This ambivalence is one of humanity’s oldest and most consistent intuitions. Technology enlarges the range of what we can do, but in doing so changes the conditions under which we choose. It gives us power over nature while threatening to subject us to systems, incentives, and purposes that no individual intended. The recurring, deep fear that permeates the history of human consciousness is therefore not simply that a machine will malfunction. It is that, by building machines powerful enough to transform the world, we will transform ourselves and lose control over what we become.

As in all things, probably the best articulation of ancient fears was the Greeks. Unlike most cultures, the Greeks had a god of technology who invented: Hephaestus, the blacksmith of Mount Olympus who created everything the Olympian gods needed. He was not merely a handyman, but also an inventor. In Greek mythology, nearly every technology, whether fire or Pandora’s box, comes from his workshop by conveyance or pilfering, and all come with a loss of control for humanity. Hephaestus himself also suffered a consequence as a result of his blacksmithing—he was lame, unable to leave his workshop. Although his loss of leg function was related to being thrown off of Mount Olympus, this was a common affliction of blacksmiths at the time because of arsenic exposure, still making it a powerful metaphor: embrace technology too much, and you will suffer the consequences, just as a blacksmith who spends too much time in his workshop loses the ability to leave.

The Greek fear of technology was not limited to merely Olympian myths; human inventors also suffered from their own creations in classical mythology. Probably the most famous tale of peripeteia is that of Icarus flying too close to the sun and melting his waxen wings. As such, although technology myths are often moral parables, there are Greek myths of explicit punishment for inventing. In the ancient story of the  Sicilian Bull, an inventor created a new torture device wherein a person was boiled alive in a bronze bull. When the inventor proudly demonstrated his invention to the king Phalaris, he became the first person on whom the device was demonstrated. Invent at your own peril. Similarly, Prometheus is punished for acquiring fire, as in nearly every other culture with a “man stealing fire” tradition.

This theme was not unique to the Greeks. Generally, the main fear surrounding technology throughout time and across various historical cultures is unintended consequences. Usually, the invention itself is what goes awry. Perhaps the most famous example is the ancient Jewish legend of the Golem, wherein an artificial being is magically animated out of mud, but escapes human control and embarks on a Yudkowsky-style non-aligned rampage. This theme repeats over and over across traditions. For example, Ilmarinen, a Finnish craftsman, creates a bride made out of gold who is beautiful but still and lifeless, leaving him unhappy, while the Egyptian god Thoth produced books with knowledge that would cause misfortunes like insanity or divine punishment. In Hindu culture, the inventor Maya Dānava builds a mirror funhouse, the Māyā Sabhā, that is so convincing that it tricks a king and triggers a war. In Petronius’s Satyricon, an inventor of flexible glass is killed by Tiberius Caesar for fear that the invention would be too valuable and put glassmakers out of work.

When technology—depicted in Clarkeian fashion as magic or crafts— is personified in a god, they are often associated with unpredictability but also creation: they are both fathers of civilization and tricksters. The Mesopotamian god of crafts and magic Enki, creates humans out of clay but also sometimes manipulates them. In Native American cultures, the Coyote spirit both steals fire for humanity from the sun and teaches them to fish. In Micronesia, Analup is the god of magic and crafts and creator of the world, but also the grandfather of the trickster god Olofat, who, like Prometheus, is credited with bringing fire down from the heavens to mankind. In a darker turn, Odin obtains his magic or crafts through painful sacrifice involving deception and even unwilling third parties. Other times, these deities are associated with unpredictable and destructive events. In Ancient Egypt, the god Ptah was the powerful god of craftsmen worshiped by blacksmiths but also feared as the cause of earthquakes, which destroyed several Egyptian cities, like Thonis and Canopus. Similarly, the Roman god Vulcan, the metalworking craftsman, was strongly associated with fire, which was the uncontrollable catastrophe of choice for Roman fears. Ogun, the Yoruba god of craftsmen, is married to Oya, the goddess of winds and storms.

Modern Fears: Technological Determinism

Comparative mythology demonstrates that these are primal emotions and recurring thoughts known by human beings across very different societies possessing very different levels of technology. It is hardly surprising then that these ancient fears were recapitulated in the work of modern philosophers of technology in the 20th century. Writing in the early 1960s, the philosopher Jacques Ellul believed that technology unfolded in a deterministic path towards decreasing human agency, comparing modern man to Esau, selling his freedom for goods and services; that is, a slave. Our material desires, he wrote, betray and distract us; “it is a matter of no real importance whether man succeeds in reaching the Moon, or curing disease with antibiotics, or upping steel production.” Instead, the “search for truth and freedom are the higher goals.” Ellul’s views are extreme enough to make one mad; this is one reason, perhaps, that his most famous intellectual progeny was the mathematician and terrorist Ted Kaczynski, also known as the “Unabomber.” 

For the German philosopher Martin Heidegger, the risk of technology is that it has a power all its own, of which we become a mere part. In “The Question Concerning Technology,” Heidegger writes of the logger who is reduced by technology into a mere cog in the industrial machine, “ordered by the industry that produces commercial woods, whether he knows it or not.” The technological logging industry itself is “never a human handiwork,” and the logger becomes subservient “as a subject relates to an object.” This should look very familiar—it is the same fears we have been facing throughout our history but told in a more mechanical, industrial fashion.

Such observations were made earlier in the Industrial Revolution as well. In 1829, Scottish historian Thomas Carlyle wrote about the “mechanical age” in his essay “Signs of the Times,” noting moral panic, millenarianism, and increasing inequality as social disruptions caused by fear surrounding industrial progress. He himself praised the mechanical wonders but feared that Scotland was being consumed by a belief in the “all-importance of physical things” and losing its humanity.

As in the ancient world, fears about technology broke into popular culture. Contemporary concerns tend to focus on machines that have their own ends and pursue them to the detriment of humans. Sometimes, like the supercomputer HAL 9000 from the film 2001: A Space Odyssey, it is a misapplication of a rule followed too literally; sometimes, as in Terminator, the artificial intelligence Skynet determines on its own a deadly prime directive. Humans have had a fascination with robots for a long time as well—it is perhaps the only modern technology to have been prophesied in mythology across cultures for thousands of years. The term “robot” comes from a 1920s Czech play, R.U.R., where the robots are abused so badly that they rise up to annihilate humanity. 

What is notable about modern stories of technological fear is that they disregard the traditional intuition that risk comes from loss of control. Instead, they adopt the James Cameron view of the world: rather than being neither good nor bad, technology is inherently corrupting and broadly destructive. Compare the story of the Golem with Mary Shelley’s Frankenstein. Frankenstein’s monster is a purely tragic figure: he is created for no real reason, does not do anything good, overpowers humanity, and dies in as much misery as he created. By playing god, Frankenstein creates pure malevolence in the world. The Golem loses control, but it is created as a protector against blood libels and does accomplish that job and is able to be stopped. The fear is the same, but the ancient nuance is lost where technology is to be feared but also to be appreciated.

Even optimistic contemporary stories are not immune from this affliction. In “Machines of Loving Grace,” Anthropic CEO Dario Amodei paints two possible futures: one where AI spins out of control, and one where machines care for us. Though meant to depict a dystopia/utopia dichotomy, these are fundamentally the same future with different interpretations. In both worlds, the machines supersede us, and we lose control. It is no wonder we fear AI: what machine is more likely to seize control than one that has its own desires?

Giving Telos to Techne

These fears are not entirely baseless. There is a determinism to technology that must be reckoned with. Arguably, we have been going down this path for a long time. Since Carlyle’s 1836 novel Sartor Resartus promulgated the idea that what defines man is making tools, we have increasingly defined ourselves in reference to technics. American historian Lewis Mumford called the totality of human technical systems the Megamachine: a societal ordering that uses humans as parts without individuality.

Even the most strident techno-optimist cannot deny that technology has an internal logic. Technologists often call this the “tech tree,” for the way in which technologies seem to branch outwards as they build on prior technical achievement. Sometimes these dependencies can be surprising, but a tech tree is often the answer to the question of why we had to wait so long. Furthermore, because technologies are built on other technologies, at places where the tree branches, the entire direction of civilization can be altered, creating a path dependency.

Technologies, however, are not just enabling but also revealing, showing us what can be next. Furthermore, we rarely leave technologies unused. Samo Burja has called these co-evolving technologies “social technologies” because they involve our frequent need to socially determine what further inventions we will need to build to use big, important technologies. The lightbulb was co-created alongside the electric utility and many of the first electrical appliances.

Melvin Kranzberg, an American historian, formulated his First Law of the history of technology as “technology is neither good nor bad, nor is it neutral.” That third part is usually skipped over, but in fact it is the most important clause. Technologies often create new possibilities, but in doing so change what we can even choose to create in the world. When media theorist Marshall McLuhan said that “the medium is the message,” he meant that certain things can only be communicated with certain technologies and thus become the communication itself, but also implied that certain messages are only possible once certain media are invented. Many technologies change the relative costs of certain choices, thus changing what economists call the production frontier. 

