Rethinking ‘Safe’ Withdrawal RatesRethinking‘Safe’WithdrawalRates

Withdrawal Rates

There is over 30 years of research exploring how much a retiree can withdraw annually from a portfolio upon retirement, with estimates generally ranging from 2% to 8% and the industry largely coalescing around 4% (i.e., the “4% Rule”). However, this rule and many modeling tools today use assumptions that do not accurately capture retiree preferences and decisions. There are three common gaps in these models: 

1. Ignoring other income streams: Many Americans receive some type of guaranteed lifetime pension benefit, such as Social Security, which provides a minimum standard of living. This means a retiree’s portfolio is generating income in addition to these guaranteed sources, thereby providing a safety net that might allow for a different portfolio withdrawal rate. 
2. Lack of spending flexibility: Traditional models commonly don’t include the desire or ability to adjust spending during retirement, since withdrawals are assumed to change only by the rate of inflation. Retirees have an ability to adjust spending based on real-life needs and circumstances, which can significantly affect spending rates. 
3. Inadequate evaluation of outcomes: Most financial planning tools today determine safe withdrawal rates by focusing on whether the goal is accomplished in its entirety and ignore the magnitude of failure using a metric commonly referred to as the “probability of success.” A better approach is to consider the total amount of the goal accomplished each year, and, if there is a shortfall, to try to better gauge the potential implications of the shortfall on a retiree. 

How Guided Spending Rates Work

PGIM DC Solutions’ guided spending rates address the above shortcomings, offering recommendations that not only tend to be higher than the traditional 4% rule, but also incorporate a more realistic depiction of retiree decision making. Our model breaks down the retirement goal by perceived flexibility, incorporates adaptive (or dynamic) spending through retirement, and relies on a more realistic outcomes metric. 

We estimate the safe spending level for a variety of scenarios to capture how differences in retiree situations can result in different guidance around spending levels, with a particular focus on spending flexibility. We assume three generic flexibility levels: conservative, moderate, and enhanced, which correspond to essential spending levels of 100%, 70%, and 40%, respectively, of the overall spending target for the specific portfolio. We also consider retirement periods from 10 to 40 years in five-year increments. 

Returns are based on PGIM Quantitative Solutions’ Capital Market Assumptions (CMAs), leveraging both the 10-year assumptions (for the first 10 years of the projection) and the steady state assumptions (for years 11 until the end of the scenario). We assume varying equity allocations that correspond to the target level of essential spending, where the fixed income portion is invested in 20% cash and 80% bonds and the equity portion is invested in 70% US large cap equities, 10% US small cap equities, and 20% international equities. 

We assume higher equity allocations for higher levels of spending flexibility, where the equity allocations for the conservative, moderate, and enhanced spending levels are 30%, 50%, and 70%, respectively. Returns for the asset classes are reduced by an assumed 40 basis point investment management fee. (See all of the assumptions here.)

These guided spending rates vary materially by retirement period and by perceived spending flexibility level. Retirees who have more flexibility around spending (i.e., enhanced) have spending rates that are approximately 25% higher than those who are less flexible (i.e., conservative), on average. These estimates are notably higher than other estimates around “safe” withdrawal rates, such as the traditional 4% rule. For example, if we focus on the 30-year period, the guided spending rate would be 5.0% for a retiree with a moderate level of spending flexibility. 

It is important to note that the guided spending rates change over time. We demonstrate this by leveraging PGIM Quantitative Solutions’ historical CMAs created since Q4 2009 (i.e., the forward-looking estimates at that point in time). Historical inflation assumptions are based on the 30-year forecast from the Cleveland Federal Reserve. The results are included in the exhibit below.

There is notable variation in the guided spending rates historically. For example, when interest rates were low in 2021, the corresponding spending rates declined. This suggests retirees should regularly revisit portfolio withdrawal rates as market situations evolve over time. 

Overall, the optimal spending rate is going to vary by retiree and should be determined based on their own unique situation and preferences.

By using our guided spending rates, retirees may find they can safely increase their withdrawal rates, potentially resulting in a more enjoyable retirement for many Americans today.