viable megaprojects, part 2

=economics =technology



I wrote a post about viable US megaprojects, and people wanted another post on that topic.



Any project has some set of costs and benefits. I defined megaprojects as being expensive, so the costs would be large, and if they're viable, then the benefits are even larger.

For such large benefits to be plausible, projects must produce results with large demand. For example, the ultimate outputs of the projects in that post were basically:

- living space
- electricity
- electric wires
- transportation
- military
- materials (including plastics)


Most of the reasons those projects haven't happened involve government:

- political gridlock and lack of vision
- political gridlock and incompetence
- political gridlock as a response to regulatory capture
- political gridlock and lack of vision
- government lack of vision and technical incompetence
- government procurement incompetence


But the last one needs some further consideration. Why would companies fail to invest in viable production of basic materials?



The term "project" represents a a conceptual bottleneck between inputs and products that may be myriad, but to determine viability it's necessary to go into more detail about those.

The main input is natural gas, and enough natural gas is available thanks to the success of fracking. But other inputs include the pressure vessels, turbines, and distillation columns of the plant itself, as well as the welding of the pipes. If there's a shortage of welders, it might not be possible to supply the pressure vessels and pipe welding necessary for a significant expansion of chemical plant construction.

The output is methanol. This can be used in several ways, including some potential new ways I've been thinking about. A major use that could be significantly expanded is methanol-to-olefins. That produces largely propylene, which is polymerized to polypropylene.

I suggested methanol production as a viable only because I think usage of polypropylene could be significantly expanded, and that involves still more detail. For example, I think polypropylene should replace PVC in piping. Obviously PVC has been used heavily, so before taking down Chesterton's fence, what's changed?

- Relative propylene costs have dropped due to improved processes.
- Phthalates leach into water, and it's become clearer that this is bad.
- Polypropylene is harder to weld with solvents than PVC, but alternative connection methods have improved.

Of course, pipe lifetime is important, and it's necessary to consider stress corrosion cracking of polymer pipes for pressurized chlorinated water, so while it's better to use polypropylene than PVC because of problems with the plasticizers and stabilizers used in PVC, there should eventually be better options. But this is just one example; polypropylene can also be used for things like house siding and honeycomb cores.


I've done a smidgen of such analysis professionally, and I'm probably better at it than the average chemical industry analyst, but that's not why I can make lists of viable projects that haven't happened yet. The actual reason is: what I'm doing is an easier problem.

What I was doing is looking at things that are in demand, looking at what's available, figuring out the best routes connecting those, and then describing key points of such routes. What companies have to do when deciding whether to build, say, a big methanol plant is different: they have to predict what prices are going to be. I have no idea how to predict the price of propylene a few years from now beyond saying it will be within a certain (fairly wide) range adjusted for inflation. Prices depend on the distribution of profits, and that's harder to predict than the overall viability or under-utilization of a route from an overall societal perspective.

The point of that post was more about failure to do projects in general than advocacy for those particular projects.

Economists talk about the "invisible hand" of markets, but Adam Smith didn't really consider investment decisions. Lots of small businesses have bought machinery that ended up being useless due to offshoring or changing economic conditions. Lots of students have studied to become, say, lawyers, and found that there was an oversupply as they graduated.

What chemical companies end up doing is making long-term contracts to buy or sell things at fixed prices. In other words, a planned economy, rather than a market. Not a centrally planned economy, but one in which the planning involves multiple companies that hide information from each other. That system tends to get outcompeted by a centrally planned economy, by which I mean a single large corporation doing everything itself. But the more significant purpose of vertical integration is preventing competitors from using things, rather than synergy or coordination. In this sense, the Chinese system is arguably more free-market than the current state of large American corporations.



From this perspective, if we want to find more viable megaprojects that aren't being done, we could consider things with a large minimum scale and a large number of processing steps that make coordination of anything new difficult.

An obvious example is aircraft development. Boeing has no new aircraft in development. They might not even be able to develop a new aircraft anymore. Meanwhile, Airbus is now outcompeting Boeing on the relative strength of existing offerings.

Development of a new blended wing body passenger aircraft and manufacturing for it could be considered a megaproject, and that seems worthwhile to me, but only if development isn't botched, and I'm not sure if America has any institutions capable of competently developing that. Maybe Northrop Grumman could do it?

But this is a different kind of failure than the original theme: loss of institutional capability for tech development, rather than failure to do projects that ways to do are already known.

Construction of skyscrapers in a city like New York is also a different type of problem: it's well-understood and potentially very profitable, but everyone with the power to obstruct construction wants as much money (or favors) as they can get.

So, is there anything else that matches the original theme? Let's see...

The US has recently had high prices for lumber because of a shortage of mills, and high beef prices despite cattle prices not increasing because of inadequate processing capacity and an oligopoly led by Cargill. Arguably those cases fit the same theme, but the undersupply of meat processing capacity was more deliberate and less based on price uncertainty.

In that case, how about...

Western N95 mask production didn't immediately ramp up once it was clear SARS-CoV-2 would be a pandemic, due to concern about low prices from overproduction. The US government could have ordered a billion N95 masks for a stockpile with delivery due once the pandemic was over; that would have spurred immediate expansion of production. It also could have maintained its mask stockpile, and maybe kept the masks in a nitrogen atmosphere for longevity.

But wait, this is forgetting about the "megaproject" part of the theme: N95 mask production doesn't require a billion dollar scale for viability.

Well, silicon ingot production fits that better. Lots of silicon is needed for solar panels, and investment has been hindered largely by uncertainty about future prices.

I could go on, retrospect, I'm not sure what the exact theme was.



Again, the point of that post was mainly about failure to do projects rather than advocacy, and I thought an extended list of projects wouldn't further that point as well as this. But that might not be what my readers are looking for. Maybe there's more demand for the kind of speculation about important future technology that you might find in Scientific American or NewScientist or r/futurology. Unfortunately, I have the rather prohibitive disadvantage of caring about viability and understanding why most of the things they write about aren't practical.

I suppose I could list a few of the major projects involving new technology that I think are viable, such as:

- Large-scale production of GaN wafers, using new processes that I don't want to talk about here because of the strategic significance and applications in eg military AESA radar. GaN is the future of power electronics, not silicon carbide.

- Growing lots of certain kinds of miscanthus and making certain biorefinery designs to convert them to levulinic acid and furfural, then processing that to polymer precursors and possibly (depending on CO2 mitigation value) fuels. I guess I could potentially just describe the design in a post, but the details are beyond the scope of this post and a bit hard for most people to appreciate.

- Large-scale production of xylylenediamine and trans-1,4-H6XDI for high-resilience polyurethanes for tires. Typical polyurethane loses more energy as it bends than polybutadiene rubber, but certain polyurethanes don't, and polyurethanes are generally much more durable.

- Production of nano-precipitated CaCO3 as a replacement for carbon black in tires. Release of carbon black from tires as they wear away on roads is actually a significant source of particulate pollution. Calcium carbonate particles are much less hazardous, and it's possible to get the same performance as carbon black (or even slightly better) if you precipitate it in very particular ways.


I can make lists like this, and I can even go into more detail, but this kind of content only has value in a certain context.



back to index