This is almost enough to make one a technological determinist. In What Technology Wants, Kevin Kelly argues that technology forms a technium, a collective organism that has its own evolution akin to biological evolution. If that is so, as author Paul Kingsnorth has argued, we stop being able to see the world beyond logical terms and get subsumed by the logic of the machines themselves—historian Charles A. Beard quipped that “technology marches in seven-league boots from one ruthless, revolutionary conquest to another, tearing down old factories and industries, flinging up new processes with terrifying rapidity.” 

Yet, the contingency of technology also makes it the most freeing force in human affairs. In the same essay discussed above, “The Question Concerning Technology,” Heidegger ends with a line from Hölderlin: “but where danger is, grows / the saving power also.” Perhaps the only way out of our technological predicament is through.

The problem with the determinist argument, as biologist Jerry Coyne pointed out in a review of Kelly’s book, is that actual evolution is not deterministic but dependent on contingencies and chance. Evolutionary processes possess emergent properties, but that does not mean that they are themselves organisms, just as we are not a single organism with the Earth, even though we emerge out of it and take part in terrestrial planetary processes. In biology and in most human affairs, the conditions of evolution are largely out of our control. They are geography, the distribution of natural resources, disease, natural disasters, differences in climate and the environment, and other aspects of our natural world that were set in motion long before we were even born as a species. The contingent characteristic of technology, and our views on it, depend entirely on framing. You could view the tech tree as unlocking possibilities, or as closing them off. In truth, the process is one and the same.

Not every philosopher was pessimistic about this situation. Aristotle is the ancient philosopher with the most developed theory of technology, and despite coming from the otherwise techno-pessimistic Greek culture, he is himself optimistic about the prospects of man and his relation to his technics. To Aristotle, technology was techne, not just scientific machinic engineering but any kind of craft or artifice. Aristotle argued that techne was distinct from physis, the forces governing the physical world, because it has no internal logic guiding its ends. The statelessness of techne means that it follows its outward purpose, or telos, which comes from us. In a way, Aristotle embraces Kranzberg’s Sixth (and final) Law of Technology: “Technology is a very human activity—and so is the history of technology.”

Aristotle was a utopian, speculating that robots would bring the end of slavery and thus anticipating, on a theoretical level, arguments that would be made thousands of years later. Nick Szabo, the computer scientist and intellectual, argued that the clock led to the end of serfdom by making it possible to calculate an hourly wage. It is no surprise that slavery ended around the time of mechanical farming or that women’s suffrage coincided with the rise of electrical appliances.

One does not require techno-utopianism to see the human element that separates techne and physis, nor to see how this puts us in control of our fate. We may not be able to control the distribution of natural resources at our disposal, but we can invent new ways to make that irrelevant. We may not be able to change our culture and our subjection to the madness of crowds, but we can invent new ways of communicating to subdue that madness. We may not be able to stop an asteroid from approaching Earth, but we can invent a way to save ourselves from the impact.

While Heidegger feared that we would become overwhelmed by the enframing technological industrial society, we could choose, like Aristotle, to see things the other way: that technology comes from us and is thus of us. Or, as philosopher Ernst Kapp put it, that technology represents a “tool so fundamentally and intimately affiliated with the human being that he finds himself beholding something of his own being in the creation of his hand, his world of representation embodied in matter.” This capacity to project our very being into the material world gives us power over our fate.

The ancient wisdom on technology emphasizes our choice to build technology around us. In the 13th century, Masamune, considered the greatest historical Japanese swordsmith, was challenged to a contest by his top student, Muramasa. They both created their greatest swords, but while Muramasa’s sword became bloodthirsty, Masamune’s only cut leaves. The ability to create, and to determine the purpose of their technology, was in their hands. The student, who forgot this lesson, created a weapon that was, ultimately, a failure. The lesson behind the legend is that the craftsman who exercises restraint and choice, and remembers that he can make even a sword peaceful, is remembered as the master of his craft.

If we feel uneasy, perhaps that is our way of telling ourselves that something may be awry inside us and how we fear that we would use the technology. After all, absorbing new technology is a change, and the impact of changes is in the future. What is unavoidable is that changing our technology will change us in ways we cannot anticipate, and that change will, ultimately, be governed by characteristics of the technology itself. That is why we are so fundamentally fearful toward technology: it is not that we fear losing control over it, but that we fear that it will cause us to lose control of ourselves.

The beauty of technology is precisely that it puts our fate back in our own hands because we can always build new technology. We are a factor—the factor—in determining which technologies we develop and how far we take them, because it is our energies and ingenuity that get applied to the development process. If we become cogs in the machine, we can build cogs to replace ourselves or build a new machine. What determines the character and nature of our technology is us. The process of creating technology, and therefore our future, is one of us choosing our own future, over and over again.

The initial response was not panic, but heavy concern. Everyone who grew up in Los Angeles knows about the Santa Ana winds—the legendary 1961 Bel Air Fire, up until now the worst fire in the city’s history, was caused by a Santa Ana wind. So when we heard about a fire caused by the winds it caught everyone’s attention. 

Santa Ana winds can move rapidly. It keeps natives of Los Angeles on a hair trigger. In an hour, everyone heard how suddenly the fire was growing by a thousand acres an hour. Then came the news that, like in the Bel Air Fire, the Palisades fire hydrants weren’t working just as the Eaton Fire, tens of miles away, was also blazing out of control. We knew, immediately then and there, that we were cooked.

Suddenly, panic and helplessness. Taking what we could before fleeing. Our family home in Malibu was rapidly evacuated. We weren’t taking chances. 

We were all looking for information but finding none. Everyone whispered something they’d heard, somewhere. The local news and the internet provided information far faster than the official sources. Cal FIRE’s website is supposed to update every twelve hours at the latest, but that sure wasn’t happening. You could look up the UC San Diego’s livestream cameras, but that only gave you limited information. 

We all waited anxiously throughout the night. We watched the dancing high flames from that UCSD feed, but you couldn’t tell from the video feed where the fire was exactly. Every Angeleno or former Angeleno was calling everyone they knew to make sure they were alright. I went to bed at 1 am, not knowing if our house was still there but knowing that, probably, this had become a lottery that promised few winners.

The next day, our neighbor snuck in and called us. His house was the only one on the entire street that survived. We learned later that of the 250 houses in the neighborhood, less than 30% remained. The next day, we learned that, tragically, someone’s mother was trapped in her house and died. I wondered if my family would have to flee; we preemptively had everyone pack their things.

Every day, there were at least two new fires. Was this arson? A terrorist attack? How did this failure happen? We heard that arsonists had already been arrested, including for the Griffith Fire. Everyone was awash in rumor. You could never truly rest until there was rain because though you might be safe from the current fires, what about the next, random 11 pm fire?

More acres have burned in Los Angeles than the entire area of Washington, D.C. or San Francisco. Two of the fires alone were larger than Manhattan with a third one nearly as large. The damage estimates keep going up, but AccuWeather puts the cost at over a quarter trillion dollars. Dozens have died, and more will be found; over the coming years, thousands more could die from having inhaled toxic smoke.

Our leaders fiddled and watched the state burn, of course. Seven of the ten worst fires have happened in California just in the past ten years. This is now the eighth. Governor Gavin Newsom, who previously overstated prevention efforts by a factor of eight, took days to bring in state troopers. We somehow don’t own any “super scooper” planes that take up water from the Pacific.

The head of Los Angeles Department of Water and Power (LADWP) said that diversity was her top priority while her deputy went to prison for corruption; the LA Fire chief, though a very brave and accomplished firefighter, put technology last on her list of priorities; Mayor Karen Bass said to find information at “URL;” someone arrested with a blowtorch was let go for insufficient probable cause—he said the blow torch was to light marijuana because he had no lighter. A perfect storm of California governance dysfunction. In Los Angeles, everyone knows that someone is to blame, but they don’t know who. And perhaps that opacity of accountability is the biggest common cause behind our state’s tragedy.

Air, Heat, and Fuel in the Urban Landscape

A fire, which requires air, heat, and fuel, typically starts with an ignition event, such as a falling power line, arson, or even friction caused by extremely strong winds. That ignition provides a spark, which specifically provides the heat. That spark must be near some type of fuel, like wood or dry brush, that can then catch fire. Once that is complete, it will then take up the open air. This is why, for example, break fires on airplanes are so dangerous; the magnesium fire creates its own oxygen, so water actually makes the situation worse.

Left to themselves, fires will typically expand radially until they are put out by a natural event, like rain, or burn themselves out by running out of fuel or hitting a natural barrier. The actual spread is a result of the flames themselves or embers, some of which get unfortunately dropped onto something that is just dry enough to react with the heat from the ember. Wind makes things much worse because it both makes the flames higher and allows the embers to go further than most people realize—in strong winds, embers can travel more than five miles.

Urban fires and wildfires are different in how they start and how you fight them. When we read about gigantic fires in California, they are usually wildfires in some type of forest. They don’t happen terribly often, but because they are in the middle of nowhere and surrounded by fuel, they are slow to detect and grow to be large. This is why you hear so much debate about brush clearance and “forest management.” For wildfires, to a significant degree, this is the name of the game. 

