Nov. 26 (Bloomberg) -- Every business day in London, five banks meet to set the price of gold in a ritual that dates back to 1919. Now, dealers and economists say knowledge gleaned on those calls could give some traders an unfair advantage when buying and selling the precious metal.
The U.K. Financial Conduct Authority is scrutinizing how prices are set in the $20 trillion gold market, according to a person with knowledge of the review who asked not to be identified because the matter isn’t public. The London fix, the benchmark rate used by mining companies, jewelers and central banks to buy, sell and value the metal, is published twice daily after a telephone call involving Barclays Plc, Deutsche Bank AG, Bank of Nova Scotia, HSBC Holdings Plc and Societe Generale SA.
The process, during which gold is bought and sold, can take from a few minutes to more than an hour. The participants also can trade the metal and its derivatives on the spot market and exchanges during the calls. Just after the fixing begins, trading erupts in gold derivatives, according to research published in September. Four traders interviewed by Bloomberg News said that’s because dealers and their clients are using information from the talks to bet on the outcome.
“Traders involved in this price-determining process have knowledge which, even for a short time, is superior to other people’s knowledge,” said Thorsten Polleit, chief economist at Frankfurt-based precious-metals broker Degussa Goldhandel GmbH and a former economist at Barclays. “That is the great flaw of the London gold-fixing.”
The U.K. capital is the biggest center for gold trading in the world, according to the London Bullion Market Association, which said more than $33 billion changed hands there each day in 2012, exceeding the $29 billion of futures traded on Comex, the New York commodities exchange, data compiled by Bloomberg show. Financial instruments including cash-settled swaps and options are priced off the London fix, according to the LBMA website.
In private meetings this year, the U.S. Commodity Futures Trading Commission, which regulates derivatives, discussed reviewing how gold prices are set, according to a person with knowledge of the talks. The FCA review is preliminary and not a formal investigation, another person said. The people wouldn’t say what’s being looked at or if regulators suspect wrongdoing.
Participants on the London call can tell whether the price of gold is rising or falling within a minute or so, based on whether there are a large number of net buyers or sellers after the first round, according to gold traders, academics and investors interviewed by Bloomberg News. It’s this feature that could allow dealers and others in receipt of the information to bet on the direction of the market with a high degree of certainty minutes before the fix is made public, they said.
“Information trickles down from the five banks, through to their clients and finally to the broader market,” Andrew Caminschi, a lecturer at the University of Western Australia in Perth and co-author of a Sept. 2 paper on trading spikes around the London gold fix published online in the Journal of Futures Markets, said by phone. “In a world where trading advantage is measured in milliseconds, that has some value.”
Pat McFadden, an opposition Labour lawmaker who sits on Parliament’s Treasury Select Committee, said British regulators need to probe any possible abuses by dealers.
“The gold market is hugely influential, and there needs to be public trust in the gold price,” McFadden said in an interview. “Question marks have been raised about the benchmark price of gold, and it’s important that regulators investigate.”
Scrutiny of the gold market is taking place as the price of the metal has fallen 26 percent this year, heading for the first annual drop since 2000. Barrick Gold Corp., the world’s biggest gold producer, plans to sell, close or curb production at almost half of its mines, and billionaire John Paulson’s PFR Gold Fund lost $630 million since the end of December, according to a person briefed on the returns.
The price of gold at today’s London afternoon fix was $1,247.50 an ounce, down from $1,693.75 on Jan. 2.
Regulators are looking into how benchmarks are set and governed across the financial system after five firms including Barclays and Royal Bank of Scotland Group Plc were fined a combined $3.7 billion for rigging the London interbank offered rate, or Libor. Investigators from Switzerland to Hong Kong are probing currency markets after Bloomberg News reported in June that traders communicated with each other and timed trades to influence foreign-exchange benchmarks and maximize profits.
There’s no evidence that gold dealers sought to manipulate the London fix or worked together to rig prices, as traders did with Libor. Even so, economists and academics say the way the benchmark is set is outdated, vulnerable to abuse and lacking any direct regulatory oversight.
“This is one of the most concerning fixings I have seen,” said Rosa Abrantes-Metz, a professor at New York University’s Stern School of Business whose 2008 paper, “Libor Manipulation?” helped spark a global probe. “It’s controlled by a handful of firms with a direct financial interest in where it’s set, and there is virtually no oversight -- and it’s based on information exchanged among them during undisclosed calls.”
London Gold Market Fixing Ltd., a company controlled by the five banks that administers the benchmark, has no permanent employees. A call from Bloomberg News was referred to Douglas Beadle, 68, a former Rothschild banker, who acts as a consultant to the company from his home in Caterham, a small commuter town 45 minutes south of London by train. Beadle declined to comment on the benchmark-setting process.
Spokesmen for Barclays, Deutsche Bank, HSBC and Societe Generale declined to comment about the London fix or the regulatory probes, as did Chris Hamilton, a spokesman for the FCA, and Steve Adamske at the CFTC.
Joe Konecny, a spokesman for Bank of Nova Scotia, wrote in an e-mail that the Toronto-based company has “a deeply rooted compliance culture and a drive to continually look toward ways to improve our existing processes and practices.”
Stewart Murray, chief executive officer of LBMA, which represents the gold and silver markets and publishes the results of the fix on its website, declined to comment, saying the group has “no jurisdiction or responsibility” for the process or its administration.
A spokesman for the association, Aelred Connelly, said Nov. 22 that the group is reviewing its own benchmarks to see whether they conform to guidelines set by the International Organization of Securities Commissions in July. Those include making prices based on “observable” deals where possible. The LBMA oversees gold forward offered rates, which reflect bullion borrowing costs for different durations and are used in loan agreements.
