Comparison
Equity Wipeout Across Loan-to-Value: The Price Fall That Removes a Deposit, and Where a Maintenance Level Arrives First
How far must the price of a leveraged asset fall before the equity in it is gone, how much of a deposit does each point of that fall remove, and at what fall does the equity ratio instead reach a stated maintenance level?
The short answer
The fall that wipes out the equity is the share of the price you did not borrow. At 80% loan-to-value the equity is gone after a 20.00% fall; at 95% it is gone after 5.00%. Nothing else enters it — not the price, not the currency, not the term, not the rate.
The percentage understates what is happening. Because the loan does not move, every point the price loses is taken entirely out of the equity: at 80% each point of fall removes 5.0× of the deposit, and at 95% it removes 20.0×. Between those two rungs the deposit halves twice over and the multiple quadruples.
- Who it applies to
- Any position bought partly with borrowed money, where the loan does not change with the price — property, and a securities position held on margin. It compares ratios, not products, lenders or venues.
- What this does not tell you
- It establishes the price fall at which the equity reaches zero, or reaches a stated maintenance level, and nothing else. It does not establish that such a fall is likely, and it does not establish when anyone would act on one.
An exact identity rather than a projection: two ratios are compared directly, so no calculation-model version, horizon, rate or return assumption enters the answer, and nothing here goes out of date.
What is being compared
Two ratios, held against each other across a ladder. Nothing varies between rows except the share of the price that was borrowed, and no amount of money appears anywhere on this page.
| Loan-to-value levels compared | 50% · 60% · 70% · 75% · 80% · 90% · 95% |
|---|---|
| Maintenance-equity levels compared | 0% · 25% · 35% |
| Quantity compared | The proportional fall in price, measured from the price paid |
| Price | None. The identity is scale-free, so no amount and no currency appears |
| Horizon | None. Nothing is projected forward and no period is compounded |
| Interest, fees and costs | Not modelled. The loan balance is held at its starting figure |
| Repayments | None. A loan being paid down lowers the ratio and is a different question |
| Likelihood of any fall | Not modelled, not estimated and not implied anywhere |
Every level on both ladders is a stated level entered into the identity, chosen so the shape can be read. Neither ladder is a survey of what any lender offers or any venue requires, no provider is named, and nothing here reports a market condition.
Where the boundary comes from
Equity is what is left of the price after the loan. The loan does not move when the price does, so a fall of f leaves price × (1 − f) − loan, and setting that to zero gives the boundary directly:
price fall that removes the equity = 1 − loan-to-value
The price cancels. That is why the answer is a percentage rather than an amount, and why it is the same for a small position and a large one. At 50% loan-to-value it is 50.00%; at 95% it is 5.00%.
Read the other way round, equity remaining is 1 − f ÷ (1 − ℓ) of what was put in, so the divisor 1 − ℓ is the whole of the leverage. It is the same 1 ÷ (1 − x) shape the emergency-buffer reference publishes in a different subject, arrived at from a different direction.
The fall that removes the deposit, and the rate it removes it at
The middle column falls evenly down the ladder and looks mild. The right-hand column is the same fact, and it does not.
How fast a deposit disappears, as more of the price is borrowed
Each point the price loses is taken out of the deposit alone, so the rate rises as the deposit thins: 2.0× at 50%, 5.0× at 80%, 20.0× at 95%. It keeps rising past the right of this frame, and at 100% there is no deposit for it to describe. Nothing crosses anything — there is one relationship and one boundary.
- At 50% borrowed
- 50.00%
- removes the equity, at 2.0× of the deposit per point of fall.
- At 80% borrowed
- 20.00%
- removes it, at 5.0× per point.
- At 95% borrowed
- 5.00%
- removes it, at 20.0× per point.
- The loan balance is held at its starting figure
- No repayment, interest, fee or holding cost is modelled
- A fall measured from the price paid
- Ratios only — no price, no amount, no currency
- Nothing here estimates how likely any fall is
| Share of price borrowed | Share put in | Fall that removes it | Deposit removed per point of fall |
|---|---|---|---|
| 50% | 50.00% | 50.00% | 2.0× |
| 60% | 40.00% | 40.00% | 2.5× |
| 70% | 30.00% | 30.00% | 3.3× |
| 75% | 25.00% | 25.00% | 4.0× |
| 80% | 20.00% | 20.00% | 5.0× |
| 90% | 10.00% | 10.00% | 10.0× |
| 95% | 5.00% | 5.00% | 20.0× |
The second and third columns carry the same figure in every row, and that repetition is the point rather than an oversight: the fall that removes a deposit is the deposit. It is a definition rather than a result, which is why it needs no assumption to hold.
The ladder stops at 95% on purpose. The shape is fully established by then,100% has no equity in it at all — the identity rejects it rather than returning a very large number — and a page per deposit percentage would be a numeric-permutation family rather than an answer.
