Is Picking Mutual Funds any Different from Picking Stocks?

Today I received an email from an advisor that continues to work with actively managed mutual fund providers to serve his clients.  He sent me a document that a fund company had produced for him showing that actively managed mutual funds have consistently beaten their benchmark ETFs, and asked me what I thought.  My response follows:

Thank you for thinking to send this over.

I think it’s very interesting that the active management vs. ETF/passive debate is important enough for this fund company to have made this document.  I think it is even more interesting that they are pushing their advisors to sell their active funds while they are also making ETFs on the side.

But look, the thing you have to understand is that there will be funds that outperform just like there will be stocks that will outperform. I could go and find a bunch of stocks that have beaten a benchmark over the last ten years and say that it’s possible to beat the market.  It's important to understand that with a mutual fund, any time you are not holding the whole benchmark there can be periods when you outperform and periods when you underperform.  I could go and find a bunch of funds that have not beaten their benchmark over the last ten years just as easily as they found ones that did.

The key though, is that you or I, or even the fund managers don’t know when they will be winners or when they will be losers.  You can go and say "ok, the XX was up 11% over the S&P/TSX last year, I’m going to put my client in that one". But you know as well as I do that past performance does not indicate future performance.  You are betting on that fund doing well on your client’s behalf based on how well it did last year; you have no idea if it will be up or down in the future.  If you don’t know what the fund will do in the future, why not just accept the market’s returns and get rid of the high MER instead of hoping that the fund happens to be up during the period that your client is invested?

Picking funds is on the same level as picking stocks in my opinion.  You can say "well this manager has tons of experience and the fund has done well in the past" and so on, but that’s no better than looking at the financials of a company and saying "oh they have strong earnings, good management, and a stable dividend I will pick that one".  There is simply too much information in both cases to make an accurate prediction AND your clients are paying high fees for you to place these bets... it just doesn’t sit well with me.

Original post at pwlcapital.com

Commentary on the CSA's recent status reports (CSA Staff Notice 33-316 & 81-323)

The Canadian Securities Administrators have been reviewing two different, but related issues for over a year now.  Both of the topics receive a significant amount of coverage in the media; they are Mutual Fund Fees and the Appropriateness of Introducing a Statutory Best Interest Duty When Advice is Provided to Retail Clients (making it the law for advisors to act in the best interest of the client).  I just want to express what I feel were the highlights of both of these papers, and offer my commentary.  I have been on three sides of this equation (investor, mutual fund salesperson, and fee based investment advisor), and as such I think that I can offer some unique insights.  The stakeholders in this discussion have been identified as the mutual fund industry, and the investors.  Their points of view are summarized below.

The Mutual Fund Industry

  • There is no evidence of investor harm that warrants a change to the mutual fund fee structure in Canada
  • A ban on embedded compensation will have unintended consequences for retail investors and the fund industry, including:
    • a reduction in access to advice for small retail investors,
    • the elimination of choice in how investors may pay for financial advice, and
    • the creation of an unlevel playing field among competing products and opportunities for regulatory arbitrage; and
  • We should observe and assess the impact of domestic and international reforms before moving ahead with further proposals.

The Investors

  • Embedded advisor compensation causes a misalignment of interests which impacts investor outcomes and should be banned;
  • Investors should at a minimum have the true choice to not pay embedded commissions;
  • We need to implement a best interest duty for advisors; and
  • We need to increase advisor proficiency requirements and regulate the use of titles.

The statement that there is no evidence of investor harm that warrants a change to the mutual fund fee structure is just about the worst reason that I have ever heard to allow a regulatory policy to remain the same.  A practice should not be allowed to continue because no harm has yet been caused by it.  My point here is especially salient when we consider that other countries across the globe are changing their mutual fund compensation structures because they have had problems in this area.  The UK and Australia have prohibited conflicted advisor remuneration "in response to unique consumer protection gaps and situations including mis-selling scandals causing harm to investors, which are not present in Canada."  So despite serious problems with the exact same thing in other countries, we don't need to change it in Canada because nothing has happened yet...seems like a good idea.

