Assuming people are homogenous (racially, culturally etc. so that nationalist emotions are irrelevant), what is the most optimal way to divide the world into different countries?
Assume also that natural resources, population and economic development are all as they are today. I haven't defined what I meant by 'optimal', that should be a part of the solution.
25 comments:
Those who love Prog... and those who don't! ;)
Those who believe in Lennon?
Well, they don't have to be that homogenous ;)
Thing is, the last time I talked about this to someone, we arrived at the conclusion that size of a country was irrelevant (of course adding some assumptions). I wanted to put it up for scrutiny and see if the reasoning was right.
How did you arrive at that conclusion, about the size of the country?
I would normally say the optimal way to divide land into countries is geographical. That would, for example, settle many of Afghanistan and Pakistan's problems in one shot. (Or so said an article I read a while ago.)
Your assumptions throw that answer out of whack, though. Could you state explicitly what about humans is assumed to be homogeneous? Language too?
Homogeneity as you're assuming it almost vitiates the need for countries entirely. In fact, I think it does so completely. Why not simply have the World, with various non-sovereign administrative regions?
In reality, those administrative regions would soon assume sovereignty and fracture along linguistic, cultural, ethnic, or irrational lines.
The question is whether your homogeneity assumption also removes the irrationality of human nature. If it doesn't, then the need to have countries arises once more, and we're back to your original question.
Are the assumptions that you discussed with the aforementioned person the same that you have stated here, Dismal?
If so, look at it purely from a strategic point of view.
People are homogeneous and they form N groups. These groups settle down together and form a country each.
Now, given that resources and economic development are as it is, there would be a systemic bias in the group formation. Some groups would be better off than the others. This would cause a gravitation of individuals towards the better endowed groups. Given that each individual is homogeneous racially and culturally, the social cost per se, of migrating will be zero. Hence more and more gravitation towards certain groups will take place. Thus, over time, there would be Imagination Land.
On the other hand, one might argue that racial and cultural homogeneity is impossible if resources and economic development (and population distribution to a certain extent) are not uniformly distributed. For example, people from the cow-belt speak hindi with a drawl when compared to people from Delhi. This can be attributed to the type of resources and economic development that they are exposed to - they are more relaxed, "let the day go on" type of attitudes as compared to the brash "oh my god no time" type of urban attitude. Therefore, the homogeneity assumption itself can be proven wrong and hence optimality in terms of partitioning the world into countries might already have reached.
Excellent comment spae. Your argument is however, too simplistic (what about administration costs, trade between countries etc.). You haven't shown that your solution is optimal by any definition.
To clarify, the reason i imposed the assumption of homogenous people is that otherwise the division will definitely take on racial, linguistic and cultural lines. I want to know other than these considerations what else is important for deciding size.
What we are assuming is that the world is filled with only free-thinking clones of yourself, in the same population densities across the world, who are rational (highly questionable assumption!) . Of course, you can add further assumptions to reach a conclusion.
Your claim is that then the world should be one big country. How is your solution optimal?
@snail: Keen observation about having homogeneity when resource endowments are different but..
The only assumption I am imposing at the start is that people are homogenous. As I said, the reason for that assumption is I want to see what factors other than racial, linguistic, cultural etc should affect a country's size. So lets assume, like I said earlier, free-thinking, rational clones of yourself.
Yes. you are perfectly right that there will be mass migration of labour.However, while the social costs is zero initially, after a point the increased population density and associated problems will prevent people from migrating (like why companies started moving away from bombay, for example).
So how would you, as God, optimally divide the world into countries?
What is your definition of optimal? It has to depend on resources, I think.
True that.
Then two countries should be the best solution? Consider the war and peace game. Nash equilibrium would be (peace,peace). Trade conditions would also be satisfied if peace is maintained.
Therefore, the division of land would have to be geographical. If I were creating earth, and humans would have to begin from scratch (something like AOE), then it would not matter whether there was equal distribution of resources. They would use different things and come up to similar technology levels at some point in time and after many unique breakthroughs in unique area.
As of now, it would require to be such that resources remain equally distributed, since development, though at different levels, would still be growing in the same direction .
*areas
Methinks this problem is too hypothetical. Such homogeneity is non-existent, and will never be existent.
But given that the resources stay where they are, differences will emerge even if you create a culture-less mass of beige-skinned people. Was it Smith who said that if you give everyone in the world the same amount of money at the end of the day you'd have rich men and poor men?
Given that socio-economic differences would exist even without differences in race and culture, how different would countries be from how they are now? You could start with countries of equal size or equal wealth per-capita or something. But you'd probably get wars of conquest anyway. Equal wealth distribution doesn't mean everyone gets identical things -- there could be a war for some specific resource.
