Circuit Breakers In Commodity Markets

Michael Giberson

I’m puzzled. Earlier this year I was involved in a project examining natural gas price volatility, and as part of the background work I examined circuit breakers in futures contracts. As it turned out, the circuit breaker on the NYMEX natural gas futures contract was the most permissive by a wide margin.

Circuit breakers are market rules that trigger a temporary stop in trading when a price moves by more than a specified amount compared to the previous day’s close. In some cases, the circuit breaker simply prevents price movements beyond the limits without a stop in trading. For example:

  • Price movements on frozen pork bellies at the Chicago Mercantile Exchange are limited to 3 cents per pound up or down from yesterday’s closing price.
  • Prices movements on corn futures on the Chicago Board of Trade are limited to 20 cents per bushel up or down.
  • Price movements on NYMEX copper futures of 20 cents per pound will trigger a 15-minute suspension of trading. Trading will resume with the circuit breaker expanded to 40 cents above or below the previous day’s closing price.

As of March 24, when I compiled the information, these price limits represented a little over 3 percent for frozen pork bellies, about 9.5 percent for corn futures, and under 14 percent for copper.

In contrast, it would take a price movement for NYMEX natural gas futures of $3 per mmBTU, or more than 40 percent of the price as of March 24, to trigger a 5-minute trading suspension. Trading resumes with a circuit breaker expanded to $6 per mmBTU; hitting that limit triggers another 5-minute trading suspension and another $3 expansion of the circuit breaker. There is no maximum trading limit during any trading day.

At the cost of a single five-minute stoppage in trading, prices would be permitted to swing as much as $6 per mmBTU above or below yesterday’s closing price. Today’s price is about $6.30 for June delivery, meaning after a single trade stoppage prices could range from $0.30 to $12.30.

This is my puzzle: Such a limit is hardly relevant to market activity, so why does NYMEX bother to have the limit at all?

The academic literature on circuit breakers focuses mostly on stock market trading, not commodities futures. Some articles suggest that circuit breakers could be destabilizing, because traders fearing a trade stoppage may rush to trade before the stoppage, a rush to trade that itself may bring about the price movement triggering the stoppage. One article worried that trading halts would interfere with the continuous trading that promotes efficient price discovery.

Other articles point out that a trade stoppage may allow markets and brokers time to issue margin calls, or otherwise adjust to unusual trade volumes or price changes. In addition, to the extent that price limits reduce price risk, lower margin requirements may be allowed.

An interesting paper by Kodres and O’Brien in the 1994 American Economic Review suggests that price movement limits may be efficiency enhancing for hedgers, but not for speculators. Such a differential impact may explain the popularity of circuit breakers on commodity markets.

However, all of these articles assume that the circuit breaker is at least potentially binding. For all practical purposes, the circuit breaker on NYMEX natural gas futures contract doesn’t bind. So why does the NYMEX have such a limit on natural gas futures contracts?


20 thoughts on “Circuit Breakers In Commodity Markets

  1. I know nothing about such things, but off the top of my head, perhaps having a circuit breaker (effectual or not) in place is fashionable or wards off regulation and costs little to implement.

    Also, I believe the price of copper has nearly doubled in the last few years. If the copper circuit breaker triggered by an absolute price move perhaps it was also nearly irrelevant relatively recently.

  2. Mike L., the futures exchanges are intensely regulated whether they have circuit breakers on particular markets or not.

    Michael G., a better measure than % of yesterday’s closing price might be a volatility measure like average true range (ATR) (the difference between today’s high and low, adjusted for gap openings (yesterday’s close was outside today’s high-low range)). In pork bellies, 3 cents is about 1 ATR. In corn, 20 cents is about 3-5 ATR. In copper, 20 cents is about 5-6 ATR. In natural gas, $3 is about 20 ATR.

    Well I take that back–the relative difference in circuit breakers measured in ATR terms is comparable to the percentages of closing prices.

