The Future Of Cities

What sets city size? That is, what determines how many people all cluster together in an urban area? On the one hand, city size increases with feasible building height and with the gains to people and businesses from interacting closely with many others. On the other hand, city size decreases with how much space folks want, and with costs to transport people and goods within a city and from outlying regions. City size increases with more and cheaper nearby non-city economic activity.

Policy also matters; poor governance and positive externalities of density reduce city size, while being the center of government for a surrounding area increases city size. The size of big cities should be limited by the reluctance of nations to let their city activity be absorbed into big nearby foreign cities. And sunk costs and coordination failures can long delay the adaptation of city sizes and locations to changing circumstances.

In the farming era, cities held only a small fraction of the population, and so their size and locations were determined mainly by nearby farming activity. However, when most folks live in cities, then nearby non-city activity matters less, and decreasing transport costs make bigger cities more economical.

So how well have today’s city size and locations adapted to industry era tradeoffs? That is, how well do cities today trade the gains from more interaction in bigger cities for the added costs of transport and reduced personal space? While we expect optimal industry era cities to be more concentrated, i.e., fewer and larger, we also expect inertia, coordination failures, density externalities, and city mismanagement to slow the transition from an ideal farming era distribution of cities to an ideal industry era distribution. So cities today are probably too many and small. But how far off are they?

One clue – alas one that that is hard to interpret – is that today (log) city size follows a normal distribution, and (log) size changes follow a random walk. Another more informative clue is that in many large nations, a big (but not too big) fraction of the urban population is in the largest city. For example: South Korea 53%, Japan 44%, Egypt 43%, Argentina 37%, Bangladesh 34%, Philippines 28%, Mexico 24% (sources here, here). This weakly suggests that such cities might be running up against a political limit – the reluctance of neighboring nations to let these cities absorb their city activity.

How should we expect cities to change in a future em era, where trillions of human emulations live in virtual reality or in tiny android bodies? Since ems are easier to transport, require less space, and interact less with rural areas, optimal em cities should be even more concentrated than industry cities. Especially if ems learn to better subsidize density, to internalize today’s density externality. And since ems require quite different infrastructure from humans, and need large and rapid changes that most cities will initially be unwilling to allow, existing industry era cities may less constrain the size and location of em cities.

Together these suggest that em cities might be quite a bit more concentrated than our industry cities. Most ems might live within a half dozen or fewer really huge cities. Which would imply that only a half dozen nations would have substantial political power, allowing for easier global coordination.

If optimal em city concentration is really high, most ems might even live in just one biggest city. An analogy in the history of brains seems apt. Some of the first brains were spread out all over animal bodies, but then brains evolved to concentrate in one small region, to minimize signal delays within the brain.

Of course one big em city could be vulnerable to bad governance, so perhaps the biggest city would change as biggest cities became badly managed. Especially if ems had better ways (e.g. prediction markets) to coordinate their city switching activities. Creates an interesting picture of a competitive world government – at any one time most world economic activity might be under a single central city government, and yet cities might compete to offer the best world governance.

GD Star Rating
Tagged as: , ,
Trackback URL:
  • Jordan Viray

    Let’s not forget the mix of urban and suburban areas as a result of subsidies to road building. That certainly distorts the dynamic with regard to perceived gains and tradeoffs for cities.

  • William Newman

    The reluctance of neighboring nations to let these cities absorb their city activity? That’s a possible explanation that had never occurred to me. Another explanation seems at least as likely to me, at least in the common case where the great city is the political center: success in business depending sufficiently strongly on connections to centralized political power that being in the political center is an important business advantage. Public services and transport can privilege the political center as well.

  • Ian

    Really interesting but wish you had pondered a bit more about the present and near future. I am very curious about what city sizes would be optimal.

  • TGGP

    Instead of the issue being neighboring country policy, might immigration policy of the host country play a large role?

  • arch1

    I have this premonitory feeling that someone is going to make the analogy with high-frequency traders clustering their computing infrastructure ever closer to that of the stock exchange in order to relentlessly shrink the ever more important lightspeed latencies. And that someone else is going to make a convincing case that the analogy is a good one.

  • ch

    How about ems not repeating human culture, but developing another one fully online?

    • Albert Ling

      Yeah I’m a bit puzzled with this notion that ems will just be “very tiny humans”. Surely if we ever do manage to emulate a mind then that experience of being will be radically different, and the way it interacts with the world will also not be comparable.

      I can imagine all ems stored in a few data-centers, controlling various external devices that require intelligence by the network, but gradually having less and less to do with “reality” and mainly solving abstract problems that require creativity and things that cant be automated.

      The majority of your day will then take place in a virtual reality. So even in situations where you have a “body”, your mind is actually stored in a Google data-center and is just controlling it by WiFi!
      So I’m seeing the real world will becoming mainly data-centers, power infrastructure and raw material extraction operations, not many cities.

  • daedalus2u

    I have a very hard time imagining ems choosing to do anything like virtual reality. Presumably they have direct access to their code, so they can modify it at will. Humans use something like virtual reality because humans can’t access their code, and so need to trick the human data processing hardware into processing data as if it was real.

    All that translation and mapping of one experience onto something else is just artifice and ems will know it. Why would they waste computation resources mapping one experience onto another and losing fidelity along the way?

    The “space” they live in will be an abstract computational space, not a 4-D space. Why would they waste computational resources emulating something fake? So they can “feel” a certain way? They can just self-modify to have the feeling, nothing else is necessary, and the feelings they get via self-modification are indistinguishable from anything they could get from a fake virtual reality.

