Description

Given the recent news around the re-opening of Three Mile Island and subsequent stock movements, I thought it an opportune time to direct some attention to AtkinsRealis (ATRL CN) - in my view one of the best ways to take advantage of the nuclear renaissance that seems to be playing out.

Summary:

• After a decade of issues ATRL is finally a clean story: an engineering and design firm with a rapidly growing nuclear business
• There are several near-dated catalysts to unlock value here, including the sale of the 407 and project wins at Pickering and Cernavoda
• The stock is cheap - once we make adjustments for the now-assured changes, the stock trades at ~8x EBITDA with no net debt
• The nuclear upside here is immense, with the potential for nuclear backlog to grow 5x over the next 2-3 years. 
• ATRL remains off-the-run for the moment, but I believe this could become *the* nuclear stock over the next 12 months

Intro/Business Description

Well, you caught me -- I have written up this company before. In April 2022 I posted a writeup about then distressed SNC Lavalin, which is ATRL's former name. That thesis took longer than I expected but played out well, we are now sitting at my base case price. I closed the position and moved onto other things. 

As a result, for a better description of the business please refer to that writeup, and I also provide a high-level history of the company. I think an important thing to point out is that, at the time, SNC was beleaguered by multiple issues, including the PR nightmare that was the "SNC Lavalin Affair", as well as two disasterous fixed price projects that cratered the income statement and burned over $C1bn of free cash flow. They had also just exited a failing oil and gas E&C business, the balance sheet was stretched, management was new, etc -- definitely a special situation.

Well, the sun is shining again. All of the fixed price E&C businesses have been exited, and the two problem projects essentially delivered. The balance sheet looks good. Cash flow is inflecting. 

So, why am I back? Reason 1:

A steady business growing at 2-5% during my last writeup is now growing at 30%+

Reason 2: As I had hoped during my initial writeup, ATRL has now committed to a sale of their stake in the 407 - one of the world's best toll roads. That process has begun, it is a fairly easy asset to value, we understand the tax considerations well, etc. At this point I think we can back it out of the enterprise value with high confidence. 

Reason 3: When you make the adjustments for the business exits, the stock is almost as cheap as it was last time I looked at it, except with much less risk and much more growth. 

Nuclear Business:

ATRL's nuclear segment makes up about 25% of the pro-forma EBITDA of the company, after removing the Capital (toll road) business, and the Linxon/LSTK businesses that are winding down. Nuclear is an engineering, procurement, development and design consultancy that manages projects across the nuclear power plant life cycle, from newbuilds to ongoing maintenance to refurbishments to decommissioning.

ATRL's nuclear segment controls the IP to the CANDU reactor system, with an "installed base" of 27 reactors globally. Anytime anything happens to one of these reactors, whether its maintenance, refurbishment, or decomissioning, ATRL has to be involved. Most of the current installed base is 30+ years old and reaching the end of its initial operating lifespan, which means they either need to be decomissioned or refurbished. Now, back in the time of my initial writeup, it was assumed most of these reactors would just be decommissioned, but everything has now changed. Life-extensions of these reactors are now the plan, and that means much more business for ATRL.

In my opinion, the CANDU is the best global reactor design. It has some concrete benefits, which include the ability to use a flexible fuel system (increasingly valuable with less centerfuge capacity globally post ukraine war), it can reload fuel without shutting down, and has a fantastic operational track record with zero substantial safety issues across the entire fleet. ATRL is today investing in building out the next generation CANDU reactor called the MONARK. This isn't some brand new design, it merely takes the core principles of the CANDU reactor and updates it with modern components. Many of the non-CANDU reactors in operation today were built with inferior technology due to "national champion" status among the designs - I expect CANDU to take market share as this next generation of newbuilds begin breaking ground. Also, Canada is an easier geopolitical partner for many countries than the US or Russia. 

