Bitcoin futures reached CME by self-certification in 2017, leaving the risk decisions to clearing firms
CME’s bitcoin futures began trading for the December 18, 2017, trade date after the exchange certified them itself. The CFTC said that was not approval. In our reading, the decisions that mattered fell to clearing members and the firms that served clients.
Looking Back revisits past weeks one at a time, up to the most recent, and asks what it meant for the banks, health systems and public agencies we serve. The series was written in 2026, with the benefit of hindsight, and each article says which week it looks back on.
Two-minute summaryThe dek and the first sentence of each section, quoted from the article
CME’s bitcoin futures began trading for the December 18, 2017, trade date after the exchange certified them itself. The CFTC said that was not approval. In our reading, the decisions that mattered fell to clearing members and the firms that served clients.
- The week On Sunday, December 17, 2017, CME Group’s new bitcoin futures contract became available for trading on its Globex electronic platform, as CME had announced on December 1.
- Listing: certification by the exchange, not approval by the regulator Under the Commodity Exchange Act and the CFTC’s Part 40 rules, a designated contract market can list a new product in one of two ways.
- Price: a regulated contract on an unregulated reference The second link was the price.
- Collateral: margin and mutualized risk at the clearinghouse The third link was the clearinghouse.
- Clients: who decided which customers could trade The fourth link was the client.
- How the decisions differed by type of firm The same listing produced different decisions for different firms.
- Objections, in order Five objections deserve an answer.
- What a leader could check These checks follow from the four parts.
- With hindsight The CFTC returned to the launch in a backgrounder dated January 4, 2018.
The week
On Sunday, December 17, 2017, CME Group’s new bitcoin futures contract became available for trading on its Globex electronic platform, as CME had announced on December 1. Trades counted for the trade date of Monday, December 18. The contract was also available for submission for clearing through CME ClearPort from the same time. It was listed on, and subject to the rules of, the Chicago Mercantile Exchange.
CME had announced the launch on December 1. Its release said it had self-certified the initial listing that day. The contract would be cash-settled against the CME CF Bitcoin Reference Rate, a once-a-day reference for the U.S. dollar price of bitcoin. CME and Crypto Facilities had calculated and published that rate since November 2016. It aggregated trade flow from major spot exchanges during a calculation window, as of 4:00 p.m. London time. The release said it was designed around the IOSCO Principles for Financial Benchmarks. Four exchanges contributed prices: Bitstamp, GDAX, itBit and Kraken.
The release quoted CME’s chief executive on the risk controls for launch. They included an initial margin of 35 percent, position limits, intraday price limits and the risk and credit controls CME offered on all its products. He said CME had worked closely with the CFTC and market participants on the design.
The Commodity Futures Trading Commission issued its own statement on December 1. It said that CME and the CBOE Futures Exchange had self-certified new bitcoin futures contracts that day, and that the Cantor Exchange had self-certified a bitcoin binary option. In the release, the CFTC’s chairman described bitcoin as a commodity unlike any the Commission had dealt with before. He said the underlying cash markets and exchanges remained largely unregulated, and that the CFTC had limited statutory authority over them.
The statement described what had happened before the filings. CFTC staff had held discussions with CME over six weeks and with CFE over four months. The exchanges had agreed, at staff’s request, to enhancements to contract design and settlement. CME had also agreed to changes in margining. The exchanges had agreed to more information sharing with the underlying cash exchanges, to help surveillance.
The statement set out what would follow launch. Staff would monitor the size and development of the market, positions, open interest, initial margin and variation margin payments. They would stress test positions. They would review the exchanges, the clearinghouses, the clearing firms and individual traders involved. If the agency found the margin held at clearinghouses inadequate, it said it could require more, including a longer margin period of risk.
The statement was explicit about one point. As with all contracts listed on CFTC-regulated exchanges and cleared through CFTC-regulated clearinghouses, the completion of the process was not a Commission approval. It was not an endorsement of the use or value of virtual currency products. The statement said it was incumbent on market participants to conduct appropriate due diligence on whether the products were appropriate, given extreme volatility and unique risks.
The National Futures Association, the statement noted, had issued an investor advisory to its members on the topic. Those members included the futures commission merchants and introducing brokers that would handle customer orders in the new contracts.
