policies are adaptable to return volatility, or if smoothing mechanisms are used and previous investment performance has been strong.

Return objective

Endowments must earn a high return to maximize the sustainable real income generated. Volatility increases the risk of losing spending power.

Constraints

Endowments have minimal liquidity needs (enough to fund current spending) and a long-term time horizon. Tax considerations and legal issues are minimal, though care is needed to maintain tax exempt status and many endowments conform to UMIFA or an equivalent standard. Unique circumstances vary by endowment.

Defined benefit retirement plans offer participants a promised benefit in future years. Although each plan is different, managers of defined benefit plans typically face similar objectives and constraints.

1. Risk objectives must consider:
   • The plan’s funded status (assets relative to liabilities)
   • The sponsor’s financial status and profitability (ability to make contributions)
   • Common risk exposures between the sponsor and the plan
   • Specific plan features (inflation protection, etc.)
   • Workforce characteristics such as average age, expected retirement age, etc.
2. The return objective is typically to earn a sufficient return, adjusted for inflation, to meet expected liabilities and minimize the need for future contributions to the plan.
3. Constraints include:
   • Liquidity, defined as the difference between plan contributions and disbursements in a given year.
   • Long time horizon for ongoing plans, though workforce characteristics may result in differing time horizons.
   • Tax exempt status for investment returns, though contributions and disbursements may incur taxes.
   • Legal and regulatory constraints on investment policies, which are typically governed by ERISA or similar legal guidelines.
   • Unique circumstances may include the sponsor’s financial condition or specific investment prohibitions or desires. For example, a union-sponsored plan may prohibit investments in firms or countries considered to follow poor labor practices.

Most asset allocation analyses are based on a mean-variance framework to compare the risk and expected return.

In the September/October 2007 Financial Analysts Journal, William F. Sharpe presented an alternative approach to asset allocation based on maximizing expected utility using a more complex utility function than that permitted by mean-variance analysis.

The alternative approach represents a generalized version of traditional optimization. If the investor’s utility preference is assumed to be mean-variance maximization the results will be equivalent. Other utility preferences could lead to other outcomes.