My submission to the CSA re Risk Classification disclosure

Re: CSA Notice 81-324 and Request for Comment – Proposed CSA Mutual Fund Risk Classification Methodology for Use in Fund Facts

The part must relate to the whole and the whole to the part and the both must know of it:

The CSA in their consultation paper fail to explain how the risk classification methodology proposed for use in the point of sale documents is to be used by investors to make informed decisions, and how advisors are to use the same to determine suitability.

When we consider that the retail investor (to whom this disclosure is aimed), is deemed to be responsible not just for the transaction, but for the sum of all transactions, the CSA’s consultation is remiss in a great many areas.

One of the great “benefits” of regulation of the transaction, for those who distribute and transact, within a limited KYC parameter to parameter suitability framework with high levels of investor responsibility for the transaction, is that you shift the subject matter complexity of the decision onto the individual investor. But this is not what actually happens. While the decision is shifted onto the investor, the complexity of the data needed to make that decision is not, and for good reason.

Importantly disclosure about the transaction nature of the relationship is also obscured. With many firms in the recent Best Interests Standards consultation claiming to already operate in their clients’ best interests, the import of the information in the mutual fund POS with respect to transaction responsibility is likewise occluded.

Unfortunately complexity of this particular subject matter is something the average individual investor cannot process. Indeed, processing standard deviations or valuation risks, correlations or relative price movements, and expected or implied returns/risk premiums to determine an optimal portfolio structure that is compatible with the client’s risk and performance preferences and liability profiles over time, is impossible for anyone, irrespective of ability, impelled to make a decision at the point of sale.

In other words, just what process are regulators assuming is happening at the time POS data is presented to the investor, and how does this data allow them to own the decision? In a blog posting on the issue, prior to the publishing of the consultation, I made the following comment:

I think in the consultation the OSC/CSA need to clearly define the process that the advisor and the client are going through at the point of sale with respect to suitability and the relevance of the fund selection and hence the fund risk profile with respect to that decision. Only by clearly defining the parameters of the process can you define the risk parameters – it is like an equation where you cannot determine the output without the relevant inputs and relationships.

Is it both the fund and the asset allocation that are being selected or is it just the fund to fit an asset allocation profile already determined, because the risk dimension in each of these two options is vastly different? In a process where the asset allocation has already been defined with respect to a richer set of risk parameters the POS risk disclosure can accommodate a more one dimensional representation of risk.

If we just look at a simple mean variance structure, we need to combine a number of assets/asset classes with different standard deviations, expected returns and correlations to come up with a diversified portfolio. We cannot just infer the decision from one input. Additionally, if we have liabilities attached to those needs, we need to select an allocation structure best able to deal with the liability profile, which requires some fairly sophisticated modelling. And, of course, we need to understand the limitations and the risks of a given construct and compensate accordingly.

A transaction without an underlying process and structure to guide it is an extremely difficult act to rationalise, and yet, by assuming that the POS is tied to a rational process capable of creating an optimal outcome, that is precisely what we are attempting to do.

So the CSA is ignoring the problem by facilitating the transaction through the over simplification of the information. At the point of sale there is no effective processing of the necessary inputs. The transaction decision is incapable of being related back to the whole, nor the whole to the decision, in most, if not all instances. The problem that appears to be addressed by regulation is how to keep the transaction wheel oiled, while performing a perfunctory, but flawed, nod to informed investor decisions and investor protection.

An argument about dessert when we have not even decided where to eat:

What we have in this consultation and its many comments, is an argument about dessert when we have not even decided where to eat. By the time you get to the simple 2 page snapshot of the fund and the point of sale, the investor should know much more about risk in general and, in general terms, how everything fits. The risks of stock market investment should have been dealt with. The profile of each transaction is obviously important, but the least important of all the information points. Besides, issues with the POS do not just rest with the risk aspect, but the return and other fundamental information. The regulators should not be thinking what A POS should look like unless they are clear on the process itself.

