June Return for a Daily-Valued Portfolio: Modified Dietz, Time Weighting and the Large-Flow Policy
A €1.2 billion balanced mandate is reported monthly. The custodian values the portfolio daily, but the firm's performance system still runs monthly Modified Dietz because the composite's other portfolios are valued only at month-end. For June the portfolio begins at €850.0 million. On 8 June the client contributes €60.0 million: the portfolio is valued at €880.0 million immediately before the flow and €940.0 million immediately after. It ends June at €941.6 million with no further external cash flows. The consultant asks for the June return, for the annualized since-inception return of the 11-month-old portfolio, and why the firm's factsheet shows 9.4% annualized. The analyst proposes reporting Modified Dietz of about 3.54% and annualizing it by 12/11 to 3.87%, arguing that daily revaluation 'makes no material difference' for a 7% flow. The composite's valuation policy was last updated in 2019.
1. Calculate the true time-weighted return for June and the Modified Dietz return, and explain the source and direction of the difference.
Read the worked answer
Split June at the flow date, because the flow is the only point at which capital changed:
- Sub-period 1 (1–8 June): 880.0 / 850.0 − 1 = +3.529%
- Sub-period 2 (8–30 June): 941.6 / 940.0 − 1 = +0.170%
- TWR = (1.03529 × 1.00170) − 1 = +3.71% (about 3.706%)
Modified Dietz: gain = 941.6 − 850.0 − 60.0 = €31.6m; weight = (30 − 8)/30 = 0.7333; denominator = 850.0 + (60.0 × 0.7333) = 894.0; return = 31.6 / 894.0 = +3.54%.
The gap is about 17 bp. Modified Dietz assumes the flow arrived at one weighted point and that the period return accrued smoothly on all capital. Here the pre-flow capital earned 3.53% over the month, while the €60m that arrived on 8 June earned only 0.17% over the remaining 22 days. Dietz effectively credits the flow with about 3.54% × 0.7333 × €60m ≈ €1.56m of gain, when it actually produced about €0.10m. Overstating the flow's gain understates the return, so Modified Dietz is biased downward here. The reverse occurs when returns are weak before a large inflow. Because daily valuations exist, the flow-timing assumption is unnecessary: the firm can and should compute the true TWR.
2. What does the GIPS standards' valuation and large-cash-flow requirement imply for this portfolio and for the composite?
Read the worked answer
The GIPS standards require time-weighted total returns and require portfolios to be valued at least monthly and on the date of all large cash flows. A 'large cash flow' is defined by the firm for each composite, documented in the composite's policies, and applied consistently; the 7% flow here is the kind of event the policy must address.
Implications:
- The firm cannot hide behind 'monthly Modified Dietz' when the custodian supplies daily valuations. For a portfolio with a large flow, the firm must revalue and compute sub-period returns. Using an approximation that ignores an available revaluation is inconsistent with the requirement and with the composite's own valuation policy.
- The valuation policy must be composite-specific, in writing, and consistently applied across all portfolios in the composite. If other portfolios are valued only at month-end, the firm must still define what it does when a large flow occurs; the policy cannot silently vary by account.
- The 2020 edition no longer permits temporarily removing a portfolio from a composite because of a significant cash flow; the answer is revaluation, not exclusion.
- The analyst's 'no material difference' argument is an assertion, not a policy. A 17 bp monthly difference compounds and is material against a benchmark spread measured in basis points.
The defensible course is to report the TWR, update the valuation policy, and apply the same large-flow definition to every portfolio in the composite.
3. Advise on the annualized since-inception figure and on the 9.4% factsheet number.
Read the worked answer
Two separate problems sit here.
Annualizing a partial year. The portfolio is 11 months old. Annualizing a period shorter than one year is prohibited under the GIPS standards, and it is statistically misleading in any case: multiplying a cumulative 11-month return by 12/11 assumes the eleventh month is representative and ignores compounding. Since-inception performance must be shown as a cumulative return for the period from the date the portfolio first met the composite's inclusion criteria. If the firm wants an annualized number, it must wait until a full 12-month period exists; even then, presenting a single year annualized is weaker evidence than a multi-year record.
The 9.4% factsheet figure. The firm must be able to reconcile every published number to the underlying record. If 9.4% is the 11-month cumulative return annualized, it is non-compliant and must be withdrawn and corrected. If it refers to a different period, portfolio or composite, the label must say so explicitly. A number that cannot be tied to the composite's calculation records is a presenting-and-reporting failure regardless of whether the arithmetic is right.
Practical answer to the consultant: report the cumulative since-inception return with the inception date, state that annualization is not presented because the period is under one year, and explain the 9.4% figure or correct it. Also check whether the composite itself has the required minimum history (five years, or since inception if shorter) and whether the portfolio's inclusion date matches policy.