The Tortoise and the Hare

Some newspapers have paid some well-deserved attention to the multi-million dollar bonuses recently handed to the executives of the Canada Pension Plan Investment Board (CPPIB) after they lost $24 billion of Canadian workers’ pension savings with their investments last year.

What has received less attention are the low long-term rates of return that the CPPIB has earned in the ten years since it was first established: an annualized rate of only 4.3% since 1999.

Before the CPP Investment Board was created, Canada Pension Plan savings were invested entirely in government bonds.  Canadians were told that we could make higher returns on these savings if they were invested in the stock market and in private investments instead.   The labour union I work for, CUPE, was one of the few organizations that opposed this move.

So what has this meant over the long-run — and how would have those investments fared if they had left them in boring old government bonds for the past ten years?  

The recently published annual report of the CPPIB shows that they have had very good rates of return in some years: 17.6% in 2004 and 15.5% in 2006.  It also boasts that it has earned $23.8 billion in investment income over ten years, even after last year’s losses.  But don’t expect them to tell us what we could have earned if these savings had been left in long-term government of Canada bonds over these years.    

Out of interest, I calculated what these returns would have been if the initial investment had been left in government bonds and if these funds and all the further net transfers to the CPPIB had been reinvested in long-term government of Canada bonds at their yield for each year since 1999.

These calculations show that leaving these investments in long-term bonds would have earned about $36.5 billion over the past decade: $13 billion more (and 50% higher) than the CPP Investment Board earned.  

(This may seem like a surprising result, given that the average long-term bond yield over the past ten years was 5%, only 0.7% higher than the CPPIB’s annualized rate of return.   But it is explained by the fact that bond rates were higher than the CPPIB’s return in early years and the CPPIB also lost a lot when its asset base was larger.)

 Many will no doubt argue that this past year was an anomaly and that returns will be much stronger in future years. However, interestingly, these poorer results aren’t just because of this past year’s disastrous returns. In five out of ten years the yield on long-term government bonds out-performed the CPP Investment Board’s investments and in each of those five years the size of the fund would have been larger. 

Last year it cost $189 million to operate the CPP Investment Board.   There would be an average of about $1,000 more in the fund for each of the 13 million contributors to the CPP if the CPP Investment Board had never been established.

Don’t get me wrong: I believe that the CPP is much preferable to private pensions plans and it also costs a lot less to operate.   I just think public pensions should be invested directly in public debt and thereby used to fund public investments.  It appears we’d all be better off if things had been kept that way.

Slow and steady wins the race: five times out of ten, and big time in the end.  

That’s my investment advice, worth $13 billion over ten years, and it doesn’t cost a dime.


  • Toby, another higher return that in my view is as important is that public investment gets higher returns for Canadians – which means it not only lowers debt but can be used to invest our economy, and by indirectly investing in creating Canadian jobs.

    So are you saying that under a Liberal govt they made these changes in how our money – CPP – is invested?

  • My question is whether government bonds would have had yielded the same rate (5%) over that time period had all the CPP been invested in government bonds. One would assume that such a huge increase in demand for bonds would naturally shift the equilibrium to a lower interest rate, and perhaps erode all of the 0.7% gain you highlight.

  • A brilliant knockout punch, Toby.

    I think David’s comment is worth considering, but we need a good sense of the numbers to answer it. I wonder if even the CPP has enough money in play to really affect bond rates in a significant way.

  • Hi David: that’s a good point and one that I certainly considered. Basic theory suggests that higher demand would increase prices and thereby reduce yields. I’m no expert on the bond market and the determinants of bond prices and yields, but it seems clear that there are other factors that have as significant an influence on bonds: short-term interest rates, inflation expectations, exchange rates, yields for alternative bonds, and other factors. While there may have been some impact from the higher demand, I don’t think it would have had reduced the impact in such a large way.

    While the differences in average returns over the ten years is not large (0.7%) and may appear to be close to the margin, the differences of yields in individual years are considerably larger. It’s these large differences in certain years and the changing profile of the investments that made this number much higher than I thought it would have been, based on the simple difference of average yields over the ten years. The higher returns from keeping funds in bonds come mostly at either ends of this period: at the beginning of the period the CPP funds invested aren’t as large; while at the end, the amounts are large, but there have been more significant factors at play.

  • Hi David once again:
    I just quickly crunched the numbers and claculated that, even if the average GoC long-term bond rate had averaged 0.7% less in each of the ten years (and so were essentially the same as the CPPIB’s annualized return of 4.3%), leaving them in GoC long-term bonds at that reduced rate would have still provided $7 billion more after the ten years. This may seem counterintuitive, but it is explained in part by the profile of the investments and reinvestments and the divergence of returns in different years. Averages can be deceiving.

  • Toby, I have been wishing someone would do this: it is excellent. Stay on this one, beat it to death, go both ways. One, what was lost by the Board as you did, and two, what we lost by going PPP instead of lending CPP funds for infrastructure.
    Send this into to ROB, and keep doing it until they give in an publish it. Remind them about the Madoff whistle blower. Putting pension liabilities at risk could hurt bad. Request a meeting with editorial boards to explain what is happening to our money.
    Hire a student to help you keep this file active. CUPE reports when they report.

  • Yep riff and riff some more.

  • Toby,

    Thanks for responding to my comment. Sounds like your math checks out 🙂

  • Dear Brother Sanger,

    Thank-you very much for preparing this analysis. this information will be useful in preparation for the forthcoming release of annual financial report of the Public Sector Pension Investment Board (PSPIB) which manages approximately $40 billion in federal public sector pension assets (inclusive of non-bank ABCP) and has a history of paying its’ executives exorbitant salaries and bonuses. Unfortunately, the PSPIB will probably only table the annual report in July under cover of the parliamentary summer recess! I will try and keep you posted on further developments.

  • Your analysis was referenced several times during yesterday’s HoC debate on the CCPIB.
    From Hansard : Wayne Marston :
    “…according to economist Toby Sanger, in the last 10 years, the CPP fund would have made $13 billion more than it did if it had been invested in government bonds, rather than in a diversified portfolio of equities, real estate and bonds.”

Leave a Reply

Your email address will not be published. Required fields are marked *