I think the question you should be asking is this: what makes you think removing racial and cultural differences will change anything about human behavior?
"I’d like to share a revelation that I’ve had, during my time here. It came to me when I tried to classify your species and I realized that you aren’t actually mammals. Every mammal on this planet instinctively develops a natural equilibrium with its surrounding environment, but you humans do not. You move to an area and you multiply, and multiply until every natural resource is consumed and the only way you can survive is to spread to another area. There is another organism on this planet that follows the same pattern. Do you know what it is? –A virus. Human beings are a disease, a cancer of this planet. You are a plague, and we... are the cure." - Agent Smith :)
I think the point I'm trying to make is missed.
I'm trying to highlight the "is versus ought" dilemma, specifically a geographical boundary is dependent on cultural, racial and linguistic considerations. I think, once we rise above such considerations, a geographical boundary ought to be dependent on some objective criteria such as welfare or wealth.
Rationally speaking, the Arab-Israeli conflict, or Pakistan and India fighting for Kashmir would not make sense, as it does not have an objective basis. By that I mean, if you take emotions out of the picture, these conflicts would seem pointless. What India stands to lose in terms of wealth or welfare of remaining populace is negligible compared to the costs imposed by retaining Kashmir.
The thing is, I am not concerned with human behavior. Economists always work under the assumption that people behave rationally in the sense that given a range of options, they choose the one that maximizes their utility. As of now, take utility to be synonymous with consumption or wealth. We can add the complications of behavior, emotions etc. later, if this leads to anything.
@han: I take it from your comment that you define an optimality as equality but you don't think the solution is stable? Wow.. equality was one consideration that never crossed my mind!! Even if wars were not permitted to happen, you are right: there would be migration and specialization that would ultimately lead to some countries being richer than other.
@Raven: I heard this quote that if you remove all the insects in the world all life will end in 50 years. Remove all humans and the earth will flourish!
All right, here's the way I thought about it (incomplete).
1. I define an optimal division as one which maximises per capita GdP or in simple terms value of output/no. of people
2. I impose the additional assumptions that
a. There are no wars between countries or countries abide by the rule of international law
b. Administrative costs are larger as the size and population of the country increases (collecting taxes, law and order)
Note that, maximizing total GDP is the same as maximising per capita GDP since, by definition, population is given.
3. A country's GDP (Y) is defined as
Y = Consumption spending + Investment + Government spending + (Exports-Imports)
Result 1.
First we rule out the possibility of having one country: One countries exports are another countries imports so in the aggregate, they sum to zero, however, if countries trade, then they specialise in the goods they export and if there is specialization the sum of the parts are greater than the whole. If you are not conviced, think of how an assembly line works: its the specialization that allows such mass production. Without it the output would be much lower. This is known as the gains from trade.
So we now have to have more than one country, because in our quest to maximize output, we needed to specialize.
I have to think a bit to continue the argument.
What makes you think the Arab-Israeli conflict is irrational? Competing for water and land resources is a pretty rational thing, in my opinion. And if the lifestyle you wish to lead means others must be poor, then you do what you need to do.
Rationality is, of course, undefined.
The scenario you paint is not very different from the current world. If you take a Marxist view of history, socio-economic conditions have always been at the root of conflict.
Those bullet points are far to economistic for their own good. Per-capita GDP can be high even with a powerful ruling class and a huge number of serfs. (How about examining the variance?) And the stipulation that everyone abide by international law is absurd -- we're trying to achieve that situation, not assume it. Any differences in GDP between countries will lead to conflict -- immigration, refugees, cross-border theft/poaching/conquering... there is no "rational" reason to accept these postulates, even as a limiting case.
I wish to make the point again that racial, cultural and religious differences are not some veneer painted on top of people that can be removed leaving all else intact. In the absence of other reasons, people will fight about...whether they like Prog or not! (Consider football hooliganism, anti-abortion vigilantes, drug wars, class wars etc.)
@han: I now see your point that the question is too hypothetical.
So let me get this one step at a time
You saying that this question does not make sense because we simply cannot remove only racial etc. differences and leave everything else intact
I had no interest in the 'achievability' of this situation. I wanted to explore the dynamics of country size and nothing else. The question could have been about Lemmings on a planet similar to Earth for instance. Maybe thats why conflict is something I have not considered.
But you do raise an interesting point. There has to be some kind of reference point. There is no way I will know how people/Lemmings will behave in the future as my assumptions will create a society whose behaviour I have no idea about.
Well anyway.. I guess that there is no way for me to separate this out to get a view of what would happen purely economically.
