The Most Expensive Part of Hydrogen May Be Everything After Production

There is a familiar way people talk about hydrogen.

They talk about the breakthrough that is coming.  The electrolyzer costs need to fall. The tax credits that may unlock projects.  The future demand from trucking, steel, ammonia, shipping, data centers, and power.  The color debates. Green, blue, and grey. They talk about hydrogen as if the central challenge is still the same one it has always been:

How cheaply can we make the molecule?

That question matters, of course.  The cost to produce hydrogen matters a great deal. The U.S. Department of Energy has made cost reduction the centerpiece of its Hydrogen Shot initiative, and its latest technical planning still targets hydrogen production at $2 per kilogram by 2026 and $1 per kilogram by 2031.  But in the same planning framework, DOE still targets dispensed hydrogen for heavy-duty vehicles at $7 per kilogram by 2028, a reminder that making hydrogen and delivering hydrogen are not the same economic problem.

And that may be where much of the industry conversation still falls short.

Because in the real world, customers do not buy hydrogen at the production source. They buy it after compression or liquefaction.  After storage.  After transport.  After transfer.  After scheduling.  After inventory management.  After all the friction and physics and logistics that stand between a promising molecule and delivering usable product.

For many applications, the most expensive part of hydrogen may not be making it.  Using the DOE targets above proves it “$2 to produce plus $5+ more dollars to use it!”

It may be everything that happens after production.

That is not a small distinction.  It is the difference between a technology that looks compelling in a model and one that works consistently in a market.

The hydrogen economy is often described in sweeping terms, a “clean” energy transition, an industrial decarbonization tool, a fuel of the future.  However, hydrogen is also something more ordinary and more unforgiving.  It is a supply chain business.  It is a reliability and a purity business.  It is a delivery business.

And delivery businesses are won or lost in the last mile.

A market that already exists, and already reveals the problem

One of the strange things about hydrogen is that it is often discussed as if it barely exists. In reality, the United States already produces and uses hydrogen at scale.  DOE says the country currently produces about 10 million metric tons of hydrogen per year, primarily for petroleum refining and ammonia production.  Likewise estimates total U.S. hydrogen production at 10 million metric tons, and about 25% of that volume is merchant hydrogen sold by industrial gas companies rather than produced and consumed entirely on-site.

That existing market matters because it shows us something important: hydrogen is not merely a future technology waiting for its first real customer.  It already has customers.  It already has industrial logic.  It already moves through commercial channels.  And yet, despite all that, the system remains highly concentrated and infrastructure-light.

DOE notes that roughly 95% of the hydrogen produced in the United States is made by natural gas reforming in large central plants. At the same time, the country has only about 1,600 miles of dedicated hydrogen pipeline; tiny compared with the vast natural gas network and largely confined to regions where major hydrogen users are clustered, such as the Gulf Coast.

That creates a revealing picture.  Hydrogen is already a meaningful industrial commodity, but much of the country still lacks the connective tissue that would make it easy to move in large volumes at low cost.  In practice, this means hydrogen often depends on a patchwork of centralized production, localized pipeline systems, truck delivery, storage assets, and customer-specific logistics.

In other words, the challenge is not only production.  The challenge is architecture.

And architecture determines economics.

The molecule is light. The burden is not.

Hydrogen is appealing because of what it can do.  It can decarbonize hard-to-electrify processes.  It can serve as a feedstock, a critical part of a chemical process to produce food, pharmaceuticals, computer chips, plastics, fertilizer and hundreds of products as well.  It’s a fuel, an energy carrier, a storage medium, and a balancing tool.  Hydrogen touches industries that boggle your mind.  A gas turbine generating electricity requires hydrogen for clean and efficient production.  Without it, the generator fails. It is versatile, abundant in potential production pathways, and increasingly central to long-range decarbonization plans.  DOE’s national strategy envisions clean hydrogen production scaling to 10 million metric tons annually by 2030, 20 million by 2040, and 50 million by 2050.

But hydrogen’s virtues at the point of use are often counterbalanced by its inconveniences in motion.

It is a tiny molecule that is hard to produce, handle, store and transport.  It has low volumetric energy density unless compressed or liquefied.  It demands specialized handling.  It does not simply ride along existing systems as effortlessly as the casual conversation sometimes implies.  To move hydrogen economically, you generally need one of three things: dedicated pipelines, high-pressure transport, or liquefaction and cryogenic handling.  None of those are trivial.  All of them add cost, and each adds complexity in a different way. DOE’s hydrogen infrastructure planning explicitly treats storage and delivery as their own major challenge area, emphasizing that different transport and storage options are needed for different scales and end uses, and that cost reduction is required across the system.