Urban wildfires are quite different. While cities exist in nature, of course, they are an oasis of concrete and steel where fires happen more frequently but do not grow as large. Ignition increases nonlinearly with human density because of the large number of sparks, but even the Great Chicago Fire and Great San Francisco Fire were both under 2600 acres, making them smaller than even several of the Los Angeles fires last week. And more importantly, because they are dense, rapid detection and response is easier. The trade-off, however, is that the human devastation is much worse than in nature because there are so many more people. 

Over a century ago, people started to live together in more density than they ever experienced before. We never really “solved” wildfires, but they didn’t impact people, so they hardly mattered. But the increased density of urbanism and power-heavy industrial activity (often at mills) transformed fires from an unfortunate incident for a barn on a farm into a conflagration that destroyed entire cities. By the 1870s, “great” fires were happening several times a decade and viewed as a normal part of life in cities. 

And then, by the 1920s, it stopped. Why? After the Great Chicago Fire of 1871, we finally got serious as a civilization about stopping urban fires. We rewrote building codes to require fire-resistant materials and metal escape ladders; we built professional firefighting forces instead of relying on local fire brigades with, literally, buckets; and we invented new technologies, like automatic sprinkler systems in 1872, motorized fire trucks with powered pumps and engines in 1910, and CTC fire extinguishers in 1912. Chicago never burned down again. 

Today, urban fires are treated as a largely-solved problem. Modern urban firefighting forces and infrastructure are designed for putting out fires in homes. In fact, firefighting is so solved that only 4% of firefighting calls are fire-related—the vast majority are medical. Yet, this model of firefighting is not adapted to the challenges we face today.

The Evolution of the Palisades Fire

The Palisades Fire started just southeast of Palisades Drive, seemingly in a backyard. At least one arsonist has been arrested so far, though we don’t yet know how this fire, which has been the most destructive, was started. The first fire report came in the morning with a call to 911. The Los Angeles Fire Department (LAFD) had a 45 minute delay in its response and only sent five engines, refusing a 1000-firefighter and 35-truck request. 

All 114 water storage tanks were full, including three backup firefighting tanks of one million gallons apiece. But these tanks are not fast to refill, and they work by using gravity to move the water downhill to the Palisades. On top of that, LAFD stupidly canceled a critical fire hydrant test just a few weeks before the fire, with no newly-planned test date. Therefore, with the massive demand on the system, water pressure dropped precipitously. 

The water pumps failed within hours. During that critical hour when the fire could have perhaps been kept from exploding, the firefighters had no water.

The Santa Ynez reservoir—built after the 1961 Bel Air Fire to prevent exactly this type of catastrophe, with a capacity 117 million gallons—was offline for maintenance. Somehow, fixing a lid has taken nearly a year and required emptying the reservoir, likely contributing to the critical drop in the Palisades hydrant pressure and, of course, meaning less water. Oh, and LADWP never told the fire department they had emptied Santa Ynez.

Oops.

Once the Palisades Fire spread to a ridge, it rapidly became uncontainable. With the dry conditions following no rain and winds at times exceeding 120 miles per hour, nothing could be done beyond damage control, at least for the first night. Ditto the Eaton Fire near Pasadena and the NASA Jet Propulsion Laboratory housed at Caltech.

By that point, no fire department could handle two gigantic fires growing rapidly, especially when they were over twenty miles apart. Once flames exceed four feet in height, non-water solutions become impossible. At the height of the Santa Ana winds during the Palisades Fire, flame heights exceeded thirty feet. It was unsafe to use water aerial vehicles, which on the fateful first night would have been the only viable way to put out the fires. 

In my own neighborhood, there were not even any firefighters that Tuesday night. The question of why it took so long to get help, with the Point Mugu National Guard and Federal Emergency Management Agency (FEMA) only arriving days later, will require a serious investigation. There are many questions to ask.

Yes, you can ask why they weren’t testing fire hydrants, though you should really be asking why we tolerated fire hydrant failures for years. We can ask why Governor Newsom disbanded a rapid fire firefighting force last year instead of doubling its size. We can ask why Mayor Bass has had more corruption scandals than fires she’s put out or why the head of LADWP hired a chief executive with no water experience. All of these acts of brute stupidity represent a short-term governance failure. 

But there is a long-term governance failure too, and that is an unwillingness to adapt firefighting to the changed urban landscape this century.

The Modern City Offers No Off Season

Over the past fifty years, people all throughout developed nations have moved into suburban areas, which fire experts refer to as the “wildland-urban interface” (WUI). As far as fire risk is concerned, these areas combine the worst aspects of wildland and urban environments. Because of humans living in density, you have frequent ignition events. But because they are near nature, you are surrounded by kindling. The environment is furthermore relatively sparse, so you can’t have the same firefighting density as in a city. Taken together this means fires can reach wildfire scale with urban frequency.

Simply put, urban firefighting forces are using an old playbook on a new, unsolved problem. 29% of the United States lives in the WUI now, and California has the highest such percentage of any state. Add to that increased dryness in Los Angeles, which is 3 degrees Celsius above pre-industrial times. The population moving to such zones over the last few decades, just as California turned itself into a tinderbox, was an obvious recipe for disaster.

Perhaps this could be addressed with contemporary techniques on its own, but modern firefighting doctrine evolved in an era where fires have an “on” and “off” season with a methodological divide between urban and wild firefighting. This is the core, fundamental problem for California: there is no “off” season anymore. Every season is fire season. Yes, there will be a high and low cycle, but the presence of material risk is now a constant.

The paradigm of seasonal firefighting extends to everything, from staffing to infrastructure to preventative maintenance schedules. 

The off season means that personnel are largely part-time. Only 35% of firefighters nationwide are career firefighters; 13% are paid-per-call and 52% are volunteers with normal jobs. Though California numbers are not readily available, on a departmental basis, California has more career fire departments than average. Despite that, California firefighters now work on a 66-hour work week—amazingly, this is a reduction from 72 hours—which includes extremely expensive overtime pay. During the off-season, firefighters work in EMS, security, train, or just take a break.

Infrastructure is also built under an assumption of an off-season. Take the unconscionable draining of the Santa Ynez reservoir. There is nothing in itself wrong with taking a reservoir offline for maintenance; after all, maintenance is a normal part of any infrastructure. And yes, it was stupid to do so during a very dry year instead of, say, last year, when there was more rain than in the past quarter century. But if your model assumes an off season, then even during a dry year, there is a rational time to do it, and that would be the off season when fire risk is low. Certainly, there is no reason to spend money on an extra reservoir if you have an off season. 

And environmental maintenance? Well, who needs that if there’s an “on” season and fires will happen anyways? California’s annual timber harvest has fallen from 6 billion board feet in the 1990s to 1.5 billion now, a 75% decrease. California’s forests are now anywhere from 80% to as much as 600% denser than historical averages, according to scientists at UC Davis. Trees are sharing more resources than they have evolved to support, in part thanks to a decades-old policy of absolute fire suppression. What this means is that there is more kindling than ever, which both makes ignition more likely to start a fire and means that when a fire takes off, it is more likely to become huge. Despite the fact that Southern California is known to be a high-risk area, there is not much currently planned thinning, and amazingly, the federal government canceled many of its controlled burns this year. Somehow, even after these fires, California’s legislature rejected a $1 billion fire prevention package.

If we say that there is no more fire season, what would it mean to take that seriously? It means that responses must be instantaneous and that every preventative tool must be utilized constantly. When it comes to stopping ignition, we cannot stop the sparks that come from civilized life; this primarily leaves us with indirect action through environmental management. When it comes to “brush clearance,” the solution must be tightly tailored to each biome. 

The most important tool, especially for high-risk areas, is usually controlled burns, which reduce wildfire risk by 60%. If you want to see the value of controlled burns, look no further than the Palisades fire, which spread into Malibu—but stopped around the Malibu Pier, which had experienced a fire just a few weeks earlier. However, it is not the case that all thinning must be controlled burns. 

Mechanical means such as cutting down trees work just as well in low-to-moderate risk areas, and will allow for more pruning near residential areas especially near power lines. In fact, in places like the Pacific Palisades, burning is the wrong strategy because the chaparral gets replaced with grass, which is more flammable, even though it is generally high-risk. Mechanical thinning has other benefits too. Unlike controlled burns, timber can be used to support Californian industry. On top of that, reducing forest cover adds 9% to California’s available runoff water, adding potentially millions of acre-feet of water to California’s water supply for free.

To allow for more preventative maintenance, there must be reform to both the California Environmental Quality Act (CEQA) and the National Environmental Policy Act (NEPA). The environmental impact statement of a controlled burn takes 7.2 years to complete, which is longer than most fire cycles. Amazingly, under Sierra Club v. Bosworth, a 2007 Ninth Circuit case, there is no categorical exclusion for controlled burns. 

Congress should exempt firefighting measures like this entirely from NEPA. The federal and state governments will need to work together because 46% of California’s 100 million acres of land are owned by the federal government. These are not just rural areas, like Yosemite, since the federal government owns land directly in urban areas, like the Presidio in San Francisco.