So, as is shown above, nuclear has accelerated rapidly - why? Well, its simple, a few projects that were going to be decommissions became refurbishments/life-extensions, notably incremental wins at Cernavoda unit 1/2 and Qinshan in China. But these weren't game-changing wins, its just that even one project can dramatically move the needle here. 

There are several drivers of ATRL future nuclear growth:

• More life extensions, with notable near-term opportunity at Pickering, as well as Korean and Argentinian reactors. ATRL expects 19 refurbishment projects on CANDU reactors over the next 10 years worth CAD$15bn, which is 9x the current backlog from refurbishments alone. 
• Newbuilds using the older CANDU design - notably Cernavoda #3/4
• Newbuild MONARK reactors
• Non CANDU work - ATRL is a global nuclear consultancy across a variety of designs, and win projects across all reactor types. I would not be surprised if they booked some of the work on three mile island, for example. There just aren't that many firms with this type of expertise and available engineers

Upcoming Project Wins:

Three potential near-dated catalysts in nuclear:

1. Cernavoda Newbuild Reactors 3&4

The Cernavoda reactor in Romania has the potential/infrastructure for 4 units but only has two operable CANDU units at the moment. The second two were never built due to the fall of the iron curtain. In May 2023, Nuclearelectrica (Romanian utility) notified the EC about a plan to build the remaining two reactors, in the current CANDU design. In July 2024 the EC approved the construction plans.

Canada and the US are offering financing for the construction project, and given the design, it seems essentially impossible that ATRL will not book a massive newbuild contract here in the near term. They control the design that MUST be used, and Canada is providing the financing. Moreover, ATRL was selected for a $750m contract for the life extension of reactor 1 – they are in poll position here. To quote management in June “we are actively negotiating units 3 and 4 today”. The cost is estimated at ~$10bn Canadian, which would likely yield ~$2-3bn for ATRL. This compares to the current backlog of $1.8bn – a huge increase. Announcement could come any day. This will use the older CANDU design so work can begin immediately. 

2. Pickering Refurbishment

The other big near-term opportunity is at Pickering, a Canadian nuclear reactor with 4 units that are set to be refurbished. In February, Ontario Power Generation switched gears and announced that instead of shutting them down they would instead refurbish Pickering’s units 5-8, which are all CANDU designs. The decision was driven by the success of the Darlington Refurbishment (another Canadian CANDU reactor), which was done by ATRL.

ATRL has already booked some “start up” work at Pickering, which increased backlog in 2Q, but hasn’t announced (or included in guidance) the formal refurbishment award. This is a C$3bn project, and I think ATRL is guaranteed to at least win some of it – and could win all of it.

Internal documents that have been FOIA’d by the press mention that resources will be retained and transferred from the refurbishment at Darlington, which is being run by ATRL. Moreover, “Our approach to planning Pickering’s refurbishment is very similar to the method we used to plan the Darlington Refurbishment Project, which remains on plan even as it reaches its final stages.” Once again, poll position for ATRL. This is another announcement we could get very soon.

3. MONARK newbuild booking

ATRL is spending $30m this year and $50-75m over the next two years finalizing the design and components for MONARK, the next generation CANDU design. CANDU historically has a great track record – in fact the best of all the reactor designs in terms of reliability and efficiency. And given Russian designs are now thoroughly out of the market, MONARK could be a good option for a lot of newbuild projects. ATRL's CEO has said he spends 50% of his time in discussions with MONARK potential customers. IR has said they expect to book a new project within the next 12 months.

A new project could be massive – probably $2-3bn for ATRL depending on size and scope. This would also prove out the next generation technology, and open up MONARK for many future newbuilds. COP28 is forecasting 1000 new reactors by 2050, and with Russia out of the market, one would expect share gains for CANDU, especially given the launch of a robust new design. If MONARK just got 5% of these new reactors, that is $C100bn in project bookings over the next 25 years. $C4bn a year would be triple the current nuclear revenue. 

