Beneath the revision confusion, CISA steel data weak

As posted here, it looked like revisions to the NBS steel production data would likely impact the intra-month CISA data given the revision in production levels.  This appears to be the case in mid-April, with the apparent surge in production largely due to a re-estimation of small mill production rates.

The huge leap to growth rates of close to 8%YoY don't make much sense in the context of what is going on in markets more broadly.  It also makes little sense that small-mill production is up ~25%YoY in the last 10-days.

The reality seems to be that current rates of production are a bit higher than what we saw in the NBS data for March.  What this means in terms of YoY growth is a little uncertain, given April 2013 could be revised as well.

So given all these revisions and changes, are these CISA data no longer useful? While they are less helpful, they are still accurate in terms of production from CISA mills.  Here production growth was not as strong as the headline, but also not particularly bad, rising 3.6%YoY in mid-April.

This stronger rate of production growth, however, meant that the destocking trend at CISA mills took a pause, with inventory in terms of days of use and in absolute terms rising in the month.

By adding CISA mill production and inventory changes we can get a snapshot of apparent consumption, which fell 1%YoY, although stock movements can be large intra-month.

Movements in iron ore inventory at ports seem to gel with the idea that conditions are soggy, but not disastrously so.  Inventory is high in absolute, although not particularly high in terms of days of use.

This may become more problematic in the coming months. Seaborne supply growth is keeping port inventory steady during peak consumption period, but production rates will fall away on the usual summer slowdown and given construction headwinds.

That is likely to weigh on iron ore prices, which have fallen quite a bit in the last week or so. This is partially due to concerns about availability of letters of credit (LOC), with policymakers concerned about financing deals keeping steel mills alive when they should be shuttering capacity.

There is a mixed range of views on how much iron ore is being used as collateral for cheap financing. Some analysts have put it as high as 30mt, while other suggest its very small.

I tend to think its the latter, given its much more difficult to hedge the risks associated with these deals than it is for copper or other commodities. Port inventories are also not particularly big, unlike Chinese bonded warehouse stocks.

Even so, more expensive LOC will impact traders ability to operate and will likely lead them to liquidate some inventory. This should see iron ore prices fall further in the short term.