|
Asset Allocation Analytics
Capital
Market Assumptions
The asset allocation decision is
correctly recognized as having the greatest impact on long-term return and
risk. There is no simple formula to determine the optimal asset mix. Asset
allocation decisions should be based on a careful examination of a number of
factors including the current outlook for the financial markets, fund
objectives, time horizon and tolerance for risk. The Asset Allocation module
in the Wilshire Compass combines fund specific factors and
advanced optimization technology to assist users in evaluating and
establishing asset allocation and rebalancing policies
The Asset
Allocation module provides access to Wilshire’s database of index returns
including over 1,400 capital market and economic time series. The index
database supports ongoing capital market research and the development of
long-term capital market return, risk and diversification expectations, which
are then used as the inputs for asset allocation optimization.
Optimization
The Wilshire Compass asset allocation
model creates an “efficient frontier” of asset mixes based on selected
asset classes, capital market assumptions and user-defined basic (single
asset class) and advanced (multiple asset class) constraints. In fact, up to
three sets of constraints can be modeled simultaneously resulting in three
efficient frontiers being plotted. Uses include showing the curve in a
totally unconstrained environment and then the shift in the curve as
constraints are placed on the optimization.
The Wilshire Compass
also allows the user to incorporate non-funded asset classes such as foreign
currencies into asset allocation work. Depending on the fund’s objectives
and asset mix, there may be a desire to reduce currency risk by converting or
hedging foreign currency positions back to U.S. dollars. From an asset
allocation perspective, the critical issue is balancing the potentially
positive diversification benefits of foreign currencies, against the
increased risk that foreign currencies introduce and the costs of
implementing a strategic hedging program. With the Wilshire
Compass users can easily incorporate their own assumptions as they relate
to currency returns and risks and the expected costs of hedging in order to
determine a fund's optimal hedge ratio.
The Asset Allocation module
also includes tools which help funds select the most appropriate asset mix on
the efficient frontier given the financial condition and time horizon of the
fund. The utility optimizer in the Wilshire Compass takes into
account the fund’s specific liability structure and calculates return hurdles
for each liability funding objective. The portfolio with the greatest
probability of satisfying the funding objective, subject to a risk penalty
related to the time horizon, is identified as maximizing investor utility.
The flexibility
of the Asset Allocation analytic allows users that are concerned
about minimizing changes in their asset/liability ratio to perform
surplus optimization. Compass has several time series that were
designed as proxies for a pension plan's liabilities which can be
used to model the impact different asset mixes have on surplus volatility.
Rebalancing
Because asset allocation
policy is the primary determinant of long-term performance, the magnitude
and timing of portfolio rebalancing decisions can significantly impact
portfolio results. A systematic and disciplined rebalancing program is an
essential component in total fund risk management.
The “optimal” rebalancing strategy is a
function of a fund’s philosophy, investment objectives, and risk
tolerance. The Wilshire Compass offers tools to weigh these
factors against back-tested results of passive and active rebalancing
strategies. To test the effectiveness of alternative passive rebalancing
methods in various market environments, the sponsor can select either
Calendar or Normal Mix strategies.
The Calendar rebalancing option will
rebalance assets back to policy targets in specified intervals (i.e.,
monthly, annually, etc.). The Normal Mix rebalancing option rebalances
assets back to policy targets when asset class weights exceed or fall
below pre-determined ranges. Users can assign specific ranges for each
asset class and measure the impact of alternative ranges on the
risk/return profile of the portfolio. Both rebalancing analytics allow
the user to reflect transaction costs in the analysis. An example of a
“Normal Mix” rebalancing over the last 20-year period is shown below.
Simulation
The Simulation Analytic uses a
statistical simulation technique called Monte Carlo simulations which
allow us to “pre-experience” the impact of capital market returns based on
the input parameters. Specifically, the multivariate simulation engine
generates simulated returns for each asset class and then combines those
asset class returns based upon the user defined portfolios to be modeled.
The simulation process can be controlled using “Sobol sequences” which
allows for convergence to true values with fewer trials.
Alternatively users may choose to evaluate the potential range of outcomes
using a parametric or cumulative density function model. Distributions of
Total Return and Market Value for three portfolio mixes (based on 1,000
trials) are shown below.
These simulation tools can also be used with the
Endowment Spending Policy Analytic to forecast the probability of maintaining
the funds inflation adjusted purchasing power.
Portfolios
may also be evaluated based upon the probability of earning a
specific hurdle rate of return, such as the return necessary to fund
the plan's liabilities.
Return
to
Product Summary
|
