A Study on Regional Governments Part III: Back to North America


I’ve been working on regional governments on and off for several years, and this time I feel like I’ve made some progress. There are a number of reasons for subdividing North America into new regional governments, as I’ve already discussed in Part I and Part II. But the main idea comes from architect and known crazy person Christopher Alexander‘s book A Pattern Language, a book about development and building patterns that goes from the very large to the very small scale. The very first pattern in the book is:

Metropolitan regions will not come to balance until each one is small and autonomous enough to be an independent sphere of culture…

Therefore:

Wherever possible, work toward the evolution of independent regions in the world; each with a population between 2 and 10 million; each with its own natural and geographic boundaries; each with its own economy; each one autonomous and self-governing; each with a seat in a world government, without the intervening power of larger states or countries.

This time I tried to be more official and used a GIS to do my work rather than Google and Wikipedia. In the past I’ve used counties as my basic geographic and population unit, but that was problematic, seeing as Los Angeles County has over 10 million people already. It’s also problematic to go down to the level of cities, towns and places, because then you run into 8 million-strong New York City, and the second you lump in any of its suburbs it puts you over 10 million. This time I went in between and used county subdivisions, which will both divide Los Angeles County up as well as take advantage of the fact that New York City, while being one city, is also five different counties.

I also used information from both the US and Canada. Initially I wanted to do Mexico as well, since there are a number of major cities on the Mexican side of the border that draw people in from the US. However, I figure that the ties between the US and Canada are much stronger and that it would be easier to integrate their populations. Also, the fact that the US and Canada make census data and GIS files fairly easy to obtain while Mexico doesn’t might have something to do with it.

My methodology goes something like this. Let’s say you’ve got Country A with Cities 1, 2, and 3 and a population of 30 million.

Process1City 1 and City 2 are the largest cities in the country, so the country would be divided between these two cities.Process2Region 1’s new population is 18 million and Region 2’s is 12 million. Since Region 1 is the larger of the two and since it is still above 10 million in population, it needs to be divided again. The second largest city in Region 2 is City 3, so new boundaries need to be redrawn between that and the two existing cities.Process3The new Region 3 took 8 million people from Region 1 and 2 million from Region 2, giving all three regions a nice round population of 10 million. At this point we no longer need to subdivide them any further.

Of course, when you do this on real land and using real borders, it doesn’t come out quite as clean. It looks a bit more like this:

Click to enlarge.

Click to enlarge.

TableThe largest metropolitan area in Anglo-America is New York City and the second is Los Angeles, so they became the first two regions. Since they were both over 10 million and New York City was the larger of the two, Chicago was the third region, then Dallas, then Philadelphia, etc.

What is apparent in this exercise is that, since the regions are based on population and not geography, the size of the region correlates to population density. The densest parts of Anglo-America, the northeast seaboard and southern California, have the smallest regions, while the sparsely populated Rocky Mountains and northern Canada have enormous regions. What I haven’t realized in past iterations of this project is that it complicates Alexander’s second pattern, the distribution of towns:

If the population of a region is weighted too far toward small villages, modern civilization can never emerge; but if the population is weighted too far toward big cities, the earth will go to ruin because the population isn’t where it needs to be, to take care of it.

Therefore:

Encourage a birth and death process for towns within the region, which gradually has these effects:
1. The population is evenly distributed in terms of different sizes- example, one town with 1,000,000 people, 10 towns with 100,000 people each, 100 towns with 10,000 people each, and 1000 towns with 1000 people each.
2. These towns are distributed in space in such a way that within each size category the towns are homogeneously distributed all across the region.

This process can be implemented by regional zoning policies, land grants, and incentives which encourage industries to locate according to the dictates of the distribution.

apl2diagramtowns of 1,000,000 – 250 miles apart
towns of 100,000 – 80 miles apart
towns of 10,000 – 25 miles apart
towns of 1,000 – 8 miles apart

The last part of that section, which describes the spacing of towns, leads to a specific size that a nation of a given population should be. I experimented with a few town distribution models and determined that based on a combination of Alexander’s population and town distribution recommendations, a region should not be less than approximately 30,000 square miles, without being more than approximately 130,000 square miles. Several of the regions I’ve created based on population alone are either too large or too small to meet these criteria.

Click to enlarge.

Click to enlarge.

