Today’s report (week ended 6 Apr): EIA: -19 bcf, EA: -11 bcf
- Like last week, today’s print fell outside of the range of forecasts. Whereas last week’s implied flow was a 1 bcf/d bearish miss versus consensus, this week’s was a 1 bcf/d bullish miss versus consensus, effectively netting one another out. With supply in the reference week up modestly—production 0.1 bcf/d higher w/w and LNG sendout up 0.3 bcf/d w/w together with GWHDDs that were some 3% lower w/w—a more pronounced reduction in withdrawal activity than that reported by EIA for the week ended 30 March and 6 April would appear necessary. To ‘true up’ to today’s larger-than-expected withdrawal, we made baseline revisions to gas heating intensity (GHI) (mmcf/d gas use per GWHDD) that is out of trend with the adjustment made to fit last week’s implied withdrawal that was much lower than expected.
Next Thursday’s report (week ending 6 Apr): EA preliminary: -27 bcf
- From Tuesday’s Storage Report, we have downwardly revised our production estimate by 0.4 bcf/d and brought down both net Canadian trade and LNG sendout by 0.1 bcf/d. Additionally, baseline revisions to GHI increased our res-com estimate.
April snow showers
The 2017-18 heating season is on track to extend into mid-April, with low temperatures and snowstorms delaying the start of injection season. All this unseasonable weather will serve to keep the April net injection as the lowest recorded this decade. A more normal month in terms of weather would have placed inventories well above where our balances indicate end-April stocks are in fact likely to end. The extension of cold has prevented a more typical softening of prices in the shoulder season. Cash prices have risen modestly, closing in the high 2.70s $/mmbtu more frequently than in what was a very cold March. In recent days a robust nuclear outage schedule has helped to support power sector gas demand as well.
While supply and demand balances are certainly tight for April, weather-aided demand can only be so supportive, even in a withdrawal season that extends beyond the traditional end of the season on 31 March. Summer prices have softened since the first appearance of the unseasonably late cold as the market seems unconcerned with ending October at inventories near 3.6 tcf, the lowest since October 2014.
The trajectory of Q1 18 sequential production growth was relatively flat. But our short-term balances (see page 3) indicate that the last several weeks have seen w/w production gains, with momentum particularly strong in the past two weeks, reaching above 78 bcf/d. Our Appalachian flow sample has also started to pick up steam for the week in progress, with readings trending at 27.2 bcf/d through today, which would best prior weekly highs. Such a return to growth has likely been helped by less competition with utilities and other storage users for local pipe now that heating demand is starting to ebb―a factor that was also likely behind the lack of sequential gains from late December until the end of March, much as it had been last heating season. Moving forward, the extension of cold into April is largely helping to accommodate more ‘room’ for US production growth this summer.
Still, even with our forecast of near 8 bcf/d y/y growth for output this summer, our balances suggest a storage carry at the end of October of just 3.55-3.60 tcf, which would leave fundamentals at a critical balancing point. If production growth is less than 8 bcf/d y/y and/or it is a hot summer, then inventories to start the 2018-19 winter will look lower than the levels normally carried into winter. On the flip side, if production growth exceeds expectations and/or peak cooling season weather comes in milder than 10-year normal, such an inventory figure to end the heating season will look healthy heading into winter.
Cash prices and futures prices at Henry Hub have been modestly supported by the extended withdrawal season, which should come to a close after the week ending 13 April, when a more typical shoulder season softness should emerge. The market’s apparent lack of concern in recent days about the impact of a late start to injections suggests we should see a bit more price softness return as we enter the headier days of the shoulder season in late April and early May. However, it will be the timing of the start of peak cooling season and the addition of Appalachian infrastructure this summer that will give us a clearer idea of just where stocks will end October.