Bitcoin futures drew a public statement from the CFTC on the day they were certified, December 1, 2017. The underlying asset traded on spot venues the agency called largely unregulated, at prices it said had at times shown extreme volatility. Their arrival on a major regulated exchange could be read as a regulatory verdict on bitcoin. The CFTC said plainly that it was not.
Our argument is that the route the contracts took, self-certification, placed the consequential risk decisions with private firms. The exchange decided to list. The clearinghouse set minimum margin. But each clearing member decided whether to clear the contracts at all, what margin to charge above the minimum, what limits to set and which clients to accept. Each broker decided whom to offer access. In our reading, the week of December 18 was when those decisions stopped being hypothetical.
For a bank-affiliated futures commission merchant, a broker-dealer or a wealth manager, this is a governance point. A listing on a regulated exchange tells a firm that the exchange has certified compliance with the law. It does not tell the firm that the product suits its risk appetite or its clients. The firm’s own committees have to decide that.
The timing sharpened the point. CME announced its launch on December 1, and the first trade date was December 18. In our view, a governance process that assumed weeks of review before a new product went live could not finish inside that window, so some firms would still be deciding their position after trading began.
By risk decisions we mean the choices that determine who loses money, and how much, if the product behaves badly. The exchange’s design and the clearinghouse’s margin set the outer frame. Inside it, each firm decides how much of the loss it is prepared to carry and which clients it lets near it. Those choices are where a bad outcome is contained or spread.
We take the decisions in four parts. Each corresponds to a link in the chain from listing to client, and together they cover the chain:
- Listing: what self-certification meant, and what the regulator could and could not do.
- Price: how the contract settled, and what it depended on outside the regulated market.
- Collateral: how margin and clearing allocated the risk among the clearinghouse, its members and their customers.
- Clients: who decided which customers could trade, and on what basis.
After the four parts we look at how the decisions differed by type of firm, take the main objections in order, and list what a leader could check. We describe the rules from their text and the agencies’ own statements. Where we say what a rule required of a particular firm, that is our reading, and firms should check with counsel.
Listing: certification by the exchange, not approval by the regulator
Under the Commodity Exchange Act and the CFTC’s Part 40 rules, a designated contract market can list a new product in one of two ways. It can submit the product for Commission approval under section 40.3. Or it can certify the product itself under section 40.2.
What a certification contained
Section 40.2, as in effect in December 2017, set out the contents of a certification. The exchange had to file electronically with the Commission’s Secretary by the open of business on the business day before listing. The filing included the product’s rules and terms, the intended listing date and a certification that the product complied with the Act and the Commission’s regulations.
It also included a concise explanation and analysis of the product’s compliance with the Act, including the core principles. That analysis had to be accompanied by, or cite, the documentation relied on. The exchange had to post a notice of the pending certification and a copy of the submission on its website at the same time.
What the regulator could do
The rule gave the Commission two tools. Under paragraph (b), staff could request additional evidence, information or data showing the contract met the Act’s requirements, initially or on a continuing basis. Under paragraph (c), the Commission could stay a listing during proceedings for filing a false certification, or during a petition to amend the contract’s terms under section 8a(7) of the Act. The decision to stay could not be delegated to staff.
Those are narrow grounds. In our reading, the rule gave the Commission no general power to refuse a certified product because it disliked the underlying market, and that is why the CFTC’s December 1 statement focused on what staff had negotiated before the filings, not on what it could block after them.
Visibility to firms before listing
The posting requirement gave firms one advantage. An exchange had to post a notice of the pending certification and a copy of the submission on its website when it filed. Firms that watched those postings could read the contract terms before trading began. But the rule provided no comment period. A clearing member that objected to a product’s design had no formal channel in the certification itself. Its routes were the exchange’s own committees and direct contact with the exchange or the regulator.
Prior approval as the alternative
Section 40.2 applies to products that have not been approved under section 40.3, the route for voluntary submission for Commission review and approval. An exchange could choose approval. It did not have to. In our reading, the choice between the two routes was the exchange’s, and an exchange that wanted speed had a reason to certify. Firms had no say in that choice either.
