In this case, the comparison to regular cars was flawed, the hybrid mileage highly inflated, which would lead people to buy hybrids instead of regular cars based on false information. It also gave people unreasonable expectations of fuel savings that they hoped would more than pay for the extra price of the hybrid. If the sticker says the hybrid gets 20 mpg more than the regular car and it really only gets 5 mpg more (regardless of what the driver actually gets in either, it's relative), you're screwing a lot of people.
Breathless accusations of some kind of eeeeevil conspiracy are for drama queens who have nothing important to do.
There's no conspiracy, just growing pains in dealing with hybrids.
You're right. My comments were aimed at the general population of people who generally moan about EPA estimates being inaccurate.
In this case, their testing methodology simply doesn't work well. Also, I'm mystified by the EPA's decision to base fuel economy estimates on emissions data, rather than a simple measure of fuel flow rates. Why not just measure what you're measuring directly?