Policy makers face growing uncertainty about the effectiveness of energy efficiency programs. This week, for instance, two reports raised concerns about questionable information on cost savings provided in point of purchase appliance labels and potentially low assumptions of how much power computers use. A report by the Haas Energy Institute at the University of California at Berkeley pointed out that federally-required yellow energy usage cost labels for air conditioners and other major appliances are based on the national average price for electricity. This results in savings being underestimated where power is more expensive and over-estimated where it is cheaper. Also, the labels for air conditioners are based on the national average for hours of operation, which vary across states\u2014even within them\u2014based on climate. Power prices can vary by as much as a 2 to 1 ratio and air conditioning usage can vary by as much as a 9 to 1 ratio, the report said. It suggests, at a minimum, providing state-specific information to consumers. Meanwhile, two reports released Oct. 28 by the California Energy Commission on power consumption by computers show they are not going to sleep or automatically turning off as much as thought because users change automatic settings and resist manually shutting down machines because they take too long to boot up. One of the studies is based on a survey involving 2,000 students and the other is based on measuring the actual power usage of 125 computers. The study of 125 computers found that in the workplace desktop computers were on 76 percent of the day, even though they were being used just 16 percent of the day.