|Shares Out. (in M):||16||P/E||0||0|
|Market Cap (in $M):||600||P/FCF||0||0|
|Net Debt (in $M):||-300||EBIT||0||0|
Thesis: Revenue for this specialized gas power plant EPC should inflect 3x in the next two years. The company’s backlog has reached $1.4 billion and will likely rise to $2.5 billion as more projects will be won over the next 12 months. As these projects are executed at close to historical mid double digit margins, AGX should generate record levels of EBITDA and earnings for the next several years, starting in FY21 (CY20). The stock should inflect higher as results improve materially starting two quarters from now (Q3.20). AGX is trading at 3x forward EV/E and 7x forward P/E and has 45% of market cap in net cash.
Argan’s largest subsidiary, Gemma, saw its backlog fall to $365m at the end of last year as it completed four power plants. Since then, it has quickly grown to $1.4B with another $1-1.5B in potential project wins within the next 12 months. This should allow for revenue to ramp from its current $200m run rate to over $1B. After speaking with management and industry insiders, we are confident that Gemma will be able to execute on these plants with at least mid double-digit operating margins as a proven best in class operator. This should lead to over $5 of EPS next year and over $6 in the out years. With roughly $19 in net cash per share (the best proxy is current assets - current liabilities) and the stock at $40, that roughly 4x P/E ex-’cash’. That is far too cheap, especially considering that Argan is likely to generate an additional cumulative $20 per share in free cash flow over the next 3 years. In essence, you are getting the business for free 3 years out.
Argan has been written up several times before, most recently as 2016 at around $33 a share, with the stock proceeding to go past $70 less than a year later. The company has not changed much since then so this section will remain brief. This is a holding company with 4 companies split into 3 reporting segments. Gemma Power Systems (GPS) and Atlantic Projects Community (APC) make up the Power Services division, which is typically 90% of revenue and essentially all of the operating profit. GPS is the only subsidiary that matters. The last two projects for APC are in their final stages in the UK, the contracts of which were only worth $100m total combined. One of the UK projects ran into trouble this year but we believe the company took an appropriately large write-off last quarter to bracket the loss. This was likely the reason for the recent drop from $50 to $40. It is important to remember that APC’s is basically a non-core UK asset and performance there does not reflect on GPS’s stellar decade long track record in the US.
The Roberts Company (TRC) was acquired at the end of 2015 and is the entirety of the Industrial Fabrication and Field Services segment. This company probably can have annual revenues of $100m and while the target operating margin is likely mid single digits, last year it was only 2%. The market was not keen on this acquisition when it happened and the potential from it has yet to be realized, but is not core to the thesis.
Finally, there is Southern Maryland Cable (SMC) which is the Telecom Services Segment of Argan. It is even smaller than TRC with revenues around $10m per year since 2013 and is unlikely to move the needle in either direction going forward.
Therefore, the future value of this business rests entirely upon Gemma, the EPC contractor that specializes in large (> 100MW) natural gas power plants. This is where all of the large projects have been won and where the inflection in revenue trajectory will come from. The CEO of this subsidiary (Bill Griffin) is a founder of Gemma, best in class operators, and the highest paid person at Argan and for good reason. Management will be discussed in greater depth below.
Natural gas power plants currently make up around only 16% of power generation in the United States, which just surpassed coal in the past few years. This has more to do with the latter’s rapid decline than an increase in demand for gas plants. However, as coal plants continue to be retired, their power contribution needs to be replaced and often times, renewable options are not sufficient, particularly in the Northeast region of the country. It is hard to make a very long term (20 year) case for growth in natural gas plants but it seems reasonable to assume that they will continue to be built during the next decade as the final 15% of coal contribution is phased out with the historically steady pace of 1% share loss per annum.
The other cause for concern about this space is the difficulty with which power plants are actually built. In less than 2 years, the field of competing EPCs for such projects has shrunk from 6 to 3. Skanska, Aecom, and Fluor all exited the space and Jacobs Engineering sold off its business. This has left Kiewet and Gemma as the two prominent players left. This shrinkage has happened as the number of projects declined and competitors overbid for plants. Fluor’s management team even came out and said the entire space is unprofitable and EPC’s working on these power plants aim to break even. These competitors exiting has allowed Gemma to build it largest backlog ever during this resurgence in the number of projects being planned. However, if competitors claim these new projects are unprofitable, how can Gemma hope to create value? This became a key question to determining the outlook for Argan.
