Margin Accounts and the Post-PDT World

Margin is a loan with a daily exam

A margin account lets you buy securities with money the broker lends you, secured by the securities themselves. The federal initial requirement for marginable stock is **50%** — **Reg T** — which is where 2:1 buying power comes from: $25,000 of your own money supports a $50,000 position. What you deposit is **equity**; what you borrow is the **debit balance**, and it accrues interest every day at the broker’s rate. From then on the account is marked to market continuously against a **maintenance requirement**. The regulatory floor for long stock is **25% of the position’s value** — this is where the arithmetic in the prediction comes from — but brokers routinely require more, and they require much more on a concentrated, thinly traded or volatile position. The requirement is not a warning system; it is the level below which the broker is contractually entitled to act. The trap is that a maintenance requirement is a *percentage*, so it rises in dollars as the position rises, and a leveraged position has less equity to absorb the same move. Two identical $50,000 positions, one of them half borrowed, fall by exactly the same percentage — but only one of them gets a phone call. The same 1,000 shares at $50, seen three ways — At entry: position $50,000, loan $25,000: Equity $25,000 · requirement $12,500 · At $40: position $40,000, loan $25,000: Equity $15,000 · requirement $10,000 — still fine · At $33.33: position $33,330, loan $25,000: Equity $8,333 · requirement $8,333 — the call ← · At $30: position $30,000, loan $25,000: Equity $5,000 · requirement $7,500 — a $2,500 deposit ← Interest runs on the debit balance whether the position is winning or losing. A leveraged position that goes nowhere is a losing position, which is why "the market only moved sideways" is never an answer to why an account bled.

The call, and what happens if you do not answer it

A **margin call** is a demand to bring equity back above the requirement, and there are exactly two ways to answer it: **deposit** cash or marginable securities, or **reduce the position** by selling something. Doing nothing is not one of the options. The call has a deadline — often until the close, sometimes into the next day, depending on the broker — and the margin agreement you signed lets the broker **liquidate positions without prior notice** if the deadline passes or if the requirement is breached sharply. That is why a call is best understood as a courtesy notice, not a negotiation. Brokers generally prefer to sell what you designate, because they are not in the business of choosing your trades; but the choice only exists while you answer. And the loan is **recourse**. If a liquidation in a fast market realises less than the loan, the deficit is a debt you owe the broker. A gap through the maintenance level can therefore leave an account with nothing in it and a balance owing — the outcome people mean when they say you can lose more than you deposited. SIPC, which covers assets that go missing when a firm fails, does not cover any of this. Liquidation happens at the worst time by construction: it is triggered by the price moving against the position, and it sells into the same move. The broker’s obligation is to the requirement, not to your thesis.

The post-PDT world

For decades the constraint on leveraged day trading was a designation, not a number. A customer who made **four or more day trades in five business days**, in a margin account, and whose day trades were more than a small share of the account’s activity, could be designated a **pattern day trader** and had to maintain **$25,000** of equity. Below it, the account could not day trade at all. The rule was widely criticised for measuring the trader instead of the risk — a count of round trips says nothing about whether the positions were $500 or $500,000 — and it was repealed. Effective **June 4, 2026**, the pattern-day-trader designation and the $25,000 floor were replaced by an **intraday margin standard**: the broker applies its intraday maintenance requirement to the **peak intraday exposure** the customer actually carries and compares that with account equity, monitoring it during the day. Brokers may phase the change in, and the transition runs until **October 20, 2027**. What that changes is the *unit of the test*. Before, the question was how often you traded. Now it is how much exposure you carry at the worst moment of the day — which is the question risk has always asked. What it does not change is the arithmetic two sections up: leverage magnifies the same price move, a maintenance requirement is a percentage, and a small account holding a large intraday position is exactly as fragile as it was. Some brokers keep stricter house requirements, and a retail account trading on a phone is still the fastest documented route from a balance to zero. No broker will extend an intraday requirement to a retirement account: IRAs cannot be margined, so the same trade in a retirement account simply cannot be made — which is a feature, not an oversight.

