Asset Allocation Flashcards
What are the three approaches to liability-relative asset allocation?
- Surplus optimisation
- Two portfolio approach
- Integrated asset liability approach
Surplus optimization
Surplus optimisation
Surplus return = (Change in asset value - Change in liability value) / initial asset value
Um = Expected surplus return - 0.005 x landa x Variance surplus return
Goal based approach
120 - age
60/40 split
1/n rule
Risk Parity asset allocation
(low risk assets higher or lower weights?)
Endowment model and Yale model
Goal based approach = sets minimum expectations and minimum required probability of success. Given 90% required success you pick the model with the highest expected return.
120 - age = % you allocate to equities
60/40 split = maintained constantly inline with the global portfolio.
1/n rule = Equally weighted portfolio with the same in each asset class
Risk Parity = Ensures each asset contributes the same amount of risk, therefore low risk assets will have larger weights
Endowment model and Yale model = large allocation to non-traditional Alternative assets. Active management.
Which asset class to endowment funds allocate most to?
Alternatives
Goal based approach return determination for
a) Institutions
b) Individuals
Goals based approach is GENERALLY for individuals and families. Stating hierarchical list of goals from essential to aspirational.
a) Institutions are based on mathematical expectations
b) Individuals are based on minimum expectations
Three risk budgeting approaches
- Total risk
- Active risk (asset allocation implementation)
- Residual risk
Tax considerations
Income vs capital
Two strategies for tax?
Interest income is usually taxed higher than capital gains.
US municipal bonds are tax exempt. Capital losses can offset gains.
- Tax loss harvesting. Realising losses against gains elsewhere.
- Strategic asset allocation = using ISAs andtax deferred accounts.
Measuring economic sentiment?
Margin borrowing
Short interest
Volatility indices
- Margin borrowing. Higher suggests investors are buying margin loans and is bullish
- Short interest = aggregate number of short sellers. A bearish signal.
- Volatility indices - measures bid-ask spread on index options. Increases with purchase of puts and decreases with purchase of calls.
Systematic hedge fund implementation
Vs
Discretionary
Systematic hedge fund implementation use computer algorithms and rules to determine which trades to make.
Vs
Discretionary use their instinct to determine when to trade.
Global macro managers
Global macro managers positions that are either directional (e.g., go long companies that are anticipated to benefit from expected interest rate hikes, and short companies that will be disadvantaged), or thematic (e.g., buy firms that will benefit from forthcoming free trade deals.) Leverage, often representing 600% or 700% of fund assets.
Single period Mean-variance optimization
Utility Equation
Risk Neutral investor landa =
Critique/Limitations
MVO identifies the portfolio allocations that maximize return for every level of risk. If the MVO analysis includes all investable risky assets, the result is the familiar “efficient frontier”
It uses ER (based on HISTORICAL DATA), standard deviation and a risk coefficient to identify the optimal portfolio. ER is sensitive to measurement errors. Correlation and Risk is not.
Utility = ER - 0.005 x Landa x SD^2
Risk neutral investoer (landa = 0) is fully high risk and would be 100% in Emerging markets for example. Low landa = high risk.
+ Inexpensive
+ Identified best risk/return trade off. Widely understood
+ Does allow for short positions
- Highly concentrated in a subset of assets
- Ignores trading costs and taxes
- Does not account for skewness and kurtosis
Reverse Optimisation
CAPM =
Weight of asset class =
Reverse optimisation Er =
Critique/Limitations
Unlike MVO which starts with Er and finds optimal portfolio. Reverse optimisation takes the optimal weights from the Global Market portfolio and uses CAPM to assess the overall return.
CAPM: ER = Rf + B x (Rp - Rf)
Note Rp - Rf = Risk Premium
Weight = Market cap % = US Equity 1,354 / Total Market cap 41,840 = 3.2%
Reverse optimisation ER = Sum of (Weight x ER)
Critique/limitations =
+ It creates a more diversified portfolio than MVO.
- Often alternative views exist regarding Expected returns and weights.
Black Litterman Model
An extension of the reverse optimisation however with adjustments based on Manager expectations and views (not simply based on CAPM).
You would look at the Er and adjust up or down.
Advantage
+ Incorporates investors views
+ well diversified as its anchored to the global benchmark
+ avoids the input sensitivity of MVO due to anchoring to the Global Market Portfolio
Disadvantage
- Does not allow for short selling (non-negativity)
LOS d Asset liquidity decisions
Issues with real estate, private equity and infrastructure plus solution?
These assets make it difficult to make capital market assumptions due to lack of accurate indexes. True risk and return representation from Private equity is difficult to track.
You can use real estate, private equity and infrastructure FUNDS as a proxy.
They cannot be readily diversified to remove idiosyncratic risk.
Monte Carlo simulation
Monte Carlo simulation complements MVO by addressing the limitations that MVO is a single period framework and can investigate many future possible scenarios including the effects of trading/rebalancing costs.
It can also paint a realistic picture of potential future outcomes, liklihood of meeting goals and potential maximum drawdowns.