I do agree that changes to the regulations could have ramifications beyond the scope of their intentions, but I do not think that these ramifications would be a bad thing.  There is some merit to the statement that unbundled mutual fund fees would reduce access to financial advice for people with small (less than $100K) portfolios.  It would not be beneficial for either the client or a fee for service advisor to enter into a relationship with only a small amount of capital to invest.  As it stands today, small retail investors are in a situation where they are limited in their options beyond investing in mutual funds with high expense ratios if they want financial advice - the only reason that financial advice is affordable with the current embedded commission model is due to the existence of the deferred sales charge (DSC).  The DSC is exactly what it sounds like; instead of paying a bunch of money (5-6% of the account) up front, you pay it behind the scenes over five or six years and never even notice that the advice you received cost anything.  This is a great idea for the advisor as they receive remuneration at the time of the sale.   It also seems to make sense for the investor because they never see the money leave their wallet, until they want to move their account or withdraw their funds before the six years is up.  If the money does not stay invested for the full deferral period, the client is on the hook to pay the fund company back for the money that their advisor received at the time of the sale.  Another issue in this area is that not all mutual funds have the DSC option; in my experience, funds with low fees that do not involve predictive management or large amounts of trading cannot be purchased in this manner.  So... small investors that want advice are likely to have their money invested in expensive funds that allow their advisor to apply the DSC to make it worth their while.  Knowing that information, I'd say that  although small investors can afford advice in the current environment, the advice that they are getting is highly biased towards funds that can be sold with the DSC.

The industry has stated that they do not think that there are conflicts of interest with the embedded fee structure because all types of mutual funds in Canada have comparable trailing commissions.  On this topic, I can speak from direct experience.  When I was operating on the MFDA platform I was very limited in the funds that I was able to offer.  Yes, there were hundreds upon hundreds of mutual funds with high expense ratios claiming to be able to beat the market available to me and the people in my office, but as I started to shift my thinking towards the use of ETFs and index tracking mutual funds I learned that I could not access them.  I remember discovering TD's index funds and being told I shouldn't use them because I couldn't make any money, and I wasn't able to use Dimensional Fund Advisiors because my dealer did not acknowledge them as a supplier.  It is true that most equity mutual funds with high fees have the same trailing commissions, but I would argue that when advisors are limited to using these funds it is difficult to act in the best interest of the client.  Based on my experiences I can safely say that the clients' best interests are not being put first.  Of course, this shouldn't come as a surprise.  In the current regulatory environment investment recommendations only have to be deemed suitable for the advisor to be legally off the hook.

This lack of a requirement to act in the best interests of the client, or lack of a fiduciary standard, is an issue on more than one level.  To open the discussion of the lacking fiduciary standard I first need to explain the educational requirements in place for an advisor to begin operating in this industry.  In order to sell mutual funds, and call yourself an advisor, it takes one exam requiring a 60% to pass and a 90 day supervised training period.  The exam is designed to take between 90 and 140 hours to prepare for.  So with that said, how is someone supposed to know if they are acting in the best interests of their client when they barely know what they are selling? During the 90 day training advisors aren't even trained to put the interests of the client first, they are trained to make suitable investment recommendations. Using the UK as an example, when they prohibited embedded commissions and increased regulations there was a drop in the number of advisors because some of them could not meet the proficiency standards put in place.  To compound this issue, the CSA cited a study stating that most investors already think advisors have a legal duty to act in their best interest.

The fact that there is a debate around whether or not advisors should have a legal duty to act in their clients' best interests is a little bit silly right?  How is this even a question?  The industry argues that the cost of an investment product should not determine suitability; although an advisor may place their client's investments in a more expensive fund they may have done so because the more expensive fund is expected to perform better in the long run...what?  I don't even want to touch that one.  History tells the story of expensive funds outperforming the market better than I can.  The industry also poses that the introduction of a fiduciary standard will give people less incentive to educate themselves on investments and place more reliance on their advisor.  We certainly wouldn't want clients to forgo educating themselves in favour of taking advice from their professional advisor.  I of course say that in jest, but it may hold more truth than it should with proficiency standards at their current level.  

