Back to the future: An old-school CIO defends old-school portfolio management:
Responding to our recent review of Ivy League endowment performance, (See: https://www.charlesskorina.com/nl51-special-report-on-big-endowment-performance/) one of our invaluable readers pointed us to a paper by Robert M. Maynard, who runs $13 billion as chief investment officer at the Public Employee Retirement System of Idaho [PERSI]. We, in turn, would like to recommend it to all of you.
“Conventional Investing in a Complex World,” appeared in the spring issue of the Journal of Investing. In it Mr. Maynard takes exception to the whole “endowment model” approach to portfolio management, raising some points that we haven’t seen articulated so clearly elsewhere.
As recruiters, we don’t presume to have any special expertise in investing per se, either theoretical or applied. We’re just observers. But we do have an abiding interest in how investment organizations are structured and staffed.
Mr. Maynard covers a lot of ground in his 18-page paper, but we were particularly struck by how clearly he coupled strategy issues to governance issues faced by boards and investment committees.
Whatever the merits of the Endowment Model [EM] as a path to better risk-adjusted returns (and he thinks it’s highly over-rated), it can’t be properly executed by most funds because of organizational constraints, or so he argues.
The EM investment vehicles, he says, are too numerous, too opaque, and too poorly benchmarked to be analyzed by even intelligent and diligent boards without very specialized experience, or by investment offices hobbled by low compensation and high turnover.
The long view from Boise; and a tidal surge in Anchorage:
Is Mr. Maynard worth listening to? Probably. He’s been in the saddle at Idaho PERS for more than 20 years now, making him one of the longest-serving CIOs we know of. He’s seen a lot of fads and fancies come and go since 1992. Meanwhile, PERSI, following his old-school strategy, has done just fine.
Mr. Maynard earned a JD from UC Davis, then, before landing in Boise, he worked in Alaska, including a hitch as Deputy Director of the Alaska Permanent Fund. Despite skipping an MBA, a read of his paper makes it clear that he understands the theory behind the EM perfectly well. He just doesn’t think it’s a strategy appropriate for most public pensions. And he would probably argue that it’s iffy even for many smaller endowments and other funds.
In the edenic early 90s, as he recalls them, portfolios were simple, transparent, and focused; relying primarily on public markets.
But then, envying the conspicuous success of David Swensen at Yale and Jack Meyer at Harvard, many institutions were seduced away from the old school. They were convinced that too much of their total portfolio risk was dominated by public market risk. The EM eschewed public market exposure, hunting for alpha among intensely-managed and illiquid vehicles like private equity, hedge funds and private real estate.
Even many public pensions were trying to emulate Big Ivys, and stodgy funds like PERSI were beginning to look like outliers. After all, EM seemed to work as promised. That is: it did, until it didn’t.
As an old Alaska hand, Mr. Maynard offers this vivid metaphor:
The Cook Inlet next to Anchorage, Alaska has the second-highest tides in the world… Occasionally, newcomers or tourists are killed as they are lured by the bounties unveiled at low tide, only to be stuck in the mud and drowned as the tidal surge appears at far greater speeds than they imagined possible… There is a great deal of similarity between the occasional Alaskan tourist and recent entrants into the endowment model of investing.
… The events of 2008-2009 showed that both theoretically and practically, the endowment model failed its first stress test miserably.
Old-school in action:
What, exactly, does a “conventional” portfolio look like these days, and how does it perform?
Here are PERSI allocations in 2013 compared to: CalPERS, NASRA (average of U.S. public pensions, per National Association of Retirement Administrators), large U.S. endowments (NCSE over $1billion) and Harvard. All are actual, not policy allocations (except for Harvard). We’ve massaged a few of these numbers to fit the NCSE categories, but they’re close enough for our purposes.
We can see that PERSI is, indeed old school, even compared to its public pension brethren. They have only half the allocation to alternatives (which, for PERSI, are private equities only), with the difference shifted to stocks and bonds. Real estate (direct investment, plus REITs) is roughly the same as other public pensions.
Big endowments, on average, have five times as much riding in alternatives and half as much, proportionately, in stocks and bonds, relative to PERSI.
How about performance?
Annualized Percent Returns (as of FY2013):
PERSI has outperformed Harvard over the most recent five years. They returned an annualized 4.8 percent, versus Harvard’s 1.7. Their smaller losses in 2008/2009 make most of the difference. Not bad.
