|Shares Out. (in M):||223||P/E||0||0|
|Market Cap (in $M):||70,200||P/FCF||0||0|
|Net Debt (in $M):||4,300||EBIT||0||0|
|TEV (in $M):||74,500||TEV/EBIT||0||0|
Sign up for free guest access to view investment idea with a 45 days delay.
The high-level thesis for Air Products (APD) is relatively straightforward:
APD operates in an industrial gas oligopoly.
The base business provides stable cash flows, a function of contractual revenue from on-customer-site infrastructure coupled with reliable and localized supply in the merchant side of the business.
They have a proven ability to pass through commodity cost increases.
Bright futures for hydrogen as a clean fuel, gasification and carbon capture provide meaningful positive skewness to the APD investment return profile.
Management have shown strong execution and are thoughtful stewards of capital. The CEO has significant stock ownership.
The balance sheet is secure.
APD retains about 40-45% of earnings (and more like 50% of cash flow) to drive growth and pays the rest as dividends.
We pencil a high teens IRR over five years (over $600 per share plus dividends), with higher probabilities above that range than below it. If hydrogen plays out as hoped, elevated returns could keep coming for decades.
There are only a handful of players with the technical chops, existing infrastructure, and global scale to compete in industrial gases and go after hydrogen, gasification, and related carbon capture in a meaningful way. These include APD, L’Air Liquide, Linde, and Messer. This rational oligopoly has historically provided for stability and pricing power that remains evident today.
Note the company has a September fiscal year. All references to years are fiscal (meaning, 2022 refers to the fiscal year ended September 2022) unless otherwise noted.
APD produces atmospheric gases (oxygen, nitrogen, and argon), process gases (hydrogen, helium, carbon dioxide, carbon monoxide and syngas, which is a mixture of carbon monoxide and hydrogen) and specialty gases. Atmospheric gases (47% of sales) are produced through air separation processes, with cryogenic being the most common. Process gases (22% of sales) are produced through other methods, like steam methane reforming or gasification.
About 52% of APD’s revenue is supplied “on-site,” meaning APD will build or acquire an on-site plant adjacent to a customer’s facilities or will connect to the customer facility through pipelines. These are typically delivered under 15- to 20-year contracts with limited volume risk and energy and raw material pass-throughs. Products include hydrogen, nitrogen, oxygen, carbon monoxide and syngas.
Around 40% of APD’s revenue is supplied through “merchant” channels. Most of this is liquid bulk that is sold under 3- to 5-year contracts using local supply chains. These gases are often drawn from excess capacity from on-site customers and are distributed using tanker trucks or tube trailers. The rest is packaged gas that is sold under short-term contracts. These are typically sold in cylinders and dewars, primarily in Europe, Asia, and Latin America.
Equipment and services make up the remaining 8% of the revenue base. This includes cryogenic and gas processing equipment for air separation, as well as an LNG equipment business.
In addition, APD has invested in many JVs, the most significant of which are in Saudi Arabia, Mexico, Italy, and India (which is the #1 player in industrial gases in India). APD is building and will operate some of the new multi-billion-dollar hydrogen and gasification projects through JVs. For instance, the new Jazan JV in Saudi Arabia is the largest from an invested capital perspective. Jazan will require a $12B total investment once completed (with fully committed financing), including 60% non-recourse project financing. APD will contribute $2.4B in equity for a 51% stake. Jazan supplies Saudi Aramco’s large refinery and power grid with power, steam, utilities, and hydrogen in exchange for a monthly fee under a 25-year contract.
End market applications for industrial gases are broad. In 2021, 23% of sales went to energy (79% of which was for refineries), 23% to chemicals, 17% to electronics, 14% to metals, 11% to manufacturing and the rest into medical, food and other. To name a few of the bigger ones:
Refiners use hydrogen to convert heavy crude feedstock and lower sulfur content of gasoline and diesel.
Chemical manufacturers use many different gases as feedstocks for many basic chemicals.
Energy production uses nitrogen to recover oil and gas and oxygen for gasification.
Oxygen is used for combustion and heating in producing steel, other metals, glass, and cement.
