Description

Great Northern Iron Ore Properties (GNI) is a trust that owns interests in land on the Mesabi Iron Range in Minnesota. The land is leased to steel and mining companies that mine for taconite iron ore. The interesting thing about GNI is that it's guaranteed to go to 0 in April 2015, as that is when the trust dissolves. You will not make a fantastic return shorting GNI; on the contrary, you're not likely to generate an IRR much above high single digits.

(Before you read any further, a word of caution so you're not wasting your time: GNI only has 1.5M shares outstanding so it is hard, if not impossible, to short in size. It's also impossible to short on the classic discount brokerage platforms, i.e. Schwab, Ameritrade, E*Trade, etc. You can short probably ~100K worth if you work with a full service broker.)

Our thought process on GNI goes something like this:

• The upside/downside/expected return dynamics here are extremely attractive:
  • Your best case scenario - if you're holding to "maturity," i.e. the end of the life of the trust - is an approximate 10% annual return over five years.
  • Your worst case scenario is making no money. Seriously, we can't find a way to lose more than a couple of percentage points (total) over the five year holding period. And if you're putting the money from the short into treasuries, you're probably up even in this worst case scenario. We'll go into detail on this in a bit.
  • Our base case is a mid-single digit return - in the 6% range - over the life of the trade.
• As an added bonus, if you are bearish on the market as a whole over the medium term (as we are), GNI will still go down if the market goes down. Go back and check the 2008 graphs of the stock if you doubt this.
• Essentially, GNI provides a way to generate solid short exposure with almost no risk of losing money, and the probability of a decent gain even if the economy and markets continue to rise in the medium term.

The GNI Business Model

GNI's business model is simple. The trust owns land. It leases the land to steel companies - Cliffs and US Steel, to be exact. The companies mine the taconite ore on the land and pay GNI a royalty based on how much ore they've mined. The contractual price per ton paid for the ore is determined very confidentially and getting the information we got from management (above and beyond what can be gleaned from company filings) was like pulling teeth unsuccessfully. Here's what we found based on filings and our conversations with management regarding pricing:

• The primary driver of pricing changes is the Iron & Steel component of the PPI, although the All Commodities component of the PPI also plays some role. When we spoke with management, they told us that while the Iron & Steel PPI is the primary driver, and the All Commodities PPI is also a driver, they do not - taken together - explain all of the variability in pricing. Other factors come into play such as mining costs, tonnage mined, and "other stuff."
  
  This was confirmed by the regressions we ran of annual royalty-per-pellet-ton-mined (royalty ppt henceforth) against the two PPI components (data is right below). As can be seen, the 125% increase in the Iron and Steel PPI from 2001 to 2008 was coincident with an approximate 55% increase in the royalty ppt. To say that they don't correlate 100% is an understatement. Based on management comments, it would be fair to apply a serious "drag" factor to the PPI drivers of pricing changes.

• When the Iron and Steel PPI tanks, as it did in 2009 vs. 2008, there is some level of minimum pricing built into the contracts that the steel companies have to pay. This can easily be seen below as the Iron & Steel PPI declined by 25% from 2008 to 2009, but pricing declined by 7.5%.
  
  As the company states in its filings "These minimum royalties can accumulate and do allow the steel and mining companies the ability to offset excess royalties (over the minimum royalty requirements) on future taconite production." We would expect this to limit any changes in pricing through at least 2011 and perhaps longer (2010 will be "fairly priced" at best given the current Iron & Steel PPI level of 210, and 2011 will allow for the steel companies to claw back 2009, assuming the PPI components rise beyond the 2008 level in 2011).

Scenario Analysis

GNI has precisely 20 cash distributions left to disburse to its trust-holders until the trust expires, i.e. precisely 5 twelve month periods to collect royalties and distribute. Additionally at expiration, trust-holders will receive in cash a final distribution equivalent to net liquid assets plus "principal charges" minus expenses and obligations of the trust upon termination. Net liquid assets is our terminology and refers to total assets minus total liabilities minus properties, while "principal charges" is GNI's term for the claim that present shareholders have on the trust properties. You can easily check the last few annual reports to see that this number does not move around very much and is presently at ~$12.5M minus those termination costs. We'll just assume no termination costs for now and use $12.5M for our scenario analysis.

We assume $2M in annual cash costs. This excludes the depreciation charges booked annually by GNI in its expenses and also adds in ~500K per year in interest and rent revenue. We're likely understating the costs by a bit here, but even if you add one or two million over the course of five years, it only increases your IRR by less than half a percent.

GNI will distribute all its royalties less expenses over the next five years. We take a look at a few scenarios below. Commentary follows.

In our base case scenario, we assume $2.70 in royalty-ppt in 2010 and 2011 and a 10% price increase annually afterword. This comes out to an average of about $3ppt over the five years. Again, this is a tad conservative as the 35% price increase over five years corresponds to a more than 70% increase in the Iron & Steel PPI off of the already high number of 2008. As for annual volume, we have a hard time seeing how volume goes much above 7M per year without an absolute rocket-ship of a recovery in auto production. If you look at the data on royalties and shipment we've presented above, you'll see that the change in pellet tons shipped on an annual basis is highly correlated to the change in annual auto production in the US. Even if we do see a vigorous overall recovery in the economy, that kind of auto production rebound is simply not in the cards.

In our worst case scenario, we assume $2.70 in royalty-ppt in 2010 and 2011, and a 20% price increase annually afterword. This comes out to about $3.50 ppt over the five years. We also assume average shipments of 8.5M pellet tons annually. We're not sure how either of these scenarios would possibly happen, but we guess it's conceivable that we go back to 2007-level US auto production and that the Iron  & Steel PPI increases by over 100% from the 2008 highs over the next few years. Still, even in this case, you come out about flat.

In our best case scenario, we assume flat pricing over the next few years and higher production. We don't think this sort of scenario is as hard to come by as our worst case, but we do find it hard to believe that the Iron and Steel PPI will be flat over the next five years.

Summary

So you have a short idea here that should generate mid-single digit returns over time and that will also be a pretty decent market hedge over the medium term. The kicker here is that it's close to impossible to lose money. Maybe not the most exciting idea, but certainly a phenomenal risk/reward (i.e. none/moderate).

Catalyst

No catalyst - just time...