Bill Bengen did retirees and would-be retirees a favor way back in 1994.

He gave them an answer to the question every retiree eventually asks: How much can I safely withdraw from my portfolio without running out of money? The answer: 4% in year one, then that same dollar amount adjusted upward for inflation every year for 30 years.

Simple, memorable, and for three decades, good enough.

To be sure, the 4% rule, which Bengen recently revised to 4.7% with additional asset classes, isn’t wrong so much as it’s incomplete, say researchers.

The rule, say experts, was designed as a worst-case survival rate (a SAFEMAX), not an optimal spending rate – and most retirees aren’t living through the worst case.

According to David Blanchett, director of retirement research at PGIM, there are three key flaws in conventional withdrawal rate models: they ignore other income sources like Social Security; they assume zero spending flexibility; and they use a pass/fail “probability of success” metric that treats a minor late-retirement shortfall the same as a catastrophic early one.

So, how much can you safely withdraw?

Well, both Blanchett and Wade Pfau, author of “How Much Can I Spend in Retirement?”, agree that the answer depends on two foundational variables: how long your retirement lasts and how your portfolio is allocated between stocks and bonds. And Blanchett goes further by adding a third variable – spending flexibility – and incorporating guaranteed income sources like Social Security into the equation.

Blanchett calls his framework “guided spending rates,” and the starting numbers are notably higher than the 4% rule. For instance, for a standard 30-year retirement, Blanchett’s rates are: conservative 4.38%; moderate 4.95%; and enhanced 5.58% – all of which are higher than the traditional 4%.

David Blanchett, PGIM

By way of background, Blanchett defined the three flexibility levels as follows:

  • Conservative: The retiree depends heavily on their portfolio to cover essential spending (housing, food, healthcare). All spending is considered non-negotiable, leaving no room to cut back. Lowest withdrawal rate, lowest equity allocation (30%).
  • Moderate: A blend. About 70% of spending is essential and 30% is flexible. The retiree has some ability to adjust if markets turn unfavorable. Medium equity allocation (50%).
  • Enhanced: The retiree has significant flexibility, with only 40% of spending considered essential and 60% discretionary. They can meaningfully reduce spending if needed without seriously impacting their lifestyle. Highest withdrawal rate, highest equity allocation (70%).

According to Blanchett, the proposed higher “guided spending rates” better reflect how retirees actually behave: they adjust spending when needed rather than robotically withdrawing a fixed, inflation-adjusted amount every year.

Consider, for instance, that in a PGIM survey of 1,500 near-retirees, only 15% said a 20% spending reduction would be devastating or require substantial sacrifices – suggesting most retirees are more financially adaptable than traditional models assume.

In his research, Pfau answered the question: What is the maximum withdrawal rate that historically never depleted a portfolio over a given period, across different stock/bond allocations?  

In Pfau’s research, for instance, a 60-year-old person with a 50% stocks/50% bonds portfolio who planned to live 40 years to age 100 could safely withdraw 3.72% in year one of retirement. Or a 70-year-old with a 50/50 portfolio who planned to live 20 years to age 90 could safely withdraw 4.98%.

How Much Do I Spend on Retirement? By Wade Pfau

Despite their methodological differences, Pfau’s and Blanchett’s numbers land surprisingly close for a 30-year retirement.

  • Pfau’s SAFEMAX at 75% stocks: 3.99%
  • Pfau’s SAFEMAX at 50% stocks: 4.03%
  • Blanchett Conservative (30% equity): 4.38%
  • Blanchett Moderate (50% equity): 4.95%

Pfau’s numbers are a worst-case historical floor – they had to survive the Great Depression, the 1970s stagflation, and every other catastrophe. Blanchett’s conservative rate is slightly higher because it factors in Social Security as a separate income cushion, meaning the portfolio doesn’t have to do all the heavy lifting alone.

Also worth reading

·   Federal student loan changes could raise payments for millions

·   White House wants federal retirement accounts for those with no 401(k)

·   5 retirement risks most Americans overlook

Several factors explain why Blanchett’s rates run higher – and understanding them helps retirees decide which framework fits their situation.

Dynamic vs. static – the most important distinction

This is the biggest reason Blanchett’s rates are higher than Pfau’s SAFEMAXs. Pfau’s numbers are static – the retiree takes a certain percent from their portfolio in year one and increases it by inflation every year regardless of what the market does.

Blanchett’s rates, by contrast, assume the retiree comes back to the table periodically and adjusts. That built-in flexibility is worth real percentage points in withdrawal capacity. You’re not locked in; you’re recalibrating.

The penalty mechanism

Blanchett bakes in a cost for cutting back, and crucially, that penalty is proportional to how essential the spending is.

A conservative retiree who cuts food and healthcare spending suffers a bigger modeled penalty than an enhanced retiree trimming a vacation budget. This is why conservative rates are lower even though you might expect a more cautious retiree to spend less anyway — the model is actually protecting them from the pain of being forced to cut essential expenses, not just the portfolio math.

The Social Security factor

Pfau’s SAFEMAX assumes the portfolio funds everything. Blanchett’s framework explicitly assumes Social Security (or pension income) is already covering a portion of spending, so the portfolio only needs to fill the gap. That alone justifies higher portfolio withdrawal rates across the board.

The goal completion vs. success rate shift

Pfau’s historical approach is inherently binary – the portfolio either lasted or it didn’t. Blanchett’s outcomes metric asks what percentage of the goal was achieved. A retiree who hits 96% of their spending goal over 30 years isn’t a failure, but a success rate model calls it one. This reframing allows for higher starting rates because small late-retirement shortfalls don’t torpedo the entire analysis.

Bottom line

Pfau gives you the floor – the rate that survived history’s worst. Blanchett gives you a more personalized ceiling – what a retiree can reasonably spend when Social Security is in the picture and they’re willing to adjust periodically.

The bottom line: treat your withdrawal rate as a living number rather than a fixed rule, and you are likely to have more spending capacity in retirement than the 4% rule suggests.

Related: AARP sounds alarm for American workers on 401(k)s, IRAs