Table of Exhibits

Exhibit 1: Price-Expenditure Lines and Market Share

Exhibit 2: Market Demand Model Findings

Exhibit 3: Share Capture Model Findings

Exhibit 4: Annual Falling-Price and Public Program Effects...

Exhibit 5: Electronic Fluorescent Ballasts Market Transformation

Exhibit 6: Estimated Wattages for Magnetic and Electronic Fluorescent Ballast Units

Exhibit 7: Shipments and Weighted Average Ballasts & Lamps Watt Reductions ......

6

13

15

17

18

21

21

Exhibit 8: Impact of Green Lights Partnership and Other Programs, 1991 through 1999 .... 22

1. Introduction

The focus of this study is on estimating the extent to which the Green Lights Partnership was responsible for mitigating greenhouse gases caused by electricity generation. The Green Lights Partnership did so through its efforts to transform the national market for energy efficient lighting equipment. Started in 1991 and superceded in 1998 by the ENERGY STAR Buildings Partnership, Green Lights was a voluntary public/private partnership that encouraged building owners and operators to adopt energy efficiency products and services.

SM

With the Green Lights Partnership, the United States Environmental Protection Agency (EPA) carried out its climate protection mission to promote energy efficiency in commercial buildings. In return for technical assistance and public recognition, Green Lights partners committed themselves to installing high efficiency lighting products when they might otherwise have installed less costly, conventional lighting products. The products that were encouraged by EPA included high efficiency electronic fluorescent ballasts, high efficiency fluorescent lamps, compact fluorescent lamps, non-incandescent exit signs, and automatic lighting controls. In addition, EPA encouraged energy efficiency practices, such as delamping in building areas that were overlit. To become a Green Lights partner, building owners agreed, through a memorandum of understanding, to make lighting efficiency upgrades over a given time period to 90 percent of all their upgradable building spaces. EPA offered no financial incentives of any kind.

Although formally merged into the ENERGY STAR Buildings Partnership, the commitments received by the Green Lights Partnership continue, to this day, to be fulfilled. The new partnership is more ambitious and comprehensive in scope, promoting whole building energy efficiency rather than just lighting energy efficiency. In the seven year operation of the Green Lights Partnership, approximately 2,000 organizations participated by becoming Green Lights partners, and another 1,000 lighting-related businesses became Green Lights allies.

The Green Lights Partnership is the first and foremost of an increasing number of public programs whose primary purposes are, through voluntary cooperation, to permanently transform the markets for energy efficient products and services. At present, most market transformation programs are designed and operated by state or regional agencies and are funded by public goods or public benefits charges collected by local electric utilities. As such, they are meant to achieve local and regional goals. Unlike these newer programs, the activities and goals of the Green Lights Partnership, and its successor, the ENERGY STAR Buildings Partnership, are national in scope. For this reason, the focus of this study is on estimating the extent to which the Green Lights Partnership reshaped the character of the national market for fluorescent ballasts. The effectiveness of EPA's program in mitigating greenhouse gases and protecting the environment is

The concept that is central to this study of market transformation is energy efficient product market share. For this study, market share is defined as the fraction of the total number of units sold that are energy efficient. Focusing on how market shares change, and moreover, on understanding the factors causing an energy efficient product's market share to change, suggests the need for a certain kind of research design. In an economic sense, a competitive, efficient market is one in which the relative prices of highly substitutable products affect a product's market share. Other factors, such as loan rates, general economic growth and public programs, are also likely to influence market shares. Hence, for this study of the transformation of the energy efficiency lighting market, the focus of the analysis is on the extent to which market share changed due to market forces and public programs.

