The term “mass customization” was in vogue 25 or 30 years ago, when U.S. manufacturers wanted to learn how to satisfy every customer’s desires while achieving economies of scale and raising quality levels to six-Sigma. They called it “agile manufacturing.”
Given the recent advent of robo-advice and “fintech,” the chief of a European financial think tank believes that it’s time to mass-customize retirement income solutions. His name is Lionel Martellini, and he’s presenting on that topic at a conference on Retirement Investing to be held at Oxford University next September.
The 48-year-old Martellini directs the EDHEC Risk Institute, a research arm of the prestigious international business school, EDHEC. Recently, he published an editorial and a March 2015 research report that should interest anybody who cares about retirement income planning or is affected by the robo-advice wave or both.
The two documents, respectively, have long titles: “Mass Customization versus Mass Production: How an Industrial Revolution is About to Take Place in Money Management and Why It Involves a Shift from Investment Products to Investment Solutions,” and “Introducing a Comprehensive Investment Framework for Goals-Based Wealth Management.”
In a recent conversation, Martellini (right) told RIJ, “We need to move to solutions, not products, and the client’s problem should always serve as the starting point for the discussion. We can already create custom solutions for pension funds or high net worth individuals, so the real challenge is to do customization on a large scale.”
Broadly speaking, Martellini advocates a two-bucket solution made up of a low-risk income-producing (“replicating”) portfolio, perhaps of TIPs, and a low-cost risky (“dynamic”) portfolio, perhaps of smart-beta exchange-traded funds, linked by rebalancing or hedging strategies that respond to changing conditions.
“The financial engineering is the easy part,” he said. “The hard question is, how do you communicate with the client? We’re starting an initiative to put together an algorithm for the end client, doing it in a nice communication way. Ultimately it’s an educational challenge.”
Martellini’s ideas about retirement saving and investing will sound familiar to anyone acquainted with goal-based investing, the household balance sheet approach, or the floor-and-upside principle. They also resemble the thinking that has gone into institutional solutions like Financial Engines’ Income Plus managed account program, and to Robert Merton’s ideas, which drive Dimensional Fund Advisors’ target-date funds.
But he recognizes that nobody has figured out an optimal way for individual advisors to deliver scalable solutions. “The missing component is in the distribution,” he told RIJ. “How do we take the clients’ inputs and turn them into solutions? That’s the feedback process. There’s a lot of inefficiency in this area now. The investment advisor has no tools to help people understand tradeoffs, you need a goal-based investing reporting tool.” He’s looking to robo-advisors to fill part of the gap, adding, “We can only hope that fintech initiatives are providing a strong push in the sense of reducing cost of distribution.”
Looking for upside
Martellini approaches the income challenge with special urgency, because low interest rates, risk aversion, and pressure on social security programs is a growing problem for retirement savers and retirees. “The whole French population is buying French bonds. They’re not generating any upside. The benefits coming from pay-as-you-go will decrease,” he said.
There was a time when many people might have described the variable annuity with lifetime income benefit as the perfect way to scale the retirement income challenge, because it combined all the necessary elements—a guaranteed income plus exposure to upside—in a single tidy package. Today, there are some who might say the same about the fixed indexed annuity with a living benefit. But he thinks those solutions, while profitable for manufacturers, are too expensive for mass consumers.
Income for life can be generated less expensively and almost as safely without an annuity, he believes. “You can basically replicate the payout by using tradable fixed income securities. During accumulation, if you’re dynamically trading between a performance-seeking portfolio and a replicating portfolio, you can do it in such a way as to generate a minimum of replacement income,” he told RIJ. That sounds something like the constant-proportion portfolio insurance method in Prudential’s discontinued Highest Daily variable annuity riders.
“You can’t replicate the longevity component, but that’s not a big uncertainty. It’s okay not to worry too much about longevity risk if you take care of the other risks. Later, when people are in decumulation, they can buy income annuities if that makes sense for them. But you’re not buying annuities for them,” Martellini added. Annuities, he implies, makes the complex challenge of mass customization all the more challenging.
If scaling customized retirement planning solutions is the general goal, it has two dimensions, he points out. There’s a “cross-sectional” dimension that addresses the needs of different investors entering at the same time and a “time-series” dimension that addresses the needs of different investors entering at different points in time.”
Some might argue that target date funds can already address those dimensions and that they are effectively distributed through retirement plans. But he and his colleagues don’t believe that TDFs have adequately solved the retirement savings problem.
Case-studies
The best illustrations of Martellini’s approach can be found in the hypothetical case studies that are described in the March 2015 report, “Introducing a Comprehensive Investment Framework for Goals-Based Wealth Management,” which he co-wrote with an EDHEC Risk Institute colleague and two members of the Investment Analytics Group at Merrill Lynch Wealth Management.
The report describes three hypothetical clients: an executive and spouse with $4.5 million who are transitioning into retirement and want to maintain a wealth level of at least $3 million; a just-retired couple, both age 67, with $2.75 million, who have no specific bequest motive and want to prepare for long-term care contingencies; a 45-year-old with $940,000 in investments and a $250,000 mortgage on a $300,000 home.
The EDHEC team approached each case by dividing the client’s assets into a personal bucket (home and ready cash), a market bucket (liquid investments), and an aspirational bucket of long-term illiquid assets. They also listed each client’s goals, noted the time-horizon for the achievement of those goals, and calculated the probability of achieving those goals with existing assets.
The complexity of the solutions to these cases suggests that it would be hard to mass-produce them. The EDHEC team admits that it would be impractical for information systems to devise a custom solution for every client. They are not the first to observe that while it might be relatively easy to mass-customize client’s investment strategies using existing theories, it will be harder to scale solutions that enable clients, or identifiable types of clients, to achieve individualized goals.
To even come close to managing the risks that endanger those goals, the solutions will have to be highly dynamic. That will require new and faster fintech, which Martellini et al describe as “an information technology system that can effectively process and update the key inputs of the framework at each point in time for each investor.” Stay tuned.
© 2016 RIJ Publishing LLC. All rights reserved.
