Why Urbanists Must Support Linkage Fees and Inclusionary Zoning: A Scalable Policy For Affordable Neighborhoods


Over the past few years a convincing narrative emerged explaining high housing costs in cities. As the narrative goes, urban housing is expensive because supply doesn’t meet demand; therefore, we need to build more housing to bring down prices. This view is bolstered by writers like Ed Glaeser (The Triumph of the City), Matthew Yglesias (The Rent Is Too Damn High), and Ryan Avent (The Gated City), each making compelling arguments for removing limits on housing in urban areas. Glaeser argues that laws regulating additional housing are directly correlated with changes in housing costs, including laws restricting building heights. Yglesias claims that limits on housing are blue America’s biggest failure. Avent goes so far as to claim that America’s economic stagnation can be attributed to housing regulation.

It’s true that expensive cities need a lot more housing. However, this narrative overlooks the differences in regulations affecting development. Specifically, it doesn’t differentiate between regulations like height and density limits, which prevent new housing, and impact fees or design review, which raise the cost of housing production. The former are limits on housing. The latter are regulatory costs.

Seattle urbanists often conflate additional building costs with limits on housing; frequently suggesting that regulatory cost, not housing limits, are the biggest impediment to affordable housing. The result of this mistake has been detrimental to urbanists’ goals, creating an adversarial relationship between urbanists and affordable housing advocates. Furthermore, blurring the lines between housing limits and regulatory costs induces urbanists to overlook the most important factor in housing affordability: land values.

This article will first show that linkage fees are misunderstood because the relationship between land values and regulatory costs is either misunderstood or overlooked. It will then show that land values are the root cause of the housing affordability crisis and linkage fees, a form of inclusionary zoning, are a unique tool to ameliorate this crisis. It will then address urbanists’ objections to linkage fees and provide empirical evidence showing that linkage fees produce affordable housing without reducing the supply of market rate housing. Finally, I will present the political reasons urbanists must support linkage fees.

Land Economics: Where Does Land Value Come From & How Does Regulation Affect It?

The Urbanist has covered land value theory before. To summarize, the price of land is connected to the amount of revenue that land can generate. If land can generate more revenue, land is worth more. Economists call this revenue ‘rent.’ Sometimes, rent is direct, like the money that a landlord receives from tenants. Other times, it is implicit, like the rental payments a homeowner avoids from owning their home. Earning more income from land often requires additional costs, such as constructing a larger building, but the greater potential income that results from those costs is reflected in greater land values. The value of land is determined by the relationship between potential income and costs. This is an old concept that has been well demonstrated:

The law of rent makes it clear that the landowner has no role in setting land rents. He simply appropriates the additional production his more advantageous site makes possible, compared to marginal sites. The law also verifies the claim by Adam Smith that the landowner cannot pass on the burden of any cost such as land value taxes to his tenants.

Perhaps most importantly, anything that reduces economic rent — e.g. regulatory costs, permitting, design review, etc. — also reduces land value by a corresponding amount.

Courtesy of the City of Seattle. Taken from the policy options report on incentive zoning reform.

This is why it’s wrong to suggest that regulatory costs are the same type of impediment to additional housing as development limits. Because of the unique nature of land, regulatory costs reduce the value of land while development limits create a ceiling for land value. If regulatory costs were reduced, would developers build more housing? Probably not. Would developers make more money? Again, probably not. Instead, landowners would recognize the land’s potential income and, in the long run, landowners would charge developers more for land. Regardless of the rent for apartments or the demand for housing, developer profits are relatively steady.

This is the most important concept for urbanists to understand: Development costs, including regulatory costs, drive down the value of land; development limits determine the ceiling of land value. Understanding this concept allows urbanists to see that reducing regulatory cost benefits landowners, not developers or renters. Alternatively, upzones produce more housing, indirectly benefiting renters, but also increase land values, massively benefiting landowners.

Piecora’s: Insane Profits With Virtually No Effort

Most of the income from higher rents flow to landowners, which is why landowners make a killing during urban construction booms. A great example is Piecora’s on Madison. Not too long ago Piecora’s sold for $10.29 million, not because their pizza is delicious but because the land beneath the pizza shop is valuable.

Courtesy of Joe Wolf

The developer who bought Piecora’s will invest in the neighborhood, add improvements to the lot, earn a modest profit and be pilloried by neighbors for closing a favorite business. Meanwhile, the landowner (who also owns Piecora’s) will abandon the business and make a 250% profit, over $7 million, simply by being at the right place at the right time.

This is urban land economics, but it raises an important question: Why does value flow to landowners instead of developers?

Why Money Flows to Urban Land Owners

It’s been shown that Walk Score is a decent proxy measurement of value. But what does this mean? High Walk Scores indicate proximity to desirable characteristics: Transit, grocery stores, schools, jobs, bars, restaurants, people and opportunity. I will call the sum of characteristics that make city-life enjoyable urban benefits. A higher Walk Score means more urban benefits in close proximity. A parcel of land close to many urban benefits is a desirable piece of land. This desirability means landlords can charge more for housing and people will pay more.

The space available in proximity to urban benefits is fundamentally limited, and this limit creates a hierarchy of desirable housing. It’s possible to create more neighborhoods with more urban benefits, but this takes time and a massive amount of investment. Many places solve this problem by living more densely or improving transit, so that more people can access urban benefits. However, there are diminishing returns on transit investment and living more densely. In Seattle, many areas in close proximity to urban benefits only allow detached, single-family homes, limiting access to urban benefits until upzoned.

In fact, as we expand upzones and urban benefits in Seattle, more places will increase in desirability and will experience land values increase. Places with high land values in proximity to locations with increasing urban benefits will see land values increase even further. In other words, adding urban benefits reinforces the hierarchy of desirable housing. This is the fundamental logic of why rents and land values decline from the center of a city outwards. Furthermore, this explains why zero-based zoning does nothing to address housing affordability; it doesn’t reduce or change the hierarchy of desirable housing.

The fundamental insight to understand is that because each parcel has unique value, landowners command a monopoly on their location. A piece of land within walking distance to a subway station is fundamentally different than one within walking distance to a bus stop. Developers compete among each other in an open market against monopolist landowners who extract all of the value (up to the point that development would no longer make sense). Ultimately, the worth of urban benefits flows to landowners.