Several advantages are obtained by narrowing the subject of this evaluation of the Green Lights Partnership to the market for fluorescent ballasts only. First, fluorescent lighting ballasts are the one product for which there are 41 years of continuous, publicly-available national manufacturer shipment data that are differentiated by categories of product quality. Second, fluorescent lamps, which are driven by lighting ballasts, are ubiquitous in commercial and institutional buildings in the United States. Third, installation of electronic ballasts is usually accompanied by the use of T-8 fluorescent lamps, another major energy efficient lighting product promoted by EPA's Green Lights Partnership. Fourth and lastly, electronic ballasts and T-8 lamps together account for the majority of the pollution prevention that can be attributed to the Green Lights Partnership.

To put this study in perspective and show its contrast with prior studies on the same subject, the next section of this report offers a brief overview of past energy efficiency program and product research. This is followed by Section 3, which contains an explanation of the theory underlying the present market transformation evaluation research design and describes the empirical models and their findings. Section 4 describes the conversion of the empirical findings to program accomplishments and climate protection impacts. The last section offers recommendations for future research activities that can be undertaken to evaluate market transformation programs.

2. Overview of Related Research Designs

Most market transformation programs differ in character from electric utility demand side management (DSM) programs. Market transformation efforts are designed to ease market frictions that hold back growth of the supply and demand of energy efficient products. They do so through research, educational, training and marketing activities that help the targeted markets expand and become permanent -- most often without offering financial incentives to any market actors. DSM programs, on the other hand, revolve around acquiring, i.e. purchasing, least-cost energy resources by issuing rebates to program participants for buying energy efficient products and services.

Over the past two decades, evaluation of the net impacts of DSM programs was the responsibility of local utilities who planned, designed and ran them. In return for providing reliable evidence of program impacts, utilities received program cost recovery, and often, lost revenue recovery

70-630 D-01--4

and/or financial rewards. The methodological approaches to these impact evaluations usually involved the collection and analysis of program participant and non-participant microdata. For studying energy consumption, everything from monthly billing data to 15-minute interval metering data of electrical circuits or equipment were collected. Also, extensive use was made of mass customer telephone and mail surveys, building audits, building engineering simulations and equipment tracking databases.

The goal of most DSM program impact evaluations was to estimate the net energy savings that could be attributed to program. This involved removing from the estimated energy consumption impacts those energy savings that would have occurred without the DSM program. The two main statistical techniques for analyzing the evaluation data and estimating net program impacts were the energy impact modeling framework and the net-to-gross method (Horowitz 1995). An important element of these methods is their capability of accounting for the extent to which participants may have been free riders or may have self-selected into the program. The markettransformative effect of a DSM programs has only in the past decade become increasing recognized as a potentially significant program benefit. As a result, quantitative techniques for rigorously estimating the market-transformative effects of DSM programs have rarely been applied to impact evaluations.

In the ongoing evolution of the market transformation program evaluation area, recent program evaluations have focused on determining the causes of resistance to product adoption and how market transformation programs have overcome, and might continue to overcome, that resistance. The geographic scope of these kinds of evaluations have been statewide or regional. Studies at the local level, or conversely at the national level, have been rare. The subject matter of this recent research has tended to be attitudes, opinions, and anecdotal information regarding the character of the product markets. Surveys of consumers and upstream market actors, such as manufacturers and distributors, have been used extensively. One of the most-widely known bodies of work of this kind are thirteen different studies undertaken in California, between 1995 and 1999, sponsored through the California Demand Side Management Measurement Advisory Committee. They provide in-depth investigations, and baseline information, for many different energy efficient products and services markets spanning the residential and nonresidential sectors.

Other recent market transformation studies include those sponsored by the Pacific Northwest Energy Efficiency Alliance and the Northeast Energy Efficiency Partnership. The subjects of these regional studies are changes in the markets for various residential and commercial energy efficient products and services. By and large, these studies shift the energy efficiency program evaluation paradigm away from the analysis of energy savings to the analysis of market baselines, market movements and their determinants. Like the California studies, they concentrate on market processes studied through surveys and interviews with upstream, midstream and downstream market actors.

In addition to the present evaluation of the Green Lights Partnership, three published studies have used Bureau of the Census national fluorescent ballasts market data to investigate the