And we are talking about the transaction here and factors which define the fund itself, as opposed to the relationship between the fund and the client’s portfolio (and the portfolio the client’s needs, risk preferences and advisor methodology) which should already have been dealt with.

The chaos and conflict of the point of sale:

As long as we have a distribution system reliant on the transaction and remuneration from the transaction, any attempt to improve information flow and education and service outcome is going to be squashed into this small space, the point of sale.

It is a fact that the transaction platform misses out huge swathes of suitability and pushes the complexity of the entire decision to this point, while also failing to provide either the processes or the information to process it properly. In this submission process, if we all attempt to agree on what risk parameter we should be using, we are ignoring the heart of the matter, which is the impossibility of squeezing everything into the point of sale where the point of sale is also the point of the transaction.

While we are faced with the impossibility of squeezing the entirety into the point of sale, any discussion about risk indicators is going to be overwhelmed by the chaos of the moment and the gaps and omissions in the process. Explanation of risk, of the fund’s risk, of the risk of investment, of the duration and depth of risk, of liquidity risk (met through structure and hence dependent on process) at this point is bound to be overwhelming and in conflict with the objectives and limitations of the point of transaction.

It is impossible to direct education about risk at this narrow squeeze point. If you look at the history of the development of the POS, it has been defined by this squeeze point – too much information and too little time.

I agree with a great many of the submissions, but they are all effectively saying the same thing. How on earth do you cram so much of importance into the point of the transaction and how do you continue to transact if everything is squeezed into this point? That is why we need best interest standards and true advice focussed processes to extricate and properly communicate the necessary information along a wider spectrum of time and process. The point of the transaction should, in both paradigms, be relatively quick and efficient.

It is the point of the transaction that is the issue and the impossibility of transacting efficiently and quickly at these points while having to squeeze an effective “groundhog day” moment of informed communication at every transaction.

A best interests’ standard would draw complexity and responsibility for the process and analysis back to the centre, to the realm of the advisor and the firm and spread the POS moment across a wider spectrum. It would also free up the point of the transaction from the tyranny and insanity of the point of sale.

Simpler explanations of risk[1] within this framework, such as that proposed within the CSA consultation would therefore have relevance within a best interests’ standard framework, where the advisor/firm has responsibility for the integrity of processes that design and construct and manage portfolios to meet individual investor risk and performance preferences for a given set of liability profiles.

Likewise the availability of more detailed statistical, fundamental and transaction information on a fund would not impede the transaction but enrich the education and understanding of investment risk and return.

And so the Catch 22 is this:

We cannot have simplicity of communication without advisor responsibility for complexity, and we cannot have transaction responsibility by the investor without processes capable of managing the complexity of the decision, which itself requires a higher level of information on inputs.

ARC + SC = TR + C

But minimum standards effectively remove advisor responsibility for complexity (where the advisor is responsible for structured processes) one the one side and complexity (the detailed information a capable investor would need to make informed well structured investment decisions) from the other, while retaining simplicity of communication and transaction responsibility. Simplicity of communication ≠ transaction responsibility.

In truth we need to separate the informed decision from the point of the transaction, to maintain the integrity of the efficiency of the transaction and to allow for the integrity of the informed decision and to separate those who are capable of making informed decisions (and processing the information needed to do so) from those who cannot and would not wish to do so.

So what is risk in the context of standard deviation?

At a basic level, standard deviation is one measure of the riskiness of an asset. If using historic data, it tells you how much the price of an asset has moved on average over a short period of time relative to the average return on that asset over a much longer period of time.

In a perfect world where markets are efficient, where prices are correctly formed and future price movements are random and independent of prior movements (i.e. a general equilibrium), risk as in standard deviation is a measure of the sensitivity of price movement to a unit of new information/shock as it randomly and independently enters the universe. Standard deviation is in fact a certainty (a physical attribute of that universe), but as to whether it effectively conveys the risks to return to which an investor will be exposed, that is a different matter entirely.

