Category Archives: Uncategorized

Is Union Station Obsolete?

This question has been bothering me for a while. The momentum behind current plans is so large that it’s probably just a philosophical question, but I still think it’s worth examining in detail both for LA to think about what we’re trying to do and as a case study for other cities that might be considering transit investments.

An Historic Mistake

A fair amount of money has already been invested in improving transit at Union Station, including the Red Line, the Gold Line, the bus plaza, the Silver Line, and Metrolink. An even larger amount of money is planned to be spent soon, including run-through tracks for Metrolink and Amtrak, high speed rail, the Union Station Master Plan, and other potential investments like the north end of the West Santa Ana Corridor.

There’s just one problem: Union Station is in the wrong place. By the time Union Station opened in 1939, downtown LA’s center of gravity had already shifted south to the Spring St Financial District. Various redevelopment schemes have tried to pull it north – notably, the development of the Civic Center (roughly bounded by the 101, Los Angeles St, 2nd St, and Figueroa St) in the 1950s and urban renewal on Bunker Hill (roughly bounded by 1st St, Hill St, 5th St, and the 110) from the 1960s through the 1980s. However, these schemes have also moved downtown’s center west, away from Union Station.

It’s also worth remembering the reason Union Station was already in the wrong place when it opened: part of the scheme was a racist plan to demolish a low-income Chinese immigrant neighborhood. The station replaced terminals operated by the ATSF Railroad at 2nd & Santa Fe and by the SP Railroad at 5th & Central. These were also inconvenient locations relative to LA’s central business district (CBD), but no less so than the current location, and the SP’s location was arguably more convenient to the Spring St Financial District.

Today, downtown’s centers of activity continue to move and south and southeast. The construction of LA Live and expansion of the convention center has made Figueroa and Olympic a busy intersection, with many hotels in the area. Residential development is taking place in many parts of downtown, but the most intense development is south of 7th St to about Pico Blvd. The Arts District and Fashion District have emerged as growing office districts despite relatively poor transit access; for example, the City Market development will be well east of the Blue Line and Red Line.

Part of this problem will need to be addressed with better rapid transit to those districts, but with renewed focus on improving Metrolink, it’s worth reviewing the best practices for commuter rail and how Metrolink might better serve downtown’s transit needs.

What Should Commuter Rail Do?

Commuter rail serves longer trips to the CBD than rapid transit, reaching deeper into suburban areas. The best systems compliment the city’s rapid transit network, and do not rely on the rapid transit network to serve last mile trips on the downtown end of the trip. This means that the terminal should be in walking distance of large employment centers downtown. Both in the US and abroad, significant investments have been made to improve commuter rail in this regard:

  • Paris developed the Regional Express Rail (RER) network, with tunnels through the central city to connect lines that formerly stub-ended at terminals towards the city’s edges.
  • Seoul Subway Line 1 operates as a hybrid RER/rapid transit type system, running through the heart of the city with closely spaced stops but reaching far into the suburbs with larger stop spacing.
  • San Francisco is preparing to spend a considerable sum of money to extend Caltrain service from its fringe location at 4th & King to the centrally located Transbay Terminal.
  • Berlin connected its regional rail through the city in a system known as the S-bahn.
  • Philadelphia connected its regional rail network with the Center City Commuter Connection, allowing riders to exit in the city’s CBD at Suburban Station and Jefferson Station, rather than the peripherally located 30th St Station.
  • Boston’s South Station is located adjacent to the CBD, with many riders able to walk to their final destination, and many lines serve Back Bay, a subsidiary business district. North Station is located further from the CBD, which is seen as a weakness, and the state has intermittently pursued plans to connect North Station and South Station with a tunnel. This would allow north side riders to exit at South Station or Back Bay, alleviating the need for a transfer to the subway.
  • Despite Penn Station and Grand Central Station both being relatively walkable to Midtown Manhattan, many transit advocates consider it to be a serious planning failure that ARC and now Gateway do not plan to connect the two stations.