Value Proposition / Competitive Advantage
Gemma has survived the trough in the power plant cycle and is poised to succeed on these future projects due several advantages developed over time:
Management discipline: they walk if cannot meet mid double digits margin projections. Turnover in core staff is low. Their average Power Services EBIT margin is stable and confirms the discipline: 17% last five years, 19% last 7 years, 16% last 10 years. There are a few low project years like FY20 (now) and FY10 as they cannot control the timing of ground breaking on projects. Management is keenly aware of rising labor costs and have bid on projects as such. To be conservative we assumed EBIT margins in power peak at 15% not the prior Panda highs of over 20%.
Great project execution through specialization: only does gas power plants and builds the simplest and most efficient plant that matches the specs; “lead with the construction in mind and not let the engineers run wild”.
A recurring customer base which prefers to work with this proven operator: competitors that left the business were working with public utilities for the most part. Gemma is willing to work with such customers, but often prefers power merchants due to better aligned incentives and faster decision-making.
Development-stage financial involvement in the project: leads to success fees and higher margins.
Lean operations: Gemma hires or subs out construction staff for each project. They have recurring labor pools in each geography they work with repeatedly. They mostly work with non-union labor. They furlough staff with some pay between projects to keep them available.
We spoke with their biggest competitor Kiewit. One executive said that any time they find themselves bidding against Gemma, they expect to lose. Gemma is very selective with the projects it chooses to bid for, but when they do, they typically win the contract. Now with so much competition gone, more projects are viable, leading to the growing backlog.
Currently, only 3 projects make up the majority of the $1.4B+ backlog (Reidsville NC $250m, Guernsey OH $900m, Harrison WV $325m). As previously mentioned, the two UK plants that are being finished contribute very little to this number. Channel checks indicate that Argan should receive the full notice to proceed (and break ground) on the large Guernsey project in the coming weeks and will start to contribute a meaningful amount of revenue in the 3rd quarter. This should be followed shortly after by a full notice to proceed for the smaller Harrison WV plant.
Furthermore, there are over $3B or so in potential projects planned to start in the next 18 months that have yet to announce an EPC. Given past relationships with various owners and developers, Gemma has a good chance to win at least $1B of that. We are tracking the following projects as likely additions to the backlog:
Chickahominy project, which Argan has already won, but has yet to receive the limited notice to proceed and therefore cannot add to its backlog. This is a massive project with $800m in value to the EPC (may be more with turbines) that just received final permitting and can close on the final step of financing.
Brooke County WV: $400m project
Harrison County OH: $525m project
There are multiple other projects we track that are further out 2020+ timeline. As Gemma starts to work through the existing book, we believe the backlog can surpass $2B later in 2019 and likely go in the mid $2B next year.
Gemma is not the only lean operation in this conglomerate, and all of Argan aims to keep its overhead as low as possible. There are only 4 people that work in the corporate office. This includes its longtime CEO, Rainer Bosselmann (76), who owns over 3% of the company. He is a deal maker that owned companies in the technology and textile spaces before this shift to construction decades ago. Rainer knows his businesses intimately, but allows the management teams of subsidiaries run their companies as they see fit, especially with Gemma. We think eventually Rainer will sell the company.
As mentioned before, Gemma’s founder and CEO, Bill Griffin (64), is the highest paid person in the company with a roughly $5m salary in good years. He owns about 2% of the company. He is the anchor to Gemma strong culture and has navigated the business through countless cycles.
Finally, there is David Watson, who joined in 2015 after Rainer’s CFO and partner of 20 years retired. He joined from a business development company and came on right before the TRC acquisition. David also runs a bare bones investor relations effort. Argan does not host earnings calls, put out investor presentations or entertain much sellside coverage. This management team is entirely focused on running its businesses and creating value.
Overall the management team owns over 8% of the company. Dimensional owns 8% and Wellington 8%.
What Happens Next
Due to a -65% decline in revenue in Q1 and an almost $30m writedown from issues with APC’s UK projects, AGX has given up all of its gains since it started to rebuild its backlog earlier this year. However, this is almost the exact same set up as the spring of 2016, the last time we owned the company. When the stock sold off on news of the TRC acquisition and a Q4 that saw the 3rd consecutive quarter of declining Gemma revenue. At that time, the backlog was just over $1B and cash per share was around $11. The stock was around $30. The following quarter, revenue from the backlog started to come in and Gemma grew sales 30% y/y with a 20% operating margin. From there, AGX doubled in roughly 6 months and nearly hit $75 at its peak.
This time around, the entire set up is even better with cash per share double what it was then and a backlog that could hit $2.5B. Even with more conservative margins projected, this should result in nearly twice the earnings generated from FY16-FY18. Once revenue starts to flow through the income statement, the stock should inflect.
