Natural Gas Trading: Right Now, It's the Wild, Wild West

In my last article, I discussed two of the major factors to this week’s run-up in natural gas - Operation Flow Orders (OFOs) and pre-configured stop orders being hit. Here, I’d like to take a look at some of the other concurrent distortions in the natural gas market.

UNG Rolling Effect

After a nearly year-long run, the United Natural Gas Fund's (UNG) assets shot past $4.5 billion at one point. Just in a two-week period in May, the ETF's assets doubled despite lagging performance (UNG share price has retreated about 55% this year.) The fund has been rumored at times to hold as much as 80% of open interest in the NYMEX front-month contract.

As reported by FT.com, this is a rollover week for UNG. The fund also has just become an active buyer of fresh futures again. Essentially, the UNG roll is another reason for the natural gas surge this week on lack of market fundamental support.


On many days, UNG has dominated in the natural gas market, but a victim of its size and structure, UNG has also become a large and predictable player that has to roll because it has no capability of taking the contract to expiry.

The fund’s swap counterparties know just when and how much it has to roll; and everything will depend on the hedging of counterparties. Accordingly, UNG usually fails to fetch a competitive price and the contango between the front and second month increases at each roll. For this reason, the fund almost always underperforms the natural gas futures it is supposed to track (Fig. 1, click to enlarge).

NYMEX Natgas Bubble

What's more, UNG said it plans to restart new issues on Sep. 28, which is also the day that the October contract expires. Trading wise, this not only means that the Oct/Nov contango will widen during the roll this week, but also that during October there could be another upside volatility to natural gas outright prices as the UNG could come back to buy outright November contracts.

Compounding the “UNG rolling effect” is that CME Group (CME) just announced much more aggressive position limits on futures contracts, and this may have prompted some short-covering. Meanwhile, Goldman Sachs (GS) recently predicted that natural gas prices will triple by this winter while a speaker at a Barclays energy conference said natural gas is the ‘trade of the year.’ With natural gas at the confluence of all these concurrent events distorting the already volatile market, a NYMEX natgas bubble as described by the Schork Report seems inevitable.

UNG Investment Risks

Though UNG has become a very popular vehicle for investors and traders to participate in the futures market, the fund has been trading at a sharp premium to its underlying net asset value (NAV) since Aug. 12, when it announced that it couldn't issue new shares because of limits on how many natural gas contracts it can buy.

The share was trading at a 16% premium to NAV on Aug. 21, but now that's down to around 4%. The premium along with the share price should continue to fall from now till the new shares get issued. As such, short interest for UNG surged 135% to 30.9 million shares in the two-week period ended Aug. 31 (Table 1, click to enlarge).


In addition, UNG itself states there is no longer any predictability to when it feels like being in an issuing cycle and when it doesn’t. This predictability issue is noteworthy enough to earn a sell, sell, sell from Jim Cramer, the host of Mad Money on CNBC.

Caveat Emptor

Therefore, from all indications, retail investors dabbling in energy markets better take heed, the above being proof of just how complex and volatile the market can be. I would agree with Cramer’s call to sell if you bought UNG at a premium to its NAV before it drops even further. Meanwhile, new investors should stay away from UNG.

For now, the best way to play the natural gas market is probably buying natural gas producers such as Chesapeake Energy Corporation (CHK) and EOG Resources, Inc. (EOG) on the dip. Since these producers typically all have aggressive hedging programs in place to protect future production, investors could benefit from their market expertise without the complexity and risk as investing in UNG.

Disclosure: No Positions

Comments

[Ricard]

"UNG has also become a *large and predictable player* that has to roll because it has no capability of taking the contract to expiry. "

Excellent article. That was my main concern with UNG. You've hit the nail on the head - it's not just the contango, but WHY the contango exists that should really give UNG investors pause. Essentially, UNG is a major cause of the contango that is making it an unattractive investment to begin with. I was wondering if futures traders were taking advantage of this situation - I would myself, but I am far too inexperienced and have yet to take the time to dabble in (for me) a new vehicle like commodity futures.

IMHO the real question is whether the NG P&Es will ever dip:

"For now, the best way to play the natural gas market is probably buying natural gas producers such as Chesapeake Energy Corporation (CHK) and EOG Resources, Inc. (EOG) on the dip."

Again, thanks for the good read. Best explanation to date on what is making UNG flop.