Perhaps the most enraging aspect of this equation is that the additional value landowners are gaining is largely contributed by the public. Tax dollars invested in transit, fire, police, schools and much more create most of this value. There is even tremendous value gained simply when people live near each other, whether it is from access to social interactions or to professional opportunities. In the end, we all create value. This value creates desirability, increasing rents and creating higher land values and a hierarchy of housing choices. The end result of this hierarchy is segregation.  Those with less money end up in areas with fewer urban benefits.

Seattle is currently investing in a subway system and the lucky landowners next to future stations won the lottery; a public investment turns into a massive private profit and the city suffers due to higher rents and segregation.

The Root Causes Undermining Efforts For A Better City

I’m not opposed to people making a lot of money out of sheer luck. It doesn’t seem completely fair, but it’s wrong to begrudge others’ luck. What should be vilified is unearned profit that creates misery. This is the dynamic of urban land values. The urban land market exacerbates four problems:

  1. Developers appear to profit from construction when it’s actually lucky landowners making bank
  2. Renters experience increasing land values through increased rents
  3. Increasing land values make it difficult to maintain mixed-income neighborhoods
  4. Increasing land values create diminishing returns on subsidized housing

Developers appear to profit from construction when it’s actually lucky landowners making bank:

The first point is one that anti-development advocates latch onto in their efforts to demonize construction. Occasionally, land doesn’t capture the value of rents. Risk-taking developers sometimes gamble on possible rent increases by paying the current market value of land, building fancy apartments, and then advertising them with high rents. If they are able to fill the apartments, they will make a killing. If they aren’t, they’ll face a loss. But this is the exception, not the rule, and many developers lose money taking this gamble. Unfortunately, once one developer gambles and wins, land values adjust accordingly and the next developer must pay more for land. In the long run, developer profits are actually steadier and mostly independent of the rent tenants pay. Most developers profit by minimizing the risks involved with building and investing in a neighborhood, not by charging exorbitant rents.

Unfortunately, neighbors see new construction with higher rents and associate this with developer profiteering. Developers take the political heat while landowners laugh all the way to the bank.

Renters’ lived experiences tell them that development increases rents:

Development does often lead to displacement. Land values increase as higher rents become possible. On a hyper local level, this likely means a property tax increase. Since a landlord is largely in the business of managing operations, nearly all the increase in property taxes are passed on to renters. Most urbanists argue that these costs would be worse if more housing weren’t built, and they’re right. But this response does nothing to actually solve the problem or address the lived experience of renters. This lived experience creates grassroots opposition to new development.

Increasing land values make it difficult to maintain mixed-income neighborhoods:

Change in Seattle’s household income composition over the last 3 years. Also an example of the common urbanist response to this problem.

All of which leads to the third point. If new neighborhood development is charging substantially more than old development, an area will likely see a boom in construction. Developers buying land have to pay increased land prices, reflecting the expectations for higher rents. Because of the upfront land cost, it becomes nearly impossible to build lower-cost units without subsidy. Any unit priced below the market rate causes a loss for the developer. This is why new development in cities is expensive, while new development outside cities can be accessible to middle-income earners. This is also why many solutions suggesting we can build below-market housing without subsidy don’t pencil out. Landowners know the market rate for rents and sell their land accordingly. In the end, neighborhoods end up with rents that are all similar and become economically segregated. This economic segregation is at the heart of the housing affordability crisis. It’s important to note that most cities have affordable housing somewhere. The real problem is that particular neighborhoods are unaffordable and economically segregated.

Additionally, this segregation has a huge impact on individuals lives. New research looking at millions of families over long periods of time indicates that proximity to urban benefits is incredibly important. The research shows that children growing up in better neighborhoods earn more, experience better health, are more likely to attend college, and generally have a better life. An economist writing on these findings says:

…the relentless accumulation of evidence is now so compelling that I believe it will sustain a new consensus. That consensus, simply stated, is that place matters.

Segregation from rising land values eliminates access to better neighborhoods for people lower on the economic ladder. It’s not an exaggeration to assert that access to urban benefits is deeply connected to many intractable, national problems.

There are diminishing returns on subsidized housing:

Lastly, many urbanists’ preferred method to achieve economically diverse neighborhoods is insufficient. Rising land values make it increasingly difficult to provide subsidized housing in the long run. Many people think a larger housing levy or state trust fund is the simplest way to support subsidized housing. This overlooks the role of land values. In areas needing the most subsidy, land values are often the highest and fastest growing portion of costs. Without intervention, land values will continue to increase and over time, it will take more dollars to provide the same number of affordable housing units.

History, Economic Morality and the Big Picture

Henry George is perhaps the most famous person to write about the unique nature of land. He became well known in the late 19th century advocating for a land tax on the unearned value accumulated in land. George made both a moral and practical appeal for this tax. First, he saw it morally superior to other taxes because the wealth from the value of land is not wealth that is earned through effort. In other words, it is unfair for a Seattle landowner to get rich because the public invests in a light rail station. Additionally, George understood classical economics. When most goods are taxed, the price of the good increases, reducing the demand and consequently reducing the supply to match the reduced demand. This is called ‘elasticity.’ Strangely enough, land is not such a good. Taxing land doesn’t reduce the amount of land in the world. Land is (for most purposes) perfectly inelastic.

economist housing values
Credit The Economist

Housing costs in modern cities present the most stark examples of wealth without effort. Currently, cities across America have vast plots of unused land, largely in the form of parking lots. These plots remain unused because landowners are speculating on the value of their land, minimizing  operational costs in order to hold the land longer. When they think the value peaks, they’ll sell. Massive fortunes are amassed on this strategy. There are publicly traded companies that depend on under-utilizing our cities; effectively harming all the residents. A land tax increases operational cost, forcing landowners to make a choice to utilize their land and collect income or sell the land.

George’s idea was so influential that many economists even today consider land taxes the most moral and effective way to raise revenue. This idea is also very popular with urbanists (see Sightline Institute for a great rundown of this) because it results in better land use. Urbanists support a regulatory cost on housing.

Concerns about land values are making a comeback, and land value taxes deserve more attention because they may eliminate some of the biggest problems in urban areas. Recent examples such as commentary decrying the surge of foreign capital into urban real estate are likely the result of land value speculation. How large is this problem? Perhaps the best response to Thomas Piketty’s Capital in the 21st Century is an analysis that looked at what types of capital are appreciating. The biggest finding suggests that increasing wealth from capital is largely due to housing.

