The reception of CPP Investments' Toronto offices in September, 2023. This year’s CPPIB annual report neglected to include the full historic data, writes Andrew Coyne.Chris Young/The Canadian Press
This time they didn’t just bury it: they omitted it altogether.
Over the years the annual reports of the Canada Pension Plan Investment Board have become longer and longer, and more and more opaque. This year’s is no exception, weighing in at a prodigious 84,000 words, most of it unintelligible – deliberately so, one is forced to conclude.
Nevertheless, there is always a tell, a little perverse Easter egg buried deep in the report for those bloody-minded enough to plow through it, wherein, as briefly and as casually as it can, the Fund lets slip the single, brutal fact that makes hash of all the boastful verbiage that surrounds it.
The pattern in recent years has been the same. Up front, in massive type, you’ll find the annual return on the several hundred billions the Fund has in its portfolio. It was 7.8 per cent this fiscal year, 9.3 per cent last year, 8.0 per cent the year before.
The newspapers dutifully report this figure, as they do the remarks of the CPPIB president, in which he modestly credits the hard work and due diligence of the board’s staff of dedicated asset managers – and of course, the Fund’s chosen investment policy of “active management.”
CPPIB sells portfolio of private equity fund interests for about $4-billion
You have to read a little deeper to find out what this means. Eight or nine per cent annual returns sound impressive, if you’re comparing them to, say, 10-year government-bond yields of three or four per cent. But pension funds don’t just invest in government bonds. They also invest in riskier assets like equities, real estate, or even private equity.
Generally, the riskier the asset, the higher the return. (Riskier assets have to pay a higher return, or no one would invest in them.) So when you’re assessing how well or poorly the CPP or any other fund has done, you have to measure it against a portfolio of equivalent risk.
And if you want to know how well or how poorly they have done as a result of their own efforts – the particular choices of assets to invest in – you measure their returns against what they would have earned had their portfolio performed as well as the average portfolio of equivalent risk.
How did the average stock perform last year? There’s no mystery to it. You look at the relevant stock index. If you’re investing globally – as you should, to diversity your risk – you look at an index of global equities. Similar indexes measure how the average fixed-income asset performed.
The people who manage the CPP fund know this. More important, they know other people know this, which is why every year the report also publishes the return to a benchmark or reference portfolio. The CPP has for many years based this on a mix of 85 per cent global equities and 15 per cent government bonds, which it claims entails about the same amount of risk as its actual portfolio.
CPPIB to sell remaining stake in European non-performing loan portfolio
The Fund takes the idea of benchmarking very seriously. Each year’s annual report goes to great lengths to explain how much “value-added” its employees have earned for Canada’s pensioners through their herculean efforts at “active management,” versus what they would have earned had they just bought the indexes, you know, like animals.
Actually, that’s a strategy, too; it’s known as “passive management.” It’s based on the observation, borne out over many years and mountains of research, that most active managers don’t, in fact, beat their respective indexes, especially after you take into account the fees they charge their customers. In a typical year two-thirds of them will fail; over longer terms, the proportion is even higher.
How much “value-added” did the CPP managers earn this year? Er, none, as it happens. Actually, they got beat by the indexes, too. Sorry, did I say beat? Correction: they got smoked. The Fund earned 7.8 per cent. That 85-15 portfolio? 13.9 per cent.
Remember, this is, by the CPP’s own account, a portfolio of equivalent risk. Yet had its managers pursued a passive investment strategy – the equivalent, statistically, of buying stocks at random – they would have earned, not the $56.9-billion of which it boasts, but $101.4-billion. Nearly twice as much.
Well, one year. Anyone can have a bad year. But the same pattern was observed last year: CPP Investments returned 9.3 per cent, but the market-index portfolio earned 13.4 per cent. And the year before that: CPP 8.0 per cent, monkeys flinging darts at the listings 19.9 per cent.
Pension funds should invest more in Canada, Senate finance committee chair says
In fact, the CPP has lately had to admit that it has been beaten by the monkeys, not just in one year or even three, but over the entire period since 2006, the year it switched from passive to active management. As last year’s report put it – on page 41 – over that nearly 20-year interval, “the Fund generated an annualized value added of negative 0.2%.”
But here’s the thing. This year’s report for some reason neglected to include the full historic data. Had it done so, after this year’s horrendous showing, I calculate the Fund’s “annualized value added” would have come in at negative 0.5 per cent.
To be fair, they did include a 10-year performance comparison. (You’ll find it on page 43.) Its verdict? While the Fund earned an average of 8.8 per cent annually, ol’ 85-15 earned 10.7 per cent – nearly two full percentage points more, on average, every year. These are staggering numbers. Take that first measure, the annualized 0.5 percentage point gap since 2007. Compounded, it means the contribution of active management has been to reduce the size of the national nest egg by more than $100-billion. Had the Fund stuck to passive management, its portfolio would today stand, not at $793-billion, but somewhere north of $900-billion.
How has the CPP reacted to this damning evidence? By changing the benchmark. The Fund now calls the 85-15 portfolio its “Market Risk Targets.” In its place, it now measures itself against a much more complex basket of indices, nearly 20 in all, including a broader range of asset types. Sure enough, the Fund performs much better against these “Benchmark Portfolios,” beating them by an average of 0.7 per cent annually over the last 10 years.
The Fund argues this is a more accurate reflection of the kinds of things it is actually investing in. But that’s just another way of saying that its chosen asset mix has underperformed the simple two-asset model – even assuming it is possible to accurately value the kinds of highly illiquid assets that now form such a large portion of the Fund’s portfolio.
And whatever the merits or demerits of different benchmarks, what is striking, and unusual, is that the CPP gets to choose its own – to set the bar for itself. Yet even then it struggles. The Benchmark Portfolios returned 13.2 per cent this year, beating the Fund by nearly as much as the old reference portfolio.
The Fund explains this away as a reflection of the “concentration of public market index returns in a set of large-cap U.S. technology and communication services companies with significant exposure to AI.” We may have been badly beaten by the market, in other words, but we were still right, because the market is caught up in a frenzy over a few trendy stocks.
Which is a version of what every active manager tells his clients after another year of underperforming the index, having gone heavily into bonds just before bonds collapsed or stayed in cash and missed a boom: sooner or later we will be proven right. Active management means never having to say you’re sorry.
What makes all of this so galling isn’t just the massive folly of plunging into active management at the very moment when other pension funds around the world (Japan’s Government Pension Investment Fund, Norway’s Government Pension Fund Global, the Dutch civil service pension fund ABP, several U.S. state pension funds) have been shifting more toward passive investment.
It’s the astronomical cost of the exercise. From around 150 employees in 2006, the CPPIB has ballooned to more than 2,000 today. Personnel costs now total over $1.2-billion, or an average of more than $570,000 per employee. Total costs, including operating expenses and management fees (but not including taxes or financing charges) now exceed $5.4-billion. Twenty years ago they were $54-million.
The executive suite is a particularly egregious case. The top five most highly compensated managers at the CPPIB make an average of more than $5-million each, including salaries, bonuses and other benefits. Twenty years ago the corresponding figure was less than a sixth of that.
You’d think the board of directors would rein this in. But the directors have ridden up the escalator along with everyone else. The chair of the CPPIB board this year will be paid more than $300,000 – three times what her predecessor was paid in 2006.
To have spent all this money, more than $55-billion in all since 2007, in the fruitless quest to beat the market – in fact, to deliver returns that substantially underperform what could have been achieved by a passive investment strategy, at a fraction of the cost – is nothing short of a massive public policy debacle. Maybe some day the Canadian media might even begin to take notice.