2009 Sep 18 04:07 AM

[Aricool]

I agree, fantastic article. The best NG trading market mechanics/dynamics I've come across. Thanks!

I have a couple futures trading newbie question in re "UNG roll is another reason for the natural gas surge this week". My first is- I expected a roll to do the opposite and actually depress the front prices because UNG would have to sell front a large volume of month contracts, and boost the next month b/c UNG would be buying a large volume of next month contracts, However, you say it works to the opposite effect. Yet, I notice that NG front month prices seem to always dive this year in the last week of the contract. If your mechanics was correct the last week should always show a pattern of artificial support or buoyancy.

Another question I have about it is that UNG (and ETNs) has been using swaps to track NG prices. I have not research how/if such swaps can move NG prices; however, since they are traded notes and futures contract I am unclear if there is an effect.

Again, excellent article.

Thanks,
Ari

2009 Sep 18 04:47 AM

[Aricool]

I have a question about the (slightly) lower than expected injections the past few weeks. You mention (and I've read elsewhere) it is b/c of OFO; however, how we be sure that it is not also due to an up tick in industrial demand since it coincides with a higher PMI and (artificial) upticks in auto and aluminum production (very NG intensive), or switching from coal to NG, or LNG being diverted away from the US when NG is < $3, etc., etc.

Do you have any thoughts on this?

thanks again,
Ari

2009 Sep 18 04:52 AM

[sst381106]

This article makes no sence. Small investors without significant experience trading natural gas should not invest in UNG or nat gas futures at this point.

Author stays that "UNG roll is another reason for the natural gas surge this week on lack of market fundamental support". UNG roll is always BEARISH event, when one of the largest nat gas players forced to sell large number of current month futures and predetermined and published schedule. Public schedule of the roll allows other market participants to frontrun the roll and position themselves on the curve before and after the roll. When short positions build in anticipation of UNG roll become too large and too obvious like in happened in September, these short positions become easy target for both speculators and traders working for UNG. Short trade in UNG and nat futures became even more crowded due to UNG trading at premium to NAV and attracted a lot of arbitrage players in addition to typical nat gas market participants. Upside action started hitting stops and blowing off shorts which resulted in upside action in nat gas prices.

2009 Sep 18 05:08 AM

[sst381106]

Swaps perform the same fucntion as nat gas futures and allows UNG to increase long exposure to nat gas. If UNG unable to buy additional nat gas futures and at the same time third party ( in case of UNG its CS) has excessive temporary long exposure to nat gas futures, third party instead of selling/hedging long exposure will sign swap agreement with UNG tracking performance of nat gas futures. Swaps should be correlated with UNG direct nat gas positions, because only net increase will in general be a bullish sign.

2009 Sep 18 05:16 AM

[Ricard]

I'd reread the author's statement about contango:

"Trading wise, this not only means that the Oct/Nov contango will widen during the roll this week, but also that during October there could be another upside volatility to natural gas outright prices as the UNG could come back to buy outright November contracts. "

You're right...it's bearish for the current month, but bullish for the next month as they begin huge purchases, thus widening contango - I understand that this is one of the main reasons why UNG is a flawed vehicle...it's subject to other players gaming their system.

Regarding the rest of your comment, I am certainly one of the 'inexperienced', so I have nothing else to add.


On Sep 18 05:08 AM sst381106 wrote:

> Author stays that "UNG roll is another reason for the natural gas
> surge this week on lack of market fundamental support". UNG roll
> is always BEARISH event, when one of the largest nat gas players
> forced to sell large number of current month futures and predetermined
> and published schedule.

2009 Sep 18 07:48 AM

[Ron2008]

Ricard, you also need to look at what happened this week. The spread for Oct/Nov NG narrowed during the roll. It went from 1.051 on the 11th to 0.997 on the 17th. Thus the contango narrowed. Opposite of what you and the author said would happen.

On Sep 18 07:48 AM Ricard wrote:

> I'd reread the author's statement about contango:
>
> "Trading wise, this not only means that the Oct/Nov contango will
> widen during the roll this week, but also that during October there
> could be another upside volatility to natural gas outright prices
> as the UNG could come back to buy outright November contracts. "
>
>
> You're right...it's bearish for the current month, but bullish for
> the next month as they begin huge purchases, thus widening contango
> - I understand that this is one of the main reasons why UNG is a
> flawed vehicle...it's subject to other players gaming their system.
>
>
> Regarding the rest of your comment, I am certainly one of the 'inexperienced',
> so I have nothing else to add.

