In my most recent pair of articles I ran screeners to find companies that offered double-digit dividend yields that might be attractive to investors looking for alternatives to mortgage REITs which fell victim, at the beginning of October, to QEternity and a substantial increase in refinances (and subsequent prepayments). The two articles presented a combined total of 14 non-REITs that seemed reasonable selections. Not all companies identified by the screeners made it, however; some got through the filters I used despite not satisfying the criteria, some had gone through changes that disqualified them (my thanks to observant readers who helped identify a couple), and yet others just "didn't feel right."
One ticker in particular kept coming up in the screener results for the second article ("Still Troubled by mREITs?") that really seemed impressive - in nearly every regard it beat all of the nine final selectees, and did so decisively. Not only would it have been the far-and-away top choice of the second group of companies selected, but it would also have placed very well among the five top choices identified in the first article ("Have mREITs Got You Down?"). However, despite the fact that its fundamental data virtually blew away the competition, I opted to exclude it from the final selections.
In retrospect I felt that it was not enough just to ignore the company, but that I had to explain why I felt (and still feel) that this company might not seem to be an effective (read: profitable) investment - at least under its present circumstances.
Great Northern Iron Ore Properties (NYSE:GNI)
Great Northern was established in 1906 as a Trust to manage over 67,000 acres of iron-ore-rich land in Minnesota. At the time of its formation it was listed on the NYSE under the ticker GNI, and 1,500,000 "certificates of beneficial interest" were created. It had, and still has, three primary functions: to manage leases to the land under its control; to collect fees and royalties from the leases; and to make distributions (in the form of quarterly dividends) to the holders of the certificates (which could be traded as shares). Other than entitling one to the payments made by GNI, the certificates endow their holders with no rights or claims.
The Trust was never intended to exist indefinitely; it was formed containing a list of 18 interested individuals, stipulating that 20 years after the death of the last of those 18 individuals Great Northern would cease to exist. Great Northern's last function would be to liquidate all of its assets, fulfill all of its obligations with the proceeds, and divide the remaining monies among the holders of the 1,500,000 shares. Upon the distribution of this final payment, GNI would cease to exist, all "shares" would cease to have any value (for all intents and purposes, they cease to exist), and the land would revert to its proper owners (currently Glacier Park Company, a wholly owned subsidiary of ConocoPhillips (NYSE:COP)).
My reason for excluding GNI from either of my lists is this: the last individual named in the Trust died on April 6, 1995. This means that Great Northern Iron Ore Properties must liquidate itself and cease to exist on April 6, 2015 - less than 2 ½ years from now. After examining the current dividend payouts I realized that there seems to be no way to recover the cost of investment at the current share price.
As illustrated in the thumbnail sketch above, shares in Great Northern are currently trading in the mid- to upper-$70's, and have been in that general area for the past month or so. What do you get for your share price? Between now and its imminent demise, Great Northern will make 10 quarterly dividend payments (4th Q 2012, all four quarters of 2013 and 2014, and 1st Q 2015). Upon its demise on April 6, 2015, Great Northern will make its final distribution. The question is, then, how much can one expect those eleven payments to be worth - and will that be enough to justify a $78.28-per-share price tag (the current share value as of 2 November 2012)?
Here is the ten-year graph tracking both the price of Great Northern and the value of its quarterly dividends:
In the past 2 years, share prices in Great Northern have dropped by more than one-half - to $60.17 on April 20, 2012, from $150.65 on December 10, 2010 - and dividends have seemed to trend along with share price, reaching a high of $5.75 in Q4 2011, and then dropping sharply to $2.25 in Q1 2012, only to rise to $3.00 in Q2 and $3.50 in Q3. Indeed, if there is a persistent characteristic in the dividends paid by GNI, it is that there is no obvious discernible pattern to them.
