Energy Star Appliances
EPA’s savings calculator exaggerates savings

Wall Street Journal reporter June Fletcher says (link temporarily available to nonsubscribers) it’s not easy being green at home because a lot of heavily-promoted household renovations and products cost more that they deliver in environmental benefits, including energy savings. “Most homeowners like the idea of going green,” she writes, ” — until they get the bill:

With home sales slumping and consumers rethinking their remodeling budgets, building contractors and suppliers are dangling green upgrades. They hope that energy-efficient systems and products made from sustainably harvested materials will hook consumers concerned about global warming, pollution and natural resources.

Yet with a few exceptions, green materials and construction cost extra, making them a hard sell. Enermodal Engineering, a Canada-based consulting firm, estimates the premium at 5% to 10%, depending on how extensively a builder uses recycled materials and water- and energy-efficient products. When Specpan, an Indianapolis research firm, surveyed builders recently for Building Products magazine, the greatest number estimated a 10% to 19% cost increase when going green.

Some consumers are willing to pay for green goods and services even if they are not cost-effective. Making significant headway into the market, however, requires that they demonstrate cost-effectiveness. Fletcher says there are two examples of green products that pass this test: home appliances and comfact fluorescent (CFL) light bulbs. Citing estimates from US EPA’s Energy Star web site, Fletcher says Energy Star clothes washers and refrigerators have a “relatively short payback” of 3.5 years and 3.1 years, respectively, and that CFLs “pay back their extra cost in about four months.”

Today we look at the claims for Energy Star appliances.

EPA’s Energy Star savings calculator for clothes washers (look for Excel file ink in the right margin) assumes that a new Energy Star clothes washer costs $500, and that a conventional model costs $300. Fletcher appears to have used EPA’s default model for her estimate of payback periods, including the assumption that water is heated electrically and not by natural gas. It turns out that the validity of the calculation depends on the validity of the modeling assumptions, which Fletcher does not question and turn out to ber seriously biased to make Energy Star appliances appear more attractive financially than they really are.

We went to Dealtime.com to learn whether EPA’s default prices were representative. More than 1,300 items are listed — many of which are variations on the same model, redundant models with different manufacturer brand badges, not full-size, or no longer on the market. Some products listed as washers are actually pedestals intended to raise the height of front-loading units. We found about 100 different models. For both Energy Star and non-Energy Star washers, the default prices in the EPA savings calculator do not reflect market prices. The median price of conventional washers is about $432 — 44% greater than the $300 default value. Indeed, the lowest price we found for a non-Energy Star washer was $366. For Energy Star washers, lowest market prices ranged from $579 to $3,295. But as price increases, clothes washers include many other features besides energy efficiency, so it’s hard to discern a representative price. For that reason, we can’t tell whether the $200 price gap is reasonable.

EPA’s model has a long list of assumptions — see the “Assumptions” tab on the Excel spreadsheet referenced above. For some reason, EPA hides many of these assumptions (columns E to M) by using white as the font color. To see them, a spreadsheet user must change the color to something visible. Doing this reveals, for example, that Energy Star washers are assumed to use 45% as much hot water (G17/K17), 69% as much electricity (G15/K15), and 70% as much less natural gas (G16/K16). If the actual savings in hot water, electricity or gas consumption is lower, then the difference in operating costs is less and the payback period for the Energy Star washer rises.

One of the most important assumptions in the model is the discount rate — the rate of interest consumers are willing to pay to consume today instead of a year from now. EPA assumes that all consumers are willing to postpone current consumption to earn a real rate of return of 4% per year. Relatively wealthy consumers may have discount rates this low, but relatively poor consumers certainly do not. Indeed, any consumer who finances the purchase of a new clothes washer on credit clearly has a higher discount rate. In December 2007, consumer credit totaled $2.5 trillion. Interest rates on credit cards are between 14% and $32%, with 20% being a representative figure. Deducting 4% for inflation yields a typical real interest rate of 16%.

Personal and household discount rates decline with income or wealth. Moreover, for many households the purchase of a new clothes washer or refrigerator is not entirely discretionary. That is, while it may be true that relatively wealthy households replace these appliances for a wide variety of reasons, relatively poor households replace them only when they fail and repair is not cost-effective. Because these events are sudden and generally unpredictable, that often leads to financing via consumer credit, which costs much more than 4% (8% with inflation).

It turns out, though, that the payback period calculated by EPA depends only on the price gap (and a toggle selecting the fuel source for hot water). EPA merely divides the price gap by the first year’s energy savings. Different payback periods are obtained by using different price gaps.

EPA’s model simply ignores the present value of operating costs over the life-cycle of the appliance. This is curious given that the purpose of the Agency’s Energy Star advocacy is to get consumers to take account of life-cycle operating costs, not just purchase prices. For this reason, EPA’s calculated payback periods are inherently flawed and misleading.

A more useful comparison might be gleaned from elsewhere in the savings calculator, which also generates present value estimates of life-cycle cost savings. But this portion of the savings calculator omits the opportunity cost of paying more today to gain future energy savings. This is another error — and a huge one, at that, especially for consumers with high discount rates. For example, if a consumer finances the added cost of an Energy Star washer on credit, he will pay interest over the life of the loan that must be deducted from realized energy savings. If a consumer does not borrow, but instead reallocates consumption from elsewhere in his budget, then the benefits of consumption that must be foregone also counts as a cost. EPA ignores this cost, or more specifically, treats it as if it does not exist.

In short, EPA’s savings calculator exaggerates the savings from Energy Star appliances because of fundamental structural errors in its “savings” model. The calculated payback period ignores future costs and benefits, and the value of benefits foregone by purchasing a more expensive appliance are ignored. Each of these errors makes Energy Star appliances appear more cost-effective as investments than they really are. Therefore, it is reasonable to infer that the Agency intended to disseminate biased savings estimates.