Internet voting in Estonia Ülle Madise member of the National Electoral Committee 4th Quality Conference Tampere 27. 29.09.2006
First Internet voting In October 2005 Estonia had the first pan national Internet Voting with binding results ~ 80% of voters had an opportunity to vote via the Internet
e Estonia Population: 1.35 Mio Everyday Internet usage: 60% Households with a computer: 40% 81% home computers in the Internet e ID card roll out: 66% ~1000 AIP
ID card project Started: 1997 Act on personal identification documents: 1999 Digital Signature Act: 2000 First card issued: 2002 990 000 cards issued: Sept, 2006
Voters authentication with ID cards Compulsory document for all residents Contains: Personal data file Certificate for authentication Certificate for digital signature
Reasons behind the case To preserve representative democracy by attempting to stop decreasing turnout trends To decrease electoral costs in a long perspective
Preferred voting means 2004 2005 Surely at voting station 43% 28% Rather at voting station than via Internet Rather via Internet than at Voting station 14% 19% 21% 31% Surely via Internet 12% 12% Does not vote at all 5% 5% Can not say 5% 5% data: Faktum, 2005
Actors Estonian National Electoral Committee and Riigikogu Ministry of Internal Affairs City of Tallinn Certification Centre Chancellery of the Riigikogu Cybernetica Ltd KPMG Baltics Ltd
Secrecy scheme
Secrecy of the vote Voters privacy cannot be guaranteed by e voting and voting by mail, therefore other methods to guarantee free expression of the voters real will are required the right to change the e vote times for changing not limited during advance voting period paper ballot priority one person, one vote
Tools for voting Voter needs: an Estonian ID card with valid certificates and PIN codes Computer used for voting must have: a smart card reader a driver for ID card (free to download from page www.id.ee/installer)
I Voting website www.valimised.ee www.valimised.ee
II Identification Put your card into card reader Insert PIN 1 ****
III Ballot completion Choose a candidate
IV Authentication Confirm your choice Insert PIN 2 *****
V Confirmation
Process leading to success Trust in new voting method A commonly acceptable approach Political agreement to introduce Internet voting and signing of Good Practice of e voting Reasonable costs: project development and realization ~ 300 000 ; Infrastructure for electronic identification system is not included
Results: Impact on turnout The first time no impact on general turnout voters need time to get used to new voting method E votes made up 1,85 % of the votes but 9% of all advance votes Turnout in municipal elections Turnout in municiapl elections 1993 2005 60 50 52,6 52,5 49,8 52,5 47,4 % of voters 40 30 20 10 0 1993 1996 1999 2002 2005
Results: No digital divide * E voting is an additional way of voting it widens accessibility Why to prefer e voting to voting in polling station? Trust in e voting Relevant computer knowledge Language What has significant impact on the decision whether to vote over the Internet or at a polling place? Age, gender, income, education do not have a significant impact. * Source: A.H. Trechsel, F. Breuer, E Voting in the 2005 local elections in Estonia,
Lessons learned Internet voting is a part of the concept of e government. It widens accessibility and encourages participation. Internet voting attracts people to other electronic public services. Suprisingly in 2005 elections 60% of e voters gave their first digital signature. Long term planning on national level to introduce widespread means for electronic authentication creates basis for further development. Existence of a countrywide electronic authentication system is vital. It reduces significantly electoral costs and raises public trust.
Thank you for your attention! Ülle Madise Member of the National Electoral Committee Further information: www.vvk.ee
e Estonia Population: 1.35 Mio Everyday Internet usage (6 74 years old): 60% Households with a computer: 40% 81% home computers connected to the Internet e ID card roll out: 66%