The Persistence of Profits in the Long Run: A New Approach

Most research on the persistence of profits models their evolution as a first-order autoregressive process. We present an alternative framework using structural time series analysis, decomposing the overall series into trend, autoregressive, cyclical and irregular components. A persistence-of-profits taxonomy is introduced, based on the level and slope of the long run trend in the final vector state. Short run persistence is tested in autoregressive and cyclical form. Rich dynamics are revealed for a sample of 53 UK companies over the period 1968-1999, around 60% exhibiting non-eroding long run persistence, and three quarters short run persistence, predominantly cyclical. Address for correspondence: School of Management and Business University of Wales Aberystwyth Cledwyn Building, Penglais, Aberystwyth Ceredigion, SY23 3DD, Wales, UK e-mail: [email protected] JEL Classifications: L12, C32 Acknowledgements We thank David Jevons and Gillian Doyle for research assistance. We are also grateful for comments and suggestions from conference participants at the IIOC, Boston, April 2003, from Dennis Mueller, and from seminar participants at UWA and the Welsh Economics Colloquium, Gregynog. We take responsibility for the opinions expressed and any remaining errors. Note: SMBA Research Papers Copyright 2003 are prepared for circulation to interested readers. They often represent reports on work in progress and should therefore not be referred to in published work without the consent of the author.