Processing Power

With new rules on the horizon blessing participant advice arrangements, plan sponsors’ attitudes toward offering employees retirement advice has shifted. “Five years ago, employers weren’t willing to offer advice because they were afraid of lawsuits, but that’s changed,” says adviser Mike Hager, President of Hager Strategic, Inc., in Washington.  

As sponsors’ attitudes changed, so has the thinking as to the nature of that advice. Giving advice is now about more than just picking the right fund, says Rod S. Greenshields, Consulting Director-US Private Client Consulting, Russell Investments. Historically, he says, the focus primarily was on benchmarking and return—how a particular fund did against other funds. Now, the focus is on how to provide the participant with level income in retirement.  

Also, as volatility proved an inadequate measure of the risk of running out of money prior to the end of retirement, says Greenshields, the probability of success or failure in the investment strategy providing level income throughout retirement is the focus of risk analysis.  

Lastly, the magnitude of any possible failure to meet goals is considered as well. We now look at just how dire the consequences will be if the plan does fail—the worst case scenario, says Greenshields. There is a huge difference, he notes, between a $1 failure and a $1 million failure. Over the last four years, he says, people saw the consequences of worse-than-expected market performance and the impact it could have on high-risk portfolios. Unless someone has a lot of assets to cushion himself, the downside circumstances need to be considered, says Greenshields. 

As the thinking surrounding retirement plans evolved, the computer models used to provide that advice evolved as well. Most models work, says Greenshields, by making predictions of the future such as expected retirement date, life expectancy assumptions, and future tax rate assumptions. In the past, predictions were backward looking, looking at historical returns. A new approach, he says, is to come up with different forecasts based on mathematical models.  

To better simulate real-world conditions, computer models also have become more sophisticated, says Greenshields, allowing for more information to be inputted. At a basic level, any sufficiently robust planning tool includes assumptions about the future behavior of asset classes, he says. Regardless of how those asset class forecasts are created, they represent the same basic set of inputs. What is most different—and what is evolving for more sophisticated players actors in this space—are the techniques and data used to construct those forecasts, he adds. This includes using more complex statistical techniques to better address the uncertainty of markets. It also can include different inputs into the forecast creation process, but that is more subtle since it is a change in data used to create other data used by the tools. For example, he says, older models may have allowed for the inputting of predictions of equity returns, e.g., 10% annually, and little else. Newer models, he says, allow for sophisticated Monte Carlo simulations to reflect the uncertainty of the future.   

Advice products now also have the capability of going to Web sites of the providers and automatically logging into participants’ accounts and drawing information from those accounts into the analysis, says Hager. In the old days, he says, participants had to bring statements from these different accounts to a session and log in all the information manually. The end result, he says, was that very few did so. Now, he says, participants can log in and have all the information from other accounts logged in automatically. This makes sense, he says, because the only way to know if participants’ assets are appropriately balanced is to know what all their assets are. Additionally, he adds, participants need to know what their entire income stream will be in retirement, whether from a defined benefit plan, Social Security, or other non-retirement assets.  

Advice products now also can give participants measurements of their success in saving for retirement. Bank of America Merrill Lynch’s proprietary advice product provides participants with a financial wellness score, says Kevin Crain, the Head of Institutional Client Relations for Bank of America Merrill Lynch in Hopewell, New Jersey. Participants can get a score, with 0 being the worst and 10 the best, as to how financially well off they are for retirement.  

Sponsor expectations also have shaped products. ­Sponsors want an advice tool to have broad capabilities, says Hager, including integration from various sources. The option for participants to implement advice immediately with the push of a button is also a feature sponsors want, he says. 

The downside to these new sophisticated tools, says Greenshields, is that they are hard to explain to participants. Explaining average returns, he says, is easy. A Monte Carlo simulation is not. The good news is that, over time, as advisers get used to explaining these new models to clients, they will develop language to convey the concepts. Any time something is new and complicated, he says, it takes a while to develop the language.