This is where the romance of hydrogen often collides with the paperwork of hydrogen.

A molecule that appears attractive at the plant can look very different after the industry has done everything necessary to make it movable, storable, and dependable.

Compressed gas has its limits.  Trucking works, but it is not magic.  A Department of Energy delivery fact sheet notes that transporting gaseous hydrogen by truck tends to become expensive at distances greater than about 200 miles from the point of production.

That single fact should get more attention than it does.

Because it cuts through the abstraction.  It tells us that distance is not a side issue in hydrogen.  Distance is an economic variable.  When hydrogen has to travel farther, the supply model changes.  Cost structures change. Reliability risks change. The entire proposition changes.

And when trucking compressed gas becomes unattractive, the industry often turns to liquid hydrogen for longer-haul movement.  But that creates its own penalties.  DOE states plainly that with today’s technology, liquefaction consumes more than 30% of the energy content of the hydrogen and is expensive.  It also warns about boil-off losses during storage, especially in tanks at small hydrogen users.

That is a remarkable burden.  Before a customer has done anything productive with hydrogen, before a truck has driven, before a fuel cell has generated power, a meaningful share of the energy value may already have been spent just making the molecule transportable.

That is not a fatal flaw. But it is a serious commercial fact.

The last mile is where confidence gets tested

Industries do not scale around molecules.  They scale around confidence.  Confidence of  supply, purity, cost and capacity to support growth in their business.

A refinery manager, specialty gas distributor, semiconductor facility, research lab, mobility operator, or industrial buyer does not wake up and ask, “What is the levelized production cost of hydrogen in a theoretical best-case system?”  They ask a more grounded set of questions.

Will it be there when I need it?

Will it arrive on schedule?

Will the purity be right?

Will my delivered price swing unexpectedly?

How much inventory do I need to hold?

What happens if the trailer is late?

What happens if the upstream source goes down?

What happens if demand spikes?

These are not philosophical questions.  They are practical questions that determine whether adoption accelerates or stalls.

It is one thing for a new energy system to be exciting.  It is another thing for it to be dependable.  Markets reward the latter.

That is why the phrase “hydrogen ecosystem” can sometimes be misleading.  An ecosystem sounds organic, diversified, and self-reinforcing.  In many places, hydrogen still behaves more like a chain, and chains are only as strong as their weakest link.

A centralized plant can operate efficiently and still leave downstream users exposed.  A truck delivery network can function well in ordinary conditions and still prove brittle when routes are disrupted, demand clusters shift, or supply tightens.  A customer can be enthusiastic about hydrogen and still lose faith after enough late deliveries, inventory headaches, or unexplained cost escalation.

This is especially true in applications that do not consume hydrogen in huge, perfectly predictable volumes.  The more distributed the customer base, the tighter the purity requirements, the smaller the volumes, or the more time-sensitive the consumption profile, the more the supply chain itself becomes the product.

The customer is no longer just buying hydrogen.

The customer is buying the burden of getting hydrogen there.

A global industry that is learning this lesson the hard way

The modern hydrogen sector is making real progress.  The Hydrogen Council and McKinsey reported in Hydrogen Insights 2024 that clean hydrogen projects reaching final investment decision globally increased from 102 committed projects in 2020, representing about $10 billion in committed investment, to 434 projects in 2024, representing about $75 billion.  The same report says companies have announced 48 million tons per year of clean hydrogen production capacity through 2030, though it estimates only 12 to 18 million tons per year of that announced supply may actually be deployed by 2030 after delays and expected project attrition.

That gap between announcement and delivery says a lot.

Hydrogen is not failing for lack of enthusiasm.  It is failing, where it struggles, in the space between ambition and execution.  The Hydrogen Council points to regulatory uncertainty, inflation, interest rates, supply chain constraints, higher-than-anticipated renewable electricity prices, and the need for enabling midstream infrastructure.

Midstream infrastructure is the quiet phrase there, but it may be one of the most important.

Because once hydrogen leaves the realm of concept slides and enters the realm of commercial deployment, “midstream” becomes another name for reality.  It means compression, storage, terminals, pipelines, trailers, handling systems, transfer losses, and all the industrial plumbing needed to turn production into dependable supply.

