Decumulation is the
hottest topic in the retirement industry today. Service providers are all looking for ways to distribute a reliable and predictable stream of income to defined-contribution-plan participants during their retirement in an effort to closely approximate the defined-benefit experience.
The beauty of a DB plan is its simplicity. The employer holds on to the investment and funding risks, while the promise for a stream of income to retirees is either professionally managed or guaranteed by an insurance carrier. Workers who are three to five years from retirement have a good approximation of their monthly income in retirement (based on age, years of service and average final employment income). Decisions regarding portfolio construction, manager selection, longevity risk and benefit-payer solvency are not necessary.
For married workers, the single decision is the percentage of monthly income the surviving spouse receives if the worker dies first. Workers rarely express concern about losing their retirement assets if either or both spouses die "too early" or question how the portfolio should be managed to assure income certainty.
The DC-plan market needs a simple, competitive, open-architecture solution that is portable, too.
In the past 20 years, Nobel laureates Daniel Kahneman, Robert Shiller and Richard H. Thaler, among other behavioral scientists, have made significant contributions to our understanding of how human physiology affects decision-making in finance and have proven that we do not always make decisions in our own best interests.
Our industry and legislators are focusing too much on investment strategies (typically backward-looking) for decumulation and very little on cognitive psychology and behavioral finance.
Studies and research have proven that heuristics — employing rules of thumb to make decisions — often lead to cognitive biases that result in poor or unintended outcomes. Since the Pension Protection Act of 2006, our industry and plan sponsors have embraced automatic enrollment and auto-escalation as the most assured way to compel retirement savings while minimizing behavioral biases. Industry insiders have long recognized that average workers are poor savers and even worse investors.
Dozens of Biases
Dozens of known biases — such as confirmation, confidence, framing, hindsight and recency biases — contribute to poor decision-making. Add to this short list loss aversion and path dependency, and the odds are overwhelmingly stacked against retirees if they are expected to make the right decisions during decumulation.
The individualization of investment and longevity risks (as opposed to the mutualization of the same risks in the case of a DB plan) presents significant challenges to responding correctly to the randomness of future events during the distribution phase of life. Every deviation from the expected (or assumptions made) can be detrimental to meeting stated objectives, such as retiring earlier than expected, living too long, a shortfall in investment returns, volatility of returns and unplanned cash needs, just to name a few.
Their retirement date is likely the most financially vulnerable date in workers' lives. They have to exchange an active paycheck for a passive one. They will gain access to the largest sum of cash they've ever received in their lives. Making the distribution decision is daunting and likely fraught with fear, apprehension and ignorance. Our industry should appreciate the power of cognitive biases and the paradox of choice at that moment when creating products and solutions.
To help guide retirees, advisers and fiduciaries should ask non-self-serving questions and offer outcome-oriented choices.
I submit that most average workers at retirement are looking for:
• An ongoing paycheck that they can count on, rain or shine.
• A 70% to 80% income replacement that will not outlive their assets.
However, the mutually exclusive mindset between asset managers and insurance companies — the main providers of drawdown products — are shortchanging a whole segment of viable solutions.
Asset managers are looking to build investment solutions to keep control of the retirees' assets by relying on their alpha-generating acumen to
deliver a "highly certain" stream of income. This can be in the form of a stand-alone managed-income product or built into the in-retirement vintage of a target-date fund.
Conversely, insurance companies are looking to capture retiree assets through
annuity products that distribute guaranteed and predictable streams of income for life, but don't offer cash-out features.
Looking to insurance companies instead to provide an income guarantee overlay on top of a diversified, professionally managed portfolio, such as a target-date fund or managed account, would be an ideal solution for decumulation. This would combine individualized flexibility and control with the benefit of risk mutualization for certainty and predictability.
Retirement income selection should be narrowed to a small, defined range of monthly income choices for the retiree and the surviving spouse. For example, the plan could provide the pre-retiree the following choices: (1) $1,000 per month for life for a single life assuming all account assets are used, or (2) $500 per month for life for a single life assuming half of the account assets are used. Or, the choice could be to ask the retiree the amount of assets he or she would like to manage personally and leave the remaining amount to provide a guaranteed income stream for life.
Philip Chao is principal and chief investment officer at Chao & Co.