What staff negotiated
The statement said staff had discussed the contracts with CME for six weeks and with CFE for four months. The exchanges agreed to changes in contract design and settlement, and CME to changes in margining. They agreed to share more information with spot exchanges and to coordinate surveillance with the CFTC. Those were agreements, not orders. They depended on the exchanges’ cooperation.
What the listing meant for a firm
For a regulated firm, the key sentence in the statement was the one saying that completion of the process was not a Commission approval. A firm’s new product committee might ask whether a product was regulator-approved. For these contracts, the answer was no. The exchange had certified compliance with the law. The regulator had not found otherwise. That is a different statement.
What a new product committee would ask
A firm’s new product committee reviewing bitcoin futures in December 2017 had a short list of matters to settle. It had to establish who regulated the underlying market and how the settlement price was formed. It had to know what the clearinghouse required and what the firm would require above that. It had to decide which clients wanted the product and which should have it, and what could go wrong operationally on the first day. Each matter had an answer in the documents available that month: the exchange’s announcement, the CFTC’s statement and the firm’s own rules. The committee’s job was to put the answers together and decide.
Who carried the residual risk of a wrong certification
If a certification were later found false, the Commission could act against the exchange. But positions already opened would remain, and clearing members would carry them until closed. In our reading, the self-certification route put the cost of any error with the exchange and with the firms that had chosen to participate. That allocation was a reason for each firm to do its own review rather than rely on the listing.
Price: a regulated contract on an unregulated reference
The second link was the price. A cash-settled futures contract never delivers the underlying asset. At expiry, positions are settled in dollars against a reference price. The quality of that reference price decides whether the contract does what it claims.
How the reference worked
CME’s contract settled against the CME CF Bitcoin Reference Rate. CME’s release described it as aggregating trade flow from major bitcoin spot exchanges during a calculation window into a single dollar price as of 4:00 p.m. London time. Four exchanges contributed at launch. The rate was designed around IOSCO’s benchmark principles.
The design addresses some risks. Aggregating several venues reduces dependence on any one. A calculation window rather than a single trade makes a spike at one moment less decisive. Publishing the methodology lets users check it.
What the reference depended on
But the inputs came from spot exchanges the CFTC’s release described as largely unregulated, over which the agency had limited statutory authority. The chairman’s statement in the release named the concerns: price volatility, trading practices, potential manipulation, flash rallies and crashes, and trading outages. It said the agency expected the exchanges, through information sharing agreements, to monitor activity on the cash platforms for effects on futures price discovery.
For a firm clearing or trading the contracts, that is a dependency it could not control. The futures contract was regulated. Its settlement price came partly from venues that were not. An outage or manipulation on a constituent exchange during the window could move the settlement price.
What a firm could do about it
A firm could not change the reference. It could understand it. Its risk team could read the methodology, know which venues contributed and watch for anomalies around the settlement window. It could size positions with the possibility of a disrupted settlement in mind. In our reading, a firm whose risk models treated the settlement price as if it came from a deep regulated market would misjudge the risk.
Settlement once a day
The reference rate was a once-a-day price. Futures traded around the clock on Globex during trading hours, but final settlement depended on a single daily calculation window. A firm’s operations team had to know when that window fell, which venues fed it and what the exchange would do if a constituent venue failed during it. Those are the operational questions behind any cash-settled contract. For this one, the inputs were less familiar than usual.
When the reference cannot be calculated
Any benchmark can fail to publish. A constituent venue can go offline, or too few trades can occur in the window. A firm clearing a cash-settled contract needed to know what the exchange’s rules provided in that case, and to read them before launch rather than after a failure. We do not describe CME’s fallback rules here, because the sources we opened do not set them out. The point is that the question existed and belonged on the firm’s list.
Basis risk for hedgers
For a client using the futures to hedge spot holdings, the reference created basis risk. The client’s spot position would be valued on whatever venue it held bitcoin. The futures would settle on the reference rate. If the client’s venue diverged from the reference, the hedge would not match. A broker offering the contracts to hedgers needed to explain that.
The surveillance chain
The CFTC said it would coordinate with the exchanges on surveillance, including receiving more surveillance information. The exchanges’ agreements with spot venues gave them some access to spot trading data. That chain ran from spot venues to the futures exchange to the regulator. Each link depended on the one before. A clearing member was outside that chain. It saw its own clients’ positions and the market price, but not the spot venues’ data.