2009 Sep 18 09:28 AM

[learnmoney]

You have done it again and got it wrong.
NG is cheap energy and if you take away the daily or very short dealing positions and without going into mumbo jumbo you have an interesting speculative investment.
Keep it simple please and by the way take a position + or minus I don't mind.
I am long at 10 and 12.5 per share but one is a dealing position and the other is a 6/12 month view position.

2009 Sep 18 09:44 AM

[Sagan100]

There are a lot of talk on bloggs and articles on UNG and others like CHK but never anything about GAZ. Any views on GAZ on premiums and rolls etc?

2009 Sep 18 11:12 AM

[LKofEnglish]

Interesting. So are prices in nat gas about to soar or collapse? Still a big believer in the natural gas utility sector with Wisconsin Energy as my number one pick. Uncertainty in the market place means "secure the contract with the end user."

2009 Sep 18 11:16 AM

[Jeffry Selvig]

UNG has ALREADY rolled into the November contract, and contrary to the author's assertion that Oct/Nov contango would widen, it actually shrunk.

2009 Sep 18 11:25 AM

[NGAnalyst]

A few comments ---

1. Contrary to the widening contango argument due to UNG roll, we actually have just seen the opposite.

2. It is hard to be sure what impact UNG roll has on prices in general, as it sells and buys at the same time.

3. I actually don't see the UNG over NAV premium necessarily going away that fast due to reasons: (1) retail investors try to jump in to catch the rally as there is generally lack of better alternatives (2) even UNG decides to offer new shares, where is the money going to be put at due to CFTC's limits? I see the premium being diluted, but not like completely removed.

2009 Sep 18 12:01 PM

[Aricool]

If OFOs were issued prior to Labor Day, which cause excess NG flows to be dumped onto the spot market, thus dropping the price, then why the bullish move on the NG price? That is, if the excess production dumped into the spot market plunged the price then that would seem to clearly indicate that industrial demand (the main swing consumer now) did not up tick and is still too low to keep NG from plummeting. If this is true, the OFO is actually far more bearish info than the bad storage situation. Continuing this line of thought, then, I am left with the question about the UNG roll. I don't like to assume coincidences when there is a causal link, and in this case it is strikingly curious that UNG's September 14, 2009 through September 17, 2009 roll exactly coincided with the massive NG price surge from 3 to 3.9. It even seems as if traders started preparing for this (e.g., covering shorts and buying long contracts) by running the price up from 2.6 to 3 the week ahead of the UNG roll. Another "coincidence" I cannot ignore is Goldman releasing a hugely bullish report smack ahead of the UNG roll as if to give license to the momentum traders to jump in on the trade and guarantee all the short stops get blown. I recall a few weeks ago there was an injection of ~52 bcf, about 10% below expectations, and that did nothing to stop the plunge NG price down another 20% or so. This all seems completely like technical market manipulation. I still don't understand, though, why UNG massively selling the front month (maybe 30+% of contract volume) would boost the front month's price.

Comments/explanations

2009 Sep 18 02:08 PM

[NGAnalyst]

Aricool, good questions asked.
My way of understanding/speculating is this --- OFOs are seen as a bullish factor in the sense that they force wellhead prices really low, thus forcing producers to shut in productions. You would think that futures are driven by spot prices, but practically they could be out of whack due to locational issues. I don't believe UNG (with no fresh money) would consistently move gas prices one way or another. I'd rather believe that GS's recommendations work better in this regard. Lastly, as I said in one of my earlier posts, the only "reasoning" behind the drive all the way down to $2.5 is that we needed bring down prices low enough to do two things: (1) tile S/D balance to make sure to avoid a storage box, and thus to (2) form THE bottom.

2009 Sep 18 03:06 PM

[Aricool]

Can anyone answer this question in re Nat gas E & P stocks:

that is, with market manipulation jacking up NG contract prices over the whole '09 curve, the E & P companies (who know the real supply/demand/timing equation) will surely hedge 100% of all '09 possible production. The rig count went up since July so they may even be bringing more supply online and selling into such additional hedged contracts. Keeping in mind that they all (largest to smallest) need more cash flow to stay above water (or even survive for the small ones) it seems like a sure bet that the E & Ps will max out what ever storage, spot demand, or hedged futures demand will profitably permit. With E & P cost of production well under $3 for most and NG futures for '09 now between $4.5-5.3 they are massive profitable. They know that NG prices into Nov. can easily go to 2 or below (have not even mentioned the LNG show stopper factor).

