The Shale Gale unleashed an absolutely remarkable transformation in the North American* gas supply picture. The figures are familiar: US lower-48 dry natural gas production increased by 15 billion cubic feet (Bcf) per day in five years, or about 30 percent. Natural gas prices fell from in excess of $8 per million British thermal units (MMBtu) to less than $4 per MMBtu - and for a time less than $3 per MMBtu.

As unthinkable as these results were just a few years ago, today they are greeted almost with a yawn. Like, that's soooo yesterday. Today's news is that we have plans for that gas: big plans, and many of them. These include, in no particular order, petrochemical plants, natural gas vehicles, electric power, exports to Mexico, and LNG exports, not to mention the occasional gas-to-liquids project.

IHS projects that, by 2020, natural gas demand in North America will have increased by almost 14 Bcf per day over the 2013 level, and by over 21 Bcf per day beyond the 2008 level - the last pre-coal-displacement and (mostly) pre-recession year - when natural gas demand in North America totaled 71.7 Bcf per day.

These are large numbers. The pace of demand increase from 2013 to 2020 is about 2 Bcf per day each year. Much of the growth from 2008 to 2013 was attributable to coal displacement - the natural gas demand created when the effective cost of natural gas drops below the cost of coal per unit of electric power generated, thus "displacing" coal from the supply stack. In 2013, this accounts for about 6 Bcf per day of natural gas demand. As prices rise with demand growth, coal displacement will fall, and will thus make supply available to the market, at a price. Without the reduction in coal displacement, the pace of demand increase would be 3 Bcf per day each year.

But domestic demand is only part of the story. Exports are growing as well. By 2020, IHS projects that North American LNG exports will reach 5.3 Bcf per day and exports to Mexico, via pipeline, will reach 3.1 Bcf per day. Combined, this means that exports will increase by 6.2 Bcf per day between 2013 and 2020. The growth in exports offsets the reduction in demand due to coal displacement.

Supply will have to grow by 3 Bcf per day each year for the rest of the decade to match the projected demand growth. At first glance, that actually seems manageable enough. After all, the pace of supply growth in the early years of the Shale Gale was about 3 Bcf per day each year in the US. Canadian supply was falling at the time, and this time around IHS expects Canadian production to return to growth on the strength of plays like the Montney. With what we know about the resource base and the improved ability to exploit it, this kind of growth appears to be possible.

The most concentrated growth occurs between 2015 and 2018. During these four years, supply will need to grow by about 4 Bcf per day each year to match the pace of demand growth, reaching a maximum year over year increase of 5.7 Bcf per day in 2017. This is more challenging to accomplish, but again, given our understanding of the resource base, it appears to be possible to achieve the necessary wellhead production growth.

Possible is one thing. Doing it is another. Many, many companies will be involved in creating the demand growth and the supply response, as well as the infrastructure to connect the two. Aligning all three as to timing and location poses a major challenge to the industry. Producers will need to increase drilling activity significantly to generate production growth, and they will need to increase it not just in the best areas but in some areas that are economically challenged at today's prices. This means that producers will need to increase the number of rigs drilling for gas, but the rigs and the crews needed may be busy drilling for oil. Perhaps most importantly, producers, in deciding what to drill, respond to actual prices that they can see, not price predictions that come with significant uncertainty. These may be prices in the futures market, where they can sell forward and hedge their production.

If the futures market anticipates the demand surge, drilling can increase in time to meet it. A typical time lag between drilling activity and production of five or six months assuming no infrastructure constraints, plus additional time to resolve those constraints, gives us a working idea of when we would need to see the futures market move up in anticipation of the demand surge. With the demand surge kicking off with retirements of coal fired generation in 2015, we would need to see this begin to happen in earnest in 2014. However, IHS projects that the average cash price at Henry Hub will be $3.86 per MMBtu in 2014, making this challenging. It may be difficult for the futures market to adequately anticipate the demand surge with sub-$4 cash prices.

If the futures market does not anticipate the demand surge, the drilling response will not occur until the cash prices have started to increase - in other words after the demand has started to materialize. Considering the time lags between drilling and production, the producing sector would then be scrambling to catch up.

Even if the producers do begin drilling in a timely manner, it is not certain that the supply will make it to market on time. This will require additional pipeline capacity to serve growing producing areas. Certainly many of the plays involved are located in areas where the existing pipeline system will be sufficient, as previously existing production declines and the new production takes its place on the system. But for newer plays like the Marcellus and the Utica, where the local production continues to achieve new highs, that's not the case.

The pipes need to be built, and that means the project sponsor will need to line up contractual support and work through the FERC permitting process. Producers tend not to sign up for pipeline capacity out of a growing supply area until it is clear that they need to do so. Given the time it will take to permit and complete a project after that conclusion has been reached, it is clearly possible that pipeline capacity growth will lag behind the need.

Those expecting a smooth, stable price path will be frustrated. Linear thinking will be punished by a nonlinear market. The problem is that, with strong demand growth coming at an uneven pace in a concentrated period, ensuring that supply growth and pipeline capacity additions exactly match demand growth will be nigh unto impossible. Periods of price volatility and higher prices can be expected when demand outpaces supply. End users should keep this in mind in setting their risk management strategies.

The reverse is also true. As the last few years aptly demonstrate, once the producing sector gets momentum going, it is hard to slow it down until and unless the price falls. In the years ahead, even amidst demand growth, if the pace of growth wanes, we can expect years characterized by softer prices and reduced volatility. Producers should consider this in setting their approach to risk management.

Both producers and consumers, in developing their risk management approaches, must consider the basis effects of new supply and new pipe. It is not enough to anticipate the price at Henry Hub if your local market sits on the wrong side of a pipeline bottleneck. With the geographic distribution of supply changing rapidly and the pipeline system changing with it, basis dislocations can occur and they can be financially painful if the signposts of change are missed.

Our analysis tells us that the resource base is there and the cost of developing it will be reasonable. But it still must be drilled in order to get production, and pipelines must be built to deliver it where it is needed. Timing those activities appropriately is difficult. That's why the real supply challenge is anticipating demand growth. * For the purposes of this article, North America refers to the interconnected US lower-48 and Canadian natural gas market. Mexico, which operates under different institutional arrangements, and Alaska, which is not connected to the continental pipeline system, are excluded.

Posted 21 January 2014

This article was published by S&P Global Commodity Insights and not by S&P Global Ratings, which is a separately managed division of S&P Global.