TheronGroup of Consultants

80/20 Golden Rule · Pareto Optimality

Worked example

INV4801: return requirements, utility and corner portfolios worked

UNISA's Jan/Feb 2024 INV4801 Question 3 worked in full: before-tax return requirements, utility-adjusted returns, corner portfolios and the tangency portfolio.

Asset allocation in INV4801 runs from the client’s needs to the portfolio that meets them: what return the client requires, how much risk they will bear, and which mix on the efficient frontier delivers it. Question 3 of UNISA’s January/February 2024 INV4801 paper, worth 25 marks, takes two clients through every step.

The question is described in our own words below; the answers are TheronGroup’s.

The case

An analyst at an investment firm is setting static asset allocations for two new clients.

  • Pfende is 45, an experienced investor who allocates strategically. He left his previous firm because the 12% return it promised could only come from taking more risk than he wanted. He needs about 3% a year for spending and puts his risk aversion at 8 on a scale of 1 to 10, where 10 is the least tolerant of risk.
  • Manjengwa is 22 and recently graduated. He likes trading on mispricings but works 60-hour weeks, so he has handed the account to the firm. He needs about 5% a year for spending and puts his risk aversion at 2.

The analyst wants asset classes that diversify well and are largely liquid, but he does not think the classes need to cover most of the world’s investable assets, and he does not think an asset needs to sit in more than one class. He has built four corner portfolios:

Corner portfolio Expected return Standard deviation Sharpe ratio
1 6,50% 5,95% 0,756
2 7,25% 8,30% 0,633
3 8,00% 11,15% 0,538
4 8,75% 14,25% 0,474

(a) 6 marks. With expected inflation of 2% and a tax rate of 25% for both clients, find each client’s minimum before-tax return requirement.

(b) 4 marks. The analyst recommends corner portfolio 1 to Pfende and corner portfolio 4 to Manjengwa. Calculate each client’s utility-adjusted return.

(c) 5 marks. Using the corner portfolios, calculate the standard deviation of a portfolio expected to return 7,5%.

(d) 4 marks. With a risk-free rate of 2,5% and leverage permitted, find the weights in the risk-free asset and the tangency portfolio that give an expected return of 7,5%.

(e) 6 marks. Discuss three factors that matter when asset classes are specified.

(a) The minimum before-tax return requirement

Each client’s spending rate is a real, after-tax need: it is what he spends while his capital keeps its purchasing power. The requirement is turned into a nominal, before-tax return in two steps. Inflation is compounded in first, and the tax is grossed up last, because tax falls on the whole nominal return.

Pfende
Nominal after tax  = (1 + 0,03) × (1 + 0,02) − 1
                   = 1,0506 − 1
                   = 5,06%
Before tax         = 5,06% ÷ (1 − 0,25)
                   = 5,06% ÷ 0,75
                   = 6,7467%

Manjengwa
Nominal after tax  = (1 + 0,05) × (1 + 0,02) − 1
                   = 1,0710 − 1
                   = 7,10%
Before tax         = 7,10% ÷ (1 − 0,25)
                   = 7,10% ÷ 0,75
                   = 9,4667%

The minimum before-tax nominal return for the coming year is 6,75% for Pfende and 9,47% for Manjengwa.

Where the marks are lost. The paper does not fix the order of the adjustments, so state the one you use. Adding inflation to the spending rate before grossing up gives (3% + 2%) ÷ 0,75 = 6,67% and (5% + 2%) ÷ 0,75 = 9,33%, and a script that names that basis is marked on it. What earns nothing is grossing up the spending rate and then adding inflation untaxed (6,00% and 8,67%): that assumes the inflation part of the return escapes tax, which it does not.

(b) Utility-adjusted returns

The utility-adjusted return penalises expected return by the investor’s risk aversion and the portfolio’s variance: U = E(R) − 0,5 × A × σ², with the return and standard deviation as decimals and the risk aversion score as A.

Pfende on corner portfolio 1 (A = 8)
U = 0,0650 − 0,5 × 8 × 0,0595²
  = 0,0650 − 4 × 0,00354025
  = 0,0650 − 0,014161
  = 0,050839  = 5,08%

Manjengwa on corner portfolio 4 (A = 2)
U = 0,0875 − 0,5 × 2 × 0,1425²
  = 0,0875 − 1 × 0,02030625
  = 0,0875 − 0,020306
  = 0,067194  = 6,72%

The utility-adjusted return is 5,08% for Pfende and 6,72% for Manjengwa. Set against part (a), neither recommended portfolio meets its client’s requirement: portfolio 1 returns 6,50% against Pfende’s 6,75%, and portfolio 4 returns 8,75% against Manjengwa’s 9,47%. The utilities stand as calculated on the portfolios the question assigns, but the shortfall is worth one line in the answer.

