Why Giant Airships Could Be a Trillion Dollar Industry

WELCOME MARGINAL REVOLUTION READERS! And thank you for your thoughtful comments. I've written responses to them at the bottom of this post. Scroll down.

One thing I plan to do a lot of on this blog is speculative quantification. I want to post a lot of best guess numbers that are subject to so much uncertainty that it verges on irresponsible to post them at all. Not biased, I hope, except inasmuch as the evidence itself is biased, as often happens; but very uncertain. Experts and aspiring experts tends to avoid saying numbers that are wildly uncertain, for good reasons. While one can surround a number with a bodyguard of disclaimers and caveats, if the number gets noticed at all, such evidential throat clearing is the first thing to vanish in references and quotations. Methodology is boring. And so, as your number circulates, it will sound more and more like a factual claim rather than a guess, and there's not much you can do to prevent it, other than to refuse to publish speculative quantification in the first place.

Yet I'm willing to pay the price of misleading people through the erosion of evidential caveats, because numbers, even highly uncertain numbers, are a crucial input to decisions, and highly uncertain guesses are better than nothing.

For example, suppose we want to know how large the giant airship industry will be, once the initial safety and technical hurdles are surmounted and giant airships are known to be safe and efficient. The state of my knowledge might, in principle, be such that if I make a few assumptions, do a little modeling, and predict that airships will be a $227 billion revenue industry, the spurious precision would make the claim a kind of pseudoscience verging on fraud, yet if I said vaguely that I thought airships could be a $200 billion industry, give or take an order of magnitude, the claim might both accurately reflect the state of my knowledge and be useful to someone's decision making, e.g., if they had vaguely thought airship industry potential revenue was a few hundred million. If so, it would be unfortunate if my scientific scruples prevented me from uttering my prediction. Yet it's quite likely and foreseeable that as the prediction circulates, the “give or take an order of magnitude” will be omitted by authors struggling to squeeze a story into a word limit, giving my prediction a spuriously precise and overconfident character, to the detriment, perhaps, of my scientific reputation, when the true number turns out to be $30 billion, or $1.7 trillion.

Compare the following three statements:

1. “Giant airships could be a multi-million dollar industry!”

2. “Giant airships could be a multi-billion dollar industry!”

3. “Giant airships could be a multi-trillion dollar industry!”

All three statements have an impressive ring about them. The amounts of money suggested are large, more than will pass through a typical person's hands in their entire lifetime. But (1) would actually imply that airships would be negligible and unworthy of investors’ attention; (2), that airships could have an impact on small niches of the economy and might yield good profit for a few small companies; and (3), that airships will transform the world economy, alter patterns of world trade and human settlement, noticeably impact most other industries, and maybe even significantly reduce global carbon emissions. So it's worth doing enough educated guesswork to mitigate the uncertainty about whether the truth is closer to (1), (2) or (3), even if we can't really know.

Let me start, then, with a table of cost and speed for existing transport modes and hypothetical giant cargo airships.

Table 1. Comparing Transport Modes
Cost per ton-mile
Typical speed (mph)
Giant Cargo Airship (projected)
60 (but with transshipment delays)

Now for the disclaimers. “Per ton-mile” estimates of shipping costs always misleadingly oversimplify. Costs tend to be concentrated at transshipment points. It matters a lot whether there’s a payload for the return trip. Modes differ in reliability, and if occasional transport delays make expensive bottlenecks, it might not matter if a mode can offer cheap service and adequate speed most of the time. Handling is crucial for some cargoes, e.g., fragile or perishable ones. Sometimes space is a more binding constraint than weight, and lightweight, bulky cargoes often pay “chargeable weight” proportional to the space they occupy rather than their actual weight. Above all, not all modes go to all destinations: e.g., trucks can go wherever there’s a road, but not over bodies of water, while airplanes can cross water but can only land where there’s a port. And so forth. Logistics is complicated!

As for the estimates of cost and speed for a giant cargo airship, suffice it to say, for the time being, that I do have some basis for suggesting that 100 miles per hour and 5-10¢ per ton-mile are attainable performance goals for a giant cargo airship, though it will take a lot of investment to get there. I’m not asking readers to trust me on that, except for the sake of argument. I plan to come back to that in future posts.

