Journal on Policy and Complex Systems
of the aggregate private sector . Namely , corporate tax reductions increase GDP when producers ’ surplus money ( increased by the tax reduction ) can be spent effectively in the market , in the form of firms ’ investment and / or consumption by executives and workers .
Third , inefficiency of government expenditure weakens the economy . In the model , the degree of inefficiency is defined as the ratio of firm subsidies to the total amount of public expenditure . In the actual system , the inefficiencies might be caused by many factors , such as public orders with higher-than-market prices , subsidies to firms in the industry , or rent-seeking behavior ( Tollison & Congleton , 1995 ).
5 . Discussions : The Validity of the Model in ABM
As described in the introduction , the validity of the ABM has been widely criticized . For example , Marks ( 2007 ) claimed that , because of the functional complexity of an ABM system , one could not assume that the factors that successfully reproduce the desired macro phenomena are necessary conditions .
On the contrary , the results of this study indicate that the necessary conditions exist for reproducing both business cycles and GDP reactions to tax reductions . The necessary conditions are the sets of factors that characterize the model structure , which can be elucidated by running a series of computer experiments where each of the factors is changed one at a time . These factors are indispensable for the model to reproduce the desired phenomenon as shown by the fact that the phenomenon does not emerge in the artificial society if any one of these factors is not included in the model in ABM . A typical example is the condition for reproducing the positive effect of corporate tax reduction . As the present research study revealed , four factors are required to reproduce the phenomenon because , among 16 possible combinations involving these four factors , only one case results in the emergence of the phenomenon , namely , the case in which all four factors are included in the model .
Moreover , by considering why such factors are required to reproduce each phenomenon , as described earlier in this paper , we can gain a better understanding of the causal mechanisms of these social phenomena . The reason for this is discussed below .
A system is a set of interacting objects and is defined as a proper relation on sets ( Mesarovâic & Takahara , 1989 ). In the case of social systems , objects that are responsible for the emergence of social phenomenon are autonomous decision-makers ( i . e ., agents , such as individuals and organizations ), because any social phenomenon is considered to emerge from agents ’ actions and their interactions . The set of the factors characterizing the agents ’ actions and their interactions is the system structure , which is defined as a set of categories of agents , their behavioral rules and relevant attributes variables . The behavioral rules and relevant attributes may include environmental factors of the system as well as factors
56