Unlike many companies that establish a set dividend which it is their intention to pay, GNI does not set its quarterly dividend until such time as it is able to determine the earnings available for distribution. This results in a measure of uncertainty in terms of trying to estimate projected distributions, as one cannot typically determine the size of a future distribution simply on the basis of the current payment. Assuming, for the sake of argument, that each of the ten remaining dividend payments are equal to the most recent one made ($3.50), one can expect a total dividend payment of $35.00 per share over the next 10 quarters.
This leaves the final payment as the only possible source of incentive to maintain a high share price. However, to stick a pin in that particular bubble, Great Northern released figures that show that - had the company made that final payment in December, 2011, the final payment would have been roughly worth $8.59 per share (this figure derives from adding the net monies to the principal charges at that time - more on this later). When we add the $8.59 final payment to the estimated dividends to be paid over the next ten quarters we reach a figure of $43.59 - again, per share.
This means that, if one were to purchase shares in Great Northern Iron Ore Properties at $78.28, receiving an ultimate payoff of $43.59, one would recover barely 55% of one's cost basis - which does not make for a good investment. In fact, for the investment to just break even, the total value of the remaining dividends and the final payment would have to increase by nearly 80%! While such an increase in value is not impossible, it is rather unlikely for any average healthy company - and Great Northern is, for all intents and purposes, a company with no future.
It was because of this gross discrepancy between share price and reasonable projected value of the shares that I felt I could not, in all conscience, include Great Northern among the mREIT alternatives I listed. Does this mean that Great Northern cannot be a reasonable investment at some price? No matter how you view Great Northern, any investment in it will involve some measure of risk, inasmuch as we cannot determine with certainty very much about the future of the Trust except for the date on which it ends.
Estimating the Value of Great Northern Iron Ore Properties
The task before us is to examine the basis for the current estimations of dividends and final payment to see if there is any reason to expect that they could increase (or decrease) over the next two-and-a-half years. It may be impractical to try to make some sense of 106 years of a company's performance, but I believe we can identify a more manageable period of time.
On October 10, 2008, Great Northern dropped to its lowest price in the past ten years: $43.81 per share. On December 10, 2010, Great Northern rose to its highest price in the past ten years (and, apparently, in its history): $150.65 per share. While the company's dividend payments at the time of its 10-year low were on the rise, they were still below the most recent payment of $3.50. The period from 2008 to the present, then, includes both a year of poor performance as well as a period of noteworthy success for Great Northern; indeed, in its dividend declaration for the 4th Quarter of 2011, the board of Trustees called the year 2011 "historic," and predicted that future years would not be able to match the level of success of that one year. (2011 Financial Report, p. 8.)
[Note: one factor that may have contributed to 2011's "historicity" is the fact that in 2010 the leases on three of its properties expired. GNI's 10K for 2011 shows those three properties to have been leased again; no doubt a significant portion of GNI's good fortune for that year was due to those new leases. There are no other leases set to expire but for one which expires in 2014. The extent to which that one lease would improve either the dividend payments or the final payment cannot be determined at present, although it seems unlikely that it could have a significant impact due to its timing in relation to GNI's dissolution.]
The graph below identifies the quarterly dividend payments since October, 2008:
Over the 19 quarters represented on the graph a total of $55.70 were distributed to each share. The dividend payments made since October, 2008, have all been reasonably high in light of the current yield - that is to say, the current yield is not an aberration, but a regular characteristic of Great Northern's stock. The annual distributions have been increasing (except for 2009, when the company noted that demand for iron ore was down due to economic conditions), and the quarterly payments in any given year seem to follow the same pattern - they tend to get larger as the year goes on, with the fourth quarter being substantially larger (this is even true for the low-paying 2009).
The first task at hand is to come to a reasonable projection for the 4th Quarter, 2012. I think it is reasonable to expect that the Q4 dividend will be at least the $3.50 being estimated on the basis of the quoted dividend yield, and possibly higher. In its 10Q for Q3 2012, Great Northern reported earnings per share of $11.01. Subtracting the $8.75 already paid out this year, the Trust is able to cover a $2.26 dividend for the fourth quarter. It is likely that the EPS will increase in Q4 due to regular revenue, and an increase of $1.24 is all that is needed to cover a Q4 dividend of $3.50. Given the data on the graph above, it would be no surprise if the Q4 dividend was, in fact, larger.