So to comply with Alexander’s recommendations, the small regions would have to have their population distributed more sparsely and their boundaries enlarged to accommodate that lower density, and the large regions would either have to become smaller and more dense, or have their populations concentrated in certain areas while others are left as uninhabited wastes, as is essentially the case in much of the mountain deserts of the western US and the arctic regions of Canada.

While I see the value of Alexander’s argument for smaller governmental units, I find his arguments for the distribution of towns a bit more dubious, especially when it comes to areas that are too dense for his recommendations. I generally feel that the best way to preserve undeveloped land is not to distribute people evenly across it, but to concentrate them all in one area and leave more of the land untouched. That’s why New Yorkers are some of the greenest people on earth; they leave the countryside alone, and are packed dense enough that they don’t have to use cars and take advantage of the energy savings of dense housing, making their environmental impact considerably lower than someone in a lower density area.

That being said, at some point I would like to take one of the Goldilocks regions I’ve created like San Francisco or Pittsburgh and try to create a distribution of towns like what Alexander recommends, just to see what it would look like. But as far as drawing new regions goes, I’m pretty happy with this version, and I don’t see myself redoing this project again.

A Federalized World Based on Global Cities


I have not been as diligent in my blog-keeping as I should be. Between my new job, my wife expecting, and me getting wrapped up in a long and tedious project, I haven’t kept up. But I’ll take this time to share my personal project with you, and then hopefully I can get back on a more regular schedule.

As I have written about before, I think that many of our political boundaries, especially in North America, are not drawn in such a way that they reflect social realities on the ground. In my most recent project, this has intersected with my interest in the concept of global cities.

Global cities are cities of international importance, whether financially, culturally, religiously, or otherwise, and is largely self-sufficient as far as resources. There are a number of ways to rank a global city, but I’ve gone with the Global and World Cities (GaWC) system. They classify cities as Alpha, Beta, Gamma, High Sufficiency, and Sufficiency world cities, and further divide the first three into plus, neutral and minus groups (i.e. alpha+, alpha, alpha-, etc.). In addition, two cities, London and New York, are so far ahead of anywhere else in international importance that they are given the rank of Alpha++. All in all, there are 296 cities that the GaWC consider global cities.

All the world cities.

All the world cities. As with all the following maps, this file was enormous and I had to shrink it to make the blog run smoothly. Click to enlarge.

My thought then became, what would the world look like with a federal system based on the rank of the nearest global city? This would mean that, first, the world would be divided in half between London and New York, the two Alpha++ cities.

Click to enlarge.

Click to enlarge.

These would then be subdivided into smaller Alpha+ regions, as London is subdivided into London, Hong Kong, Paris, Singapore, Tokyo, Shanghai, and Dubai; and New York is divided into New York, Chicago and Sydney.

Click to enlarge.

Click to enlarge.

These Alpha+ regions are then subdivided into Alpha regions (e.g. London is divided into London and Amsterdam):

Click to enlarge.

Click to enlarge.

then into Alpha- regions (e.g. Dublin breaks off from London):

Click to enlarge.

Click to enlarge.

then Beta+ regions (e.g. Hamburg and Copenhagen break from Amsterdam):

Click to enlarge

Click to enlarge

then Beta regions (e.g. Manchester breaks from London):

Click to enlarge.

Click to enlarge.

then Beta- regions (e.g. Birmingham breaks from Manchester):

Click to enlarge.

Click to enlarge.

then Gamma+ regions (e.g. Glasgow and Edinburgh break from Manchester while Bristol breaks from Birmingham):

Click to enlarge.

Click to enlarge.

then Gamma regions (e.g. Leeds breaks from Manchester):

Click to enlarge.

Click to enlarge.

then Gamma- regions (e.g. Southampton breaks from Bristol):

Click to enlarge.

Click to enlarge.

then High Sufficiency regions (e.g. Liverpool breaks from Manchester and Newcastle from Leeds):

Click to enlarge.

Click to enlarge.

and finally individual sufficiency regions (e.g. Sheffield breaks from Leeds, Aberdeen from Edinburgh and Nottingham from Birmingham).

Click to enlarge.

Click to enlarge.

From here, regions could be administered as outlined in Alexander’s “A Pattern Language.” Cities of higher level regions would be capitals for the smaller regions. So just as the United States currently has a capital in Washington, DC, each state has a capital, and each county has a seat, so London would be the capital of its Alpha++ region, Paris the capital of a smaller Alpha+ region under London’s administration, Amsterdam over an even smaller Alpha region, etc. I tried to put it all on one map, below, but the labels, especially in dense areas like Europe, got really messy, so I only used the lowest-level labels. Please cross-reference the maps above.