So, with all the above backdrop, the question I have is related to price -vs- storage dynamic; that is, can the NG market be manipulated like that of oil and keep NG prices very high (e.g., above 3.5) until the storage is almost or completely full, or is it really driven by suppliers and producers setting prices based on supply/demand. The way the Sep. contract immediately fell from 4 to 2.5 argues for a supply/demand driven market, and the recent move from 2.5 to 3.9 argues for a manipulation driven market. Just before the month's contract ends does NG front month price always have to reflect true supply/demand dynamics b/c the speculators have roll/sell out there position to avoid taking delivery. What am I missing here?

thanks!
Ari

2009 Sep 18 03:14 PM

[Dian L. Chu]

Below was excerpted from the US EIA Sep report www.eia.doe.gov/emeu/s...
"Industrial natgas consumption declined by 12% YoY in the first 6 months of 2009. EIA expects natural gas consumption will increase slightly in the commercial and industrial sectors in 2010 due to improved economic conditions and low prices. The anticipated addition of new coal-fired generating capacity and rising natural gas prices limits the potential for significant increases beyond the forecast 2009 level in natural gas consumption by electric generators."
So, the increase is not expected to give too significant boost to the slumping demand. However, this could change depending on the economic recovery pace.
Thanks for your comment.

On Sep 18 04:52 AM Aricool wrote:

> I have a question about the (slightly) lower than expected injections
> the past few weeks. You mention (and I've read elsewhere) it is b/c
> of OFO; however, how we be sure that it is not also due to an up
> tick in industrial demand since it coincides with a higher PMI and
> (artificial) upticks in auto and aluminum production (very NG intensive),
> or switching from coal to NG, or LNG being diverted away from the
> US when NG is < $3, etc., etc.
>
> Do you have any thoughts on this?
>
> thanks again,
> Ari

2009 Sep 18 04:13 PM

[PairsTrader]

Yea, right..
If the author takes a positon, the you will turn around acusing her of pumping existing position.
As for keeping it simple, let's see you publish an article here, (i.e., NOT Instablog) ??
There's no pleasing everyone, so sStay your course and keep up with the great work, Dian.

On Sep 18 09:44 AM learnmoney wrote:

> You have done it again and got it wrong.
> NG is cheap energy and if you take away the daily or very short dealing
> positions and without going into mumbo jumbo you have an interesting
> speculative investment.
> Keep it simple please and by the way take a position + or minus I
> don't mind.
> I am long at 10 and 12.5 per share but one is a dealing position
> and the other is a 6/12 month view position.

2009 Sep 18 04:21 PM

[Ron2008]

Speculators have been massively short NG futures all summer. The Oct NG contract had the highest open interest of any contract this year. Peaked at over 186,000. Probably a lot of specs using Oct for their short contracts.

So I suspect that the current rise in front month prices is due to the specs unloading their short Oct contracts before expiration. They did a lot of it during the ETF roll so that they would have someone selling large quantities of contracts. But they had far more shorts than the ETFs had longs, thus the price rise. On Friday, after the UNG roll was over, the Oct NG was up 0.32 but Nov was only up 0.20. More narrowing of the contango.

Open interest (OI) in all NG contracts dropped by 24,888 on Tuesday through Thursday. Shorts (specs) getting out and longs (UNG selling more Octs than the number of Novs they were buying because of the price difference) getting out. Thus the drop in OI.

As far as manipulation, the Producer/Users have less than 16% of the NG futures contracts. That is the lowest of all contracts reported by the CFTC in their DCOT report. Worse than oil. That makes it very easy for futures to not have to reflect current cash NG prices because the majority of the trading is specs with specs. A very low percentage of contracts delivered.

Oct NG OI will go from 186,000 contracts to less than 2,000 contracts where delivery is made/taken. Aug had less than 500 deliveries.

Manipulation possible? You betcha.

