In contemporary egalitarian societies, it is considered desirable for parents to distribute resources (i.e., time and money) equally among their children. But how does a strictly equal distribution of resources (per investment interval, such as day or week) affect the cumulative distribution of resources? Does it produce genuine equality? And how do other investment strategies (e.g., always attending to children’s immediate needs, independent of fairness considerations) perform in comparison?

Such questions are difficult to answer. Field studies would have to be run for decades in order to muster the necessary data, and they would rely on parents’ and children’s subjective reports.

How does simulation help?

A powerful alternative is an “agent-based” simulation approach. It starts with reasonable assumptions about the behavior of the agents involved (working, sleeping, and recreation times; income and financial requirements; etc.). These specifications allow “individuals” to be characterized in a way that is understandable to computers. “Families” can then be composed of individuals who vary in key dimensions (e.g., high or low income, single parent or not, number and age distribution of children).

The development of resource allocations over a period of decades can then be simulated within a few seconds, and different family characteristics and allocation strategies can be compared.

What is the outcome?

The scientists found that the results of their simple simulation model reproduced the pattern observed in the available real-world data. Specifically, their counterintuitive finding is that an equity motive produces a fair distribution at any given point in time, but an unequal cumulative distribution of investments across time*.

A basic rule of modeling is proper validation and verification (V&V) before results are generalized to the real world. The V&V process relies primarily on further conceptual analysis and empirical data. Since the complex simulation scenarios presented in this exhibition go far beyond the original model* and real-world data is scarce, their results have to be taken with a grain of salt. Nevertheless, because the assumptions underlying the models seem reasonable, they can be used to generate new hypotheses, ideally leading to innovative research questions**.