[Note: Great Northern is obligated to pay its earnings to shareholders (among other things) in order to maintain its status as a tax-free trust. Therefore, its earnings-per-share is a fairly accurate determinant of what its dividends will be. This can be illustrated by its dividend cover (the extent to which EPS is adequate to pay its dividends - determined by dividing EPS by dividends):
As you can see, GNI plays it close when it comes to paying out dividends, even if it means accreting funds from elsewhere when there is a shortfall, as in 2009 and 2010.]
GNI is currently guaranteed a minimum of more than $6,000,000 per year in royalties from the lands it has under lease. Actual royalties paid have averaged nearly $20,000,000 per year over the past 5 years (2007 - 2011). Through the first three quarters of 2012, royalty payments have been $19, 483,824, up over $1,000,000 from the same period in 2011. Q3 royalties, however, are down nearly $1.3 million, with Q3 2012 at $5,178,144 - down from $6,473,243 in Q3 2011. Total royalties for 2011 were $26,614,880; Trustees were estimating that total royalties for 2012 would not reach that level.
These considerations would seem to place something of a cap on actual earnings per share while still leaving some room for a Q4 dividend somewhat higher than $3.50. Furthermore, it seems that Great Northern's Trustees do not expect any significant increase in income in the foreseeable future. This would seem to confirm the total dividend value for the remainder of GNI's existence would be approximately $35.00, with a possibility of being nominally higher.
There is a caveat to be brought up, however. With the exception of 2011, each year's total dividend distribution has been lower than $14.00, and Q4 2012 would have to pay a dividend of $5.25 to reach the $14.00 projection. This is not impossible, but it does raise the possibility that the $3.50-per-quarter projection may be somewhat higher than can be realistically expected - that is to say, the estimated $35.00 in future dividend payments may just as easily be somewhat lower as higher. For this reason, I take the $35.00 figure to be the best estimate that one can reasonably expect.
If dividend payments are not going to make up the difference between share value and the projected payments, then we must turn to the final payment to determine where to make up the difference between share value and expected returns; however, as it turns out, there is little hope of that happening.
The "final payment" is made up of two sources: the "Principal Charges Account" and the Trust's "net monies." The Principal Charges Account is not an asset or liability of the Trust, but is, rather, a court judgment requiring that shareholders be paid by the reversioner [that is, ConnocoPhillips ] with respect to leases held by U.S. Steel (litigation about this compensation was carried out between 1972 and 1982). The Principal Charges Account consists of attorney fees and expenses for the litigation (a fixed amount of $1,024,834), costs of surface lands (the value of real estate that is subject to change, but worth $6,606,815 at the end of 2011), cumulative shipment credits (a charge against the Account which is variable, but worth $2,297,652 in 2011), and cumulative asset disposition credits (variable charges against the Account, valued at $372,124 in 2011).
In 2011 the Principal Charges Account had a value of $4,961,873. Because the Account has variable elements to it, however, it is subject to change, as shown in the following table:
[Note: for the duration of this discussion I will include the year 2007 while disregarding 2012. The reason for this switch is to avoid comparing an incomplete 2012 analysis based on incomplete data with the full-year data of the other years; it will also prove interesting to see the data from 2007 in comparison with what follows it.] While the 2011 balance is higher than both the 2010 and 2009 balances, note that the 2010 balance is approximately $90,000 lower than that of 2009, and that the 2011 balance is $75,057 less than the balance in 2007. Because the Principal Charges Account is variable, and particularly since it seems as likely to decrease as increase, it does not appear to be the candidate for a dramatic increase in value over the next two years.
The other element of the final payment is termed (by the Trustees) "net monies," and represents what is left of the total assets of Great Northern after all liabilities have been paid, and after the value of the property held by the Trust is subtracted (the properties revert to Glacier Park Company). (2011 Financial Report, p. 14.) The following table provides the figures for the years 2007 through 2011:
Note that, as with Principal Charges, net monies does not follow a pattern, but increases and decreases on an annual basis. Note also that the data for 2011 is less than the data for 2007 by more than $200,000.