Click to enlarge.

Click to enlarge.

Is this a viable solution in the short term? Are straight lines the perfect boundaries? does this necessarily reflect cultural lines on the ground as best as they could be drawn? No, but it does reflect something of the reality of urban and regional hierarchy that exists in the world, which could be one element in creating a world that better reflects its economic and cultural realities.

A Study on Regional Governments Part II: the British Isles


I haven’t posted in a few days because I’ve been completely wrapped up in this project. Hopefully today and in the coming days I can make up for it a bit with some original content. For starters, I wanted to revisit the idea of regional governments.

I am still a big fan of Christopher Alexander, despite the fact that I’ve talked to a number of people who have met him in person and say he’s basically insane. In his most well-known work, A Pattern Language, the very first pattern is “Independent Regions,” where Alexander says:

Metropolitan regions will not come to balance until each one is small and autonomous enough to be an independent sphere of culture…

Therefore:

Wherever possible, work toward the evolution of independent regions in the world; each with a population between 2 and 10 million; each with its own natural and geographic boundaries; each with its own economy; each one autonomous and self-governing; each with a seat in a world government, without the intervening power of larger states or countries.

Alexander lists a number of reasons for this conclusion, including that a citizen of a smaller nation like Denmark can have direct access to the Minister of Education, while in a large country like the United States this is impossible. This, combined with my own opinions about the arbitrariness of state and county lines and the modern reality of metropolitan economies led me, a few years ago, to undergo an effort to redraw the entire United States based on regions of 2-10 million people.I produced three maps, including the one above (which I redrew once I learned something about design and how awful my colors were), which showed what the country would look like with minimums of 2 million, 5 million and 8 million. The problem with using minimums was that many regions ended up way over the 10 million mark. Los Angeles County, for example, has over 10 million people in it already, and when you add the surrounding counties which couldn’t form their own regions, Los Angeles became enormous.

The way I came up to address this was, instead of starting with all the towns in the US with over 100,000 people and seeing which ones could amalgamate 2 million people, I thought I would try another technique. I would start from the top, with America’s largest cities, and assign areas to them, based on which city it would be easiest to walk to based on Google’s walking directions. Once an area had been assigned, if it was more than ten million, I would take the second largest city in the assigned area and re-divide it; Thus, New York would start with 300 million people, but then be divided between New York and Los Angeles. The half of the country covered by New York, would still have the most people, so the next division would be Chicago, then Houston, etc., until each region had less than 10 million people.The problem is, the United States is an enormous country, with over 3,000 counties, and dividing the country county-by-county takes a very long time.

Another major issue I came across was that, just as state and county boundaries are arbitrarily geometric, so are our national borders. It is silly to think that, just because there is a line between them, that Mexicans don’t participate in the economy of El Paso, or Americans in the economy of Vancouver, or Canadians in the economy of Buffalo. So to really create a model that would take into account the realities of economies that straddle current international boundaries, it would be necessary to incorporate multiple nations, or even the entire continent, because when you get down to it, the hardest borders we have are oceans.

I wasn’t ready to attack that sort of a project yet, so I wanted to work on a pilot; I needed something that was smaller than the US, but still fairly urban, that was completely surrounded by water. I decided to test my new technique on the British Isles.

The British Isles consist of a number of islands, the principal ones being Ireland (Eire) and Britain. Politically, there are two sovereign states in the Isles: the Republic of Ireland, covering most of the island of Ireland, and the United Kingdom, covering Britain, Northern Ireland, and most of the other islands in the Isles. The UK is further divided into four constituent countries (England, Northern Ireland, Scotland, and Wales) and three crown dependencies (the Isle of Man in the middle of the Irish Sea, and the Channel Island dependencies of Guernsey and Jersey). Each of these countries and dependencies are largely responsible for their own internal affairs (all except England have their own elected legislatures; England’s internal affairs are governed directly by the UK Parliament), but international representation and defense is managed by the UK. Although this is certainly closer to the model supported by Alexander, some regions are too large (England especially), while others, the dependencies in particular, are too small.

From wikipedia.org.

