On the fault (in)tolerance of coordination mechanisms for distributed investment decisions

The efficient allocation of scarce financial resources lies at the core of financial management. Whenever humans are involved in the allocation process, it would be reasonable to consider abilities, in order to assure efficiency. For the context of coordinating investment decisions, the competitive hurdle rate (CHR) mechanism (Baldenius et al. in Account Rev 82(4):837–867, 2007 ) is well established for allocating resources. This mechanism is derived from an agency model, which, as is the nature of agency models, assumes agents as being fully competent. We employ the agentization approach (Guerrero and Axtell in Emergent results of artificial economics, Lect Notes Econ Math, vol 652. Springer, Berlin, pp 139–150, 2011 ) and transfer the logic behind the CHR mechanism into a simulation model, and account for individual incapabilities by adding errors in forecasting the initial cash outlay, the cash flow time series, and the departments’ ability to operate projects. We show that increasing the number of project proposals, and decreasing the investment alternatives diversity (in terms of their profitability only), significantly decreases the fault tolerance of our CHR mechanism. For misforecasting cash outlays, this finding is independent from the error’s dimension, while for larger errors in forecasting cash flows, and the departmental ability, the impact of diversity reverses. On the basis of our results, we provide decision support on how to increase the robustness of the CHR mechanism with respect to errors. Copyright Springer-Verlag Berlin Heidelberg 2015