The economics and algorithmics of an integral settlement procedure on B2B networks

We develop the economic and operational foundations of a new method of financing companies’ financial obligations. In this new banking business model, a network funder sets an optimal combination of netting and financing. Given a network of companies and their respective invoices, and under the condition of a full settlement of the invoices, the netting procedure consists in applying a multilateral netting algorithm to the network, conceived as an oriented multi-graph. This algorithm introduces a new method of exploration: during regular periods of time (i.e. monthly, weekly, or even daily sessions), the set of invoices is found that maximises the amount of debt offset, given a quantity of loanable funds. From a systemic point of view, the algorithmic exploration of the multigraph is subject to optimisation constraints. The exploration finds those configurations which allow the network funder to manage a policy trade-off between the maximisation of both the total value of invoices and the number of companies involved in netting, and the minimisation of the amount of financing needed to settle payments in full. To test our method, we use an empirical dataset from Infocert (electronic invoices operator) consisting of more than 60,000 companies. The policy trade-off shows that it is economically significant and feasible for a network funder to reduce the financial need of about 50% of companies by about 45% of the total amount of their financial obligations.