Municipal government in these countries is, in many ways, similar to that of the US: In Ireland, Scotland, and Wales, the country is divided into counties which then are made up of parishes (cities and towns). Northern Ireland’s traditional counties have been subdivided into smaller districts, but the principal still holds. England, on the other hand, is a bit messier. It is divided into nine regions, which are then divided into two types of counties: metropolitan (for the more urbanized areas of England, such as Birmingham and Leeds) and non-metropolitan (less urbanized). These counties are divided into districts, and then into parishes. In addition, there are a number of “unitary authorities,” which work similarly to the independent cities of Virginia. Greater London is it’s own beast, on an administrative level similar to that of a county. The map I decided to use features counties, metropolitan districts, and unitary authorities.

Administrative divisions of the UK. From wikipedia.org.

I should point out that what follows is mostly a geographic experiment. Differences in linguistics, religion, and history were not weighed. Some of the boundaries in these countries are where they are for very good reasons, and many residents of Ulster may not want to be lumped in with the Republic of Ireland, or people west of Offa’s Dyke may want to remain in a united Wales despite being closer to English Liverpool than Welsh Cardiff. That being said, the geography may show things differently.

With this more top-down method, the larger regions emerged first, so the first thing that came up was a set of regions less than 10 million people:

I think that this map does a fair job of reaching Alexander’s goals. It coincides well with many of the traditional regions of the Isles: Ireland, Scotland, Northeast and Northwest England, the East and West Midlands, and the Southwest are well defined. There are some anomalies. First are those caused by ferry services between the islands. For instance, while Dublin is geographically the closest major city to Belfast, it takes much less time to take a ferry between the two cities than to walk. However, if you’re already taking a ferry, it takes less time to ferry from Belfast to Liverpool than to Dublin. By the same token, it takes less time to ferry from Anglesey in Wales than to walk to Liverpool. In reality, these anomalies probably wouldn’t exist.

Another interesting thing is the smaller cities around London. Unlike in the North, where major cities form a string from Liverpool to Manchester to Sheffield and Leeds, London is sort of on its own, and it falls on smaller cities such as Luton to create regions around it. This is most pronounced with Woking, a town of just over 60,000 people, and not even the largest in its region (Reading is much larger), but the largest that could anchor a region of 2 million. It would be interesting to see what would happen to these cities if they were given a larger administrative role.

The largest city that really gets the short straw in this analysis is Bristol. At about 430,000 people, Bristol is one of the top ten cities in the UK, but always ended up on the wrong side of the dividing line. It started out being a part of Birmingham, but when Southampton split off from London, it got sucked into its sphere of influence. It is hard to say if something like this would happen in real life.

Next, I took these regions and subdivided them until the subsequent regions were either less than 4 million and thus couldn’t be divided into two regions of 2 million, or when, even being above 4 million, it was impossible for the second city in the area to support a population of 2 million. For instance, so much of the population of Ireland is based in Dublin and its nearby coast that the second largest city, Cork in the west, was unable to amass the 2 million needed to be a region.

Some of the anomalies from the first map go away, while news ones arise. By creating a Belfast region the ferrying issues on the first map go away. However, Anglesey is still closer to Dublin than Liverpool. The introduction of Cardiff also creates a Welsh region, although much of the north is still closer to Liverpool. Bristol still draws the short straw though, as it is now shifted into Cardiff. The problem of small cities that are really part of a larger region spreads from London to Birmingham, where Wolverhampton forms its own region when really it is very much a part of greater Birmingham. The North is generally much more fractured. This is also the first time that the unitary authorities pose a problem; for instance, Derby is closer to Leicester, while Matlock, the administrative center of Derbyshire, is closer to Sheffield. In all, I think the larger regions better reflects the reality on the ground, the creation of Welsh and Northern Irish regions notwithstanding.

Even in this case, there are some if the same issues with international boundaries. For example, the Channel Islands are both closer to France than England. Would it be better for them to be administered by France? What about the Faroe Islands, administered by Denmark but much closer to Scotland? These questions notwithstanding, I think I’ve learned some things for another American experiment; work with maximums, but don’t worry too much about getting down to 2 million; just getting past 10 million will get you pretty far.

The 38 States of America


The 38 States. From tjc.com.

Someone recently posted a link to this map and it’s accompanying explanation on my old blog as a comment on my regions project. I know that I’m not the only person to think about how meaningless state boundaries are in the United States, but I was unaware of this effort by C. Etzel Pearcy. It may be the most thorough I’ve seen. Although some of the regions would not fit Alexander’s criteria for independent regions (the ones in California, Texas and New York, to name a few, certainly have populations over 10 million), it is an interesting effort.

%d bloggers like this: