Let the College Fund Velociraptors Roam Free

Points of Return

Keep It Simple, Stupid

It's taken a decade since the Global Financial Crisis, but maybe we are on the verge of realizing that investment has been made far too complicated. An important new piece by Richard Ennis, former editor of the Financial Analysts Journal, inveighs against the Failure of the Standard Model of Institutional Investment and argues instead that big university endowments, and the many other institutional investors who follow them, should opt for something simpler.

As discussed in this space two weeks ago, endowments have steadily grown more complicated and ambitious over the last few decades. This involved a model of working out a portfolio allocation between different asset classes, and then looking for good managers to pursue winning strategies in each. For a period when hedge funds and private equity flourished, in the years before and after the bursting of the dotcom bubble, the big endowments (most famously Yale's) did vastly better than might have been expected from a standard split between stocks and bonds. Since the GFC, however, alternative investments haven't delivered so much, even though endowments continue to pour more money in. 

Ennis shows that investing in alts worked very well for endowments for about a decade, and then badly for another decade. This shows the excess returns (over a benchmark of bonds and stocks) the big endowments have recorded over the last 46 years:

He confirms the intuition that this can primarily be attributed to their continued love affair with alts. The GFC, which ushered in an era of rock-bottom interest rates, looks like the turning point:

But Ennis isn't just arguing that they made some poor asset allocation decisions, or timed their investment into alts wrongly. This isn't like a critique of a manager who missed a turn in the market. Rather, the point is that the expense of making life this complicated always required very strong returns. By choosing elaborate asset allocation models, and spending much time finding managers, endowments are wasting money in two ways. First, they are paying too much in fees, and second, they are spending too much time looking for managers when the prospective additional return should obviously be very low. 

The more managers you add, and the more they are genuinely active, the less you reduce risk, and the more your portfolio comes to resemble a giant tracker fund (only with exorbitant fees). By the time you add your 20th manager, the chances are that you are already up against the law of diminishing returns, as Ennis shows:

If hiring as many as 20 different managers is a waste of money, this is bad news for the big endowments, because Ennis reports that on average they have more than 100:

According to NACUBO [the National Association of College and University Business Officers], large endowments had an average of 108 managers each in 2019. That is up dramatically from an estimated average of 18 in 1994, when the remarkable run of endowment performance was getting underway. We estimate that large public pension funds average more than 100.The funds typically use multiple managers for each of their asset classes.

The great run of success beginning a quarter of a century ago benefited, it appears, from "first-mover advantage." There were relatively few hedge funds and private equity managers around, trying strategies that nobody else had yet attempted, and aiming at market anomalies that had yet to be exploited. Generally, the only people able to go into business as alternatives managers in those days were seriously good at what they did. Adding a few of them to an endowment portfolio was a great way to juice up returns. Now, none of those things is still true. Even the most skillful managers must eke out extra returns in a world where many of the most profitable anomalies have already been discovered and minimized.

Thus the last decade has witnessed a "failure of institutional investment strategy," and not merely a period when institutions were unlucky. "There is no reason, in other words, to expect the performance pendulum to swing in the other direction." So what should come instead? Ennis suggests an approach that was much discussed in the aftermath of the GFC, but since seems to have disappeared. Funds should have a big weighting or core in passive investments. This, he suggests, masters the trifecta of maintaining diversification, taking advantage of any mispricing in markets, and keeping costs under control. 

The advantages of passive investment are well enough known. So are the disadvantages. Many trustees would balk at leaving their endowments or pension funds wholly dependent on the prevailing direction of markets. As such long-term institutions genuinely have great advantages over others, it is incumbent on them to try to do better than match the index. And of course there is the issue of price discovery. As is often pointed out, markets with all passive investment couldn't function; nobody would be attempting to spot which securities were over- or under-priced, and nobody would be there to react to important news.

Ennis suggests an approach to minimize these objections. I think he succeeds. The "Passive Core" could have "whatever characteristics you want," with allocations between stocks and bonds tweaked according to risk tolerance, while geographic diversification could also be done according to taste. Alternatively, Ennis proposes an approach that would minimize the career risk for managers and trustees who often feel themselves to be judged solely against their peers. This would "mimic the market exposures of your peer group, virtually ensuring you earn the return of the peer group plus a significant net-return advantage as a result of saving on investment costs of 1-2% per year." Ennis suggests that this would more or less guarantee that the Passive Core would be in the top quartile of any peer group, so it would definitely appeal to a lot of investors — although I am not sure myself that "crowd-sourcing" investment decisions like this is ever a good idea.

As for the active side, it would be wholly unfettered, with no particular limitations on asset classes or styles. Ennis says: 

Every investment decision would be made exclusively on its merits, i.e., its potential to contribute alpha net of cost. In this vein, the essential identity of each asset would be its estimated contribution to total fund alpha, which is all that really matters in any institution's ultimate reckoning of its overall performance

So trustees could look far and wide for managers who offer an opportunity for performance above and beyond the market. Using an analogy from biology, the passive core is like large lumbering herbivores, while the active managers would be left to behave like nimble and aggressive velociraptors.

