Large and small, pension plan sponsors are adopting in-house tools, partnering with technology research institutes or relying on investment managers for technological support with asset allocation, risk management and other aspects of investment management.
The growing interest coincides with an increasingly complex investment, demographic and regulatory landscape. A 2020 Deloitte report on the importance of technological innovation in the pension investment space noted pension funds across the globe are facing ageing member populations and pressure to optimize their costs, take up new investment strategies to boost returns and integrate environmental, social and governance factors into their investment and asset management practices.
Read: How will technology change investment landscape going forward?
“Over the years, we’ve seen the market evolve and [plan sponsor] clients are becoming more sophisticated and asking more from partners,” says Frédéric Kibrité, vice-president and director on TD Global Investment Solutions’ asset liability and passive fixed income team. “The investment landscape and strategies being offered — and the pension regulatory environment — have grown more complex over the years. And there’s a need for more delegation — many corporate plans want to focus on their core business and rely on their strategic partners to help manage their pension assets.”
Finding the right risk level
Defined benefit plan sponsors have steadily increased their exposures to private markets and alternative assets as a way to find new sources of return in a previously ultra-low-rate environment. Tracey Grant, BlackRock Inc.’s head of institutional client business in Canada, says the firm expects that trend to continue, given the macroeconomic and market volatility. The trend also ties into a growing interest in adopting digital risk management tools. “As more institutions allocate towards private markets, robust analysis is increasingly essential to understand and manage total plan risk.”
For example, BlackRock’s Aladdin technology presents pension plan sponsors with their public and private holdings in one view, offering a clearer understanding of their risk exposure and how their portfolio performs against a benchmark. In the current market environment, it allows users to stress test their portfolios and track their liability profiles in the face of interest rate hikes, says Ultan Geraghty, the firm’s head of Aladdin client business in Canada.
At the same time, as plan sponsors increasingly turn to other investment strategies for new sources of return, they must also assess new investment characteristics and asset correlations — meaning plan sponsors could end up with different risk-return profiles than their targets, according to an online fact sheet from Fiera Capital Corp.
Read: A look at designing and implementing long-term pension investment mandates
Caroline Grandoit, the firm’s global head of total portfolio solutions, says it aims to address this complexity with a risk factor-based asset allocation tool that’s born out of a decade of academic science around volatility. “[Volatility] isn’t just a unique number, [but] exposure to forces in the market that are shared experiences for strategies.”
Fiera has identified 10 common economic forces — or risk factors — that it believes drive all investment strategies’ risks and return: developed market growth, emerging market growth, real rates, inflation, credit premium, slope premium, commodity, currency, real assets and liquidity. Using information from plan sponsors, including their risk tolerance, valuations, cash flows, upcoming liquidity events, any annuity purchase plans and whether the plan has indexed benefits, the tool can create a portfolio that better matches a plan’s risk tolerance and provide a clearer view of their exposures, says Grandoit.
“We can take a number we’re all quite commonly used to saying — the volatility of investment — and break it down to say, ‘OK, your volatility is 30 per cent coming from developed market exposure, 10 per cent from [currency], five per cent from illiquidity and so on’ . . . That gives insight to the client as to where they’re exposed and they can say [whether they’re] comfortable with the exposures. In an inflationary environment the past year, let’s say a plan sponsor has indexed exposure, we’re able to confirm through the risk factor model on the asset and liability side that the portfolio will be able to sustain that inflationary environment and keep the [plan sponsor] on target for their objective.”
The tool allows pension plan sponsors to stress test their portfolios against hypothetical market outcomes like interest rate increases or a financial crisis. It uses a regime-switching element, which projects portfolio performance across thousands of scenarios and market types, and varying correlations between asset classes. “Is it a high-volatility market or is it a normal market? Is it a turbulent economy or a regular, growing economy?,” says Grandoit. “From there, we’re able to say, as a starting position for most probabilistic outcomes, ‘You’re in a high volatility [market], so taking a more conservative approach would be appropriate for the short term.’”
Read: Pension plan sponsors advised to stress test as regulators shift focus to second wave
TDGIS employs stochastic modelling — a financial model that forecasts the probability of multiple out-comes under different conditions, which is used to account for unpredictability and randomness — to help plan sponsors find the right strategic asset mix based on their risk profile and future liabilities, says Kibrité.
While he says most plan sponsor clients hold more assets than just fixed income, the firm considers it particularly important to optimize those liability-hedging assets in the portfolio construction stage, by integrating an assessment of plans’ credit and interest rate sensitivities, assessing key asset-liability metrics, such as duration and yield, and referring to TD’s internal credit ratings for public and private credit, which are embedded in its tool.
Artificial intelligence, real insights
Technologies like AI and machine learning, aided by quality data, have the potential to enable company- and sector-specific insights that can generate alpha, says Sadiq Adatia, chief investment officer at BMO Global Asset Management.
The firm uses AI to read and filter through large quantities of information — such as reading through quarterly and annual reporting across a sector — to find emerging trends in which it might want to dig deeper, he adds.
Read: OMERS joining think tank to find uses for AI in the pension sector
The Ontario Municipal Employees’ Retirement System is employing AI — or what Sami Ahmed, the pension fund’s data, analytics, AI and technology leader, likes to call “augmented intelligence” — in that way. His team has been helping the pension fund’s investment teams pull signal from large and complex data sets. “We have been working with them to address the opportunity to make sense of it all.”