· Is standard deviation a measure of the uncertainty of achieving an expected return?

o No. It cannot be because expected return is a fixed dynamic used to define the standard deviation of return. So standard deviation in itself is not a measure of the risk of failing to achieve the expected return but a measure of the way price reacts to changes in certain relationships.

· Is standard deviation a complete picture of the uncertainty of price movement, around an expected return, over a given time period? Does it fully reflect the degree to which, on route to an expected return, a price can move up or down and in unexpected ways?

o No. It cannot be because a standard deviation is not only a certainty regarding the expected average of random price movements (in a general equilibrium) around the expected return, but it is also likewise assumed to be fixed (because the physical properties of the universe are stable). In fact, within the efficient market model, both expected return and the standard deviation of expected return are assumed to be fixed (they are directly related to stable physical properties of the equilibrium) and are therefore expressing certainty as opposed to uncertainty. The boundaries of price movement are a given as are the range of shocks/new information likely to impact the universe, as is the expansion (return) of the universe itself.

· Or, is it a linear measure of the sensitivity of price to a specific unit of change in the underlying economic and market equilibrium, itself reflecting unit changes in random, independent variables?

o In other words, it does not express anything about the range of outcomes or the certainty of those outcomes, but the way in which a given change in an array of market and economic equilibrium relationships impact price. As a regression it focuses on an underlying relationship between a unit of change in the environment and a ∆ in price.

Standard deviation is a measure of the rate at which price changes (markets respond) in response to a unit change in the environment (the rate at which new information/shocks enter the universe). It is a measure of sensitivity.

But much of price movement is due to changes in relative demand amongst assets, and hence much of a standard deviation is not a risk per se if other assets which are less closely correlated with that asset are held in the same portfolio. And on its own, it tells us little as to whether the expected return is worth taking on the risk.

Standard deviation on its own does not tell us anything about the uncertainty of price movements (be it their size or their probability of occurring) or the uncertainty of events surrounding price movements, or whether it is a good or a bad risk to assume.

If we are operating in a general equilibrium then what is more important than standard deviation for the assessment of risk is the probability distribution of individual period returns (price movements) and how the pattern of returns could evolve over time given assumptions of random independent price movements (i.e as with a Monte Carlo simulation).

Uncertainty is to do with the randomness and independence of each piece of new information, meaning that while you may know the average sensitivity and the distribution of price sensitivity points, you never know what part of the distribution is going to hit you at any given point in time.

What is uncertain therefore is what part of the distribution of price movement is going to hit the investment at any point in time and the potential size of these movements. This information is not available to the investor in the current format, even if they could understand it. Therefore relying on standard deviation as the sole information point about risk does not inform the investor about the actual range and impact of outcomes that could affect them. And note that we are operating in a paradigm where the investor is responsible for the investment decision and the advisor need only select a suitable investment consistent with a limited range of parameters that need not necessarily be in the client’s best interest.

Unfortunately, if we are not in a general equilibrium, and if markets are not efficiently pricing existing structural relationships, then historical standard deviation and return will tell us even less about the risks that could hit the investor – both return and risk are no longer certain, but highly variable and dependent on when and where in the cycle you are and how far out of equilibrium key structural market and economic relationships are – which means the duration and depth of negative return periods are likely to exceed those defined by the randomness of a normal probability distribution and are themselves material in defining the risk of investment per se, as opposed to just the risk of investment in an isolated asset.

In other words standard deviation (from historical data) tells us only how risky an investment has been on average but gives less than a full picture of the risks they could face, especially when investors have high financial demands on their assets. Cue the environment of the last 15 to 20 years (1996 to 2014). And even if they had a full picture of the risks they could face, they would need to be able to process them into a structure able to manage them.

Informing investors about the reality of investment risk (the full picture including the interaction of assets and liabilities) is the responsibility of the investment advisor and the firm, and not the responsibility of the POS document. Providing a structure that can better manage the range of risks likely to hit a portfolio, for a given risk and liability profile, should also be the job of the investment advisor, unless the investor has categorically taken responsibility for this role.