Viewed this way, LA may be about to embark on a plan to invest a lot of money in upgrading Metrolink and Union Station, without resolving a fundamental weakness: Metrolink relies on the Red/Purple Lines and Regional Connector to make last mile connections downtown. This increases crowding on the rapid transit network, and is an inefficient use of vehicle capacity, with peak loading driven by a large number of passengers who ride the system for only one or two stations.

Los Angeles RER

A plan to increase Metrolink frequency should look at options to resolve this weakness. Fortunately, it does not need to be done at the same time, and could be done in the future if desired, though as time passes the sunk costs of investments in Union Station are going to increase.

The central feature of such a plan would be a tunnel from the Ventura/Antelope Valley Lines north of Union Station through the CBD and connecting to the Orange County/91 Lines somewhere near Redondo Junction. There are many potential route options but the general destinations to be served should include the Civic Center, Bunker Hill, the Financial District, LA Live, and the Fashion District. Five stations would likely be too many, but these destinations can probably be served by three stations.

From Mission Junction, the tunnel would run generally southwest towards the Civic Center, and then follow one of downtown’s north-south arterials. All of them between Figueroa St and Spring St should be analyzed to determine the best option, excluding Flower St and Hill St, where the cost of going parallel and beneath the Regional Connector or Red Line would likely be too high. Figueroa St or Hope St would serve the Financial District and LA Live best, but provide poor connections to the Civic Center and places east of Hill St. The opposite would be true of Broadway and Spring St. Grand Ave or Olive St may prove to be the best; well-placed stations could provide good connectivity to all destinations and the rapid transit network. A Civic Center/Bunker Hill Station between 1st and 2nd could connect to both the Red Line and Regional Connector. A Financial District/LA Live Station between 8th and 9th could provide underground connections to 7th/Metro.

The tunnel would then turn and run southeast to Redondo Jct. The options here are less numerous and may have less appealing trade-offs. Everything between 9th St and Washington Blvd should be studied. 9th St offers the most direct route to Redondo Jct and provides a good location for a Fashion District Station around 9th & San Pedro. However, the further north the turn east is made, the less convenient the route would be for serving LA Live. 11th or 12th St might be a good compromise, as it would allow the turn to be further south while still allowing for a good Fashion District location. The location of the turn might be tricky with all the development going on in the area now. Washington Blvd would make it possible to connect to the Blue Line, and would offer the possibility of grade separating the Blue Line at the same time, but is too far south to adequately serve the Fashion District.

Below are some general concepts that show possible locations for the tunnels and stations.


In a future post, I’d like to look at the branches of Metrolink in more detail and discuss what makes sense for each. For now, at least consider the possibility that Union Station may not be the right place for LA to concentrate transit investments.

Update (9/26/18)

Per some good comments on the post and Twitter, here’s a revised map, that has the tunnel serve Union Station (a much better plan) and shows the proposed Union Station run through tracks (pink dashed line). Also, here’s a link to a larger version of the original graphic and the revised graphic.



Zoning and Immigration Restrictions

The impact of California’s high housing costs on refugees has rightly gotten some play in the media lately. CALmatters ran an article detailing the challenges faced by people trying to earn a living in a new country with high rents looming over their heads.

The inability of California to welcome more refugees due to high housing costs highlights that the crisis arises not just from inadequate tenant protections, but from a lack of supply. Tenant protections and rent stabilization will help people who already live here, but don’t do anything for people who are trying to escape desperate situations elsewhere.

Many Californians like to think of our state as welcoming to refugees, but we can’t fulfill that ideal unless we build more housing and make the state more affordable. And the local policies that prevent California from being affordable spring from the same poisoned mindset as the Trump Administration’s immigration policies, devised by Steven Miller, a SoCal son we’d all like to disown.

Research by Jessica Trounstine, associate professor of political science at UC Merced, finds that whiter neighborhoods are more supportive of restricting development, and that cities that were whiter in 1970 are more likely to have restrictive land use in 2006. As racial resentment drives whites fearful of demographic change to support Trump’s cruel practices and proposed cuts to immigration at the national level, so it drives the push to restrict development at the local level.

