Qn 6 of R35 EOC is confusing. It asks to identity which attribution analysis method is suggested with two statements. The relevant parts of the vignette is below.
Stephanie Tolmach is a consultant hired to create a performance attribution report
on three funds held by a defined benefit pension plan (the Plan). Fund 1 is a domestic
equity strategy, Fund 2 is a global equity strategy, and Fund 3 is a domestic fixedincome
Tolmach uses three approaches to attribution analysis: the return- based, holdings based,
and transaction- based approaches. The Plan’s investment committee asks
Tolmach to (1) apply the attribution method that uses only each fund’s total portfolio
returns over the last 12 months to identify return- generating components of the
investment process and (2) include the impact of specific active investment decisions
and the attribution effects of allocation and security selection in the report.
Question 6 is:
Of the three attribution approaches referenced by Tolmach, the method requested by the committee:
A is the least accurate.
B uses the underlying holdings of the actual portfolio.
C is the most difficult and time consuming to implement.
Answer is A:
A is correct. The committee described a return- based attribution, which is the least accurate of the three approaches (the return- based, holdings- based, transaction- based approaches). Return- based attribution uses only the total portfolio returns over a period to identify the components of the investment process that have generated the returns.
The answer seems to just take in point (1) which is clearly returns-based.
But point (2) in the vignette does point to a holding’s based approach as well: include the impact of specific active investment decisions and the attribution effects of allocation and security selection in the report.
Anybody can shed light on what I may be missing here?
Computing the security selection effect using, say, Brinson-Fachler, is not holdings-based; it’s returns-based.
Greetings friend! What point (2) seems to be saying is the committee is asking Tolmach to break down the returns by allocation effect and security selection effect. This will give the impact of specific active investment decisions “generally/overall for the time period,” whether they were good or bad decisions to make. For example, the allocation effect shows how specific active decisions to overweight or underweight sectors performed versus simply tracking the benchmark. The security selection effect shows the cumulative result of the active stock picking capability of the manager within sectors, due to their cumulative specific active security selection decisions within a sector versus the benchmark’s performance. While both will give information about the result of active decisions, they don’t necessarily track to your holdings or transactions item-by-item.
So point 2 is just describing the security and allocation effects, while trying to draw your attention to the words “impact of specific active investments” and getting you to forget the point 1 wording “uses ONLY each fund’s total portfolio returns…”
Answer A describes return-based approaches. They are the easiest to create but also the least accurate. Answer B refers to holdings-based approaches. They are of the 3 approaches in the middle for difficulty and accuracy and are based on actual holdings. Answer C describes transaction-based approaches. These are the most accurate because they comprise the actual transactions in real-time (as opposed to holdings-based approaches which are based on beginning of period holdings but don’t track to subsequent intra-period holdings changes in real-time), but they are also the most time consuming and difficult to track/implement.
Cheers - good luck - you got this
I got it. Thanks S2000magician and Greybeard_The_Elder!
Appreciate the help!