San Francisco and Silicon Valley pride themselves on being on the cutting edge of new trends—the area’s famed ‘move fast and break things’ ethos extends to everyday conversations from barstools to brunch tables. Many San Franciscans are indeed proud of their part in ushering in the future. But in an ironic way, this yearning for the future is an old trait for the city. Henry George, a 19th century American economist, described the excitement San Franciscans felt as they anticipated the completion of the transcontinental railroad in the way modern audiences might anticipate a major AI announcement from Google or Apple:

I remember, after having come down from the country, sitting on Christmas eve in the gallery of the old American Theatre, among the gods, when a new drop-curtain fell, and we all sprang to our feet, for on that curtain was painted what was then a dream of the far future, the overland train coming into San Francisco.

The completion of the railroad was poised to connect the city to the rest of the country, promising to stimulate even greater development. Many believed correctly that it would transform San Francisco from a gold rush outpost to one of the top metropolises of the country. However, even in the midst of this excitement, George foresaw the limits on who would benefit:

[A]nd after we had shouted ourselves hoarse, I began to think what good is it going to be to men like me? those who have nothing but their labour? I saw that thought grow and grow; we were all—all of us, rich and poor—hoping for the development of California, proud of her future greatness, looking forward to the time when San Francisco was to be one of the great capitals of the world; looking forward to the time when this great empire of the West was to count her population by millions, and underneath it all came to me…What about the masses of the people?

What about them? They are members of the community, and their work and social value is what makes the value of the city. But they don’t get a larger slice of the pie when it grows; they don’t own any share of the common venture of the city. In fact, they have to pay rent for their share, so when the city grows, any gain in quality of life or economic position comes right out of their paycheck and into the landlord’s pocket. Even a small business like a restaurant sees much of their gain slurped up into competition for space. Most of the gain goes to passive landowners, who didn’t even have to do any work for it. There’s an incentive problem here, and a misallocation of everyone’s resources.

The growth of the digital economy centered in San Francisco is an event to match the scale of the arrival of the railroad. In the world of technology, the city is one of the great capitals of the world; the population just of the Bay is in the millions, and the Bay’s hinterland reaches to tens of millions. But activists, analysts, and residents are still asking George’s question: what about the masses of people? While median incomes may be higher in San Francisco than elsewhere in the country, housing costs eat up much of the difference—while household income is 65% higher than the median in the U.S., rents are 78% higher. The city and the counties around it are running up against an old problem, one which has not been alleviated by the dramatic economic changes—that of land, and the hard limits on its availability within traveling distance. This scarcity has expressed itself as astronomical rents and a seemingly intractable homelessness problem, even in the midst of some of the fastest economic growth in the world. And as much as the city prides itself on bringing about the future, the truth is that it’s been wrestling with this same problem for over a hundred years.

George himself predicted that increased economic prosperity alone would never fix San Francisco’s socioeconomic ills. In 1879, when the city was still much less developed than the Eastern seaboard, he wrote: “When San Francisco reaches the point where New York now is, who can doubt that there will also be ragged and barefooted children on her streets?” The Bay Area today bears out this prediction. The only missing part is the children, ragged or otherwise. Given the persistence of the problem, and its consistent nature, it is worth examining the explanation George offered, and the solution he proposed.

This brings us to the more immediately practical element of governance: policy.

Policy discussion is usually terrible for a simple reason. For the average person, policy discussions represent bloodless statistics, disembodied proposals, and an infinite scroll of historical trends and qualitative analysis. In other words, they seem disconnected from the actual experience of life. Daily work, families and houses, and the flowing currents of warm bodies that make up a city in which a person actually lives, seem to exist in a different universe.

This disparity did not exist for Henry George when he was living in San Francisco as it began its ascent. His discussions of land, tax, and resources were inseparable from a vision of cities as engines of wealth, shared in by the same people who created it. By the very dint of its focus on land, George’s proposals demand the perspective of building up a city, or even a country. It restores policy as real decision-making, which animates the life of families, workers, communities, and cities. As San Francisco and many other cities endure the barriers foreseen by George, the same discussion offers the chance to re-center the life of the city as the object of policy.

