In urbanist circles, one thing that is bound to come up is the issue of impact fees. Generally, market-focused urbanists can't stand impact fees. As they will tell you, this limits the ability to develop, and simply increases housing costs for the units that are in line for production.
It's a similar line of thought to levying taxes or fees on businesses. If you do it, they won't happen, and then everyone loses. Of course, reality tells us otherwise. In Seattle, we have secure scheduling, a higher minimum wage than the rest of the state, took the lead on sick and safe time, and enforce our laws through the Office of Labor Standards. Yet with all of these costs and regulations, businesses continue to thrive in Seattle.
Looking across the state at cities that do levy impact fees - there is no evidence that impact fees have an adverse effect on the rate of construction. Even in Seattle, limitations on development appear to have more to do with zoning restrictions and laborious processes to develop than anything else.
Personally, I view impact fees for what they are: a means of revenue to ensure that necessary infrastructure comes with growth. Had the Master Builders' Association not been donors to the campaign to defeat I-1098, I would give their arguments more cred. But we are limited in revenue streams, they worked to keep one more progressive option from being implemented - so let's work with what we've got.
Conversely, there are many who would seek to levy such high impact fees as to completely stifle development. Much like any outside influence, there is, on a more probable than not basis, a breaking point wherein increased fees would necessarily discourage development in the city, and developers would look elsewhere for certain types of projects (with others getting through fine, and the increase in cost being able to be borne by the target audience of that development. More on that later, if I remember).
There are constitutional considerations. Anyone deep into the weeds of land use policy should be familiar with Nollan v. California Coastal Commission, holding that a taking (such as a land-use permit application requirement putting limitations on use of property) must further a genuine public interest (among other things). Dolan v. City of Tigard took it a step further, requiring that exaction to find a nexus with the proposed development (ie: while it may be in the public interest for a bike path to be included in a redevelopment, there must be evidence that the redevelopment will be the cause of greater bike use). The legality more generally of developer impact fees were contemplated in Koonts v. St. Johns River Water Management Dist., where the Court affirmed that impact fees are constitutional inasmuch as they follow the requirements of Nollan (furthering public interest) and Dolan (nexus to the development). These all stem from the Takings Clause of the Fifth Amendment to the constitution.
In Washington, there are clear statutory considerations. RCW 82.02.050, initially implemented as part of the Growth Management Act (GMA), requires that such fees "only be imposed for system improvements that are reasonably related to the new development." The fees may only be spent on facilities that are part of a "capital facilities plan element of a comprehensive land use plan." The specific facilities allowed are identified in RCW 82.02.090, and are limited to four things: public streets and roads; publicly owned parks, open space, and recreation facilities; school facilities; and fire protection facilities.
So put plainly, in Washington State, and in accordance with the U.S. Constitution, impact fees may only be assessed for new or expanded roads (not transit), new parks, new schools, new fire departments, and only when the new development can be reasonably construed to be causing the need for these new roads, parks, schools, and fire departments. An added note: the only fee collected may be that which is a reasonable percentage of the new need created (ie: if a new park were to cost $1,000,000 to purchase and develop, and the new development would be expected to utilize 10% of it, then that would set off a chain of limitations, and the remaining cost would be necessarily borne by the remaining taxpayers).
Ultimately, I do support impact fees. As a matter of diversification of revenue streams, where they are applicable, we should look to implement them for necessary system improvements. At the same time, I believe that urban areas should have a ceiling on how much can be charged for impact fees, and that ceiling should be the floor for suburban and exurban areas. Discouraging sprawl is a must if we want to protect our natural environment, while providing transit improvements that reduce reliance on cars.
But I'm also a realist. I think there is a way to identify capital investments in roads that can be done without expanding them, but repaving and adding in some separate funds to improve transit mobility. However, the direct impact to a new structure for this I am unsure would be enough to raise significant funds compared to the existing use. That doesn't mean we shouldn't implement the fee, but we also shouldn't expect new development to pay for capital backlogs that have existed since before a new structure has been built.
This extends to park facilities. Depending on the type of development, there may well be need for a new park, or improvements to an existing park to better meet the needs of the expected new community if it crosses a tipping point for the needs of the existing community. Think playgrounds - if new developments are expected to bring new kids, and the existing park doesn't have a playground, but there are kids in the neighborhood, and the new development's kids will push the total kid population over the mark to justify adding a playground, then a fee for the amount of kids to hit that tipping point may well be in order (should be in order, if you ask me).
Where this tends to get most play: we should use developer impact fees to build more schools. Setting aside the total facility needs of the school district, there is a major flaw with this argument, and it couples with additional complaints about development: there aren't many 2+ bedroom units being built (family-sized units - and if you don't think 2 bedrooms is a family sized unit, GTFO, you classist jerk). The nexus requirement, as codified by the statute, simply do not support using impact fees in Seattle for new schools. And the limitations on housing types - not allowing rowhouses, no duplexes/triplexes in historically SF zones - are constraining the ability to produce more family size units that might be affordable to young families.
There remains a valid argument on this part in particular. If we were to implement impact fees on units of housing reasonably calculated to bring more families with children to Seattle, would that dissuade development of more family sized units? Frankly, I don't think it would. A further review of how impact fees have been used for school construction might shed light on exactly how this interplay has worked. But there aren't impact fees for school construction now, and we're not seeing a boom in 2+ bedroom units. There are other problems at play.
Naturally, some or all of an impact fee will be incorporated into rents or sale prices of homes. Some of the arguments against such fees is that by increasing rents to cover fees, they will only exacerbate the affordability crisis. Seeing as how so many projects in the city aren't being designed to actually be affordable to most folks, I'm not buying it. There remains a significant market for higher-cost homes in Seattle, and that's beating out the market for folks in the 80-120% Area Median Income range. I can't see any collection of impact fees really having a dire impact on those of us in the 80-120% range (or even up to 150%).
That said, I also believe this is something that could be taken into consideration - whether to collect an impact fee on a building that will have a set amount of units affordable for 12+ years for folks in the 80-120% range. Unlike MFTE, this wouldn't be exempting from regular property taxes (and subsequently spreading out the collection to everyone else), and could be utilized as another incentive for more family-size units to be produced, and more production of homes for folks in the 80-120% AMI range.
In sum: everyone loves to talk about impact fees, and all of the politicians are going to promise to use them. But these aren't some silver bullet to finance capital improvements necessary for our city, or to pick up the slack of years of lack of adequate or equitable investment. Should they be looked into as part of a broader exploration into our revenue policy? Of course. And if their collection would bring in more than what it would cost to collect, there should be implementation of policy to do so. But don't let politicians pretend impact fees are a panacea for revenue woes. And if they try to tell you otherwise, they're either ignorant of the limitations, or lying.