Hacker News new | past | comments | ask | show | jobs | submit login
Does Adding Expensive Housing Help the Little Guy? (marketurbanismreport.com)
54 points by DINKDINK on Sept 28, 2017 | hide | past | favorite | 56 comments

If you're interested in this topic, you might also like the recent Econtalk episode, "Philip Auerswald on the Rise of Populism," http://www.econtalk.org/archives/2017/09/philip_auerswal.htm.... The spread of housing restrictions via zoning has had much larger impacts than is commonly appreciated. I find it interesting especially that Piketty's famous results are really a result of capital accumulation to landlords: https://www.brookings.edu/wp-content/uploads/2016/07/2015a_r... who benefit from supply restrictions.

I've been writing about (and to some extent working in) these issues for a while and come at them from another angle here: https://jakeseliger.com/2015/09/24/do-millennials-have-a-fut...

Thomas Sowell is brilliant at highlighting the unexpected nature of many parts of economics as well as the countless government policies initiated in the name of helping the poor or lower class that end up either doing little to help or, more often, ultimately helping the upper-middle class and above the most.

He helped me grow beyond a cliche idealistic libertarian worldview into something much more practical and based in real world policy.

I highly recommend his 'Wealth, Poverty and Politics': https://www.amazon.com/Wealth-Poverty-Politics-Thomas-Sowell...

Or for something more lightweight see his book 'Basic Economics': https://www.amazon.com/Basic-Economics-Thomas-Sowell/dp/0465... --- Despite it's name it's not an economics 101 guide, it basically a teaching-through-example guide by listing policy after policy that were implemented in the realworld, for ex:

- rent control in NYC/Toronto in the 1970s which severely reduced access to affordable housing, disincentivized building maintenance, incentivized arson, and gave countless upper/middle class residents cheap rent for beautiful properties

- various industry licensing pushed by market incumbents not protecting consumers, such as interior designers requiring 4yr bachelors degree to choose the colorscheme of an apartment

- etc, etc

He digs into their good intentions but unintentional self-defeating side-effects. Many of which have countless analogies to today (see Uber vs Black cabs in London).

After reading it I'm hardly surprised Japan's economy has stagnated, to the point where it's almost so obvious it saddens me. They are one of the worst proponents of the type of heavy handed economic-intervention he critiques.

Just adding another plug for Econtalk. They cover a lot of different topics and invite people that really know how things work. I remember listening to a podcast on the housing crisis and walked away with a great understanding of how it all fit together.

The Hoover Institute and Russ Roberts have a neoliberal/libertarian slant, but he does a great job in inviting people to his program that he doesn't agree with, but he asks great questions and gives them an open forum.

This just in: woman who has staked her career on a pro-housing political platform announces that market-rate redevelopment is good for everyone.

The question is not “will adding N market-rate units help?” because that’s trivial. Of course it will. The question is: “what if you have to knock down N affordable units to build M market-rate units?” This “simulation” doesn’t even come close to addressing that question.

Like it or not, that’s the situation in San Francisco.

> The question is: “what if you have to knock down N affordable units to build M market-rate units?”

No it isn't. No one is seriously proposing knocking down N affordable units unless they're also including N+X affordable units + M luxury units in the new structure. Any such plan would be a massive waste of time because it has zero chance of being approved.

This article is addressing protesters against "The monster in the mission!!!!1" who marched against replacing an empty warehouse and a Burger King with 16 stories of housing. It is also addressing people like Dwyer who thought they were untouchable and important enough to work against building more housing without suffering any consequences.

I've noticed some attitudes starting to shift out in Sunset as lifetime SF residents find their kids coming home from college or getting married, yet being forced to live hours away. Suddenly the anti-development stance is less attractive when it means never seeing your grandkids.

Ultimately I don't think the Bay Area can muster the political will to solve the housing problem. Too many people benefit from eternally-rising housing costs or have a "screw you, I got mine" attitude. The state government will have to step in and remove local zoning control.

The Bay Area will never solve this on their own. The only solution is to repeal the cap on property taxes at the state level to force current owners to pay based on assessed value. And that's one hard sell.

Repealing Prop 13 is a transitional measure to a full land value tax, collecting all rents on the unimproved value of land. Then you've really got it solved.

> N+X affordable units + M luxury units in the new structure

How does that work in a way that someone that really needs the affordable unit gets it, versus, say, a thrifty young (well paid) engineer?

Are they held back only for people with HUD vouchers or something? Or is "affordable" just a relative term?

Asking because I've never lived somewhere like SFO.

"Affordable housing" is a very specific term - it means city-subsidized housing that is restricted to tenants within a certain income bracket (and usually applicants are selected by lottery). There are sometimes requirements that new developments must set aside a certain % of units as affordable housing.