What remains is to determine the total distributions that result from adding the Principle Charges with the net monies, and seeing what sort of distributions-per-share result. Two things need to be mentioned before we do so, however. First, the final distribution figures do not include the "expenses and obligations of the Trust upon its termination," referring to costs the Trust may incur in the process of being liquidated - any such expenses and obligations would be subtracted from the final distribution figures. Second, there always has been, and can only be, 1.5 million shares of Great Northern Iron Ore Properties - there is no chance of dilution of shares, or splitting or reverse-splitting of shares that might affect the final distribution.
The following table contains the total distribution funds available for the years 2007 - 2011:
As one would expect, the estimated total distribution figures from each of the past five years vary from year to year, and do so in such a way as to make future predictions virtually impossible. The amount available from year to year is either an increase or a decrease from the year before. And again, and overall, the figure for 2011 is shy of the figure for 2007 by nearly $300,000. The estimated distribution per share is displayed for each year in the following graph:
As one can see, there is no surprise here. The estimated distribution varies from year to year, increasing in some, decreasing in others; and again we have the foreboding decrease from 2007 to the "banner" year of 2011.
Summary and Further Observations
At this point, it would be very easy to simply say "GNI is going to pay approximately $43.00 per share from now through its April 6, 2015, demise. Buy its shares now and you will likely lose more than $35.00 per share." To be honest, however, there are other considerations which require our attention. The past several years have been difficult, and not only in the U.S. Globally, economies have been stuttering and the effects of an insecure global marketplace will likely have unpredictable consequences for those companies leasing and mining land controlled by GNI. The American economy is gradually recovering as are other economies, but this does not necessarily mean that there will be record demands for iron ore.
Moreover, the process of mining iron ore is subject to many variables: labor conditions, weather, costs associated with the mining process, transportation of the iron ore once mined, all have a role in the amount of royalties Great Northern will receive over the next 2 ½ years. To be sure, GNI does not get involved with the mining process, but the royalties it receives comes from those who do. The miners' success or failure in their endeavors translates directly into the success or failure of Great Northern Iron Ore Properties.
It is impossible to predict the nature of business over the next 2 years or more. One can only attend to the observation made by the Trustees of GNI that 2011 was an excellent year for the Trust, and that the next few years are not expected to be quite as good. With that in mind, given the rather erratic nature of business over the past few years, and the similarly erratic nature of GNI's performance over that period, I find it hard to believe that the next ten dividend payments and the final dissolution payment will amount to much more than we have notice so far; that is to say, that the next 11 payments made by GNI to certificate holders should amount to less than $45.00.
It is virtually a fore drawn conclusion that any money paid to acquire a certificate of beneficiary interest in Great Northern Iron Ore Properties is money that is gone - all that remains is to collect the payments to be made from that point on. It is not a case where you reclaim your share price plus the dividends and final payment - all there is will be the monies paid out by GNI. Once the dividends have been paid and the final distribution sorted out there is nothing left to the "shares." That makes the decision to purchase GNI's certificates a very final act. (Consider the people who, two years ago, paid $150.00 for each certificate. They have seen - so far - $27.50 from their investment, and will see an additional $43.59 for the remainder of GNI's existence. Their $150.00 becomes $71.09 - less than half their initial investment.)
To my mind, if purchased before the next ex-dividend date, a share of GNI is worth approximately $35.00. The net gain the investment would realize would be approximately $8.50, a little better than 10% per year. Each quarter that goes by drives that purchase price down by approximately $4.00 (again, to my mind). Someone who is inclined to take more risk might be willing to pay more for their shares now, but I believe that anything over $45.00 borders on the reckless, and would seem to be more of a gamble than an investment.
But then, only time will tell.
All material for this article was drawn from GNI's 10Q's and 10K's on file with the SEC, financial reports released by GNI, and from GNI's website. Charts courtesy YCharts.