Q: Can I raise my pledge-borrowing rate (to 4%, 6%) as my assets grow?

In 30 seconds: No. Pledge borrowing must hold to a withdrawal rate, not to “what fraction of total assets my accumulated borrowing is.” A 433 allocation supports borrowing about 2% a year, 70/30 about 3% — beyond 3% you must shift into high-dividend (cash is air). “Accumulated borrowing ≤ 20% of total assets” is a hard ceiling, not a withdrawal rate — the market can fall 80%, and even if you only borrowed 20%, once assets shrink your borrowing ratio blows past the line and you get margin-called into bankruptcy. To live better as assets grow, the right move is to rebalance: park twice your accumulated debt in the cash position to hedge it, redo 433 on the rest, then take 2% of the new total.¹

Variants (how people actually ask)

“I’m on 433, borrowing 2% a year via pledge — can I raise it a bit each year, say to 4% or 6%?”
“As long as accumulated borrowing stays under 20% of total assets, I can borrow however I like, right?”
“My assets have grown and I’d like a nicer life — can I borrow more?”
“20% is the safety line, so isn’t borrowing straight to 20% fine?”

Why “under 20%” is not a safe bet

The error is treating a ceiling as a withdrawal rate. James’s breakdown: you have NT$50M and borrow at 2% to spend; along the way the market falls and NT$50M shrinks to NT$15M — now the money you borrowed in the early years, as a fraction of remaining assets, “will of course exceed 20%.” You can’t assume assets never fall; they do. So the point was never “will I cross 20%” but holding a 2% withdrawal rate from the start, so borrowing accrues slowly enough that even a deep drawdown doesn’t push you over the margin-call line.¹

Extreme case: even if you only borrow to 20% and the market falls 80%, your NT$50M becomes NT$10M while you owe NT$10M — the broker liquidates your stock and you’re bankrupt. Whether you go bankrupt is a function of the allocation, not of “whether you can stomach it.” (Same principle as the ≤20% pledge line in the three lines of defense and the pledge threshold in borrow, never repay.)¹

The fix: rebalance as assets grow, don’t raise the rate

To raise your standard of living after assets grow, you rebalance rather than dial up the withdrawal percentage. James’s worked example:¹

Start at NT$50M, 433 allocation, borrowing 2% a year via pledge = NT$1M, and just live on that.
Say 5 years later assets have grown to NT$100M; over those 5 years you’ve accrued roughly NT$6M of borrowing.
Rebalance: first move twice the accrued debt — 6M × 2 = NT$12M — into the cash position (e.g. short-duration 00865B) to hedge that debt.
Redo 433 on the remaining 100M − 12M = NT$88M.
Now you can take 2% of the new NT$100M = NT$2M. Your living standard doubles from 1M to 2M — driven by assets growing plus rebalancing, not by forcing the rate from 2% up to 4%.

In other words: fixed withdrawal rate, re-allocate as assets grow — that’s the safe path to a better standard of living. To verify, run CLEC’s backtest tool (pCloud “newcomer handbook,” updated weekly).¹

More precisely: the withdrawal rate is computed on the starting assets, not re-derived from today’s total

00572 gives the conclusion (“you can’t raise the rate”); 00534, half a year earlier, states the sharper rule behind it — a member summarized it in one sentence and James confirmed it on the spot:²

The safe method: the withdrawal amount is fixed by your assets and rate at the start (e.g. NT$50M × 2% = NT$1M in the year you retire), and you draw that amount thereafter.
The dangerous method: re-deriving the rate from today’s total assets each year — you draw more as markets rise, and you have already raised your baseline by the time they fall. “That becomes rather dangerous.”
James’s reply: “If you want to change the autopilot, then you have to change the situation” — i.e. you may reset the baseline, but that is a deliberate, one-off rebalance (the process in the section above), not an automatic drift with market value.

This is the heart of the question in this page’s title: the answer isn’t “you can’t raise it,” it’s “you can’t raise it automatically — only by rebalancing, deliberately.”