On Sep 18 03:14 PM Aricool wrote:

> Can anyone answer this question in re Nat gas E &amp; P stocks:<br/>
>
> that is, with market manipulation jacking up NG contract prices over
> the whole '09 curve, the E &amp; P companies (who know the real supply/demand/timing
> equation) will surely hedge 100% of all '09 possible production.
> The rig count went up since July so they may even be bringing more
> supply online and selling into such additional hedged contracts.
> Keeping in mind that they all (largest to smallest) need more cash
> flow to stay above water (or even survive for the small ones) it
> seems like a sure bet that the E &amp; Ps will max out what ever
> storage, spot demand, or hedged futures demand will profitably permit.
> With E &amp; P cost of production well under $3 for most and NG futures
> for '09 now between $4.5-5.3 they are massive profitable. They know
> that NG prices into Nov. can easily go to 2 or below (have not even
> mentioned the LNG show stopper factor).
>
> So, with all the above backdrop, the question I have is related to
> price -vs- storage dynamic; that is, can the NG market be manipulated
> like that of oil and keep NG prices very high (e.g., above 3.5) until
> the storage is almost or completely full, or is it really driven
> by suppliers and producers setting prices based on supply/demand.
> The way the Sep. contract immediately fell from 4 to 2.5 argues for
> a supply/demand driven market, and the recent move from 2.5 to 3.9
> argues for a manipulation driven market. Just before the month's
> contract ends does NG front month price always have to reflect true
> supply/demand dynamics b/c the speculators have roll/sell out there
> position to avoid taking delivery. What am I missing here?
>
> thanks!
> Ari

2009 Sep 18 10:33 PM

[Aricool]

Dian, thanks for your reply. However, they were just forecasting a soft 2010, and indicated the known drop in '09 1H- no mention of recent industrial NG demand trends. Thus, seems to me that this does not rule out the lower injection was due to higher demand and/or lower production, and not the OFO issue.

Interesting that they were assuming a 57 bcf/wk injection to get max storage by end Oct, but we've been seeing upper 60's. That seems to get there a week earlier.

BTW, seems like the EIA is expecting an U-shaped recovery. I wonder how much we can trust their price forecasts when they don't even include LNG as a major factor keeping 2010 prices low. They just mention the shale plays- "However, upward price pressure next year is limited by the sensitivity of natural gas use in the electric power sector to higher natural gas prices and continued expansion of U.S. natural gas production from shale formations" They only say LNG grows in 2H '09 will double next year, but no mention of price impact.

Also, they talk of a planned 10GW of coal power plants, yet, I would expect the fear of any kind of cap-and-trade would strongly push new power plants to be gas fired, or at least dual capable.

Ari-

On Sep 18 04:13 PM Dian L. Chu wrote:

> Below was excerpted from the US EIA Sep report www.eia.doe.gov/emeu/s...
>
> "Industrial natgas consumption declined by 12% YoY in the first 6
> months of 2009. EIA expects natural gas consumption will increase
> slightly in the commercial and industrial sectors in 2010 due to
> improved economic conditions and low prices. The anticipated addition
> of new coal-fired generating capacity and rising natural gas prices
> limits the potential for significant increases beyond the forecast
> 2009 level in natural gas consumption by electric generators."<br/>...
> the increase is not expected to give too significant boost to the
> slumping demand. However, this could change depending on the economic
> recovery pace.
> Thanks for your comment.
>
> On Sep 18 04:52 AM Aricool wrote:

2009 Sep 19 02:50 AM

[Aricool]

NGAnalyst:


On Sep 18 12:01 PM NGAnalyst wrote:

> A few comments ---
>
> 1. Contrary to the widening contango argument due to UNG roll, we
> actually have just seen the opposite.
>
I agree.

> 2. It is hard to be sure what impact UNG roll has on prices in general,
> as it sells and buys at the same time.
>
the other variable is how they play their swap contracts on the ICE vs futures on the CME.

> 3. I actually don't see the UNG over NAV premium necessarily going
> away that fast due to reasons: (1) retail investors try to jump in
> to catch the rally as there is generally lack of better alternatives
> (2) even UNG decides to offer new shares, where is the money going
> to be put at due to CFTC's limits? I see the premium being diluted,
> but not like completely removed.

As far as I can tell, UNG is a ponzy scheme. They hide/spread the NAV losses w/ the new money that comes in used to buy the new contracts. I read that they trying to get limit exemptions and they plan to go temporarily over limits so that their fund does not decay to zero w/ this steep 30% contango. Seems like a game of chicken with the CFTC. BTW, contrary to popular belief the CEO of USO (who does UNG too) said DXO did not close b/c of CFTC limits, but b/c the ETN swap contracts premium got too expensive to be profitable.

Ari-

2009 Sep 19 02:59 AM