What also concerns me, inter alia, here is that a definition of the advisor’s responsibility to educate is ignored (although implied by the omissions of the POS), yet the advisor’s ability to sell and to transfer responsibility is emphasised. There is indeed a conflict between a business model where the onus is on the transaction and where the information an investor needs to make an informed decision, if they possessed the necessary expertise and resources, is incomplete, unstructured and provided, if they are lucky, at the point of the transaction.

Again we come back to the point of sale and the need to free the transaction from the informed decision and to optimise the integrity of both.

Investors still need to take responsibility, but they need a simply communicated investment process that aligns investor needs and risk profiles with a portfolio able to manage the risks to the ability of those assets to meet those needs in the context of the investment discipline and portfolio construction methodology used.

The risks and realities within the context of a disciplined approach to managing both assets and financial needs still need to be communicated to investors, to a) better manage expectations and b) to allow adjustment of the risk and return profile of a portfolio to better suit risk temperament and investing realities.

In other words education about risk does not start and end at the standard deviation in a POS document but covers the gamut of the nature of risk and return, how portfolios are constructed and managed and how the portfolio aligns with their own financial profile.

But what would allow a professional investor, and advisor or an investor with more sophisticated tools and processes to make an informed mutual fund transaction decision? Well, that it is a good question, and as I noted in the opening of my submission, it is neither posed nor answered. The implication is also that this risk disclosure represents the minimum standard of information an advisor would need to determine suitability.

To be honest a mutual fund company could easily provide access to a whole host of other important data (both statistical and fundamental) on their funds and advisors a higher level of education on the risks and realities of investment.

But, at a fundamental level I believe that the problem is more to do with the way in which the retail investor is viewed and treated. This is not about advice or responsibility, but about keeping the retail investor in a place where transactions as normal, within the current process, can continue. Note the simple fact that there is no mandated benchmark comparison in the POS fund facts document, and no benchmark eliminates the ability to provide effective risk as well as return comparisons.

This submission is directed at the framework in which the POS disclosures are delivered as opposed to the pros and cons of the indicator itself, because I feel the issue of the indicator is a red herring to the real issue. How do we keep the integrity of the transaction while raising the integrity of the decision? It cannot be by filling the POS with all the information needed to make a decision, but a solution can be found in best interest processes which fundamentally offer no obstruction to the transactional integrity of the capital markets.

Standard deviation is a fair summary of the risk of an investment, in terms of the average deviation of price from return, especially when it is crudely compared to the standard deviations of other investments and asset classes. I have no issue with simplicity in a relationship that effectively manages the complexity. But to force the consumer to rely on standard deviation as a stand alone measure of risk, given current minimum standards, is troubling and unfair.

I would of course like to see a more detailed supporting information sheet comprising more comprehensive statistical analysis, fundamental data (market cap, sector, style, yield, P/E, PB relative to an appropriate benchmark), transaction data as well as more graphical representations of risk for those investors who really do wish to make their own transaction decisions and take responsibility for them, as well as providing an educational resource for those investors who wish to learn more about their investments and investment per se.

As a final aside, I would have to say that I concur with many of the points made in those submissions with wider vistas: IAP, SIPA, CAC, Kenmar, Hallet, PMAC, Fundata, Prof. Sinha, CCC, Mcfadden, Ross, Schalle, Gourley. The message that transcends all these arguments is that risk is more complex than one indicator alone and that standards governing the current retail advisory relationship are a very large part of the risk equation, but one that appears to be keenly ignored by regulators, apart from the two lukewarm submissions on best interests standards and mutual fund fees.

Kind regards

Yours sincerely

Andrew Teasdale, CFA, BA Hons. Econ.


[1] We know that many investors do not use the information in the simplified prospectus because they have trouble finding and understanding the information they need. Research on investor preferences for mutual fund information, including our own testing of the Fund Facts, indicates investors prefer to receive a concise summary of key information. Financial literacy research further reinforces the need for clear and simple disclosure. – http://www.osc.gov.on.ca/documents/en/Securities-Category8/ni_20130613_81-101_implementation-state-2-pos.pdf

Leave a Reply