The research concludes that policies restricting development are effective at maintaining segregation. They are, in effect, border walls put up at the edge of neighborhoods. So if we want California to be welcoming to refugees, we need to change the exclusionary land use policies that were enacted to keep people out.

Who Can Telecommute?

Telecommuting has been held out as a solution for America’s transportation and land-use woes for my entire adult life and then some. Ever since the rise of the internet and the first tech bubble of the late 90s, it’s been proposed that eventually, the technology would be so good that you do your job from anywhere. That way it wouldn’t matter if Los Angeles were expensive, because you could just live in Spokane and do your job from there.

The experimental evidence we have from the last 20 years does not support this theory. An organization that promotes telecommuting found that between 2005 and 2017, the number of people working from home at least half time in the United States rose from 1.8 million to 3.9 million, or 2.9% of the  US workforce. While that number is not nothing, it effectively represents only a 1.5% reduction in the demand for transportation for work trips. Over that same period, total employment in the US rose by 12.5 million, or 8.5%. So despite the growth of telecommuting, overall employment grew much more quickly and consequently the demand for transportation continued to increase.

The continued success of large US metros relative to mid-size and small metros also suggests telecommuting is having little impact. A recent spate of “millennials are doing X” articles have tried to push the idea that growth is now shifting to mid-size cities, but it’s amazing how many articles there are citing Pittsburgh, when Allegheny County has lost population every year since 2013. In fact, given how high housing prices have had to rise in large metros to force growth elsewhere, I wonder if you could make a case that the internet has actually increased the agglomeration effects of living in a large city.

Finally, while there are clear benefits of telecommuting (lower housing and transportation costs for the employee, lower office and overhead costs for the employer), there are productivity costs as well. I’ve found in my personal experience that it’s much easier to supervise employees and coordinate a large project team when everyone is in the same office. Being in different offices in the same city is worse, and being in different time zones is even worse.

While I’m not a great manager by any stretch of the imagination, I would like to think I am at least marginally competent and that my experience in this regard is not unique. The single biggest benefit of being in the same office as other people is that it makes it impossible for them to avoid you. Emails can go unanswered, calls can go unreturned, but if I show up at your desk asking you about something, you have to respond. This is a pretty common thread in human relations – it’s much easier to keep up and keep in touch with people that you see than people you don’t.

Finally, there’s the simple fact that living in a bigger city give you more amenities. You might be able to do your job from an RV in rural Nevada, but you won’t be able to access the amenities of Las Vegas or LA.

Even throwing all of that aside, it’s worth asking who could theoretically telecommute. Chances are, if you’re reading this blog or City Observatory or Curbed or City Lab, you have a job that you could telecommute to do, or at least have a lot of friends who do. But that is not a very representative sample of American employment. The five biggest major industry sectors according to the BLS are professional and business services (20.1 million jobs), state/local government (19.4m), health care (19.1m), retail trade (15.8m), and leisure and hospitality (15.6m). My guess is that if you’re reading this, you work in professional and business services, easily the category that best lends itself to telecommuting, and is still subject to all the problems above. Some of these sectors, like health care and retail trade, have almost no potential for telecommuting, as do other major sectors like manufacturing, construction, and transportation/logistics.

Let’s do a quick analysis on the major industry sectors with low, medium, and high guesses for the percentage of people that could telecommute. All numbers are BLS 2016 data.


At the low end, something like 10% of people will be able to consistently telecommute. That suggests we’ll see more growth in telecommuting, but it still won’t have much impact on transportation demand or housing prices. Even at the high end, if 25% of people were able to telecommute tomorrow, that would put the number of employees not telecommuting at about the same as total employment in 1993, and of course, the economy is still growing.

None of this is to say that telecommuting is per se bad, or that you shouldn’t do it if it works for your personal situation or your business. But if you’re counting on telecommuting to fix traffic and housing prices in places like Los Angeles and San Francisco, it’s not going to work. We should be planning cities for how people want to live, not hoping for technological solutions to bail us out of the problems created by a few selfish actors.