    • Albert Ling

      with the “bottom-up” approach to creating ems, we don’t assume to know the “source code”. It is entirely possible for a working emulation to work without our understanding the theory of consciousness. That’s the whole point of Robin’s last post on brute forcing our way into making ems. So if as an em you’d look into your source code, you’d still see a chaotic mess (just like we see today when we look at neurons, or at DNA). Sure, we can make some statistical inferences of mechanism, and tweak a little at the margins, but it seems harder than you are saying.

      And even if it were so, you’ve just stated that mapping reality onto a virtual environment just wastes computational power. How then are you proposing that future ems gonna live their lives? Is the alternative of tiny android bodies less resource intensive? I mean, it seems vastly more resource intensive, but I’m not a physicist so I can be wrong here.

      • daedalus2u

        Why would ems want or need bodies?

        I thought the whole premise of the singularity was that ems would be designed and so could be designed to be more intelligent. If ems can’t be designed, then they can’t be designed to be more intelligent and so no signularity happens.

        I suppose you could design ems to have android bodies, just like you could design horseless carriages to have buggy whip holders to whip the horses that are not needed. The question is why would anyone do that? The mechanical part is going to be the most difficult part and is something needed in addition to what ever computational stuff is needed.

        The first ems won’t have android bodies because there first has to be an em before it can be interfaced with an android body.

      • Salem

        The point about ems is that they are a form of AI that is not designed. You could design a super-intelligent AI, but Prof Hanson’s point is that it would be cheaper and easier to emulate a human brain (hence ems). So he thinks that at least the first wave of advanced AI is likely to be em.

        Ems would be more “intelligent” than humans not because of better design, but better hardware. With sufficiently powerful hardware, you could run an em’s software “brain” much faster than any real human, and this, in Professor Hanson’s estimation, approximates to them being more “intelligent”. Hence possibility to create lots of creatures much more intelligent than humans extremely quickly. Hence, singularity.

      • Salem

        And presumably, ems would want and need bodies for the same reasons that humans do. Emulating a human will also emulate his psychological needs, etc.

  • Douglas Knight

    Why does the low cost of transporting people increase city size? Yes, it increases city diameter, but it seems to me to have a lot of effects both pro and con.

    Another factor is the cost of communications. You might count this under “transporting people,” but it seems pretty clear to me that low cost of communication should encourage dispersal. Thus ems might have no need for cities (cf Albert Ling, above). If they live fast enough, then the cost could be light speed latency, which would then encourage centralization.

  • Rafal Smigrodzki

    Robin, I just wanted to make sure you are familiar with Greg Egan’s ideas on the subject, e.g. in “Diaspora” – he does come to somewhat similar conclusions regarding population density of ems, and sparkles with ingenious ideas (even if his prose is a bit wooden at times).

  • David Clausen

    The work of Geoffrey West (i.e. is relevant to this discussion. He argues cities obey consistant scaling laws with interesting contrasts to biological scaling. Some of his later work also covers the difference between the scaling of complex organisms like cities compared to corporations. Both corporations and biological organisms seem to produce sublinear returns to scale while cities have superlinear returns. His work suggests that today’s cities are drastically smaller than they should be, and we still have large productivity gains from urbanization ahead.

  • Robin Hanson

    Jordan, yes other policies distort city sizes.

    William, I mentioned the political center factor in the post.

    TGGP, yes immigration barriers could reduce city size.

    ch, Albert, even with virtual lives and different cultures there’d still be physical interaction gains. Tightly clustered data centers are cities.

    Douglas, larger diameter cities contain more people. Yes communication is a form of transport. As long the comm cost, including time delay cost, is lower for folks closer, there is still a reason to clump.

    Rafal, I’ve read Diaspora.

    David, I’ve read West.

  • gwern

    I wonder how power and cooling will play into this. Existing cities like NYC have major infrastructure issues (look at NYC’s tunnel most recently), which additional density would aggravate. Datacenters have similar issues, forcing companies like Google to locate datacenters solely based on where they can find cheap power and cooling, like near dams or in Iceland.

    I don’t know if West’s data really plays into this. Presumably the outliers in his data are the largest existing cities like NYC or Tokyo-Yokohama, and they can’t tell us whether they *can* get bigger, economically, until they *do* get bigger.

  • Pingback: Today’s Links: October 20, 2011

  • JenniferRM

    The western half of the US was still being colonized as the modern railroad/machine-gun/telegraph economy was coming into place and is dramatically “clumpy” compared to urban distributions in the eastern half of the content. Maps of lighting at night make this vivid.

    California has relatively recent historical urban development so its infrastructure has mostly been deployed with modern optimization pressures in mind: no city cores based on roman military encampments, no walls to avoid being “sacked” by barbarians, but lots of strip malls… There are about 37.2 million people in the state, and the combined population of the counties of Los Angeles, Ventura, Orange, Riverside, and San Benadino (which contain a vast metropolitan conglomeration, plus bedroom/vacation communities) is about 18.3 million giving a probably overestimate around 49% of the total population of the state. South Korea and Japan are in the same ballpark, and both are notable for historically recent wars that involved a substantial reboot of their urban infrastructure afterwards.

    I wonder what China’s urbanization patterns will look like in 80 years?

    This weakly suggests that such cities might be running up against a political limit – the reluctance of neighboring nations to let these cities absorb their city activity.

    Agreement with political limits but I think they are diffuse, heterogeneous, and mostly accident in their effects of urban expansion. In California, squinting from the outside, there are basically two “cites”: SF and LA. Both are constrained in their expansion by bodies of water, mountain ranges, and vast tracts of state and federal land. If Camp Pendleton didn’t exist then I would expect San Diego and Los Angeles to have merged already, and San Diego and Tijuana are basically a single city sharply bifurcated by the US-Mexico border.

  • Pingback: Overcoming Bias : Future Wealth Inequity