Major operating costs include electricity (for atmospheric gases), natural gas (for steam methane reformers) and liquid and solid hydrocarbons (for gasification and truck fuel). APD mitigates these exposures through contractual pricing formulas, surcharges and cost pass-through and tolling arrangements.
APD claims to have a roughly 15% global market share in industrial gases, with strength in hydrogen (where they claim to be the world’s biggest), oxygen, nitrogen, and helium. Most gases are produced at or near the point of use given the difficulties with storing gases at low temperatures. As such, APD has a large footprint of manufacturing and distribution locations (450 in Americas, 250 in Asia, 200 in Europe and 15 in Middle East and India).
As mentioned, L’Air Liquide, Linde and Messer are the major competitors (customers produce a lot of industrial gases in-house as well).
Management is solid. Here is a slide from their investor deck on shareholder value that captures their philosophy well:
Management incentives are reasonable. Annual cash bonuses are driven by EPS, while long-term comp is 60% performance shares based on 3-year total shareholder return, and 40% restricted stock. Per the proxy, the CEO owns 533k shares of stock and 290k options/RSUs that vest in the near-term. The common stock alone is worth nearly $170M at current prices.
The Chairman and CEO, Seifi Ghasemi, is a focused operator and thoughtful capital allocator. He leads a team that is people-focused with an attention to culture. For instance, Ghasemi said he met with 2,000 employees during June and July to share the company’s strategy, answer questions and gather input. He planned to talk to all 20,000 employees in small groups over the course of the following year. He has shown his management principles in every quarterly slide deck since becoming CEO in 2014. His principles have guided him well, as he has overseen a period of dramatic operational improvement. He oversaw significant cost cuts and divestment of lower margin non-core businesses (electronic materials Versum in 2016 and specialty additives Evonik in 2017). This is consistent with his prior position leading Rockwell, which he focused on lithium through noncore divestitures and then sold to Albemarle for a nice premium and a 30% ongoing stake in the combined business. He also aggressively positioned APD for growth in emerging markets like China, India, and Saudi Arabia. In 2018 he announced a plan to invest over $30B over a 10-year period ending in 2027, which has since increased to $36.5B. Ghasemi targeted 10% annual EPS growth when he took over as CEO. Over the past eight years he has delivered 11% EPS and 10% dividend growth (now 40 consecutive years of dividend growth), with excellent stability:
Along the way, Ghasemi significantly improved APD’s balance sheet, which now sits at 1x net debt to EBITDA (this includes related party debt from JVs where APD is on the hook). They target staying within an A/A2 debt rating (which would allow for up to 3x leverage).
It’s worth noting that Ghasemi does not like to buy back stock. He views buybacks as a way to juice EPS in the short-term, without creating true long-term value. He believes the shares would need to be significantly cheaper for a buyback to result in better economics for shareholders vs. deploying it into organic growth like building new facilities. As he said on the Q4 call, “Now Air Products has entered a new phase of our company's evolution, in which we expect a steady stream of meaningful contribution from these new projects going forward and for years to come. By choosing capital deployment over share buyback, we believe that we have traded quick gains in the near-term for greater reward in the future.”
While industrial gases are a commodity, customers are often willing to lock in long-term contracts at premium pricing to make sure they have continued access. This is because industrial gases are critical to their operations but are a small fraction of overall costs. Reliability of supply is therefore at least as important as cost.
As discussed, about half of APD’s revenues come from long-term take-or-pay contracts based on equipment that is located on-site or piped into customer facilities. The customer typically provides the inputs for gas production from their facility and APD uses its integrated equipment to produce the desired gas output for a fixed fee. These revenues are sticky due to contractual terms and the high cost for customers to switch providers.
While more vulnerable than the on-site business, the merchant business also generates solid returns as customers typically enter multi-year contracts and rely on APD for on-site storage and vaporization. The strength of the merchant business is evident in the pricing power that APD has displayed this past year by passing through increased input costs without a hitch (though it can operate on a bit of a lag and requires some forecasting to capture well).