In a country where urban land makes up most of the value of land, the rising value of housing is likely due to rising urban land values. So what is the connection to linkage fees? Put simply, a land value tax would recapture the unearned increase in land values. It would also discourage speculation. A linkage fee would likely have many of the same effects, capturing land value increases for public benefit and decreasing the return on land which should discourage speculation.

Linkage Fees Are Inclusionary Zoning: An Essential Tool Of Housing Affordability

The term ‘inclusionary zoning’ is derived from the long battle to end exclusionary zoning practices. Exclusionary zoning laws were designed to purposely exclude groups of people from neighborhoods, usually racial minorities. These laws included racial covenants, redlining, and many other practices that are responsible for the segregation we still see in our cities today. To overcome this and meet the promises of the civil rights movement, activists began pursuing inclusionary zoning. The first instance of inclusionary zoning was implemented in Montgomery County, Maryland, a racially and economically diverse suburb of Washington DC.

Racial segregation in Seattle. Courtesy of The Demographic Research Group at the University of Virginia. You can explore the maps yourself here.

Inclusionary zoning laws vary widely but generally require private developers to include a certain number of affordable housing units in new development. A municipality might require that 5% of all units in new buildings be affordable, perhaps by lowering rent to less than 60% of AMI. There are often opt-out clauses that allow developers to pay a fee instead of building units. The fee goes to the municipality and is used to build affordable housing. Seattle’s linkage fee policy would require developers to pay a fee or opt out by building affordable units. The linkage fee would act like an inclusionary zoning ordinance.

Inclusionary zoning was invented to produce affordable housing and help integrate communities. It is scalable by design, increasing the costs of development as well as affordable housing in proportion to the amount of total housing built. But the policy adds a critical benefit beyond building units and integrating neighborhoods; a benefit not seen in other affordable housing tools. It passes regulatory costs to landowners, throttling land value increases. Overall, inclusionary zoning captures some of the increasing land values in a growing city to produce affordable housing and economically integrated neighborhoods, value that otherwise would increase rents to profit landowners. As we will show later, it does all this without reducing the amount of housing built.

Dissipating The Problems Undermining Development

If you acknowledge the evidence that regulatory costs are passed to landowners, linkage fees will effectively solve the problems that undermine building a better city. Most importantly, they will create a political dynamic that will encourage more development.

Developers appear to profit from construction when it’s actually lucky landowners making bank:

New construction and upzones will create affordable housing. Everyone will know that new development is building more affordable units and private developers will get the credit. This is done by reducing landowner profits but it will appear as if developers are using their money to make a better city.

Renters’ lived experiences tell them that development increases rents:

With linkage fees, the lived experience of renters becomes very different. As developers calculate their costs and needed returns, they bid less on land and landowners can no longer make a killing. This means that increases in property values will be more uniform across the city. Existing buildings in neighborhoods with construction won’t see as steep property tax increases. Consequently, renters in existing buildings won’t see rents spike due to property tax increases. The lived experience connecting development to increasing rents becomes much different.

A desirable city finds it increasingly difficult to support mixed-income neighborhoods due to high land values:

Linkage fees necessarily require new construction to be mixed-income. This won’t be a panacea for segregation, but it will go a long ways towards resolving inequities. This is something that won’t happen without government policy. Without market intervention, there will always be a hierarchy of housing desirability, a hierarchy of costs, and segregation.  Many of the oldest, most intractable problems in our cities and country are connected to segregation. Currently, the overwhelming solution is to subsidize people to live where they want. This is expensive, reducing the number of people receiving subsidies. It has also shown extremely limited success.

There are diminishing returns on subsidized housing:

Linkage fees avoid the diminishing returns on subsidizing the construction of affordable housing. The amount of housing is scaled to the growth of the city. If the city is experiencing a boom, more affordable housing will be built. Rather than taxing people and using those dollars to buy diminishing amounts of land and produce less and less housing, linkage fees always produce the same amount of housing relative to the amount of development. Additionally, since it slows the growth of land values, trust fund dollars go further.

Linkage fees are a policy tool that unites urbanist constituencies:

Most importantly, this type of policy will align the interest of affordable housing advocates, developers, environmentalists, and renters. In order for linkage fees to produce affordable units, an area has to encourage private development. This means development limits, such as single-family zones and height and density limits, hinder the production of affordable housing. The arguments made by anti-development advocates are persuasive because development increases land values, displacing residents. By disconnecting these two mechanisms, anti-development advocates will be marginalized. Furthermore, urbanist political constituencies will have an incentive to eliminate housing limits and encourage development. When private developers build affordable housing, new construction becomes the path to a better city.

Addressing Urbanist Concerns

The linkage fee proposal in Seattle has been mocked and vehemently opposed by many Seattle urbanists. Some Seattle urbanists, including the Downtown Seattle Association, even pursued a lawsuit to require an environmental impact study that might prevent the fees. The lawsuit targets the process, but is intended to obstruct the policy. Additionally, a group that includes Vulcan, one of Seattle’s largest urban landowners, is developing opposition to further regulatory costs.

Opposition to regulatory costs are the result of conflating regulatory costs with development limits and misunderstanding the effects of regulatory costs on land values. If urbanists understood the theory of land economics and looked at the evidence, they should come to the conclusion that linkage fees will not reduce development.

To address Seattle urbanists’ concerns it’s necessary to understand that they believe increasing the number of people who choose to live in the city is paramount. They know the benefits of urban living are huge. Many more people would live in the city if they could find acceptable housing at the right price. Most urbanists have the goal of building as much housing as possible.

With that in mind, urbanists’ opposition to linkage fees usually falls into one of these five categories:

  1. Increased development cost will be passed on to renters, increasing rents
  2. Increased development costs will make some projects unaffordable, reducing production
  3. Even if increased development costs are passed on to landowners, fewer landowners will sell their land at a lower price, reducing housing production
  4. It’s unfair to put the burden of providing affordable housing on developers
  5. Why spend all this energy and effort on linkage fees when we can focus on building enough housing to drive down prices?

