Construction Physics

It’s become fairly common knowledge that Yellowstone National Park, in addition to being incredibly beautiful, is sitting on top of an enormous supervolcano that catastrophically erupts every few hundred thousand years.

(note: all quotes are from Nemet’s How Solar Energy Became Cheap unless otherwise noted.) Welcome to Part II of “How did Solar Power Get Cheap?” To recap Part I, the modern solar photovoltaic (PV) cell was invented at Bell Labs in 1954. Early markets were almost entirely satellites, followed by other remote power uses such as navigation buoys and offshore platforms.

(Note: unless noted otherwise, quotes are from John Perlin’s “From space to earth: the story of solar electricity”) Solar photovoltaics (PV) have become one of the cheapest sources of electricity. Lazard’s estimate of unsubsidized levelized cost of energy (LCOE), the average cost of electricity generated over a plant’s lifetime, has utility scale solar PV cheaper than anything except completely depreciated natural gas plants and wind in the very windiest locations.

OpenAI recently released an updated version of their GPT large language model, GPT-4, and have incorporated it into their ChatGPT chatbot. In case you've been living under a rock, ChatGPT is a chatbot that uses large language models, which use artificial neural networks with many billions of connections between the neurons, and are “trained” on huge volumes of text and other data.

Last week we looked at trends in skyscraper construction speed for New York and Chicago, finding that New York has gotten significantly slower at building skyscrapers over time. Chicago, on the other hand, has declined in speed less steadily, and currently builds skyscrapers much more quickly than New York does. The obvious next step is to look at skyscraper construction speed around the world. How do modern New York and Chicago compare with other cities, both in the US and abroad?

It’s good to be able to build things quickly. The faster you build something, the quicker the benefits from it accrue, making it more valuable. Additionally, you can reallocate your resources to another project, letting you build more things overall. Building things faster also generally means building them cheaper. If you’re building slowly, you’re probably building inefficiently, spending more time and effort than is really required.

When talking about (the lack of) construction productivity growth, or the fact that we used to build things much faster than we do today, commentators frequently mention the safety of the construction workers. On this view, construction speed/efficiency and worker risk are a tradeoff, and as a society (for better or for worse) we’ve opted to reduce risk, making construction work safer at the cost of making it slower and less efficient.

This week we’re continuing our investigation of productivity trends in US construction. We previously looked at single family home construction, and noted that the number of hours required to construct 100 square feet of single family home has slightly increased over the past 50 years. While I think this is a useful metric of construction productivity, it has some drawbacks. The primary one is that the content of a single family home has changed over time.

The modern world uses shocking amounts of steel - in the US, we make roughly 575 pounds of steel per person per year. At the peak of US steelmaking in the late 1960s, it was closer to 1500 pounds per person, which is roughly how much China makes now. Modern steel is produced by two different methods. The first uses an electric arc furnace to melt recycled steel scrap. The second method is to make new steel out of iron ore.

In the New York Times, Ezra Klein investigated the recent Goolsbee and Syverson paper on construction productivity we recently looked at. Klein suggests that the stagnation in construction productivity might be the result of organized special interests increasingly leveraging their influence over various parts of the construction process.