Description

Finally getting the chance to formally write this one up. You can view my previous comments on MAXR under glgb913's MDA write-up.  

Summary:

Maxar is a leading space company that has multiple catalysts to drive a potential 300%+ re-rating. The stock has moved strongly from the bottom but still has only recovered a fraction of its former price. A new satellite constellation to launch next year will dramatically reduce capex and provide for new growth opportunities. In addition, the space infrastructure business is in the early innings of a turnaround. As a result, Maxar should see multiple benefits beginning in 2H21, increasing cash flow and reducing leverage ultimately causing a re-rating in the stock.

High Level Points

1)      The equity has 3-5x upside from the price today

2)      The imagery segment has sole source contracts with the US Government

3)      Post the launch of Worldview Legion in 2021 the capital intensity of the business will decline

4)      The Space infrastructure business is growing again

5)      Will see a transfer of value from debt to equity as de-leveraging plays out

6)      One of only a handful of pure-play space companies that trade publicly

7)      Maxar is an M&A target

Background:

Maxar was created through the acquisition/merger of MacDonald Dettwiler (MDA) & Digital Globe (DGI) in 2017. MDA owned a Canadian space robotics & communications equipment business and Space Systems Loral (SSL) a manufacturer communications satellites. Digital Globe is a provider of satellite imagery and data services for Government and Commercial applications. At the time of the acquisition, MDA wanted to reposition its business towards the US Government and owning DGI was a key piece to becoming further ingrained in the US Government space & intelligence workflow.  However, MDA took on a large amount of debt at a time when SSL was beginning to suffer as order flow for GEO communications satellites began to slow meaningfully. This caused SSL EBITDA to decline and as a result working capital became a large cash drain on the business. Leverage began to grow and the equity started to compress rapidly as this slowly became a potential bankruptcy situation.  The failure of the Worldview 4 satellite which was a growth driver for the imagery business was the final nail in the coffin and the stock bottomed at $4/share from ~$65 a year prior.

Soon after, Maxar replaced the management team and refocused on the core imagery business while working to return SSL to profitability and reduce leverage. In 2019 Maxar reduced run-rate SG&A by $65 million and in 4Q19, three significant news items were announced; 1) Maxar sold non-core real estate in Palo Alto for $291 million 2) Maxar issued $1 billion of new secured debt to push out near term maturities 3) Announced the sale of the MDA robotics business to a Canadian buyer for $765 million at a multiple much higher than they were trading at the time (closed in 2Q20). These three items took bankruptcy risk off the table.

Today Maxar operates in two segments:

1)      Earth Intelligence – The former Digital Globe business which includes satellite imagery + data analytics for Government & Commercial customers

2)      Space Infrastructure – Consists of SSL and focused on the manufacture of satellites and other space programs

Detailed Thesis

1)      New satellite constellation Worldview Legion is a game changer:

Historically, the average fully launched observation satellite cost was about $650 million (taking an average of WV1-4). This was the full cost for launching a single satellite (including insurance, launch, ground station, startup infrastructure, etc..) which also carried a tremendous amount of risk to the company if one of those satellites failed as there is little redundancy. We saw this with the loss of WV-4, which forced the company to move customers to a similar resolution satellite (WV-3), but missed out on the future opportunity to fill up empty capacity on WV-4.

The fully launched cost of WV Legion is approx. $600 million and includes 6 satellites. This is a game changer to the company for multiple reasons:

A.      Lower capex & more predictability in the business model

A permanent reduction in capex will also lead to more predictability in the business. Legion is expected to go up in two launches in 2021 and I expect capex to decline towards maintenance levels ($125 million) beginning in 2H21.  Typically a capex holiday follows a large launch for two to four years as the company harvests the new operational satellites, before ramping up capex for the next satellite build. With 6 satellites in the constellations vs 1, Maxar will have no need to plan a large new expensive constellation. As capacity is required, Maxar will add additional satellites to the constellation, which costs much less than $100 million each. In addition as I will explain in my next point, Maxar may not need its own balance sheet to build new satellites in the legion constellation.