In general, the legislature has resisted CEQA exemptions that would allow projects to proceed at a normal pace. SB 1003, which would have provided an exemption for burying power lines, died in committee. SB 1159, which would exempt vegetation clearance projects, did too, as did AB 1554, which would have exempted fire prevention methods like controlled burns from CEQA. The governor’s programs meant to run-around this regulation were so ineffective that groups were told to just use CEQA instead. There have been other legislative failures, too. For example, Governor Newsom vetoed AB 2538, which would have allowed Cal FIRE to retain seasonal firefighters. 

Similarly, California must plan for much more water accessibility, both in terms of surge capability and a higher baseline water availability. Major urban centers should not just have one reservoir, but at least two for redundancy. All fire infrastructure must go through an audit and, if need be, rebuilt. California throws 21 million acre-feet of water away each year, which is more than enough to fill firefighting reservoirs and use for irrigation to wet our forests. The money is already there—California voters already approved $2.7 billion for reservoirs with Prop. 1 back in 2014, none of which have been built a decade later. All that is missing is the state capacity to do so. 

Unfortunately, you can’t just take water from the San Joaquin Delta arbitrarily. For example, 15% of the environmental outflows are done to allow water used for urban and agricultural uses to remain at acceptable quality, like salinity control against inflows from the San Francisco Bay. However, technology can save the day here. Desalination will be required to meet some of these needs so that residential and farm uses can come from desalinated water and the remaining water can be used for irrigation and firefighting. At-scale filtering can also help make sure that water is used appropriately. This can not only be used to double California’s number of reservoirs, ensuring that major urban centers have redundant options, but also be used for irrigation, which can help fight fires by reducing the dryness of potential kindling. 

Beyond that, other than regular PSAs about cigarette disposal and not smoking in nature, the only clear way to reduce ignition directly is to deal with homelessness. The homeless cause 54% of all fires in Los Angeles. That number jumps to 80% for downtown fires. In San Francisco, such fires have doubled in the past five years. Throughout California, the homeless plague highway underpasses with fires, some from cooking and some from derangement. Many of the arsonists arrested during these fires were homeless. Unfortunately, the eternal summer of endless fire season means the time for tolerance is over.

Staffing Shortages Can Be Overcome With Technology  

The problem with all of this is personnel. If we are going to embrace the death of the off season, we need more people, but we already struggle to meet our current needs. The U.S. Forest Service has lost 20% of its firefighting manpower in recent years. Attrition for firefighters is terrible. Currently, 35% of firefighting jobs are vacant in California. Overtime pay is nearly a third of firefighter pay. 

Among America’s largest cities, Los Angeles has the second-fewest firefighters per capita. Only San Diego—which was battling its own major fires at the same time—has fewer. Astoundingly, despite a larger budget than ever, Los Angeles now has fewer firefighters than it did 36 years ago despite serving four times more calls, and was facing budget cuts that might have “impacted” its ability to respond to catastrophe. And it isn’t just firefighters: when the LA fires broke out, fire trucks were simply out of service for want of sufficient maintenance staff. 

If there is no more fire season, then a workforce that is only 35% professional and built around fire season surge capacity from volunteers isn’t going to work. The maintenance, repair, and system monitoring needed will require a surge of staff. Ditto environmental management, like brush clearance and forest management. While this is something that the firefighters can do during the low season—and it is safer than firefighting, which should make it easier to recruit for—the math is difficult to square. This is leaving aside the question of whether dysfunctional organizations can even make use of increased staff effectively.

But thankfully, through using software, automation, and electronics, we can make up for this shortfall. Technology must become every fire department’s highest priority, not its lowest, and technology cannot just mean IT.

The first, most important priority must be rapid detection and response. The Palisades Fire started around 10:30 am, but it was not detected by a plane using infrared until the afternoon, by which point the fire was over 770 acres in size. The Cal FIRE website, which could have provided critical real-time information to volunteer firefighters and concerned residents alike, was updated infrequently, requiring dependence on independent apps like Watch Duty. This is mind-boggling, particularly since the analytics for predicting which blocks should be on a red-flag warning are so accurate.

Sensors are now incredibly cheap. A camera with a machine learning algorithm written by a high schooler can detect smoke. There are more sophisticated sensors now that integrate cameras with other sensors, like gas and spectral sensors, made by companies like Torch and Pano. These should be all over urban areas. While a mass distributed sensor system may not be entirely practical for 100 million acres of wildfire prevention, it may be more practical to use satellites and pole-mounted cameras with wider ranges and less sensitivity. Today, fire detection is almost universally a human-triggered endeavor. Urban fire response is triggered by a 911 call, while in forests “fire lookout” is an actual job where park rangers sit in tree houses and look for smoke with their eyes. A ludicrous arrangement given that labor could be freed up with cheap, solar-powered cameras on poles installed every 100 acres.

Better, more highly proliferated sensors can also allow for better analytics. Weather systems are already highly accurate—in fact, Mayor Bass received warning of high-risk Santa Ana winds last Thursday, nearly a week before the Palisades Fire. (She still chose to go to Ghana instead.) The by-the-minute predictions are less accurate because there is not enough data; firefighters must rely on visual analysis or infrared planes, which must be flown. This can not only allow more data to be provided to citizens but also to firefighters, so they can more accurately get ahead of a blaze. This will allow for better planning and maintenance—imagine if, instead of testing fire hydrants on a set schedule, there was 24/7, by-the-hydrant analytics through water pressure sensors, for example; this would reduce the need for more bodies. 

It would also allow for better surge provisioning. When Florida had its most recent hurricane, Governor Ron DeSantis was able to prepare 42,000 linemen—that is not a typo—to restore power. While a hurricane is more predictable than a fire, better analytics could make that type of action a new possibility for fire.

Sensors must be met with a rapid response. For that, sensors can be used not only to coordinate people but also drones. There are simply not enough firefighters to verify and respond to every call, especially in an emergency situation like the Los Angeles fires. But drones can cheaply respond to an almost infinite number of fires. They can be specialized firefighting drones from companies like Rain or modified commodity drones. They can provide direct video feeds and even autonomously drop water to put out spot fires. For ignitions that are growing rapidly, even more drones can come and spray fire retardant to slow the growth of the fire in time for firefighters to arrive. Instead of using an error-prone manual evacuation alert system, why not use an instantaneous, automated system to lower false alarms? Drones from companies like Kodama can also be used to help cut the cost of thinning forests. On the more experimental side, fire departments should investigate cloud seeding, both to increase background moisture and potentially for firefighting when containment becomes impossible.

More must be done to spur the firefighting technology category–firefighting tech has only attracted $100 million in venture capital and is less than five years old as a category. There is already an effort underway with the XPRIZE Wildfire Prize, but there should be more such prizes. Firefighting departments committing to buying the solutions of the winners of such prizes, and putting out rapid competitive requests for proposal, will help a lot. On the regulatory side, the Federal Aviation Administration (FAA) should consider an immediate Beyond Visual Line of Sight (BVLOS) exemption to the FAA regulation that restricts drones, Part 107, for firefighting drones, and explicitly allow them in the upcoming Part 108 regulation when it is issued, which will allow things like delivery drones. 

What this will ultimately mean is a massive increase in capital spending and effective procurement processes, which is a change from the way firefighting institutions work today. Today, the LAFD budget is 93% salaries. LADWP spends less than 10% of its budget on capital. Every institution must assume it will have to invest in major projects. SoCal Edison, the utility company, will need to invest dramatically in undergrounding cables, which can cost seven times more than fire-resistant pole techniques, but it will have to be done. The LAFD will have to make nine-figure investments in sensor blankets and drones and maintain an ongoing line item in its budget. The LADWP has known about issues with the firefighting water system and put off a pumping system for a decade; it will have to overhaul its process to react quickly and boost the number of projects. Every place with a fire hazard will have to undergo a similar analysis and invest in projects like these.

This will require a more predictable budget, though we must audit our current spending. The current cuts are counterproductive. Yet for all the money we spend—LAFD’s 2024-2025 budget is approximately $819 million with over $100 million in requested additional spending while Cal FIRE has a $4 billion budget—we still have crumbling fire stations and secret budget cut negotiations that would cut sixteen stations. How can a city that spends nearly a billion dollars per year have fire hydrants that fail? How can it have 1960s technology instead of modern tooling? We need to invest in a top-to-bottom review of water and firefighting infrastructure and audit the finances of every major fire department. Just because the firefighters are incredibly brave does not mean that the fire departments themselves are blame-free.

Yes, California needs to throw the bums out, and even put some of them in jail. And yes, we will learn over the coming weeks about many operational and logistical failures, not to mention arsons. But with sufficient technology and planning this should have never happened. However a fire starts, California should be able to rapidly contain and extinguish it. 

We are investing more and more money and getting worse and worse results. It is not just incompetence and corruption. Even the right managers and staff would not be effective with the wrong tools. California is not the only state at risk. All of the Western United States, from Colorado to Texas, faces the same issue. In fact, the same can be said of other countries with similar patterns of urban development, like Australia, Canada, and even Mexico. 

