BWX Technologies (BWXT): A Leveraged, Speculative, and Expensive Growth Stock that Might be Worth It
BWX Technologies (BWXT) has been at the top of my research pipeline for a while now. I wrote about the company – when it was the combined company that is now split into BWXT and Babcock & Wilcox Enterprises (BW) – a few years back. You can read my report on the combined Babcock & Wilcox in the Singular Diligence archives. Today, I’m not going to talk about the business – which is described in great detail in that report (see the “Stocks A-Z” tab). Instead, I’m going to talk about price.
I’ve talked before about how I need to check off 4 points about a stock. One: do I understand it? Two: is it safe? Three: is it good? And four: is it cheap? If a stock clearly and definitively fails any of these 4 criteria – it’s not something I’m going to want to buy. Since I wrote a report on Babcock a few years back – and since BWXT is the part of the old, combined Babcock I felt I understood best – I definitely think BWXT is something I can understand. I also think it’s a high enough quality business. The big concern with safety is debt. The company does not have an investment grade credit rating. However, the business itself is very safe and very predictable. So, analyzing the debt load is really just a matter of arithmetic. You can judge that part as well as I can. The more interesting question is price. On the surface, BWXT does not look cheap. It has almost never looked cheap. And so: the quickest way to disqualify this stock would be to show that it is, in fact, too expensive to consider at $53 a share.
BWXT had a missile tube issue last year. The stock price declined. And it hit a low around the start of this year. The stock has since rebounded though. We can look at the year-to-date return in the stock as an indicator of how much more expensive it’s gotten. The stock started 2019 around $39 a share. As I write this, BWXT is at $53 a share. So, it’s 36% more expensive. Obviously, the market as a whole has done well in January and February. But, it hasn’t done anywhere near as well as that. So, we’re talking about a substantial rebound in the stock price here. I had put BWXT on my research pipeline before that rebound. So, the question is: at $53 a share, is BWXT too expensive for a value investor to even consider?
The company has debt. And, normally, I’d start with an enterprise value based price metric (like EV/EBITDA or Enterprise Value / Free Cash Flow). However, I’m trying to eliminate BWXT from consideration here. I strongly believe the business is a good, safe (when debt is kept manageable), predictable business. It might be worth a very high multiple of EBITDA, free cash flow, etc. So, starting with something like EV/EBITDA might give us an inconclusive answer. Instead I’ll start with a measure that is most favorable to the possibility that BWXT might be cheap. Remember, if we prove it’s clearly far too expensive a stock for a value investor to consider – we can eliminate it from the research pipeline right here and now. But, if the answer we get is that the stock looks pricey but also looks a lot better than most businesses – that’s not a definitive answer at all. So, we have to start by giving BWXT the benefit of every doubt we might have. We can then work our way back from that extremely favorable calculation to something more reasonable. And, if BWXT passes the most favorable test – that is, the lowest possible price hurdle I can think for a stock to clear – then, I should keep it on my research pipeline and re-visit the stock at some future date. But, if BWXT seems clearly overpriced even when I give it the benefit of every doubt I might have – then, we can eliminate it right now.
So, I’m going to use EPS. Using a P/E type measure instead of something like Enterprise Value / Free Cash Flow is very favorable to BWX Technologies, because the company has quite a lot of debt.
So, let’s stick to a P/E calculation. BWX is guiding for pretty fast growth. And we want to take that into account. It might have a high P/E ratio this year – but, if it’ll be growing EPS by 10% a year while the S&P 500 returns only 5% a year (I’m using that just for the purpose of illustration here) – then, it could still be a good investment. For that reason, I’m not going to talk about the stock’s P/E ratio in 2018. Instead, I’m going to try my best to look out about 5 years and see what the stock’s current price divided by the EPS it may have in 2024 would be.
As of the last quarterly press release, BWXT’s management said its 2018 guidance was for “non-GAAP EPS in a range of $2.23 to $2.27”. Because this article is just a first hurdle for BWXT to clear – we’ll assume the number is $2.27 a share and that this number is effectively equivalent to a GAAP number. BWXT also reiterated its long-term guidance. This is something the company has been saying for a long time. The exact quote is: “…the company anticipates an EPS compound annual growth rate in the low-double digits over a three to five year period based on a robust organic growth strategy and balance sheet capacity.”
Note that last bit: “balance sheet capacity”. BWXT already has about $700 million in net debt. That’s about $7 a share in net debt (since the company has a little under 100 million shares outstanding and a little over $700 million in net debt). This doesn’t include pension liabilities (which the company also has). So, the $7 a share in net debt figure is a bit conservative. We’re talking about a company where debt is already 3 times expected EPS and they expect to borrow more to buyback more shares. That’s clearly what “balance sheet capacity” means in the EPS growth guidance. In other words, BWX’s debt adjusted stock price would be more like $60 a share – not the $53 a share market price I’m using. We’re giving the company the benefit of the doubt as if it can always keep this much debt. And then management is actually going beyond that and saying that part of its EPS growth over the next 3 to 5 years will be fueled by additional debt growth.