B.      Potential new business opportunity (Satellite-as-a-Service Model) will extend lead vs competition

Maxar currently has contracts with 12 foreign governments and is allowed by the US Govt to get engage with up to 15 on an approved list. Foreign governments task capacity on Maxar’s existing satellites to quickly receive imagery in certain strategic areas of the world. For foreign governments there are limits to satellite access as they have to share with other customers. Through Maxar’s SSL entity, the company can engage with foreign governments to build a legion satellite and the imagery side of the business can manage it for that government customer exclusively. This would allow Maxar to add an additional satellite to the constellation at little cost and increase data collection that can be used for other customers in their network, further widening their imagery moat. In addition, this would have the benefit of increased overhead absorption at SSL. Currently no competitor offers 30cm imagery even though Airbus is working on their own 30cm constellation called Pleiades Neo which is expected to launch ahead of Legion.

C.      Improved revisit rate + redundancy 

Small satellites in low earth orbit are characterized by high revisit rates but work at low resolution and image less landscape. Legion will have a revisit rate of up to 15 times/day which will allow Maxar to compete with the revisit rates of small sats. We have seen some very public missteps at small satellite companies and post legion, I can’t see small satellites competing head to head with Maxar. Small satellites will likely become more of a complementary offering. In addition the increased revisit rate will allow for better data collection for autonomous driving applications

Legion video: https://www.youtube.com/watch?v=hmp06v9i0es

Autonomous driving:  https://digitalglobe.wistia.com/medias/04pvm8ejic

A loss of a Worldview legion satellite won’t be as impactful to the company given the lower cost of replacement and redundancy within the constellation. This flexibility only increases as Maxar adds new satellites to the constellation over time. When Worldview 4 failed the company took a $162 million impairment to write down the book value of the satellite, but received $183 million in insurance proceeds. Although the proceeds were a nice offset, the missed opportunity to produce revenue on that satellite dramatically outweigh the benefit from insurance.  

D.      Massive de-leveraging opportunity 

Legion will bring incremental capacity over areas of interest  and allow for new sales models which should increase EBITDA meaningfully at Maxar’s imagery segment. The Earth intelligence segment (Imagery) is expected to produce over $500 million of EBITDA this year. I expect legion to add an incremental $100 million of EBITDA through 2023 ($50 million/year in depreciation + $50 million in EBIT). This does not include additional US government service add-on contracts or incremental business from foreign governments which could add another $50 million in EBITDA. I expect 2023 EBITDA in this segment to approximate $650 million, growing at a 9% CAGR from 2020.

With permanently lower capex and visibility into EBITDA growth, I would expect a very meaningful re-rating in the equity as leverage declines and cash flow grows. Maintenance capex is in the $125 million range vs capex spend of over $300 million today. This will free up $175 million of incremental cash flow + another $150 million from EBITDA growth at Earth intelligence = $325 million cash flow swing. The company is already bearing the interest expense for the legion build and taxes will likely be negligible due to existing NOLs. Also this does not include further opportunity at SSL which I will touch on later.  

E.       The US Government will renew the enhanced view contract at an equal or higher value 

When Digital Globe was an independent company there were a series of short seller write-ups (of which you can find here on VIC). One key part of the short thesis was that the enhanced view contract which at the time made up over 50% of revenue was going to get competed away by emerging small satellite companies. Today the enhanced view contract is about 25% of revenue. The contract has yearly renewal options and expires August 31, 2023. At some point in the next few years this contract will have to get renegotiated. The US government has been trying to work with small satellite companies given the high revisit rate which is useful in situations where information needs to be updated quickly even though the quality is not as high. Legion will provide an option for the government to receive the highest quality imagery more frequently (15x/day). Currently no small sat company has the capability to compete with Maxar’s Legion offering.  With no technically comparable competition (https://blog.maxar.com/earth-intelligence/2019/satellite-constellation-choices-for-earth-observation-business-success), the US Government will likely renew the contract and increase the scope where Maxar will serve as the prime and likely integrate other small sat companies’ complementary offerings into their platform. Pricing will likely be higher given the increased scope which will likely include software & more powerful data analytics. It is possible that the government would be willing to fund their own private legion constellation which would be another big positive for Maxar.