South Bay Boulevards

Though I live in Glendale now and have written extensively about Palms, one of my favorite parts of LA County is the South Bay. Development and land use patterns in this collection of under the radar cities like Gardena, Torrance, and Lawndale are genius – though perhaps we should say accidental genius.

For readers outside of LA, the South Bay is roughly the area south of the 105 and west of the 110 – though Carson extends east of the 110 and is South Bay in my mind, and the portions of the “beach cities” (El Segundo, Manhattan Beach, Hermosa Beach, and Redondo Beach) west of PCH may or may not be “South Bay” depending on how you conceptualize things.

This area was one of the first prime areas of post-war suburbanization in LA, and has the features you’d expect of such a place: great climate (what people usually think of as SoCal’s climate), and centrally located with great access to both employment and amenities like the beach. In fact, the South Bay is the Boomer-era mythos of SoCal, created and popularized and immortalized by the Beach Boys out of Hawthorne.

So what do I love about the South Bay? Its anonymous and eponymous boulevards, and the surprising amount of low-rise density like dingbats and two-story podiums, that create wonderful diversity and make it possible for many, many different people to live life and pursue their dreams. One theme of this blog has been a celebration of Palms, but much of the South Bay is Palms on a grander scale. And as Palms is one of the most diverse neighborhoods in LA (and especially on the Westside), the South Bay is one of the most diverse parts of LA.

South Bay Boulevards

Start at the beach and cruise east on any of the South Bay’s boulevards and just take it all in. The diversity of these cities is reflected in the diversity of business establishments you’ll see as you go. And you can go for miles and miles. I recently got the opportunity to do this on Rosecrans Ave and on Manhattan Beach Blvd, and earlier this year I did it a few times on Artesia Blvd and Western Ave. In the past when I lived in Palms and my partner lived in Torrance, I spent a lot of time on Hawthorne Blvd and Torrance Blvd. But you can picky any of them and have the same wonderful experience. I love how far you can go and go and just keep going, looking at each small business and each apartment building, each one representing a person getting to try to make their own way in Los Angeles.

Naturally, the South Bay is poorly understood, its density overlooked by aesthetic and urbanist observers who focus on its auto-orientation and single-family neighborhoods. To them, I would say, I implore you – look at how many people the South Bay is working for, and ask yourself, what can we do to make more places like the South Bay? And what can we do to make the South Bay work better for more people, and make sure lower income people don’t get priced out?

So why did I call the South Bay accidental genius?

Well, if you go back to the history of housing and development in Los Angeles between 1945 and 1965, it’s pretty clear that they were not trying to create medium-density integrated suburbs. Pretty much the opposite, in fact. They created land use patterns that enabled the South Bay’s diversity by accident, despite their efforts to the contrary, and you could probably interpret some of LA/OC’s suburban fringe development between 1965 and 1990 as an attempt to “fix” what didn’t work about those efforts.

Looking at the South Bay’s land use patterns, what the planning was, what the intent was, and what actually happened can provide good lessons on what we should do with current land use planning. It would not be good to say we want to “go back” to the planning regime that created these land use patterns, because that regime was discriminatory against people of color. But in the post-war era, many schemes were tried to make and keep neighborhoods white, and some of them “worked” much better than others. Very low density schemes produced neighborhoods that today, 70 years later, are much whiter than the high density schemes.

No one should be under the delusion that exclusionary zoning is what caused racism; the causation, of course, runs in the opposite direction.

However, you also should not be under the delusion that undoing the exclusionary zoning policies that most successfully perpetuated racism and segregation would not help, or that it would not be worthwhile on the merits. We have enough history to know which policies keep neighborhoods segregated, and we ought to make those policies nothing but history. The South Bay may not have evolved in a way originally envisioned by post-war suburban planning, but it offers many ideas on land use policies that could be incorporated into a vision for a denser, more progressive, more inclusive Los Angeles.

In Praise of Remodeling

Just a short note that I was reminded of recently when I came across a project in LA that proposes to convert a manager’s office into a one-bedroom apartment in an existing building: subdividing old housing is very good, but it has been curtailed by zoning.

