CFL Energy Misinformation

Do CFLs really save energy? It depends on where you live. You certainly don't save it if you live in a northern state, where the heat from incandescents that is claimed to be “wasted” is worth more for heating your home than anything you save on CFL lighting.

I live in Minneapolis, where the local electric utility, Xcel Energy, has advertised you save $50 “every time you replace an inefficient incandescent bulb with a CFL.” The company also claimed you can “Get CFLs for a s little as $1. For a store near you, visit [our website].” I checked the stores listed on that website, then visited the nearest ones whose websites listed CFLs for $1 and couldn't find any. I visited Home Depot, Rainbow Foods, Dollar General, Family Dollar, Lund's and Ace Hardware. (Wal-Mart did not list CFL prices on its website.) A few of the stores' websites listed a CFL for $1 or 99 cents, but on visiting the stores I found the the stores no longer sold CFL bulbs at those prices. For example, Family Dollar's website showed 3 CFL models for $1 apiece and one for $1.50, another for $2. My visit to the store showed they didn't sell any for a dollar. In fact, they no longer sold the ones for $1.50 or $2.00 either. They were now selling 60w (equivalents) for $4 per bulb or a two-pack for $7. They also were selling a 100w CFL for $6.00 and, for the same price, a 60w “Armorlite” (which I explained previously is a CFL inside a shell looking like an incandescent, which offers some protection against the CFL causing a fire or the glass “twisty” exploding. See http://scienceandpublicpolicy.org/images/stories/papers/originals/cfl_fires.pdf)

I went to Home Depot to check on whether the Philips 60w CFL was available for $1, as shown on their website. It wasn't. Nor were the two EcoSmart bulbs its website showed at $1.06 each. Instead, the store now had two EcoSmart 60w CFL models at $2.24 (4-pack for $8.97) and $1.74 (4-pack for $6.97).

The Lund website showed one for 99 cents, but that turned out to be out-of-date information and the store no longer carried that bulb. Some other models on their website were also no longer available, and all the bulbs they did carry were no bargains.

I next visited Ace Hardware. Their website listed a 60 watt-equivalent CFL for $1.40. That, of course, is well above the “dollar or less” Xcel had been advertising, but since so many other stores were no longer carrying the cheapest bulbs listed on their websites, I thought I'd check on this one priced at $1.40. It turned out Ace no longer sold it. But then the manager (named Craig) called my attention to another bulb display, a special: GE 5-pack of 60w reduced from $12.97 to $4.99. The sign said: “Sale ends 3/31/12.” So here were bulbs for 99 cents (compared to the pre-sale price of $2.59 each). Craig said, “You better buy them if you want them, because when the sale ends [in 3 days] they are going back up to $12.97. You won't see them at this price again. We can't afford to sell them at that low price. We do so only because Xcel Energy pays us to do it. After we get the money from Xcel, all these bulbs are going way up in price.”

That was certainly an alarm bell! I had previously read in the Wall Street Journal (referenced in the Science and Public Policy Institute paper linked above) that Pacific Gas and Electric, the largest utility in California was subsidizing CFLs. The WSJ said PG&E was selling bulbs for $1.30 compared to $4 for unsubsidized bulbs. PG&E was not only using ratepayer funds to subsidize the apparent low cost of the bulbs but also collected state subsidies of $104 million in rewards and incentives for promoting the bulbs. So the taxpayers as well as the ratepayers were paying for these CFLs in ways that are not included in all the claims about how much money customers were saving on them. I don't know if Xcel Energy was receiving government subsidies as PG&E did, but it certainly is using ratepayer money to give a false impression of how much the CFLs are costing its customers. It was about 8 o'clock in the evening when I was driving home from Ace hardware on March 28. I turned on the news on my car radio and heard that the Minnesota Public Utilities Commission had approved a 2.7 percent increase in the rate Xcel customers will be charged.

After that, I lost interest in making the detailed energy comparisons, using degree days of heating and cooling throughout the year, which I expected to prepare and send to Xcel. It seemed pointless to go through that exercise when the true cost of the bulbs is not known because it is disguised under Xcel subsidies to the retailers and paid for by higher costs to the Xcel ratepayers (and perhaps taxpayers.) Now, however, it occurs to me that, even without the detailed cost study I originally envisioned, a convincing case can be made against Xcel's claims of CFL cost savings, just on the basis of lost incandescent heating in the winter versus the cost of air conditioning in the summer.

The official (legal) heating season in Minneapolis is September 15 to May 15. Those are the dates between which landlords are legally required to provide heat to renters. Of course, there may be some days in September after the 15^th, or in May before the 15^th, when no heat may be necessary; but it is far more likely that heating will be required in the latter half of September than that air conditioning will be required in the first half of that month. The reverse will be true in May. I'll come back to the months of September and May a little later, but first I want to discuss the winter and summer months, which are more clear cut.

Certainly after October 1 it may assumed there is really no demand for air conditioning and nearly all days will require at least some heat. A similar situation may be assumed at the other end of the heating season. I don't think there is any dispute about that.

