Novagold (NG) caught the eye of some pretty big fish in 2010. After closing deals with John Paulson and George Soros to raise $175 million in capital, it has now struck the fancy of Jim Cramer, who has called it the "best long-term play on gold" out there. The stock had a stellar 2010, surging from around $6 to its current price just below $14. But with meaningful production still 4-5 years away, this is a tough one for most investors to evaluate. However, by breaking down Novagold’s core assets and using a bit of present value analysis, we begin to see why Paulson, Soros, and Cramer are so excited about Novagold.
Novagold has two projects well along in the development stage, as well as another significant project that is in the very early stages of development. Its Donlin Creek deposit, located in Alaska, is primarily a gold asset that is 50% owned by Novagold (the other half is owned by Barrick (ABX)). The Galore Creek mine, located in British Columbia, is a 50/50 joint venture with Teck Resources (TCK) that will produce a significant amount of copper, gold and silver.
Finally, its Ambler project in Alaska, already contains very significant amounts of copper in the limited amount of exploration that has been done on the property. Donlin is currently updating the 2009 feasibility study in preparation for permitting, which should bring the mine into production sometime in 2015 or 2016. Meanwhile, Galore Creek is undergoing the pre-feasibilty study, but due to the faster approval process in British Columbia, it is expected to be in operation before Donlin Creek.
Of note, both the feasibility study for Donlin and the pre-feasibility for Galore are expected to be completed in the first half of 2011, providing significant catalysts for further upside in the stock price in the near term. In particular, the use of an up-to-date gold price and copper price should meaningfully enhance the size of Novagold’s economically recoverable reserves and thus the value of the company.
Given that Novagold is still several years away from production, valuing the assets is more art than science. However, we know that Barrick (Novagold’s partner at Donlin) offered $1.6 billion for Novagold in 2006, an offer which was voted down by shareholders. Subsequent to that offer, Novagold has succeeded in raising over $200 million in capital and has significantly expanded its resource base from just over 12 million ounces of gold measured and indicated (per the 2005 Annual Report) to over 27 million ounces measured and indicated today.
How much might this resource base be worth in time? Well, Novagold’s CEO recently noted that Red Back Mining was acquired by Kinross (KGC) for around $1000 per ounce of measured and indicated reserves. Now, we have to grant that Red Back is producing over 400k ounces of gold in 2010, but the expected cost structure of both Red Back and Novagold is similar (just under $400/oz).
If we assume that Novagold is going to ramp up production in 2016 and use a 6% interest rate (the likely cost of issuing long term debt for one of the many miners rated BBB) over five years, the discounting factor (denominator) for Novagold’s five years without production is equal to 1.34. Dividing $1000 by 1.34 yields an equivalent buyout price of about $745 per ounce of measured and indicated gold reserves. Based on 27 million ounces M&I, that would put Novagold’s present value at $20 billion in a takeover.
Now, I’ve been doing this long enough to know that nobody is ever going to pay that sort of premium in a takeover. Novagold could easily be worth that kind of price as it moves into the production phase in a few years, but until it joins the ranks of the million ounce per year miners, a more modest valuation is appropriate. But let's call $20 billion a glimpse into what could be in a four years time, and run a few more numbers to get a realistic idea of what a takeover price could look like today.
For mining companies, I like to use present value analysis because it doesn’t take a lot of guesswork to know where the earnings will come from many years down the road. Once we make a few key assumptions about the gold price, copper price, the amount of recoverable reserves, and the overall cost of bringing those reserves to market, estimating the present value of those earnings streams is pretty easy.
As far as the gold price is concerned, I have chosen to use $1400/oz, the level around which the current spot price has been trading, as well as a $4.50/lb copper price. Later in this piece we’ll talk about Novagold’s leverage to upside to rising gold prices. With regard to reserves, I have assumed that Novagold’s total inferred reserves will be recoverable on the basis that the inferred reserves are very likely to be recoverable due to the fact that the gold price is more than $400/oz higher than the most recent feasibility study.
In reality, what is more likely to happen is that the amount of recoverable reserves will grow by about 20% when the new gold price is factored in, but I have chosen not to include that assumption to make the analysis more conservative. Finally, I have assumed an overall cost of production of around $850/oz, which encompasses $400/oz cost of production and $450/oz for depreciation and other costs. To my mind this is conservative, but consistent with comments by Anglogold’s CEO last week on industry costs (accounting for NG’s lower cost of production, which is pegged at just under $400/oz in the feasibility study).
Finally, I have assumed that the capital required to bring Donlin Creek and Galore into production will be raised by selling the Ambler property as Novagold’s CEO recently expressed his desire to do. As a result, Ambler’s copper resources have not been included in the analysis below.
Given those assumptions, let’s crunch the numbers. It will quickly become apparent that while Novagold is predominantly thought of as a gold miner, more than 40% of its future earnings will in fact come from copper. So we take Novagold’s 33 million ounces of indicated gold reserves, multiplied by $550 (our $1400 gold price minus the $850 in total costs), and we see that Novagold’s total future income from gold sales is just over $18 billion.
Then we take into account Novagold’s copper resources (which have been meaningfully rounded down to account for getting rid of Ambler). Based on our assumptions above, the net worth of copper will be $2 per pound. Assuming 7 billion pounds of copper reserves yields another $14 billion in future income. Put the two together and you have $32 billion in future income based on today’s prices for copper and gold. But what is that worth to a buyer in today’s market?
Deciding on a discount rate is a pretty crucial decision, and 25 basis points can make a major difference in the valuation for long term assets such as Novagold’s mines. I have tried to be conservative in discounting Novagold’s future earnings in the selection of both the time period and the interest rate. Given that Donlin Creek is supposed to be in operation for approximately 25 years plus four years to bring them to production, I have decided to use 29 as the exponent in discounting future earnings. This can be refined once the feasibility for Galore Creek is complete, but 25 years should be conservative with regard to that project.
As for choosing an appropriate interest rate, I have steered clear of the risk free rate (long-term Treasuries) for two reasons: first, it is at an historical low, and second, it does not adequately reflect the cost of capital of an acquiring company. Instead I have chosen to use a 6.5% long term interest rate. This corresponds to the rate at which a BBB rated miner like Barrick Gold would be able to borrow money, and thus reflects the true cost of capital of a debt financed acquisition.
Putting it all together, when we discount Novagold’s future earnings of $32 billion over the course of 29 years at a rate of 6.5%, you get a valuation of around $5.2 billion. For those of you disinclined to do the math, that equates with a stock price of about $23 per share. That would be a pretty massive premium over today’s stock price, but it is probably around the price where Novagold shareholders would be willing to do a deal. And as we know from Barrick’s attempted takeover back in 2006, bid prices can easily be raised once the bidding begins. So the bidding might well start at $20, but it doesn’t make much sense for Novagold to do a deal at a price below $23.
On a final note, I just wanted to touch on Novagold’s leverage to upside in the gold and copper price. Given the same assumptions above, every $100 increase in the gold price adds some $500 million to the present value of Novagold’s future earnings. Every 50 cent increase in the copper price adds roughly the same amount in present value. That is some impressive leverage, and gives investors a good sense of why this is such a great long term play on both metals. That might be the real reason why Paulson and Soros are still holding onto all of their shares despite the fact that Novagold has already doubled.
Disclosure: Author long NG