Blog

Last month, I reposted a commentary on LinkedIn written by someone who had previously worked at the Canada Pension Plan Investment Board.  The author’s beef was not just with the below-benchmark performance but how the benchmark is calculated and how the benchmark for compensation might be manipulated so that the underperformance still delivers significant bonuses to staff.

Some quickly wrote off the rant as being written by a disgruntled former employee or at least someone that doesn’t understand why everything at CPPIB is fine and everyone who works there deserves the overwhelming gratitude from the lowly Canadians lucky enough to have these folks managing our collective $800 billion retirement nest egg.

The commentary had just enough specifics to raise questions in my mind if we are in fact getting our bang for the buck for the 2,000 (or maybe more) employees working for us over at CPPIB.  The more I dug in, the more opposition I got back suggesting that I am going to embarrass myself going down this rabbit hole with Mr. Conspiracy.  As I often do, the more people say ‘please look away, nothing to see here’ the more I want to pull back the curtain to find out how smart the Wizard really is.  Here we go…

Investment Benchmarks

There are two schools of thought on investing.  One is to actively build a portfolio of investments that the investor thinks will provide the best return subject to the level of risk they think they can withstand.  The second school suggests that investors cannot outsmart the market and the correct answer is to passively invest in ‘index funds’ that return whatever the market average returns.  This latter approach is much less expensive to operate but comes with a greater feeling of helplessness as you no longer have any influence over investment outcomes beyond choosing the mix of index funds.  For the record, I believe in active investing.  Andrew Coyne at the Globe and Mail is on record as believing in index investing and often criticizes the CPPIB for its underwhelming performance over the last decade or so.

If you follow the passive/index investing philosophy, the name of the game is low fees, and the primary performance measure is around tracking error.  If you follow the active management philosophy, the name of the game is alpha or value added, net of fees.  Where active management gets tricky is when it comes time to measure the value added.  The main question is value added compared to what?  The investment management industry has largely settled on the idea that the ‘what’ is the index you could have bought at a low fee and the value added is how much more return the manager generated than the index.  It gets a little messy when managers are working with a complex investment policy where many indices must be combined to create an overall benchmark.

CPPIB Performance

I have gone back to read pieces of the this year’s CPPIB annual report – I know I am weird, but I find the actuarial reports on the CPP easier to read and understand.  Investment reports to me always seem to try and tell a good story regardless of how the managers have actually performed.  Here is an example:

“Over the past five years, investment selection generated annualized value added of 0.1% net of expenses, relative to the Benchmark Portfolios.      This result reflects the Investment Portfolios’ deliberate construction to be more broadly diversified than the Benchmark Portfolios – a design intended to deliver more resilient returns across a range of market environments, but which created a structural headwind under unusually concentrated market conditions in the past five years.”

So, what you are telling me is that we haven’t accomplished much for the past five years, but good news – the past five years was not the type of investment environment in which we expected to succeed.  After looking at the annual report I could neither get excited about the investment success at CPPIB nor be angry or scared that everything is off the rails. 

As I was reading, I was empathetic to the writers selling the story at CPPIB, not only do they want to send a positive message, but they need to give some details which will bore most readers.  But at the same time, it’s impractical to give enough detail to satisfy the actuaries and investment experts in the audience.  So, the ‘goldilocks’ effort gives almost everyone an unsatisfied feeling.

Last week, I came out of a 3-hour meeting with a few CFAs reviewing the performance of a group of active managers selected for a DC pension plan.  Interestingly, looking at the universal underperformance of the active managers, one CFA opined, it isn’t that active management doesn’t make sense, it is this concentrated market that doesn’t make sense.  As someone who subscribes to active management it sounded rational to me but if you are an indexer, it surely sounds like excuses.

The Funder’s Benchmark

I have often lauded the actuarial reports on the CPP prepared by the Office of the Chief Actuary.  Having read hundreds of pension plan funding actuarial reports in my career, I have a sense of what goes into a good report.  The OCA reports are detailed and rigorous.  Thoughtful questions are considered and answers are provided.  This isn’t to say that the future outcomes for the CPP are guaranteed to be good.

Like all actuarial work, it’s a prediction of a future world that is more likely than not to represent the actual future within a reasonable margin of error.  But when we look out 75 years – there is no way to predict with any certainty what the world will look like.  To frame what 75 years looks like – think back to 1950 (many of us were not alive) and look at how the world has changed since then.  By 1950 we had the printing press and the steam engine – and the transistor had just been invented.  Look at where the transistor has taken us in 75 years and it’s hard to imagine where AI will take us in 5 years, let alone 75.

What does the OCA think about the future?  They think that the CPPIB should be able to earn a real return of 4% per annum – on average.  Remember the real return is what you get after you subtract from the return you earned the inflation that makes money look like it’s worth more than it really is.

My question is, shouldn’t we just dispense with all this benchmark mumbo-jumbo and just agree that we will consider the CPPIB returns over 4% per annum to be the goal?  I am fine looking at 5- or 10-year averages.  I said this to my friend Don, and he subscribes to the benchmark approach used by CPPIB because there is no investment we can buy that guarantees a 4% real return.  If we have to use published indices as our benchmark, I think we should dumb it down a lot for folks like me – and they should be indices of publicly traded stocks and bonds.

Compensation

Investment management is one of the better paid professions in the world.  I get the specialized education and the employment risk you face if you do not deliver results.  Sometimes managers are fired prematurely because few funders are willing to wait 10 years to measure results.

But as I dug in on this topic, it really is the compensation discussion that is more interesting than the investment performance conversation.  Call me in another five years and we can look at 10- and 15-year returns to see how the CPPIB is doing.  Five years is just too short to reasonably measure whether our managers are succeeding or not, and the further the underlying investments diverge from those in the benchmark, the longer the period you need to determine if you are winning.

The real question for me is whether the fund performance benchmark is the right way to compensate all the employees at CPPIB.  It isn’t even clear to me if every employee is eligible for performance bonuses and if they are all based upon the same formulas and metrics. 

I am all for tying compensation to results – IF the person being compensated can drive that result.  In this case I just don’t know.  The problem we have is we can’t really measure results over a short period like five years, but how do you recruit someone today if you tell them that their compensation will be anchored to the results over the past 10 or 15 years?  So now you are stuck – you need to have a short time frame to measure results, but you can’t measure results over a short time frame.

What I do think is that paying the CEO $5 million+ to get a result that is close to the index doesn’t sound right to me.   The poor CEO is in a tough spot – defend your compensation and you look defensive – ignore the concerns and you look arrogant and out of touch with the regular Canadians paying your salary.  We may have an agency problem here and to me it is up to the directors to do a better job justifying the compensation they are directing away from the pockets of Canadians to the select few that have landed these jobs managing our money. 

I don’t know if I would want the job on the board of the CPPIB, but I feel somwhat confident that my vocal skepticism around the status quo being our best way to do things will ensure no one calls to invite me to join.