It costs the same to raise the temperature a given amount as to cool it, given the same temperature differential to the outdoors. That is, if the outdoor temp is 50 degrees, it takes the same amount of energy to raise the indoor temp to 70 degrees as it does to cool the indoor temp to 70 degrees if the outdoor temp is 90 degrees.

On a winter day when the outdoor temp is 20 degrees, your furnace must work more to heat your house to 70 degrees because the heat lost to the outside is greater because of the 50 degree indoor/outdoor temp differential. But you will never see a 50-degree temperature differential in the other direction; you will never see 120 degrees in Minnesota in the summer. Similarly, we have winter days below zero, which your heating system will raise the indoor temp to 70 degrees, but you will never see 140 degree temps in Minnesota that your air conditioning system has to cool to 70 degrees.

Last year there were 14 days above 90 degrees, which also happens to be the official 43-year average. Typically these are all in the months of June, July, and August (though there are occasional exceptions, as we have just seen by two 90+ temps in late May.) Basically, however, we should compare the heat lost from CFLs for seven winter months (October through April, inclusive) to the air conditioning cost for three months (June through August). The comparison is not even close. There are seven months of heating and only three months of air conditioning, and the heating requirement is vastly greater than the cooling requirement: there will be, on average, only 14 days above 90 degrees, but the number of days below 50 degrees will be many-fold larger than 14. In fact, there will be many more than 14 days below 40 degree and even 30 degrees—with no summer counterpart the same distance above 70 degrees. And there will be many days below 20 degrees and some even below zero.

The case for CFLs is not improved by including the months of September and May in the calculation. In fact, they make the case worse for CFLs. Frost in not unknown in May—even in the latter part of the month. And, as is the case with the winter months, there is no comparable high (e.g. 108 degrees) to offset a 32-degree low. Lows in the 40s are common in May; most of the first two weeks in May this year had lows in the 40s. And Sunday May 20—even after the 92-degree high—we saw a low of 47 degree. Moreover, on May 21 the mean temperature was 58 degrees, which is the normal mean for that date—three weeks into the month! Also, the normal high for that date is 71 and the normal low 51 degrees. Thus the normal high for 3 weeks into the month is only one degree above our reference of 70 degrees while the normal low is 19 degrees below. The normal high for May 31 is 74 degrees, and the normal low if 54.

For May 2012 in Minneapolis, the number of heating degree days during the month was 148 while the number of cooling degrees days was 65. So the amount of heating energy required in this month was more than twice the amount for cooling—and this was in a year when May was warmer than normal. I have not gathered the numbers for September, but decades of living in Minneapolis leads me to believe those heating/cooling numbers would not be appreciably different from May. In any case, September and May would not be months in which the air conditioning demand would be greater than heating demand. The heating value lost from CFLs during the seven winter months is lopsidedly of greater economic value than the cost of air conditioning during the three summer months. Factoring in the months of May and September makes the comparison even more unfavorable for CFLs. What is occurring in Minnesota will also be true in other northern-tier states and probably to a lesser extent in states bordering them to the south.

The claims of energy efficiency, not only by Xcel but by others, are based on the commonly advertised CFL bulb life of 10,000 hours. But PG&E, the largest electric utility in California, found bulb life is overstated by a whopping 49 percent! Since PG&E serves millions of customers, its experience with bulb life represents a very large sample and is, therefore, probably accurate.

Using the realistic shorter bulb life, plus the fact CFLs could not be purchased locally for “a dollar or less,” as advertised, plus the fact the heat lost from replacing incandescents is worth more than the energy saved from CFL lighting, and you can see the claim you can save $50 “every time you replace an inefficient incandescent bulb with a CFL” is utter nonsense. This does not even include the fact that the incandescents will be used less in the summer—when the days are longer—than in the winter when their heating is more important. Nor does it include the cost of recycling or of transporting burned out bulbs to collection points—not to mention the 2,301,000 CFL bulbs that have been recalled by the U.S. Consumer Product Safety Commission because of danger of causing fires, explosions or lacerations from flying glass. (Note: additional examples of fires and explosions and broken-glass hazards beyond those mentioned in the SPPI report can be found at http://amlibpub.blogspot.com/2012/04/cfls-more-fires-product-recalls.html).

A further cost that is not included in economic comparisons with incandescents is the cost of special containers for transporting even unbroken bulbs—containers which, according to Dr. Brosseau, are required by state and federal workplace safety requirements to keep mercury vapor emissions below permissible exposure levels. (See Science and Public Policy Institute paper cited above.) Failing to comply with this standard leaves businesses vulnerable to legal hazard, says Dr. Brosseau, yet CFL advocates never mention this.

The really important issue about CFLs is human safety. Yet the government, CFL producers and energy companies—including Xcel—continue to promote these dangerous products—and even subsidize them!—while ignoring or trying to dismiss the dangers, including mercury. And they are promoting them with advertising statements of energy efficiency and cost savings that are not even true.