Another argument that the industry poses against implementing a fiduciary standard is that it would be expensive, and these costs would be passed to the client.  To this I say that clients are better off paying more for advice when the advice has their best interests in mind than getting cheap advice that quietly pads the pockets of their advisor.

One of the biggest problems the mutual fund industry would face if a fiduciary standard and unbundled commissions were implemented is that they only sell...mutual funds.  If the most suitable product for a client was not a mutual fund, or like in my experience the most suitable product was a mutual fund without embedded commissions, how can the client's best interests be served?

Personally, I think that the fact that there is a debate around these issues at all is ridiculous.  I worked with some great people when I started in this business on the MFDA platform, but it did not take me long to realize that there were many pieces missing if I was going to be acting in the best interests of my clients.  There a lot of good people in this industry that genuinely think that they are doing what is right for their clients; it is an issue of education on the part of the advisor. I look forward to seeing how all of this turns out, though it doesn't make any difference to me.  I am already operating on a fee-only basis within a firm that has a self imposed fiduciary standard.  

Everyone else will get it right eventually.


What the Automotive Industry Can Teach Us About Mutual Funds

Up until General Motors (GM) went bankrupt in 2009, their business practices created a beautiful juxtaposition for the business practices of Toyota.  GM had been able to rely on their size, market share, and Japan's non-existence to remain profitable for many years.  This fostered an environment of complacency in their ability to perform their core activity...creating vehicles.  In 2006, GM's profitability was mostly derived from financing the sale of their vehicles while they lost an average of $2,300 on each sale; their core activity had become financing while they were still stuck with making cars.  They were using inefficient manufacturing techniques to create a wide array of uncompetitive vehicles.

In contrast, it's no secret that Japanese automotive manufacturers have absorbed the market share that GM was slowly giving up; in 1960 GM held 50% of the North American market, while they only held 19% in 2009.  There were some very clear factors that contributed to this shift in the market, and, strangely enough, I believe that the same principles can be applied to the current state of the mutual fund industry.  The principles are very basic; efficient manufacturing practices, and competitive products.

Efficient manufacturing practices were revolutionized by the Japanese with Just-in-Time manufacturing (JIT).  American manufacturing methods employed economies of scale to keep the cost of machine hours down; they would have a large production run of a single part and then change the machinery to produce the next part.  The problem with this was that the time and cost associated with changing the machinery was cumbersome.  In the event that a manager suddenly needed a small run of rear doors when the machines were set up for front doors, they would be paying a premium for those machine hours.  JIT made the manufacturing process flexible by implementing production machinery and a labour force that was interchangeable; managers no longer had to pay a premium for unexpected production runs because multiple machines could be quickly set up for a small production run of any part.  These interchangeable parts in the manufacturing process eliminated the high fixed costs of any single production run.  Parallels can be drawn between JIT in auto manufacturing and the trading done within a mutual fund.  

Active fund managers will conclude through research that it is the right time to purchase a certain amount of a given security.  When they fill this order, they will be paying a premium to obtain the specific amount of securities at a specific time due to the nature of bid-ask spreads; at a given time spreads will vary, and filling the kind of large market order that an active mutual fund would place will result in paying varying prices for different available lots.  In contrast, a portfolio designed with holding diversified across asset classes and geographies would not need to accept an undesirable bid-ask on any individual security because they would not be looking to purchase any particular security at any exact time.  A mandate may require a certain weighting of domestic small cap stocks, but the individual securities could be anything fitting that description.  This approach lowers overall trading costs; trading costs are not something to be overlooked either.  It would not be unusual to see an actively managed small cap fund with a trading expense ratio as high as 1%.

