The parable of the seven fund managers

This parable is based on a conversation that I had with a friend and mentor last week. Readers of my earlier posts will recognize the moral of the story. It went something like this.

There once was a pension fund with poor performance. In an effort to turn things around, the fund’s Trustees decided to set up an internal investment team. Their first step was to hire a Chief Investment Officer (CIO).

The fund decided that the best way to do this would be to engage a highly-paid executive recruiter, who took out a prominent advertisement in a national newspaper. The many responses were then carefully sorted first into a “long list” and then into a “short list”. Shortlisted candidates were interviewed multiple times – by the fund’s Chief Executive Officer (CEO), the Chairperson, and several of the directors. References were checked and psychometric testing was carried out.

After an exhaustive search, an impressive candidate with a background in investment consulting and asset management was selected. The appointment of the new CIO was accompanied with a press release and extensive media coverage by the industry press.

Eventually, a team of investment staff were hired – using a similar rigorous process to the one described above for hiring the CIO – and the internal investment team was now ready to get to the bottom of the fund’s performance issues.

The CIO and his team used a quantitative risk models to perform returns-based and holdings-based analysis on the fund’s investment managers, asset class portfolios and the fund’s overall multi-asset strategy. This included both ex-post (historical) and ex-ante (forward-looking) analysis.

The first asset class portfolio to be investigated was equities. Following the advice of their asset consultant, the fund’s current portfolio contained 7 investment managers spread across the following investment styles.

  • Index
  • Core
  • Value
  • Growth
  • Quant
  • Small Capitalization
  • Absolute Return

The result was a portfolio with a low-overall tracking-error (even though the tracking-error of several of the individual managers was quite high), low-active share and a high-level of redundancy.

Redundancy is when the under-weight (relative to a market benchmark) positions of one fund manager are offset by the over-weight positions of another fund manager. For example, the value manager fund manager is buying a stock because it is cheap, while the growth manager is selling the same stock because its forecast earnings growth is low.

Or the core manager might be buying a stock that has just been included in the benchmark index, while the small capitalization manager is selling the same stock because its market capitalization has increased and it has graduated out of the small capitalization universe.

The overall result is that the gross performance 7-manager equity portfolio performs almost identically to the market benchmark over the medium-to-long term. Naturally, after investment management fees, trading costs and taxes, the net return to the fund’s members is significantly lower than the benchmark return.

In other words, the fund could significantly improve its results over the medium-to-long-term by investing the  entire portfolio in an index fund. This would allow the fund to.

  • Save money on investment management fees.
  • Cut trading costs close to zero.
  • Reduce capital gains taxes paid.
  • Deliver better long-term returns to members without taking the risk (i.e. potentially under-performing the market) of active fund management.

Excited by this result, the new CIO presents these results to the CEO and the Trustees who are both confused and disappointed by this conclusion. They explain their reasoning to the CIO.

We hired you, following a lengthy and expensive recruitment process, to improve the fund’s returns. We hired an internal team. What’s more, we even went out of our way to stir up fanfare in the industry press by issuing press releases and giving interviews – telling all who would listen to expect great things from our internal team. And the best you can come up with is for the equity portfolio to be 100% passive? Is that what we’re paying you for, to just give up?

The chastened CIO then explains that there is another way to improve the fund’s returns.

We could reduce the number of fund managers from 7 down to 3-4. This would eliminate much of the redundancy between our fund managers. It would also save on fees as we could negotiate lower investment management fees for larger investment mandates. The active share and the tracking error of the overall equity portfolio would also increase. The equity portfolio would then have a better chance of out-performing the market after costs.

The CIO also adds that there’s an important caveat.

Having a portfolio with 3-4 managers will have a higher probability of beating the benchmark over the long-term (as opposed to having almost no chance with 7 managers). But it will also increase the risk of short-to-medium term under-performance. The reason is that no investment strategy, no matter how good, works all of the time.

All other things being equal, having 3-4 managers increases the chance that the overall portfolio will under-perform the market over the short-to-medium term if one or more of the fund managers are experiencing a period of poor performance.

The Trustees are undeterred.

We understand the risks. Our members are investing for the long-term. So we can be patient, our members are in a position where they can ride out any short-to-medium-term ups and downs in performance so long as they are better off over the long-term.

All of which is true, at least in theory but unfortunately not always in practice. With the approval of the Trustees, the CIO and the investment team progressively reduce the number of fund managers from 7 down to 4. Two years later, 2 of the equity portfolio’s 4 fund managers start to under-perform the market. The under-performance continues for two years. The long-term performance of both managers is still good, despite under-performing over the short-term.

In the meantime, several of the Trustee directors have changed. And other directors have begun to notice the performance of other equity portfolios run by peer funds that are doing better. After 3 years of underperformance, pressure starts to build within the fund to “do something”.

What is the CIO to do? Investing in an index fund would be an admission of total failure. Holding fast to the current strategy will be tough as it may appear to the Trustees as if the CIO has run out of ideas (even if the original idea might still be a good one if only the Trustees have patience). And sacking the under-performing managers means realizing a loss on investment, something that most people find emotionally painful.

The only other option is to “partially redeem” or reduce the investment in the under-performing fund managers in order to invest in fund managers that can hopefully turn the equity portfolio’s performance around.  Of course the fund managers selected will have performed strongly in the recent past. After all, nobody sacks one “loser”  just to hire another “loser” do they? Everyone wants to hire a winner that they can feel good about*.

Before long, the fund’s equity portfolio again holds investments in 7 fund managers…

 

 

* Very few investors stop to think that today’s “losers” weren’t losers when they were hired and that the same thing may happen to today’s “winners”. Which is why Benjamin Graham and David Dodd quoted Horace in their classic investment book Security Analysis: “Many shall be restored that are now fallen and many shall fall that are now in honour”.

 

 
Advertisements

One comment

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s