Can your broker explain the perp liquidation waterfall without slides?

Every time a market tries to “make something safer” by bringing it onshore, the same confusion shows up. People talk about permission, venue, and logos. They talk about regulation like it is a safety certificate you can stick on a landing page.

But safety is not a certificate. Safety is plumbing. It is the boring machinery that keeps working when price moves faster than humans do.

That is why I read the excitement around CFTC-cleared perpetual futures on stocks with a very specific kind of respect and skepticism at the same time. Respect, because moving leveraged products into a clearer rulebook matters. Skepticism, because a perpetual future is not a new wrapper. It is a continuous risk engine, and it will punish anyone who treats it like “just another contract.”

The question I would ask any broker, platform, or sponsor rushing to offer regulated perps is simple: can you explain your liquidation waterfall in plain language, without slides, and without hand-waving? If you cannot, you are not launching a product. You are shipping risk.

The product never stops, so your risk never stops either

A dated future has a reset built in. You get expiry. You get roll. You get a forcing function that cleans up exposures, even if it is messy.

A perp does not give you that courtesy. It is always on. Risk is not something you reconcile at the end of the day. It is something you carry minute to minute, across volatile periods, across correlated names, across sudden liquidity gaps.

CoinDesk frames these contracts as a safer onshore alternative to offshore venues, now that they are CFTC-cleared. That may be true in a legal sense. Operationally, it just moves the responsibility onto participants who are used to different rhythms.

The hard part is not listing. The hard part is running margin, funding flows, and liquidation as a single system.

Margining has to be rebuilt for a contract that never expires

The moment you offer perps, margin is no longer a parameter. It becomes a design choice with second-order effects.

You are taking a view on:

• How quickly you reprice risk intraday, not just at end-of-day.
• How you treat correlated portfolios, where individual positions look fine until they all move together.
• How you handle gap risk, when the market jumps over your “safe” thresholds.

Most broker stacks are not built for that level of continuous repricing. They are built for products with natural pauses: session boundaries, clearing cycles, or at least a user base that sleeps at the same time.

With perps, margin is not “set and monitor.” It is a living mechanism. And you have to decide which pain you prefer: tighter margin that reduces your tail exposure but makes the product unattractive, or looser margin that sells well until the first violent move turns your support desk into a crisis room.

This is where regulated perps will surprise teams. The rulebook may be clearer, but the tolerance for sloppy risk will be lower, because the whole point of bringing it onshore is trust.

Funding rates are not an index, they are cash operations

The trap with funding is that it looks like an abstract market mechanic. It is not. It is cash moving between accounts, repeatedly, intraday or at scheduled intervals, forever.

CoinDesk points to the need to manage intraday funding-rate flows. That single sentence hides most of the work.

Because once funding becomes real flows, you have to answer uncomfortable questions:

• Where does the funding debit land if the client is under-margined at the funding timestamp?
• What happens if the client cannot pay but still has an open position?
• How do you represent the charge in statements, taxes, and dispute handling without confusing the client?
• What is your fail-safe if the funding calculation pipeline breaks at the worst time?

If you have ever watched a “small” mismatch between a ledger and a risk engine turn into days of manual clean-up, you know why I am stubborn about this. Funding is not a chart. It is a mini payroll system that never ends.

This is also where treasury discipline stops being a finance department topic and becomes a product requirement. If you liked my earlier lens on treasury KPIs that actually matter, perps are the same story under more stress. You are either built to reconcile fast, or you are built to argue later.

Liquidation is not a button, it is an engineered process

Most teams talk about liquidation like it is a feature toggle. Turn it on, set a threshold, send an email.

In reality, liquidation is a waterfall with moral consequences and financial consequences. When it works, nobody thanks you. When it fails, you own the loss and the headlines.

Here is the standard I use: if you cannot explain the liquidation path in plain language to a smart non-trader, you have not designed it yet.

At minimum, you should be able to articulate:

• The trigger logic: what exact conditions start deleveraging, and how often you evaluate them.
• The sequencing: partial reductions, full close, or position transfers. In what order, and why.
• The pricing method: mark price, index, last trade, blended feeds. When do you switch.
• The controls: kill switches, throttles, and what you do when liquidity is thin.
• The loss allocation: who eats slippage and residual losses when the market gaps.

I have built broker tooling in IBHQ, and the most consistent lesson is this: teams are happy to sell leverage, but strangely allergic to writing down the exact mechanics of how leverage gets unwound. That allergy is where disasters come from.

CoinDesk calls out that brokers need to treat liquidation as an engineered process. I agree, and I would push it further. Liquidation is part of user experience. It is also part of your solvency story. If the only time a client understands your rules is when they are getting closed out, you built a complaint factory.

Surveillance has to become real-time, not ceremonial

Perps amplify reflexivity. Leverage tightens feedback loops: price moves, margin changes, forced selling starts, price moves again. The product does not create manipulation, but it makes manipulation cheaper and faster.

That is why surveillance cannot be end-of-day reporting dressed up as “monitoring.” CoinDesk points to the need for real-time monitoring for manipulation, correlated squeezes, and liquidity gaps across venues. That is the correct direction, and it is also where most projects under-budget.

What changes when you take this seriously is that surveillance is no longer a compliance afterthought. It becomes part of your market integrity design:

• Cross-venue signals, because the perp price will be tugged by what happens elsewhere.
• Concentration alarms, because a few positions can dominate liquidation risk.
• Latency and feed quality checks, because bad prices create fake liquidations.
• Playbooks for halts and throttles, because “we will look into it” is not a response plan.

This is the part people miss: surveillance is not only about catching bad actors. It is about catching your own system when it starts behaving dangerously.

A practical readiness test before you list anything

If you are considering regulated perps, I would not start with a roadmap. I would start with a readiness review you can complete in a week, and I would insist on written answers.

Three questions. If any of them produce vague language, you are not ready.

1. Can we explain, in one page, how a client gets liquidated? Include triggers, sequencing, pricing sources, and what happens in a fast gap.
2. Can we reconcile funding flows the same day, every day? Not “eventually.” The same day, with exception handling and audit trails.
3. Can we detect a correlated squeeze while it is building? Not after. While it is building, with clear actions we will take.

If you pass those three, then yes, start thinking about go-to-market.

If you do not, your best move is not to ship anyway. Your best move is to narrow scope, reduce leverage, and build the machinery. The marketing will come later. The trust will not come back later.

My opinion: regulated perps will reward boring competence

There is a version of this story where “regulated perps” become a slogan and everybody rushes in. That version ends with a few spectacular failures and a lot of damaged trust.

The better version is quieter. A few teams do the unglamorous work. They treat margin, funding, liquidation, and surveillance as first-class systems. They write the rulebook in plain language. They test the edge cases before the market tests them.

That is who should win. Not because they are more compliant. Because they are more honest about what they are building: a risk engine with a rulebook.

If you are the person accountable for launching this kind of product, make one decision now. Decide whether you are in the leverage business, or in the trust business. If you pick trust, the plumbing becomes the product.