Ideas to Action:

Independent research for global prosperity


Views from the Center


On Monday March 15, Paul Romer gave an impassioned presentation here of his proposal that donor countries add a new tool to their toolkit for helping the world’s poorest – the establishment of “charter cities”.  As you can learn in more detail here, such cities are conceived as contracts between three parties: a poor country which provides the land, one or more rich countries which establish the rules and norms and invest in the infrastructure and entrepreneurs/residents/workers who would choose to move in to have access to opportunities for investment or work that, because of the rules, are superior to those outside the city.

While many people at the luncheon seemed intrigued by Romer’s proposal, it was passionately rejected by a few on several grounds. 

First, the host country would be jealous of its land and its workers.  Second, the cities would fill up with young male risk-takers, with no benefits going to women or children.   Third, those entering such cities would not have a long-term stake and thus would seek only short-term payoffs.  And fourth, such cities would deprive individuals of their human rights by forcing them to live to a set of rules they had not chosen.  These critics dismissed Romer’s point that residency in the city is optional, saying that in the 21st century, city residents would eventually attain the power to change the rules.  These critics seemed to believe that the possibility of eventual popular influence on the rules is a fatal flaw, which condemns any charter city to fail.

First, Romer provides many analogies to his idea from the real world, including Singapore, Hong-Kong and the industrial zones of China and Madagascar. I think of how private investors and the Philippine government converted the US military base Subic Bay into a successful economic zone called Subic Bay Freeport and I wonder whether Cuba, the Caribbean Region and the US will someday benefit from a Guantanamo Freeport. The persistent success of these experminets, despite the potential that any of them could fail for one of the four cited reasons, already demonstrates that the idea is plausible.

Second, think of the charter city as a multi-player game, in real time, in which the norms/rules can be changed by the players either by voting, or more plausibly by a drifting consensual change in behavior. The structure will persist provided that no player or collective of players gains more from behavior contrary to the rules, than from observing them. This will be true if the rules are self-reinforcing or if some other feature of the system rewards their observance. (The most trivial example of a self-reinforming rule would be the convention that we drive on one side of the road and not the other. Any individual driver choosing to drive on the opposite side is unlikely to survive long enough to be emulated.)

CGD Open-Sourse Chess

As an analogy that is closer to (our) home, though admittedly a bit flippant, consider the establishment of “Open-source chess” here at the CGD. One day a chess board and chess pieces appeared in the lunch area of CGD. On the board was a set of rules which read as follows:

Rules for CGD Open-Source Chess

  1. This chess board is reserved for group or “open-source” chess. People wanting to play an individual game should use their own board.
  2. Anyone can play.
  3. Any individual player can make a move no more than once a day.
  4. Once an individual player has moved a White or Black piece, that player is honor-bound to play only on that same “team” until the end of the game.
  5. A card next to the chess set indicates whether the next move is for the Black team or the White team. After you move a piece for the designated team, TURN THE CARD OVER.

Think of the chess board as a “charter micro-city.”  The CGD plays the role of the host country donating its land.  The beneficent donor provided the infrastructure and established the rules.  The players have opted to move in to “work” at the chess board.  Like a real “charter city,” CGD Open Source Chess is a multi-player game and potentially subject to all four of the vulnerabilities cited by the critics at Romer’s talk.

Will this “charter micro-city” work?

In only a few days there are signs of success – and of the city’s vulnerability to the four potential sources of failure.

Success: There has been an enthusiastic in-migration of previously clandestine chess players who are contributing their time and effort to the micro-city’s product, which in addition to job satisfaction probably includes creative stimulation which spills over into other CGD activities.

Lots of junior staff stop by and contribute a move, then flip the card.  Others kibitz, looking for strategic opportunities to jump in the game.  Some of the senior staff wander by and gaze thoughtfully at the board now and then.  A few contribute a move if no one is looking.

Vulnerabilities: But one can see the seeds of crabgrass sprouting between the cracks in the pavement.

First, the host country is jealous.  Senior staff look askance.  Has some of CGD’s “labor force” been diverted from its work for the CGD?

Second, others point to the fact that few women are playing chess as a sign that the “city’s” benefits are unequally distributed by gender.

Third, individuals who contribute only one move per day tend to go for short-term gains (the “gotcha” move: White - P x N) without considering the long-term consequences (whoops: Black - B x P, check).

And fourth, the new residents (the players) are CHANGING THE RULES.

It’s true.  A recent stroll by the Open-source Chess table, where the mid-game advantage seems to be slightly in favor of black, reveals that SOMEONE HAS CROSSED OUT RULE NUMBER 3.  One of the new residents thinks the game is so much fun that one person should be able to make several moves each day!

So is CGD’s open-source chess doomed to failure?  We’ll have to wait and see.  Preliminary evidence suggests that the first three sources of vulnerability will not, by themselves, sink the experiment.  Senior staff have so far been skeptically tolerant.  A few female staff are contributing and more might do so as time goes on.  The fact that players go for short-term gains produces the off-setting benefit that games are faster – and thus more likely to attract the women.  But what about that eroding rule?

The rule of one move per day per employee is not self-reinforcing, like driving on one side of the road.  However, like the host of a real charter city, which competes in export markets, CGD competes in a marketplace of ideas.  CGD cannot afford to host this “charter micro-city” if it reduces our productivity.  Furthermore, like a Romerian charter city’s government, CGD’s management is unlikely to wait passively as a process of consensual rule change erodes its productivity. So I think it likely the experiment will survive.  If a few people step outside the norm and take multiple moves per day, most will see the good sense of restraining themselves in the interest of the “city’s” survival.  They may even bring social pressure to bear on the few infringing enthusiasts.

My guess is that in a Romerian charter city, as in the Singapore and Subic Bay models, consensual rule changes will be limited by the enlightened self interest of the participants.  Those who occupy the lower ranks of the social hierarchy in such a city are likely to be better off than they would be if the city fails.  I think this perception would serve to keep the city on track and rule changes within bounds that preserve the city’s productivity advantages.  The potential advantages are so massive that the experiment certainly seems worth a try.

Related Topics:


CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.