An Examination of Alternative LQ-Based Approaches to Computing Regional Input–Output Coefficients

Abstract In this paper, we examine alternative methods of computing regional input–output (IO) coefficients, with an emphasis on their relative accuracy and the complexity of the computations required. Our focus is on the well-known FLQ (Flegg’s location quotient) approach. Although the FLQ formula often yields satisfactory results, the need to specify values of the unknown parameter δ in this formula presents an obstacle to its implementation. After examining the FLQ’s conceptual foundations, we develop a possible new approach that obviates the use of this parameter. Instead, the new formula, the hyperbolic tangent LQ or HTLQ, incorporates an alternative proxy for regional self-sufficiency in terms of a new parameter $$\mu $$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mi>μ</mml:mi> </mml:math> . We argue that, in most cases, it would be reasonable to set $$\mu =0.5$$ <mml:math xmlns:mml="http://www.w3.org/1998/Math/MathML"> <mml:mrow> <mml:mi>μ</mml:mi> <mml:mo>=</mml:mo> <mml:mn>0.5</mml:mn> </mml:mrow> </mml:math> . We test our proposal using the 2005 and 2015 Korean survey-based interregional IO datasets and contrast our estimates with both survey-based values and the results from several other techniques. The results suggest that the new formula can yield more accurate estimates of regional IO coefficients and multipliers, and in a more straightforward way, than is possible with the traditional FLQ.