Nerds of Steel

Lots of people involved in the steel industry have access to steel pricing information to follow trends in the industry or to drive their own steel selling or buying decisions. There are at least seven well-known publishers of steel prices. Their steel industry experts survey steel buyers and sellers, analyze the data, and come up with a market price. But beware taking these prices at face value. If you take a close look at the prices and compare the prices supplied by different companies, you may be surprised. And now that pricing systems are being developed to support price discovery in the exchanges launching steel futures, we ask ourselves what these prices really mean, what specific products they are referring to and how accurate they are.

I have been tracking Purchasing Magazine’s hot rolled sheet prices and decided to look at how the US domestic hot roll price compared to the European domestic hot roll prices. Over the last two years, if you compare the Purchasing Magazine prices in US$ per short ton, the US price started $113/ton higher than the Metal Bulletin European price in October 2005 and ended $109/ton lower in September 2007. The Metal Bulletin European price rose $200/ton over that period and the Purchasing Magazine US price decreased $22/ton.

You might think that freely-traded commodity steel prices should move in the same direction in different regions of the world, but this data does not illustrate this. This could be explained by a shortage of hot rolled coil in the EU which can’t be supplied by imports because of high freight costs or some type of import restrictions. There are no obvious product definition problems, though MB expects us to trust their good reputation on this since it doesn’t give any easily accessible user guide with detailed definitions. This surprising price trend made me wonder how US hot roll prices from different publishers compare.

Metal Bulletin’s “HR Coil US mill FOB Midwest” price is consistently higher than Purchasing Magazine’s US FOB Midwest price, on average over the 2 years US$18/ton higher and rising up to US$47/ton higher in January 2007. However, the prices from the two companies are well correlated, rising and falling at the same time, which could mean that the Metal Bulletin price is for a higher quality product.

Maybe European-based publishers use a pickled hot rolled product, accounting for the price difference, while Purchasing Magazine’s is a non-pickled product. We can check this by making another comparison of prices. Steel Business Briefing is another well-respected (European-based) and steel-focused price provider. They also have a hot rolled coil US FOB Midwest price, which turns out to be on average much closer to the Purchasing Magazine results.

The average difference between the Purchasing Magazine prices and the SBB prices is only $3/ton. But as you can see from the graph, one reason the average difference is low over the two year period is because the SBB price seems to go higher than the MB price at the price peak and lower than the MB price at the price trough. In fact the range of difference between the prices runs from SBB at $35/ton higher to $20/ton lower, while in the Purchasing Magazine/MB comparison the difference ran from $47/ton higher to $6/ton lower, the range almost the same.

Possibly SBB has a different way of calculating the average price from all the price inputs they get which causes the peak and trough prices to be higher; including outlying data far from the average could do this. In the end the SBB and Purchasing Magazine prices at least seem to be for the same product, while the MB price may not be. If these well-known publishers can’t come up with close to the same answers, how will the companies that will provide prices to the planned steel futures exchanges suddenly get it right?

Platts (which is developing the pricing system with the LME, the London Metal Exchange) gets off on the right foot by clearly defining exactly which hot rolled product the price is for. They include the grade, which is a pickled product, so that clarifies some of the uncertainty we have with the SBB/MB definition. However, and worrisomely, it says it will easily normalize different product prices to this definition based on published list-price differences for different qualities. This is not very convincing, even if they claim to already do this for price discovery related to other metals on the exchange. Of course, the LME will list billet futures first rather than hot rolled coil so they will have more time to develop a recognized price.

World Steel Dynamics (using the SteelBenchmarker system with NYMEX, the New York Mercantile Exchange) does not specify the grade of hot rolled coil and says “for near-term delivery commodity-grade product to mid-size buyers”, which is vague considering they will be the first exchange to begin hot rolled coil futures trading by the end of 2007 (they claim).

We have good reason not to trust the steel prices the traditional suppliers offer for some of the surprises mentioned above. The steel futures exchanges are going to have to do a much better job of convincing us that their prices are credible. For hot rolled coil, there is a lot more work to be done.