After the 1871 fire, every city in the world marveled at Chicago’s practices and copied them. Let’s do the same with California in 2025.

This is strange because governments are the one thing the Bay Area does not lack. The region is made up of 101 municipalities distributed among nine large counties. In addition, numerous agencies and institutions operate across these boundaries. One might expect that the fight for talent and capital in the region would force these myriad administrations into competence, but this never seems to happen.

In reality, the large number of governments is the root cause behind why nothing ever seems to get fixed in the Bay Area: the economy is regional, but governance is local. There exists a fundamental mismatch in scale that means no one really represents the interests of the overall Bay Area, nor has the power to govern it coherently. This is causing a fundamental breakdown. The only solution is to collapse this disparity.

Too Big for Its Bridges

The Bay Area has a highly integrated regional economy that is structurally comparable to the more populous New York City tri-state area. Employers and people are both distributed around the Bay. Facebook is based in Menlo Park, but its large office 32 miles away in San Francisco holds about 10 percent of its workforce. Stripe moved from San Francisco to a different city—South San Francisco—almost on a dime. Before the pandemic, San Francisco proper would have a 20-percent population surge during the weekday, tech giants had fleets of buses that shuttled 34,000 employees daily, and the Bay Area Rapid Transit (BART) system carried 126 million passengers a year. 

It is not just tech workers who are mobile. Firefighters, teachers, and service workers often don’t live in the municipalities they serve. The Bay Area is so integrated that the main beneficiary of the COVID-19 San Francisco exodus has been not Austin or Miami but other cities in the Bay Area: IRS data shows that four of the top five counties San Franciscans moved to are in the Bay Area. 

The Bay Area was not always this way. In the past, its geography kept it highly segmented. Early on, railroads drove development in the East Bay, Gold Rush shipping in San Francisco, and farming in the Peninsula. It was not until after industries like denim and aerospace had emerged that infrastructure finally got built to connect the region. Highways funded by the Federal Aid Highway Act of 1956 connected cities all over the Bay Area, turning once-burdensome trips into drives no longer than an hour. A commuter rail system was planned from 1957 onward, and BART began operations in 1972, followed by Caltrain in 1985. 

The results benefited businesses throughout the region and greatly contributed to the Bay Area’s wealth. Thanks to a high degree of integration, a modern startup in Berkeley can easily travel to Menlo Park to raise venture capital, recruit engineers from San Mateo, San Francisco, and San Jose, and run its manufacturing facility in Fremont. The Bay Area as a whole thus became extremely attractive to young engineers, who moved en masse in search of their fortune.

However, all this also created the conditions for a much wider influx of people. Today, the Bay Area leads the nation in regions where workers have to “supercommute” at least 3 hours a day. It is not uncommon to have an Uber driver who lives in Sacramento, and he’ll have to brave the fifth-worst traffic in the world. Meanwhile, out-of-town arrivals from across the country fuel homelessness, leading to a situation so poor that one UN envoy compared San Francisco to India.

The root cause of this dysfunction is that while the economy is regional, the politics are local. The Bay Area’s 101 municipalities are home to as many mayors and councils. The entire metro area measures 6,900 square miles and is home to 7.8 million people, a little less than New York. The Bay Area as a whole, however, has no single government responsible for the whole region.

There are pan-regional entities, like the Bay Area Rapid Transit Authority and a voluntary planning organization with no enforcement power called the Association of Bay Area Governments. Certain independent bodies have borders that overlap with municipalities in non-obvious ways—for example, over 100 independent school districts—but these entities have little authority over local governments and vice versa. Large agencies operate with no oversight to become slush funds that double as springboards to higher local politics. Even criminals know this, running crime rings across different cities in the Bay to “hop” jurisdictions. Even if they are working together, 50 different police forces are easier to evade than a single, unified department.

The lack of any central agencies makes it impossible to coordinate, and on many matters, there is no incentive to coordinate at all. This state of affairs also eliminates all stakes because everyone knows that no project will be completed during any particular official’s term, turning every plan into an opportunity for grandstanding and graft. In 2020, San Francisco Supervisor Aaron Peskin couldn’t support Measure RR, a small tax to save Caltrain during the COVID-19 pandemic, without extracting a concession. The result is a Venn diagram of governance where no one knows who is accountable to anyone.

Economic integration thus masks extreme regional variation. Monthly average rents vary wildly but are all expensive; Vallejo has the low end at $1,759, while Menlo Park carries the high at $4,395. This increases turnover and makes it difficult to attract talent, especially service workers. Instead of coordinating on housing the homeless and creating economies of scale, each city is responsible for its own homeless population with wildly different shelter costs. Thanks to different tax policies, the same company’s tax bill may be as much as 1300 times higher—that is not a typo—based on a 40-mile drive.

This results in two obvious problems. The first is a lack of economies of scale. The research is relatively clear that cities have economies of scale, especially with capital projects like transit and sanitation. But without coordination, millions of dollars are wasted on unnecessary government salaries and corruption. With so many different systems, each municipality and trans-Bay entity must negotiate separately, which is one reason BART is spending nearly $130,000 per new fare gate.

The second problem is that the large number of entities increases the number of veto points, a term for political operators that can block a decision. As the number of veto points increases, the chances the game works go down. Holdouts can bring the probability of project completion to zero, creating what Francis Fukuyama calls a “vetocracy.” In a coordination game where everyone defects, things don’t get done. For example, many cities opt to build commercial rather than residential properties because the former is heavily taxed, and the expectation is that someone else will build the latter. The problem is that there is no coordination, so everyone just suffers high rents and long commutes.

The Lessons of New York

San Francisco is still a place people love and want to live in, and there are people who dream of something grander for the city. One widely-discussed new vision, from Y Combinator’s Garry Tan, is called “San Fransokyo,” a name borrowed from the movie Big Hero 6. This concept draws both on an idealistic, hyper-urban vision of San Francisco and on practical lessons from some of Tokyo’s successful policy choices, especially reliable mass transit and mass high rises. It is certainly more ambitious than the San Francisco Chronicle’s idea that there is something to be emulated from, say, Pittsburgh’s experience. 

Ideal as it may be, there is something standing in the way of San Fransokyo: the reality that San Francisco and Tokyo operate at very different scales. San Francisco has a city population of about 800,000 people and 49 square miles; Tokyo has about 14 million people and 841 square miles. San Francisco, on its own, just isn’t big enough. But another city—one much closer to home—has more immediate and practicable lessons for what strategy can overcome the Bay Area’s crippling localism. That city is New York. 

What we now know as New York was once a region divided among five different counties and numerous municipalities. It became a single city through a process called consolidation, and that process was the vision of one man: Andrew Haswell Green.

Green became involved in civic life early in his career under the tutelage of his mentor, Tammany Hall antagonist Samuel Tilden. He cut his teeth on the Central Park Commission, where he created a staggering number of notable institutions that still endure today, like Central Park and the Metropolitan Museum of Art. There, he also saw firsthand the problems stemming from the lack of consolidation. Over the years, the region became more tightly economically integrated, especially Manhattan and Brooklyn. Rich Manhattanites would move to Brooklyn to avoid high housing prices, and Brooklynites, proud of their independent culture and industry, would sell their goods to Manhattan while avoiding the excessive regulation of smaller municipalities like Queens.

Green observed that this caused a great disparity between the tax base and the beneficiaries of the city’s services, creating a mismatch that “produced a character of citizen who is a drudge in his field of work and a dummy in his sphere of citizenship.” This caused a breakdown where the cities were unable to coordinate even on mutually beneficial endeavors, like infrastructure. The Brooklyn Bridge’s previous success had residents across the region clamoring for more such infrastructure, but city governments instead fought over who would claim the benefits. The equivalent of Peskin’s recalcitrance over Measure RR in 2020 was the norm of the New York government in this period. But Green called the whole situation a “sham.” To him, New York was already a unified city, just not in government.

There was recognition that a lack of consolidation was hurting the ability of the region to provide public goods, especially infrastructure. As early as 1857, the New York state legislature passed a resolution recognizing the need to create such a city. Green described New York’s governance as a “travesty upon government,” explaining that “there is not, in the world over, another like area so disturbed by multiplicity of conflicting authorities.” The New York Times, in endorsing the referendum, dubbed the burgeoning metropolis a “commonwealth of divided municipalities.”

Green argued that shared resources without shared governance led to a tragedy of the commons, as the commons were depleted by politicians “relegating their custody to the irresponsible charge of all without permitting them to the special concern of any.” Green decided to make the consolidation of New York City the cause of his life, advocating on its behalf for nearly 30 years and ultimately chairing the Greater New York Committee. By 1894, when consolidation was put up for a vote, New York was deeply integrated but straining more than ever from the lack of a single governing authority. 

