A Data-Driven Model of a Firm's Operations With Application to Cash Flow Forecasting

A firm’s cash flow from operations is a function of the contemporaneous and lagged values of its operational variables---sales, operating cost, inventory, payables, etc. Estimating this function is important for forecasting and managing cash flows. However, cash flow forecasting is a challenging problem. In this paper, we propose a generalizable and data-driven model of a firm's operations to disentangle this endogeneity and estimate causal impacts among variables. By estimating our model using quarterly public financial data from S &P's Compustat database for 1990-2020, we obtain several results. First, we provide evidence that cash flow has both endogenous and lagged relationships with sales and inventory. Second, we show that lagged operational variables significantly improve the accuracy of cash flow forecasts compared to an autoregressive model of prior period cash flows alone. Moreover, cash flow also helps improve forecast accuracy for sales and inventory. Third, our model helps quantify the short- and long-run impacts of structural shocks in variables on the entire system. These estimates are useful to assess the effects of exogenous macroeconomic shocks such as the Great Recession on future cash flows and operational variables and provide a joint distribution of variables that can be used as an input in operational planning.