Collateral: margin and mutualized risk at the clearinghouse
The third link was the clearinghouse. CME’s contracts are cleared by its clearinghouse, CME Clearing, which stands between buyer and seller. Clearing members stand behind their customers’ positions at the clearinghouse. If a customer cannot pay, the clearing member must.
Clearinghouse margin
CME’s December 1 release said initial margin at launch would be 35 percent. The CFTC’s backgrounder of January 4, 2018, said CME’s initial margin was in the end set at 47 percent. The CFTC’s December 1 statement said that if it found margin inadequate it could require more. Initial margin is the clearinghouse’s first protection. It is meant to cover the loss on a defaulting position over the time needed to close it.
For a contract on an asset with large daily moves, the margin level matters. Too low, and a default leaves a loss the clearinghouse must cover from other resources. Whether CME’s figure was high enough was a judgment about bitcoin’s volatility over a short horizon, made for an asset with little history in regulated markets.
Price limits and halts
CME’s release listed intraday price limits among its controls. Price limits slow trading when prices move sharply, which can reduce disorderly moves. They also mean a firm may not be able to close a position at the moment it wants to. For a clearing member managing a defaulting customer, a halt delays liquidation while the underlying spot market, which has no such limits, keeps moving. In our reading, a risk model that assumed continuous liquidation would understate the risk of a contract with price limits on an underlying that trades without them.
Mutualized resources
If margin is not enough, a clearinghouse draws on other resources, including a guaranty fund to which clearing members contribute. That fund is shared. A large loss on bitcoin futures could, in principle, draw on contributions from members who never cleared a bitcoin contract.
In our reading, that concern lay behind one clearing member’s call, reported in the CFTC’s later backgrounder, for a separate clearing system for virtual currencies. The backgrounder said the self-certification process did not provide for the creation of separate guaranty funds. It said separate funds had come about in the past through negotiations between clearing members and clearinghouses, not by CFTC action.
In our reading, this is where the listing decision reached firms that chose not to participate. A bank-affiliated clearing member that declined to clear bitcoin futures still contributed to a guaranty fund that stood behind them. Its only levers were its voice on the clearinghouse’s risk committees and its decision whether to remain a member.
The clearing member’s voice
Clearing members are not passive. Clearinghouses run risk committees on which members serve, and members can raise concerns with the clearinghouse and the regulator. The CFTC’s later backgrounder pointed to members’ active participation in those committees as one way they could protect themselves. In our reading, a member that wanted different margin or a separate fund had to use that voice before launch. Once trading began, changes were harder.
House margin and limits at the clearing member
Clearing members could charge customers more than the clearinghouse minimum. They could also set position limits and decline orders. The CFTC’s rule on clearing FCM risk management, section 1.73, required each clearing FCM to establish risk-based limits in the proprietary account and in each customer account. The limits could be based on position size, order size, margin requirements or similar factors.
The rule also required order screening. For electronic access or automated execution, the FCM had to use automated means to screen orders against the limits. It had to monitor adherence intraday and overnight. And it had to stress test positions that could pose material risk to it at least once a week, under extreme but plausible conditions.
What section 1.73 meant for a new product
Section 1.73 did not mention bitcoin. It did not need to. When a clearing FCM decided to accept bitcoin futures, the rule required it to have risk-based limits for those positions, screening that applied them and stress tests that covered them. In our reading, a firm could not comply by applying its general limits to a product whose risk profile differed sharply from its usual contracts.
So the rule turned the decision to clear bitcoin futures into a set of documented choices. What limit for each customer. What stress scenario. What house margin. Those choices had to exist before the first order, because the screening had to be automated.
Concentration in a new market
A new contract starts with few participants. A single large client can hold a significant share of open interest. If that client defaults, its position may be large relative to the market’s depth, and closing it may move the price against the firm. Section 1.73’s limits could be set with that in mind, for example by capping any client’s position relative to total open interest. That is a design choice for the firm, not a requirement of the rule.
Liquidity and settlement
Section 1.73 also required clearing FCMs to evaluate weekly their ability to meet initial margin requirements, and their ability to meet variation margin in cash. Bitcoin futures could generate large daily variation margin flows. A clearing member had to pay the clearinghouse on time even if a customer was late. For a firm with modest capital, a sudden move could create a funding need it had not planned for.