We view Argan as one of the best opportunities in this late cycle economy as it is at the low point before a large profit inflection and is uncorrelated with the overall market. However, valuing it is not easy given the large upcoming profit increase and the large cash balance but we think it’s conservatively worth $70-80/share. Note that the stock peaked at $70 last time (early 2017) and now we project the earning power and cash cushion to be almost double. There’s no close comp to this so we are using a wide set of STRL, TPC, PRIM, MDR, GVA, KBR, SNC:CN, FLR.
We look at valuation in multiple ways:
Simplistically, you are getting the business for free now at $19/cash + $20/share in earnings in the next 3 years = $5.60 in FY21, $7.50 in FY22, and $7.10 in FY23 (FY20 is irrelevant and we ignore it but we project $1.75.)
Earnings will not collapse after FY23 as we have not counted 3+ other plants that we know of right now and more will be added in three years time. The company could be making $6/share for a while. At 10x and assuming half of the current net cash is excess you get $70. Comps trade at 9-12x forward P/Es and have substantial debt.
Applying a trough 8x multiple to $7.50 of peak earnings plus future excess cash you get to at least $80/ share.
Applying 6.5x EV/E and current cash (which will go up) you get to $70-85/share using our EBITDA projections of $135m, $180m, $170m. Comps trade at 5-9x EV/E.
Full Notice to Proceed on Guernsey, OH project and then Harrison, WV should cause revenue to grow massively in Q3. A very small portion may hit earlier, but we expect Q2 (ends July) to be the last low quarter before the inflection point.
Future very likely contract wins announced on Brooke County WV and Harrison County OH inflect revenue further in Q4. We also believe that Reidsville starts generating revenue by Q4
Future likely projects wins like Chickahominy (gets financing done), Anderson County SC, Windham CT start contributing next year
Return of some excess cash to shareholders via a special dividend once working capital needs for major projects are met (could be as early as this year).
Project financing risk. Due to permitting and other delays multiple large projects are competing for financing this year. Our channel checks show that this is progressing as debt financing is being raised and equity follows, with turbine manufacturers pitching in as well.
Ability to handle 6-7 large projects simultaneously. The company has done this before in FY 2016-2019 with the Panda projects.
Project execution risk. Cost overruns would materially harm project profitability and lead to lower margins and possible penalties. Gemma has a stellar record of execution so we think this risk is low.
Much higher natural gas prices could make such plants uneconomic. Again we think this risk is low.
Post coal replacement cycle natural gas will eventually give way to cleaner energy sources. We think this is 10-20 years away.
See Catalyst Section Above
|Subject||Exelon build gone bad|
|Entry||08/13/2019 07:50 PM|
What happened there?
can u roughly translate what $1.4bln in backlog means for annual revenues?
How specialized are these power plants or are they basically the same box for every build? Ie, the more the plants are built alike the less likelly to run into any problems.
Also, what about the other companies invovled in these plants ( eg, https://guernseypowerstation.com/project-development-facts/ ). Who pays AGX? Apex? How does Apex or others affect Argan's ability to execute?
|Entry||08/19/2019 02:15 PM|
Zipper, the Teesside project is being built by Technicas, a significantly larger EPC. APC (Argan’s UK subsidiary) was a subcontractor for Technicas with a limited scope of their assignment around $75 million and they were building the boiler of the plant. At this point, AGX took close to a $30 million charge on the project. Given that it was a first estimate (though described as well scrubbed), it’s possible that AGX may need to take additional small charges -- maybe $5-10 million, it’s hard to know.
The Teesside problem is not helpful to the story in the short-term and attracts attention to a bad project in an environment where other E&C are having large losses. However, we do not think it should matter in the medium or long-term for Gemma GPS. Teesside is supposed to be the world’s largest biomass plant, whereas Argan’s bread and butter core competency (and everything in the backlog) is the natural gas plants in the U.S.
|Subject||Re: Exelon build gone bad|
|Entry||08/19/2019 02:15 PM|
The Exelon West Medwey II project initially was supposed to be completed by mid 2018. GPS had not done business with Exelon before this project. Exelon is asserting timing delays. GPS says the delays were caused by Exelon and also led to cost increases, and GPS reached “first fire” on both power generating units at the plant, so basically finished the project. They claim that only performance tests and other checklist items were left to achieve substantial completion of the project. Thus in Jan ‘19 GPS sued Exelon for breach of contract (didn’t pay) and has $17m receivable. Management thinks this case will be tied up in court for about a year.
As far as backlog conversion to revenue, we think that Argan is set up to have three $1 billion+ years starting in FY21 (ends Jan ‘21). This comes from the $1.4B in stated backlog as well as 4 other major projects ranging $250-700m as described in our write-up. We believe the backlog will increase over $2B in the next 2-3 quarters.