    As for why the circuit breakers differ so much….

    If Kodros and O’Brien are correct, one would expect price limits to be tighter in markets that have more hedging and less speculating. I don’t know the breakdown of commercials vs. speculators in pork bellies, corn, and natural gas, but my guess is speculators are heavier in pork bellies than in natural gas, which would be the opposite of the Kodros and O’Brien expectation. (Caveat: I haven’t read their work). You can look up the commercials vs. speculators report at the CFTC website if you’re interested.

    My guess for why the circuit breakers are set where they are is that floor traders in each market, back in the day when the exchanges were setting up circuit breakers, pushed for whatever limit most benefited them. E.g. maybe back in the day most pork belly floor traders traded a style that benefited from tight circuit breakers and most natural gas floor traders had a style that benefited from loose ones. Hmm, maybe–but why the very loose natural gas circuit breaker instead of none at all?

  3. I know nothing about the frozen pork belly market other than what I found on the CME website, so I can’t hazard much of a guess on the role of speculators there. Actually I do have the CFTC Committment of Traders (COT) data around here, so I can look this up.

    However, in the case of the NYMEX natural gas futures market, I have the CFTC COT data handy. It shows a fairly dramatic fall in the percentage of open interest held by commercials over the past five years (from about 90 percent in December 2000 to near 50 percent in February 2005). For comparison, from the April 1990 beginning of trading for the contract on NYMEX, commercials tended to hold an average of just under 90 percent of open interest. Coincidently, or not, the beginning of the fall roughly coincides with passage of the Commodity Futures Modernization Act of 2000.

    Actually, I like your thought of looking at the role played by floor traders. A circuit breaker may be a way for the floor traders to help “maintain an orderly market” during times of dramatic price moves (and typically higher volumes).

    I would guess that, in the case of natural gas, the rules-making folks at NYMEX wanted no circuit breaker at all, but thought that such a proposal would attract unfavorable regulatory attention. As an alternative, they just raised the limit high enough so as to accomplish the same result.

    [Two recent studies of investor participation in energy futures markets (by NYMEX and by the CTFC itself) tend to discount the usefulness of the COT data for analyzing the effects of ‘speculator’ involvement. Both studies rely upon internal data collections not publicly available. But that is a subject for another post.]

  4. Just FYI: currently the 15 min circuit breaker in Gold is $75 (at a Price of around $400, but 2 years back $200) and $1.50 in Silver – there are some other caveats, see the NYMEX site. Until these breakers were implemented in 1991, the NYMEX had price movement limits (Gold $10 to $25 and Silver $.20 to $.50) between 1975 and 1985 – I believe there were no limits between ’86 and ’91.
    During the period of the Hunt Silver manipulation (1979-80) these price limits were in effect for almost every day for 2 months. However, I did not find any other period when these price limits were reached or breakers were enacted as severely as during the Hunts.

    So why circuit breakers? I am by far no expert in commodities but perhaps, breakers are set at such extremes because margins are set fairly liberally in gas. Breakers, I believe, are set so that margin accounts can balance their deficits and if there is no immediate need for adjusting the margin account then breakers are not needed.

    Lastly, and this may be a stretch, but they may also tend to prohibit excessive manipulative behavior (such as hammering which has to be performed at a rapid pace and will cause such high price fluctuations).

  5. Re: Thomas G

    I had the gold and silver circuit breakers in the table I compiled a few months back. I would have posted the whole table but I don’t know how to render the table in Movable Type. As of late March, the NYMEX gold circuit breaker was equivalent to about an 18 percent price move, and the NYMEX silver circuit breaker was about 22 percent.

    For comparison, other NYMEX energy futures (crude oil, unleaded gasoline, heating oil) trade with circuit breakers in the 15 – 20 percent range. Corn and wheat futures on the CBOT tend to have price movement limits in the 8-10 percent range. Cattle, frozen pork bellies, and live hog futures trade on the CME with price movement limits in the 2-4 percent range. (Note in some cases the price movement limits are more complex, i.e. 2 cents per bushel if the contract is trading below 60 cents, but a 3 cent limit if the price is over 60 cents. That sort of thing.)

