CPPIB Gets 'Lucky' With Active Management?

The National Post published another hit piece from conservative commentator Andrew Coyne on how the Canada Pension Plan Investment Board (CPPIB) is lucky with active management, for now:

In 2006, the Canada Pension Plan Investment Board announced it had switched from a “passive” to “active” investment strategy. No longer would it merely seek to replicate the performance of the broad equity and bond market indexes in its own portfolio, as it had earlier been required to do; now it would free its managers to pick and choose, buying and selling at particular times in an effort to “beat the market.”

In pursuit of its new mandate, the fund has undergone a massive expansion in staffing, compensation, and costs. The number of employees has gone from 164 in the year ended March 31, 2006, to 1,661 in fiscal 2019. Total costs have grown from $118 million to $3.3 billion annually, or from 0.12 per cent of assets to 0.83 per cent. All told, the fund has spent about $22 billion over the past 13 years, or more than six times as much as if it had kept spending the same amount, in proportion to assets, as it did in 2006.

What a relief to find it was all worth it! Or so claims the Parliamentary Budget Office in a new report. Having set out to determine “whether [the CPPIB’s] active management strategy resulted in higher returns compared to a passive strategy,” the PBO finds that, indeed, the fund’s annual returns under active management were 1.2 per cent higher, on average, than they would have been under passive, contributing an extra $48 billion, in total, to the fund’s holdings.

This is an extraordinary finding. The CPPIB itself, not known for keeping its light under a bushel (along with everything else, the length of the fund’s annual report has ballooned, from 25,000 words in 2006 to 90,000 words in this year’s doorstopper), does not make this claim. Its 2019 annual report estimates the fund’s $392 billion portfolio is $29 billion larger than it would have been had it stuck with passive management. On average, it claims to have beaten the market — measured by a “reference portfolio” of stock and bond indexes — by just 0.6 per cent annually.

What explains the discrepancy? One part of it seems to be the asset mix the PBO chose for its benchmark. The CPPIB doesn’t just invest in the public markets. Much of its portfolio is in more esoteric, less frequently traded assets whose values are harder to assess: private equity, real estate, public infrastructure projects and so on. There’s general agreement that it’s significantly riskier than a portfolio with a comparable mix of debt and equity, but made up entirely of publicly traded assets.

That’s important because, other things being equal, a riskier portfolio would be expected to perform better than a safer one. (Riskier assets have to pay better, to induce people to hold them. Or to put the matter another way, if you want to do better than the market average, you have to take on more than average risk.) If the CPPIB were to consistently earn a higher return than a benchmark of equivalent riskiness, it would have something to brag about. But outperforming a safer portfolio might well be underperforming, on a risk-adjusted basis.

So while the CPPIB nominally holds just 56 per cent of its portfolio in equities, for comparability its benchmark would have to hold a much higher proportion of equities. The fund itself has belatedly admitted this: its reference portfolio is now made up of 85 per cent equities, up from just 65 per cent in 2015. And yet the PBO’s “baseline” case, the source of its claimed 1.2 per cent annual premium for active management, is measured against a passively managed portfolio with 70 per cent equities. At 85 per cent, the claimed gains dwindle to 0.6 per cent.

Worse is the PBO’s treatment of costs. The return that matters, as any mutual fund investor knows, is not the gross return but the return net of costs: one of the reasons actively managed funds tend to underperform passively managed is because their costs are so much higher. Bizarrely, while the PBO study deducts transaction costs and management fees — which are now, respectively, 17 times and 44 times what they were under passive management — from the CPPIB’s returns, “operating expenses were assumed to be the same under either approach.”

But this is absurd. Operating costs, at $1.2 billion, are now 22 times what they were in 2006 — five times as much, relative to assets. Assuming instead that they were the same as under passive management would greatly overstate the CPPIB’s net returns, and exaggerate any premium to active management.

But leave that aside. Even if the PBO study had not used an unrepresentably safe benchmark, and even if had not understated the CPPIB’s costs — even, that is, if the CPPIB had actually beaten the market by the amount claimed — it still wouldn’t make the case that this outperformance “resulted” from the switch to active management, still less prove that active management is superior.

Suppose, as an alternative scenario, the CPPIB’s managers had bet the fund on the fifth race at Woodbine, and suppose their their horse had won. It wouldn’t mean either that betting on the horses was a good investment, or that the CPPIB knew how to pick horses. It would just mean they got lucky.

The case against active management, amply supported in the research — at least two-thirds of investment managers in any given year underperform the relevant benchmark — is not that nobody ever beats the market. It’s that they do not do so reliably, predictably, or systematically. Probability theory teaches us that, if you toss a coin 10 times, it should most often come up heads five times, tails the other five. But it’s entirely possible for it to come up heads 10 times in a row, without discrediting probability theory.

The CPPIB has bet billions of dollars over the past 13 years on the proposition that it can consistently beat the market. It’s possible that it has, so far. Will it still be ahead of the game 13 years from now? We shall see. But by then the money will have been spent, and the managers paid, and it will be too late to ask for it back.

I read Andrew Coyne's long rant yesterday and it confirmed what I've long known about him, he's a conservative political pundit who sounds fine on television commenting on politics, but is completely clueless about the CPPIB and how its active management strategy is critical for ensuring the long-term sustainability of Canada's retirement system.

Keep in mind, Coyne has published nearly a dozen opinions over the last half decade or more (and was passionately opposed to the CPP reforms in general in the 1990s) criticizing active management, so when he calls CPPIB "costly" and "lucky" in a national newspaper, it really irritates me because unlike reporters, he's a well-known commentator who is not held responsible for spreading lies and disinformation on an important topic he knows very little about.

