Accounting for Technological Change in Regulatory Impact Analyses: The Learning Curve Technique

Abstract

Executive SummaryRegulatory impact assessment is formally required by the U.S. and many other nations in order to help governments weigh the costs and benefits of proposed regulations, particularly as they compare to those of alternative actions and other government priorities. 1One of the "best practices" of regulatory impact assessments, as established by the OECD, is to use estimates of costs that are grounded in economic theory. Economic theory indicates that changes in compliance costs should be expected over time as a result of factors related to technological innovation. But many U.S. regulatory impact assessments have traditionally employed a practice that is in conflict with this expectation: they take current estimates of the costs of complying with a proposed regulation and project that those costs will remain unchanged over the full time period that the regulation would be in effect.There are a number of indications in the literature that ignoring technological change is an important explanatory factor underlying the often-observed tendency for regulatory impact assessments to seriously over-estimate the costs of compliance with new environmental, health, safety, and energy efficiency regulations. The regulatory reform movement has described the appropriate modeling of technological change as an element of "an ideal cost-benefit analysis," and in February 2011, this issue was raised by the U.S. government agency tasked with establishing and enforcing best practices in regulatory impact assessments -the Office of Information and Regulatory Affairs (OIRA) in the Executive Office of Management and Budget. At that time, OIRA called for federal agencies "to use the best available techniques to quantify anticipated present and future benefits and costs as accurately as possible," which include "identifying changing future compliance costs that might result from technological innovation or anticipated behavioral changes." OIRA, the OECD, and the academic literature do not currently specify the "best available techniques" for accounting for innovation in regulatory impact assessments, however. This report seeks to inform this discussion by focusing on the most widely available technique, which is a learning curve-based cost adjustment method. In vehicle regulation, the Environmental Protection Agency (EPA) and the National Highway Transportation Safety Administration (NHTSA) apply a component-based learning curve approach that the EPA began developing in the 1970s. In energy efficiency regulation for appliances and other products, the Department of Energy (DOE) has applied a whole-product learning curve-based price adjustment approach since 2011. This report: (1) provides an overview of some of the major findings of the academic literature on learning curves in order to inform an assessment of a "best" approach to a learning curve-based regulatory impact assessment cost adjustment technique; (2) describes the EPA-NHTSA and DOE approaches to this technique; and (3) assesses these approaches against the criter...

Modeling technological change and its impact on energy savings in the U.S. iron and steel sector

“…The Boston Consulting Group applied learning phenomena to analyze the relationship between the average unit price and cumulative output of 24 industrial products in 1968. Since then, this approach has been used in empirical studies in a wide range of sectors, including manufacturing [3,21], consumer products [22,23], energy supply technologies [24][25][26][27][28][29][30][31][32][33][34][35][36][37][38][39][40][41], energy demand technologies [42][43][44] and environmental control technologies [45][46][47].…”

Using learning curves on energy-efficient technologies to estimate future energy savings and emission reduction potentials in the U.S. iron and steel industry

“… manufacturing (Argote and Epple, 1990;Nadeau, 2010),  consumer products (Bass, 1980;Teng and Thompson, 1996),  energy supply technologies (Criqui et al, 2015;Goldemberg et al, 2004;Hettinga et al, 2009;Hong et al, 2015;OECD/IEA, 2000;Junginger et al, 2006;Li et al, 2012;McDonald and Schrattenholzer, 2001;Neij, 1997;Neij, 1999a&b;Neij et al, 2004;Nemet, 2006;Rose and Joskow, 1990;Rubin et al, 2006;Schoots et al, 2008;van der Zwaan and Rabl, 2003;Wene et al, 2005;and Yeh and Rubin, 2007),  energy demand technologies (Desroches et al, 2013;Weiss et al, 2010) and  environmental control technologies (Rubin et al, 2004;Taylor et al, 2013;and Yeh et al, 2007).…”

“…Using price instead of cost may be useful when price-cost margins are constant over time. This, unfortunately, requires a number of unlikely things to stay constant in an industry (Taylor and Fujita, 2013). Whereas cost changes occur over time due to changes in input prices and production efficiency, price changes can occur from several other factors (such as government subsidization, various marketing decisions, funding allocated to R&D, and commercialization initiatives).…”

Retrospective North American CFL Experience Curve Analysis and Correlation to Deployment Programs

“…Experience curves are a common framework for assessing technology learning and cost reduction with increasing production volume (Taylor, 2013). These curves are empirically found to follow a power law, with the rate of cost reduction a power law function of cumulative production volume:…”

“…Its definition should not be interchanged with the term "learning rate" since progress ratio typically refers to the percent cost remaining after a doubling in production, i.e. 100% minus the learning rate, which is the opposite of a learning rate (Taylor, 2013). If we assume the quoted progress ratio in fact refers to a learning rate of 10%, which is likely considering the comparison to Iwafune's learning rate of 20.1%, issues still remain.…”