That would be better than the status quo, certainly. It would also be easier, politically, than variable tolls for all drivers. But variable tolls would produce larger net welfare gains.

A MSA is (generally) much bigger than what one could a "city" if looking at a satellite image. The contiguous area of built up, urban development is called a city's "urbanized area." The Census Bureau uses an algorithm that calculates the density of each census block and assigns it to an urbanized area if it exceeds a certain threshold. There are complicated rules for jumping open space, water, etc. I don't believe that San Marco is in Austin's urbanized area, although it _could_ be -- I think the CB excluded it just because it didn't want to merge any more urbanized areas. Metropolitan areas are generally much larger and typically include a lot of vacant land and, sometimes, satellite cities separated from the central city by open land. But those satellite cities are included (or, technically, their counties are included) if they send a lot of commuters to the central city.

Almost all new development in the core today is VMU- or VMU-type development on underdeveloped CS properties. (That's the game on South Lamar today.) That's because you can do little else as a matter of right. I'm a big fan of VMU development, but I don't think we can meet the demand for all new multi-family in the core by corraling it all onto a few core transit corridors. My understanding of that Burnet Rd project is that it's basically a VMU project without the ground floor commercial. Put differently, it doesn't allow any density that you couldn't get in a CS-MU-V district. (The height limit in CS is 60', and neither VMU nor MF-6 have any minimum site area requirements.) I wasn't privy to negotitions between the neighborhood and the developer, so I don't know what traffic concerns the neighborhood had and won't try to address them. But for both practical and philosophical reasons, I usually disagree with neighborhood demands to be "protected" against traffic from new residential. Practical: (1) neighborhoods overestimate the impact of new traffic, which causes them to make unreasonable mitigation demands; (2) past mitigation by the city has made a horrible, disconnected mess of our street network. Philosophical: why do we always have to screw over renters for the sake of incumbent homeowners?

I don't put much weight on the trend line at all. The first reason I plotted the trend line was to show, vividly, that density does _not_ explain affordability. Some people claim that denser cities are intrinsically more expensive. I've never believed that to be true. The real point of this piece is, "There's really not much association between weighted density and affordability." That argument has three subparts: (1) the association between density and affordability has a low R-squared; (2) the association is even weaker when you exclude California cities; and (3) if there were a true, positive association, NYC, which is much, much denser than anywhere else, would be much more expensive than it is. NYC in particular is often cited as evidence that density makes places more expensive. Plotting the densities this way is a useful rebuttal. If density does make places more expensive, then NYC (especially considering its restrictive housing regulations) should be much more expensive. One could argue, of course, that home prices are a function of the log of density or some other relationship, but that's really just another way of saying, "Density alone doesn't much impact the cost of housing."

What M1EK said. There's still fairly moderately-priced housing at the southern end of South Lamar that has not flipped to more expensive housing. Will it? I don't know. But the hundreds of units now being or about to be built on South Lamar reduce the chance of that housing flipping.

I think you somehow misread me. I agree there is no practical difference between an apartment complex being pushed into a higher bracket through renovation and its being demolished and rebuilt as a more expensive complex. I, too, would consider both of those to be "filtering," and I've made that argument before. I made it specifically in supporting Post's rezoning application on South Lamar back in 2007. Some folks were arguing that the application should have been denied in order to preserve the housing as affordable. My point was that other apartment complexes on South Lamar had already been renovated and converted to condos -- i.e., had "filtered up" -- and there was no reason to doubt this would happen to those, too. (At least Post was proposing to add an extra 120 units or so of net density.) The City ultimately can't control whether specific housing units filter up or down. The other projects now under construction on South Lamar are not instances of filtering, since they are going up on commercial lots that contained no housing units. They will ask high rents. They won't be creating the demand for those high rents, but that won't prevent people from blaming them for making the area more expensive. Austinites tend to get cause and effect backwards. Finally, "filtering" is not a euphemism for "gentrification." (For starters, housing economists were using the term before "gentrification" was coined.) "Filtering" denotes the process by which housing moves up or down the quality ladder. "Gentrification" means the process by which lower income residents of an area are displaced by an influx of wealthier newcomers. When a neighborhood gentrifies, the housing generally filters up; conversely, the fact that the housing throughout a neighborhood is filtering down is a sign that the neighborhood is getting poorer. But individual housing units can and do filter up or down the quality ladder without a neighborhood gentrifying or doing whatever's the opposite of gentrifying. Keeping the distinction straight is important. People who don't tend to underestimate the benefit of new housing supply, even expensive new housing supply: it discourages the renovation or replacement of existing affordable units.

Interesting. thanks

I don't have a link.

There were a bunch of vacancies in 2010, true, so a new count would be useful if you were interested in, say, convincing AISD there are enough students to keep schools open. But we also see __relative__ declines in population -- e.g., a decline in children's share of population, a decline in children's share of population relative to the Austin average, and a decline in the number of children per occupied housing unit. These measures effectively control for vacancies. They show the decline in children to be a real phenomenon, in other words, and not just a product of the recession.

That might have been a problem with the graphing function. Look at my grapg of austin 2000-2010 a few posts back. The numbers weren't the same. But the census spreadsheet graph wasn't showing the difference properly.