powered by NetLogo
view/download model file: dct-s.nlogo
This model tries to explore whether there would be a tax competition among countries (counties) if people could move with no restriction other than a cost of emigration and if there is a democratic choice of redistribution in each country.
At a first glance it may seem obvious that there must be a tax competition among such countries as long as the cost of emigration is low. The reasoning might be like this: if there is more redistribution in one country then in the other one, its more productive inhabitants (who pay more taxes than receive transfers) would leave the country. The least productive inhabitants of other countries would come to enjoy this country's generous transfers. This should force the country to lower both its taxes and its transfers, i.e. to lower its redistribution.
However, this reasoning is false for a simple reason: it assumes the country is an entity that reacts to the outflow of the net payers and inflow of the net receivers in such a way as to keep the first and discourage the later by a change of the tax/transfers system. But this is not true if the level of redistribution is decided democratically and if the voters are only boundedly rational.
In such a case, if an inhabitant emigrates from one country to another one, it makes two effects (both in his former and his new country): 1. few taxes are paid and few transfers demanded in his former countery, and more in his new country and 2. it changes the number of voters in both countries---and also the structure of their electorates.
Let us suppose that one inhabitant in one country believes she could be better off in another country (e.g. because she is a net payer and the tax rate is lower in the other country). If she emmigrates there, the tax revenue in her former country decreases, and the transferes must decrease there as well. However, this need not to force the other inhabitants to decrease the tax rate. Quite contrary, they may even raise the taxes because one person opposing the tax increase was eliminated. If this happens, other inhabitants of the country may emigrate, and the process goes on---possibly forever.
This section could explain what rules the agents use to create the overall behavior of the model.
The "Setup" button sets the stage up. Several parameters affect the model initialization. The "#-of-agents" says how many agents will be created. The "default-lazyness" says how much the agents dislike the labor; the higher number, the more they prefer leisure. The "consumption-cost-of-emigration" sets a treshold that must be climbed for the agent to emigrate (this is substracted from his consumption).
The "number-of-initial-tax-haggling-rounds" says how many rounds of haggling about taxes will be used in each country in the set-up process. The more, the more is likely that there will be the tax rate preferred by the median voter. The "number-of-tax-haggling-rounds" means the same for each tick.
The "Go" button starts the simulation. You can visualize several things (see the radio-button).
This section could give some ideas of things for the user to notice while running the model.
Try to move the "comsumption-cost-of-move" and observe the behavior of the model. Notice that there are faze transitions.
There are some ways in which I plan to extend the model but I don't want to discuss it here. :-)
The model involves no special NetLogo feature.
I'm not aware of any other similar model but I have to admit that I've skipped the literature review. In a sense, the model is similar to Schelling's famous Segregation model (available in the NetLogo Models Library). In both his and my models a move of one and only discontent person can create a long serie of further moves of other persons. However, if the "consumption-cost-of-move" is small enough, this model need not have a steady state.
This section could contain a reference to the model's URL on the web if it has one, as well as any other necessary credits or references.