F.       Vricon acquisition 

Maxar started the Vricon JV w/SAAB in 2015 where Maxar contributed satellite imagery while SAAB contributed IP for 3D map modeling. Recently, Maxar purchased the 50% they didn’t own for $115 million (10x trailing EBITDA). Vricon is a SAAS subscription business that uses Maxar’s imagery to create photo-realistic 3D images. Most customers are in defense & intelligence, but there is an opportunity to broaden its appeal to commercial customers (ie: Telecommunications for 5G placement in urban centers). Legion will feed more imagery into this Vricon application which in turn will produce more 3D model data output and provide a more valuable offering for customers.  This offering is not available anywhere and is proprietary to Maxar.

In 2019 Vricon did $40 million of revenue and grew 90% YoY. I would expect this business to grow rapidly over the next few years at very high incremental margins which should support growth in the Earth Intelligence Segment. The Vricon acquisition will allow Maxar to consolidate and report cash flow coming from the business onto their financials, previously the profitability was backed out of the cash flow statement given an inability to control cash distributions from the entity.

2)      Space Systems Loral (SSL) will no longer drag on profitability and the outlook is changing 

The sell side is imputing zero to negative value for this business. Other than the loss of WV-4, the decline in the GEO sat order market combined with high debt was the main reason the stock fell precipitously from $65.  Now the GEO sat order market appears to have bottomed and there are prospects that have dramatically improved its outlook.

A.      Backlog beginning to grow (beneficiary of C-Band spectrum sale)

The space infrastructure segment continues to be impacted by lower GEO sat orders in previous years. In 1Q20 book-to-bill was above 2.5x as the company booked orders for 4 GEO communication satellites to be delivered in 2022 for Intelsat. These bookings ($700 million) are expected to be cash flow positive over the life of the build and guarantees a second year of book-to-bill over 1x. This should allow them to reach profitability in future years as explained by the CEO on their 2Q19 call:

“our near to medium-term goal is to shape the Space Solutions talent base so they can run at breakeven adjusted EBITDA with $500 million in annual revenue. And we think we can accomplish this with one to two 1300-class orders a year along with a steady drumbeat of other civil, commercial, and government work across our Legion-class and robotics offerings.”

B.      C-band opportunity could help bridge the gap for a refresh in GEO orders 

The FCC's recent actions to clear C-Band spectrum for future 5G applications will require new GEO satellites to cover U.S. territories; spectrum license owners in U.S. (Intelsat, SES, Telesat) will be spending around ~$2.0BN on these assets by FY2022 and Maxar has historically garnered 1/3rd of GEO orders. Intelsat needs seven new satellites to facilitate the C-band transition. Maxar has already won four of these satellites, Northrop Grumman won two, and there is one more outstanding to be awarded. In addition, there are three more satellites left to be awarded; two for SES & one for Eutelsat – these will be required to be built here in the US given the political sensitivity around the C-band process.

More information here: https://spacenews.com/c-band-alliance-pegs-satellite-spectrum-clearing-costs-at-3-3-billion/

There has been a delay in new GEO orders partially because of confusion over rapid technology changes due to emerging LEO constellations. These constellations have yet to be fully proven (ie: Oneweb collapse - http://interactive.satellitetoday.com/via/oneweb-special-edition/industry-leaders-sound-off-on-onewebs-collapse/) and likely will not be able to replace GEO sats for certain applications. It is possible that SpaceX’s Starlink constellation will render existing satellite businesses obsolete, but that has yet to play out.