Moreover, his comment is an affront to all of Canada's large public pensions which have received international acclaim for the way they manage pensions to ensure they are sustainable over the long run.

Anyway, it's clear Coyne has an agenda and never bothered to read about the history of CPPIB:

CPPIB emerged out of the realization in the 1990s that the CPP fund was unsustainable primarily because changing demographics were leading to fewer workers supporting a growing number of retirees. Federal and provincial finance ministers acted boldly in creating the CPP Investment Board and in making other changes to the CPP. This new organization, operating at arm’s length from governments, would invest the funds of the CPP to help ensure its long-term sustainability.

In 2006, CPPIB launched active management strategy to broadly diversify investments by asset class and geography. In particular, given CPPIB's long-term liquidity profile, a decision was made to invest more assets in private markets all over the world (real estate, infrastructure, private equity and other assets) to capture the long-term risk premiums of these illiquid asset classes.

The shift into private markets is in line with CPPIB's investment strategy which is structured to be resilient in the face of wide-ranging market and economic conditions.

Importantly, by investing in public and private markets all over the world, CPPIB has realized significant gains over its passive benchmark (Reference Portfolio).

A recent PBO report states that having the money in the Canada Pension Plan fund actively managed by investment experts has been worth nearly $50 billion in extra returns since the mid-2000s and the extra returns from active management justify the costs:

“Given that active management requires more personnel to conduct the required research, and also involves more transactions, this strategy comes with increased complexity and additional costs,” the PBO report says. “The goal is that in the long run, after netting these costs, the strategy will achieve a higher return than an identified benchmark.”

So far, so good, the analysis found: In each of the 12 full years the investment board has been using the active-management approach, including in the recession at the end of the 2000s, it’s brought better returns than passive investing.

Even when all the extra costs of active management of the Canada Pension Plan fund are accounted for, its experts’ wheelings and dealings are worth an extra 1.2 per cent in investment returns in an average year. That’s added up to $48.4 billion in extra investment profits.

Coyne dedicates most of his comment on costs, however, for a business like CPPIB (and it is a business), costs don’t exist in isolation, they are borne to create value. $239 billion of net income (after all costs) over the recent 10-year period with a net return of 11.1%. None of this is mentioned in his article or that based on independent projections by the Chief Actuary of Canada, the Fund should be just above $300 billion today yet it stands just under $400 billion, net of all costs.

He also criticizes the PBO for using a passive benchmark made up of 70% equities to measure these gains but even if you use CPPIB's actual Reference Portfolio which is made up of 85% Global public large/mid-cap equity (including Canada and emerging markets) and 15% Canadian Federal and Provincial Governments Nominal Bonds, you'd see the Fund has generated $29.2 billion of compounded dollar value-added, after all costs, since the inception of active management at April 1, 2006. This is hardly insignificant.

Equally important, he fails to realize that by shifting more assets into private markets all over the world, CPPIB has created a more resilient portfolio over the long run which isn't beholden to the vagaries of public markets.

Have a look at CPPIB's asset mix as at March 31, 2019 (from fiscal 2019 Annual Report):

As shown, only 33% is invested in Public Equities, 10% in Government bonds and absolute return strategies. The rest is in Real Estate (12%), Private Equity (24%), Infrastructure (9%), and Credit Investments (9%) which also include Private Debt.

The shift into private markets is completely lost on Coyne but if you go ask the average Canadian if they invest all their money in stocks and bonds, most will say no, their house is their most important asset. Others invest in rental income if they can to diversify their investments.

CPPIB has huge advantages -- a long investment horizon, more certainty of assets and the size and scale to invest in private markets  -- so not only can it afford to take on illiquidity risk, it would be highly irresponsible and completely against the fiduciary duty of its senior managers to manage these assets in the best interest of their contributors and beneficiaries if they didn't invest in private markets all over the world, capitalizing on their competitive advantages.

Investing in commercial real estate or a toll road or a significant position in a tech company is high risk for the average investor. It is absolutely appropriate and prudent risk/return for a global investor like CPPIB working for multiple generations – for decades and decades, to do so.

Canadian retail investors cannot buy commercial real estate, infrastructure and private companies all over the world, or invest in the best hedge funds all over the world, which is why I once commented that CPP is infinitely better than crummy RRSPs.

Again, all this is lost on Coyne, and to make matters worse, he cites spurious academic research which holds true for most mutual funds but is irrelevant to CPPIB because a huge chunk of its portfolio is in private markets which by definition are less efficient than public markets and that's where most of the value-added is being generated at Canada's large pensions over the long run.

Private markets are not immune to market shocks and carry their own set of risks, but since pensions have a long investment horizon, they can easily wait out a cycle to realize big gains on these investments over the long run if they invest wisely, keeping costs low by co-investing alongside their trusted partners and diversifying their holdings across the world.

Strikingly, Coyne managed to file a long comment about active management without once noting diversification which is by far the most critical thing CPPIB is doing.

Lastly, Coyne doesn't understand that CPPIB is the envy of the world and just how lucky Canadians are to have a national pension fund managed at arm's length from the government with one mandate: to invest the assets of the CPP Fund with a view to achieving a maximum rate of return without undue risk of loss.

Below, Mark Machin, president and CEO of the Canada Pension Plan Investment Board, offers his global economic insights and discusses the plan's investment strategy and his thoughts on CPPIB's 20 year anniversary. You can read more about CPPIB's fiscal 2019 results here and why its CEO deserved his big bonus here.

If you ever meet Mark, you'll see not only is he one of the nicest people in the pension industry, he's extremely intelligent and very well prepared. Luck has nothing to do with CPPIB's success, and 13 years from now, Andrew Coyne's latest drivel on CPPIB will be proven spectacularly wrong.