To begin addressing these concerns, it’s important to understand that there are only three possible reactions to linkage fees: The added costs could be eaten by the developer, passed on to the renters, or passed on to the inputs for development, most likely the land. Any outcome that doesn’t reduce the production of market rate housing but does increase the amount of affordable housing would be a win for most progressive Seattleites and urbanists. For that reason, I will focus on whether or not any of the objections reduce the amount of housing built.

Increased development cost will be passed on to renters, increasing rents: 

This hypothetical outcome is not how the rental market actually works. Landlords can’t set rents according to development costs. Rents are determined by the relationship between the supply of housing in a market and the demand for that housing. If the supply/demand relationship allowed for higher rents, landlords should already be charging more regardless of linkage fees. Linkage fees can’t affect rents unless they affect the supply/demand relationship and I’ll address the supply issue shortly. Lastly, if developers resolve this problem by passing the fees to renters it means there is no reduction in housing construction.

Increased development costs will make some projects unaffordable, reducing production:

This hypothetical also misunderstands how the market works. Developers would seek to lower their costs before giving up. This logic is demonstrated in Vulcan’s quarterly survey predicting that linkage fees would likely reduce bids on land. Developers that do nothing, sitting on their capital, would not be rational economic actors and would lose market share to developers bidding less on land. Ultimately, the same amount would be built as developers bidding less gained more market share. This article will later show that supply is not constrained which means developers don’t stop building projects. This leads to the next objection, what happens if developers bid less on land?

Even if increased development costs are passed on to landowners, fewer landowners would sell their land at a lower price, reducing housing production:

Live chat now about affordable housing in Seattle The Seattle Times
Seattle Times Chat on Housing Affordability

If bids on land decrease, won’t landowners sell less land? This objection is most clearly aligned with classical economics but it misunderstands the unique nature of land. The answer is a definite ‘no’ and this is because land is perfectly inelastic and heterogeneous. In fact, even Smart Growth Seattle points us to evidence that linkage fees don’t constrain supply; that is, the fees don’t constrain the amount of land that landowners sell. Other urbanists have begun to state this fact as well.

And this makes logical sense when you consider that landowners treat their land like an asset. When all developers bid lower on land, landowners can’t convert their land to some other economic good or take it to a different market. They don’t sell more when their land is more valuable or less when their land is less valuable. In fact, there might even be a negative relationship between land sales and values, in which land sales increase as land values drop. This is because the decision to buy or sell an asset is based on expected future value, not current value. If you expect your land to increase in value, then it makes sense not to sell it. The inverse is also true. In fact, linkage fees may have a significant effect on land sales; as the fees are implemented sales should accelerate because landowners would expect their property to become less valuable. Overall the evidence is strong and The Urbanist has made an empirical case for this, showing property values are not connected to the number of property listings.

Even if there were a stronger correlation between land sales and land prices (which, again, there isn’t), the value gained from redeveloping most properties in Seattle is so much greater than current land income that additional costs from linkage fees wouldn’t discourage the sale of land and subsequent development. Even in the case of interim use properties (a property generating income on an under-utilized lot) where the gap is smaller, income from an interim use diminishes over time and the sale value can be greatly increased with upzones. When considering the hypothetical that landowners earned nothing from selling, they would still have an incentive to better utilize their properties in order to earn more income. This suggests that land with interim uses still has a strong incentive to be better utilized. In fact, if landowners expect less money from the sale of their land, they may depend more heavily on the income from the land. This means reducing sales prices may reduce speculation and increase utilization. Lastly, the money raised from linkage fees goes back into building units, increasing the money used for development. Without linkage fees, all the money earned from Piecora’s sale left the development market.

It’s unfair to put the burden of providing affordable housing on developers:

I sympathize with the sentiment in this objection. I find it equally dispiriting that developers are scapegoats for Seattle’s problems. It seems like nothing more than a convenient, populist appeal when political leaders say we should increase developer fees. But a pragmatic solution to this problem is a solution that makes developers heroes while extracting value from landowners. Linkage fees capture the anger aimed at developers and redirects it towards something constructive. If you seriously want to change the narrative about developers, it seems logical to give everyone the impression they are building tons of affordable housing with their profits.

Why spend all this energy and effort on linkage fees when we can focus on building enough housing to drive down prices?

The final objection is usually the last resort among linkage fee opponents. It concedes the fee wouldn’t affect housing supply, would produce affordable units, and might even integrate neighborhoods, but suggests it’s a distraction from increasing Seattle’s housing stock and therefore a distraction from solving the affordable housing problem. Urbanists will typically try to refocus the conversation by saying something like, “what we really need is more developable land” or “we should have zero-based zoning.”

I agree that the linkage fee won’t solve all our problems. Even if the amount of affordable housing we build from this point on matches the nexus between growth and the need for affordable units, we would still have a huge deficit. We should absolutely focus on removing or reducing housing limits. We should absolutely advocate for building more housing.

With that said, if we only build more housing without addressing access to urban benefits we will never solve the affordable housing crisis. Higher land values make it unprofitable and nearly impossible to build new, low-cost housing without subsidy and consequently produce mixed-income neighborhoods. The hierarchy of housing desirability is reinforced the more we build. This means just building more housing increases, not decreases, segregation, rendering urban benefits inaccessible to large segments of the population. If we want a city with affordable housing in every neighborhood, we need inclusionary zoning before land values increase even further.

Second, housing limits won’t be removed simply by stating the urbanist argument more loudly. Urbanists must build political coalitions. Linkage fees unite diverse political groups in the shared interest of increasing development capacity. This political unity is exactly what urbanists need to reduce development limits.

Evidence That Inclusionary Zoning Doesn’t Reduce Housing Production

At the heart of this long argument is the assertion that linkage fees won’t reduce housing production. I’ve offered a narrative explanation and economic logic for why this is true but we should all be skeptical of logical arguments without evidence. In fact, these explanations were motivated by empirical evidence showing linkage fees won’t reduce housing supply and might even increase production. There are five academic studies examining empirical evidence about the impact of inclusionary zoning on market rate housing production. Out of these five studies, only one found a relevant negative impact. That study, published by the libertarian think tank Reason, failed to acknowledge larger market forces, drastically oversimplified its models, and could be reasonably accused of intentionally misleading its audience. You can read a criticism of the study here.

The other four studies found inconsequential or no impact on market rate housing production. The first study found the most negative evidence. The Furman Center for Real Estate and Housing Policy at New York University compared jurisdictions with and without inclusionary zoning around San Francisco, Boston and Washington D.C.