Another thing. I think that maximizing per capita GDP is a worthwhile objective. An objective of minimizing variance (creating equal countries) suffers from the problem that resources are not evenly distributed and so like you said you will end up with inequality anyway.
Han, perhaps you could try suspending your disbelief and approaching this as a hypothetical question? It's an attempt at construction of a simple model. It can help in understanding reality as well as in subsequent construction of more advanced models. Whether humans are in reality or could ever be homogeneous is not in issue here. I'm not criticizing you; I respect your responses and would like to see them applied to the problem rather than its grundnorm.
On the other hand, Sadman has not framed his question very well either, failing even to state the definition of optimal, upon which the entire model will depend.
People's responses here in that regard are quite indicative of their inherent outlook on life. I find it quite interesting.
Han takes the concept of entropy that he was exposed to as a physics student and (correctly) applies it mutatis mutandis to human behaviour, something he is studying at its very fundamentals.
Dismal falls into the trap of all economists and starts talking about macroeconomic quantities and international trade. The Hairy Snail could have taken him on, but opts to talk instead about a dead hippie. Oh no, wait-- he does take him on later on. I don't quite understand what they're arguing about.
Raven chips in with a reference to our common love, and surprises with a pre-condemned (by virtue of being from cinema) insight into human behaviour that I would have expected someone like me to bring up, not him.
And as for myself, I attempt to analyse the problem from the point of view of jurisprudence, or legal theory. GDP and the other economics that I barely believe in would not have entered my definition of optimum at all. The purpose of law (as differentiated from its object) is to regulate human behaviour and relations inter se. Speaking on a global scale, the ideal situation (the optimum) of which Dismal speaks is, from the point of view of legal theory, one that minimizes international conflict.
Or is it? I need to think more on that before I can proceed with my thesis-- or, at any rate, proceed while bringing to bear my unique viewpoint on it. International law is, like any border region of a science/art, a very murky area with strongly justified conflicting viewpoints. It has been called "the vanishing point of jurisprudence," and I'm not sure I don't agree.
Besides, I'm rather drunk, and I can't concentrate any more.
I do appreciate the Discussion Topic, though. Sorry about the rather useless comment. I hope to come back to the Topic when I am less drunk and have less exams. And may there be more Topics in future! Saale bhehnchod.
Where's Tee?
I should be working, but this thread is good fun. Very apposite comment, Perakath! Not drunk-sounding at all. (Though I don't know what you mean about entropy here.)
I like the idea of creating toy models to help our thinking. However, in science this usually serves a didactic role -- Feynmann's thought experiments were based on intuitions developed after the mathematics was explored.
What I mean is, suppose we construct a hypothetical model of country size effects independent of other factors. What then? The economic axioms we use are largely suspect from a predictive point of view to begin with. They assist post hoc analysis. So will analysing a toy system using iffy axioms give us any traction with the real problem?
Consider a mathematical analogy. Reducing dimensions can simplify certain problems. However, reducing from a nonlinear to a linear system will not only simplify, but also give you qualitatively incorrect results.
Such toy models are then like certain branches of pure maths -- interesting in themselves, but of no clear utility.
So we could restate the problem as one applying to video games. The benefit would be that you could study Age of Empires or Starcraft and examine how humans behave in restricted game conditions.
Hahaha... I loved the swipe at Maths there! Hell is a Real Analysis class when you haven't prepared for it. Number theory would have driven me up the wall if I'd studied it, I'm sure.
Good point, your point. My point (or Point of View) is that for people like me, who are novices in logic and rational deduction, the simplistic models provide a nice training ground in which to stretch one's reasoning muscles.
(I didn't know that about Feynman and the other thought experiments, btw. I REALLY disliked the thought experiments we 'learnt' in UG physics, especially the ones to do with optics. "Thought experiment"-- was there ever a worse oxymoron?)
What I meant about entropy-- your thesis that there will always be conflicts between humans, that there will always be inequality among nations, and the implicit corollary that nothing we can do will stop inequality (which is one of the moral justifications I give myself for not feeling bad about the existence of poor people, something I blogged about recently), reminds me greatly of the concept that a closed system will almost always progress from order to disorder, albeit if only because our definition of "disorder" is satisfied by so many more states of the system than there are states we label "order". (Physics can be such bullshit sometimes!)
And finally, we need look no further than our own imagination to construct worlds of little homo-men. Video games are the scourge of humanity. Bah.
Re: entropy. The Second Law only holds for a closed system, and the universe is the only known closed system (by assumption, mostly). While the planet seems mostly closed, the sun's radiation and magnetic effects have massive implications, as do the effects of the moon.
