A STUDY OF THE ROLE OF REGIONALIZATION IN THE GENERATION OF AGGREGATION ERROR IN REGIONAL INPUT-OUTPUT MODELS* Michael L. Lahr Center for Urban Policy Research, Rutgers, The State University of New Jersey, New Brunswick, NJ 08901-1982, U.S.A. E-Mail: lahr@rci.rutgers.edu Benjamin H. Stevens Benjamin H. Stevens died December 9, 1997. For more information, see Miller (1998). ABSTRACT. Although the need for aggregation in input-output modeling has dimin- ished with the increases in computing power, an alarming number of regional studies continue to use the procedure. The rationales for doing so typically are grounded in data problems at the regional level. As a result many regional analysts use aggregated national input-output models and trade-adjust them at this aggregated level.In this paper,we point out why this approach can be inappropriate. We do so by noting that it creates a possible source of model misapplication (i.e., a direct effect could appear for a sector where one does not exist) and also by finding that a large amount of error (on the order of 100 percent) can be induced into the impact results as a result of improper aggregation. In simulations, we find that average aggregation error tends to peak at 81 sectors after rising from 492 to 365 sectors. Perversely, error then diminishes somewhat as the model size decreases further to 11 and 6 sectors.We also find that while region- and sector-specific attributes influence aggregation error in a statistically significantly manner, their influence on the amount of error generally does not appear to be large. 1. THE GENERAL AGGREGATION QUESTION Economic researchers have always had the problem of aggregating data to a “reasonable” level so that economic interactions could be analyzed effectively. As Leontief states, “the practical choice is not between aggregation and non- aggregation but rather between a higher and lower degree of aggregation” (1949, © Blackwell Publishing, Inc. 2002. Blackwell Publishing, Inc., 350 Main Street, Malden, MA 02148, USA and 108 Cowley Road, Oxford, OX4 1JF, UK. *Some of the discussions in this paper have been presented in more primitive form as early as May 29, 1987, in Binghamton, New York. The first thoughts on the subject first appeared in 1990 as RSRI Discussion Paper No. 130. We gratefully acknowledge the comments of Rodney C. Jensen and Ronald E. Miller on that rendition. In 1993, equipped with a full set of simulation results, it was republished as RSRI Discussion Paper No. 134 with the author order reversed. We gratefully acknowledge the comments of an anonymous referee and Erik Dietzenbacher, in particular, on that paper and its revision. In incorporating the two latest sets of comments to form the present paper, Lahr worked without the benefit of his senior author’s wisdom, charity, or approval, and yet all the while recalled his different points of view on some of the paper’s aspects, even though they were echoes of conversations over three years old. Received December 2000; revised May 2001; accepted August 2001. 477 JOURNAL OF REGIONAL SCIENCE, VOL. 42, NO. 3, 2002, pp. 477–507