Who Owns the Number?

If prices can be con­structed, some­body writes the con­struc­tion rules. For most of the numbers that matter, that some­body is a small organ­i­sa­tion almost nobody outside finance has heard of, working from a method­ol­ogy doc­u­ment almost nobody has read.

Modern finance runs on ref­er­ence numbers. A mort­gage resets against a rate. A pension fund owns a slice of an index and is judged against it. A mine sells its gold at a bench­mark. A futures con­tract settles against a ref­er­ence rate. A lending pro­to­col on a blockchain liq­ui­dates a bor­rower when an oracle says the col­lat­eral has fallen far enough. None of these con­tracts names a trade. Each names a number, and a pro­ce­dure for pro­duc­ing it.

Once a number decides who pays whom, the pro­ce­dure stops being a tech­ni­cal­ity. Somebody spec­i­fies the inputs. Somebody spec­i­fies the rule. Somebody runs it and pub­lishes. Every one of those roles creates an incen­tive, and every incen­tive a failure mode. Governance, the unglam­orous busi­ness of who is allowed to do what and who checks, becomes a piece of market infra­struc­ture as real as a match­ing engine.

The rate that was asked, not traded

LIBOR, the London Interbank Offered Rate, is the case every­body learns, for good reason. From 1986 it was pub­lished under the British Bankers' Association, and for decades it sat under­neath an enor­mous range of con­tracts: inter­est rate swaps, syn­di­cated loans, float­ing-rate notes, student loans and American adjustable-rate mort­gages. The US reg­u­la­tors' own working group esti­mated in 2018 that around two hundred tril­lion dollars of finan­cial con­tracts ref­er­enced US dollar LIBOR.

Nothing ever traded at LIBOR. Each morning a panel of banks was asked, in effect, a hypo­thet­i­cal ques­tion: at what rate could you borrow funds, were you to do so by asking for and then accept­ing inter­bank offers in a rea­son­able market size just before 11 am London time? Each bank sub­mit­ted a number. The highest and lowest quarter of sub­mis­sions were thrown away, the middle was aver­aged, and the result was pub­lished for each cur­rency and each bor­row­ing period.

The design had a logic. Banks did lend to each other unse­cured, but not in every cur­rency and every matu­rity every day, so a survey of informed esti­mates could cover gaps that trans­ac­tions could not. Trimming the extremes meant one strange answer could not move the result much.

The design also had two flaws that nobody had to be a villain to exploit. The sub­mis­sions were judge­ments, so they could not be proved wrong. And the banks sub­mit­ting them held posi­tions whose value depended on the answer.

What went wrong

Two kinds of pres­sure turned out to act on the number. In 2007 and 2008, a bank that sub­mit­ted a high bor­row­ing rate was telling the world it was in trouble. Some banks sub­mit­ted lower rates than they could really borrow at, pro­tect­ing their rep­u­ta­tions at the moment it mat­tered most. Separately, and for years, deriv­a­tives traders at some banks asked the col­leagues who made the sub­mis­sions to nudge them up or down a little on days it suited their posi­tions, and in some cases coor­di­nated with traders at other banks.

In June 2012 Barclays settled with British and American author­i­ties for a com­bined sum of around £290 million, and the inter­nal mes­sages pub­lished along­side made the mechan­ics painfully con­crete. Other banks fol­lowed with far larger set­tle­ments. The UK's Wheatley Review rec­om­mended in September 2012 that sub­mit­ting to and admin­is­ter­ing LIBOR become reg­u­lated activ­i­ties. ICE Benchmark Administration took over as admin­is­tra­tor in February 2014, and sub­mis­sions were anchored to trans­ac­tions wher­ever they existed.

It was not enough to save the number. The unse­cured inter­bank lending LIBOR was sup­posed to describe had shrunk to the point where most sub­mis­sions could not be tied to real trades. Regulators pushed markets to move to rates built from trans­ac­tions. The US dollar LIBOR panel ended on 30 June 2023.

What replaced it, and what it solves

The replace­ment for dollar LIBOR is SOFR, the Secured Overnight Financing Rate, pub­lished by the Federal Reserve Bank of New York since April 2018. Transactions go in, not answers: the rate on overnight loans secured by US Treasury secu­ri­ties in the repo market, taken as a volume-weighted median across trades that in 2026 have run to roughly three tril­lion dollars a day.