My criticism of optimality is not based on cynicism regarding progress towards equality. A casual glance at history will reveal that for vast swathes of humanity, life is better than it was for their great-grandparents -- along pretty much every dimension. Black people have a rough time in the US, but they are no longer slaves. Class may linger in the UK, but compared to the 1920s it has been eradicated. The caste system is Kerala has also taken a sound beating since the 1870s or so. And workers' rights have improved vastly in the past 100 years. In the 1880s child labour and 7 day 18 hour-a-day workweeks were common in the west.
We may not have reached our ideal goals, but all the little victories along the way add up.
@spae: That was a pretty sharp comment whether you were drunk or not! Interesting that your idea of optimality revolves around minimizing international conflict.. I would have never thought about that. If that were the goal, I don't know if any reorganization of countries would be able to achieve it.
@han: well, yes this is a toy model. I wouldn't go so far to say that such experimentation is completely useless. I realize its far from ideal but economics really is this complex of interactions all working simulatneously. Imposing restrictions will help isolate relationships. Rather than thinking of it as a linear approximation of a non linear relationship, I'd think of it as a partial analysis, something like given a 3D figure, holding z constant to see the relationship between x and y. Like slicing through a bowl and finding a parabola. That parabola is not the whole story but it is a part of it.
Even the mathematics is just functional representation of assumptions. The reasoning and logic is the same. This is the way all economics has been done. The only way to meaningfully turn this on its head would be to provide an alternate way to study it.
If we don't even try this we only reach the conclusion that nothing we do distributionally can improve human welfare as there will be an inherent tendency to fight for resources or something like that.
Alternatively, if we find that the size of the country that maximizes GDP depends only on say the distribution of natural resources, or population density, I'd say thats a useful starting point. We could then relax the assumptions we've made one by one and see where and why this result breaks down.
Hmm. I'm a bit suspicious of how much this kind of social science has in fact established, for the following reason: economic (and in general social-scientific) reasoning is used to convince policy-makers, businessmen and others to undertake certain courses of action individually, in the hope that their synergies (if you'll pardon the expression) will result in a desirable outcome. However, in the absence of control experiments, it is unclear if these courses of action and their outcomes can be attributed entirely to the reasoning used to convince people. The same reasoning usually has little predictive value. In important ways, each company or country is on a unique trajectory in time and space that is not part of any distribution. Also, commerce and statecraft were happening in the absence of theoretical bases for thousands of years.
I guess you already know what I'm talking about, so I'll stop being a buzz-kill!
So what's the first-pass intuition? Given a region with a certain population and resource distribution, how best should they be governed? Do large centrally-ruled countries have lower "operating" expenses than small independent states or loose federal republics? I don't know the data, but I imagine defense would be cheaper if it were centralized for a vast area. However, large powerful countries spend disproportionately highly on defense.
What about inequality? There are small rich countries and small poor countries. Rich countries with no resources and poor countries with plenty. If we were to plot this kind of data, we would probably find no trend lines.
If you define optimality by per capita income, what's on the x-axis? Countries of pretty much any size can have pretty much any per capita income. You might come to the conclusion that being in a white Western democracy -- irrespective of size -- is the best solution to optimal per capita GDP!
So the question is... what is the question? Is there some type of data that has not been collected yet that might reveal a size-related trend? Is there a definition of optimality that will confer a trend onto the data?
True, there would be nothing better than the ability to do control experiments. That being said,you'd be surprised how 'successful' this approach has been in the past. Its a little shady but, the thing is that we look at aggregated data. We are looking at things on average. So if a majority of people act in a certain way and we predict it, that would be a 'success'. There is room for error. The Popperian question of falsification is a big problem in my opinion.
The solution for the Great Depression, Successful management of Inflation all these years, currency crisis management have been evidence either that these models have got some of the basic relations right. It could be chance too, I don't think that it can be completely ruled out.
Anyway, here's another way to look at the problem which may make things easier. Let's look at it at the margin (a fav phrase). Given a country, if it were to increase/decrease its geographic area by a small amount, there would be costs (that would reduce GDP) and benefits (that would increase GDP). What we need to know is what outweighs what. If benefits are larger, we should increase the size.
So the trick is to specify the costs and benefits properly. Like you said, inequality could be considered as a cost as it would take away from GDP in the form of lawlessness or something. So to start, we would look at the simplest forms of costs and benefits and add further variables to complicate things.
Does this approach make sense?
Makes sense, but is there any appropriate precedent? The British Empire got smaller, but we can't really say how its GDP was affected because of WWII. The US got bigger and richer at the same time. Going back in history, we know that the time of an Empire's largest geographical extent is not necessarily its "Golden Age".
In order to develop some intuition about the changes that might occur from marginal increase/decrease, what information do we use?
Post a Comment