The alternative to the 4% Rule isn't a different number. It's a different mechanism.

By Stefano Starkel

Special to Financial Independence Hub

Ask for an alternative to the 4% Rule and most people offer you 3.5%, or 5%, or a dynamic band. Those answer the wrong question. The rule’s most important feature isn’t the number: it’s the assumption buried in its first sentence.

William Bengen’s 1994 study said this: Withdraw about 4% in year one, adjust that dollar amount for inflation, and across the worst U.S. sequences he tested no portfolio was exhausted before 33 years. He has revised it upward since: to roughly 4.7% in his 2025 book. The number was never the fragile part.

Notice how you get the cash. You sell. The 4% Rule is a selling rule, and most Retirement calculators inherit that frame: the portfolio is the only lever, drawing it down the only way to reach it. So “what’s the alternative” is really two questions: a smarter way to size withdrawals, or not drawing the pile down at all?

Why the number isn’t the weak point

The fragile part of a fixed-withdrawal plan is Sequence-of-Returns risk. I run a leveraged, income-oriented book myself, and I’ve watched how brutally the order matters.

Round numbers: A $1,900,000 portfolio. $76,000 withdrawn at the start of each year, held flat, then that year’s return applied. (The real rule inflates the withdrawal; flat isolates the sequence effect, and understates the damage.) Two retirees, same three returns, opposite order.

Same returns, same withdrawals, and B finishes about $80,000 behind: a gap that compounds with every subsequent return. The reason sits in year two: B took that $76,000 from a portfolio already down to $1.28M, a 6.0% bite against A’s 3.5%. Selling a fixed amount into weakness turns a paper dip into spent-and-gone principal. Over a full retirement, that early-sequence damage is what empties portfolios: not the headline rate.

Changing the mechanism, and what it costs

One family keeps selling, but flexibly: guardrails soften Sequence Risk at the cost of variable income. Still decumulation.

The other changes the mechanism: borrow against the portfolio rather than sell it, so the assets stay invested and nothing is forced to be realised in a downturn. This is what people mean when they say the wealthy never sell. It is not a free lunch : and for a Canadian reader, considerably less free than the American version.

The loan is callable. FINRA’s mandated margin disclosure is blunt: the firm can sell your securities without contacting you, you don’t choose which ones, and you aren’t entitled to extra time. Canada is no gentler: TD Direct Investing tells clients it may sell holdings “potentially without any prior notice,” and decide which. That is the bottom-of-the-market sale the strategy exists to avoid, arriving on the broker’s schedule.

It accrues interest, at a floating rate. If your return doesn’t beat the borrow rate you destroy capital faster than selling would: and you’ll be wrong about that exactly when the collateral is falling.

Interest on money borrowed to live on isn’t deductible here. Deductibility turns on the purpose of the borrowing: earning income from a business or property. Groceries don’t qualify.

There is no step-up at death. “Buy, borrow, die” works in the U.S. because heirs inherit a stepped-up cost base. Canada deems you to have disposed of everything just before death, and the deferred gain lands on the terminal return. The endgame that makes it coherent south of the border isn’t there.

It reaches only non-registered money. Pledging RRSP assets as security for a loan includes their full fair market value in your income that year. For most Canadian FIRE portfolios, that rules out the majority of the balance.

Drawdown risks running out. Borrowing risks a call and carries a running cost. Neither is safe. I’m not recommending the second : I’d argue it’s the harder to operate.

What survives both models

Whichever you pick, the outcome turns on whether you measure the variable that kills it, and act on a written rule instead of a mood.

Selling: your number is annual draw divided by portfolio value. Not 4%: yours. Recalculate after a drawdown, when it rises silently and nobody writes to tell you.

Borrowing: use your broker’s own margin-equity figures, not loan divided by total portfolio. Non-marginable holdings count as zero collateral, so a homemade ratio reads safer than theirs : and theirs is the one that triggers. Then stress it: mark the collateral down 35% and recompute. If that breaches, your rule is already too loose.

A plan you only examine when you’re frightened isn’t a plan. It’s a mood with a spreadsheet attached.

Stefano Starkel founded and exited an eCommerce brand and a SaaS company, and now runs a leveraged, multi-strategy portfolio as a one-man family office. He writes at incomestead.com, the governance tool he built to hold his own book to its rules on margin, allocation and income. He writes education, not financial advice, and is not a registered adviser. This blog was adapted for Findependence Hub and based on a longer version that ran on his site on July 21, 2026.