Now let me bring in a couple more numbers:

World GDP: $80 trillion

Transportation’s share of US GDP: 8.9%

Transportation fixed assets in the US: $4.9 trillion (derived from Table 5-1 here)

Now, suppose that in the next couple of decades-- a rough minimum for how long it might take a giant airship industry to scale up to something like maturity-- world GDP rises to $100 trillion. And transportation comprises the same share of GDP in this future world economy as in the present US economy, 9%. And-- here’s the big if-- giant airships capture a lot of market share in freight, comparable to trucks for trans-oceanic trade routes, and some market share in passenger service, and end up accounting for 15% of world transportation GDP.

Why 15%? I didn’t exactly pull it out of a hat. I’ve reflected and studied it quite a bit from a lot of angles. But it would take a lot of sophisticated modeling to make it more like a formal estimate than a guess. Still, hopefully Table 1 is enough to prime the reader’s intuition to accept the number tentatively. After all, giant airships look competitive in cost and speed terms, and unlike trucks and rail, they are airborne and should be able to cross oceans. Their huge cargo bay is an advantage. So I think 15% could be plausible. Let’s proceed.

15% of world transportation GDP to airships × 9% of world GDP is transportation × $100 trillion in world GDP = $1.35 trillion in revenue for giant airships in this hypothetical future

So maybe (3) is closest to the truth. Airships could be a trillion dollar industry.

“Trillion dollar industry” is a great phrase that I'll probably repeat often, and hopefully get others to echo, so I'll stop for a moment of disambiguation to clarify what it might mean. It could mean-- to focus on the most salient potential meanings-- a trillion or more dollars of revenue, of market capitalization, or of profit. Never mind the last one, though. Profit is a small fraction of revenue in most cases. A trillion dollar profit industry would have to be much bigger than what I’ve described. That’s not a claim I want to put forward. The fast-and-loose calculations above embody a scenario in which giant airships generate a trillion dollars of revenue. Possibly, though, the assumptions are a bit aggressive. Much of the transportation industry is a local, last mile affair, people commuting to work for example, and airships couldn't really do that job in cities anything like those we know. And airships seem unlikely to dominate passenger travel over any distance because they are so much slower than airplanes. Suppose, then, we scale back our assumption to airships capturing 5% of world transportation GDP. Now airships, earning $450 billion a year, aren't a trillion dollar industry by revenue. But market capitalization is typically three times revenue or so, and airship transportation is likely to be a rather capital intensive transport mode. So the value of the airship industry’s capital stock, the airships themselves and ports and hangars and so forth, ought to be around $1.5 trillion. So it's still a trillion dollar industry, in a different sense.

Let's start over with a different question. How many giant airships will be built? This time I'll focus explicitly on freight. The great chart below (source here), from BTS, shows how freight in the US was distributed among transport modes by tons, miles, and ton-miles.

It doesn't give total ton-miles of US freight shipping, so in the deliberately slapdash fashion meant to keep the reader keenly aware of the extreme roughness of these calculations, I'll eyeball it and say US freight is around 5 trillion ton-miles per year. If future world GDP is five times current US GDP, and ton-miles of shipping per dollar of GDP is similar, there might be 25 trillion ton-miles of world freight shipping. If giant airships capture 20% of that, 5 trillion ton-miles of shipping will be performed annually by giant airships.

Now, let's assume -- I have my reasons, but trust me for now -- that the payload of giant airships will be 100 to 500 tons, and their usual speed, 50 to 100 miles per hour. Suppose they'll fly 200 to 300 days per year, 15 to 20 hours per day. Multiply points near the middles of these ranges and you'll get around 50 million ton-miles per year per airship.

5 trillion ton-miles ÷ 50 million ton-miles per airship = 10,000 giant airships to carry all this freight

To help you wrap your head around that, New York has about 0.3% of world population. If it were serviced by a share of the world’s giant airships proportional to its population, there might be about 30 giant airships docking in New York City or gracing its skies over the course of a day or week. Giant airships would be hundreds of feet long, and rather tall and wide for their length, unlike airplanes. They would be very conspicuous and alter New York’s skyscape.

