“Pension actuaries are well positioned to play an important role in developing solutions to the lifetime income challenge. In this session we explore emerging solutions and strategies to optimize the effectiveness of retirement savings.”
As retired pension actuaries, we totally agree with the first sentence above. And since the basic actuarial principles used in the Actuarial Approach advocated in this website to develop a spending budget in retirement are almost identical to those used by pension actuaries to develop pension plan contributions, we are pleased to see more pension actuaries entering into this discussion.
In this post, we will discuss:
- how the basic actuarial principles we endorse in this website for spending budget development are similar to those used by pension actuaries for pension plan funding and
- our rationale for the few modifications to the basic pension funding model we suggest.
When we first started blogging in 2010, we shamelessly borrowed the same basic actuarial principles we had used in our careers as pension actuaries to help individuals (and later couples) develop reasonable spending budgets in retirement. These basic actuarial principles include:
- Making deterministic assumptions about the future
- Reflecting the time value of money
- Reflecting the concept of probabilities
- Reflecting mortality
- Use of actuarial present values
- Use of a generalized individual model that compares the present value of assets with the present value of liabilities
- Periodic gain/loss adjustment to reflect experience different from assumptions (annual valuations), and
- Conservatism
We also borrowed the concept of risk assessment from pension actuarial practice. Similar to the requirements of the recently released Actuarial Standard of Practice 51, Assessment and Disclosure of Risk Associated with Measuring Pension Obligations and Determining Pension Plan Contributions, we encourage individuals and couples to model deviations in assumed experience to quantify risks in retirement and assist in their financial planning.
It is important to note that pension actuaries do not generally use stochastic models to develop pension plan contributions. For this purpose, actuaries generally make deterministic assumptions about the future knowing that actual experience will emerge in the future and the actuarial valuation process will automatically adjust the plan contribution requirements for that emerging experience. The exact same principles apply for the Actuarial Approach. And while some financial advisors and academics reject the use of deterministic assumptions in favor of more “sophisticated” stochastic assumptions, we believe the actuarial process employing deterministic assumptions and automatic adjustments in future valuations is preferable for personal budget setting purposes for the current year (just as the same approach for determining pension contributions for the current year is preferable to using stochastic assumptions). We maintain that if pension actuaries believed stochastic modelling produced a more robust annual pension contribution, then actuarial practice would have long ago migrated to using stochastic assumptions and modelling for that purpose.
This is not to say that the use of stochastic assumptions and stochastic (or Monte Carlo) modeling is not also useful (for personal finance or pension plan funding). If reasonable assumptions are employed, stochastic modelling can certainly be helpful in looking at the potential benefits of currently implementing various investment or spending (contribution) strategies by providing probabilities of success associated with each strategy. However, we are saying that for purposes of current year budget determination (and current year pension contribution determination) it is generally not as effective. One of the primary reasons we believe this is because the automatic adjustment mechanism built into the Actuarial Approach is a powerful feature. All things being equal, if assumptions about the future are too optimistic, future spending budgets determined in subsequent valuations will be reduced. And if assumptions about the future are too pessimistic, future spending budgets determined in subsequent valuations will be increased. And, just like with pension funding, it is important to allow this automatic adjustment process to function properly by not over-smoothing the results from year to year.
Additionally, the use of deterministic assumptions enables the user to more easily calculate the present values of more complicated expense patterns required under the Actuarial Approach. So, for example, the user of the Actuarial Approach can relatively easily incorporate expected non-recurring expenses into the total spending budget calculation.
Minor Modifications to the Basic Pension Funding Model
Other than the obvious differences that the Actuarial Approach develops a spending budget for one or two individuals for the current year and the pension valuation develops a contribution requirement for a pension plan for the current year, the primary modification we have made to the basic actuarial principles used in pension funding is to recommend the use of relatively conservative investment return and mortality/longevity assumptions. The assumptions we recommend for determination of the Actuarial Budget Benchmark (ABB) are consistent with assumptions used by insurance companies to price inflation-adjusted life annuities (admittedly based on very limited data). This same pricing concept was employed by the Society of Actuaries as a reasonable way to think about how much real dollar lifetime income can be provided by a specific 401(k) balance in their article, “Big Question, When Should I Retire?”
While the use of such relatively low-risk assumptions is not necessarily in conflict with the basic actuarial principle of conservatism (as noted above), some pension plan actuaries, notably public pension plan actuaries, use assumptions that anticipate expected returns on risky assets to discount pension plan liabilities. And while our default spreadsheet assumptions may be overridden to permit users to do the same thing, we don’t recommend anticipating (or pre-recognizing) these higher expected returns until they are actually earned. As discussed above, the Actuarial Approach process automatically adjusts for experience more or less favorable than assumed in future valuations. So, if you use overly pessimistic assumptions today, your spending budgets will be automatically increased in the future as the more favorable experience emerges. We believe this is a more prudent strategy for individuals and couples rather than pre-recognizing the higher expected returns and developing higher current spending budgets.
We believe the use of our default recommended assumptions for retiree budget development is consistent with general philosophy of if one is going to err, one should err on the conservative side. Individuals are not governmental pension plan sponsors with the ability to raise taxes. Therefore, individuals tend to prefer relatively safe approaches to those involving the possibility of significant spending reductions. The 4% Rule is just one example of this search for safer decumulation strategies.
Conclusion
We recommend that individuals and couples use the same basic actuarial principles for their spending budgets as are used by pension actuaries for pension funding. We believe that a spending budget developed using the Actuarial Approach (and recommended assumptions) is generally more robust than spending budgets developed using either Strategic Withdrawal Plans (SWPs like the 4% Rule or the IRS RMD approach) or Stochastic (Monte Carlo) modeling. We fully understand why financial advisers may prefer Monte Carlo models, particularly those models that assume healthy risk premiums for equities and encourage their clients to invest in equities or other risky investments. We, however, believe that retirees should be skeptical of any spending budget that pre-recognizes expected future returns on risky investments.
We encourage other pension actuaries to play a role in developing solutions to the lifetime income challenge. We hope, however, that this role does not find them straying too far from the basic actuarial principles they learned as pension actuaries.