New York’s consolidation ultimately passed through a combination of carrot and stick, excitement and fear. Boosters advocated for consolidation to create, at last, an American city with the scale and power of London or Paris. On a more practical level, advocates for consolidation focused on infrastructure. Even Brooklyn, the only county with meaningful opposition to the goal, implicitly conceded the point through its actions. It had previously merged its fire department with Manhattan, and the Brooklyn Daily Eagle, the main opponent of consolidation of the city as a whole, self-servingly urged the consolidation of the water supply, which was Brooklyn’s greatest resource need at the time. And besides, Brooklyn itself had annexed all nine towns in Kings County. 

At the same time, New Yorkers feared being overtaken by Chicago, which was rapidly growing. The pivotal event in the public mind was when Chicago beat out New York to host the 1893 World’s Fair. Everyone remembered how New York had sapped prestige from once-mighty Philadelphia when it eclipsed the City of Brotherly Love in size in 1835. Increasingly, New Yorkers viewed consolidation as the only defense against the same thing happening to them. 

The success of the push to consolidate New York shows the power and necessity of a positive, definite vision in large projects. The consolidation of New York only occurred because Green offered a practical, specific argument. He believed the economies of scale from consolidation would result in cheaper mortgages, lower taxes, better jobs, and better infrastructure. The establishment of underground subways had been undermined by infighting for years and was one of the great promises of consolidation. Green painted vivid pictures of the harbors and bridges that could only be built when his goals were met. 

Visions like that are rare today because they require thinking about what one actually wants and how to accomplish it. Picking priorities in a world of finite resources means picking winners and losers and making enemies in the process. Negative critiques of a problem are easy because most people can agree that an obvious problem is bad. Almost no one in New York in the 1890s or San Francisco today disagrees about problems—it’s the solutions that divide people and cause established factions to block reform.

In politics, it is quite common for visions to end up as nothing more than broken promises. Consolidation was one of those rare causes that over-delivered and started doing so immediately after its achievement. It was no small project—voters approved consolidation in 1894, and it was completed four years later in 1898. That year, unified New York became the second-largest city in the world by population. 

Green had been strongly motivated by the idea that a single mayor rather than five would be able to break logjams and complete public projects. After consolidation, New York had three mayors in rapid succession; each of them kept a major promise of the cause. Its first mayor, Robert Van Wyck, pushed the creation of the first subway through in his first year and provided funds for the proposed Brooklyn Tunnel. His successor, Seth Low—who had been the mayor of recalcitrant Brooklyn—fulfilled the promise of using increased scale to lower taxes and fight graft in various city agencies, especially the police. The third mayor, George McClellan, Jr., opened the first full subway line, as well as Grand Central Station, the New York Public Library, and other iconic New York institutions.

The strength of a united New York City also empowered mayors to resist the power of Tammany Hall, a machine of political corruption that had once seemed hegemonic. Mayors like William Jay Gaynor, who was hand-picked by Tammany Boss Charles Murphy, now had the strength to renege on their deals and support civil servants who protected projects like the expansion of the subway. Green himself, who once struggled against corruption to build the Washington Bridge across the Harlem River, created institutions like the Bronx Zoo for the benefit of all the New York boroughs.

Today, New York is still a center of culture, finance, and power. It is even the home of the United Nations. Green’s vision of an “Imperial City” has survived for nearly 130 years.

The Vision for a Greater San Francisco

San Francisco finds itself in the same situation today as New York did over a century ago. The Bay Area is straining under the tension of cultural, economic, and political realities that are artificially separated by multiple jurisdictions. The only sustainable resolution is to collapse those realities into one.

In the Bay Area, the idea of consolidation is not entirely new—the first attempted consolidation occurred when the city was ready to rebuild in the wake of the 1906 fire. San Franciscans had noticed the growth of Oakland, which by the end of the nineteenth century had absorbed eleven municipalities, including Temescal and Claremont, and had tried to annex Berkeley and Emeryville. A number of prominent locals decided to try to follow in New York’s footsteps. Together, they formed the Greater San Francisco Association. But this initiative offered a much smaller basket of goods than Green had: a bridge across the Bay to replace slow ferries, lower taxes, and, in a Californian twist, a municipal water district that would be supplied by the Hetch Hetchy Valley in Yosemite.

The consolidation required a state constitutional amendment called the Wolfe Amendment. A San Francisco Chronicle op-ed in favor of consolidation noted that “people take pride in being in a big town. The once satisfied Brooklynite or Staten Islander likes to hail from New York.” However, the advocates of consolidation in the early 1900s failed to make the case that a Greater San Francisco would be as great as New York. The Oakland Chamber of Commerce led the charge against consolidation, and it won. Only San Francisco, San Mateo, and Marin County voted in favor of the 1912 amendment.

Over the following decades, the Bay Area watched Los Angeles grow from a secondary rival to a metropolis that soon eclipsed it in population and cultural prestige. New York, meanwhile, became a global powerhouse of wealth and influence that San Francisco simply could never match. 

But at the time of the vote, none of this was obvious or predictable. The fact is that consolidation was not compelling in the Bay Area of 1912. Economically and politically, the conditions simply did not exist. Consolidation would have been a major effort for its own sake rather than the resolution of a long-festering problem.

The Bay Area was becoming connected but was not yet irretrievably linked; an Oakland Tribune op-ed arguing against consolidation pointed out that a mere six percent of Oakland residents earned their living in San Francisco. There was hardly any infrastructure connecting the wider Bay economy. None of the major bridges had yet been built; Oakland and San Francisco were primarily connected through ferries. Outside of the two major metropolises, the Bay Area’s economy was largely agricultural, not industrial. There were no economies of scale to be had.

Moreover, there were no compelling projects to unite anyone behind other than the Hetch Hetchy Reservoir. As historian Mel Scott observed, “the movement for a metropolitan water district thus became identified with the political consolidation movement.” Many viewed it as a simple cash grab to avoid issuing bonds for a project that didn’t need consolidation and they were largely right: Hetch Hetchy provides water to San Francisco to this day, no consolidation required.

Most importantly of all, there was no local Andrew Haswell Green. A Tribune op-ed accused the Greater San Francisco Association of being formed in 1907 as an insidious plot well in advance of consolidation. Ironically, the problem is that it was formed too late. Green had pushed for his Imperial City for decades and been a forceful advocate with a vision of what he would accomplish.

The Bay Area of 1912 is long gone, and in its stead are all the conditions that make consolidation necessary. The Bay Area is no longer defined by a battle between two mid-size cities surrounded by rural farmland and a few wealthy vacation homes. Today, it is straining under the weight of its internal divisions. It is time to give consolidation a try again. Unlike in 1912, there is no need for a Wolfe amendment. Article XI of the California Constitution has permitted consolidation since 1988. The legal tools to create Greater San Francisco are readily available.

The Bay Area’s nine counties would form the equivalent of “boroughs” in the new city. The most important element of a Greater San Francisco would be the creation of a single city administration. Instead of the highest regional government being BART, the new city could have a real government with a single mayor. There would be representation from each region, but this should also be an opportunity to avoid the mistakes of the past. Moving from 101 municipalities to one would also come with enormous benefits at the government level. Many Bay Area cities today are hamstrung by weak mayors who are at the mercy of various city legislators and special interests; consolidation can be used as an opportunity to wipe the slate clean.

The new Greater San Francisco would also be able to engage in rational city planning at the right scale. Right now, every city has to have a housing element that tries to impose a housing plan, even in extremely small towns. A bigger city can rationalize this planning process by having a greater diversity of townships—keeping some areas rural by absorbing them into a larger city with a single housing element while placing homes in places that are meant to be dense.

The removal of redundant veto players will also allow for big projects to be completed on reasonable timelines. In the 1960s, San Mateo was able to kill the southern BART expansion, severely kneecapping its ability to be a true pan-Bay transit system. This was only possible because San Mateo had the right to control its borders and public purse. In a Greater San Francisco, there would be no more rogue towns killing grand projects. 

This new city would be similar in size and population to New York City, with an economy larger than that of most countries. Rather than being unwieldy, it would provide for the simplicity of a single, coordinated response with economies of scale. It would be a big economy with manufacturing, like Tesla’s Fremont facility and the Santa Clara semiconductor complex, the South San Francisco biotech cluster, and even See’s Candies main facility, resulting in a more balanced overall economy. The police would be able to coordinate and break up crime rings as part of one large, single department instead of 50 departments, with authority that would reach across the entire Bay. A company looking to manufacture will know the zoning rules without worrying that things will change if they move ten miles away.

A single Greater San Francisco government for the whole Bay Area could also achieve dramatic economies of scale. One large sanitation department can get a much better deal on equipment than many small ones can. Currently, there are over two dozen transit agencies throughout the Bay Area, and construction is horribly delayed and expensive; the Central Subway, a mere 1.7-mile line, cost $1.6 billion and took 11 years to complete.

Combining into a single bureaucracy can allow for the streamlining of thousands of unnecessary government officials, saving not only government salaries but, more importantly, saving decades—the time wasted by piecemeal politicians who have the incentive to block projects in the current vetocracy hellscape. The surplus politicians and bureaucrats are likely the biggest barrier to this new vision, not the legal restrictions.