It might be a very specific term clearly defined that way by certain government programs or NGOs, but most people using it just mean housing that people with low incomes can afford as a reasonable portion of their income.

In San Francisco it absolutely means that.

There's often a weird negotiation/ battle that goes on between developers and the city over what percent of units in a new building ate affordable vs market rate. It has nothing to do with them being small or Spartan or anything like that, and everything to do with income requirements and rent caps put on them

I think it's not used like that in regards to SF because no one is proposing plans that would make that possible without government rent control.

Most people discussing it on the internet do not actually live in SF. This here is the internet. I just checked. I am not in San Francisco and I am talking about it.

Sure, I don't live in SF either, but afaict SF is the most discussed city with regards to housing affordability.

The comment I was replying to did not specify "in this case" or "in this context" or "when we are talking about San Francisco." Saying that the term is very specific and specifically means X without, yourself, specifying the conditions under which it means X reads as holier than thou, not as let me helpfully clarify this technical term for you.

I actually studied housing issues in college because I was pursuing a BS with a concentration in housing as I wanted to be an urban planner. I am fully aware of the fact that the term is used that way in some situations and stated so in my comment. But, I write a fair amount about affordable housing and that isn't how I use it.

Good communication starts with context. This is the internet. It is not San Francisco.

Fair points, I'm sorry for my somewhat pithy original reply.

There are essentially no development projects in SF that don’t involve demolishing existing housing. While I grant you that some such projects have (ridiculous) community resistance, these still aren’t the norm.

In most places in SF, the resistance comes when housing is being replaced and/or the developer doesn’t want to build low-income units (instead preferring to donate to the low-income housing fund). It’s not a clean story.

In both Oakland and SF, there's been significant resistance to projects that propose to replace things that are not housing at all with things that are. If memory serves, the "Monster in the Mission" was proposed for an empty lot. I saw significant resistance to putting a tower in a literal hole in the ground at MacArthut BART.

In SF, the resistance often seems centered on the idea that however much subsidized housing the developer proposes isn't enough for the interest groups fighting it. Often the push is for 100% subsidized, plus assorted "community benefit" donations to non-profits.

Yup. There is another project trying to get approved that will replace a parking lot and a laundromat (the owner owns the land and says his business has been going down for year).

Jesus Christ, you wouldn't believe the opposition to it. The preschool next door will be shaded for 2 more hours! The proposed mural doesn't align with the community.

Unbelievable! This would bring another 90 units to the Mission with 15 of them for low income residents.

People opposing the tower at MacArthur BART claimed it would block views. Of what? The lovely highway and classic BART station architecture?

Even 100% affordable housing gets objected to: https://sf.curbed.com/2016/10/6/13189882/1296-shotwell-affor...

To be honest, people like to extol the virtues of Bay Area liberalism, but to me the housing policy in the Bay is one of the most systematic (and regulated) form of prejudice. Cities like SF are losing their minorities, their middle class, all to fatten the paycheck of existing landowners. Almost the entire black population has emptied out of SF. It's one of the most pervasive form of class discrimination I've seen.

It comes down to incentives. With the current policies in place, there isn't much incentive for incumbents / existing residents to allow more housing.

If you are a home owner or landlord, reduced supply results in asset appreciation.

If you are a renter with rent control, then building more housing in your neighborhood may not seem like much of a benefit either.

Policy makers are always looking for piecemeal solutions, e.g. let's give teachers a rental subsidy, tweak the % of affordable housing that must be build, etc.

Lawmakers really need to create incentives so incumbents have a reason to want more housing.

  Almost the entire black population has emptied out of SF.
Bayview / Hunters Point is still part of S.F. I have black friends whose parents still live in the same homes for 50+ years.

This is patently untrue, especially weighted by number of units. There may be a lot of small rebuilds of single family homes into duplexes, but the projects that add the great bulk of housing units to the city are usually built on empty lots, parking, or small-scale commercial/industrial. 1515 Van Ness, 1 Oak, the 16th St. BART project, etc.

If people cared about affordable housing, why are there protests again projects which are 100% affordable housing? You can see the hypocrisy here:


"because that’s trivial. Of course it will."

Many in San Francisco (and other cities) argue against this. It's certainly not universally accepted.

"The question is: “what if you have to knock down N affordable units to build M market-rate units?” This “simulation” doesn’t even come close to addressing that question."

That's because housing units are virtually never knocked down in San Francisco, so this question basically never comes up. Almost all new housing is built on warehouses, parking lots, gas stations, and so on. See my analysis of all recent housing built in the Mission District:


“Almost all new housing is built on warehouses, parking lots, gas stations, and so on.”

That may have been true, but the city’s own analysis shows it isn’t true anymore, and it certainly won’t be true by the time we’ve built enough to lower market rents (yet another reason why this “simulation” is nonsense).

