Price Tag! - Price Reporting Agencies and the Shrinking World of Fixed-Price Deals

The latest natural gas transaction data from the Federal Energy Commission (FERC) shows the natural gas market is increasingly relying on published index prices for transacting physical volumes for day-ahead and month-ahead deliveries. Index prices — volume-weighted averages of all eligible prices reported to index publishers by location — are considered representative of the market and mitigate some of the perceived price risk associated with “fixed-price” deals, in which the price is independently negotiated between counterparties. But in order to make their indices representative and grounded in market reality, publishers — or price reporting agencies (PRAs) — rely strictly on prices from those independent fixed-price deals to set the index in the first place. As more of the deals done are based on index, what happens to the index itself? Today, we continue our review of natural gas transactional data and what it says about how the market is evolving. 

As we explained in Part 1 of this blog series, PRAs such as Natural Gas Intelligence (NGI), Platts and Argus have a long-standing role in the physical natural gas market as benchmark price-setters, collating transaction data from trading counterparties and spitting out index prices that companies can then use as benchmarks for pricing the commodity at any given location. But only certain types of trades are eligible to include in the index, namely physical natural gas transactions for next-day or next-month delivery and they must be deals done at fixed prices between buyer and seller, not index or long-term deals. Besides fixed price deals, the only other type of deals included in the preparation of indices — only in the month-ahead index — are based on a fixed differential (or “basis”) to the Henry Hub (e.g. minus $0.15/MMBtu under Henry), the primary pricing point for natural gas in the U.S. and delivery point for the CME/NYMEX futures contract. Those companies that choose to participate in the index-making (it’s optional) send their transactional data to PRAs — including volumes, prices, timing, etc. — at the end of each trading day. The PRAs pull all the prices into database systems, and crunch all the numbers to come up with an average (or some other mathematical midpoint) of all reported trades on the daily basis for the next-day market and on a monthly basis for month-ahead prices. The resulting prices are then used by the industry as the settlement price for contracts tied to the index.

For many years, the PRAs were the only ones accumulating and disseminating this kind of price data. But in the early 2000s, the process ran into manipulation issues and a few indictments later (see Part 1 for more background on that sordid history), FERC dipped its toes in this world of commodity index pricing. Starting in 2009, per Order 704, FERC began collecting and reviewing the natural gas transactional data on an annual basis under Form 552. The purpose was not to collect every single deal happening in the market, but instead to hone in on the deals that are either impacted by index prices (the index-priced deals) or that are eligible to be included in the index — essentially, fixed-price deals for delivery of physical natural gas in the day-ahead or month-ahead period. The latter of those are transacted during the last five business days of the month (called “bidweek”). The big difference between PRAs and FERC is that Order 704 requires pricing data to be submitted from all companies meeting a size threshold that conduct these types of PRA-eligible deals, not just the ones who elect to participate in PRA indices. Last year (2015), 676 individual companies filed these reports with FERC. However, FERC did cap reporting at companies who buy or sell a minimum of 2.2 trillion British Thermal Units (TBtu) or 2.2 million (2,200,000) MMBtu of “reportable” natural gas purchases or sales in the reporting year. That’s about 6 MMBtu per day, which is a relatively low threshold for most of the market participants doing the majority of these deals. There are also some other types of deals collected under Form 552, such as NYMEX trigger transactions which are contingent on a futures contract. At the same time, there are also fixed-price deals that it does not include, namely long-term deals. Nevertheless, FERC’s 552 data is the most comprehensive and publicly available dataset on U.S. natural gas transactions that either impact or are impacted by PRA indices.

We want to thank our good friends at Natural Gas Intelligence (NGI) for crunching the data and letting us use the analysis here so we can see what it can tell us about the health of the natural gas market and of PRA indices.

In Part 1, one of our findings was that there may be a problem brewing for PRAs who rely on those fixed-price deals to set an index price as the market seems to be doing more and more of its business based on index prices. While that’s good news for PRAs’ index sales (which is big business), it also makes index-creation tougher — an index that’s used for nearly 80% of traded volumes in the physical natural gas daily and bidweek markets. While total volumes, and volumes for each of the deal types, were up in 2015 to the highest levels since at least 2008 (blue bars in the left graph in Figure 1), the percentage of fixed-price or basis deals (the kind that are eligible for index-making) were down (right graph in Figure 2). Moreover, putting 2015 data into historical context, it’s clear that the latter shift is a trend that’s gradually deepened over the years. Of the total, fixed-price (green bar in the right graph) and basis deals (orange bars in the right graph) combined made up only about 21% of total transaction volume in 2015. That’s down from 31% five years ago. Another 1.5% of transaction volumes in 2015 were on miscellaneous deal types (yellow bars), leaving a whopping 78% of deals done at index-related prices (blue bars in the right graph). The volumes for that last deal type are up about as much as fixed-price deals are down in the last five years.