APD’s deep expertise in industrial gases and its global scale of operations (including significant pipeline infrastructure) make for a high barrier. For instance, APD can leverage its existing investment in hydrogen pipelines to build new environmentally friendly hydrogen facilities that would be uneconomic without the pre-existing pipes.
The quality of APD’s business is most apparent in capital deployed and return measures. Sales growth and margins are a red herring due to the large amount of pass-through revenue relating to commodity cost changes as well as price changes in the merchant business made in reaction to commodity cost changes. The significance of JVs to APD’s results further complicates the accounting. We therefore focus on growth in capital deployed to gauge business growth, and ROIC to gauge profitability.
Management’s stated hurdle is to generate at least $0.10 of after-tax unlevered profit for every dollar of investment, or a 10% after-tax unlevered ROI. Some of these projects are riskier than others, as many are based on long-term fixed fee contracts while others include end market demand risk. Management does not give specifics around how much of a new project’s capacity is locked up in long-term contracts prior to investment, but management seems thoughtful around this and does not risk investor capital lightly.
The proof is in the pudding. APD has produced an attractive and stable ROIC (pretax) of around 18-20% since 2017, with 12% annual growth in capital deployed. They have accomplished this result despite holding significant amounts of cash needed to fund committed projects. This has resulted in a stable 15-17% ROE (after-tax). These figures likely understate the magnitude of the true returns as maintenance capex tends to run at less than half of depreciation expense. This dynamic results in part from GAAP depreciation guidelines, under which APD depreciates production assets associated with long-term supply contracts in line with the underlying contract term. In addition, there are projects in development that have driven up invested capital without contributing any profit (yet).
Management’s growth strategy centers on providing solutions for the world’s energy and environmental challenges through gasification, carbon capture and clean hydrogen. APD has developed core expertise in these areas. In 2018, management laid out a plan to invest over $30B (including project-related and other debt at 3x EBITDA) into these growth areas over ten years and have since increased the bogey to $36.5B. They have spent $11B so far and have close to $16B remaining in backlog. This leaves approximately $10B yet to be committed between now and 2027. Here are the major committed projects in the pipeline:
Gasification is a way to make syngas from coal, high sulfur liquids or natural gas. Customers use syngas to make chemicals, diesel fuel, high-end olefins, or power. This technology has been used since the 1800s. It’s adaptable to various hydrocarbon feedstocks, including high-sulfur coal. There are no smog-causing particulates and the carbon dioxide it produces is concentrated in a capture-ready stream. It has a low operating cost and becomes more economical on a relative basis as oil prices increase relative to coal or gas.
Gasification is appealing to countries with large coal resources (like China, Indonesia, South Africa, and India) or natural gas reserves (like the US, Algeria and Uzbekistan) that would like to reduce dependence on imported oil to produce liquid fuel and high-end chemicals. In addition, it provides refineries around the world with a use for high sulfur bottom-of-the-barrel liquids that can no longer be used as fuel for ships post the IMO 2020 maritime law change.
Coal is far and away the feedstock leader for gasification processes in use today, followed by petroleum, gas, and others. This is not surprising given the abundance and cheapness of coal in many countries that lack access to other domestic hydrocarbon resources (like China and India). Chemicals are the largest end-use application, followed by liquid fuels, power and gaseous fuels (with the latter set to grow considerably through planned projects).
APD became the global leader in gasification in 2018, when it acquired Shell’s and GE’s technologies. Shell’s dry feed IP has been used to build over 200 gasifiers since the 1950s and has 120 currently in operation, with a focus on bottom-of-the-barrel refinery applications. The coal-water slurry feed IP from GE (formerly Texaco) is in even more widespread use, with nearly 300 gasifiers operating or under construction. After these two technologies, there are a handful of smaller players like Lurgi, ECUST, SEDIN and Siemens.
Carbon capture opportunities exist primarily in gasifiers and hydrogen plants, whereby carbon dioxide is removed and stored, typically belowground. Furthermore, the 45Q tax credit (which was recently increased through the Inflation Reduction Act) should provide a meaningful windfall to APD given how much carbon it already sequesters and plans to sequester as part of previously agreed projects (like the Louisiana blue hydrogen facility – see below). Under the new law, the US offers a tax credit of $85 per ton for carbon dioxide sequestration. Other countries may follow suit.