Our analysis finds no evidence that IZ programs have had an impact on either the prices or production rates of market-rate single-family houses in the San Francisco area. In suburban Boston, however, we see some evidence that IZ has constrained production and increased the prices of single-family houses. The number of affordable housing units produced under the suburban Boston IZ programs, and the estimated size of the programs’ impact on the supply and price of housing are both relatively modest.

While the conclusion may seem negative, a modest impact on single-family homes in one of the three jurisdictions, the full study indicates that this impact disappeared in 2 of the 4 models they used. Additionally, it doesn’t indicate that there was less housing produced overall.

Another study examining jurisdictions in Los Angeles and Orange County using multivariate regression analysis found no negative impacts:

Our research also suggests that critics of inclusionary zoning misjudge its adverse effect on housing supply. Contrary to the claims against inclusionary requirements, we found no statistically significant evidence supporting the purported negative effects of inclusionary zoning on housing supply.

Yet another study looked at jurisdictions in California with and without inclusionary zoning between 1988 and 2005, with surprising results:

The analysis found that inclusionary zoning policies had measurable effects on housing markets in jurisdictions that adopt them; specifically, the price of single-family houses increases and the size of single-family houses decreases. The analysis also found that, although the cities with such programs did not experience a significant reduction in the rate of single-family housing starts, they did experience a marginally significant increase in multifamily housing starts.

While this study found negative impacts, it was on the price and size of single family homes, not the supply of housing. In fact, the most statistically relevant finding was an increase in multi-family housing, the opposite of claims that inclusionary zoning reduces development and density.

The fifth and final study also found no impact on housing supply. The research examined 28 municipalities in California, some with and others without inclusionary zoning laws, between 1981 and 2001. Seattle hired the author, David Rosen, to help develop the linkage fee program.

An analysis of these data shows that for the jurisdictions surveyed, adoption of an inclusionary housing program is not associated with a negative effect on housing production. In fact, in most jurisdictions as diverse as San Diego, Carlsbad and Sacramento, the reverse is true. Housing production increased, sometimes dramatically, after passage of local inclusionary housing ordinances.

From four studies, most findings showed no impact on housing supply. Two studies found statistically relevant measures that inclusionary zoning increased supply. Part of two studies found negative impacts on single family homes, a good outcome for urbanists.

There is further evidence if we expand our scope beyond inclusionary zoning and look at impact fees. Impact fees are very similar to the linkage fee policy. The traditional view of impact fees are that they reduce housing production. Most of these views are based on old studies. Newer studies are more accurate in their measurements and have diverged from the traditional views. For example, a study looking at impact fees in Florida found nearly a one-to-one relationship between impact fees and land values:

The results show that an additional $1.00 of fees … reduces the price of land by about $1.00.

First, we find that the difference in the effect of an additional dollar of real impact fees between new and existing housing is small and statistically insignificant. Second, although we are unable to measure the property tax savings that are expected by homeowners from the imposition of fees, we do find that higher fees reduce millage rates and the present value of the associated property tax savings are in line with the estimated effects of fees on housing prices. Finally, impact fees are found to reduce land values. (emphasis is mine) While the increase in housing prices is found to cover the current level of fees, developers purchasing land in the present have no assurance that rising housing prices will cover future fees, especially given the upward trend in total fees over time. Hence, impact fees add additional uncertainty to the development process, causing developers to reduce the amount they are willing to pay for land.

Another study again broke with conventional wisdom and suggested that impact fees actually increase the amount of housing produced in areas they are applied:

The results show that non-water/sewer fees increase the number of completions of all sizes of homes within inner suburban areas and medium-sized and large homes within outer suburban areas.

This is likely due to the reduction in land prices and the increase in demand due to urban benefits built with the fees.

To sum up, we have one study from an ideologically-driven organization suggesting some negative effects of inclusionary zoning, four studies in peer reviewed, academic literature demonstrating at least no impact on market rate housing production and possibly an increase. The economic theory and empirical research point in the same direction. Inclusionary zoning doesn’t reduce housing production and might even increase it.

The Political Landscape For Urbanists Going Forward

I was initially opposed to linkage fees. I began to doubt this view because the most urbanist politician in Seattle proposed the policy and it was supported by affordable housing advocates. It should give urbanists pause when their political views require that people who spent their careers working on housing affordability must either be willfully ignorant or actively disinterested in affordability. This seems implausible. As my skepticism grew, the urbanists’ blind spot became clear; we often conflate housing limits with regulatory costs and misunderstand land economics.

This fall, a new and drastically different city council will run Seattle. Right now, the campaigns are largely in disarray. There are more than 40 candidates and most are still forming policy positions. Still, there is nearly universal agreement; Seattle is facing a housing affordability crisis. Nearly every candidate will formulate solutions, which they will likely pursue if elected. Now is the time to push candidates for the best solutions.

Urbanists and affordable housing advocates agree that we need much more housing. Lesser Seattleites and affordable housing advocates understand land values are the largest long-term obstacle to affordable housing. Lesser Seattleites would solve this problem by lowering the ceiling on what can be done with land and advocating for housing limits. Affordable housing advocates would like to capture land value increases for public benefit and create mixed-income neighborhoods. Urbanists, following the lead of those conflating regulatory costs and housing limits, have seen little political success so far. In fact, it appears the Lesser Seattleites are a larger constituency and winning the political battle. Urbanists are faced with a choice.

We can remain the smallest voice in this debate. We can continue to conflate regulatory costs with housing limits. We can continue to ignore the problem of increasing land values. We can continue advocating only for policies that lead to displacement and segregation. We can expend our energy fighting against regulatory costs when we should be fighting for reduced housing limits. We can continue to use narratives that explain-away evidence rather than seeking to understand. We can continue to give people the perception that we are adversaries of affordable housing and integration by opposing a policy that evidence shows would be beneficial.

Or, we can expand our political constituency by joining with affordable housing advocates. This means urbanists must first support a policy capturing land value increases, and only afterwards work to remove housing limits. This also means shifting our attention away from total market deregulation and refocusing on removing housing limits.

It’s not coincidence that the Seattle City Council’s leading urbanist, Mike O’Brien, spearheaded a year-long effort to craft this thoughtful and strategic policy. The first vote on linkage fees passed 7-2 in support, but three of the council members voting in favor are retiring, and it’s not clear if a final vote will take place before the election.

