Bill Sharpe, a Nobel laureate in economics, once said about asset decumulation in retirement - “It's…the hardest problem I've ever considered”. Needless to say, the problem only gets harder for most other people. So, when someone tackles this problem head on and provides a comprehensive guide on how to solve the retirement income puzzle, it demands the attention of anyone who spends any time on retirement income planning. In ‘Safety-First Retirement Planning: An Integrated Approach for a Worry-Free Retirement’, Dr. Wade Pfau provides an excellent outline of a broad swath of financial products available to retirees in the marketplace and lays out a “safety-first” approach to generate retirement income using those products. Retirement income planners have much to gain from this book.
Dr. Pfau talks about two distinct approaches to generate retirement income. First, a probability-based approach or investment approach which uses a mix of stocks and bonds to generate retirement income. The other approach, “safety-first”, includes insurance products like annuities, in addition to stocks and bonds, to provide added income protections. He argues that the latter approach is more “efficient” because it provides higher lifetime spending and greater legacy value.
The appeal of the “safety-first” strategy lies in its approach to funding retirement expenses, which can be broken down into two components – discretionary and non-discretionary. Dr. Pfau argues that non-discretionary expenses should be funded by income sources which are guaranteed for life and not subject to the ups and downs of the financial markets. If existing sources of guaranteed income such as Social Security benefits and defined benefit pensions (if any) cannot cover all the non-discretionary expenses, then retirees should annuitize a part of their assets to fund the rest of their non-discretionary expenses. On the other hand, discretionary expenses could be funded by riskier assets such as stocks and bonds and the returns that they provide. In times of turmoil in the markets, retirees should adjust their discretionary expenses to suit their portfolios.
“Safety-first” is a sound approach to retirement income planning, but it's not without its challenges. First, although annuity payouts are not subject to the fluctuations of the equity markets, they face another type of market risk, namely interest rate risk. Annuity payouts are directly related to nominal interest rates. If the nominal interest rate is low when the annuity is purchased, then the monthly payments will be low. So, the interest rate risk becomes a timing risk. Dr. Pfau acknowledges this risk and suggests possible solutions such as laddering income annuity purchases over time. However, we are living in an extremely low interest environment. The average nominal interest rate on a 10-year treasury bond was 6.03 in 2000. At the end of April 2020, it stood at 0.63. Such low interest rates cannot provide enough livable income. This is reflected in retirees’ behavior. A recent studyFootnote 1 found that a one-percentage point increase in the interest rate translates to a five-percentage point increase in the probability that a retiree chooses an annuity over other draw down options available in a retirement plan. Dr. Pfau notes that even in low interest environments, annuities have an advantage over bonds because of the mortality credits, but retirees need to decide if that's enough or should they seek higher returns through equities so that they don't have to downgrade their lifestyle.
The other challenge with the “safety-first” approach is the categorization of non-discretionary expenses. At any point in time, expenses can be classified into discretionary and non-discretionary, but these categories tend to evolve with time. Someone entering retirement at 65 might have a mortgage payment, a car payment, monthly prescription payments etc. which can all be classified as non-discretionary expenses. But what are the chances that the same payments will be there when she is 80 or 90? Non-discretionary expenses related to food and clothing tend to go down with age. Household size and the associated non-discretionary expenses can also change due to marriage, divorce or death during the retirement years. But the annuitization decision made at age 65 cannot be altered at a later age to adjust to the needs arising from these changed circumstances. As a result, a retiree might find herself over-annuitized or under-annuitized.
Setting aside these challenges, “safety-first” is an intuitive and practical approach to retirement income planning. Including annuities or insurance products more broadly to address the retirement income needs is very useful, particularly when low-frequency high-impact risks such as a prolonged long-term care stay, or an extremely long lifespan can make or break a retirement. But one of the limitations of Dr. Pfau's book is that it restricts the discussion of annuities to the traditional annuity marketplace. There might be other more “efficient” ways to purchase an income annuity.
Social Security is the main source of annuity income for most Americans. By some estimatesFootnote 2, Social Security benefits account for nearly 30%-60% of net worth of retirees. So, a large portion of retirement wealth is already annuitized for Americans and there is a valid question of how much additional annuity do retirees need. Setting aside that concern, if following the “safety-first” approach a prospective retiree finds that there is a gap between her expected non-discretionary monthly expenses and expected monthly annuity income including expected Social Security income, is buying additional annuity in the marketplace the best annuity solution? A recent studyFootnote 3 compared immediate annuities, deferred annuities and using retirement assets to delay Social Security claiming (which is equivalent to buying additional inflation-indexed annuity) and found that for a vast majority of retirees delaying Social Security provides higher lifetime utility.
Another piece missing in Dr. Pfau's “safety-first” approach is debt management. While funding of non-discretionary expenses remains in focus, a push for how to lower the non-discretionary expenses in the first place is absent. Whether retirees should pay-off their mortgages, car loans etc. with their savings instead of buying annuities to service such debt is an important question. Getting rid of such debt payments lowers the need for guaranteed income and might close the gap between non-discretionary expenses and existing annuity income. Mortgage pay-offs increase legacy value as well.
To sum up, in “safety-first” Dr. Pfau lays out the foundations of a sound and intuitive approach for retirement income planning and provides a possible solution for a notoriously hard problem. However, it might be a tall claim to call it a “worry-free” approach as the silver bullet of retirement income planning does not exist. But the foundations of the “safety-first” approach can be used in conjunction with additional elements to build a more robust retirement income planning process.
