Personal post-retirement risk and the future of Social Security

The following are comments by Anna Rappaport from the Academy’s forum on the Future of Social Security, held on September 5, 2018, in Chicago, Illinois. Anna served as a discussant for a presentation by Henry Aaron on his proposed set of reforms to assure the long-term financial balance of Social Security.

I agree with Henry Aaron about the changes in the environment that lead to an increase in personal risk over time. And the level of personal post-retirement risk is increasing.

I want to share research that reinforces the importance of Social Security. The Society of Actuaries (SOA) Research on Post-Retirement Risk tells us about what people know about post-retirement personal risk. That in turn tells us about the importance and role of Social Security.

The SOA risk research includes a series of 9 bi-annual public surveys and several sets of focus groups and in-depth interviews, as well as an analysis of the resources of retirement age middle Americans. The SOA learned:

• That many middle-income retirees have little in the way of financial assets and most lower income retirees have almost no financial assets. For many of these people, Social Security is either the only source of regular income or the vast majority of regular income. Except for those with defined benefit (DB) plan benefits, this group gets very little additional regular income, and those benefits are declining.

• SOA focus groups and the surveys show that for many middle-income and middle asset-level retirees, planning is often short-term cash flow focused.  These individuals tend to look at their expected cash flows over the next 2 to 5 years, and whether their income will pay their regular expected expenses. This type of planning leaves a number of gaps – it fails to consider longer-term risks and unexpected expenses; it fails to focus on the longer-term. For people focused on this paradigm, major risks such as long-term care and health care are not explicitly included in such planning.

• The same groups who are using cash flow focused retirement planning tell the Society of Actuaries that they wish to hold onto their assets – which are often modest. Rather than having a plan to spend down assets, they wish to hold onto assets to the extent possible. The assets are viewed as a combination of an emergency fund, possible funding for long-term care, and amounts that can be left as a bequest.

• The SOA has seen a consistent pattern of responses in several sets of focus groups and interviews. These studies included in 2013 focus groups with retirees retired less than ten years, in 2015 focus groups with individuals retired 15 years or more, and in 2017 in-depth interviews with individuals who are 85 and over.

• I observe that while most of the planning models in use today call for people to maintain their pre-retirement living standards, retirees are thinking about things differently. Many of them are willing to cut expenses and are cutting expenses substantially. Many of them seem satisfied after having done so. They are very frugal. However, Social Security is critical to them to allow them to manage at the level that they do. I believe that for very many lower and middle income Americans, cutting Social Security benefit levels would likely lead to the difference between being frugal and managing satisfactorily vs. a life that includes a lot difficulty. Social Security has allowed them to manage with substantial reductions in spending, but further cuts would not be desirable.

• It is clear from all of the surveys and the focus groups that there are a number of individuals at retirement age with gaps in knowledge about retirement planning, and relatively few of them are real long-term planners considering a wide range of issues. Other sources document challenges with financial literacy. Social Security is extremely valuable because it does not require individual decisions (except for claiming) and participation is automatic.

I have observed, as Henry has, the shift away from defined benefit (DB) plans and to defined contribution (DC) plans. Unfortunately, many Americans have neither type of plan, and do not have any employer supported retirement savings. While Individual Retirement Accounts are universally available, relatively few Americans take the initiative to use them on their own without employer support.

Within DC plans, there are usually a variety of payout options available, but most of them do not strongly encourage lifetime income payouts. The Society of Actuaries, working jointly with the Stanford Center on Longevity, has sponsored several research projects looking at lifetime income in DC plans. Actuaries generally believe it is important to encourage such use of DC funds.

But for now, the reality is that relatively few DC participants are using their account balances to secure lifetime income. All of these reasons also serve to make Social Security important.

It is critical that the groups who work on Social Security reform look at the economic situation of individuals in retirement by economic status – and how important Social Security is to different groups. What I have told many colleagues over the years is that we should recognize that for a portion of the population, Social Security and public benefits are what they will have in retirement.

I would like to mention two more points:

Henry mentioned increased longevity and also the increasing disparity in longevity. In spite of this, we have a system with ever-growing periods of benefit payment. I personally believe that one of the reasons for the decline in DB plans is that we failed to recognize how important is was to address this ever-growing period of benefit payments. This also applies to Social Security and it should be considered moving forward.

Henry mentioned the changes in the family that have taken place since the design of current spousal benefits. My personal view – held for many years – has been that there is significant inequity between single and dual earner families. There also remain significant differences in retirement security by gender, with women being less secure since they are often the longest lived, but they have had lower lifetime wages, are very likely to be alone in their final years, and they are more likely to need long-term care while they are less likely to have a family caregiver. I applaud discussion of spousal and caregiving issues, and hope that we can realistically focus on the range of family situations and what will best fit a wide variety of needs.