George began his work by identifying a flaw in the dominant economic models—classical and Marxist—and by extension, in the decisions of those governing the development taking place around him. Building on work by David Ricardo and John Stuart Mill, these paradigms treated land like any other form of capital. However, unlike factories or railroads or any other physical capital, no matter how high the rental price of land rises, no more can be brought to market. As Mark Twain put it, “Buy land, they aren’t making any more.”

Since land doesn’t respond to the normal incentives that encourage the creation of more capital or labor, it requires more careful stewardship than capital or labor to achieve efficient use. The results of the lack of this attention to land are visible everywhere, but are most critical where land is in the highest demand. Valuable land is used inefficiently or not at all, sunk costs and landowner resistance make it difficult to change that use even when the inefficiency has become obvious, and physical restrictions make it nearly impossible to bring new land to market to replace it. For a city to develop, land must be put to denser and more intensive use, or it will be stuck with sprawl, high rents, and limited growth. Achieving that development, given the constraints of limited supply, sometimes requires a more deliberate approach to growth.

This is exactly the struggle facing San Francisco today.

The best way to bring about this more careful stewardship within a market economy is to change the incentives embedded in our tax structure and the barriers in zoning and codes. San Francisco is a perfect place to start. All policy requires a subject, a determined goal-seeking actor for whom that policy is crafted, and by whom it is executed. Let’s imagine a hypothetical San Francisco city government, full of energy and confidence with a political mandate to truly govern, which was gunning for the city’s rejuvenation. Such a government could take strides toward its goal by transitioning from sales and property taxes to land value taxes. Under a land value tax, the city essentially charges a community fee for the exclusive occupation of a plot of land. A land value tax differs from the familiar property taxes in that it disregards the improved market value of the land though buildings and such, and instead assesses the tax burden as if it were an empty plot. The overall goal: to encourage more economic development of land, without penalizing that development by taxing the improved value.

“But,” the skeptical voices of San Francisco’s leadership would likely inquire, “what incentives have you changed? Show us the mechanism!”

For George himself, the justifications were philosophical, as well as economic. He adhered to a liberal tradition, which tied in property to the virtue of labor put into its creation. Since land is not created by human labor, property in land cannot be justified by those same arguments.

But whatever one’s views on this tradition, there is a strong economic logic to fall back on. There’s little question that individuals respond to incentives and attempt to avoid taxes. As such, taxes on windows reduce windows, taxes on cigarettes aim to reduce smoking, and so on. With the exception of taxes levied specifically to reduce the prevalence of a particular consumer choice (called Pigouvian taxes), this change in consumption in response to taxation is referred to as deadweight loss. Naturally, however, the quantity of land in a given jurisdiction, unlike the quantity of labor or improvements, cannot change, and thus deadweight loss is avoided.

Another result of this unchanging quantity of land, referred to in economics as inelasticity, is that the burden of taxation lies almost entirely with the owners of land, and not with renters. When a tax appears to raise the price of a good—as with cigarettes or gasoline—it is the result of less of the good being produced. After all, if producers could sell the same quantity of gasoline for a higher price, they of course would—but the ability of their competitors to undercut them limits the potential prices. The rise in price associated with a tax occurs as the tax levied increases effective production costs and creates a supply shock, reducing the overall quantity of the good supplied and raising prices to reflect that lower quantity.

With land, on the other hand, a tax cannot produce a supply shock; the supply of land is fixed. If the market for land will support a $25,000 yearly rent for a plot before the tax, it is because that’s the price where the quantity supplied and quantity demanded are equal. After the tax, the quantity supplied has not changed, and so the equilibrium point of supply and demand will remain at the same price. The price may even decline, as landlords opt to put more rental space on the market to cover the tax, increasing supply. In this way, a tax on land is quite unique, and functions differently than a tax on buildings, rentals, or any other asset that can be created or destroyed by normal economic processes. While the nominal value of the land may fluctuate and be impacted by taxes, the real quantity is static.

This logic in and of itself does not fully justify a policy. It only outlines a possible tool which can be wielded by those in the position to govern. The context in which someone wields it and the goals for which it is used—these elements, altogether—create the real policy.