The fix dates back to September 1919, less than a year after the end of World War I, when representatives from five dealers met at Rothschild’s office on St. Swithin’s Lane in London’s financial district. It was suspended for 15 years, starting in 1939. While Rothschild pulled out in 2004 and the discussions now take place by telephone instead of in a wood-paneled room at the bank, the process remains much the same.
At the start of the call, the designated chairman -- the job rotates annually among the five banks -- gives a figure close to the current spot price in dollars for an ounce of gold. The firms then declare how many bars of the metal they wish to buy or sell at that price, based on orders from clients as well as their own account.
If there are more buyers than sellers, the starting price is raised and the process begins again. The talks continue until the buy and sell amounts are within 50 bars, or about 620 kilograms, of each other. The procedure is carried out twice a day, at 10:30 a.m. and 3 p.m. in London. Prices are set in dollars, pounds and euros. Similar gauges exist for silver, platinum and palladium.
The traders relay shifts in supply and demand to clients during the calls and take fresh orders to buy or sell as the price changes, according to the website of London Gold Market Fixing, which publishes the results of the fix.
Bank of Nova Scotia provides clients with updates as the fixing proceeds through a page it makes available through Thomson Reuters Corp. terminals. Thomson Reuters itself only receives and publishes the official fixing price after the call has finished, said Kate Reid, a spokeswoman for the company. Konecny, the Nova Scotia spokesman, didn’t provide any details about the bank’s service. Caminschi, the University of Western Australia professor, said the information on Nova Scotia’s feed is delayed and often incomplete.
Bloomberg LP, the parent company of Bloomberg News, competes with Reuters in providing news and information as well as currency-trading systems.
David Govett, head of precious metals at Marex Spectron Group Ltd., a closely held commodity broker in London, said the benchmark gives clients an opportunity to buy or sell large amounts of gold in a single transaction anonymously, without having to turn to the futures market.
“The fix is a very efficient way of doing it,” he said. “It’s very open, it’s very transparent and it’s a good thing.”
A trader at one of the banks that sets the price defended the process, saying it’s structured to minimize opportunities to exploit the difference between the spot and fixing price of gold. He asked that neither he nor his firm be identified because he wasn’t authorized to speak publicly.
Caminschi and Richard Heaney, a professor of accounting and finance at the University of Western Australia, analyzed two of the most widely traded gold derivatives: gold futures on Comex and State Street Corp.’s SPDR Gold Trust, the largest bullion-backed exchange-traded product, from 2007 through 2012.
At 3:01 p.m., after the start of the call, trading surged to 47.8 percent above the average for the 20-minute period preceding the start of the fix and remained 20 percent higher for the next six minutes, Caminschi and Heaney found. By comparison, trading was 8.7 percent higher than the average a minute after publication of the price. The results showed a similar pattern for the SPDR Gold Trust.
“Intuitively, we expect volumes to spike following the introduction of information to the market” when the final result is published, Caminschi and Heaney wrote in “Fixing a Leaky Fixing: Short-Term Market Reactions to the London P.M. Gold Price Fixing.” “What we observe in our analysis is a clustering of trades immediately following the fixing start.”
The researchers also assessed how accurate movements in gold derivatives were in predicting the final fix. Between 2:59 p.m. and 3 p.m., the direction of futures contracts matched the direction of the fix about half the time.
From 3:01 p.m., the success rate jumped to 69.9 percent, and within five minutes it had climbed to 80 percent, Caminschi and Heaney wrote. On days when the gold price per ounce moved by more than $3, gold futures successfully predicted the outcome in more than nine out of 10 occasions.
“Not only are the trades quite accurate in predicting the fixing direction, the more money that is made by way of a larger price change, the more accurate the trade becomes,” Caminschi and Heaney wrote. “This is highly suggestive of information leaking from the fixing to these public markets.”
For derivatives traders, the benefits are clear: A dealer who bought 500 gold futures contracts at 3 p.m. and knew the fix was going higher could make $200,000 for his firm if the price moved by $4, the average move in the sample. While the value of 500 contracts totals about $60 million, traders may buy on margin, a process that involves borrowing and requires placing less capital for the bet. On a typical day, about 4,500 futures contracts are traded between 3 p.m. and 3:15 p.m., according to Caminschi and Heaney.
The trader at a fixing bank said there’s little money to be made from buying and selling gold derivatives during the process because information from the call is disseminated into the wider market so quickly. Arbitrage opportunities also are limited because the chairman will adjust the price during the call if there are moves in the spot and futures markets, he said.
Govett, at Marex Spectron, said it’s common for people to try to exploit the difference between the current price of gold derivatives and the final fix.
In terms of timing and money available from such arbitrage, “it is quite small, but you’d be amazed at the amount of people who try to do it,” said Govett, a trader for 30 years.
Abrantes-Metz, who helped Iosco formulate its guidelines, said the gold fix’s shortcomings may stretch beyond giving firms and clients access to privileged information.
“There is a huge incentive for these banks to try and influence where the benchmark is set depending on their trading positions, and there is almost no scrutiny,” she said.
Abrantes-Metz said the gold fix should be replaced with a benchmark calculated by taking a snapshot of trading in a market where $19.6 trillion of the precious metal circulated last year, according to CPM Group, a New York-based research company.
“There’s no reason why data cannot be collected from actual prices of spot gold based on floor or electronic trading,” she said. “There’s more than enough data.”