Cameron Schuler, chief commercialization officer and vice-president of industry innovation at the Vector Institute, says these algorithms can also be used to identify slow-moving macro shifts — such as, hypothetically, the U.S. becoming less of a consumer-based economy or a growth in multi-generation families — that have investing implications. However, he notes large-scale data sets are necessary to identify these kinds of trends — as well as more sector-specific ones — and investment managers need to ensure the data is high-quality and that they understand the baseline.
While many uses of AI tend to be more about “tweaks at the margins” and involve making quick moves that many DB plan sponsors may not have the ability to do, the technology can also help when it’s time to rebalance, says Adatia.
“If the pension plan has latitude in terms of the exact timing, you can utilize [that], based on what you’re hearing and seeing from your models, to say, ‘Maybe now is not the right time because 85 per cent of people may be rebalancing.’ And you might want to wait . . . and let the market move and then you can sell at that stage.”
Read: AIMCo enters partnership to build AI, machine learning applications for investment management
Chad Langager, CEO of Alpha-Layer, a past partnership between the Alberta Investment Management Corp. and Alta-ML Inc., says machine learning can also help plan sponsors better forecast the returns of illiquid asset classes. In early 2022, the company performed a use case for the AIMCo to improve its risk estimates for illiquid assets, which are valued infrequently and difficult to compare to publicly traded assets. AlphaLayer used machine learning techniques to represent monthly data for illiquid assets based on their adjusted quarterly data and removed the seasonal factors, such as year-end accounting measures.
On the operational side, AI can streamline repetitive or mundane tasks, says Langager, pointing to AlphaLayer’s past project for the AIMCo of automating the capital calls process for its real estate holdings — reducing what was normally a 20- or 25-minute process to one minute — or checking forms for “fat-finger errors” like an incorrect number in a term sheet.
Information on demand
Lewis Gascoigne, director of investment at Eckler Ltd., says the firm is seeing plenty of appetite among pension plan sponsors for up-to-date performance, funding status or target monitoring tools that will allow them to see the status of their investment portfolios on any given day and support them in their roles as fiduciaries.
Currently, “static and sterile” reporting is done on a quarterly basis and, even then, it can take weeks or months to pull together a report on a fund’s performance for the previous quarter, he notes. Real- or near real-time monitoring technologies give plan sponsors the opportunity to build in key triggers that allow them to make quick adjustments once those thresholds are met, such as taking some risk off the table when equities outperform bonds and bringing the plan up to a particular funded status.
Read: Sounding Board: How the APS Corp. uses analytics for more efficient pension services
“If you could have real-time, daily or weekly updates about your investments, you can move quickly if something changes in the markets . . . and protect members,” says Tom Lambert, Eckler’s associate director of investments. “With a lot of [plan sponsor] clients, they have to think about the long-term targets of their plans to meet the benefits of members.”
Gascoigne and Lambert, both from the U.K., say this real-time monitoring is much more common across the pond, which they attribute to the maturity of U.K. pensions, many of which are working towards annuity buy-ins or buyouts, as opposed to Canada’s younger market.
To use or not to use
Technological support may not be necessary for every type of pension plan.
Via Rail Canada Inc. hasn’t felt the need to implement technological solutions to support its work, says François Quinty, director of investment management. Instead, it has developed internal tools to track the fund’s performance and its key metrics, such as duration risk. The organization’s “very mature” $2.5-billion plan is heavily invested in investment-grade bonds for liability matching purposes. “Tech is helpful when you’re larger.”
• The growing interest in using tools and technology for investment management coincides with an increasingly complex investment, demographic and regulatory landscape for DB pension plan sponsors.
• Asset allocation tools that incorporate regime-change or stochastic modelling help plan sponsors match their portfolios to their risk tolerance and account for market volatility and unpredictable outcomes.
• Real-time performance modelling tools and more granular reporting can support pension plan sponsors and their boards or committees in their fiduciary duties.
As an example, he notes Via Rail has roughly 20 privately held partnerships compared to larger funds that may have hundreds to manage and need a team to oversee them. “We know what our exposures are and they’re easy to aggregate. When you’re small enough, it’s only a nice-to-have, in a sense. We’re pretty happy having more simplistic architecture. Our reality still allows for that.”
Read: Via Rail Canada restructuring fixed income allocations in response to challenging markets
However, he also noted Via Rail occasionally turns to a third-party firm for its performance measurement service and to provide statistics like the fund’s standard deviation and its relative returns against peer pension funds. It also uses consultants on an ad hoc basis for determining metrics like the fund’s surplus risk if it’s conducting a policy review.
Looking to the future
Adatia expects the adoption of technology in the investment management space will continue to grow and for applications to become more sophisticated.
“At some point, I believe you’ll get to a stage where the data will be telling you on a pretty regular basis [when your portfolio is] optimized [or] not optimized [or whether] there’s a trade required to bring it back,” he says, noting investment professionals will still have the final say based on the insights they receive.
“You’ll have to lay out your objectives pretty clearly and your rebalancing thresholds carefully, . . . but I think that’s where the future is going.”
Kelsey Rolfe is a Toronto-based freelance writer.
Download a PDF of the 2023 Top 40 Money Managers Report.