To this end, there should be no "silos" or fixed slices in a pie chart to be offered to managers offering one particular kind of asset exposure. Rather than run "beauty contests" of different managers all offering to manage the same mandate, he suggests "scouting" for them. Endowment staff would keep their eyes peeled for investment opportunities, and talk to plenty of people to see if they can spot good and aggressive managers. The critical point is:

not to allow these areas of interest — be they hedge funds, private equity or emerging markets — to morph into silos in the course of implementation. There is simply no advantage in maintaining rigid compartments within the portfolio itself. Doing so only encourages diversification within them; at the same time, having a fixed set of silos, or compartments, impedes genuine creativity

Acting this way, and keeping to a small list of managers, increases the chances of making a genuinely good return for the endowment and those who rely on it, while also ensuring that big institutions do their job of allocating capital well and identifying market inefficiencies. It also gives active investment managers a strong incentive to be aggressive, have the power of their convictions. In the process, the endowment would pay out much less in fees.

It certainly seems like a good idea to me. 



There are three weeks to go until the U.S. election. Polls continue to show Joe Biden leading by a margin that should translate into a comfortable victory in the electoral college. His advantage in the most important battleground states is statistically significant, while his overall lead in polls of polls is almost into double figures, according to RealClearPolitics:

Investors have finally started to believe that Biden is going to win. In betting markets, which saw the race as a tie at the beginning of last month despite a clear Biden lead in the polls, the Democrat is now perceived to have a two in three chance:

Does this mean political risk is over? No. Three important questions remain:

  • Who will win the presidency? (Probably Joe Biden – 33% chance that he doesn't.)
  • Will the election be so close that it is contested? (Less likely than it was, but more likely than the market currently bets.)
  • Who will win the Senate? (Probably the Democrats, but it's tight.)

For Biden to lose, either something has to happen in the next three weeks, or the polls must be wrong. He is gaffe-prone and a major blunder raising doubts about his ability to do the job could still change things. The president has the power to create more "October surprises," and a strong incentive to do so. And the supreme court nomination battle could add a new layer of unpredictability. Hearings begin this week, and will feature Democratic vice-presidential nominee Kamala Harris trying to force a "You can't handle the truth" moment out of the nominee, Amy Coney Barrett. This should be exciting political theater, with high risks on both sides.

For the polls to be wrong, Republicans have to be systematically more likely not to tell the truth about their intentions. One or two papers suggest this might be happening, although neither indicates that "shy Trump voters" could make up more than about two percentage points of the difference.

Then we come to the question of a contested election. Biden needs to win by four percentage points or more in the popular vote to win the electoral college. Add in the confusions caused by the pandemic, the heavy use of postal voting this year, and the two armies of lawyers ready to fight any recount to the last ballot, and the chance of an awful mess remains very real. 

In the Senate, polls suggest that Democratic candidates have their noses ahead in enough states to win 51 seats. But each race has its own dynamics. The Democrats' candidate in North Carolina admitted to an extra-marital affair last week, giving the Republicans a new lease of life. On such small things can very consequential decisions rest. This is how the Predictit market for North Carolina shifted after the revelation:

Where does that leave us? Joe Biden has a better chance of winning than Donald Trump. But there is plenty of uncertainty. Best to keep hedging against the possibility of a contested election, or of congressional gridlock afterward.


Survival Tips

Having quoted the "You can't handle the truth" scene, I'd like to to recommend A Few Good Men . It's currently streaming on Amazon in the U.S., and is one of those movies that seems to get better with time. It's been much parodied over the years, but I'd like to direct you to this brilliant pastiche, written by former colleague Izabella Kaminska, of how Jean-Claude Trichet reacted to German questioning during a European Central Bank press conference in 2011:

Reporter: What is your answer to German people and economists who want the return of the DM?
Trichet: You want answers?
Reporter: I think the Germans are entitled.
Trichet: You want answers? (SHOUTING)
Reporter: Germans want the truth! (SHOUTING)
Trichet: *You can't handle the truth!* (SHOUTING)


It gets better from there. Have a good week.

Like Bloomberg's Points of Return? Subscribe for unlimited access to trusted, data-based journalism in 120 countries around the world and gain expert analysis from exclusive daily newsletters, The Bloomberg Open and The Bloomberg Close.



Before it's here, it's on the Bloomberg Terminal. Find out more about how the Terminal delivers information and analysis that financial professionals can't find anywhere else. Learn more.



Popular posts from this blog

अभिषेक बच्चन ने कहा- पापा ने मेरे लिए कभी कोई फिल्म नहीं बनाई, उन्होंने कभी मेरी मदद नहीं की

The speech that sunk the world's biggest IPO

James Gunn Confirms ‘The Suicide Squad’ Panel For CCXP Worlds