4 ms·
The biggest problem with density is _cost_. Even if we "liberalize housing regulations", density in America still means housing prices 2x-4x higher than current
by maxsilver 10y ago
The biggest problem with density is _cost_. Even if we "liberalize housing regulations", density in America still means housing prices 2x-4x higher than current averages.
Wages can't support that -- most people can barely afford housing today as it is. Even higher tech industry wages can't support that in many places. (Even in my small-town-Midwest city, tech wages don't cover "downtown" cost of living)
Until we get serious about planning density for people who work for a living, this problem will only get worse. If you just let cities get bigger on their own with "less regulations", developers will build cities full of penthouses (as they do today). Sure, millionaires will save some cash, but 99% of the population will still forever be priced out, and therefore be driving in from elsewhere
- woah 10y agoYour comment makes no sense. You are telling us that increased supply drives up prices?
- ArkyBeagle 10y ago"Land rents" are the increase in value of a plot of land caused by its location. As more people fill an area, network effects raise production. This means people have more money to spend on land to have access to this network. So yes. Land Is Different.
- maxsilver 10y agoYes, because housing isn't just about any "supply", it's the right kind of supply. In car terms : If companies only built Lamborghinis and Ferarri's, cars would never be affordable for regular people. Even if you built 1 million of them, you'd never end up with the Yaris or Camry or other regular cars that the market actually needs. Increased supply can drive up prices, if your increasing the wrong kind of supply. This is the same problem we see in cities today. Developers build exclusively super-ultra-high-end housing. If we loosen restrictions, Developers will just build more exclusively-ultra-high-end-housing. Sure, your wealthy folks will save 5% on housing costs as supply increases and the cost of "ultra-high-end" housing drops. But the 95% of regular folks will still forever be priced out of the market. It will take 60+ years before ultra-high-end housing becomes affordable enough for a regular person to afford it. To solve the gentrification problem this article mentions ("to stop gentrification steadily erode opportunities for ordinary, middle-class people") we have to start by building urban housing for ordinary, middle-class people. Something no city in America is currently doing.
- aninhumer 10y ago>If companies only built Lamborghinis and Ferarri's, cars would never be affordable for regular people. This is nonsense. If companies produced exclusively luxury products in excess of demand, they would become affordable because that's the only way they'd be able to sell them. Of course that wouldn't be profitable, so they don't do that. Houses aren't expensive because they're "luxury", they're expensive because they're scarce. Building "urban housing for ordinary, middle-class people" doesn't help much if there are still lots of rich people who want to live there.
- hx87 10y ago> If companies only built Lamborghinis and Ferarris, cars would never be affordable for regular people. The question is then /why/ would companies only build Lamborghinis and Ferraris when there is such a huge market for Hondas and Fords. To extend your analogy, if all manufacturers were collectively allowed to build only 10,000 cars a year, and each one of those were subject to neighborhood review and potential lawsuits, /of course/ they will build only luxury cars!
- maxsilver 10y ago> The question is then /why/ would companies only build Lamborghinis and Ferraris when there is such a huge market for Hondas and Fords. Because the market for high end is much more profitable. The question isn't "how do we ensure builders are profitable as possible". We know how to do that, everywhere does that every single day. The question (from the article) is "how do we prevent cities from eroding opportunities for ordinary, middle-class people". My answer is "we need to just start building housing for ordinary middle-class people". Everyone wants to frame this in economic terms that make it rational for a for-profit company to do this. That seems silly to me, we all already agree it's not the most profitable action. But that's ok -- a city government's goals are supposed to be in the interest for it's citizens and region, not in the interest of profitability. > if all manufacturers were collectively allowed to build only 10,000 cars a year, and each one of those were subject to neighborhood review and potential lawsuits, /of course/ they will build only luxury cars! This assumes everywhere is San Francisco. Even where there is no building limits, there is no shortage of land, and there's comparatively no review process -- density is still 2x-4x the cost. Even out here, developers make so much more money selling 1 high-end unit, than 5 mid-range units, that the vast majority never bother with low-end. Most "free markets fix everything" arguments hinge on this future hope that builders eventually exhaust the high-end market and look to the working-class market. I don't think that will ever happen -- every builder in the nation could continue building high-end-only indefinitely and still never run out of work in our lifetimes.