Defaults and liquidation
The rule required clearing FCMs to evaluate, at least once a quarter, their ability to liquidate positions in an orderly manner and to estimate the cost. For a new contract with uncertain liquidity, that estimate was hard. A firm accepting bitcoin futures needed a view on how quickly it could close a defaulted customer’s position, and at what cost, in a market only days old.
Clients: who decided which customers could trade
The fourth link was the client. Futures are traded through futures commission merchants and introducing brokers, registered with the CFTC and members of the National Futures Association. Some are affiliates of banks and broker-dealers. Each decides which customers to accept and which products to offer them.
The regulator’s message to participants
The CFTC statement said that it was incumbent on market participants to conduct appropriate due diligence to determine whether the products were appropriate. It warned of the potentially high level of volatility and risk. The NFA had issued an investor advisory to its members, including FCMs and introducing brokers involved in virtual currency futures. The statement said NFA would closely monitor its member firms trading the product.
Decisions a broker had to make
A broker offering bitcoin futures had to decide several things. Which account types could trade. Whether retail customers could trade, or only institutional ones. What disclosures customers received. Whether existing futures customers needed a new approval to trade the contract. What limits applied. Each decision was the firm’s. None was set by the listing.
Staff who answer clients
Clients ask the people they deal with, not the risk committee. Brokers, relationship managers and service staff needed to know the firm’s position before the first call. They needed to know whether the firm offered the product, to whom, on what terms and with what warnings. A firm that decided its policy but did not brief its front line would see that policy applied inconsistently. Short written guidance for client-facing staff, issued with the decision, closes that gap.
Clients who already held spot bitcoin
Some clients would come to the futures already holding bitcoin elsewhere. For them, the futures might be a hedge or a way to add exposure. A broker could not see those spot holdings unless the client disclosed them. A margin call on a short futures hedge would require cash, even if the client’s spot holding had risen in value. Explaining that liquidity mismatch was part of offering the product fairly.
Disclosures and conduct
The risks the CFTC named, from volatility to the unregulated spot market and the possibility of outages, are the substance of a disclosure. A broker that described bitcoin futures as a regulated way to trade bitcoin, without explaining that the price came from unregulated venues, would give an incomplete picture. In our reading, the CFTC’s statement supplied the outline of a fair disclosure.
Clients of affiliated firms
For a bank holding company with a broker-dealer, an FCM and a wealth management business, the client question crossed entities. The FCM might clear for institutional clients. The wealth business might receive client requests to buy. The broker-dealer might be asked about related funds. A consistent firmwide position needed a decision at the level above any one entity.
Electronic access and automated screening
Section 1.73 required automated pre-trade screening when a clearing FCM provided electronic market access or accepted orders for automated execution. Retail and active-trader platforms connect clients directly to the market. For a firm offering bitcoin futures on such a platform, the limits it chose had to be configured in its screening system before launch. A limit that existed only in a policy document would not stop an order. The configuration was the control.
Monitoring after launch
The first weeks of a new product show how clients use it. Complaints, margin calls, failed payments and unusual trading patterns are early signals. A firm that set limits at launch and then left them unchanged would miss what it learned. The CFTC said it would monitor positions, margin and variation margin flows across the market. A firm could do the same for its own book, and adjust limits and eligibility as the product matured.
Records
Each of these decisions belongs in a record: the new product approval, the risk committee minutes, the limit settings, the disclosure text and the client eligibility rule. If the product later caused losses or complaints, the record would show what the firm decided and why. That record is also what an examiner from the NFA or a prudential supervisor would ask to see.
How the decisions differed by type of firm
The same listing produced different decisions for different firms. Here is how the four links applied by type.
Bank-affiliated clearing FCMs
For a clearing FCM inside a bank group, the decision was whether to clear at all. Clearing exposed the firm to customer defaults and to mutualized losses. Not clearing still left it exposed to mutualized losses through the guaranty fund. The decision also touched the bank group’s risk appetite and its relationships with regulators. The CFTC’s backgrounder later said some clearing members had chosen not to trade virtual currency futures, and many had required substantially higher initial margins.