The longer we wait to act, the more expensive our city gets and the harder it is to reverse. Urbanists need to be on the right side of this battle by vocally supporting linkage fees. This sign of goodwill would go a long way towards building a lasting coalition. After the linkage fee battle is won, urbanists can approach affordable housing advocates with credibility and work together to remove housing limits.

Editor’s Note: Changes have been made to the quote of the first study referring to impact fees. A number of readers responded that the study indicated impact fees increased housing prices. This is a misunderstanding of the abstract which I unsuccessfully attempted to simplify by focusing on the finding that costs are passed on to landowners. The study purposely looked at an area with impact fees and increasing housing prices in order to understand whether the two were connected. I’ve provided a fuller quote from the conclusion of the study that shows the  author’s finding; housing prices increased due to lower property taxes. This is contrary to earlier studies which attribute the increase to impact fees.

We hope you loved this article. If so, please consider subscribing or donating. The Urbanist is a 501(c)(4) nonprofit that depends on donations from readers like you.

Owen does servicing and consulting for a software company to pay the bills. He has an amateur interest in urban policy, focusing on housing. His primary mode is a bicycle but isn't ashamed of riding down the hill and taking the bus back up. Feel free to tweet at him: @pickovven.

Inline Feedbacks
View all comments

This is exactly the economic argument we needed. Thanks for researching and writing this!

Patrick Goetz

So, you’ve convinced me that urbanists should be advocates for linkage fees, which is nearly a 180 degree shift in my thinking. I was oscillating on my support for a local density bonus point program which basically introduces linkage fees triggered by additional entitlements. Allow me to offer a few friendly amendments:


Patrick Goetz

Excellent, comprehensive article! You’ve changed my thinking on linkage fees.

One minor point: This is a dead link:


Thomas Gonzales

Hi! I really enjoyed reading this piece! I especially appreciated incredibly clear explanation of land value theory. However, I wasn’t totally convinced by your argument towards the end. I am still a little split about affordable housing programs. Specifically, I am wondering if you have looked at this academic study: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2638698

Also, do you have any other thoughts about land taxes or other ways to capture increases in land value?


Great article. Conceptually this is an interesting idea. But, I’m still a little skeptical that the cost of the linkage feel will be offset by falling land prices.

We also need to study what will happen to the real estate prices in the rest of the city. Say hypothetically the city imposed a $1 billion fee on all new development. Nothing new would be developed. But, we would still have a growing city with rising demand. Prices for existing residential (and non-residential) buildings would soar.

Now of course, that won’t happen. But, it also seems unlikely 100% of the linkage fees will be borne by landowners.

I agree with you on the political merits of tying subsidies for affordable housing to new market rate construction. Perhaps developers could pay linkage fees in exchange for much greater density than current zoning allows?

Sarajane Siegfriedt

The study was done for the City Council by the consultant Rosen mentioned in the article. He looked at various fee levels and gave the Council a range from which they chose a low end.


Thank for a very well researched and well thought out article. I have no doubt that there are plenty of places in Seattle where “the value gained from redeveloping … properties … is so much greater than current land income that additional costs from linkage fees wouldn’t discourage the sale of land and subsequent development.”. Your example of a little one story pizza shop where a big building is to be built is a great one. There are other examples all over.

But to imply that this exists for *all* land is, of course, silly. So the question is — how much land is in that category, and how much isn’t? How much land is in the category that Chris listed below? The assumption is very little — so little to matter, but lacking an assessment of the land where these fees would be applied, I stand unconvinced that this is good policy.

More importantly, you extend your argument for linkage fees (which is a reasonable one based on the above assumption, which again may exist in the area where linkage fees will be charged) with an argument against the impact of regulatory costs (or construction costs in general) on development. This is absurd. If the price of building an addition dwelling unit to my house triples, I’m not building it. If the cost of building a townhouse triples, it won’t be built. Rent just isn’t that high (yet). This is precisely the type of housing that is likely to lead to lower priced rent. This is because the construction costs are so cheap. But not if other, artificial costs are attached to the construction. The biggest one is parking. This city is in the middle of a huge growth spurt, but we have very little ADU development, in part because this city (unlike similar ones) requires the developer to add parking. The same is true for low rise zones. In the first case, it simply isn’t worth the effort (an owner will build only if he/she expects to make a huge profit and the parking requirement cuts into it). In the second case, the owner of the single family house is better off just keeping it as a house (just as the owner of a two story warehouse is often better off just keeping it as is). Construction costs matter, especially for low density property and at the margins of profitability — which is where you are likely to get more affordable housing.

So count me in the group that thinks all of this is a big distraction. If we really want more housing in the city (which would then play a big part in housing being affordable) we should do the following:

1) Eliminate all parking requirements. As I said, this will have the biggest impact on the low end (ADUs and low rise).

2) Further liberalize the ADU laws. Even with the parking change, our laws are way to restrictive compared to Vancouver BC (or even Portland OR) — http://daily.sightline.org/2013/03/15/adus-and-donts/.


So, you are basically citing a handful of studies on linkage fees (that you admit showed mixed results) as evidence that the cost of construction has no impact on construction. Really? So we could raise linkage fees to a million dollars — even on an ADU — and it would have no effect whatsoever?

Martin H. Duke

Excellent piece!

What are the implications for property taxes more broadly? Are they a Pigovian tax where the revenue collection itself drives down values and increases affordability?

I think I agree with the fundamental tenets of this argument, though it seems to me that a linkage fee focused on housing discouraging housing as a use of land vs. commercial applications. That is, a tax on new housing encourages more office buildings, shops, malls, car dealerships, and parking lots vs. enlarging housing supply. I think property taxes, though less politically popular, would avoid these problems.

michael goldman

The linkage fee proposal would apply to all commercial property including apartments, condos, offices, and retail. So it wouldn’t make non-multifamily development less appealing than multifamily development.

Martin H. Duke

Well that removes some concerns!


If I understand the article correctly, property tax increases function very differently from linkage fees and would not make housing more affordable. According to the article, regulatory charges like linkage fees will not be passed onto renters, whereas property tax increases are “nearly all . . . passed on to renters.” It is not obvious to me why landlords could pass tax increases onto renters but not linkage fees, but it seems at least plausible given their different operation. Property taxes are universally applied, so when they go up, they affect the entire market, whereas linkage fees would be one-time charges on individual properties at the time of development.