Or suppose you're living in Alaska, which lies (counter-intuitively) on the direct route between the mainland US and the Pacific Rim of Asia. The US has about one-fourth of world GDP, the Pacific Rim, about one-third, and while I don't have figures handy, I'd guess that between one-twentieth and one-tenth of world trade by value flows between the US and the Pacific Rim, much of which would be captured by airships since it's a transoceanic route. So it might keep 10% of the world's airships busy. A thousand airships might crisscross Alaska once a week or so on the way from the US to Asia and back. For some Alaskans at least, airships flying overhead would become a familiar sight. And since they'd likely want to stop for fuel, or hydrogen, along the way, airship refueling and reflating could become a big Alaskan business.

How much would each giant airship cost? I think $30 million might be a reasonable first stab. That would put the value of giant freight airships at $300 billion, well short of our “trillion dollar industry” threshold. But we’re only talking about freight now. There are other applications, and it would take other kinds of tangible and intangible capital to comprise the whole capital stock of the industry. So I think the “trillion dollar industry” standard is still in reach.

Perhaps you find it as fun as I do to think about this stuff, but wonder whether it's worthwhile. Does it matter now if giant airships, with a large dollop of investment dollars, could grow into a trillion dollar industry?

Yes, because the trillion dollar industry of the future is the source of the massive potential financial returns that should motivate investors to fund the launch of the industry today. Not that it's a simple matter, even if you believe today in the trillion dollar industry of the future, and history proves you right, to invest today so as to reap those future rewards. You might have the right dream but invest in, or found, the wrong company. But a rising tide like that can lift most boats. Investors who believe in that airship future should be eager to buy a stake in it today, if only they can figure out how to do so. And that conclusion doesn't change much if my forecast of a trillion dollar industry is an order of magnitude too high.

Incredulous? You should be! Wisdom has a knee-jerk skepticism of all manner of large claims. That was my initial reaction too, when I first heard this pitch years ago. If this were such a great idea, someone would have done it already, is a lazy and complacent attitude to take, but it’s often right. But try to justify your skepticism. Obviously, I’ve left a couple of placeholders for future argument: I need to justify my 5-10¢ per ton-mile figure, for example. What else am I missing? Try to articulate the reasons for your instinctive doubts. You might find them harder to defend than you think.

If giant airships become a trillion dollar industry, that will affect practically everything. A lot of company founders and early stage investors will become very, very rich, the Jeff Bezoses of the second half of the 21st century maybe. Some industries, and some regions, will face devastating competition, while others will face brilliant opportunities. A lot of relative prices will shift, and more of them will shift favorably than unfavorably for most people, because technological progress is a positive-sum game. Jobs will appear and job descriptions will change. Industries upstream and downstream from airships will grow. I wouldn’t rule out the possibility that giant airships could be important enough to non-negligibly lift the global rate of return on capital.

So it’s important to know whether a trillion-dollar giant airship industry is really a possibility, and if so, to make it happen.

UPDATE: On reflection, I'm dissatisfied at how I flaunted the fast-and-loose character of the above calculations. Readers might reasonably wonder whether I'm wasting their time. It might help to mention that I've interacted with a number of airship companies over the years, and exchanged information that might arguably be confidential, and more recently, I've signed Non-Disclosure Agreements with a couple of airship companies. So I know more than I can say here, and the scattered estimates that seemed safe to cite won't seem like a complete picture to the uninitiated. My hope is that sometime soon, actual airship company representatives will agree to be interviewed by me for this blog, and some of these estimates can get some substantiation from their own words. And one motive for writing this post was to publish what I already know before airship companies share more with me, since the NDAs permit me to continue to use information I knew before it was divulged under the NDA. So I'm telling what I know now, so that I can prove what I didn't learn through confidential channels.

One airship company that has been somewhat more explicit than others with some of its quantitative estimates is Skylite Aeronautics. This page provides estimates of the energy efficiency of the GeoShip, the giant airship that Skylite hopes to build, compared to jets, ships, rail, and trucks. Mike Voorhees, the designer-- and a former colleague of mine at Loft Aeronautics-- here cites estimates of $200 million to build a GeoShip, declining to $100 million over time thanks to learning-by-doing and economies of scale in production. The GeoShip, as envisioned by Voorhees, is very big, 1,000 tons payload, and made from top-quality materials, with a lot of special functionality, e.g., for autonomous landing. My $30 million is a guess based on downsizing the GeoShip somewhat and economizing on its functionality to make an airship optimized for long haul cargo transportation.