The U.S. Department of Energy’s own national hydrogen strategy acknowledges this directly.  It notes that regional networks prioritizing shared, open-access infrastructure can reduce the delivered cost of hydrogen by lowering transport and storage costs.  It also says hydrogen hubs are designed as networks of producers, consumers, and connective infrastructure in “close proximity,” and that co-location of supply and demand can reduce the need for new long-distance infrastructure and lower the cost of early market growth.

That is a subtle but important admission from the center of the policy conversation: proximity matters.

Not someday. Now.

Why the industry keeps underestimating this

Part of the reason production gets the most attention is that production is easier to model and easier to market.

Production has a clean story.  Lower electrolyzer cost.  Better efficiency.  Cheap renewable power.  Better utilization.  Tax credits.  More capacity.  Bigger plants.  Larger hubs.  Lower cost per kilogram at the source.

That is a compelling narrative because it points toward scale, and scale is one of the great seductions of industrial strategy.  Bigger plants promise better economics.  Larger networks promise eventual lower unit costs.  Centralization feels efficient.

And in some cases, it absolutely is.

But centralization also assumes that the cost and difficulty of moving hydrogen can be overcome cheaply enough, reliably enough, and widely enough to preserve the production advantage all the way to the customer.  That assumption deserves more skepticism than it usually receives.

The history of industrial gases offers a useful lesson here.  Economies of scale matter, but so do logistics.  There is a reason industrial gas businesses are obsessed with proximity, route density, utilization, storage, and customer clustering. In gases, distribution is not a footnote to the business.  Distribution is the business.

Hydrogen only amplifies that truth.

And that is why the delivered cost matters more than the headline production cost.  A cheap molecule at the source can still become an expensive product at the point of use. A technically successful plant can still feed a commercially weak network.  A celebrated production breakthrough can still leave customers wrestling with the same old questions about delivery windows, trailer availability, storage losses, and margin stacking.

None of this means centralized hydrogen cannot work.

It means centralized hydrogen is not the whole answer.

The question the industry is slowly being forced to ask

For years, hydrogen conversations have tended to revolve around supply.  How do we produce more?  How do we produce it more cheaply?  How do we clean it up?  How do we finance it?

Those are worthy questions.

But the next phase of the industry may revolve around a different one:

Where should hydrogen actually be made?

That question changes everything because it shifts attention from the molecule to the map.

If the economics degrade as hydrogen travels, then distance is not incidental, it is structural.  If transport and storage costs meaningfully influence delivered price, then location becomes part of the production strategy.  If supply resilience improves when production and use are closer together, then infrastructure design becomes as important as generation cost.

And once you see the problem that way, a new possibility starts to emerge.

Maybe the winning hydrogen systems will not always be the ones with the cheapest molecule at the biggest plant.

Maybe, in many cases, they will be the ones that eliminate the most friction between production and use.

That does not mean every customer should make its own hydrogen.  It does not mean pipelines are irrelevant.  It does not mean centralized production disappears.  It means the industry may need a more distributed architecture than many people expected, one shaped not just by theoretical scale economies, but by the stubborn realities of compression, trucking, storage, purity, uptime, and distance.

In other words, the future of hydrogen may depend less on producing more molecules somewhere, and more on producing the right molecules closer to where they are needed.

Hydrogen’s next breakthrough may be logistical, not chemical

The hydrogen industry understandably loves technology stories.  New electrolyzers. Better catalysts.  Novel storage materials.  Advanced liquefaction.  Lower-carbon pathways. These are important and necessary.

But sometimes industries do not unlock their next stage of growth through a scientific miracle.  Sometimes they unlock it through a simpler, harder insight:

The system works better when you stop making the product travel so far.

That is not as glamorous as a new process breakthrough.  It does not fit as neatly into headlines about gigawatts and tax credits.  But it may be the kind of insight that turns hydrogen from an exciting sector into a durable market.

Hydrogen does have a production challenge.  It has a policy challenge.  It has a capital challenge.

But it also has a geography challenge.

The farther the molecule must go, the more things have to go right.

And in business, systems that require too many things to go right usually end up costing more than expected.

That may be the quiet truth emerging beneath the noise: the most expensive part of hydrogen may not be making it.  It may be everything that happens after production, every mile, every handoff, every compression step, every storage decision, every scheduling constraint, every layer of uncertainty between supply and use.

If that is true, then the industry’s biggest opportunity may not be to think only bigger.

It may be to think closer.

And once that idea takes hold, a more important question naturally follows:

If the burden of hydrogen grows with distance, why are we still so comfortable assuming it should be made far away from the customer at all?