Sooner or Later? – Part 3 - How Flow Data Provides Transparency Into Natural Gas Production

In all sorts of commodity markets, buyers and sellers would give their eye-teeth to have access to accurate daily supply and demand data.  Access to such data would provide insight into the utilization of transportation assets, transportation patterns and ultimately --- the holy grail of commodity markets – price.  What if there was a commodity market where you could know supply and demand on a daily basis?  Well there is.  And it is the natural gas market.  Gas market analysts have access to the luxury of pipeline flow data that (in the right hands) provides reasonably accurate estimates of daily supply (including production) and demand. In today’s blog, we explain how the natural gas industry uses flow data to track gas production trends in real time.

Recap

In Part 1 of this series, we compared the pros and cons of U.S. gas production data from the Energy Information Administration (EIA): the Natural Gas Monthly (NGM), and two forward-looking reports, the Short-Term Energy Outlook (STEO) and the Drilling Productivity Report (DPR). The EIA historical production data is considered the benchmark and for good reason. But the big downside of this data is that it is lagged by two months so the latest NGM, for instance, only provides actual estimates through July 2015. The STEO and DPR data provide projections largely based on history, to extrapolate that data forward. However current market dynamics of rapidly improving productivity and a disconnect between drilling rigs and production render such backward looking projections somewhat moot. Another way to look at gas supply (and demand) in more real-time that we introduced in Part 2 – is pipeline flow data. Using daily, real-time flow data from our friends at Genscape, we showed that production from the Marcellus and Utica basins combined has continued to grow since July. Today we’ll show you how we built the Appalachia production flows using Genscape’s natural gas flow database.

First a brief primer on what flow data is, where it comes from and some of its pitfalls (because no one data source is perfect).

What is Flow Data ?

Pipeline flow data is a collection of daily gas volumes nominated by market participants to either be received from or delivered into natural gas interstate pipelines (pipes that cross state lines) at  thousands of individual meters across the U.S. Flow data makes the natural gas market uniquely transparent and it shapes the way that fundamental analysis gets done nowadays in the natural gas business. Aggregating all that data into a common format provides some extremely powerful and actionable market intelligence. For instance, if you know the receipts on every pipeline in a given area that originates from production‑type facilities ‑‑ natural gas plants, gathering systems, etc. ‑‑ and add them all up, then you can estimate total production in that area. Same thing for deliveries – if you total flows from all the end-use and utility delivery points on the Eastern Seaboard, you can estimate total demand in that region. In this way, you can get a pretty good idea about supply and demand on a near real‑time (daily) basis. At RBN we crunch this data through our models to understand flow and capacity and their potential impact on the gas market – supply, demand, and especially basis differentials.

Where flow data comes from

It's possible to do all this on a daily basis because of a very arcane rule, FERC Order 587c, put in place about 20 years ago and put in its final form in 1997. The Order implemented a whole raft of rules that pipelines have to follow. One of those rules was, "Standard 4.3.6," developed by the North American Energy Standards Board (NAESB), which requires pipelines to post their operationally available capacity on their websites, also known by their legacy name as “Electronic Bulletin Boards” (EBBs) from the pre-high speed internet days. FERC implemented this rule in order to level the playing field between the pipelines’ own marketers and everyone else. This way all market participants would enjoy the same transparency.

Interstate pipelines are required to post daily scheduled volumes at each one of their meter locations. That includes all the receipt meters, all the delivery meters, all the storage meters, and in a lot of cases compressor stations along the pipe. All of this data is publicly accessible on pipeline websites.

However, as much as FERC and NAESB tried to standardize how flow data is reported, each of the 100 and some odd interstate gas pipelines in the U.S. posts the data in their own way – meaning different formats. This is where data analytics companies, such as Genscape and PointLogic Energy, come in. They collect the nominated volumes at each of the thousands of meters multiple times a day and associate each meter with useful metadata (describing the type of nomination, facility, timing and location). The metadata allows analysts to aggregate the volumes by type (ie. production, demand, storage) and geography (ie. state, county), among many other attributes, in order to piece together big picture trends.

Some pitfalls of flow data

Now, the catch. As ingrained as the flow data now is in industry analysis, it is not without its pitfalls.

One of these is that the data is only publicly available for pipelines that are regulated by FERC – the interstate pipelines. A good deal of natural gas goes directly into local distribution systems or intrastate pipelines and is therefore not reported. The individual states, not FERC, have jurisdiction over these pipes and do not require flow data to be publicly posted. FERC tried several years ago to extend its jurisdiction to these pipelines in the name of transparency. For a brief period they were successful and the industry got a peek at flows on those pipes. However, the courts struck that down before long and flow data today primarily consists of scheduled flows on interstate pipelines. This constraint often limits the ability to parse out different types of demand in some locations, for instance, whether gas from a local distribution company is being delivered to a consuming end-user or into a storage facility. These gaps affect some regions of the U.S. more than others. But interstate flow data captures the vast majority of gas volumes moving around the country. There are also ways to make reasonable guesses about the data that you can't see. And proprietary models do sometimes “gross up” the flow data to fill in the gaps. Nevertheless, due to the small gaps, raw flow data should be regarded as a “sample” of total flows, not 100% of volumes.

Another issue related to flow data analysis is that it can be volatile and susceptible to the impact of maintenance and weather events, which can create noise and temporarily mask macro trends. However, this aspect of it is also beneficial to physical traders who want to understand and anticipate the impacts of short-term events.

Despite its drawbacks, the real-time nature of flow data makes it a valuable information source for market participants. With flow data, you can tell where gas is headed to, where it is being used, and how fully pipelines are being utilized. And you can do it daily.

In the example that follows we’ll walk through how to aggregate thousands of individual meters to get to the total Appalachian shale gas production we showed you in Sooner or Later? Part 2, using Genscape’s gas flow database. In this case recall that we presented an estimate of overall regional production based on flow data that we could compare with EIA data in the DPR report.