Non-bank FCMs and introducing brokers
Smaller FCMs and introducing brokers serving active traders faced client demand directly. Their risk was concentrated in their customer base. Section 1.73’s limits, screening and stress testing applied to clearing FCMs. Introducing brokers do not clear, but they depend on a clearing FCM that will accept the orders, and that FCM’s limits flow through to them.
Broker-dealers and wealth managers
Broker-dealers and investment advisers that are not futures firms faced a different question. Their clients might ask for exposure. Advising a client to trade futures, or buying them in a managed account, required a futures account through an FCM and an assessment of suitability under the firm’s own standards. In our reading, an adviser’s fiduciary duty applied to the recommendation in the usual way. The listing changed none of that. The adviser still had to understand the product, its costs and its risks before recommending it.
Proprietary trading desks
Some firms might trade bitcoin futures for their own account, to hedge or to make markets. Section 1.73 required risk-based limits in the proprietary account as well as customer accounts. For a bank-affiliated firm, proprietary trading also raised questions under other rules on bank trading activities, which firms would need to check with counsel. The listing did not settle those questions.
Asset managers and funds
Fund managers registered as commodity pool operators could use the futures in pools. Others might consider them for funds regulated under securities law. In each case, the manager’s own investment, valuation and liquidity processes decided whether the contract fit. A fund that could not value or exit a position in a stressed market would face problems the listing did not solve.
Market makers and liquidity providers
Firms that make markets in new contracts provide the liquidity others rely on. For bitcoin futures, a market maker would want to hedge in the spot market, which meant holding or trading bitcoin on the unregulated venues the CFTC described. For a regulated firm, that hedge brought custody, counterparty and venue risks the futures alone did not. The decision to make markets was therefore also a decision about spot market exposure.
Banks and credit unions without futures businesses
Banks and credit unions without futures operations had no direct exposure from the listing. Their questions came later and from other directions, such as customers who traded through other firms or deposits from businesses in the sector. We leave those to other weeks.
Fintechs and new entrants
Firms that had built businesses around spot bitcoin could read the listing as a sign of acceptance. But the CFTC’s statement denied that the listing endorsed the use or value of virtual currency. A fintech that described its products as regulated because futures now traded on CME would be overstating what had happened.
Objections, in order
Five objections deserve an answer. We take them in the order they bear on the argument.
1. The CFTC’s heightened review was effectively an approval
Staff negotiated design, settlement, margin and surveillance changes before the filings. That looks like approval. But the agency itself said it was not. The legal difference matters: the Commission made no finding that the product was appropriate, and it disclaimed endorsement. A firm relying on the review as approval would be relying on something the regulator had expressly said it was not giving.
2. Clearinghouse margin was high, so clearing members bore little risk
Initial margin was high. CME announced 35 percent, and the CFTC later said the level was set at 47 percent, more than ten times the margin required for CME corn futures. But margin protects the clearinghouse against a defaulting member, not the member against its customer. A member faced a customer’s loss beyond the customer’s collateral, the funding of variation margin and the shared guaranty fund. High clearinghouse margin reduced one risk. It did not remove the others.
3. Firms could simply decline to participate
Firms could decline to clear or offer the contracts, and some did. But declining was itself a decision that needed governance, and it did not remove exposure through mutualized clearinghouse resources. It also had commercial effects if clients took their business elsewhere. The decision was the firm’s either way.
4. The product was small, so the governance burden was disproportionate
The market was new and small at launch. But the governance burden of a new product does not scale only with its size. Limits, screening, stress tests and disclosures have to exist before the first trade. In our reading, the right size for the effort was set by the product’s risk profile, which was unusual, not by its early volumes.
5. The real risk was in the spot market, which the listing did not touch
The spot market remained outside federal market regulation, as the CFTC said. But the futures connected regulated firms to that market through the settlement price. Before December 2017, a CME clearing member had no bitcoin exposure through that clearinghouse. After it, the member’s clients could gain exposure through a contract it cleared, and in our reading the member shared in the clearinghouse’s mutualized exposure whether or not it cleared the contracts. The listing brought spot market risk closer to regulated firms’ books, which is why the firms’ decisions mattered.
What a leader could check
These checks follow from the four parts. They are our suggestions, not requirements.