But for a government with the goal, means, and authority to make San Francisco into the engine of innovation and wealth that it should be, the tool is incredibly powerful. By the power of incentive, the land value tax removes a number of barriers which currently undermine the development of beautiful, livable, and environmentally friendly communities.

Several shifted incentives would bring about this outcome. First of all, there is a transfer of wealth from detached low-density homeowners to higher density apartment and commercial landlords, as they now pay the same taxes, but the latter gets more rental income. This immediately produces an incentive to either build higher density buildings and find higher density uses for existing buildings, or sell to developers who will. Of course this only works if the costs and legal barriers to building are low enough to not be prohibitive. But those barriers, like building codes, zoning, NIMBY vetocracy, and rent control, would also have to be tackled for a complete developmentalist program. Our hypothetical developmentalist city government would have to deploy more tools than just the land value tax to find and break through every barrier to development. But the land value tax and its intended behavioral changes are a powerful foundation and guiding intention for those secondary policies to build on.

This real-world complexity of the behavioral structure of society is part of why armchair marginalist policymaking can’t actually produce desired outcomes on its own, and why policy discussion has that strange detached feeling to it: there are always many barriers to any particular change, some hidden. Real change has to be pursued by a persistent and powerful agency, not just one-off policy recalibrations. The intention and capacity to govern is a more important determinant of desirable outcomes than any given policy tool.

Second, the land value tax would represent a transfer of wealth from landowners to renters, through the increased density of housing supply that it would stimulate. Renters who now spend in many cases over 50% of their income on rent would suddenly have much more disposable income. When a single cost looms that large in people’s cost structure, even relatively small changes can have large effects. That disposable income would translate into wealthier, more livable communities, and higher economic viability for local commerce and services. Further, a land value tax would decrease the value of land as a passive speculative investment, which would have the effect of reallocating capital to more productive purposes, and partially mitigate the asset inflation problems that make home ownership inaccessible to younger buyers. Again, this serves the goal of a more flourishing and livable city.

Land is underused in San Francisco. To secure its future development, the city has to change the incentives surrounding land and capital use. For every development project, the land value tax incentivizes the investment of more capital in order to use less land over time. Space-saving construction—like underground instead of surface parking, extra stories on buildings, and the like—typically costs more up front. In the current tax code, where assessments are based on full improved value, these sorts of developments carry a heavier annual tax burden. In San Francisco, the last sale of the property is used to assess its value, plus a maximum yearly increase of 2%, and that value is taxed at 1.18%, regardless of which portion is land and which is improvements. However, the San Francisco County assessor estimates the total land stock in the city to be $110 billion, nearly half the total property tax role. Therefore, a 2.5% land tax could entirely replace the current property tax.

Shifting to a land-based tax system would encourage developers and remodelers to put more money into the initial construction of their buildings in order to get as much value as possible out of a smaller space. Tall buildings downtown and dense neighborhoods like Chinatown and the Mission District would see savings, while Sunset and other neighborhoods dominated by single family homes would see an increase in taxes. I did some calculations, using data available from the city, to estimate the changes. Of a few sample properties I looked at, the biggest savings were in a block in Chinatown, which would pay 43% less. In Clarendon heights, on the other hand, taxes would increase by 48%.

This would obviously be an immense shock to many neighborhoods, and the best approach would be to gradually phase in the changes and potentially offer deferrals for homeowners on fixed incomes who were hit harder by these taxes. As time passes, however, we can expect that neighborhoods would evolve to get more use out of each expensive square foot. If zoning allows, the single family residential neighborhoods would be nudged to evolve into the sort of mixed use neighborhoods common in the Mission District or Chinatown, two-to-five story buildings predominating and commercial and residential uses nearby one another, ultimately making more walkable, less automobile-oriented neighborhoods. And it is likely that zoning will follow—as landowners start to pay by the square foot, they can be expected to start lobbying for more freedom to make use of their limited space. Shifting the incentives of landowners is perhaps the most important outcome of a land value tax, as it would better align their interests with those of renters, since both would benefit from more capital investment and densely utilized land.

