Technological Change and Investments on a Rugged Landscape - An Agent Based Simulation

In this paper, we present an agent-based simulation model of technology investment by heterogeneous and interacting financial agents. Investment decisions are explained by the topology of the technology landscape, the agents' capability to receive and interpret incomplete landscape information, and their investment capacity. We are particularly interested in the effects of different information-sharing an co-investment network structures among agents on the rate and direction of technological change. We model these agents as to observe emerging technologies on a technology fitness landscape, and select potential investment targets according to their perceived risk-adjusted returns, where risks are a function of the technology's maturity and the returns of the achieved technology performance. Subject to imperfect information and bounded rationality, financial agents are heterogeneous in their view of the landscape determining the potential investment targets they are able to spot, as well as in their forecasting ability determining the accuracy of their prediction of achievable technological fitness. Assuming a trade-off between search radius and forecasting ability, the population of financial agents will consist of more specialized investors with a narrow view on the landscape but high forecasting ability within this area, and more generalized ones who can search a large area but have a low forecasting ability. We observe which configuration of financial agents lead to high rates of technological change and diversity. In a next step, we introduce investor networks and allow agents to co-invest together in order to pool financial resources and get access to their forecasting capability in a specific technological domain. We compare which investor network structures and compositions lead to the high rates of technological change and diversity on a given technology landscape. Results from a Monte Carlo simulation indeed indicate networked investor population to outperform isolated investor performance, an effect that tends to increase with complexity of the technology landscape