Integrating input-output (IO) analysis and econometric model has become increasingly popular for it combines the advantages of both models (Isard & Anselin, 1982; Moghadam & Ballard, 1988; Beaumont, 1990; Coomes et al., 1991; West, 1991; Treyz, 1993; Israilevich et al., 1996; Rey, 1998; Motii, 2005). There are a number of strategies through which an IO analysis is merged with an econometric model. This paper examines two strategies to build an integrated input-output econometric (IO-EC) model, namely coupling and holistic embedding, from both theoretical and practical perspectives. Each strategy is analysed based on their simulation and forecasting performance. Both strategies are applied to investigate the economy of the Illawarra region of New South Wales, Australia. This paper conducts a series of ex-post forecasting experiments. Each strategy is examined based on their forecasting performance and the accuracy of their impact analysis. Each strategy is applied to a hypothetical scenario of increased government expenditure on knowledge sectors, to investigate their significance in terms of sectoral employment, value added, income, and gross regional output.