3 ms·
What are the factors preventing real estate developers from buying up land and building larger complexes to meet the demand? Seems like an opportunity to make
by cgb223 9y ago
What are the factors preventing real estate developers from buying up land and building larger complexes to meet the demand?
Seems like an opportunity to make a lot of money
- deadmetheny 9y agoZoning laws that make it illegal to build tall buildings and insane NIMBYism from property owners who benefited from Prop 13.
- santaclaus 9y ago> What are the factors preventing real estate developers from buying up land and building larger complexes to meet the demand? Local property who oppose new development in order to restrict supply, primarily.
- johan_larson 9y agoHere's a good summary: https://techcrunch.com/2014/04/14/sf-housing/ https://techcrunch.com/2014/04/14/sf-housing/ Basically, there are a lot of regulations governing what you can build and where. And those regulations have a constituency that benefits from them, namely everyone who already owns a home in the Bay area. If you bought a nice single-family home twenty years ago, you're sitting pretty. You live in a nice area that probably isn't going to change very much (thanks to the regulations), and the steeply rising value of your real estate (thanks to the regulations) is going to fund your retirement very nicely.
- azernik 9y agoNot sure how much of this primer on Bay Area politics you need, but here we go: Mostly local governmental opposition. Local towns want businesses (tax revenue!) but not residential development (costly services!). Local homeowners who vote for said town governments also have an interest in higher home prices. What makes this possible for them is the lengthy public hearing process involved, allowing local activists to delay construction for years for BS reasons; and the discretion given to elected officials not just to make policy, but also to override it. In recent years, Bay Area activists (under the slogan of YIMBY - "Yes In My Backyard") have been successful in both pushing housing development through direct local advocacy (especially effective in places with lots of voting renters, like SF proper), and in getting the state to exert its power more over local government. The background to this being that the state has since 1982 had a law trying to combat this phenomenon - the Housing Accountability Act. It requires cities to publish city development plans for public inspection, requires those zoning policies to meet certain requirements in terms of housing quantity, and then to actually make zoning decisions that comply with those policies. However, private individuals or organizations need to bring suits against cities to make it effective, because the US is crazy and prefers lawsuits to actually hiring professional regulators. So activists have funded lawsuits (see the more recent entries at https://en.wikipedia.org/wiki/California_Housing_Accountability_Act#Litigation_involving_the_HAA https://en.wikipedia.org/wiki/California_Housing_Accountabil...) against cities that have obstructed development. They also got the CA state legislature to strengthen the HAA this year, putting more of the burden of proof on the cities to show their justification for development denials, making cities pay attorney's fees when they lose a case, and allowing judges to fine cities for violations.
- wahern 9y agoThe fiasco that was the appointment of the acting director of the Consumer Financial Protection Bureau (CFPB) and the absurd ping-pong policies of the Federal Labor Bureau (FLB), FCC, and (presumably) the CFPB exemplify why in the United States civil suits are the better enforcement mechanism, especially in the modern era. They typically lead to more stable regulatory policies. Heck, even the behavior of zoning boards exemplifies this. The civil actions are being used by YIMBYs precisely because the regulators--the zoning board--are neither obeying the spirit nor letter of the law, having become puppets of the elected officials or worse--wagging the dog. A system of centralized regulators only works when you have a strong and independent technocratic bureaucracy. Because of the extreme divisiveness of modern politics, especially at the federal level where regulatory agencies have become turf for epic policy battles, that simply doesn't exist in the U.S. Indeed, the use of civil actions for enforcement is often chosen precisely to try to deflect criticism aimed at proponents of a regulatory framework, and protect the framework from interference by opponents. This is common at both the federal and state levels, and used by both parties. Sure, civil actions are more costly. But much like democracy you pay a price for a certain amount of stability. Because of the size, diversity, and peculiar politics of the U.S., we often must resort to such approaches where, e.g., Norway, Japan or the EU would not.
- azernik 9y agoYou create a centralized bureaucracy by hiring bureaucrats and giving them the final say. The ping-pong policies of the FCC, FLB, and shortly the CFPB result from legislatures insisting on a system where political appointees are the ones who make the final vote on regulatory decisions. This results in a situation like the current FCC, where votes are party-line and regulatory action depends not on the written policies but on who won the most recent election. Similarly with zoning boards and the YIMBY suits - local zoning policies were made by elected city councils, and the YIMBY suits and state legislation are trying to force cities to pass legislation and then apply those rules without political interference.
- wahern 9y ago> You create a centralized bureaucracy by hiring bureaucrats and giving them the final say. Americans don't like doing that, especially conservative Americans for whom "centralized bureaucracy" is literally synonymous with Communism. In many states even judges and sheriffs are elected. My point is, if you want to shield the regulation from politics your best bet (although by no means guaranteed) is a framework that creates a private right of action. > Similarly with zoning boards and the YIMBY suits If zoning boards mechanically applied zoning ordinances like they're supposed to, things wouldn't be nearly as problematic as they are now. Ask any developer; they'd be thrilled if they only had to contend with the letter of the zoning ordinances. FWIW, in the business world private rights of action are usually preferred over a centralized regulator, at least if they have to choose their poison. Centralized regulators usually _prohibit_ certain business practices prophylactically long before any substantial harm arises. Whereas with private rights of action you get to do whatever you want, whenever you want, until you _actually_ harm someone. Historically this is one reason why some economists argue the United States and other Common Law jurisdictions that relied on civil actions were more innovative. Similarly, such jurisdictions were long considered more business friendly as regulatory policies were more predictable. A central regulator can decide to block your business on a whim, whereas with a civil action there has to be _actual_ harm as well as refusal to settle damages outside of court. Both can involve a lot of red tape and a long paper trail. The important distinctions are 1) whether the government steps in before or after substantial harm has occurred, and 2) how closely penalties reflect actual harm. Moreover, the basic issue is how reliably you can price risk into your business plan. A business can reasonably plan expenses for civil suits. A regulator can shut you down, and all the money in the world won't stop it. But other Common Law jurisdictions have slowly shifted enforcement to Continental-style centralized regulators. The U.S. has too, but it's been more chaotic, inconsistent, and (predictably) much more corrupted by politics. There are many reasons for this, but in any event there's no reason to think that Americans can do centralized bureaucracy any better than we do now. It's important to recognize and accept built-in political and societal constraints when engineering regulatory frameworks if you want to maximum benefit and minimize harm.