Okay. So, we’re being extremely easy on the company by using a $53 stock price here and by including all of management’s EPS growth expectations – even the part that will be supported by additional debt. What does “low double-digits over a three to five year period” mean?
Again, we want to be favorable to the stock in this first article where we decide if BWXT makes the cut and deserves additional consideration. So, I’m going to assume that low double digit growth in EPS over 3 to 5 years translates into a 10% growth rates in EPS sustained for 5 full years.
Let’s do the math on that. We start with $2.27 in EPS guidance for this year. We then compound that $2.27 at 10% a year for 5 years. That gives us $3.66 a share five years from now ($2.27 * 1.10 ^ 5 = $3.66).
We can then use the $3.66 a share estimate for 2024 and capitalize it at some P/E ratio. Today, the S&P 500 is pretty pricey. Historically, a normal P/E had been around 16. BWXT is a better than average business. Let’s us a P/E ratio of 20 as fair. That’s actually more than fair, because BWXT has debt today and might have even more debt in 5 years. So, $3.66 a share times 20 equals $73.20 a share.
So, how much of a capital gain would you have if you bought BWXT stock today and sold it in 5 years at 20 times an expected EPS of $3.66 a share? That is, you buy at $53 and that becomes $73 in 5 years. That’s a 6.7% compound rate of growth in the stock price. The dividend yield would add another 1% to the stock return. Add 1% to 6.7% and you get 7.7%. Round it up: it’s 8%. So, there you have it. That’s a pretty average – historically – return in a stock. And that’s counting on management’s guidance of (as I translate it) 10% EPS growth for 5 years coupled with an ending P/E ratio of 20. Adjusting the P/E ratio for debt would make it more like 22 or higher (probably higher, since BWXT seems to be planning to use debt to buyback stock).
BWXT is such a predictable stock that we could actually go further than this. For example, the company has given us 30-year guidance – it doesn’t call it that, but that’s kind of what it is – for the U.S. Navy’s expected orders for both submarines and aircraft carriers through the 2040s. Those are real expected orders. We’d have to add inflation on top of that. If we were to do that and extend our EPS estimate out beyond the next 5 years for another decade on top of that – we’d actually get a very, very similar answer. In fact, it looks to me to be an almost identical answer. If you assume something like 3% inflation and include the dividend yield BWXT now has – you’d again get something more than 6% but less than 8% as your annual return in this stock all the way out through 15 years or so instead of just the 5 the company gives guidance for. You can go out a full 30 years – but, that makes the starting stock price less and less important. If you really can predict 30 year growth, ROE, etc. and find it all very satisfactory – then, today’s starting price on the stock is relatively unimportant. Basically, value investing doesn’t work as well as pure “quality” / “growth” investing if you have a holding period of more like 30 years than 15 years. It just doesn’t make sense to think much about value when holding a stock for over 15 years. That fact doesn’t worry me here though – because, I doubt anyone reading this would really buy BWXT today and hold it without ever selling for 15-30 years.
This suggests that – if everything goes right – you might get a return that matches (a pricey) S&P 500 over 15 years or so. But, you’re not going to beat the S&P 500 by buying and holding BWXT at today’s price of $53 a share (as long as we don’t go crazy in assuming Warren Buffett type holding periods of 30 years or something).
Unless…
The company does have other potential avenues for growth. These could matter a lot. Probably not within 5 years. But, it would matter a lot within a 15 year holding period. And this is where the calculation gets tricky. For a company with a low return on net tangible assets – such additional growth avenues wouldn’t change the calculation much. If you have a 10% annual return on equity and a 10% EPS growth rate in the first 5 years and 6% growth rate after that with that same 10% ROE – it doesn’t make a huge difference if the company pays dividends, buys back stock, makes acquisitions, or grows organically. In the long run, your return on equity becomes the stock’s destiny.
But, what if a stock has a pre-tax return on net tangible assets closer to 100% than 10%? Then, the financial engineering – like debt fueled stock buybacks – and the various growth avenues start to matter. See, if BWXT can deliver all those reactors for all those subs and carriers without retaining much of what it earns – then, it can grow EPS faster than what I’ve just laid out. It’s not going to grow faster than “low double digits” since that is only what the company is guiding for even over 5 years. However, I could be quite wrong about the growth rate being as low as 6% (nominal) and more like 3% real over the following 10 years. Or, if the company doesn’t grow very fast – the dividend will grow very fast. The issue here is the company’s very, very high ROE within its core business. BWXT could grow a lot without retaining a lot of earnings. That’s actually the recipe for a successful Warren Buffett type 30 year plus holding period. But, that complication doesn’t just kick in for years 15-30. If a company’s ROE is high enough and it raises its dividend, buys back stock, acquires things, grows in new ways, etc. – you can get surprisingly high 15 year returns despite a high starting price in the stock.
