Size matters

Among the most interesting features of the modern world is the differing quality of various governmental units throughout the world.   The Danish, Swiss, and Singaporean governments appear to work very well.   The  Chinese, Indian, US, and Russian governments appear to work somewhat less well.  All assuming that serving the populace is the goal, and correcting for per capita GDP.


——Relevant Aside——
In my day job, I’m a software process consultant, focused on software process delivery.  And I’ll happily pair in and sling code with you in whichever language you prefer, though it might take me a couple minutes to catch up on Lisp or Smalltalk.   Among the most notable features of my job is that I’m asked to go into very large organizations (who can afford me), in order to help them try to capture the Agility of very small organizations who did things right.

And the single thing I’ve learned in the last 10 years doing this thing (’07.  Yikes) is that size matters.   Transformations of large organizations is fundamentally different than transformation of smaller organizations.   Indeed, there are layers and layers and more layers of difficulty in turning organizations of 100,000 people, as compared to 1,000 people, or 10 people.

In the Agile community, one of the founders of the movement, Alistair Cockburn, built, not a methodology, but a framework for choosing methodology based on project/product characteristics.     A 5 week project, 3 person project has an appropriate methodology involving ONLY 1 room and a door that closes.    A 10 year, 300 person effort might ought to have a bit different set of controls.    If you use the 500-week method for the 5 week project, it will probably take 45 weeks.   If you use the 5-week method for the 500 week project, you’re 99.3% guaranteed to fail utterly.   Yes, other factors matter besides length and manpower, but those are pretty strong indicators for what you ought to do.  The factor I personally use first to choose between methodologies is:   What’s your goal:

Faster delivery?  Ease of change?  Better large-group Coordination?       Each goal has more and less appropriate approaches.

So… where’s the analysis of Organization size as it applies to governance?
Notice anything about the best governance in the world?   There are 3 countries, each listed by someone as brilliant governance…and they’re each tiny.    Smaller than metro Chicago. Perhaps someone should study this phenomenon.   And perhaps one should abandon hope that a large group of 320 million can get the quality governance one expects of a 5-million person city-state, a 5-million person small country or a 8-million person aggressively federalist system where most decisions are made at the 50,000 person city level.

Yeah…so all that is another plug for federalism.   Size of governing unit matters.   Probably via payoff.   In 5 million people, all of whom live within 30 miles of one another (Singapore), you have a bit different of a problem than 320 million people who mostly, but partly don’t, live within 3000 miles of one another.

Communication layers, Public Choice strength on population, financial impact of regulations, anonymity and reputation concerns, common culture…they’re all very different between Denmark and India, even Singapore and China.


I’m in favor of separatism.   I’d love for California to secede.   And Texas.   And then for Austin to secede from Texas.   And then Cedar Park to secede from greater Austin.    But in the absence of my preferred solutions, I’ll take real federalism.  Swiss variety, modeled on historical American, preferably.

But we’d all do a lot better if we stopped pretending that families could work the same way as companies, or as cities, or as states, or as nations… Size matters a lot.   Any discussion of policy that doesn’t take size into account appears, to this commentator, insane.