I've been bashing density bonuses lately because they raise the price of density downtown.
But there are rare occasions when they make sense. The new development off Barton Springs Road, near Zilker Park, is one of them. We ought to use density bonuses to encourage developers to maintain the stock of restaurants along Barton Springs Road.
Development sometimes poses a collective action problem. Each development would benefit if everyone provided a public good (good sidewalks, retail, etc.), but no developer has an incentive to provide it unless he has an assurance that everyone else will provide it too.
Take the sidewalks downtown. Developers have an incentive to under-invest in sidewalks. Most of their benefits are captured by passers-by rather than the condo's residents. Even if a developer decides to spend lots on his building's sidewalks, he has no assurance that any other developer will do the same. It doesn't do any good to have good sidewalks along one block if the rest of them suck. Thus, left to themselves, developers usually build narrow, cheap sidewalks.
But everyone is better off if every developer builds good sidewalks -- wide, lined with trees and benches, with awnings to break up the sun. Good sidewalks become public places that encourage loitering, which makes the area safer and encourages retail, which encourages more foot traffic, which provides more safety and retail . . . . Good sidewalks enhance the value of the entire area, including the value of individual condo developments. Developers are better off if they are required, or at least incentivized, to build them. That gives them the assurance that they won't be throwing away their money. This is the point of the Great Streets program.
The same goes with retail and restaurants. It doesn't do much good to put retail in the ground floor of one building if no one else does the same; it takes a critical mass of retail to generate the necessary critical mass of foot traffic. The same is true to a lesser extent for restaurants. But if everyone has retail or a restaurant, and the location is good for retail and restaurants, then everyone is better off.
The city should be very cautious about intervening to solve collective action problems because it usually does not know that there is a collective action problem. When the city mandates improvements that no one wants, even in the aggregate, then the city is just wasting everyone's money.
But we know that Barton Springs Road is a great place for restaurants: It is lined with thriving restaurants. That will be part of the appeal of the new condos.
It doesn't look like either the Zilker Park or Barton Place projects will leave any room for restaurants or other retail, though. 
They will freeload off the restaurants already there. But as other condos move in and displace restaurants, there is a real danger that the strip will lose the critical mass of restaurants that makes the strip a vibrant place.
Density bonuses make sense here. Give the developers a carrot -- modest increases in density (I don't want skyscrapers there, either) -- in return for space for restaurants or other retail. The condos will be more valuable because their residents will be assured of a continuing vibrant restaurant scene. And the rest of us won't lose Barton Springs Road as a public space.
Update. I should have done a little more homework. For some properties on Barton Springs Road, there is already a (relatively modest) incentive to provide a mixed use project. The lack of mixed-use in the Zilker Park Lofts may be the consequence of neighborhood meddling, though.
As Shilli pointed out in the comments, Barton Springs Road is a Core Transit Corridor. This means that the new commercial design standards apply. It also means that the Vertical Mixed Use regulations -- assuming they ever take effect -- will also apply because all of the properties along Barton Springs Road between South Lamar and Robert E. Lee are zoned commercial. Normally, a developer would have to provide ground-floor pedestrian-oriented uses in order to put a residential development in a commercial district; i.e., VMU solves the "collective action" problem.
There's a wrinkle (as there always is with our Land Development Code). This area is in the Butler Shores subdistrict of the Waterfront Overlay District. Multi-family is already permitted in the WO Districts. In some, ground-floor pedestrian uses are required, but in Butler Shores, they are required only for properties that abut Town Lake. However, there are (relatively modest) bonuses to encourage mixed use in WO districts. Unless I'm misreading our labyrinthine land-use code (which is entirely possible), a developer can add residential floor area equal to 60% of the base maximum -- essentially, a 1.2 FAR for residential in a CS district. Not as generous as VMU (which has no FAR maximum with affordable units, only compatibility requirements), but still it's something.
So why won't the Zilker Park Lofts be mixed use? The Statesman doesn't say, but here's the story as best I can tell: One of the tracts was subject to a conditional overlay imposing a ridiculously low FAR. The developer sought a zoning change, to Downtown Mixed Use, because he needed a higher FAR and he wanted to put a mixed-use development there. After lenthy negotiations with the Zilker Neighborhood Association, though, the developer agreed to a conditional overlay limiting the development to 10,000 square feet of non-residential uses and 650 trips/day. 650 trips/day doesn't leave much room for a commercial use -- the site plan TIA projected 550 trips/day from just the residential uses.
Thus, it looks like the local neighborhood association negotiated a trip limit that effectively precluded any mixed use.
I admit I'm reading between the lines; anyone with personal knowledge of the negotiations should feel free to correct me. The developer initially wanted more units; perhaps he intended to abandon pedestrian-oriented uses when he couldn't get the density he wanted. However,I have to say I was suspicious when I read the Zilker Neighborhood Association crowing about its cooperation. It clearly had an opportunity to negotiate for mixed use. At best, it passed up the chance. At worst, it actively opposed mixed use through trip limits.
