We live in a time of the highest rate of technological advances ever. Technology permeates through every facet of our life and the collective computational power of our modern society is mind-blowing.
As a wealth management industry, accuracy is paramount. Regulators have rightly forced our industry to embrace this principle throughout all operations. One area of specific interest for me is the Mifid II requirement which states that Investment firms ‘shall inform the client where the overall value of the portfolio, as evaluated at the beginning of each reporting period, depreciates by 10% and thereafter at multiples of 10%, no later than the end of the business day in which the threshold is exceeded’.
When I look at the guidance from the European Securities and Markets Authority (Esma) and the industry response to the 10% rule, I am astounded by approaches taken to comply with this requirement which at best are ‘haphazard’, but also could be considered ‘careless’.
Here are a couple of reasons why.
Estimation Calculation Methods
The proposed methods of calculating the 10% drop scenarios are estimates and estimates only. Some of the methods (such as Modified Dietz) are reasonable approximations and others that will be used can simply be described as ‘rough as guts’.
The mathematical fact is that there is no simple algorithm that can calculate the returns on a portfolio with irregular capital movements. Peter Dietz came up with his methods (Simple Dietz and Modified Dietz) of performance estimation in 1966 because of computational limitations at the time.
Both algorithms provided a reasonable estimate which required low computational resources. To accurately calculate the return on an individual portfolio with irregular capital movements over a short period (i.e three months) we would need to adopt a Money Weighted return algorithm. This requires an iterative calculation method to hone in on an actual rate of return.
Yes, it does require more computational processing and it does require a little extra effort to implement, but as we live in a world and work in an industry with lots of big computers and smart people, surely this shouldn’t be too hard.
Here is an example of how this would work:
Client starts a quarter (90 days) with £100,000 and 10 days into the quarter the client withdraws £80,000. Therefore, one day before the end of the quarter the balance is £17,220 (see table).
Estimates are just estimates. Some are reasonable and some are so far from the truth as to make them senseless.
Interestingly, the guidance provided by Esma follows the worst possible estimation method. This is indirectly acknowledged in its Q&A paper when it states: ‘Firms should avoid any behaviour that might incentivise clients to add investments/cash that is invested for the purpose of avoiding the reporting on a portfolio depreciation.’
If the performance reporting was done using a reasonable estimation method, then this would not be possible. It is only the questionable calculation guidance that makes this possible.
So why are some in our industry using these calculation methods?
In short, I think that DFMs can delegate the task of reporting but it appears that they cannot delegate due care and the responsibility of doing the job properly.
Large, ugly assumptions
A number of platforms in the industry seem to be calculating the performance of client portfolios with some inaccurate assumptions, such as, if you work out the performance of the model over the period, all clients in the model will have the same performance. It is also assumed clients who are in a model (and DFM) on a given day have been in that model for the entire period and when a client is in a model (and DFM) all of their assets held are aligned with the model.
If any of these assumptions are adopted, the resulting calculations may be wrong. Why is this?
Clients are transitioned into and out of models all the time and clients are transitioned between DFMs all the time. Clients deposit into and withdraw from their accounts all the time. This can skew the calculations even further when transitioning between models.
Clients often hold assets outside of a model. For example: as a result of timing of realignment trade instructions when moving between models or when models are realigned; as a result of holding closed funds that are unable to be realigned; as a result of holding ‘sacred cow’ investments that are not to be traded on instruction from client but are still held within the account.
In summary, regulation has stated that we need to accurately identify when performance drops a client’s returns below 10% in a reporting period.
DFMs do not have the data to calculate individual portfolio returns and are reliant on platforms to do this on their behalf.
Some platforms are really taking on this task seriously and some are just doing something that’s quick, cheap and easy. The side effect of this approach is that some clients will be reported when they shouldn’t be and some will not be reported when they should be.
The only loser in this is the DFM and their reputation.
Ray Tubman is the CEO of consultants, FinoComp