Forecasting Nuclear:

Clearly a lot of moving pieces on the nuclear side and I don't want to try to be overly precise. A few ways to look at it:

1) IF ATRL books only Cernavoda/Pickering, and keeps the rest of its nuclear business flat, ATRL's backlog will grow by 280% and revenue will grow over 20% each year for the next 4 years as these projects come online

2) If you add one MONARK newbuild, backlog grows a further 30% and this adds 10% to growth for 7+ years. 

3) If you start booking multiple MONARK newbuilds (and we apparently need 1000 (!) new nuclear reactors), the numbers begin to get silly and you can grow 30-50% for a decade in this business with no cyclicality and no fixed price risk. And there's really no reason to think this *won't* happen. I feel like you want to own this stock when they book the first MONARK newbuild. 

If they book Pickering/Cerdanova and 1 MONARK, I think 2027 Revenue in nuclear will be ~C$2.5bn, EBITDA of C$340m, growing 20%+.  

Remaining Business and Catalysts

I know I've spent all my time on nuclear thusfar, but the remaining business at ATRL is very high quality. 75% of the pro-forma EBITDA is an engineering and design consultancy specializing in infrastructure with a global presence. ATRL's business has no fixed price risk and grew revenue and EBITDA by 12% in 2024. It has very little cyclicality and comps well to WSP CN and STN CN who both trade around 13x EBITDA. Even if nuclear growth didn't exist it would look like a steal here in my opinion. It has been clouded by the many moving pieces but ATRL has operated this business well for decades and continues to book high quality contracts globally. 

This segment has a C$12.2bn backlog of projects growing 19% y/y. Big trends for this business are aging infrastructure/energy transition/global urbanization. This should be a steady grower for years. 

A few more catalysts on the horizon that could really clean up the story further and allow the market to finally recognize the growth engine that is nuclear:

1. Full Wind-down of two fixed cost problem projects – Eglington and Trillium

The cash flow inflection is finally here, with the company guiding for a significant step up in cash in 2H24 as these projects basically go to zero. They have been delivered and we are in the final commissioning stages – I believe the landscaping is even done at this point. I believe the backlog for these projects is now down to $10m or less and will be zero by end of the year. I think when the market sees the free cash flow inflection occur, it can safely put these items to bed. It’s been a long time coming and while maybe everyone is now comfortable, it will be nice to be done formally.

2. Sale of the Concessions business and closure of Linxon

ATRL’s concession business consists of a 6.5% stake in highway 407 and a few smaller concessions. The 407 is one of the best toll roads in the world. It basically encircles Toronto and is a critical commuting tool for the growing city.

 As an example, a trip from Vaughan to Brampton during rush hour is only 13 miles but takes 90 minutes during rush hour on city streets. It takes 22 minutes on 407 and costs about $5.

As is obvious from the above, the toll road has had a lot of success raising prices, as Toronto grows and its wealth grows, the 407 becomes increasingly valuable.

The valuation is a function of traffic, toll growth and discount rate. The toll growth situation is very good, as the road automatically raises prices by CPI each year, but then has a discretionary kicker thereafter. For example, last year prices went up between 11% and 19% depending on the time of travel.

I conservatively estimate terminal pricing growth of 3% and a discount rate of 6%, which yields about $C2bn after minimal tax leakage due to NOLs (the company has confirmed the tax situation). The remaining concessions probably yield about 400m or so, bringing the total concession value to 2.4bn.

The issue here is solely the timeline. The other partners in the 407 are the Canadian Pension Plan (CPPIB) and Ferrovial (international concessions business). While the contracts here are confidential we assume these parties have a ROFR on any transaction, which makes the dynamics difficult. Upside is one of these players offers a decent price (this happened last time), otherwise the bid process will likely drag until mid 2025. But, the nice part is – this is easy to value, the value probably goes up over time, and a deal will get done.

Linxon is simpler – they have been improving this business and it now makes a little bit of money, but its likely not worth much and I've attached zero value to it -- it is a JV with Hitachi that just mucks up the financials. If ATRL doesn’t get a decent price for it they will just shut it down, further cleaning up the income statement.