Expand that to ten years, though, and Harvard wins. PERSI returned 7.8 for ten years, versus 9.4 for Harvard (and 11.0 for Yale). Note, however, that they still beat the average college endowment. (The NCSE numbers here are preliminary for FY2013, and are averages for the whole NCSE universe, not just the big ones. PERSI and NASRA certainly wouldn’t look as good against the over-1 billion endowment group.)
Read his paper carefully and you will see that Mr. Maynard nowhere says that he expects to consistently outperform the Ivys. What he does say, pretty emphatically, is that his strategy is “very viable.”
What is “viable” in Mr. Maynard’s playbook?
He says: “The goal should be to survive comfortably with the highest odds over the very long term. As long as an institution enjoys long-term stability, produces returns comparable to the market, and produces returns at least in the pack of institutional investors, there should be no reason to be jealous of the few that hit home runs.”
By his own standard, then, PERSI is certainly viable. They’re ahead of their NASRA peers in 10-year performance. They also beat the Mellon Median Fund — one of their internal benchmarks — over ten years; and miss it by only a hair over five years.
Succinctly, PERSI’s strategy has been to make its money by losing less when public markets drop, then holding on and getting close-to-market returns when they rally, and in the calm stretches between the booms and busts. (He has a pretty sophisticated argument in the paper about how and why that’s possible, which we won’t try to summarize here.) That was the plan Mr. Maynard articulated when he came aboard twenty years ago, and it’s worked pretty much as intended. In FY2009, when Harvard lost 27.3 percent, and Yale lost 24.6 percent; PERSI “won” by losing only 16.0 percent. And, they did it with no liquidity panic.
How is this done in practice? Here’s where the cockroach comes in.
The Way of the Cockroach:
In the old-school model, with your money mostly in liquid, public-market vehicles, with relatively fewer managers and fewer moving parts, you can take maximum advantage of rigorous re-balancing:
The cockroach lives in a very, very complex environment, with one of the best long-term success rates of any creature. Yet it has only one defense mechanism: running in the opposite direction from a puff of air. The equivalent for the investment world is, at the core, a very simple structure founded upon public market diversification with one basic defense mechanism: See a volatile movement and react in the opposite direction (i.e., rebalance into it).
But public funds can only follow the way of the cockroach in real time if they are dealing with relatively simple and transparent investment vehicles, benchmarked to real-time market indices. When things go wrong with opaque, illiquid vehicles, the cockroach is stymied. A more complex organism is required.
That famous Markowitzian free lunch from diversification is only available to investors who can rebalance in a consistent and disciplined way. OSM funds [like PERSI] were able to rebalance using independent, third-party pricing after March, 2009, buying equities at the bottom and minimizing their losses. Endowment-style funds, loaded with hedge funds, illiquid commodities and portable alpha devices, couldn’t; and they suffered the consequences.
As Mr. Maynard puts it:
If you don’t rebalance to asset types, you will get no diversification return benefit. If you can’t rebalance to an asset type, you cannot get diversification return benefit.
Even if you have the will to rebalance rigorously, you must also have the ability. Private assets have limited rebalancing opportunities, and it gets worse as volatility rises. You can’t cash in on their supposed non-correlation when you need to.
They are useless for diversification at exactly those times when the diversification benefits would be greatest.
Mr. Maynard ticks off the virtues ascribed to EM investors, and finds most of them unconvincing.
For instance: It’s undoubtedly true that if you can invest in the top tier of private equity, VC, and a few outstanding hedge funds, you will do very well. But this hasn’t been easy even for the elite endowments with their old-boy networks. For most other funds, it’s been an invitation to high fees and mediocre returns.
The relative simplicity of the OSM also improves risk control, which is closely related to the way of the cockroach. That “puff of air” the cockroach reacts to – an unexpected deviation from expected risk and return — has to have a clearly perceptible direction and magnitude, or the cockroach won’t know where to run, or how fast.
For “sophisticated” EM portfolios, risk must be managed actively through quantitative models and risk-control systems: e.g., value-at-risk, and so forth. This is the only way to get a grip on what’s going on in highly-leveraged and black-box strategies.
VaR made a lot of money for the consultants who sold it, then crashed repeatedly in real life. It was invented in the back offices of JPMorgan in the early 1990s then, ironically, and most recently, it helped JPM walk into that $2 billion trading loss in London in 2012.
See, for instance:
The best risk control lies in being able to see the whole portfolio easily and being able to spot deviations from the expected without difficulty. In the OSM, he thinks, quantitative models are best used as a supplemental, and not a primary risk-control device.
You want the endowment model? You can’t handle the endowment model!