It solves the problem LIBOR died of. A median of that many real trans­ac­tions is very hard to push, and nobody has to guess. It also changed what the number describes. SOFR is overnight and secured against gov­ern­ment bonds, so it con­tains almost no bank credit risk. LIBOR, what­ever its faults, tried to describe what it cost a bank to borrow unse­cured for months. Every bench­mark reports a market, and first it picks which one.

Gold went through its own reform. For almost a century the London gold price was fixed by a small group of banks on a call, in a process dating from 1919. In May 2014 the UK reg­u­la­tor fined Barclays £26 million after a trader manip­u­lated the fix on one day to avoid paying a client. From 20 March 2015 the fix was replaced by the LBMA Gold Price, an elec­tronic auction admin­is­tered by ICE Benchmark Administration, run twice a day at 10:30 am and 3 pm London time, with bids and offers from direct par­tic­i­pants and an audit trail.

Judgement never leaves

Transactions are not the only answer, and in some places they are not an answer at all. The most famous stock index in the world, the S&P 500, is not a list of the five hundred largest American com­pa­nies chosen by formula. Its members are selected by a com­mit­tee at S&P Dow Jones Indices, apply­ing pub­lished eli­gi­bil­ity cri­te­ria and its own dis­cre­tion. Trillions of dollars in index funds buy what­ever that com­mit­tee adds.

Closing prices, the numbers that value most funds each evening, are increas­ingly set in a closing auction, a single match­ing at the end of the day into which a large share of index-fund trading is delib­er­ately con­cen­trated. The rules of that auction, how imbal­ances are pub­lished and when orders may be can­celled, decide the number that becomes the day's offi­cial price.

What a good bench­mark looks like

After LIBOR, reg­u­la­tors wrote down what they had learned. In July 2013 IOSCO, the inter­na­tional body of secu­ri­ties reg­u­la­tors, pub­lished prin­ci­ples for finan­cial bench­marks, and the European Union later turned similar ideas into law in its Benchmarks Regulation. The prin­ci­ples are unglam­orous and precise, and they read like a list of every­thing that went wrong.

A bench­mark should be anchored in observ­able trans­ac­tions wher­ever pos­si­ble. Its method­ol­ogy should be pub­lished, and changes to it con­sulted on. The admin­is­tra­tor should manage con­flicts of inter­est, keep records, and be subject to over­sight by some­body inde­pen­dent of the people who make sub­mis­sions. When data is thin, the bench­mark should say so and explain what it does instead. And it should be designed with the pos­si­bil­ity of its own end in mind, so that con­tracts which depend on it can find a replace­ment.

None of that guar­an­tees a good number. It moves the ques­tion from whether the people making a number are honest to whether the process makes dis­hon­esty dif­fi­cult and visible, which is the only ques­tion a user of a bench­mark can actu­ally check.

The same problem, in new archi­tec­ture

Blockchains repro­duce all of this in soft­ware. A lending pro­to­col has no way to see a price by itself. It must be told by an oracle, a system that reads venues off-chain, applies filters and weights, and writes a number on-chain that con­tracts then treat as fact. The oracle is an admin­is­tra­tor, its code is a method­ol­ogy, and its sources are a panel.

The failure modes are recog­nis­able. In October 2022 a trader bought up the thinly traded token of the Mango Markets pro­to­col on the venues its oracle read, watched the reported price mul­ti­ply, and bor­rowed against the inflated value, leaving with roughly $110 million. He was later con­victed of fraud; in 2025 a federal judge set the con­vic­tions aside. Whatever the law finally says about the trader, the oracle did what it was designed to do. It reported the price the market showed. The market it chose to read was small enough to own for a few minutes.

That is the thread through every case in this article. Nobody owns a number out­right. But some­body always decides which market it listens to, and what it ignores, and whether the people who can move it are also paid by it. Reading those rules sounds like home­work for spe­cial­ists. It is also the only way to know what a number is actu­ally telling you.

For nearly four decades some­body asked a ques­tion at 11 am every London morning, and the answers set the price of an enor­mous amount of money. Whenever a number matters, find out who is asking now.