UPDATE 2: It just occurred to me that published estimates for the GeoShip provide a numeric basis for the crucial 5-10¢ per ton-mile estimate in Table 1. At technological maturity, its estimated cost is $100 million, its payload, 1,000 tons. The top speed mentioned is 140 kilometers per hour, which I'll round off to 100 miles per hour for ease of calculation. (Other assumptions could be tweaked to get the same results.) If we multiply:

Payload: 1,000 tons
Speed: 100 mph
Days of service per year: 250
Hours of service per day: 20

We get 1,000 × 100 × 250 × 20 = 500 million ton-miles of production capacity per GeoShip per year

So the capital cost per ton mile produced per year, if the required rate of return is 10%, is

$100 million ÷ 500 million × 10% = 2¢ per ton-mile. Over and above that is fuel costs, but the energy efficiency of the GeoShip keeps those low, and other costs like crew salaries and rent on port facilities and marketing etc. If labor, fuel and miscellaneous triple the basic capital cost, we get 6¢ per ton-mile, which lands in my estimated range quite comfortably.

UPDATE 3: This post got linked from Marginal Revolution! Thanks Tyler! And MR commenters provided a good deal of mostly skeptical commentary. It's a useful inventory of common objections, and an opportunity to rebut a few of them.

One theme was WIND. Stated most aggressively...

mkt42: Bizarre. Over the past couple of decades we've had a couple of waves of people who claimed that zeppelins and other dirigibles were going to make a comeback, usually during times of rising fuel prices when airplanes became relatively expensive. I'm not sure what's inspired this latest article, given that fuel prices are low at the moment.

Do you want a transportation system based on ultra-cheap, totally renewable energy? I've got one: sailboats.

And there's a reason why we don't use sailboats to move cargo anymore: wind.

And as others have already commented, wind is also the reason why dirigibles are not a serious choice. 
... this objection is just ignorant. In the 1930s, the Graf Zeppelin flew a million miles and provided regular passenger and mail service across the Atlantic. Yes, the old giant airships failed, but not because of wind; rather, because of the Hindenburg disaster and then competition from jet aircraft. But put more moderately, as here...
rayward: A problem with airships is that they cannot "fly" (fly seems like a misnomer since they are slow - it's like a sailboat "race") in windy conditions.

RatInPutinsMaze: Airships suffer from intermittency. They can't be used or deployed in adverse weather. And they are more limited than even aircraft.

Routine cargo deliveries require a robust service that has a high availability. There's no indication that airships can meet that requirement. To be competitive they would have to be much cheaper than the competition. 
... this is an important point, though still overstated. "Slow" is relative. The Graf Zeppelin's top speed was 80 miles per hour, and modern designs anticipate higher top speeds. Airships will always be slower than airplanes, because the need to contain a huge volume of lifting gas means that they can't have streamlined shapes like planes, and they suffer from more drag. Jim Birch, who seems to get it, puts the wind issue in the right perspective.
Airships will absolutely require meteorologically optimised flight paths. Airlines do this routinely now but with airships the requirement is critical, for fuel usage, travel time and possibly even making some targets. This requires high quality, detailed forecasting but we more-or-less have that now. Landing in high/gusty winds could be a problem, as a balloon pilot will tell you. I would guess that airships would be unsuitable for some climatic regions, eg, high latitudes southern hemisphere, but this depends. Airship economics will be cheap delivery, not guaranteed timeliness.
Bravo! Yes, giant airships would be more weather dependent than modern airplanes, more comparable perhaps to the great sailing ships of old, but they have plenty of other strengths to compensate.

Another theme was lifting gas, and the problem that of the two possible lifting gases, helium is a scarce, expensive, non-renewable resource, whereas hydrogen has safety issues:
rayward: And then there's the image of the Hindenburg disaster. I know, the Hindenburg was filled with hydrogen. It was designed to be filled with helium, but because of trade restrictions, helium was in short supply, so hydrogen was substituted. Today's airships would never be filled with hydrogen.
Abelard Lindsey: The problem with airships is wind and Helium supply. Helium comes from Natural gas wells, but is more limited in supply. Because it is a small molecule, it easily leaks like Hydrogen gas.
This issue needs a full post, but briefly, I think helium scarcity would prevent the kind of giant airship industry I'm envisioning here (unless helium proves to be a lot more abundant in the Earth's crust than is currently known, and high helium prices might provide the impulse needed to discover that; or possibly fusion power could arrive at last and make helium as a by-product); but the risks of hydrogen may prove manageable. The Hindenburg didn't burn up because of hydrogen but because of flammable paint.