Let’s stipulate that the Endowment Model over the long haul (even with fat tails like 2008/2009 included) may generate somewhat higher absolute returns, but that the old-school model is still “viable” for Mr. Maynard and his constituents. So what?
Our takeaway is that applying the EM to a public pension (and probably a lot of other funds) is simply not a practical possibility. Even if they wanted to emulate the Ivys, they can’t get there from here. Their institutions can’t effectively execute the EM strategy.
Mr. Maynard doesn’t put it quite as emphatically as Jack Nicholson’s splenetic Colonel Jessup in A Few Good Men, but what we infer is that OSM portfolio management is the only kind most publics can effectively handle. Fortunately, he says, it’s good enough to meet the needs of an institution that’s not too greedy.
Organizations that follow complex and opportunistic strategies typically have investment staffs of dozens, if not hundreds, often regardless of the size of the assets. A successful long-term strategy with this type of approach requires consistently maintaining the resources necessary to implement it. That has often proved to be a stumbling block for many public and private institutions…
Salaries of public-fund investment personnel cannot approach those in the private sector, and attempts to provide for any incentive compensation often fail entirely when the market delivers losses – even if those losses are much less than expected for a particular fund. This is not just a public entity problem; corporate funds and university endowments also run headlong into it.
There is also the problem of continuity, which is exacerbated by over-complex strategies:
Investment disasters regularly occur because of the inability to maintain the overall investment focus.
(All of this had a familiar ring for us. We’ve previously tried to quantify the turnover among CIOs and its implications. There are about 1,300 funds of all kinds in the U.S. big enough to justify a full-time, professional CIO. And the typical CIO holds his job for five to eight years. That implies that one-fifth to one-eighth of the slots in the total population will turn over in any given year. That is: annual churn for the whole population will be between about 12.5% and 20% or about 160 CIO openings per year.)
See our research here: https://www.charlesskorina.com/346/
Mr. Maynard points out that relatively high turnover among both public-fund CIOs and board members makes it very, very difficult to maintain a consistent approach through even one investment cycle, let alone over the “long term.
An institution’s investment approach has to be one that can be administered by the people likely to fill boards and staffs in the present and passed along into the future. Most board members are intelligent and well-rounded people, he thinks, but with no special background in investing. And, at the staff level, limited compensation means long-timers are more the exception than the rule. Boards and staffs are rarely around to suffer the consequences of their decisions.
All of Mr. Maynard’s observations in this area are consistent with our experience as recruiters and consultants to institutional funds.
Are there other “conventional” portfolios out there like the one Mr. Maynard runs?
Probably several, but we can point to at least one. Way back in Skorina Letter No. 21 we interviewed Ken Lambert, who ran the $22 billion Nevada PERS fund. We noted that:
Ken Lambert…runs a plain-vanilla, mostly stocks-and- bonds portfolio that would have looked right at home in the Eisenhower administration…and he explained that his real secret sauce, if he has one at all, is his proprietary re-balancing formula.
You can read it here:
We’ll have some more to say about Ken and the Nevada fund in an upcoming issue.
And I alone am escaped to tell thee:
Despite the average 5-year turnover of CIOs, Mr. Maynard is still at his post. Like Melville’s Ishmael, he’s survived a long, stormy, cetaceously-challenged voyage that scuppered many another mariner.
He was lively and gracious when I spoke to him briefly about his paper.
Here’s an obvious question, Bob: If “conventional investing” is good enough, then why are so many funds, especially public pensions, still loading up on unconventional investment vehicles?
It is an obvious question, but a good one. The conventional model I advocate will probably produce more volatility in the short run (1 to 5 years) and maybe somewhat lower returns over the very long run as compared to the endowment model. But it can be successfully executed by a very small staff and it can be explained to a non-expert board and public. I believe it will produce real returns of 3 to 5 percent (neglecting inflation) over the long run. And its bias toward domestic equities reduces inflation risk.
But the conventional model won’t work if you insist on real returns of 6 percent or more over the long run. If that’s what your board demands, then beating the market becomes an absolute requirement. Then the endowment model becomes the only game in town. Some funds are obviously in a position where they feel they have to reach for those higher returns and persuade themselves that it’s doable. I think most of them will be disappointed.
Note to readers:
Mr. Maynard’s paper in the Journal of Investing is posted behind a pay-wall and is only available to subscribers. However, an earlier version is accessible to our readers on the website of the Brandes Institute.
The earlier version has a variant title, and some of the charts aren’t as up to date, but the text is identical.