Assuming the article is correct about this behavior, then property tax increases would presumably not drive land prices down — rather, they would drive rent prices up, thereby decreasing affordability.

I don’t really understand where land taxes would fit in here. My intuition is that their effect on land values would be similar to the effects of a full property tax, but the article seems to suggest that they would have similar effects to those of linkage fees.

Anyway, I thought the article was fantastic and thought-provoking, but I could still use some clarity on those points. Property taxes do seem like a natural way to try to recapture value from external improvements with fewer perverse incentives (e.g., even if it is true that linkage fees will not discourage new development overall, they might encourage less efficient forms of development, such as new single-family homes, which are not covered by the current linkage-fee proposal). If property tax increases to pay for affordable housing all got simply passed onto renters, they would be counterproductive, but I’d like to know more about why that is.

Sarajane Siegfriedt

The reason the two are different is that the linkage fee is paid once, as part of permitting, is built into the project’s pro forma and is financed over decades as capital cost. Like rent, property taxes are paid every year and come out of annual profits as operating costs. The linkage fee is known up front, whereas property taxes seem to rise every year, especially due to levies to pay for necessities not covered by Tim Eyman’s 1% properrty tax cap. We should avoid candidates like Rob Johnson (D4) and banker and FDIC regulator Kris Lethin (new in D5) who represent developers when they say levies are the best way to pay for affordable housing (insufficient) or more density, including backyards will open up other land. Lethin also said the answer to building more school classrooms is to take adjoining park lands.

Northgate is below its growth goals because it has been waiting since the bottom fell out of real estate in 2008 for lenders to see higher rents, making the business case. Thornton Creek sat nearly empty for almost two years of rent-up. Now, the Mullalys are ready to redevelop the low-income two-story buildings west of Target. Two large buildings have been completed on Northgate Way from 5th to 8th Ave. NE. The coming of Light Raile and economic recovery have made all the difference.

Sarajane Siegfriedt

A 5% linkage fee paid once at permitting becomes part of a 30+-year mortgage, capitalized as part of the cost basis. Property tax increases are limited to 1%/year, courtesy of Tim Eyman, so we have to resort to levies to cover all sorts of services and maintenance. A property tax in the form of a levy is an annual operating cost, passed directly to renters, both residential and commercial. Seniors on fixed incomes are being taxed out of their homes and are close to being maxed out on levies.

Mark A

No fan of Timmy, but one could argue that the property tax limit did not lead to the levies – it was the inability to prioritize city spending.

Sarajane Siegfriedt

If one did, one would be wrong, the state and local budgets are over-dependent on sales tax. When the bottom fell out of retail spending due to the Great Recession, the need for housing, healthcare and other services and interventions went up. The 1% cap disregards population growth and increasing needs such as homelessness and an increasing senior population. Obviously, sidewalks and local streets cannot be maintained under this lid. We cut back on parks maintenance , etc. What we needed was a summer jobs program for teens and other counter-cyclical stimulus infrastructure investment. This is what has led us to using levies for operations costs and obsolescence, such as the EMT communications gear

Sarajane Siegfriedt

Not true. With human services needs rising and sales tax revenue dropping in a recession, plus high population growth, the 1% cap is completely arbitrary. With no local revenue alternatives, both councils have relied on levies to fund even normal maintenance.

Sarajane Siegfriedt

The linkage fee would apply to all new construction, both residential and commercial. It’s a small marginal rate, one-time 5% fee paid at permitting and financed over the life of the mortgage. The rate is at the low end of what the consultants’ study showed was feasible. Now if we could just get Vulcan to agree…

Matt the Engineer

Wow. Excellent work. My disagreements are getting smaller.

I still have a fundamental unease with charging fees per SF of new construction – we generally tax what we want less of! Even if most/all of these fees make it down to the landowner, if they’re high enough they should discourage construction – even if you buy this land at a discount because of these fees, if you can find a server farm, stuff storage building, or other lower SF use, it might pencil out better than housing.

Take your math in the graphic to its extreme, where fees are so high that land is free. Surely that would stop development (why give you my land? I’ll just leave it as a parking lot and get done income). At what point does that effect kick in? Is it linear starting at low taxes? If not, why not?

I also feel that if we stop cramming all of the multifamily housing into 13% of our land area land prices for MFs would drop dramatically. But that fits in with your last Urbanist complaint. Convince me linkage is really a land value tax, and maybe we can do both.

Dan Bertolet

Owen, regarding point #2 in your above comment, any landowner would prefer to maximize income from their land, but just wishing for that won’t make it so. Converting a property to high density housing requires a complex redevelopment process, at the core of which is a lending decision made by a bank based on risk versus return. A linkage fee will reduce the return relative to the risk, and make it less likely that a project will be funded. Therefore I think it is pretty clear that in the case where the landowner and developer are the same entity, a linkage will reduce the development of housing.

You agree that if the fee was big enough to push the property value to zero, then the owner of an income generating property wouldn’t sell. At the same time, your central claim is that linkage fees would have zero impact on property sales for redevelopment. How do you explain that contradiction, i.e. if the upper limit results in zero sales, then as you lower the fee, how does it transition to having no impact on sales (this is the question Matt also asked)?

I haven’t had time to really get my head into this, but I’ll take a wildass guess that you may be reaching an incorrect conclusion about the effect of the fee on sales because the “perfectly elastic” theory applies only to the land itself, and not to the income generating uses on the land. Does that make any sense? And so the greater the income from the land relative to the value of the land, the more likely it will be that a linkage fee will impede sales and curtail housing production.


As I said above, I think the theory is similar to raising the minimum wage with a hot market — you are basically “skimming off the top”. The tax is not high enough to actually cut into sales, or cut into construction. I think that is a reasonable theory, especially in this market (where demand for housing is very high and artificially constrained). This is where the studies come in — studies based on actual examples (in markets that are relatively hot).

Where I take issue is with the idea that you could raise the fees to astronomical levels, or the idea that a fee has no effect at all. With both minimum wage and linkage fees this is simply not the case. Raise the minimum wage by ten cents an hour and you probably don’t increase unemployment very much — in fact you might increase it. One out of a thousand owners can’t pay an employee, but everyone else has more money to spend (the redistribution adds to more wealth). Raise the minimum wage by a million dollars an hour and of course you lay people off. The key is to find the “sweet spot”, and Owen thinks he has found it (with this fee). I’m not convinced of this, but I think it is a reasonable theory.