    Trading in Ethanol began on the CBOT in late March of this year, and as befits an ag-energy product the price movement limit falls between the energies and the grains at about 12 percent.

    There seem to be patterns here, but I’m not sure just what it all means.

    Re: Continuing the response to Raymund

    Looking at the CFTC COT report numbers for Frozen Pork Bellies, there is no apparent trend in commercial/non-commercial participation over the Jan 2000 – Feb 2005 period. (There is a lot more movement with commercials swinging from as high as about 75 percent to as low as 10 percent of reported positions — i.e. ignoring positions that fall below the CFTC reporting threshold — but unlike the near steady fall of the natural gas commercial participation, in frozen pork bellies the swings are all back and forth.)

    I think that the CFMA of 2000 exempted agricultural commodities from the general deregulation of OTC derivatives trading, which may account for the lack of any significant trends.

  6. A related question is: why are their price limits on the futures contracts, but not the option contracts? I remember standing in the cotton options ring when the cotton futures moved limit up. It was very painful. I was short and covered by buying deep in the money (no extrinic value) calls at a level over where the futures locked up from other locals. The market in synthetic futures was very, very illiquid: five contracts up, thirty points wide. The synthetic futures traded half a limit over where the futures were locked up.

    Something to consider, I believe those commodities which have fatter tailed distributions have no price limits or wide price limits. Do price limits truncate the tails of the distribution of those commodities which have them or are do those commodities naturally have thinner tails? You could test that by looking for trades in options (sometimes it was deeps, sometimes it was conversions–so know what to look for) that have no extrinsic value and took place past the future price limits to get an idea of what the maximum futures price should have been.

  7. I heard a remark when a particular market moved 20% in a day, one broker told his clerks, “Don’t pick up the phone!”.
    Brokers refused to answer the phone in 1987 stock market crash, got in trouble and the SOES system was created as a result.

    A five minute break for natural gas futures is for the floor brokers to collect themselves and the customer orders without having to deny customers access to the market. Without it, you would have brokers not accepting customer orders leading to the death of a floor based trading system.

  8. I heard a remark when a particular market moved 20% in a day, one broker told his clerks, “Don’t pick up the phone!”.
    Brokers refused to answer the phone in 1987 stock market crash, got in trouble and the SOES system was created as a result.

    A five minute break for natural gas futures is for the floor brokers to collect themselves and the customer orders without having to deny customers access to the market. Without it, you would have brokers not accepting customer orders leading to the death of a floor based trading system.

  9. I’m inclined to agree with Mike G: the exchange found it would be politicallly useful to have a circuit-breaker in natural gas to avoid other regulatory impositions which might be worse. Economically, it makes little sense, but politically it makes a tremendous amount of sense.

  10. I’m inclined to agree with Mike G: the exchange found it would be politicallly useful to have a circuit-breaker in natural gas to avoid other regulatory impositions which might be worse. Economically, it makes little sense, but politically it makes a tremendous amount of sense.

  11. Breaking the commodities market online

    With the popular hype going around the Freakonomics book and blog, I’ve been adding a lot of new sites to my RSS reader. Knowledge problem, by Northwestern Prof Lynne Kiesling and Michael Giberson, is probably my new favorite read. Especially…

  12. foreign currency trading

    To create a website from the ground up may be very confusing at first on how getting traffic. You know that your success comes down to your ability to generate traffic to your web site. Remember, If you don’ t have traffic to your site, it’ s similar t…

  13. Free Forex Demo Account

    There are many good books and references that are relevant to club investing. Some suggested readings might include Peter Lynch’ s One Up on Wall Street , Beating the Street , and Learn to Earn , Robert G. Hagstrom’ s The Warren Buffet Way , and Graham…

Comments are closed.