But you can’t talk about changing tax assessments without talking about how that would actually work. The current system is a mixture of legislated rates and mechanisms, and some discretionary assessment power by the office of the assessor. Changing that would be a difficult coordination and legislation problem under the current system. The ideal outcome would be again a mixture of transparent and easy-to-compute mechanisms and some discretionary judgment. The task of actually making that change would involve considerable study of the problem, including the details of what the city bureaus are actually capable of, and some institution building. As such, it is again not the domain of bloodless policy knob tuning from the sidelines, but of a competent, powerful, and determined government grappling with the problem in an active way.

The density that would result from this kind of policy isn’t an end in itself. Its first and most obvious major effect is increased housing supply, leading to some combination of lower rents and higher population, and in turn supporting more productive activity overall. It’s also a tool for re-imagining the city—its built environment and its flow of life. And there is an explicit vision of the city which predominates here. Density is the large-scale effect of a city organizing its environment to prioritize homes, transport, workplaces, and public green spaces. It prioritizes a greater number of people having access to the necessities of life and building families, rather than only that proportion able to purchase or rent single-family houses with an insurmountable price tag. An especially important goal is bringing workers and jobs closer together. If every individual has more jobs within a given commuting distance, they are more likely to find employment that meets their financial and lifestyle needs. If a given location where a job is has more nearby housing supply at lower rent, again it is easier to choose a walking commute. This is a major reason why increasing density in cities is associated with increasing economic mobility.

It’s a common trap to think about policy decisions as merely responding to the data provided by markets in labor, real estate, and the like. But in fact, these markets lie downstream of law, policy, and the city’s built environment. Most companies are choosing cities based on the environment they provide. Historically, this becomes especially transparent when the world’s largest companies encourage large cities to court them, as Amazon did when it pushed cities to try and out-compete each other to become the location of Amazon HQ2.

When considering markets as downstream phenomena, it becomes obvious that big cities have leeway in choosing which markets to encourage. Markets are tools, not states of nature. For example, small-and-medium enterprises (SMEs) depend on reliable, low-barrier legal institutions, and markets for demand and labor supply. Since they often operate on lower profit margins, these changes can often determine whether or not they operate at all. While many small towns have seen their SME sector hollowed out, out-competed by the country’s Walmarts, cities provide a much more sustainable environment for SMEs to operate in. By encouraging a built environment which lets SMEs, employees, and customers live closer to each other, San Francisco could diversify the sizes and sectors represented in its economic life, as well as increasing the capacities of the larger tech players that call the city home.

From the perspective of city government, greater density also means it is easier to provide city services. The cost of city services like roads, water, and even fire departments and police is dependent not just on population but on the physical distances involved. A less dense city requires more per capita spending on pipes, roadways, and requires more fire and police stations, than a denser one providing the same services. This is one reason why more densely populated cities typically are more efficient at providing city services. A study published in the Government Finance Review in 2015 found, unsurprisingly, that “communities realize savings from the spatial efficiencies created by more compact development (higher densities) and the ability to make use of existing infrastructure when growth follows an “infill” strategy” of increasing the area available for housing and businesses within the existing boundaries of the city.”

In other words, cities should view density as not only a social and economic benefit, but also as budgetary savings when it comes to growth. The same level of growth will be easier to service in a higher density city. As city residents use more and more land, they stretch city budgets and require more spending on city services. Tying local taxes to underlying land value, not property value, ensures that the city can properly absorb the costs of this kind of growth when it does occur; denser growth will be more concentrated in the municipal core, and require less expansion of transportation and services This density is also a tool for ecological stewardship: denser cities permit lower per capita carbon use, less paved area per person, and greater preservation of habitats and open spaces. This is already visible in San Francisco: Sunset and Richmond districts produce around 40 metric tons of carbon emissions per household, while the densest zip codes produce fewer than 30 metric tons.

While San Francisco is currently a disaster when it comes to public order and cleanliness, a city government which explicitly sought to create safe and family-friendly public green spaces would ensure that density is not only economically and environmentally friendly, but creates neighborhoods full of life and social connection.