Reconfiguring buildings that already exist is one of the cheapest ways to create new housing units. Unsurprisingly, it is one of the primary ways that affordable housing was created in the past. Single-family housing was converted into boarding houses. Once-fashionable homes on Bunker Hill in downtown LA became cheap apartments, providing many affordable housing units until they were demolished for urban renewal. Row houses that were single family became one apartment per floor, or several apartments per floor, or even single-room occupancy units.

Now, this wasn’t the only way affordable housing was built in the past. Many buildings were purposely built as market-rate affordable housing, from LA’s dingbats to New York’s tenements. But remodeling an existing building is likely going to be cheaper, for structural reasons you can’t really change. This includes:

  • The building already exists, so its capital construction cost has probably already been recovered, and the amount of construction needed is smaller.
  • Because you don’t have to demolish an entire building and start from the ground up, the units can be delivered more quickly, and the carrying cost is lower.
  • Because fewer resources and skills are needed to do this type of construction, more people can do it. This lets more people participate in development and makes it easier to finance.

All of this means that new units in existing buildings can be delivered at lower cost than new units in new buildings. That lets housing supply be more responsive to prices, making new units viable at lower rents.

Of course, all of this is intuitively obvious to a lot of people in crowded but disinvested neighborhoods. People don’t have to work out the theory to see that there are other people who need housing, and that they can provide it by remodeling space in a building they already have. That’s why there are so many unpermitted units all over Los Angeles, from garage apartments to unpermitted additions to illegal subdivisions of commercial structures.

And of course, rather than seeing this human ingenuity as a way to solve the housing shortage, we send out code enforcement to demolish these units when we have nowhere to tell the people who live there to go. It’s like making pushcart vending illegal and then telling people to go stand in an unemployment office line instead.

The City of LA’s Unpermitted Dwelling Unit (UDU) ordinance was a huge step forward in this regard, because it acknowledges that “these units add much-needed affordable housing to the City of Los Angeles.” The UDU ordinance allows a unit to be legalized, provided that they pass plan check and correct any building code deficiencies, and that one unit on the property is dedicated as an affordable unit for 55 years.

However, for political reasons, the UDU ordinance only allows this for units occupied prior to December 10, 2015. If you could build a new illegal unit and then legalize it through the UDU ordinance, this would create an incentive for people to start building more illegal units. But of course logically, if the units are providing “much-needed affordable housing” – housing that is so badly needed we are willing to legalize the ones that exist – the problem is not that people would build illegal units, it’s that the units people build would be illegal!

UDU ordinance aside, subdividing old housing into multiple apartments generally doesn’t fit into the modern conception of planning. Neighborhoods are planned to be built once and then never change, regardless of what social and economic conditions change around them. If you happen to own a single-family home in a multi-family zone, a very rare situation in LA these days, you could do this. But most places in coastal California are already built out to their maximum permitted density (or in many places, more than their permitted density due to downzoning).

It’s a shame, because upzoning would probably lead to not only an increase in new building construction, but to remodeling of existing buildings. By not allowing it, California is leaving some of the most affordable potential housing units on the table.

Missing Middle Musing

I recently had the opportunity to travel to both Philadelphia, a city I don’t know very well, and Boston, a city I lived in for 12 years before moving to LA in 2012. In Philadelphia, I was fortunate enough to have @riccoja show me around Northern Liberties and Fishtown; in Boston, I spent a lot of time walking around the city looking at how things had changed (or not changed) since I left. (I also visited DC, but really just the tourist thing and didn’t get to see much new development.)

I came away from my trip with two main thoughts on housing in Los Angeles:

  • We have systematically made it harder and harder to build missing middle (link) housing in LA. This is unfortunate, because I don’t think we can solve the housing crisis without building a lot more of it.
  • The homelessness crisis is the most visible and critical facet of the housing crisis. Even being away for just 10 days, it was shocking how bad things are in LA when I came back.

Row Houses and Ridiculously Narrow Streets

For historical reasons I don’t have the time to research or go into here, Philadelphia seems to have a strong development cultural emphasis on single-family housing, regardless of how small the lots or buildings are. Philadelphia is also cool with really narrow streets. This leads to a pattern of narrow streets with narrow two to four story row houses fronting right on the property line.

Unlike other American cities, it’s still possible to do development like this in Philadelphia, and there’s a ton of infill row house development in Northern Liberties and Fishtown. The only other place that allows townhouses like this is Houston, I think, and it’s certainly not on streets this narrow!

My biggest impression from Philadelphia is that LA needs to unleash this type of development. The virtues of small scale development like this are many:

  • It doesn’t require lot assembly, so there’s no delay for that and never any need for redevelopment agency shenanigans.
  • It’s cheaper per unit to build, because there is less common area and amenity space than a large building.
  • It’s cheaper per development to build, which means a larger number of people can build them. This both increases housing supply and increases community input into how development occurs. It also makes it more likely that small local developers, who know the community well, can participate.
  • It’s less complicated to build than a large building, which means a larger number of contractors and construction workers can be employed in its construction. This increases employment opportunity and lowers costs.
  • It’s faster to build than a large building, which means it can respond to housing needs faster than large developments. This also lowers costs because carrying costs are directly proportional to the time between when the developer buys the lot and when the building opens.
  • It creates small retail spaces, which increase the diversity of local business and creates opportunity for a wider range of small entrepreneurs.

All of this means a faster, larger, and cheaper response to the need for new housing. This is both good on the merits and good politics, because slow and expensive responses to the need for housing lead to more widespread disillusionment and agitation for radical action.

New Districts and Static Neighborhoods

In contrast, development in Boston seems to be focused in a handful of new districts: the Seaport, North Point, Harvard’s campus in North Allston, Station Landing, the area near North Station vacated by demolishing the Central Artery, Assembly Square, and Kendall Square. There was a little infill development (a couple buildings in the North End, a house in Brighton Center whose architecture my friends complained about) but it seemed to be the exception.

Now, there’s nothing wrong with replacing parking lots and abandoned rail yards with new neighborhoods. In fact, it’s pretty great. But it almost necessarily involves a lot more central planning from the city, big institutional developers, and big institutional finance. That means projects that move much more slowly, have a higher cost per unit, and are less likely to happen before the next market downturn. North Point is great… but it was one of the first projects I worked on when I lived in Boston after graduating, and it’s just getting built over a decade later. The same could be said for Assembly Square, Station Landing, Harvard’s campus, and the Seaport. These projects will happen when demand is high enough but they take a long, long time.

My friends in Boston are white collar professionals and public employees. They’re not housing insecure by any means. But they didn’t hold new development in Boston in high regard, and it’s somewhat understandable. It’s not really for them; it’s for people just a little higher up the income ladder. Comparing to Philadelphia, I can’t help but think that Boston would be better off if it were also building a crap load of small scale development like townhouses and small apartment buildings in places not quite as central as the new districts but still T-accessible.

Golden State Squalor

Returning to LA, I was more convinced than before that we need to allow more missing middle development here. If it were up to me, I would rezone every RD, R2, and R1 zone in the city to something like an “RD1.2” or “RD1” townhouse zone, with townhouses allowed by right with a minimum lot area of 1,200 SF or 1,000 SF. (I’d also eliminate parking requirements and reduce the impact fees.)

But more immediate, and more appalling, was returning to LA and walking around downtown, bearing renewed witness to the tide of human misery that floods downtown every night, retreating east of the double yellow line on Los Angeles St and south of 3rd St during the day, a thousand sidewalk washers in its wake.

I didn’t notice the absence of homeless on the east coast, but I noticed the presence when I got back, and it’s a complete embarrassment to LA and California that a state so rich should have so many people living on the streets. The numbers are overwhelming, the state of public health on the streets is atrocious, and yet still, proposals to relieve the situation face NIMBY opposition.

It’s hard to put into words how bad things are in California compared to other US cities, so let me not mince words in my final thoughts. We need to be out there trying to do something every day to fix this problem. And if you oppose new housing, especially permanent supportive housing or temporary shelters for the homeless, you are a bad person, and I hope you feel bad about yourself, because you should. I’m not a religious person but if this crisis doesn’t move you, by god, what would?

