How advisors can increase a client’s annual retirement income by optimizing their personal retirement strategy
Key point
A standard approach to retirement income planning is gathering data on a client, budgeting their expenses, and then an advisor piecemeal testing various levers in isolation to find a strategy with a reasonable probability of success (PoS). This is a time consuming process that is unlikely to yield the best plan for a client. Even if the advisor is testing a limited set of options such as three Social Security claim steps, five investment strategies, five annuity purchase options, two pension settlement options, and whether or not to consider a Roth conversion at a single targeted tax rate, that is still 300 possible strategy combinations.
The number of potential lever combinations for many households is in the tens or hundreds of thousands. Each of these should be tested over hundreds or thousands of possible future market scenarios and at various levels of income to find the maximum income a household can safely spend. This requires powerful software to evaluate the options and quickly return results so the advisor can focus on communicating with the client.
Example household
Edward (age 62) and Samantha Perch (age 60) are both planning to retire in 2022. They have accumulated $1.9MM split between taxable non-qualified accounts and retirement accounts (qualified tax-deferred and Roth). Samantha also has a $400K pension which can be taken as a lump sum at her retirement or converted into $2,150/month lifetime income.
The Perch family wants their base spending to be $120K/year (growing 1% slower than inflation) with each needing $5K/year in additional health care expenses (growing 2% faster than inflation). Following a typical strategy, Edward and Samantha can only safely spend $112K/year, 7% less than they want. A typical strategy, followed by most Americans, will keep the investment allocation constant after retirement, claim Social Security upon retirement, not purchase lifetime income or annuities, and take the pension as a lump sum. “Safely” means that the household has a 90% probability of success to the specified life expectancy.
Downsides of piecemeal approach
Today, most advisors spend a lot of time testing different combinations for the client but can never reach the level of analysis of a powerful simulation system. Evaluating levers in isolation is inefficient and ineffective. Advisors waste their time searching through possibilities instead of allowing software to find the optimal solution for them. It can also lead to advisors giving substantially different advice to similar clients based on the process of evaluation used since it is not standardized. This is an issue especially for organizations as the same client could get different answers depending on which advisor they see or even from the same advisor on different days.
By evaluating levers in isolation and finding the safe income while optimizing only one particular individual lever, the amount an advisor can improve over the base is limited. In the Perch’s case, the biggest improvement is found by adjusting Social Security (see Figure 1 below). This increases annual safe income by $17K. Although this is a good improvement, it is less substantial than the $35K increase possible with co-optimization across all the levers (see Figure 2 below). In the Figures below, TQD is tax-targeted qualified disbursals and CGM is dynamic capital gains management.
Figure 1
AIDA co-optimization of levers
As discussed in the post Optimized Personal Retirement Income Strategy, a unified income strategy co-optimizing across all available levers to build a guaranteed income base and implement dynamic, tax smart strategies will give the Perch family, and any client, the best plan for retirement. Powerful software can give these answers to an advisor instantly, in a client-friendly, digestible format.
The answers found will be consistent across advisors for a given set of inputs, not dependent on compartmentalized knowledge and approach. Scalable, best practice solutions across an organization’s platform is both legally defensible and in the best interest of clients.
Figure 2 below shows the cumulative increase in safe income by adding the optimal strategy for each lever to the previous optimized levers (from left to right). The far right bar shows a cumulative increase for the Perch family, when all levers are co-optimized, of $35K in additional annual safe income.
Figure 2
Optimal strategies can also be personalized per client. By using constrained optimization that includes a client’s particular preferences, the optimal plan can be solved for. This is important if a client has a particular need or preference on types of products or strategies, such as the client doesn’t want to consider an annuity purchase or doesn’t want to delay Social Security claiming until age 70. As advisors work with clients on their income sources and preferences, they often need to adjust the plan. If the advisor has to manually re-evaluate levers individually for each change, the amount of time spent continues to pile up. Using Income Discovery, the advisor simply updates the constraints and in a single click, receives a revised answer showing the new optimal strategy for the client.
1 Safe retirement income is the amount of income that has a 90% opportunity of being available into a retiree’s mid-90s. Safe retirement income consists of all sources of income (not just guaranteed sources): Social Security benefits, pension payments, lifetime income payments, withdrawals from investments and any other incoming cash flows, such as part-time work and rental income. Safe retirement income is increased by Social Security retirement benefit claim deferral, purchase of annuities, changing investment portfolio asset allocation and other strategies. The amount of increase and the optimal strategy varies depending on the individual circumstances, annuity payout rates and capital market return expectations. Download the white paper for limitations of the analysis, details of a hypothetical client case and assumptions.
« Back