A developmentalist city government would have to counter opposition to both the land value tax itself, and the density which it is used to achieve. The above focus on social and green spaces helps to alleviate one of the major anti-density concerns: that density itself will be burdensome and unpleasant. This response is visceral—the metaphor of packed sardines is frequently invoked, and at the mention of residential density Americans especially start to imagine either 19th century tenements or dystopian sleeping tubes. This is a falsehood in and of itself: total amount of living area in a denser development can in fact be greater than in a less dense one, because taller buildings and more efficient parking allow every square foot of city space to host two or three square feet of offices, apartments, or shopping.

But the fact is that the forms of density, and the experience of it, are themselves downstream of policy. Just as the land value tax is a tool to create density, so too is density itself a tool to create a great and prosperous city. It should not be taken as an end in itself. One simple policy is to allow ultra-localized voting controls over the type of building projects which occur, with supermajority voting to ensure that a single objector does not stall the process.

A denser city seems worse for automobile transportation, but it gives rise to opportunities for a more socially and environmentally healthy transportation system. As people live in denser areas—and more mixed use ones—they tend to use automobiles less frequently, or forgo owning them entirely. The hassle of starting and parking a car simply isn’t worth it for most people taking short trips, and the percentage of trips that are short is much greater in a high density, mixed use development. This less frequent travel can actually translate to less congestion as density rises. This not only reduces the environmental impact of cities but also provides numerous health benefits, both to those who now forgo driving in favor of more active options, and to everyone breathing the air in the city. The social benefits of more regular face-to-face contact with neighbors also cannot be discounted.

But the greatest obstacle toward unleashing San Francisco’s productive energies, and to building a city in which families can live and thrive, is not simple policy objections. Rather, it’s the massive coordination problems which beset most of California’s state and city politics. Making these outcomes possible requires coordination among many stakeholders: California would have to reduce the scope of Prop 13, the voters of San Francisco county would have to agree on a billion dollar change to their tax system, and zoning boards would have to react in a timely manner to allow land use to change to reflect the new system.

It’s difficult to see how San Francisco could get the job done by itself, without support from the state level, and even the federal level. Ideally, the different levels of government would be united on what is needed to build the future of American cities—at the very least, they wouldn’t get in each other’s way. But if the city were able to employ the land value tax and achieve the reforms discussed above, it would be a milestone in the city’s life. A strong precedent would be set for recognizing the nature of land as a commons in law and in the tax code. Moreover, it would do so in a way that balances the collective stewardship of a commons with the autonomy of residents. The taxation needed to cover public costs would focus more on the use of common resources, and proportionally less on the real proceeds of individual labor.

A city government which understands that these markets exist downstream of policy and higher-level coordination would escape one of America’s most fatal governance traps. Cities from San Francisco to Detroit have hollowed out as governments have shrugged their shoulders, using the ‘realities’ of the market as an excuse for inaction. If one of America’s greatest cities proved this to be no more than a convenient fiction, it would be increasingly difficult for other cities—not to mention state and federal governments—to excuse their political sloth. Indeed, those who successfully carry out such policies at the city level would likely gain tremendous opportunities at these higher levels.

Reformers are often accused, not always unfairly, of ignoring the reality of individual wants and needs in favor of an idealized vision of society that fits their policy, but bears little resemblance to actual conditions. A Georgist land value tax isn’t like that. It starts with two facts. The first is the finitude of the land in San Francisco; the second is the power of well-aligned incentives. The land value tax brings policy in line with reality. But in doing so, it doesn’t limit the city’s potential—it unleashes future growth. It does not force residents to subsume their own good for that of the collective, but instead rewards the efficient use of community assets.

From the perspective of city government, using a land value tax as a structuring framework for a larger initiative for densification and rebuilding represents a shift away from technocratic policy-mongering. Contemporary American governance is plagued by a surplus of grifters, offering a disparate array of optimizations, marginal improvements, and success metrics. This is policy done by wonks, not visionaries. San Francisco must have a clear and bold vision of what it must become. Policy should be the result of reverse-engineering that outcome to figure out the path to success. A prosperous and growing metropolis demands density and its benefits, which in turn requires that the fixed land supply be properly managed. The land value tax is the right approach, as Henry George foresaw generations ago.

Such a change would fuel the city’s future growth and incentivize the heights of civic achievement. More importantly it would ensure that all San Franciscans stood to gain from the realization of this potential. That is a city worth building.