CFL Power Factor may vary from below 0.5 to over 0.9 depending on type of integrated ballasts (traditional magnetic or electronic high-frequency), and also depending on ballast manufacturer and quality of ballast.
Lighting Research Cente: Power Quality, which includes tables & graphic illustrations of how CFLs, computers and other non-heating appliances distort power supply harmonics, explaines the difference between incandescent (incl. halogen) and fluorescent (including CFL) lamp effect on the power system:
“Incandescent lamps, toasters and other heating devices usually have a power factor of unity = 1. (…) Resistive loads such as incandescent lamps actually reduce voltage harmonics.”
“Poor power quality can damage the distribution system and devices operating on the system. (…) High frequency electronic ballasts operate at frequencies ranging from 20 to 60 kilohertz (kHz). The harmonics produced by these ballsts are correspondingly high frequencies and can interfere with some communication equipment including radios, intercoms, and cordless phones. Devices that use power-line carrier signals, such as synchronised clocks and control modules for building energy management systems may also experience problems if harmonics exist at frequencies close to the carrier signal.” 
However, according to a study at the Vienna University of Technology (cited by the Swedish Energy Agency ), distortions may vary with the actual situation and depend on the CFL in relation to other appliances and other CLFs, e.g. distortions may decrease if CFLs are of different brands, but increase if they are of the same brand. 
But this is not the main issue. As demonstrated by SaveTheBulb CFLs Real Power Used, CFLs with poor power factor may use up to twice as much power as claimed! 
Looking for a second opinion on this astonishing revelation, I made a few calls.
– According to a representative at Swedish Osram, an 11W CFL may in fact use around 18W! The integrated ballasts also use about 2W.
– The Swedish Energy Agency says this sounds about right. That a 15W CFL may have a PF around 0.5 and a VA of 30.
– Vattenfall, one of Europe’s biggest energy suppliers, said they are well aware of the difference between active and reactive effect and that they bill larger customers for their VA, whereas smaller customers are only billed for the active effect (= watts used). But to compensate for reactive effect from home electronics and CFLs with poor power factor, a generalised extra fee is included in the standard price per kWh! When I asked what will happen if more people start using significantly more CFLs, the reply was that utilities will have to compensate by making this extra hidden fee higher! This, of course, is nothing customers are informed about (unless one asks, and knows what to ask).
To get a Canadian EnergyStar label “an average of 10 samples tested must be greater than 0.5” – which is not overly reassuring as it needs to be at least over 0.85 in order to not create distortions and tax the energy supply with more than the watts it is marked with.
Technical Editor Margery Conner of EDN also found confirmation of CFLs poor power factor (0.57 in her own CFL, and 0.45-0.50 in a Luminaire Testing Laboratory test) and wanted to know what EnergyStar plans to do about it:
“I emailed Peter Banwell of the EnergyStar program and asked if EnergyStar was considering making minimum PF a requirement for Energy Star compliance. He replied, ‘We looked at this in detail several years ago and decided against it, though there are a couple of utilities that still support the idea. We may take this up in the future, as the market share grows, but right now it is still in the noise in terms of impacts.'” 
1. Lighting Research Cente: Power Quality
2. Fyra frågor om lågenergilampor, Swedish Energy Agency, 1999.
3. CFLs Real Power Used
4. “Utilities suffer from CFL’s poor power factor”
See also: “The Hidden Costs of CFLs”
Sylvania – Power Factor