Part of the reason that New York politically consolidated despite such barriers is that it had been going in that direction for decades at an institutional level: Manhattan had already consolidated the Bronx, Brooklyn had already merged some of its departments with Manhattan’s, and there was already a desire to merge the region’s harbors. The Bay Area already has some regionally governed transit systems and bridges, and lower-level institutional consolidation is a more readily achievable goal.

One obvious place to start is the police. Bay Area residents are much more afraid of crime than they were just a few years ago, and crime rings are known to cross the entire Bay Area. In particular, a shared desire for more police coordination with established cities like Redwood City and San Jose could follow such a pathway. Merging police departments and giving them a mandate to crack down on criminal organizations operating across the Bay Area would be a decisive and visible signal of what full political consolidation could achieve. All the different fire departments and transit agencies that currently don’t work well together could all be brought under one roof, as Bay Area citizens see the power of consolidated government and wish for more.

This step can, and likely should, occur before political consolidation. In the case of New York, prior consolidations made the process easier and more obvious, contributing to a sense of inevitability and removing some of the opponents. Working together would show the benefits of consolidation to the entire region in its area of greatest concern and create a powerful constituency in its favor. It may also make sense for some of the smaller towns to combine as an intermediate step.

Consolidation also makes infrastructure development possible on a scale that its opponents could never counter-promise or swoop on delivering. The most dramatic of these is a seamless transit system, which was one of the major carrots of New York’s consolidation as well. Bay Area residents have long been plagued by broken transit and terrible commutes, and polls find that 83 percent of voters would support merging Caltrain and BART. It is no wonder that at a 2013 conference on consolidation, the participants concluded that transit integration would be a good place to begin.

With some of the worst traffic in the world and the highest percentage of super-commuters, voters would likely view a world-class transit system with higher speeds and frequency as highly motivating. Some nonprofits like Seamless Bay Area have put out a Vision Map of what an integrated system could look like; they are a great start but can stand to be even more ambitious. With consolidation, such a system could arguably be built faster and cheaper per mile than without such a move. A 2011 study found that Bay Area transit agencies had lower operating margins than larger, more centralized peers. This is the power of economies of scale and the power to just make heads roll rather than shift blame.

Within the current political structures of the Bay Area, there would be winners and losers from consolidation. Other Bay Area reform movements, from YIMBYs to the tech billionaires building a city in Solano County, have generally tried to focus on mobilizing the professional classes. The reason it’s easier to try and build a whole new city rather than flip even a few local governments is because there has been no deal available toward which to turn entrenched political interests.

What matters is not money but the ability to build these alliances. Green advocated for and achieved New York consolidation over 30 years without a great deal of wealth to his name. What he possessed was extensive personal experience in the institutions of New York. Green did not just advocate but actually sat on commissions to implement projects that advanced his vision step by step. That willingness to engage with the actual work of sitting on boards is often what’s missing from today’s aspiring Andrew Haswell Greens. He knew the ropes, the names, and the possible alliances. And in the end, he succeeded.

The battle for Greater San Francisco will be long and bruising, and the involvement of several charter cities means that the state legislature will need to ultimately back the effort. But it would not be starting from ground zero as it was a century ago. What matters is a clear vision of what is at stake and a charismatic champion to advance it. Consolidation saved New York from stagnation: without it, Manhattan and Brooklyn would have remained mid-sized cities. Rather than eclipsing Chicago and London, they would have been the Eastern Minneapolis-St. Paul.

As things stand, the scattered cities of the Bay Area will continue to buckle against the contradictions that emerge when there is no scale. Greater San Francisco can embrace the powerhouse the Bay Area has become and make a golden future possible.

But with the global transition to electric vehicles underway, there is now a threat looming over the future of the German automotive industry. Europe’s largest economy is more dependent on gas-powered cars than is commonly realized, and because of how its economy is structured, transitioning to electric vehicles will be more difficult than most people imagine.

The transition to electric cars is going to completely upend Germany’s economy—and probably not for the better.

Germany’s Automotive Reliance 

On paper, the German auto industry represents approximately 10 percent of the country’s GDP. But even that figure is an underestimate of its economic centrality. A Germany Trade and Investment report found that the auto industry generates nearly 24 percent of all German domestic revenue and 35 percent of its R&D. Automotive employment is usually described as employing 800,000 people, but that is merely the number directly employed in motor vehicle manufacturing—another 1.8 million jobs are indirectly tied to the automotive industry, with some estimates even higher. Volkswagen, Germany’s largest car company, is also Germany’s largest employer overall. Most notably, Germany’s automotive employment estimates have remained stable even as manufacturing employment has fallen by some 10 percent since 1991, meaning that cars are a higher share of manufacturing employment than twenty years ago.

The German economic system is efficient partially because it is based on a highly tracked society that starts sorting workers at the earliest possible ages. As early as 10 years old, children are put onto tracks that determine the rest of their lives: whether they go to university or learn a trade. From there, Germans that are not on the university track are predominantly taught the skills needed to fit into the industrial complex, rather than a more comprehensive education.

If they go to university, they will likely earn a functional degree designed to meet the needs of a Volkswagen or a Merck. If they don’t, they will probably work at a Mittelstand, a small or mid-sized business that accounts for 52 percent of German GDP, 70 percent of employees, and 82 percent of trainees, in contrast with the United States, where only 46.4 percent of all employees are employed by small businesses. Mittelstands are often family-owned; on average, the owner is 57 years old and the company is 70 years old. But with this stability comes brittleness—Germany’s long-term unemployment rate is well above the OECD average.

The German automotive sector relies heavily on the Mittelstands for industrial power components, specialized parts, and chemicals, and they contribute up to 70 percent of the value-add for a manufactured vehicle. Without this automotive demand, many specialized Mittelstands would go out of business, causing significant knock-on effects. German engineering is well-regarded not just because of its quality but also because of its value—the customer pays a reasonable price for an excellent product. Without the subsidy of automotive demand, German products in other industries that share manufacturing supply chain nodes would have to command premium prices to make similar profits, putting them in competition with higher-end goods.

Cars are Germany’s most differentiated products: they leverage the unique strengths of German innovation to create products with an outsized surplus value. Because of this, auto industry patents make up 47.5 percent of Germany’s patent applications and Germany exports 75 percent of the cars it produces. That makes cars the greatest share of German exports at a whopping 15.3 percent, which is 1.5 times more than the next largest category: chemicals.

Germany’s economy is structured around an assumption of high exports, especially in the automotive industry. Germany has benefitted from the cheap euro of the 2010s because its export-driven economy allows it to maintain a significant current account surplus. One paper found that the top driver of this surplus was a high demand for capital goods—cars. Without the auto industry, it would be impossible to maintain it, and by extension fund Germany’s welfare programs.

Different Manufacturing

Germany’s reputation for automotive engineering prowess is so potent that it is widely assumed that German engineering can be applied to any domain. German companies cultivate this assumption with promises like BMW’s commitment to spend $34 billion dollars on “future-oriented” technologies.

But this is not always the case—manufacturing is not a generic input into a production function. Building battery electric vehicles (BEVs) is vastly different from building internal combustion engine (ICE) vehicles. The reason is that BEVs and ICE cars are built around highly distinct—and incompatible—propulsion systems.

Because of the vast number of specialized parts, ICE cars depend on supply chains that are more specialized than they first appear. Modern ICE engines are very complex, with specialized mechanical subsystems for everything from compressing air, combusting fuel, and managing exhaust to balancing the vibration of the engine and changing gear ratios. As such, a combustion engine has 30 different types of main parts, from engine valves and intake manifolds to piston rings and spark plugs. Each of these has its own requirements like specific lubricants and operating fluids, each with their own supply chains.

By comparison, BEVs are much simpler. Their massive battery packs are heavier than an ICE drivetrain, but they drive their high-efficiency motors directly, and the axles are directly driven by the motors with very simple, usually single-speed transmissions. As a result, an electric drive train may have just 17 moving parts compared to more than 200 for an ICE, which means that BEVs require very little maintenance. BEVs are therefore 30 percent less costly to maintain than ICE cars.

This complexity difference is related to the large efficiency difference. Besides the inherent thermodynamic inefficiency of converting combustion into motion, the mechanical nature of an ICE powertrain causes significant energy to be lost to friction and heat, and puts tight constraints on the thermodynamic design. As a result, ignoring heat loss, only 40 percent of the energy converts into kinetic energy compared to 85 percent or more for BEVs. The most recent data from the U.S. Bureau of Transport Statistics say that the average light-duty passenger vehicle has an average fuel efficiency of 39 MPG, in comparison to the EPA’s 2021 rating of 70 MPGe (an equivalent for electric vehicles) for the Porsche Taycan S, which has the lowest energy efficiency of any electric vehicle. The very high efficiency and cleanliness of converting electricity to motion in BEV powertrains means much simpler cooling and lubrication systems.

However, the reduced complexity of BEV propulsion systems is deceptive. Making an electric car is anything but simple.

If BEV manufacturing was simple, all the auto companies would be catching up with Tesla in a few years. But the difficulty of BEV manufacturing comes from different needs in software and supply chains. Battery packs are not one giant battery, but many batteries daisy-chained together. This requires complex software to prevent temperature problems, balance discharge, and optimize efficiency. In a sense, BEVs are as much software as they are cars, but software is usually considered one of Germany’s weaker areas. Indeed, Volkswagen’s BEV program was delayed for years by faulty software and still launched with hundreds of bugs.

The batteries themselves introduce their own supply chain complexities for car manufacturers. Batteries are a specialized branch of chemical engineering; Bayer can’t just transition its skills in, say, fertilizer to lithium-ion. New BEV factories now commonly involve a co-located joint venture with a battery company for this reason, also helping to reduce batteries’ high shipping costs. 

Even the basic car platforms are different. BEV chassis are designed around placing the extremely large and heavy battery on the bottom of the car, under its floor. This change in basic form factor will obsolete the billions of dollars invested in current vehicle platforms with different design assumptions. The Economist reports that Volkswagen will have to create an entirely new chassis platform, and it will not fully roll out until 2030. That means that much of Germany’s nearly century-old investments in expertise and refinement must be replaced, reducing Germany’s otherwise huge first-mover advantage.

All this together upends decades of German factory floor design and supply chain relationships. It creates a new manufacturing line where the core competency is diluted away from the original equipment manufacturers (OEMs) to the battery partner. After all, the largest battery facilities in Germany are not German—they are American and East Asian. If the German automotive industry will try to naturalize battery production, they will have a lot of catching up to do.

The performance of German BEVs underlines how behind their industry is. Consider the example of “core efficiency,” a BEV metric that adjusts range to take into account weight class and battery size, isolating efficiency improvements due to engineering. Setting aside Tesla, which is in a league of its own when it comes to core efficiency, the best-performing manufacturers are American and East Asian; German manufacturers are completely at the bottom. BMW and Porsche are among the absolute worst in their weight class. Mercedes does better than the other German manufacturers but is still less efficient than Ford or Hyundai. It is staggering how absent German technology is in this mix.

Comparing Tesla to Mercedes is particularly instructive. Take the Model S, Tesla’s flagship. The Model S was Tesla’s first mass-production car, first released in 2012; today, it retails for $99,990 dollars. The longest-range model was released in 2017 and continues to undergo improvements. The EPA rated the Model S at 402 miles in range years ago, which is comparable to the average ICE vehicle range of 412 miles. The Mercedes EQS, the top Mercedes BEV, costs $102,310 and has had the advantage of learning from years of Tesla’s mistakes. The EQS is rated for 350 miles. In other words, Mercedes found that their expertise in building ICE vehicles doesn’t transfer over to building an electric car as good as a Tesla, or even a Hyundai.

Batteries tell a similar story. Germany excels in chemical engineering and is home to companies like Bayer, BASF, and Henkel. But battery manufacturing is quite complex, and non-German companies specializing in their development have run circles around their German competitors. Tesla co-develops batteries with Panasonic. Mercedes must buy its batteries from CATL, the largest Chinese battery manufacturer.

More importantly, none of these batteries rely primarily on German technology. Even if some are being made in Germany through joint ventures, they are more like outposts for American and Asian companies to access the European market. The core intellectual property is not German, and the value will mostly accrue outside Germany.

Mark Fields, the former CEO of Ford and now the CEO of Hertz, explained his company’s mass purchase of Tesla cars: “Tesla is the only manufacturer that can produce EVs at scale.” Ford, Fields’ old employer, seems to be proving him wrong with their Lightning line of electric trucks. But Ford is also American, and it’s telling that German firms aren’t even part of the conversation. 

Why Germany Is Vulnerable

The source of German value is the many specialized Mittelstand suppliers that manufacture the powertrain which currently makes up around half of an ICE vehicle’s value. But the consulting firm PwC estimates that this is where the core change in the BEV value chain will be seen. The powertrain will shrink from about half of the car’s value for ICE vehicles to between 34 and 40 percent for BEVs, and will have completely different components. Instead of complicated engines, the bulk of the new powertrain value will come from the batteries themselves, implying that the differentiating value of German cars is at risk.

Mercedes and BMW, as companies with distinctive products beloved by millions and brands worth billions, will not disappear. But they will fall behind and shrink. And the automotive industry isn’t just one part of the vibrant tapestry of German engineering—beyond just the Mittelstands’ relationship with the automotive industry, car manufacturing has an outsized impact on Germany’s whole economy. Its competitive difficulties will strike at the very heart of the German economic model, and reforming it could entail sweeping changes everywhere from welfare to the educational system.

But the current economic model depends on a sector-wide consensus between the government, industry, and labor unions that leads to full employment and a dependence on export strength. This also means that industries coordinate to secure their interests, regardless of what that means for Germany as a whole, like when German employers and unions recently worked together to oppose Russian gas embargoes. Jörg Hofmann, the head of Germany’s largest union IndustriALL Global Union, or IG Metall, threatened a “ruckus” if auto companies pursue “destructive company strategies” that include economically necessary changes, like factory closures.

The power of Germany’s historic industries is not to be underestimated. Even in the face of climate change and Germany’s Paris Accord obligations, the country decided to shut down its remaining nuclear plants and replace them with coal, one of Germany’s oldest industries. ICE-dependent companies are even more powerful and may delay Germany’s BEV investment pace just as other countries are speeding up—all to protect the present at the cost of the future.

Employees will also bear the brunt of this energy transition, in part because of how highly regimented the German workforce is. Someone whose entire education led them to a specialized career will have difficulty retooling for another part of the supply chain, let alone a different career. Germany’s engineering system is considered the best by many Europeans, but when those engineering skills are in less valuable fields, the lack of flexibility becomes a weakness and not a strength. A report by the Boston Consulting Group and Agora Verkehrswende forecasted that in 2030 Germany would have approximately the same number of jobs as today even though they also predict a nearly 50 percent turnover, with half those jobs classified as having “partly high training requirements.” It is difficult to square those two predictions. 

It would be difficult for any economy to absorb such drastic change, let alone one as structured as Germany’s. The government-backed German Economic Institute reported that the government’s Kurzarbeit furlough scheme, which pays 60 percent of a worker’s lost wages if an employer reduces their hours instead of laying them off, was not enough in the face of COVID-19; if the German auto industry shrinks and causes many Mittelstands to fail, the cost of hundreds of thousands of Germans on the dole will be challenging to absorb. 

Where the Future Lies

The fundamental issue is that the BEV revolution strikes at the heart of Germany’s differentiating value. This is not the end of German OEMs or evidence that Germany has no engineering or value advantages. But in the near future, with simpler electric vehicles, it isn’t unreasonable to imagine that high-end car companies will focus more on marketing if their engineering cannot catch up. It’s been done before—Ferrari has a higher market capitalization than Fiat Chrysler, which spun it off. But nearly half of Ferrari’s revenue is already high-margin merchandise.

The impending transition to electric platforms does not mean that German companies will be unable to make cars, but it minimizes a core source of differentiation that allows the German industry to maintain pricing power and sales dominance, which means less prosperity to spread around the rest of the economy. German automotive sales growth in the 2010s was driven by China, which plans for EVs to be 40 percent of its market by 2030. In 2022, Volkswagen lost the top spot in Chinese sales to BYD, a homegrown brand, for the first time. In electric vehicles other than Tesla, the top ten brands in China are all local. If this trend continues as China approaches its EV goals, that portends poorly for Germany’s export model—they just won’t be able to keep up.

This threat to the German export model could also change the country’s relationship with the European Union. Germany depends on exports to drive its current account surplus, which itself relies on a weak Euro. Some economists have considered this a threat to the Eurozone economy as a whole, but Germany’s strength has nonetheless allowed it to exercise outsized control over the European economy, like during the Greek debt crisis. The rest of Europe would likely relish the opportunity to lessen German economic influence to weaken the Euro and allow more lenient debt rules, accelerating Germany’s decline. This would put Germany in greater competition with France for pan-European influence. If it proves impossible to shore up its competitive export machine, Germany may choose to engage in another sphere of competition. The most obvious one is military affairs, where Germany is already starting to increase its spending as a result of the war in Ukraine. Conveniently, military spending also functions as a domestic subsidy for its current industrial base.

But perhaps the most important risk of the BEV revolution is to the German psyche. Indeed, after the First World War, historian A.J.P. Taylor observed that “the most serious blow inflicted by the war economically was to men’s minds, not to their productive powers.” German identity is built around their remarkable work ethic and their long devotion to their historic industries, and modern Germany’s sense of pride is built on applying those virtues to automotive exports, not just its economy.

Just as losing their historic coal and iron industries in the 1920s led Germans to doubt their governance model, so too may the decline of the car industry throughout the 2030s lead Germans to question their economic model like the U.K. in the 1970s. Then, the British experienced an economic slowdown and hollowing out of historic industries like shipbuilding, resulting in waves of union strikes, civic unrest, and ultimately the Thatcherism of the 1980s. It isn’t unthinkable that a similar economic situation could lead to similar disturbances and overhauls. Whatever the outcome of this transitional period may be, the German economy of the future will be vastly different than that of Germany today.