Skyscrapers solve several really hard problems. For example, dense sidewalks have higher throughput than city streets or even subways, unfortunately people don’t walk very quickly and most sidewalks see little use. However, the number of places you can walk to directly correlates to building height. So, in effect someone walking surrounded by skyscrapers is doing the equivalent of highway speeds when surrounded by 6 story buildings.

Which is why NYC sidewalks get used so much, they are a very useful form of transportation. Which has positive knock on effects in terms of human health etc.

> However, the number of places you can walk to directly correlates to building height.

Except every floor doesn't have an equal effect? Lower levels and rooftops are going to host the majority of destinations, while mid-levels are most likely to be homes and offices that are adding to the pool of people able to walk to those destinations.

The ground floor of a 6 story building can’t support a full set of shops with just the people inside the building. This shows up as cities like Paris having clusters of destinations and then seemingly dead space streets filled with buildings only residents visit. It can still be “walkable”, but you’re walking farther with fewer destinations within walking distance.

It’s a basic economic reality is if a restaurant needs X customers to be viable a city will have some constant * population/X restaurants. Be they in 5 story buildings or 50.

A restaurant's capacity isn't limited to the amount of customers that make it viable. There can be enough population to support two restaurants, but if the first restaurant has the capacity to serve them all and the customers prefer it, only one of them will actually survive.

So once you hit the point where the foot traffic retail is all viable, there's a stretch where there's not a real need for new businesses and any new competitors would face a large disadvantage from location.

Not that there aren't ways to design around that to make very high density areas viable, but it's not a simple line graph correlation.

Businesses can always acquire new customers by diversifying away from what the competition is providing. This means that larger markets can inherently "support" a higher variety - there simply isn't a point where new business must inherently fail. This is not a new observation; it has been made repeatedly since ancient times.

But it's going to take a whole hell of a lot of special sauce to survive when you're the first random mid-level restaurant competing against places that can attract customers through line of sight.

You can't simplify away the varying value of certain land for certain uses and still pretend you're doing economics.

Clearly I overly simplified things, but the point was doubling a cities population should roughly double the number of restaurants. That doesn’t hold for towns, but you see multiple McDonalds in the same city because being the closest McDonalds is a useful difference. Similarly at scale you see more specialization such as Thai vs Chinese vs Pho etc etc.

Add cost and you can have almost unlimited density without inherently limiting the number of restaurants much like how a mall can have 100 stores all selling different kinds of clothing.

It's important to note that this does not just apply to retail business, but to jobs and economic/social interactions of all kinds. Don't get me wrong, increased density is expensive all other things being equal, which is why the highest density can only be sustained in but a few very unusual places. But this is the stuff that actual agglomeration dynamics are driven by.

I can see how walking between buildings and then riding elevators is a good local maximum, but I wonder if there’s another local maximum where you drive between buildings but climb stairs once you get to them.

If nothing else, skyscrapers are expensive; I don't think they would ever pencil out as the default mode of housing for most people. The 5-story model you talked about is pretty good - I went to college in a neighborhood like that - and that's what I'd like to see displace the suburbs for the greatest number of people in economically unspectacular places.

Still, I think it's important to have skyscraper cores and we should build them closer to the economic limits than the present legal limits.

There is no shortage of space for people. Only a shortage of places people actually want to live. Everyone flocks to mega-cities because that's where the jobs are.