3. Eglington/Trillium Settlements

ATRL has two long-standing fixed price projects at Eglington and Trillium in Ontario. These were multi-billion-dollar legacy projects that turned out to be a nightmare for the company due to COVID. Essentially, during COVID ATRL was unable to staff the projects because of many fewer work days, and because of local regulations around disease transmission in Ontario. In addition, the supply chain nightmare created by COVID further delayed these projects.

The Ontario government (both city and province) essentially told ATRL to deliver the projects on the original time and budget, and basically stopped paying them after the time/budget had been exceeded. As such, ATRL ate over $C1bn in cash costs to deliver these projects, which are finally done.

The contracts on both projects have specific carve outs for extraneous events including a global pandemic, although to date the state/local government has refused to settle with ATRL. Not much detail is public, but we do know the claim about is “many hundreds of millions of dollars.” In addition, there are arbitration clauses here which will kick in if settlement talks fail.

ATRL would like to get to a settlement, and each project could be settled separately given they are different government agencies on the back end. Otherwise arbitration will be required, but that could be good – ATRL could get a huge award in that case. The timeline is highly uncertain – doubt we see a settlement in 2024, but 2025 could be a year of meaningful progress here and certainly this is an option not contemplated in today’s share price.

Valuation

The key piece of analysis here is to take the model and make the following adjustments to understand the pro-forma business:

1. Remove the after-tax proceeds from the sale of the concessions business from the enterprise value

2. Remove the concessions/linxon business from historical/projected income statement

3. Remove the fixed price projects, now finalized, from historicals

I think these are reasonable adjustments and reflect changes management has already made, to get a better look at the historical trends and future projects for what the business will actually look like. I've had to make assumptions around earnings and free cash flow, etc:

Taking my 2025 estimates -- which do not really reflect the growth potential in longer-dated nuclear projects, and give them the WSP/STN multiple, you get $81/share, up 50% from here. I think if the market begins to realize the growth inherent in the nuclear business the multiple could/should expand further from there. 

Risks:

1) We don't know the timeline on the 407 sale, these processes can sometimes take a while. But, they have announced a sale and started the process.

2) Execution on margins. The stock sold off a bit last quarter because they modestly brought down nuclear margins. This is just because of some of their new projects had more procurement mix (which is lower margin), but the total EBIT dollars was much higher given the growth. I think their LT guide of 12-14% is reasonable for nuclear and I have assumed it in my model. 

3) Lumpy project bookings -- these are big projects with uncertain timelines

4) Pickering - Westinghouse is being aggressive here, even though this is in ATRL's backyard. They will have to get some revenue here, but maybe split the project with Westinghouse. THat said, zero in the backlog for this at the moment. 

Disclaimer

This document is for informational purposes only. All content in this report represents the author's opinion. The author obtained all information herein from sources believed to be accurate and reliable. However, such information is presented “as is,” without warranty of any kind — whether express or implied. All expressions of opinion are subject to change without notice, and the author does not undertake to update or supplement this report or any information contained herein. This report is not a recommendation to purchase the shares of any company. The information included in this document reflects prevailing conditions and the author’s views as of the date submitted, all of which are accordingly subject to change. This document does not in any way constitute an offer or solicitation of an offer to buy or sell any investment, security, or commodity. Any or all forward-looking statements, assumptions, expectations, projections, intentions or beliefs about future events included in this document may turn out to be incorrect. Any investment involves substantial risks, including, but not limited to, pricing volatility, inadequate liquidity, and the potential complete loss of principal. Investors should conduct independent due diligence, with assistance from professional financial, legal and tax experts, on all securities, companies, and commodities discussed in this document and develop a stand-alone judgment prior to making any investment decision.

Catalyst

1. Booking Cernavoda Newbuild

2. Booking Pickering Refurbishment

3. Final wild-down of fixed price projects

4. Sale of Highway 407/Concessions business

5. Book newbuild MONARK reactor

6. Potential settlements for fixed price project losses