Where the marks are lost. Work in decimals. With percentages, 0,5 × 8 × 5,95² = 141,61, and the “utility” becomes −135,11%. The answer is a decimal return, 0,050839, read as 5,08%.

(c) The standard deviation of a 7,5% portfolio

A target return between two adjacent corner portfolios is reached by combining those two corners. The 7,5% target lies between corner portfolio 2 (7,25%) and corner portfolio 3 (8,00%).

Weight in corner 3  = (7,50% − 7,25%) ÷ (8,00% − 7,25%)
                    = 0,25% ÷ 0,75%
                    = 0,3333
Weight in corner 2  = 1 − 0,3333
                    = 0,6667

Check: 0,6667 × 7,25% + 0,3333 × 8,00% = 4,8333% + 2,6667% = 7,50%

Standard deviation  = 0,6667 × 8,30% + 0,3333 × 11,15%
                    = 5,5333% + 3,7167%
                    = 9,25%

A portfolio of 66,67% in corner portfolio 2 and 33,33% in corner portfolio 3 earns 7,50% with a standard deviation of 9,25%.

Where the marks are lost. Use the two corners that bracket the target, not corner 1 and corner 4. The weighted-average standard deviation is the approximation the corner portfolio theorem is examined on: adjacent corner portfolios are very highly correlated, so 9,25% is close to the true figure and is, if anything, slightly above it. Say so in a line rather than leaving it unstated.

(d) Leverage on the tangency portfolio

With a risk-free asset available, the best risky portfolio is the one with the highest Sharpe ratio, and the investor reaches any target return by lending or borrowing at the risk-free rate. The Sharpe ratios in the table are struck at a risk-free rate of 2%, so they are recomputed at the 2,5% given here.

Corner portfolio Sharpe ratio at 2,5%
1 (6,50% − 2,50%) ÷ 5,95% = 0,6723
2 (7,25% − 2,50%) ÷ 8,30% = 0,5723
3 (8,00% − 2,50%) ÷ 11,15% = 0,4933
4 (8,75% − 2,50%) ÷ 14,25% = 0,4386

Corner portfolio 1 remains the tangency portfolio.

Weight in tangency portfolio = (7,50% − 2,50%) ÷ (6,50% − 2,50%)
                             = 5,00% ÷ 4,00%
                             = 1,25
Weight in risk-free asset    = 1 − 1,25
                             = −0,25

Check: 1,25 × 6,50% + (−0,25) × 2,50% = 8,125% − 0,625% = 7,50%
Standard deviation           = 1,25 × 5,95% = 7,4375%

The client invests 125% in corner portfolio 1 and holds −25% in the risk-free asset, that is, borrows 25% of the portfolio’s value at 2,5%. The levered portfolio earns 7,50% with a standard deviation of 7,44%, against 9,25% for the unlevered mix in part (c): the same return for less risk, which is the point of allowing leverage.

Where the marks are lost. The negative weight is the answer, not an error: −25% in the risk-free asset means borrowing. And the tangency portfolio must be chosen at the risk-free rate the part gives. Here the ranking does not change between 2% and 2,5%, but the script has to show it.

(e) Three factors in specifying asset classes

  1. Diversification. The classes should not be so highly correlated with one another that one duplicates another’s risk; a class that adds nothing to diversification adds only cost. The analyst’s wish for a well-diversified set meets this factor.
  2. Preponderance of investable wealth. Together, the classes should cover most of the world’s investable wealth, so that the opportunity set is not needlessly narrow. The analyst’s view that most investable assets need not be covered fails this factor, and it restricts the efficient frontier the clients can reach.
  3. Mutual exclusivity. Each asset should belong to one class only, because an asset counted in two classes distorts the correlations and makes the optimised weights meaningless. The analyst’s view that an asset need not sit in more than one class is consistent with this factor.

Two further factors can be used where the question allows: the assets within a class should be homogeneous, and each class should have the capacity to absorb a meaningful share of the portfolio without hurting liquidity, which is where the analyst’s preference for liquid assets belongs.

Where the marks are lost. Six marks for three factors means two marks each: one for the factor and one for applying it to the case. A list of factor names with no reference to the analyst’s views earns half.

Studying INV4801

Return requirements, utility and the corner portfolio theorem come back in different cases across the sittings, and each turns on a short formula applied in the right order. The INV4801 course holds eighteen parts of notes and five CFA supplements, 18 examination sittings and 12 assignments worked, three exam packs, and a formula sheet and memorisation pack, at R3 000 a year.

Get in touch

Tell us your module

Send the module code and your sitting. We reply with exactly what exists for it today, a free extract from that module, and what it costs.

Students

Name your module code and sitting. You get a free extract from your own module — not a generic brochure — before any payment.

Send an enquiry

Free extract

A full worked example and a note extract from the module you are sitting, sent the same week you ask.

Request the extract

Employers and professional bodies

In-house governance, risk and internal-audit training, and bulk licensing for staff in part-time postgraduate study.

Corporate enquiry