Every lending protocol advertises its loan-to-value ratio, and the number is always presented as generosity. Seventy percent sounds better than forty. More borrowing power, more capital efficiency, more of your own money available to you. If you are shopping on that number alone, Pillar looks stingy: forty percent against Apple, thirty-five against Nvidia, thirty against Tesla. We want to explain why those numbers are what they are, because the reasoning is the product. Anyone can offer you a higher ratio. What they cannot offer you is a market that stays open.
Here is the problem in one sentence. A tokenized equity follows an asset that stops trading on Friday afternoon and does not start again until Monday morning, and the loan it backs is live every second in between. Nothing about that is fixable. It is not a latency problem, so a faster oracle does not touch it. It is not a modelling problem, so no prediction curve helps. For roughly sixty-four hours every week, and longer over a holiday, there is a price your collateral is worth and no way for anyone to transact at it. The gap between Friday's close and Monday's open is not risk that a protocol takes on; it is risk the calendar manufactures and hands to whoever is holding.
Think about what actually happens in that window. A company reports earnings after the bell. A regulator opens an investigation at nine at night. A supplier warns, a trial fails, a product recalls, a chief executive resigns. By the time the market reopens the stock is fifteen percent lower, and it never traded at any of the prices in between. There was no moment at which a borrower could have added collateral in time, and no moment at which a liquidator could have stepped in early. The first price anyone can act on is the one after the damage.
Now run the arithmetic on a loan against that. If you borrowed seventy percent against a position and it opens fifteen percent down, your loan-to-value is instantly above eighty-two percent. You are underwater before you have had coffee, and the liquidation that follows is not a consequence of a bad decision you made. It is a consequence of a limit someone set for you, in a product that was competing on the size of that limit. If you borrowed thirty percent against the same position, the same gap moves you to a little over thirty-five percent. You are fine. You were always going to be fine, because the headroom was there before the news was.
That is the entire argument. The conservative number is not caution for its own sake and it is not a lack of confidence in the asset. It is the recognition that a fifteen percent overnight move in a single large company is an ordinary event, not a tail event, and that a lending product should be sized for ordinary events rather than surprised by them. We would rather explain a lower limit once than explain a liquidation to someone who did nothing wrong.
The differences between markets follow from the same logic. A broad-market ETF borrows at fifty percent because it is a basket: a single company's disaster is diluted across hundreds of holdings, and the index does not gap the way its worst constituent does. Apple and Microsoft borrow at forty because they are enormous and liquid, but they are still single issuers, and a single issuer can miss. Tesla borrows at thirty because its historical overnight moves are simply larger, and pretending otherwise would not make them smaller. None of these numbers are opinions about whether the company is good. They are statements about how far the price can travel while nobody is able to trade.
There is a version of this product that would look more impressive and serve you worse. It would push the ratio up, market the capital efficiency, and absorb the liquidations as a cost of doing business — borne by borrowers, not by the protocol. The liquidations would cluster on Mondays. They would be concentrated in exactly the users who borrowed the most against the thing they believed in most. And every one of them would be defensible on the terms the protocol had published, which is a different thing from being right.
We would rather be boring on this specific number. Boring is what a thirty percent loan looks like on the Monday it does not get liquidated.