GPS tries to design the simplest and most efficient plan that matches the specs. These power plants have a generally consistent build with some bespoke engineering based on certain conditions that have to do with the clients’ wants and needs as well as geography. As a simple example, interconnection systems can use either water or air cooling. Overall, we agree that the more plants Gemma builds, the less likely problems are to arise. Most of these upcoming projects are for owners that were previous clients, which also increases our confidence in smooth execution.
Below is an indicative chart that shows you all parties involved in a given power plant build. Gemma is a contractor (upper right corner) that gets paid by the plant developer/ owner, which is Apex in the case of Guernsey. The developer can affect the contractor’s ability to break ground by delayed notices to proceed due to permitting and financing issues. Once the contractor has broken ground, the plans are typically all sorted and it up to the contractor to finish with the agreed upon specs and before the set end date. Payments are made along the way as milestones are delivered. The developer secures the equity and debt financing and controls the payments along the way on strict contractual terms. At times Gemma gets involved early on with development loans for repeat clients.
|Subject||Re: Re: Exelon build gone bad|
|Entry||08/19/2019 02:43 PM|
|Subject||Re: Project Starts/Financing/Risk|
|Entry||08/21/2019 05:14 PM|
Q: 1. Do you know if Guernsey or Harrison have closed on their equity financing? Any sense as to who is the likely financier?
For Guernsey our channel checks suggest that the entire financing is about to close. We heard the debt was 2x oversubscribed in late July with 20+ lenders involved. It was a $950m L+300 5 year construction loan and 125m revolver. Investec was the bookrunner. We heard it was syndicated (ICBC, KB Kookmin, Nomura, China Merchants, etc.) suggesting again that the equity was circled. MS is equity advisor and we heard Global Infrastructure Partners and John Hancock among possible investors. We expected groundbreaking by August but it slipped to September. We think this is the next FNTP which should start to unlock the value in the stock.
Harrison County (WV) is supposed to be just behind Guernsey. It has all the permits and is working on the financing but we do not have details yet. We expect the FNTP on this in Q4 or next year.
2. Any idea of what is going on with Reidsville? It has been in backlog for a while but to my knowledge, Gemma has not done any work yet.
Reidsville is smaller and we heard the owner was waiting on an additional PPA to get the equity financing. They have at least two PPAs signed already. NTE did the same for their former Kings Mountain project in NC in 2016. NTE is a repeat customer and Gemma have high confidence it will get done. We heard Credit Agricole and Sumitomo are lead debt arrangers for a $660m package. Turbines are from Mitsubishi and we heard they would be participating in mezz or equity. Guggenheim Partners did the equity for Kings Mountain but we do not know if they are involved in this deal. We got the sense that GPS would not care if this slips into year end given the other bigger projects expected to close in late summer-fall. We expect FNTP on this in Q4 or next year.
3. What the Chickahominy??? They announced this project 14 months ago and it has yet to go into backlog. I thought receiving the air permit (which they did 2 or so months ago) was the final hurdle. Any idea if they closed on the equity financing of this one either?
Chickahominy’s air permit (now received) was also what we were waiting for in the short term. A lot of people are focused on this one given the larger size and the length of time it’s been discussed, but this project is still not in the company’s backlog, reflecting a slower expected timeline. Turbines are from Mitsubishi and again we heard they would be participating in mezz or equity. The Balico team is ex Cogentrix and very experienced. Management hopes to get the FNTP in the Fall but in our projection groundbreaking starts next year.
There are two other projects (Brooke County WV and Cadiz OH) which we think may get started even sooner.
Overall, it’s fair to say that the equity financing is the most difficult component of the process. In addition, the required steps have generally taken longer than we initially expected. Nonetheless, we think that the majority of the projects will be completed.
|Entry||08/22/2019 10:25 AM|
Any idea what their plans are for the cash?
from the business description in their filings:
At the holding company level, we may make additional acquisitions of and/or investments in companies with potential for profitable growth. We may have more than one industrial focus. We expect that acquired companies will be held in separate subsidiaries that will be operated in a manner that best provides cash flows and value for our stockholders.
|Entry||08/23/2019 10:01 AM|
They require having net cash (in this case the right proxy is current assets - current liabilities) for bonding requirements as a small percent (e.g. 12%) of outstanding contracts. They now have serious excess cash. We think they may pay a special dividend of say $5/share, perhaps this fiscal year. The amount of excess cash will go up in future years as cash flow comes in. There’s a positive dynamic here if some of projects do not get underway - their excess cash in that case could be $10-15/share on lower bonding needs. We think it is very unlikely that they make an acquisition.