The same identity, at a level above zero
A securities position held on margin is not usually left alone until the equity reaches zero. It is measured against a stated maintenance level — the equity as a share of what the position is currently worth. Requiring that share to equal m gives the same identity with one more term:
fall that reaches the level = 1 − loan-to-value ÷ (1 − maintenance level)
At m = 0 it returns 1 − loan-to-value exactly, which is the column on the left below. That is why this is one page and not two: the property reading and the securities reading are the same arithmetic at two levels, not two mechanisms that resemble each other.
| Share of price borrowed | 0% maintenance | 25% maintenance | 35% maintenance |
|---|---|---|---|
| 50% | 50.00% | 33.33% | 23.08% |
| 60% | 40.00% | 20.00% | 7.69% |
| 70% | 30.00% | 6.67% | Below the level already |
| 75% | 25.00% | At the level already | Below the level already |
| 80% | 20.00% | Below the level already | Below the level already |
| 90% | 10.00% | Below the level already | Below the level already |
| 95% | 5.00% | Below the level already | Below the level already |
Read across a row and the fall required shrinks as the level rises, because a higher maintenance level is reached earlier. The wipeout fall is therefore the most generous threshold there is: any stated level above zero arrives before it, and never after.
Two rows carry named states rather than figures, and they are the reason the two markets are not the same decision. At 75% borrowed against a 25% maintenance level the position sits exactly at the level before the price moves at all, and above that ratio it is already below it. A borrowing level that is unremarkable when the threshold is zero is outside the ratio entirely once a maintenance level is stated — which is not a claim about what any venue permits, only about what the two numbers entered imply.
A ratio reaching a level is not a forced sale
Everything in this section is a statement about a ratio. When, whether and how anyone acts on that ratio is a different matter entirely, and this page does not reach it. A call may or may not be issued, there may or may not be time to meet it, some or all of the position may be closed, the price used may not be the price quoted here, and fees and slippage may make the outcome worse than the arithmetic.
Those are lender and venue rules, not arithmetic. They differ between firms, between products and between markets, and they change. Nothing on this page names one, models one or should be read as predicting one.
A threshold is not a probability
Everything above is conditional. It says what a fall of a given size would do, and says nothing whatever about whether such a fall is likely, imminent or plausible for any asset. A 5.00% threshold is not a 5.00% risk, and no figure here can be turned into one: there is no volatility in this model, no horizon, no distribution and no history.
Nor does the threshold decide anything on its own. Two holders at the same ratio can be in entirely different positions depending on:
- whether the holder can meet the payments while the price is down
- whether they have to sell at the low price, or can wait
- recourse — whether the lender can pursue assets beyond the one financed
- the cost and availability of refinancing at a higher ratio
- transaction costs on the way in and on the way out
- how quickly the asset can be sold at all
- what else the same money could have been doing
No level of borrowing is presented here as sensible or reckless. A thin deposit buys a larger position with less money and removes the cushion faster; which of those matters depends on circumstances this page does not know and cannot see.
What this comparison does not determine
The figures on this page are exact arithmetic on two stated ratios. They are not market conditions, quoted requirements, historical falls or a forecast of anything. The comparison cannot determine:
- whether a price fall of any size is likely, imminent or plausible — nothing here is a forecast
- when a lender, broker or venue would issue a call, close a position or force a sale
- what a close-out would cost in fees, spread or slippage, none of which is modelled
- interest, arrangement fees, insurance, maintenance or any other holding cost
- the effect of repayments, which lower the ratio over time and are held fixed here
- tax of any kind, and no wrapper, allowance, jurisdiction or rate is modelled
- inflation, so every figure here is nominal and every fall is a nominal fall
- whether any level of borrowing is prudent, affordable or suitable for anyone
- what any real lender, broker, exchange or product actually requires
It also holds the loan still. A repayment mortgage's balance falls over time, so its ratio improves and its wipeout fall widens; a position whose loan accrues interest moves the other way. Both are real and neither is modelled here, because the identity is a statement about one moment rather than about a path.
None of this is financial, legal or tax advice, and no product, provider, lender, exchange or platform is recommended, ranked or named anywhere on this page.
Where these figures come from
Every number above is produced at build time from price fall that removes the equity = 1 − loan-to-value and its maintenance-level form, applied to the two ladders in the comparison set. Nothing on this page is typed by hand.
There is no calculation-model version behind them, and that absence is the point. The threshold is a subtraction and a division on two stated ratios rather than a projection, so it inherits no horizon, no assumed return and no rate. Its regression suite holds every published figure against a second, independently written implementation of the same identity, and asserts that the maintenance form collapses onto the wipeout fall at a zero level exactly — two implementations that agree are evidence, one checked against itself is not.
The build refuses a loan-to-value of 100% rather than reporting a very large multiple for it. There is no equity at the outset there, so there is nothing for a fall to remove, and printing a boundary that is not a quantity would be worse than printing none.