Another commenter, Ryan C, seems friendly, but then sells giant airships short for no apparent reason. He writes:

#1: the author is right, in that transforming a chunk of the valuable transportation industry would be worth a lot.

Airships have been a commercial/military technology for over 100 years, and they’re not getting any better, are they? I recall serious commercialization efforts in the 1980s, to go along with all the other serious efforts dating back to WWI. The only difference I can think of today is there might be more private equity money available, if the missing piece is a lot of startup money.

Don’t bet on airships to beat trains: if there is money in it, we can already make trains go 100 mph.

Don’t bet on airships to beat ships: we already have a slow/fast split for air and shipping across oceans, so at best I see airships poaching a bit from airplanes and a bit from ships.

There might be a niche for airships, but I predict it is only a niche: something like replacing heli-logging, or doing runs to the Far North, where any form of transport is challenging.

My guess, if the perfect working cargo airship arrives, is that it would take away a modest amount of business from ships and aircraft, and dominate several niche roles, mostly taking over jobs we do with helicopters.

And yeah, they are very challenged by wind.

Ryan C. needs to take a long, hard look at Table 1. If airships can undersell airplanes by a factor of more than 10, while outpacing ships by a factor of five (plus being able to go over land), why in the world would one dream that they would  only "poach a bit from airplanes and a bit from ships?" Trucks and trains dominate overland transportation. Ships and airplanes only compete for small slices of cargo at the very high and low ends of the value spectrum. Most shippers prefer the middle options in cost/speed terms. Giant airships would be to trans-oceanic routes what trucks and trains are to overland routes, the middle option. Why shouldn't we expect them to grab a similar market share of trans-oceanic shipping? And if they did, you have yourself a trillion dollar industry. Maybe Ryan C. doesn't trust my cost and speed estimates and just didn't get around to saying so. As it is, his skepticism makes little sense.

As for competing with trains and trucks, giant airships wouldn't need to, in order to become a trillion dollar industry. And yet it's likely that they would. Giant airships could beat trucks on cost, though they're inferior to trucks for last mile service. That points to intermodality, with long haul by airship followed by last-mile distribution on trucks. Unless stores are built so giant airships can land on the roof and unload, or the last mile service is provided by drones, which are interesting possibilities. As for rail, it's cheap if you've already built the tracks, but how much railroad construction happens these days? A lot of major cities aren't served by rail.


  1. > 5 trillion ton-miles ÷ 50 million ton-miles per airship = 10,000 giant airships to carry all this freight

    This should be 100 000, no?

    1. And this takes the value of all airships to $3 trillion.

  2. Sometime in the 1970s I sat next to a fellow on a commercial airtight who was very enthusiastic about the cargo capabilities of large zeppelins. Maybe he meant large airships. I hope they work.

  3. This is the sort of analysis where stating your minimum and maximum estimate instead of only your central estimate would be much more convincing.

    Unfortunately, in transport technology, it is quite common for entire modes of transport to capture close to zero percent of the market. Hydrofoils, hovercraft, ballistic or glider aircraft, road trains, etc all have nearly zero percent of the market despite being contenders in the cost per ton-mile and speed contest, on the surface of things. That leads me to think that transport technology has relatively strong winner-take-all effects, network effects and path dependency.

    That's the reason why scepticism of your estimates is warranted. Basically, the real estimate of how much of the market airships will make up is either ~0 or ~15%, with very little chance of anything in between. Either the technology takes off, and the presence of airships and airship stations encourages the creation of more in a vicious cycle until they saturate their market, or the technology doesn't take off because of existing sunk costs in vehicles, infrastructure, expertise, etc and it stays a niche technology. It's unstable.

  4. First, What is your estimate that there will be 10 giant airships moving freight profitably 10 years from now?

    The low hanging fruit market is probably California to Hawaii freight. Choice is air at $1 per ton mileand 8 hours, or ship much cheaper and 1 to 2 weeks, and there is lots of traffic. Fill freighters at $1 a ton mile (2,500 miles, or $1.25 per pound / trip, trip takes a little over a day. Can you make money ?


Post a Comment

Popular posts from this blog

Ryan Cooper on "The Return of Airships"

Big is Beautiful: The Mind-Boggling Size of Giant Airships