It seems like a good idea at first glance.  The major problem with it is that it’s actually very hard to put a value on land independent of what’s on it; apart from stuff like agricultural fertility and proximity to water, land is mostly valuable for what it’s next to, which is to say buildings.

But that is a pretty radical move (even if we came close to doing it during the New Deal). It would mean the only remaining specialized programs needed would be for the disabled and sick, which is a huge group of course, but a National Health Service would reduce that problem substantially as well.

And pollution is hardly exclusive to high density: traffic jammed freeways are a symptom of low density suburban developments as much as high-density urban areas with good transit. Plus, it’s pretty trivial to turn that externality into a cost that’s paid by the party engaging in it: congestion tolls.

Besides, low density has plenty of its own externalities: increased energy usage leading to increased pollution; the need for wider, straighter and faster roads, carrying more traffic (which aren’t exactly welcomed by the people who live next to them); increased utility costs that are borne by all subscribers; requirements on businesses to provide large amounts of parking space; and so on.

That’s not exactly what a negative externality is. It’s when costs are not fully borne by a person making an economic decision. With traffic congestion, the “economic decision” is whether or not to drive/walk/etc. 
With noise, it’s the decision to do whatever activity is directly causing the noise. Density is a factor in how common these decisions are, and thus how large the externalized costs are. However, these external costs do affect the appeal of living in a particular place, and if the place becomes less appealing, the density of housing units should stabilize. So I think Yglesias’s point isn’t exactly nonsense.

“People worry that a denser neighborhood will have more traffic and more noise. Generally speaking, they’re correct. But all this means is that allowing higher density will be a self-limiting process. Balancing the different costs and benefits involved in denser building is, after all, precisely the sort of thing that relatively free markets are good at.”

But this is nonsense.  Traffic (congestion) and noise are examples of negative externalities of higher density.  Negative externalities are costs that are not reflected in market prices.  That is why free markets are not good at producing optimal density.  Either these costs must be internalized through a “density tax” of somesuch, or density must be limited through regulation to prevent external costs from becoming excessive.

We’re in agreement that inclusionary zoning is a much better avenue for providing low-income housing than project-based housing. And I don’t have any problems with the stated goals of inclusionary zoning, but I think that IZ supporters are too often ignoring its unseen consequences. Again, to use the DC example because that’s what I’m most familiar with, here 5% of new housing developments with 10 or more units or newly renovated housing must be affordable for people 50% of the AMI and an additional 5% for people making 80% of the AMI. (I’m pretty sure these figures are still accurate — someone please correct me if these are wrong.) By requiring developers to provide below-market-rate housing, this policy drives up the price of market-rate housing and also reduces the supply of new housing from what it otherwise would be. Who does this policy hurt most? Anyone making less than 50% of the AMI and those making up to 80% or just over who are not able to secure price-controlled units.

Also, from a practical standpoint I think it’s silly that IZ typically applies to new or newly renovated properties. In most cases, these would be a neighborhoods’ most expensive housing units in a free market, so why are we seeking to make them go for the lowest prices? Vouchers would give us more bang for the buck if the goal is providing access to desirable neighborhoods.