I found comments from Eutelsat CEO Rodolphe Belmer to be interesting regarding the GEO market:

“we have not changed our view in terms of the advantage of GEO satellites for the time being. Even in terms of connectivity services, particularly with (Very High Throughput Satellites) VHTS technology, we think they are by far, the best value proposition for the telecoms market. This is because of the kind of price per Mbps they provide as well as the massive capacity they can bring. They can unlock demand in the mobility segments as well as the broadband to consumer segment. We are truly convinced of this.”

“We think, for the moment, that LEO constellations come with significant economic challenges as proven by what has happened to the likes of OneWeb. It is very difficult to derive strong profitability from LEO projects at the moment. It does not mean LEO is not an interesting technology, or that LEO won’t be integrated in the overall communications portfolio in the longer term. We think LEO will be effective in delivering some services and complementing GEO. The issue is timing. Today, the economic challenges of LEO are very, very difficult to overcome.”

Source: https://www.satellitetoday.com/business/2020/05/20/eutelsats-belmer-hopeful-of-bright-future-after-promising-results/

The GEO replacement cycle will happen at some point but likely not until 2022+. It typically takes 3-4 years to build a GEO sat and has an operational useful life of around 15 years. From 2006 – 2010 approximately 150 GEO sat contracts were awarded. Assuming a four year build time and 15 year useful life, orders to replace these sats will likely come in 2022-2026. If SSL was to win 20% of these orders (a lower share than they historically received), they will likely win around 30 new contracts or a cadence of 6/year well higher than their 2/year breakeven.

C.      Potential to win Telesat LEO Constellation 

Maxar already has a deep history building GEO & LEO constellations. The company has built more than 280 satellites. Over 90 geostationary satellites are currently on orbit and nearly 100 small satellites for low earth orbit (LEO) have been built or are under construction.

Telesat is looking to build a LEO constellation of 298 satellites and have narrowed a winner down to Maxar or Thales/Airbus partnership. It is expected that this contract can cost upwards of $3 billion CAD and multiple launches are expected to be in the 2022-2023 timeframe. This win would be not only improve the operating performance of SSL but demonstrate that Maxar has a strong capability in manufacturing LEO satellites. Over time this could make SSL agnostic to GEO or LEO constellations.

D.      Civil Space program awards 

Last year, Maxar won a $375 million contract with NASA for the power and propulsion element (PPE) of the Lunar Gateway. This is a development award and is expected for a launch in late 2022. In January a $142 million contract was also awarded to Maxar for technology to demonstrate in-space assembly using a robotic arm (Spider). This is part of a larger Maxar program (Restore-L) which looks to refuel spacecrafts.

Another potential opportunity is Maxar’s partnership with Dynetics for the Human Landing System, which if selected, would be a key part of the NASA’s strategy of returning to the moon on a more permanent basis.

These are just a handful of awards that position Maxar’s space infrastructure business as a key asset aligned with NASA’s strategy over the next ten years.

E.       Burn off of bad contracts 

As the GEO market was beginning to collapse, the previous management team aggressively pursued business. The CEO was fired over mismanagement and now the percentage of revenue from underperforming contracts is set to run off by 2021. Underperforming contracts at SSL accounted for over 25% of revenue in 2019, 15% of revenue in 2020 and will be <5% in 2021. This should help turn the corner towards positive EBITDA with a goal of getting to 10%+ EBITDA margins.  

3)      M&A Target

The sale of MDA allowed Maxar to become a majority US business with little ties to Canada. In 2008 Alliant Techsystems (acquired by Orbital Sciences and now part of Northrop Grumman), tried to buy MDA for $1.3 billion but was blocked by the Canadian Government. Today, a potential Maxar bidder would have little risk of being blocked particularly if it is a US based acquiror as a result of the MDA divestiture.

There are many companies that would be interested in acquiring Maxar which would provide strong exposure to the space economy.  The FY21 budget request saw bi-partisan support for 30% growth in funding space related priorities. With space force emerging as a new division of defense and NASA’s long term goal of putting a man on Mars, there are expectations for further government spending.

Northrop Grumman (NOC) – Acquired Orbital ATK for 14.5x EBITDA in 2017. Their space business is focused on the manufacture of satellites, launch vehicles and classified payloads. Approximately 20% of revenues are generated from the space economy.  They potentially would be interested in acquiring Maxar to consolidate the satellite manufacturing industry and to gain exposure to new intelligence contracts for imagery. Northrop also maintains a strong balance sheet and could easily purchase Maxar in a highly accretive deal given the cost of capital difference.

Boeing (BA) – They have a large defense business with satellite manufacturing capabilities mostly for defense customers. Airbus has an imagery business that competes with Maxar, but Boeing is absent from this market. In order to more directly compete with Airbus, Maxar could give them the opportunity to scale their satellite manufacturing business and gain exposure to higher growth imagery segment. However, given issues with the 737 Max program and the commercial side of their business, this likely would not happen for some time, if at all

Lockheed Martin (LMT) – Also have their own satellite manufacturing capability but could leverage SSL to gain scale and take out a competitor. Their make-up is similar to Northrop in that ~20% of revenue comes from space with a focus on launch vehicles.

L3Harris Technologies (LHX) – Approximately 17% of revenues come from space with a focus on small satellites and other communication payloads. An acquisition here make a lot of sense given LHX’s exposure to defense communications, intelligence & surveillance.  The imagery segment could bolster those capabilities and the satellite manufacturing business could vertically align LHX closer to communications customers.

SpaceX – SSL could boost their small sat starlink build. In addition they could leverage some of Maxar’s existing technology/R&D work and package it together with launch to provide a unique offering unmatched by the competition. They would be a fully vertically integrated space player – could build a satellite/payload, launch it, and manage it throughout the life of the asset. Also Tesla is striving towards autonomous driving and Maxar’s imagery capabilities could help them advance that initiative further.

Amazon/Blue Origin – Has already won the award to launch the LEO constellation for Telesat. Could be interested in acquiring Maxar for some of the same reasons as SpaceX – to vertically integrate. Maxar’s contracts could help the company get closer to NASA & defense partners. Amazon has aspirations for autonomous vehicles (recently acquired Zoox) and imagery data is key for autonomy. AWS is already a partner to Digital Globe. 

Private equity or management buyout – Could see a situation where they use their cost of capital as an advantage to buy the company and harvest the cash flow given the end of a capex cycle early next year. After owning for some time, can reintroduce this to the market with a better balance sheet and refined growth opportunity.

4)      General Corporate & Capital Structure Opportunity

A.      Recent debt raise & ability to structurally lower interest rate in 2022 

Maxar recently issued $150 million of 7.54% senior notes due in 2027 and used the proceeds to repurchase $150 million of 9.75% fixed 2023 notes at 112.45% of principal.  This transaction will lower Maxar’s interest expense and extends the maturity profile. There is approximately $885 million still outstanding on the 9.75% notes that are due in late 2023. Currently, they trade at 110 which implies a 6.5% yield. Post a Legion launch, Maxar will begin producing meaningful levels of FCF and will likely call the rest of the 9.75% notes and lower interest expense further. At that time I would expect Maxar to refi in the 6.5% range consistent to where their existing bonds trade which is an overall savings of almost $30 million/year (VSAT bonds trade in the 5-6% yield range). These savings will help further accelerate de-leveraging.

B.      Canada dual listing removal will improve US liquidity 

Currently, Maxar trades on the TSX & NYSE as a result of MDA originally being listed in Canada. This is unnecessary today as Maxar is a US based company. A consolidation of the listing onto NYSE could bring renewed interest and further liquidity in this overlooked name. If this was to happen, I could see Maxar doing a secondary roadshow to place Canadian holder shares that need to be sold with longer term oriented US based institutions.  This would also serve as a catalyst for more US based sell side firms covering the name (only one US based sell-side analyst follows Maxar).

C.      Strong management team dedicated to the company 

Maxar has enhanced the board of directors and changed the management team. The new CEO Dan Jablonsky was promoted after spending over seven years at Digital Globe. He has a background in the US Navy as well as SEC legal experience. Moving Dan into this key strategic role repositioned the company to focus on the Imagery business. Dan has successfully transitioned the company away from a near bankruptcy to a return to growth. The company also replaced the chairman of the board with a former Digital Globe alumni as well as added a former member of Digital Globe’s board to the directorship.  In 2019 the board redesigned the long term incentive plan to focus on total shareholder returns vs peers and a cash adjustment leverage factor with the goal of reducing leverage. These incentives should continue to drive management toward a goal of 3x leverage.

Valuation:

Earth Intelligence:

The Earth intelligence segment is expected to produce over $500 million of EBITDA this year. Reiterating my comments above, I expect legion to add an incremental $100 million of EBITDA through 2023 ($50 million/year in depreciation + $50 million in EBIT). This doesn’t include additional US government service add-on contracts or incremental business from foreign governments which could add another $50 million in EBITDA. I expect 2023 EBITDA in this segment to approximate $650 million, growing at a 9% CAGR from 2020.

Space Infrastructure:

The Space infrastructure segment is expected to post negative $35 million of EBITDA this year due to an $18 million hit from COVID-19 operating inefficiencies and a $14 million impact due to a manufacturing error which was identified during the final stage testing process.  Both these costs are one time and excluding them would get you to breakeven EBITDA this year. FY21 should have easier relative compares as these one times go away and we should start to see the benefit from backlog growth. Projecting out to 2023, I expect Revenue of $750 million at an EBITDA margin of 10%. I assume very little revenue growth (3% CAGR) and mostly operational improvements.

Corporate, Other & Eliminations:

I would expect management to control costs at the corporate level as they have demonstrated an ability to reduce overhead fairly dramatically (expected to decline 25% vs FY19). I have $65 million of corporate & $25 million of EBITDA eliminations related to the Legion build in FY20 for a total of negative $90 million of EBITDA. As Legion delivers, I would expect eliminations to decline dramatically while corporate to remain relatively the same.

Based on my estimates shown in the chart below, I would expect an over 300% re-rating in the stock towards $70 over the next 3 years. As leverage drops from 6x to under 4x (this does not include FCF generated from FY20 through FY23), the multiple should re-rate to 10x EBITDA. The average defense peer trades at 13x FY2 EBITDA (KTOS 20x, NOC 12x, LMT 11x, AJRD 9.5x, LHX 11.5x, LDOS 12.5x), so 10x EBITDA isn’t a stretch. In addition, Maxar is one of the only pure-play publicly traded space plays – so there is some scarcity value here. Virgin Galactic (SPCE) another space play trades for 10x FY23 revenue multiple.

In terms of FCF, I estimate that Maxar will produce over $600 million from FY21-FY23 which is 60% of the current market cap today. That alone is probably worth another $10/share not accounted in the above analysis and would see leverage come down under 3x by end of FY23.

Risks:

1)      Worldview Legion launch failure

2)      Failure of an existing observation satellite in orbit

3)      Space Infrastructure head-fake, recent orders are not enough to bridge the gap to a GEO market refresh

4)      Inability to refinance large debt maturities in 2023/2024

5)      SpaceX Starlink makes it more difficult for GEO customers to compete

Catalyst

1) Successful Legion launch

2) Space Infrastructure turning EBITDA positive

3) Strong FCF growth as capex declines

4) Renewal of Enhanced View contract at a higher $ value to Maxar