Who's going to provide the funding for high tech start-ups? A model for the analysis of determinants with a fuzzy approach

The aim of this paper is to understand which factors influence the financial structure of Italian young, high-tech, innovative firms, and to attempt to formulate a predictive model to determine the ideal financial strategy for a given entrepreneurial project. Venture capital is the most relevant form of financing for high-technology start-ups in the United States and is frequently cited as crucial in the technological leadership of the US economy. However, banks are also moving toward establishing a role in capital provision, making innovative offers to meet the financial needs of start-ups, especially in bank-centric countries such as Italy. Is it possible to build a robust and ordered set of determinants of the financial strategy of new technology-based firms? Is it possible to gather them in a model that allows a rigorous analysis? Is it possible to summarize the analysis in a synthetic value of orientation to one or other form of financing? Through a systematic review of the literature and comparison between investors, we have taken the first step toward answering these questions. This work develops a methodology to solve the problem and builds a provisional fuzzy-set based tool to permit the rationalization of the relevant information and effectively support the reduction of qualitative evaluation of complex phenomena into simple and measurable dimensions. The structure of the model is hierarchical but simple. We consider, as the first level of main determinants (sub-dimensions), the figure of the entrepreneur, the nature of the project, the financial scenario, and the market characteristics. For each of these, we provide deep insights about their relation with finance. We have verified the value of this approach in the context of ten business cases, by matching the financial strategies undertaken by entrepreneurs and the obtainable conclusions with the fuzzy tool. However, the definition of a robust, predictive model requires more consistent empirical validation, which we intend to develop from this work.