- Does the firm’s new product process distinguish between products a regulator approved and products an exchange self-certified?
- For each cash-settled contract the firm clears or offers, does the risk team know the reference price methodology and its input venues?
- Are house margin and position limits for new contracts set from the firm’s own stress tests, with the analysis recorded?
- Does the firm know its share of clearinghouse guaranty fund exposure, and does it use its voice on clearinghouse risk committees?
- Is there a written rule on which client types may trade each new product, with disclosures matched to its risks?
The checks need owners. The new product process belongs to the chief risk officer with the business. Reference price analysis belongs to market risk. Limits and stress tests belong to the FCM’s risk function under section 1.73. Clearinghouse exposure belongs to treasury and the firm’s representatives on clearinghouse committees. Client eligibility and disclosures belong to compliance.
The checks apply to any novel listed product, not to bitcoin alone. Each asks whether the firm made its own decision or borrowed one from the listing. A firm that can answer them for every new contract it touches has a product process that does not depend on regulators to approve what exchanges list.
With hindsight
The CFTC returned to the launch in a backgrounder dated January 4, 2018. It explained that the self-certification process had been designed by Congress and prior Commissions to give the initiative to exchanges. It said the process did not provide for public input, separate clearing guaranty funds or value judgments about the underlying spot market. It said grounds to stay a certification were limited, such as a false statement, and that none were evident in the CME and CFE filings.
The backgrounder described the heightened review in seven elements. They included substantially high initial and maintenance margin at the clearinghouses, large trader reporting thresholds at five bitcoins or less, and information sharing agreements with spot platforms. It also listed exchange monitoring of cash market data, inquiries down to trade settlement level, coordination with CFTC surveillance and coordinated launches. A footnote said initial margin had in the end been set at 47 percent for CME and 44 percent for CFE, more than ten times the margin on CME corn futures.
The backgrounder also addressed clearing members directly. It said major banks and brokerages that are clearing members could look after their own commercial interests in three ways. They could choose not to trade virtual currency futures, as some had done. They could require substantially higher initial margins, which it said many had done. And they could take an active part in clearinghouse risk committees. It said the CFTC’s hands were tied by statute from requiring a separate clearing system or guaranty fund. That is consistent with our argument: the agency itself located the risk decisions with the clearing members.
On May 21, 2018, two CFTC divisions issued Staff Advisory 18-14 on virtual currency derivative product listings. It applied to listings under either self-certification or prior approval. It set out staff expectations in five areas: enhanced market surveillance, coordination with CFTC staff, large trader reporting, outreach to stakeholders and clearinghouse risk management. It said a well-designed surveillance program included information sharing with the spot markets that make up the settlement price.
The advisory gave clearing members a larger part in the process. It said staff expected an exchange to solicit comments and views before listing a virtual currency contract, and that the exchange should consider consulting clearing members and FCMs, including those that did not plan to clear it. It said a submission should consider including an explanation of substantive opposing views from that outreach and how the exchange addressed them. In our reading, that was a response to the governance gap the December 2017 launch exposed.
The longer arc shows in a decision by another agency. On January 10, 2024, the Securities and Exchange Commission approved proposals by three exchanges to list spot bitcoin exchange-traded products. Its order rested in part on the CME bitcoin futures market. It found, based on correlation analysis, that fraud or manipulation affecting spot bitcoin prices would likely similarly affect CME bitcoin futures prices. So a surveillance-sharing agreement with CME could reasonably be expected to help detect fraud and manipulation that might affect the spot products.
The futures market that opened on December 17, 2017, thus became, in the SEC’s analysis, part of the protection for investors in the spot products. The order also recorded the limits of that protection: spot bitcoin does not trade on CME, and CME does not surveil spot bitcoin markets.
What held up from our argument is the allocation of decisions. The regulator said it was not approving, and later said clearing members could protect themselves through their own choices. What changed is the process around those choices. The 2018 staff advisory said exchanges should consider consulting clearing members before listing and explaining their objections in the filing. In our reading, that gave the firms that bore the risk a clearer voice, though through staff guidance, not a rule. For a financial firm in 2026, the practical point is unchanged: an exchange listing is the start of the firm’s own product decision, not the end of it.
Spotted an error? Tell us and we will check it and correct the article where needed.