Agent based modeling of macroeconomics
This code contains the backbone for running the ABM models of a simulated world with market exchange. Further components available for collaborations. The simulation assumes three components:
-
Firms: They are the production power of the simulation. They can hire staffs, make product, buy ingredient from the market and sell their product. They can choose to use the profit to expand their companies or use it for investment.
-
City: The city contains two roles. One is the role of the state, that it controls how much "cash" is flowing in the society by deciding the amount of purchase through its liability. It is also responsible for collecting. The second role is the workers. Firm employ the workers for their production. The workers receive the payment and spend it in daily household items, which are also coming from the firms. In this simulation, the workers are assumed to have the same salary.
-
Market: The centralized place for trading. All firms and workers can buy what they need from the market. The trade is done in the implicit way. The market records first how much demand and supply occurs at that day. Then update the price. And the firms and workers pay for the trade using the updated price. This guarantees that if there are some unusual demand or supply from the users, the user will have to pay a premium for changing the market.
-
World: In a more generalized context, we can consider trade not only inside a city with the market, but also among cities. In that case, we need to generalize the market structure that it can accept trade from different cities inside the simulation. This allows firms sell their product in other cities other their theirs. For inter-city trade, we can also introduce the concept of Forex that for profit or expense done in other cities, the firms need to exchange for the local cash. This generates the forex market wh 5884 ich can be further established.
In this simulation, the price is assumed to follow the non-equilibrium pricing scheme, where the price changes according to the local demand and supply of that day. It is possible that this may over-estimate the price variation for some items. However, to make sure the code can finish one simulation within an hour. I choose this simplified scheme instead of the more complicated but realistic market book scheme.
Some of the components are implemented in the previous (but not so stable) version and after checking that the system is stable against different setting, I will put them back in this version.
-
A tax system with public welfare: In the current version the city has almost no role in the society except printing money. In fact, a more realistic system should also consider the macro-economics, how willing the workers are spending money, and this will lead to consideration of how "good" the society is, for example, education, health care and so on.
-
An investment tool: Since this version contains the backbone of the code, it still needs a good investment channel to remove excess cash in the market. The investment can also generate secondary employment which also affects the global employment rate. The investment tool can be viewed as also the indicator to the sentiment of different cities.
-
A more generalized product chain: In this version the minimal product chain (only 3: Water, wheat and bread) is included to make sure the firms and workers can trade successfully. To create a more complex system where more non-linear relation appears among firms, we can include a more genearlized product chain.
Enjoy simulation! 😃