The asset allocation material in the level III curriculum is largely conceptual but you do need to be able to do some basic calculations to find the optimal portfolio and other risk-return concepts.
Download the last three years’ essay exams and guideline answers from the CFA Institute’s website to follow along with the notes below.
A lot of the conceptual material revolves around the different approaches to asset allocation, i.e. Mean Variance, Resampled Efficient Frontier, Black-Litterman, Monte Carlo, ALM, and Experience-based.
Understanding the advantages and disadvantages (looking at past essay exams, something the Institute loves to test) will help you get an idea of what each is trying to do and in what situations it is most appropriate.
Be ready to choose the most appropriate approach and list a few strengths/weaknesses on the exam.
Remember, strategic asset allocation is long-term whereas tactical allocation is short-term. Target weights are set in the policy portfolio and then can be adjusted through tactical allocation.
Asset-Liability Matching (ALM) models liabilities to allocate assets to best meet these financial obligations.
It is considered less risky than asset-only asset allocation which only targets the highest level of return for a given risk tolerance.
The item set for Asset Allocation on last year’s CFA essay exam was fairly brief. It was worth 20 points, or about 5.5% of your total exam points.
Skimming the questions quickly, we see that we probably already know much of the answer without even reading the vignette.
You will need to refer to information in the case within the answers, but the core part of the answers is basic strengths/weaknesses kind of material.
A. Advantages to using Resampled Frontier are stability and diversification (relative to MVO). From here, we just need to pick out why Finnegan might want stability and diversification (i.e. reduced turnover and lower risk).
The advantages to Black-Litterman are ability to incorporate viewpoints and diversification.
The advantages to Monte-Carlo are the ability to model different inputs (like changing tax rates) and path-dependent modeling.
B. Anytime you see ALM, note how important liabilities are in the client or institution’s profile.
Institutions that are quasi-trusts (banks and insurance) or clients with low risk tolerance often prefer ALM because it specifically addresses their liabilities in the model. This provides a greater assurance of meeting those liabilities.
C. The human capital piece of asset allocation often depends on two variables: whether the client’s income is bond or equity-like, and their time horizon.
The Institute likes two ‘scenarios’ here, a capital market professional (who is going to have a risky, equity-like income correlated with the stock market) and a university professor (who, given tenure, is going to have a low-risk, bond-like income stream).
This will determine the appropriate riskiness of financial capital and asset allocation.
Again, some basic and conceptual stuff which should be easy points on the test. Start with understanding the concepts, strengths and weaknesses, and then move on to the formulas.
Only a few weeks left. Let me know if you want any questions covered.
‘til next time, happy studyin’
Joseph Hogue, CFA