But if the fee was extremely high, then of course it would squelch development. Set it at a billion dollars and you can forget about development. The land isn’t useless, either (I would buy it all for a dollar). The land simply goes down to the value of its current use — which, in many cases (even that pizza place) — is pretty high. This is why there are plenty of places all over the city where development doesn’t reach the height limit. Often this is because the current use is pretty high compared to the ideal use and the cost of construction is too high (no sense in replacing a four story building with a six story building).

Even for brand new buildings, we have plenty of examples in this very city where constructions costs play a part in building height. If builders always simply built to the maximum height, then this entire article would be a lie: http://www.theurbanist.org/2014/09/02/85-foot-and-125-foot-height-limits-are-a-missed-opportunity/
But it is obvious (just by checking the zoning records and looking around) that both play a part. To quote the article “Given this limitation, every real estate developer must ask themselves,
“Are two extra floors (20 extra feet) worth the hassle of concrete and
steel?” For most developers, the answer is “no“. Instead, they choose to buy and develop lots with NC3-65 zoning.” In other words, building higher is more expensive, and thus usually not worth it. This is exactly the scenario we are talking about.

Again, I think it is reasonable to suggest that these particular linkage fees are not that high, and thus will have a minimal impact on development — it will simply cut into the property owners profits; and that this reduction in development will be more than made up for with new units constructed from the fee. Where I take issue is with the ridiculous leap made in this article (which is simply unnecessary and hyperbolic) that costs play no part in development, anywhere.


I agree that the notion that linkage fees will never discourage production is somewhat oversimplified. In theory, the price of a piece of land should be a discontinuous function based on the expected future value of *each* of its possible uses. It’s not just a basic supply and demand curve — the price will be the maximum of several different curves, and there will be slope discontinuities where circumstance change to make one use more profitable than another. I think the article makes a strong case that, in situations where new development is clearly the most profitable use of a piece of land, linkage fees won’t discourage that development or make it less profitable — the fees will just result in a lower price for that land. At the same time, in cases where new development wouldn’t have been profitable in the first place (as compared to say, the rents from the existing use of the land), then linkage fees won’t deter development because there is no development to deter. But there obviously will be cases right on the boundary where development would be barely more profitable than the existing use. In those cases, linkage fees cannot substantially decrease the price of the land, because the the price cannot drop below what the existing use would command. In that case, the development may no longer be profitable and will not happen. The higher the linkage fees and other regulatory charges, the more projects will fall within that zone on the boundaries of profitability, and the more development will be deterred. How large that effect would be and whether it would be noticeable or problematic in practice seems like a purely empirical question, and while the data presented in the article seems promising, I agree that it does not seem definitive to prove the categorical statement that “linkage fees won’t reduce housing production.”

That said, I still found the article extremely persuasive as to the broader argument that linkage fees could be a good idea, mostly in terms of the politics. I agree that the market is not going to solve the affordability problem on its own — we likely need to subsidize affordable housing and we will obviously need money to do that. By tying funds for affordable housing to new development, that could make upzones much more palatable. Relaxing zoning regulations will lead to more new development opportunities, so fewer pieces of land will fall right on that boundary where the best options for new development are only slightly more profitable than the existing use. So if linkage fees can be used to form a broader political coalition of urbanists and affordable housing advocates in support of both upzones and subsidized affordable housing, it does seem like it would be at least possible to end up with the best of both worlds.


I get why that would be true for inclusionary zoning (where, as I understand it, the affordable housing would be built in the new development itself), but why would it also be true for linkage fees? Is there something in the current linkage fees proposal that ties the new affordable housing to the neighborhoods from which the fees were originated?

Mark A

I am all for mixed-income neighborhoods but what is the track record of taking an established neighborhood from high-income to mixed-income?

Sarajane Siegfriedt

What is the point of looking at the extreme limits of the curve? Why not focus on the marginal effects on housing development of a 5% linkage fee, as proposed?


Addressing Urbanist concerns #2- Assuming a single level parking lot in Seattle CBD, you are correct linkage fees are not likely undermine supply. But that is not the predominant condition of properties that may be affected by the legislation. But in places with mid-rise zoning (NC zones, MR zones, etc.) where the base condition is SF houses, or a strip mall, it absolutely could affect supply. Even in places where the land lift currently seems substantive (ie existing SF vs. NC-65 in roosevelt, for one example recently), it takes a major inducement value, not to mention patience, to assemble and for redevelopment. In northgate, redevelopment values just barely clear the average strip mall, and linkage fees are supposed to hit this area, which is a well documented case of being below its growth goals. Same in Rainier Valley and ID.

Sarajane Siegfriedt

Not at all the same. Timing is everything. Northgate is below its growth goals because it has been waiting since the bottom fell out of real estate in 2008 for lenders to see higher rents, making the business case. Thornton Creek sat nearly empty for almost two years of rent-up. Now, the Mullalys are ready to redevelop the low-income two-story buildings west of Target. Two large buildings have been completed on Northgate Way from 5th to 8th Ave. NE. The coming of Light Raile and economic recovery have made all the difference. (Sorry for the duplication. Tried to delete the first post.)


This is a very good analysis, and make me more of a supporter of linkage fees. However, I have one criticism. The author assumes that apartments are the only possible use of land, when in reality it is not. Land value comes from the most profitable use (defined by total value of improved property minus construction cost of such use), whether its apartments or not. If the potential profits from building apartments is diminished due to linkage fees, other uses (such as condos, offices, hotels, grocery stores, auto dealerships, restaurants, nursing homes, etc) may become more profitable than building apartments. This would slow the development of new apartments because something else would get built instead. So, when determining what the linkage fee should be, an analysis must be done of what the next most profitable use would be after apartments, and make sure the linkage fee doesn’t bump apartments down to being the second most profitable use.


Yeah, I realize now after reading the link below that the linkage fee would apply to all development in non-single family zones. Thus, my point about the linkage fee making non-apartment uses more favorable and reducing housing as a result doesn’t apply. That’s why I deleted the comment. I didn’t realize you’d already replied to it.