San Francisco simply doesn’t have that many vacant parcels of land.

> That may have been true, but the city’s own analysis shows it isn’t true anymore

Alright, this is three times you've said and been called out on the same exact lie. You're going to need to provide some sources, or at the very least some references to sources, to back up your extremely bold and most likely false claims. What specific housing is being targeted for being knocked down for which projects?

Nobody “calling me out” has provided any evidence, so I’m not sure what you’ve been reading, but OK:


By my math, that’s 769 legal units removed or merged between 2012 and 2015.

AKA, it happens all the time. So let’s talk about your definition of “lies”...

Would you link to this analysis please? I hope you will support your claim that new housing is replacing old housing.

Please list which housing developments have destroyed existing housing stock. Here's the catch, you can't because all of it in the past 5 years has been on existing commercial property (gas stations, commercial businesses), so your central argument that market rate housing is taking land from income-controlled subsidized housing doesn't hold.

Side note: >woman who has staked her career on a pro-housing political platform announces that market-rate redevelopment is good for everyone.

What's the point of this statement? Do you equally scoff at groups that delay housing construction to increase subsidized housing units and ludicrous when they advocate that slowing down housing construction benefits certain groups?


Lots of legal housing has been demolished or merged over the past five years. It happens all the time, you just don’t hear about it on pro-development blogs.

I have no particular problem with development, but Sonja Trauss is not an objective commentator, and this “simulation” is little more than opinion dressed up as data.

The question here is about housing demolished in order to build new housing.

Merged units are not in this category. Merged units are when someone buys a duplex (or bigger) and removes all but one of the kitchens in the building and thereby makes one big house out of what used to be a house that was split into 2-4 apartments. That is not an example of housing being demolished in order to build new housing.

Here is the city's 2015 housing inventory: http://default.sfplanning.org/publications_reports/2015_Hous...

It has more specific numbers, like in 2015, the city built 3,000 units and tore down 25. 15 of those were single family houses, which is not a category of housing that is affordable by any measure in SF in 2015. Vastly more affordable units are lost in SF through merger, Owner Move In evictions or just the previous tenant moving out and the new tenant being charged more than by demolition.

It very much confuses the housing debate to believe or claim that every time a new apartment building goes up, it means by default that dozens of low income people were kicked out to make that happen. That is just not true.

You've just linked a chart that you claim shows "lots" of housing demolished or merged over the last five years, but these numbers are tiny, representing far less than a single percent of the city's housing stock.

The vast majority of new housing construction happens without the removal of any previous housing unit.

You can’t evaluate it as a percentage of all housing. By that standard, even new construction is a “tiny” percentage of the whole.

SF doesn’t have much infill development left. That’s the point. Arguing about what people did in the past isn’t relevant today.

You can run the simulation yourself, but the bottom line is if the total number of houses grows, more people get housed in all income groups. That is a tremendously powerful message for city supervisors. They only have to look at one thing "After this change will there be more or fewer houses?" if the answer is 'more' then the change will be positive for people trying to find housing, if it is 'less' then the change will be negative. A very simple and objective proposition.

Say someone proposes to demolish a 100 unit SRO hotel in the Mission and replace with with a high rise luxury condo building with 250 market rate units. There will be a big protest about how gentrification is pushing the people out of their own neighborhood and where will the people in those 100 SRO units move to? etc etc etc.

What this tells you is that after adding an additional 150 houses all income groups will have more housing opportunities period. That might seem non-intuitive but the model works.

The politics in some parts of SF favor people whose response to the simulations will tend to be "That's nice, but what if all of those units were permanently subsidized?"

You can't demolish 100 unit SRO in the mission, rightly so, without a development agreement that would relocate the current SRO tenants into a new building without interruption.

There's no reason to have this kind of thing be an example scenario because that's not how housing gets built. SF isn't tearing down 100 unit SROs, they're tearing down one story commercial.

The “simulation” is garbage. I’ve pointed out one limitation, but there are many others. For example: rental markets are not auctions. Landlords set prices, and renters pay them, or do not.

What happens when you replace 150 SROs with market rate condos? You make 150 of the poorest people in the city homeless, and you remove a tiny bit of competition (maybe) at the higher end of the market. (You’ll note that this post doesn’t discuss who is evicted.)

At a minimum, it is the responsibility of the simulator to show that their simulation reconstitutes reality before it is to be trusted for any form of prediction.

The only “powerful message” here is that Sonja Trauss is running for office on a pro-development agenda.

It is a model, not a simulation, and it is traditional to argue the flaws in the methodology or assumptions of the model if you seek to invalidate. What I heard you to say was: “what if you have to knock down N affordable units to build M market-rate units?”

And the model exactly models this, showing that as long is M > N, everyone will will get more housing. If you run the simulation you can see how it does this, market rate housing captures people who can afford it, they move out of mid-tier housing, which captures the folks who want mid-tier, which frees up affordable units that were being occupied by people who could move up.

They're removing subsidized houses to build market-rate ones? How does that work? Is it the same owner demolishing their own house and replacing it - surely a violation of the conditions they had to agree to to get the subsidy. Why would the government pay a subsidy, then allow the very thing the subsidy was supposed to prevent? Does that money get refunded to the government - maybe allowing them to spend it on another subsidized house somewhere else?

Not “subsidized”...affordable.

A landlord has 3 older units that rent for $x. She wishes to redevelop the property to create 6 new units that rent for $3x. Net result is that 3 units have been created, but 3 people have been displaced. Likely permanently.

This is the heart of the redevelopment debate in SF.

$x is a rent controlled price? By rebuilding, she loses the requirement to keep the rent low? You're right, that's not a subsidy since it's the landlord that pays for it.

In most cases, when you improve a property, you then charge more because it is improved. It isn't unreasonable on an individual basis, but the end result is that affordable units disappear and do not get replaced. Rinse and repeat and pretty soon there simply aren't any affordable units.

Subsidized is an accurate description of the financial arrangement surrounding the unit in a way that "affordable" is not. Why not use it?

I know, I know. Politics, right?

Yeah but people are moving out of less desirable and housing to move there, opening up new housing down market.

In SF M >> N, to the extent you can assume N =~ 0. The results hold.

There is no reason to assume that building new housing requires tearing down affordable units. That's the point of using the word "adding".

You can extrapolate from her result, that if M is significantly greater than N then it helps.

No, you can’t. You displace people, who then have to look for housing. Moreover, there are hard floors on new construction rent. It’s not linear, and the “simulation” isn’t even remotely accurate.

If you replaced every low income unit in SF with twice as many market-rate apartments tomorrow, you would raise those individuals’ rent to some value near market rate.

> If you replaced every low-income unit in SF with twice as many market-rate apartments tomorrow, you would raise those individuals’ rent to some value near market rate.

Assume there are 100,000 rental units in SF and 50,000 of them are low-income units. If overnight you replaced those with 100,000 units that will float to market rates, you are correct that the displaced individuals would now have no units available that are significantly below market rate. However, the market rate will not be the same as what it was previously. It will go down based on a 50% increase in supply. From an economic standpoint, the same thing should happen regardless of what percentage of low-income housing is being replaced. Whether or not this goes down to the artificial price they have in low-income housing depends on how much more supply is being introduced into the market.

> there are hard floors on new construction rent.

If you are saying that the price set for "affordable" housing is so low that new construction couldn't even break even at that price, then there is a whole different set of problems involved in addition to supply and demand.

From the linked "Residential Nexus Analysis":


>The prototype rental unit, also drawn from the Sensitivity Analysis work program is also 800 square feet and rents for $2,500 per month or a little under $3.20 per square foot per month. New rental units are not feasible in today’s market; however, the inclusionary program will be in place beyond the current market cycle and must anticipate development of rental units in the future. The assumed rental rate is higher than is achievable in the current market except under extraordinary circumstances (luxury projects in premier locations, etc.). The rental rate has been estimated as the required minimum level for a project to be feasible, given total development costs, conventional financing terms, and typical operating expenses. The household living in this unit is likely to be paying approximately 30% of income on rent (not including utilities). This translates to a household with a gross income of $102,000 per year.

This is a rather interesting claim that I haven't heard come up in this debate before. I'm not sure if it's just because it made more sense to sell condos in 2007 (before November) or because construction costs are actually so high you can't build apartments that cost less than $2500/month and/or $3.20/sf.

It seems that research into cheaper building types, and regulatory initiatives to permit those, ought to be part of a solution. Although since the majority of the value of a property in SF is often the land, a land tax might be called for as well.

Doesn't assuming that demand is exogenous ensure that their model will return the result that they want? It seems that how exogenous demand is is the central question in the housing debate; if SF is so desirable that building more luxury apartments just brings in new rich people, then this model is useless.

(I think, and hope, that it's not. But I think the author danced around the central question by building this model.)

I think Sonja's study does more than "prove a point" - it also illustrates exactly _how_ "luxury" housing helps everyone, which is a huge sore point in local politics. Understanding of the nature of how flexible preferences and different levels of wealth interact is still a useful exercise.

It would be very difficult to devise a study to prove whether exogenous demand is real, but a casual examination of San Francisco suggests that it is not - SF has built very little housing and still consistently out-attracts many other metro areas.

Yes, exogenous demand is the central question in the housing debate.

That's one of the uses of a model, to think about one aspect of a process at a time.

Applications are open for YC Winter 2022

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact