Christmas in Cambridge: HMC gives themselves a raise

We don't know the details of the HMC compensation plan, but we know, in a generic way, how these things work.  The board hires one of the big compensation consultants (from among Mercer, Watson Towers, Aon Hewitt, etc.) to recommend an industry-standard plan.  Then the comp committee and board tweak and add features to suit their specific needs, run it past the lawyers, and sit down to negotiate with the lucky devil across the table.

As they craft a plan, they have two overt objectives and one implicit one: First, they have to reward performance to get the talent they need in the first place.  Second, they want to keep the talent they've got, so the plan has features which encourage retention.

The third, less overt, goal (or constraint) is to keep the numbers low enough to avoid blowback from Certain Sensitive Constituents.  The latter pinches a lot harder at, say, a public pension fund, than at an Ivy endowment.

Among the large nonprofits we're aware of, the bonus is based primarily on average investment return for a recent (rolling) multi-year period.  In the U.S., it's typically a three-year average.  The big Canadian public funds seem to prefer a four-year average.  Subtract benchmark return from actual return over that period and you get the percentage "value added" which is the basis for computing the bonus in a given year.

Part of that bonus is paid out immediately, but a substantial piece is usually held back (deferred), and not vested or payable until some later year.  That's the retention feature: a manager who departs before her bonuses vest could forfeit them.

In practice, it's complicated.  There may be thresholds, caps, clawbacks, etc.  And the bonus isn't based solely on returns; the board may also impose other, more subjective goals for managers.

We like to use the most recent 5-year annualized returns to get a sense of how well the current investment managers are doing their jobs; it suits our purposes as executive recruiters.

However, today, let's first turn back the clock to look at the three-year returns as of 2010 for HMC and a few other endowments.  That's the performance period which is probably most relevant for salaries paid in calendar 2010, the latest available in most cases.  Then we will look at Harvard's returns for the three years ending 2011, which is most relevant to the 2011 payroll just disclosed for the HMC team.

Here are the five biggest U.S. endowments, with annualized performance in fiscal years 2008-2010, and bonuses paid in calendar 2010.  The absolute returns are all negative for this period, but apparently everyone beat their benchmarks.

HMC had annualized returns of about -4.2 percent, fourth-highest (poorest) among the Big Five (Princeton did a hair better than Harvard, despite the rounding which makes it look like a tie).  As we see, Ms. Mendillo still got the biggest bonus and biggest paycheck among the Big Five U.S. endowments.

Bruce Zimmerman at UTIMCO had the best returns by a good margin.  But his bonus was only one-third of Ms. Mendillo's.

Advancing the calendar one year, here are three-year annualized returns as of 2011 for the same group:

All the three-year returns have improved markedly, but Harvard is now fifth out of five, and the only one with a negative number.  Still, apparently, the value-added by HMC's managers improved over the previous year: enough to add about $1.7 million to Ms. Mendillo's bonus.

It will be a few months before all 2011 salaries are public, and we can fill in those other boxes.  But, if history is any guide, we'll still expect Ms. Mendillo's comp to be the highest in this group again.

Among all U.S. endowment leaders, only Narv Narvekar at Columbia University was paid more than Ms. Mendillo in 2010.  He earned $4.4 million that year.  But Columbia's 5-year return as of 2012 was 4.9 percent, versus only 1.2 for Harvard.

With more than $30 billion AUM, Harvard is almost sui generis in the U.S.  It's half-again as big as Yale, to whom it's usually compared, and Texas (UTIMCO) and almost twice as big as the other major endowments.

If we're going to benchmark Harvard's pay and performance, maybe we need a bigger boat.

One other place we can look for highly-compensated investment managers is among Canada's big public funds.  They have a reputation for hiring outstanding investment staff and paying them well.  (We intend to take a closer look at these funds in a future letter.)

Meanwhile, let's see how they stack up against the largest U.S. endowments, including Harvard, re pay and performance.  It happens that we do have more recent comp numbers for the Canadians: for both 2011 and 2012 for most of them.

Some of them are pension funds, per se.  But others, as Jacques Demers, President and CEO of OMERS Strategic Investments explained to me last week, contain a hodge-podge of funds; some pensions, some not.  Those include Caisse, BCIMC and AIMCO.

The first chart gives asset size; the second lists total compensation; and the third ranks the funds by performance.  The results are interesting.

First, AUM:

N.B.: We haven't bothered to convert the Canadian dollar amounts. The small difference between the Loony and Yankee dollar is immaterial for our purposes.

Now, let's look at compensation:

Note that not only Ms. Mendillo, but also most of her senior staffers make more than the leaders of these much larger (and, mostly, better-performing) funds.

In most cases, the chief executives of the Canadian funds are also the designated (or effective) chief investment officers.  At OTPP and OMERS, the CIO is a direct report to the CEO.

Now, here are the performance numbers for the past six years for all twelve funds.  We've reverted to our usual five-year annualized returns as of 2012.

Three of the Canadian funds surpassed all of the largest U.S. endowments; and Harvard ranks twelfth out of twelve in this group on a five-year basis.

US endowments are in brown, Harvard is in red, and the Canadians are in black.

Six years performance numbers for all twelve funds.  Five-year annualized returns as of 2012.

Any reader who wants a clear explanation of how a CIO compensation plan works could, for instance, take a look at this section of the OMERS annual report:

http://annualreport2011.ac.omers.com/uploads/OMERS_AC_CDA_2011.pdf

Or, the compensation discussion in the CPP annual report, here:

http://www.cppib.ca/files/F2012_-_YE/CPPIB_2012_Annual_Report_-_English_-_Online.pdf

Both funds proclaim that they're committed to a pay-for-performance approach to compensation for all employees, with the proviso that the return must be achieved with an acceptable level of risk and that compensation policies don't encourage undue risk-taking.  CPP notes specifically that they have to compete for talent globally, with compensation to match.

Performance-linked compensation, which is also calibrated to control risk-taking, seems to be working for them.

There are things to learn from our Canadian neighbors.  Some U.S. nonprofits could do better on the performance-linking part.  And certain Wall Street firms should certainly ponder that controlled risk-taking part.

We aren't moralists or social critics.  Those fields are already over-crowded.  We're just humble headhunters, and we don't think there's any Platonic standard for how much anyone should be paid to do anything.

The organizations we talk about are all run by responsible people with access to all the information they need to do their jobs.  A Canadian Crown Corporation, a U.S. public pension, or the board of an Ivy university each operates in its own peculiar context.

Harvard, for instance, and their donors should do what they think best with their own money, including how they choose to compensate their people.  Maybe the glory days of Jack Meyer and decades of 15% annualized returns will come again.  In which case, no one is going to quibble about a few million more or less paid to the people who make it happen.  Certainly not us.