Impact Fees Are Bad, Taxes Are Good

One of the more nefarious impediments to housing construction in California, in my opinion, is the proliferation of impact fees on new development. Impact fees first became widespread in the 1980s, a consequence of Prop 13 bankrupting local governments’ ability to pay for public improvements like parks and schools. The state allowed cities and counties to start charging community district fees, also known as Mello-Roos fees, whereby each unit in a new subdivision pays a fixed amount to cover the construction of new public facilities. These fees only apply to new construction in a particular district.

Other development impact fees, charged on all new construction in a city regardless of location, have also been expanding. For example, the City of LA charges a park (or Quimby) fee on all new construction to pay for new parks in the city, and recently enacted an affordable housing linkage fee, where all new construction pays into the city’s affordable housing fund. Impact fees are popular among conservatives and progressives alike, the former because it keeps taxes lower on a majority of voters, and the latter because it symbolically punishes people that progressives don’t like, such as developers and the people who can afford new housing.

Unfortunately, impact fees are bad. They are a regressive and inefficient way of funding improvements, and they unintentionally hurt many people as a result. People who support progressive outcomes should oppose impact fees and support general broad-based taxes instead. Let’s take a closer look at why.

Impact fees often form a large portion of the cost of new housing. For example, a recent study of impact fees found that some cities, such as Irvine and Fremont, have impact fees of over $60,000 per multi-family unit. That translates to about $450/month in higher rent  — just to pay for impact fees.

Progressives tend to not care about this, because they view these taxes as falling on big bad developers and on the people rich enough to afford new housing. However, by raising the price of the cheapest new construction, it enables landlords to raise rents to higher levels without any competition from new construction. Since the vast majority of housing is existing units, and there are relatively new units, impact fees are in effect taxes on renters for the primary benefit of landlords.

Here’s a visualization. Supposed that every year, you have one new housing unit built for every 50 existing units (an optimistic assumption in California). For simplicity’s sake, assume all housing units are equal, and the city has 10,000 housing units. If new construction rents for $2,000/month, prices for existing units will not rise higher than $2,000/month because they will be undercut by new construction. This year, we would expect 200 new housing units to be built.


Now, let’s impose a $20,000/unit affordable housing impact fee on new construction. The cheapest new construction will rent for now is about $2,150/month. That allows landlords to raise the price for existing units to the same level. The net effect is that we raise $4,000,000/year for affordable housing construction (enough for about 8 units at California prices) and we increase landlord profits by $18,000,000/year ($150/month x 10,000 units * 12 months/year). So our “impact fee” is really a 7.5% tax on renters, and we give over 80% of the revenues generated to landlords. Is this really what we wanted to do?


A better approach would be to enact a much smaller tax on everyone that would generate the same amount of revenue. In this example, a 1.67% tax on rent on all units would generate the same amount of money for affordable housing, and none of the money would go to landlords.


Of course, having made it clear that the cost is actually being paid for by renters, the next logical question is why we should be taxing housing in this way at all. Now, this model is clearly an oversimplification; dynamic effects are hard to predict, but it stands to reason that at least some of the cost will effectively be paid for by people selling land, since higher costs of new construction will make people less likely to build new housing, lowering the demand for land. And taxing all units does not lead to the perverse outcome of rewarding landlords for doing nothing.

But it is still logical to conclude that if you increase taxes on new housing, the cost of housing will rise. It is not revolutionary economics to say that if you want more of something, you shouldn’t tax it. Increasing the cost of new housing during a severe housing shortage is just poor policy.

Really, tying taxes or fees for public improvements to new construction in this way is not progressive at all. There’s no reason funding for affordable housing has to come from anything to do with development. The progressive way to do it would be to tax things that are bad and use the revenues to fund things that are good. A carbon tax whose revenues are used to develop affordable housing would be great. So would a congestion tax used to develop affordable housing. A broad property tax that applies to everyone would also be fine. But impact fees should be avoided if possible, unless the goal is increasing landlord profits.