Callcredit Information Group's 2009 Regional Electoral Roll Opt Out Report found that the figure had increased nearly 10% year on year since 2008, when 22.9% of councils had more than half their residents opt out.

The credit referencing company found that more than 60% of residents have opted out in 13% of councils, up from 7.8% in 2008.

This latest 2009 survey comes hot on the heels of a December report that suggested an overall average opt-out rate of 43.8%.

Chief executive Mike Green said: "These results will provide yet another profound blow to companies trying to drive business by targeting localised marketing campaigns.

"Marketers will have to look to sources of data other than the edited electoral roll to top up their acquisition campaigns," he added.

Ealing had the highest recorded opt-out rate of 82.29 %, followed by Epping Forest with 78% and South Gloucestershire with 76.51 %. The lowest opt-out rate was in North Tayside & Angus with 9.94%.

CDMS marketing manager Richard Higginbotham said: "The increase in the number of people opting out gives a reason to focus on the data you hold about customers to target the right individual at the right time with the right product."

He added that the Merseyside-based marketing communications firm had recently acquired data company Transactis in order to improve its data pools and minimise its reliance on the edited electoral register.

Caroline Roberts, Direct Marketing Association director of public and legal affairs, said: "A recent survey carried out by the Local Government Association and the Association of Electoral Administrators showed that 98% of electoral officers would like to see a change in the law to stop the direct marketing industry from having access to the edited electoral roll.

"The regulations state that by law, all electors must be given a fresh choice every year as to whether or not they'd like to opt out of the edited electoral roll," she added.

"But it is our understanding that some councils positively encourage voters to opt out."

She said this could explain the significant disparity between the highest and lowest opt-out rates.  

"The increase in opt-out rates will also be counter-productive as an edited register with a high opt-out will make data verification more difficult, thereby making it harder for marketers to target accurately and effectively," said Roberts.