Lowering the costs associated with producing either a car or a mutual fund is beneficial to both the customer and the manufacturer, but if the product is not competitive, the cost savings do not drive any value.  In 2006, consumer surveys about vehicle value showed that people perceived Toyota vehicles as being better designed and built, costing less to own, and lasting longer than GM vehicles.  This higher perceived value combined with lower manufacturing costs allowed Toyota to grow its margins while GM’s margins became negative.  GM boasted a massive portfolio of brands and vehicles, each one with a wide array of available features.  This lack of focus and internal competition made it very difficult for GM to produce any single excellent product; in contrast, Toyota has only two brands (Toyota and Lexus), and the vehicles that each brand produces have fewer options.  Toyota’s focused approach to the continuous development of excellent products combined with their low costs allowed them to surpass GM as the number 1 auto manufacturer in the world.  As GM’s margins were squeezed by high manufacturing costs and competition with Toyota’s low prices, the quality of their products continued to suffer.  Near the end, and since their re-birth, GM has eliminated or sold many of the brands in their portfolio.

In the context of mutual funds, the industry is in a similar state as GM was before they went bankrupt.  Current funds in general are not of a high quality, they are costly to produce and costly to own.  Much like GM, fund companies are managing massive amounts of mutual funds.  On average, mutual fund companies manage 117 funds, a huge contrast to the average of 1.7 funds managed in 1951.  Just as GM had to eliminate many of their brands to pursue a more focused effort, mutual fund companies have been going through the same kind of consolidation.  Between 2001 and 2012 there was a 7% per year failure rate for mutual funds; failing funds would be absorbed into funds with strong records to mitigate the evidence of their poor performance.  If a mutual fund company were to take the same approach to their product portfolio that Toyota has taken in manufacturing vehicles, it would produce a few high quality, engineered, and focused products.

So, I have drawn a comparison between the automotive industry and the mutual fund industry.  Why do we care?  We care because those who cannot learn from history are doomed to repeat it.  If I went back in time and had a choice between investing in GM and investing in Toyota, I think it’s obvious where my money is going.  If I wanted to use this analogy to decide where to put my money today? Dimensional Fund Advisors would be my Toyota.

Investment Evolution

I have spent the last two weeks doing a lot of reading.  I will soon be entering a new professional environment with new philosophies on investing, and I made it a priority to educate myself on the broad governing concepts that form the investment style of this new firm.  In the readings that I have done I have come across information that I already knew, information that I did not know, and new implications of concepts that I already understood.  Below is a summary of my thought process as I gained an understanding of three-factor asset class investing:

The Efficient Markets Hypothesis (Fama, 1966) asserts that current securities prices reflect all available information and expectations.   Based on this theory, stock mispricing should be considered a rare condition that appears with randomness, and active management strategies cannot consistently achieve alpha

The Efficient Markets Hypothesis states that investors may be best served through passive, structured portfolios using asset class diversification to manage uncertainty and position for long term growth

Diversification helps reduce uncertainty and control risk

Diversification across asset classes allows for efficient portfolio management and flexible trading

Global diversification of overall investment strategies (conservative, moderate, aggressive etc.) created by combining asset classes can minimize the volatility caused by the inherent randomness of returns

A globally diversified portfolio should include asset classes that are exposed to different macro risk factors, with different dimensions of risk and return across the globe

Using the size of a country’s stock market relative to the world’s total market value rather than economic data (population, GDP, consumption etc.) to assign asset classification removes emotional distortions caused by economic statistics

Severe negative volatility (7% or greater decline in monthly value) has, historically, been experienced simultaneously by domestic large cap, domestic small cap, international developed markets, and emerging markets equities only 3.4% of the time between 1988 and 2009.  This negative correlation shows the importance of global diversification

In any given year, a small subset of stocks may contribute a large portion to the market’s overall return; with the CRSP 1-10 Index from 1926 to 2012 as a sample, removing the top performing decile of stocks reduces the total return from 9.6% to 6.3%.  Because it is impossible to consistently pick winners before the fact, it is logical that the optimal portfolio will maintain broad diversification

The three dimensions of stock returns are the equity market (complete value weighted universe of stocks), the company size (market cap.), and company price (BtM)

Returns of the CRSP 1-10 Index show that the smallest capitalization stocks produce both superior returns, and also the highest standard deviation in returns

The size effect of superior returns has been established over the longest available time period (1926-2012), but over shorter periods returns of small caps have been significantly above, and below the S&P 500