Search Copying and distributing are prohibited without permission of the publisher
Email a friend
  • To include more than one recipient, please separate each email address with a semi-colon ';', to a maximum of 5

  • By submitting this article to a friend we reserve the right to contact them regarding AMM subscriptions. Please ensure you have their consent before giving us their details.

Can past scrap pricing predict future movements?

Keywords: Tags  The Metrics System, ferrous scrap prices, John Ambrosia

History has been called everything from knowledge to a weapon, from cyclical to bunk. But is it prologue? Which end of that spectrum does ferrous scrap pricing history fall on if it is to be used to try to understand market dynamics?

One possible way to answer that question is to look at how the market has performed historically in each individual month. Or to put it another way: Over time, do particular months exhibit specific outcomes that might be useful in looking ahead? Using the first 14 years of the 21st Century as a sample, here are the average scrap prices for each month from 2000 through 2013.

Well that was easy. August clearly is the strongest month for scrap prices, while November is the worst. Done, right? Not so fast. There are two things wrong with those conclusions.

First, as the old joke goes, if you sleep with your feet in the oven and your head in the freezer, a statistician will tell you that, on average, you’re pretty comfortable. The numbers are averages, and although 14 years is a decent sample, such an approach doesn’t tell the whole story.

Second, this approach also ignores the impact of the biggest fiscal event in recent memory: the Great Recession of 2007 to 2009 (and beyond?). So here are those averages again for 2003 through 2007--the five years leading up to the economic collapse of 2008--and for 2009 through 2013, the five years since (I’m going to leave out 2008 as its wildly high and low prices tend to distort the overall averages):

Whatever else the fallout of the recession might have been, scrap prices appear to have permanently moved into a higher rent zone.

But there is another way to look at monthly markets beyond averages: how many times a given monthly market has increased or decreased over time. This table shows each month, and its number of up or down markets from 2000 through 2013:

Now this is interesting. Some of these months are very strong and only rarely have experienced any significant price decreases. Clearly, the spring and the fall are the weakest times of the year for prices, and the winter and summer the strongest.

Finally, looking at trends of data not in these tables reveals some telling facts about monthly market behaviors in recent years:

  • January has risen seven years in a row, from 2007 to 2013.
  • February fell four of the five years from 2009 to 2013.
  • June has produced a sideways market four times, the most of any month.
  • December has risen eight years in a row, from 2006 to 2013.

The middle of the year is most likely to generate a run of sideways or near-sideways markets. Examples include June-to-September 2011, May-to-October 2002, May-to-July 2007, March-to-September 2001 and June-to-September 2000.

This history suggests that at the start of 2014, January should have increased and February should have fallen. Both of those occurred, extending January’s streak to eight up years in a row. However, it also suggests that March should be sideways to up, and that seemed unlikely as of mid-February. But remember, these are useful only as statistical probabilities, not as forecasting tools.

Have your say
  • All comments are subject to editorial review.
    All fields are compulsory.