The withdrawal rate sets your Beta, not the other way round

00538’s “flexible rebalancing 2.0” turns the withdrawal rate into a table, and it is the quantitative skeleton of this page:³

Withdrawal rate	Corresponding Beta
2%	β 1.0
2.2%	β 0.9
2.4%	β 0.8
2.8%	β 0.7

The order is rate first, Beta second: “it should be that your withdrawal rate determines your Beta.” (See Beta, 聰明再平衡法.)
3% is the red line: “Once the withdrawal rate goes over 3% it’s no good … you’d best use high dividend.” Above 3%, the answer isn’t more leverage — it’s a different tool.
This explains why this page says “433 supports ~2%, 70/30 ~3%” — different allocations support different rates, and once the rate rises, fewer allocations remain viable.

Answer log

2026-07-04 (00572): A member asked, “I have 20× annual expenses saved, on 433, borrowing 2% a year via pledge — can I raise it to 4%/6% as long as accumulated borrowing stays under 20%?” James called it “wildly wrong” and walked through the NT$50M→NT$100M rebalance (move twice the accrued debt to cash, redo 433 on the rest, then take 2%). The same Q&A is printed on that week’s deck.¹

⚠️ Withdrawal rates and pledge safety lines vary with your assets, cash flow, leverage, and local borrowing terms; this page summarizes James’s spoken principles, not personalized investment advice. Before acting, run the backtest tool and understand the four withdrawal tiers in cash is air and the margin-call risk in the three lines of defense.

2025-11-29 (00542): Gives the 2.5% divide — anyone needing to withdraw more than 2.5% of retirement assets a year should consider partially switching to high dividend (QQQI); under 2.5%, the QQQ / 00662 / QLD growth allocation is fine.⁴
2025-11-01 (00538): The withdrawal-rate ↔ Beta table (2% / 2.2% / 2.4% / 2.8% ↔ β1.0 / 0.9 / 0.8 / 0.7), and the rule that above 3% you should move to high dividend.³
2025-10-04 (00534): The sharpest statement of this question — the rate is computed on the starting assets; re-deriving it from today’s total each year becomes dangerous. To change it is to deliberately “change the autopilot.”²

🔎 This answer is assembled from four sessions: 00534 gives the method (compute on the starting assets), 00538 the table (the rate sets Beta), 00542 the divide (above 2.5%, switch to high dividend), and 00572 the conclusion and the rebalancing process (you can’t raise it automatically, you rebalance). Reading only 00572 suggests the answer is “no”; the full answer is “not automatically — only by rebalancing.”

CLEC James, 長篇 00572 “Investor’s Declaration of Financial Independence,” 2026-07-04, pledge withdrawal-rate Q&A @29:30–34:30; the matching deck Q&A page is in raw/docs/簡報資料/00572… (“You can use 433 and borrow 2% a year via pledge — yes! But you must not arbitrarily raise the pledge borrowing rate; you’ll go bankrupt!”). Transcript in raw/transcripts/長篇/00572…. Amounts are James’s spoken examples. ↩ ↩² ↩³ ↩⁴ ↩⁵ ↩⁶
CLEC James and a member, 長篇 00534 “Retirement allocation and withdrawal rate: staying alive matters more than returns”, 2025-10-04, rate computed on starting assets vs re-derived from today’s total @3:24:00–3:24:30 (“if you want to change the autopilot, then you have to change the situation”). Transcript in raw/transcripts/長篇/00534…. This passage is a member’s restatement confirmed by James, not a full statement by James himself. ↩ ↩²
CLEC James, 長篇 00538 “Flexible rebalancing 2.0”, 2025-11-01, withdrawal rate ↔ Beta @23:30–24:00, @29:00, “above 3% you’d best use high dividend” @33:30 (deck p.46). Transcript in raw/transcripts/長篇/00538…. The table reflects the model output at the time and moves with model and market assumptions. ↩ ↩²
CLEC James, 長篇 00542 “The 2.5% retirement withdrawal-rate key…”, 2025-11-29, the 2.5% divide @06:00–06:30. Transcript in raw/transcripts/長篇/00542…. ↩