As Thijs van de Graaf of the IMF recently said, “If the 1990s were the decade of wind, the 2000s the decade of solar, and the 2010s the decade of batteries, the 2020s could launch us toward a next frontier of the energy transition: hydrogen” (https://www.imf.org/en/Publications/fandd/issues/2022/12/hydrogen-decade-van-de-graaf). Hydrogen opportunities for APD are primarily in production and distribution as well as developing global infrastructure to support new applications (like trucking and bus transit).
Unlike oil and gas, hydrogen is manufactured and can be produced anywhere there is electricity and water. Hydrogen has several important characteristics that give it great potential as an environmentally friendly fuel during and after the energy transition away from dirty fuels. For one, it is abundant (though isolating it is tricky) and can be stored indefinitely in tanks and salt caverns. For another, it’s combustible. That means it can power engines and replace fossil fuel gas used for heat in manufacturing. Finally, it is transportable via pipeline and across oceans as it can be liquified or shipped as ammonia. Ammonia, a common ingredient in fertilizer, is produced by combining hydrogen with nitrogen (both of which APD produces). It is easily produced, shipped, and then cracked back into hydrogen at the point of use. This allows for global transportation of hydrogen.
Hydrogen has several applications that look most promising. These include production of steel, power generation and heavy transportation (large trucks, buses, airplanes, ships, and trains). For instance, steel production is a heavily polluting process where iron ore is melted in a blast furnace with coke (a coal derivative) or charcoal to extract oxygen and leave pure metal. In “direct reduction of iron” (DRI) process, hydrogen can remove oxygen from solid iron ore with much less heat intensity. In transportation, hydrogen has appeal over battery electric for heavy transport mainly due to the prohibitive weight of batteries. Another application is to use hydrogen to convert triglycerides into diesel (renewable diesel refineries require 4-5x as much hydrogen as a conventional refinery per barrel of production).
As per the International Energy Agency (IEA), industry represents the biggest growth opportunity, followed by grid injection and power:
In essence, anything that can be well-serviced by electricity (passenger cars, HVAC, cooking, lighting, etc.) will likely not be a good candidate for hydrogen. But all the heavier stuff (heavy transportation, industry, etc.) is.
There are currently three main ways to produce hydrogen.
The traditional method is to mix natural gas (mainly methane) with steam to cause the carbon to separate from the hydrogen. Because this method produces significant carbon dioxide, it is not considered especially useful for the clean energy transition. This is called “gray hydrogen,” and it’s traditionally been used primarily for refining fossil fuels and manufacturing metals.
The second method is to produce gray hydrogen but capture and store the carbon. This is called “blue hydrogen.” Blue hydrogen can be done with only 5% of the carbon dioxide emissions of gray hydrogen. This is accomplished by using autothermal reforming (ATR), instead of the traditional steam methane reforming (SMR). ATRs enable carbon dioxide produced in the processing of natural gas to be captured as steam, which allows for easy carbon dioxide capture. In an SMR, by contrast, half of the carbon dioxide is released into the atmosphere as flue gas and is not easily captured.
The last method is to use an electrolyzer, whereby hydrogen is separated from oxygen in water by running an electric current between catalyst-played electrodes. This method produces no emissions directly, and if the fuel source used to power the process comes from renewable sources (like wind, solar, hydro, or geothermal), it is especially clean. This is called “green hydrogen.”
All three types of hydrogen are indistinguishable from each other once produced (they are all just fungible hydrogen molecules). The differences are only in the fuel and process required to produce them, and therefore their carbon intensity (around 12-13 kg of carbon per kg of gray hydrogen, 2.5-3 kg for blue hydrogen and 0.4 kg for green hydrogen, per management) and cost. These carbon differentials will translate to value differences in the marketplace as users will pay more for the lower carbon products (for instance, in the US customers receive carbon credits for purchasing green and blue hydrogen). As scale economies grow and technology improves, analysts expect the cost of green hydrogen to drop from $5/kg towards $1/kg, even without subsidies. This would make it competitive with gray hydrogen. As per Nature: