Interesting article by Pensions Expert around investment consultant manager selection, with input from Roger Brown of IC Select. Pension trustees have an important, but difficult, role to play in the good governance of their scheme. As an industry, pension experts should strive for increased transparency and clear presentation of salient information across all areas of impact.
Markets have since undergone profound change, however, and as Pensions Expert reported in April, high levels of dispersion in equity returns signalled the beginning of a period of volatility that should have allowed active managers to shine.
And yet, the next iteration of S&P Dow Jones’s Indices Versus Active Funds Europe scorecard shows that the majority of equity active fund managers underperformed the S&P Europe 350 in Q1 this year.
European funds lost 15.5 per cent in March compared with a 14.1 per cent drop in the benchmark. The first quarter of 2020 saw fund returns down 22.7 per cent, while benchmark returns were down 22.4 per cent.
“While active managers’ performance in the quarter may appear broadly level with the benchmark, the chances of choosing an active fund that outperformed were not,” S&P associate director Andrew Cairns says.
“The majority of Europe Equity fund managers were unable to beat the benchmark in either March or Q1 2020 as a whole, with 66 per cent and 57 per cent underperforming the benchmark, respectively.”
Few dispute that active managers play a role in the efficient allocation of capital. But when their results paint them as more of a social good than a must-have in some (but not all) markets, why are they still so popular?
A key driver of active’s enduring popularity is the investment consulting industry. Consultants have been branded by one study “the kingmakers of institutional asset management”.
In the UK, 74 per cent of £7.7tn total assets under management are managed on an active basis, according to a 2019 estimate by the Investment Association.
In 2018, an investigation by the Competition and Markets Authority estimated that 73 per cent of pension schemes purchase investment consultancy services. Assuming the consultancy advises on all of the schemes’ assets, the Financial Conduct Authority estimates that investment consultants advise on a minimum of £1.2tn of pension scheme assets.
Investment consultants spend huge amounts of resource on manager selection, with teams of more than 200 at the largest.
But a series of academic papers, the most recent of which landed in 2018, query the basis on which the performance of recommended managers is measured, and highlight the paucity of evidence made available for external analysts to make an objective evaluation.
“The reaction wasn’t great,” says Roger Brown, founder and director of IC Select, of research published by Blacket, which he likewise founded and directed, back in 2006. The goal of the research project was to measure the effectiveness of active managers and investment consultants. We found that, on average across the industry, [their recommendations] don’t add any value.”
Some of the findings of the Blacket research were startling. While 70 per cent of pension schemes “felt the overall value they received from their investment consultants was good or very good”, the study found that 80 per cent of the mandates it analysed underperformed against the objective, net of fees.
One reason the findings were poorly received and gained little traction, to Mr Brown’s consternation, is that they had the potential to offend, not only the investment consultancies, but trustees as well.
“One of the issues is the more we drilled down and measured investment consultants, by default, you’re also measuring whether the trustees are adding value,” Mr Brown says.
Among the report’s findings: 58 per cent of final decisions by trustees “selected one of the poorer performing managers from a shortlist and 42 per cent selected one of the better performing managers”.
Blacket found: “Past performance plays a significant role in a consultant’s manager recommendation for equity mandates. Forty-eight per cent of recommendations were top quartile and 32 per cent were second quartile over the previous five years.”
Of course, citing any document produced 15 years ago will rightly be open to the rebuttal that times have changed. Consultants interviewed by Pensions Expert say that, now at least, past performance does not factor into their decisions to anything like the degree it once did, and Mr Brown agrees the shift to other attributes does constitute a change.
But on the question of whether investment consultants add value in manager selection, he argues that the new focus on investment philosophy has yet to translate into proven performance.
“Not all, but most of the manager researchers, when they go through this process, they score the manager on how well they do their investment philosophy,” he explains.
“You can look at whether there’s any predictive power, from whatever the factor is, to subsequent performance. And I have asked some of the consultants if they’ve ever looked at that. And almost nobody says they’ve bothered to look at it.”
Mr Brown says he can recall one researcher who did admit to having studied the question, and they were “honest enough” to admit that they couldn’t find any relationship.
More recent academic work has borne out both the veracity of Mr Brown’s original criticisms, as well as his contention that many of the problems he identified remain.
Tim Jenkinson, professor of finance and director of the Oxford Private Equity Institute at the Saïd Business School, has co-authored two papers in the past six years with bearing on this topic.
“The question in our mind was why institutional asset managers or institutional investors seem to be slower to move to the passive products than retail investors. We would have thought that they might have been quicker to do so,” he tells us. “And so we started looking at the sort of advice they were given.”
They found that, while investment consultants market themselves to institutional clients by claiming that their manager recommendations add significant value, this claim is not data verified.
Per the published paper: “On average, consultant-recommended investment products perform no better than other products available to institutional investors. Our results indicate that, over our 10-year sample period, the portfolio of all products recommended by investment consultants delivered average returns gross of management fees of 5.4 per cent per year (5.1 per cent after management fees).
These returns are, on average, 0.3 per cent a year lower than returns obtained by other products available to plan sponsors but not recommended by consultants.”
“The question in our mind was why institutional asset managers or institutional investors seem to be slower to move to the passive products than retail investors. We would have thought that they might have been quicker to do so”
TIM JENKINSON, SAÏD BUSINESS SCHOOL
How, then, do consultants justify making the outperformance claim? The answer recalls a joke among economists: When asked, ‘How are you?’, the correct reply is, ‘Compared to what?’
Mr Jenkinson says: “Many of the benchmarks against which the products are judged are of themselves not particularly demanding,” he explains.
Depending on how indexes are constructed, it can be impossible for an active manager not to beat the benchmark. But that says nothing about the consultant’s ability to select the best manager.
“So what we did... is we just compared the performance of recommended products against the ones they didn’t recommend,” says Mr Jenkinson. “At that point, the outperformance for most of these managers recommended by consultants disappeared.”
Mr Jenkinson calls one of the more unusual tactics used by one consultant “the opportunity to forget losers”. By only including managers that had been recommended by the company for all of the five-year period commonly used to state performance, it was effectively able to eliminate any recommendations that had prematurely turned sour.
“It’s bizarre,” he says, “and you’d only ever spot it if you really dug deep into the way they construct things.”
A third controversial tactic identified by Jenkinson et al was the inclusion of simulated and backfilled returns. Simulated returns are what an active manager or investment consultant tells a client they would have achieved had that product been used or available in the past, while backfilled returns occur when past performance is added into a database.
Backfilled returns in particular are known to create an upward bias, called variously ‘instant history bias’ or ‘backfill bias’, in which past positive results are included more often than past negative results. Backfilled returns comprised 42.1 per cent of the sample used in the research by Jenkinson et al.
“In all cases we found that investment consultants, when they were comparing their track record for choosing particular asset managers, included simulated and backfilled returns,” Mr Jenkinson says.
Taken together, these behaviours go some way towards explaining the discrepancy uncovered between what investment consultants claim they can achieve and what they achieve in reality.
“The difference between recommended and non-recommended products returns or excess returns over benchmarks... is in most cases negative but not statistically significant and not very different from zero. This suggests investment consultants have, on average, no real ability to pick investment products,” the paper concluded.
“The difference between claimed and actual [performance] ranges between 1.6 per cent and 2 per cent per year depending on the specification and whether we do the calculations before or after management fees. This pattern is not driven by a single consultant or a single asset class, but seems to be present across the board to varying degrees.”
These significant variances risk undermining trust, and needlessly distract from the other good work done by active managers and investment consultants.
Pensions Expert asked a variety of investment specialists and consultants who research, select and utilise active managers to explain the value they bring to pension fund clients.
Willis Towers Watson’s head of manager research, Chris Redmond, acknowledges that there are some areas in which passive is more suitable but nonetheless maintains that active is particularly effective in times of macroeconomic uncertainty.
“Typically what we’ll find is where dispersion is highest, where the market is uncertain about the future path, where the macroeconomic situation is tricky... that typically is a better environment for skilled active managers to differentiate themselves,” he says.
“And you’ll see a greater dispersion of outcomes from managers and obviously the goal is to find those who are capable of sustainably, consistently, ideally over the long term, delivering good performance.”
Suzanne Lubbe, manager researcher at Mercer, says: “I appreciate that active management can also disappoint, and the so-called passive options available to investors have opened up a world of possibility,” says Ms Lubbe. “I say so-called, because many of these options, such as factor indices, involve active decisions.”
Besides which, she adds, some asset classes, such as fixed income and alternatives, remain difficult to access via passive.
This has led some consultants to recommend active only where they see it as justified, or where there is no alternative.
“We have a preference to use passive where we think it’s appropriate,” Ben Gold, partner and head of investment at XPS, explains. “If you take equities, for example, then the vast majority of the research we do and the recommendations we make to clients are passive. And that reflects our view as a house that we think it’s very difficult for active managers to consistently add value, especially after the higher levels of fees they typically charge.”
“In other asset classes, it varies. And in some asset classes, you don’t have the choice,” Mr Gold adds. This is especially true of corporate bonds, he explains, where the goal is to filter out “the worst companies, that have a high likelihood of defaulting on their debt”.
“We think it’s very difficult for active managers to consistently add value, especially after the higher levels of fees they typically charge” BEN GOLD, XPS
Ms Lubbe describes the selection process of active managers “that give us confidence of outperformance over a full cycle”, focusing on characteristics “that have been proven to deliver alpha”.
“Claims of greatness need to be backed up with examples,” she says, emphasising that past performance is not a significant factor. Instead, face-to-face meetings “to test the manager’s responses” and “pick up on those nuances and those intangibles that can be the secret ingredient for success” are employed.
Kempen Capital Management’s head of investment strategy, Nikesh Patel, sounds a similar note.
The company favours “stable, strong, specialist teams with a clear culture” in their organisation. He says: “We prefer engaged long-term shareholders – fund managers with ‘high conviction’ holdings, a high active share and a high risk awareness that such an approach demands.
“To be a good manager you have to have the investment philosophy of value, momentum and stability,” he says.
LCP partner Dan Mikulskis tells us that measuring success “is one of the hardest decisions in the business, especially when you’re having to end your relationship with a manager that’s not doing well.”
Stability and expectations are key, he argues; more so than measuring managers against a benchmark, measuring them against an agreed-upon process is important. This is something with which other commentators agree – in fact, Mr Patel says outperformance can be deemed undesirable if a manager achieves it by deviating wildly from expected practice.
As for how much the research and selection process costs, figures were somewhat hard to come by. Many say that while they control costs closely, research can be expensive.
Meanwhile, Ms Lubbe says: “Let’s say that research is important to Mercer, as evidenced by the fact that the research team is more than 200-strong, across the globe.”
Mr Redmond estimates where manager, custodian and broker fees in bond allocations typically add up to as much as 100 basis points, the money charged by research teams represents a fee in the region of “one, two basis points, something of that sort”.
What was remarkable about the studies, in particular those in which Mr Jenkinson was involved, is that they were among the first of their kind. This is because they were only possible due to the authors having access to a unique dataset provided by the FCA.
Much of the information used by Jenkinson et al was not otherwise available to the public, external researchers or analysts, and seldom would it have been shared with the consultancies’ client base.
Last year, the CMA published ‘The Investment Consultancy and Fiduciary Management Market Investigation Order 2019’. Though designed to address competition issues involving investment consultancy services that also provide fiduciary management services, several of its provisions have broader applicability.
The CMA now demands that trustees set strategic objectives before receiving services from an investment consultant, which must in turn disclose information about their recommendations and the advice they give to customers.
“I think that, for whatever reason, the CMA kind of backed off and said [to the consultants] you can carry on doing what you want, though it’d be nice if we had some sort of agreed way of doing it”
ROGER BROWN, IC SELECT
However, Mr Brown feels it could and should have gone further, calling for standardised performance calculations and the introduction of benchmarks for advisory clients
“I think that, for whatever reason, they kind of backed off and said [to the consultants] you can carry on doing what you want, though it’d be nice if we had some sort of agreed way of doing it. It was really quite soft, and I don’t think it’ll make any difference at all,” he says.
He says that other financial services sectors make their data available for public scrutiny, but that investment consultants have escaped an obligation to be examined in the same way.
“Unfortunately, the CMA didn’t actually recommend that. Ultimately, they said that the data had to be made available in a way that made performance comparable… it’s not really in the raw form we would have liked.”
For charts and other articles please use the hyperlink below