Two regimes, and the price of the loan

Everything above described **strategy-based margin**, the regime almost every retail account sits in. It works position by position: the federal rule governs what you may borrow to open, FINRA’s rule governs what you must keep, and both are flat percentages applied to each holding on its own. It is simple, transparent and blunt — a hedged pair of positions is charged as though the hedge were not there, because the requirement is computed from the legs, not from what the legs do together. **Portfolio margin** is the other regime, and it is a different kind of rule entirely. Instead of flat percentages per position, a risk model simulates the whole account against a grid of worst-case moves in prices and volatility and charges the largest modelled loss as the requirement. The regulatory floor is lower — around 15% of equity — because the model replaces the crude percentages with something closer to the actual risk, and brokers often let accounts above roughly $100,000–$125,000 elect it. The gap between the two is widest exactly where a hedged position lives. Under strategy-based margin, a covered call is a long position charged at the maintenance rate plus a short option charged on its own; under portfolio margin the model recognises that the short call offsets the stock and charges a fraction of that. The same is true of a collar, a long straddle, or an index hedge against a concentrated book — expensive leg by leg, cheap as a book. The model is not being generous. It is noticing that the two positions cannot both lose the whole amount at once. Two caveats stop that from being free money. It is a privilege the broker grants and can withdraw, usually gated behind options-spread approval and a higher account minimum. And a risk model is only as sound as its scenarios: the 15% floor exists precisely because a model can be confidently wrong in a way a flat percentage cannot, and a pair of positions the model calls hedged can come apart when the correlation it assumed breaks at the worst possible moment. Portfolio margin rewards genuinely diversified books and punishes ones that only look diversified in the model. Then there is the loan itself, because leverage has a price and the price turns a multiplier into a hurdle. Brokers quote a rate tied to a short-term benchmark and tier it down as the balance grows, so a small debit pays the top tier and a large one pays much less — often several points apart. Two accounts holding identical stock can therefore have quite different carrying costs, and neither can tell from the position alone. The arithmetic is worth doing before the trade. Borrow $25,000 at 8% against a $50,000 position and the loan costs $2,000 a year, which is 8% of your own $25,000 — so the stock has to appreciate roughly 8% just for the account to stand still before leverage does anything at all. Above that point the leverage amplifies the excess: a 12% gain on the position is $6,000 against $2,000 of interest, a 16% return on your equity rather than 12%. A 12% fall is $6,000 plus the $2,000, which is 32% of your equity. The same asymmetry, mirrored. There is a mechanical asymmetry between the two outcomes that is easy to miss. You may repay the debit whenever you like, so the cost of leverage falls the moment you are right — a winning position can be de-levered and the interest simply stops. You cannot do the opposite. A losing position grows the balance, because the loss is financed by the loan, and the requirement rises in relative terms at the same time. Leverage is cheap to hold and expensive to be wrong in, and that is the honest reason so many accounts that read the direction correctly still were not around to collect on it. The one-sentence version for a learner: **margin changes the denominator**. Every return, every drawdown and every requirement is measured against your equity rather than against the position, so a leveraged result and an unleveraged one are not comparable numbers — which is the arithmetic behind almost every dazzling leveraged record and almost every blown-up one. Before electing portfolio margin, check what the broker actually charges for it. The requirement falls, but the financing rate often rises, and a lower margin requirement funded at a wider spread can be more expensive than the position it replaced.

Sources

• Regulation T — initial margin for marginable securitiesFederal Reserve — Regulation T
• Maintenance margin requirementsFINRA — Rule 4210
• The intraday margin standard that replaced the pattern-day-trader rule, effective June 4, 2026FINRA — day-trading and intraday margin rule change
• Margin loans, liquidation rights and recourse debtFINRA — Investor guidance on margin accounts