In fact, we know about a couple possible ways BWXT could grow much more than I’ve anticipated over 15 years. There’s a very small contract they have for working on testing the possibility of a nuclear propulsion system for a manned mission to Mars. This is absolutely irrelevant at present. But, if there was a Mars mission planned for 10 to 20 years from now and the propulsion system chosen was nuclear – then, that would become meaningful to BWXT. I have no clue what the odds are of that happening. It’s probably closer to a 5% chance than a 50% chance. But, I can’t say whether it’s closer to a 1% chance or a 10% chance.
It’s so speculative – I just can’t account for it.
There’s another much less speculative – but, still way more speculative than what I usually consider – program that could drive meaningful growth over the next 15 years. This is the Nordion medical isotopes business. BWXT has a plan to produce technetium-99m generators. Technetium-99m is the world’s most commonly used medical isotope. Its used in tens of millions of diagnostic procedures each year and BWXT gives some information on the likely market size (it’s a close to $3 billion global market for the actual end product – the generator market is estimated to be $400 million). The speculative aspects are marketing related. I just don’t know how good BWXT would be at getting this stuff to market and taking share. However, on the technical aspects – we’re talking about a product that is produced from another product (I don’t want to get into all the technical aspects of the chain of production here) that is the result of the fission of highly enriched uranium. Certainly, BWXT is as knowledgeable as anyone when it comes to working with highly enriched uranium.
Highly enriched uranium is anything over 20% concentration of uranium-235. Naturally occurring uranium is less than 1% uranium-235. Commercial reactors for the sort of thing BWXT is now looking into with the medical isotopes would be 25% or so concentrations. The most common civilian nuclear reactor design runs on uranium at less than 5% uranium-235. Naval reactors are probably 50%+ concentrations. While early U.S. nuclear weapons were sometimes 85%. Certainly, BWXT has a ton of experience in highly enriched uranium (BWXT’s work is probably been mostly 50-90%) as compared to other companies that have more experience with low enriched uranium (like 3-5%).
Another question I don’t know the answer to is the fact that BWXT’s planned process might be considered safer from a nuclear non-proliferation perspective. The company claims that’s true. They may be right. Certainly, the concentrations we are talking about here (25%+) would be sufficient to build a nuclear weapon. It would just have to be a very, very big bomb. Historically, weapons grade enrichment has been 80%+ (that’s the concentration the U.S. had enriched to at the time it bombed Japan). So, we’re talking about a number far off from that. But, unlike say nuclear reactors operated by utilities the limitation here is practical rather than theoretical. It’s true that, in theory, highly enriched uranium (though nowhere near weapons grade) could – if you built a weapon massive enough – be used in a nuclear weapon. So, there’s definitely a proliferation concern with greater than 25% enriched uranium (what we’re talking about for the production of medical isotopes) that there isn’t with less than 5% enriched uranium (what we’re normally talking about with nuclear power plants operated by utilities). So, could BWXT develop a process that a government – like the United States – prefers and even privileges legally over competing sources of molybdenum-99 (the product used to produce technetium-99m)?
Maybe.
But, other people are already producing all of the needed supply. They may be doing it less efficiently than the process BWXT plans. But, as far as a proliferation risk – if it exists, it already exists and has existed for a very long time (I think these medical isotopes have been in existence since the 1950s).
What I’m saying is this: the whole medical isotope thing is still very, very speculative. But, it certainly sounds like a very smart long-term strategic move into an adjacent area of competence for BWXT.
The reason why I spent so much time talking about the medical isotopes business is because – if successful – it won’t produce meaningful amounts of earnings during the 3-5 year guidance period BWXT has laid out. In other words, BWXT is predicting 10%+ EPS growth without this business kicking in. Therefore, my prediction of 6% growth in the Navy business from 2024 and beyond wouldn’t be a good long-term prediction for the total BWXT if this isotope business takes off.
For this reason, I can’t 100% rule out BWXT as being too expensive. If you think about how much repurchases, dividends, acquisitions, etc. it could have while also growing 6% a year for about as far as the eye can see (you can look at the company’s investor presentation in June of 2018 for a slide that includes a 30-year U.S. Navy plan that runs through the 2040s) – I can’t rule out 10% long-term returns in this stock even at today’s high price.
But, I think expecting anything beyond about 8% annual returns in this stock is speculative. It might happen. But, at today’s price – I’m not sure that even if you buy and hold for 15 years, you’d really do better than 8% or so.
Nothing I see in the price tells me that this is a safe way to do better than the S&P 500. The company is more leveraged and growth is somewhat speculative. This may be a safe way to expect 6-8% returns over the next 5-15 years. But, I don’t see it as something likely to produce 10%+ returns. I certainly don’t know enough about medical isotopes to bet on strong growth beyond the next 5 years here.
I can’t fully eliminate it today. But, at $53 a share – this is definitely a pass for now.
Also, because it has been several years since BWXT spun-off BW – I don’t really think this stock qualifies as an “overlooked stock” in the sense that it’s eligible for inclusion in the accounts I manage. For that reason, I’m less likely to re-visit this one. But, I think it’s a stock you may want to keep watching. At over $50 a share – I think it’s too expensive. At under $40 a share, it’s a stock you might want to add to your portfolio and hold for the next 5 years.
Geoff’s initial interest: 40%