I've told my readers more than once that a system like Colorado's, in which elections are largely run by partisan elected officials (the Secretary of State at the state level, and a partisan elected clerk and recorder at the county level, outside of Denver . . . I don't know what they do in Broomfield), is a disaster in the making.
Don't believe me. Hear it from an election law specialist. An alternative proposal is set forth here. He proposes:
1. Universal voter registration conducted by the federal government which issues ID cards with biometric ID (such as a fingerprint). The ID would not be required to vote, because the biometric information would be stored in a central database. Change of address cards at the U.S. Postal Service and death certificates would be cross-referenced.
2. Nonpartisan election administration. A bipartisan supermajority (say 75%) would be required to appoint a state election administrator, who in turn would appoint local election administrators. Australia and Canada are proposed as models.
3. Increase the ease with which pre-election court challenges can be filed (under doctrines such as standing and case and controversy rules), but bar challenges that could have been foreseen prior to the election which were strategically not brought.
Universal Voter Registration
I personally think it wouldn't be hard to do (1) at the state level, as opposed to the federal level, respecting the general principle that election administration is conducted by states rather than by the federal government. This would also alleviate fears of a national ID card. Interstate cooperation and U.S.P.S. cooperation could still be mandated. Indeed, government sponsored voter registration can be done at the local level even if not required and rather than producing a race to the bottom, tends to produce interjurisdiction competition to improve voter registration.
I agree with the model of (2). I've spoken with judges in Michigan, who are elected locally, about whether, if there was a shift to an appointment system, they would favor local appointments or a state appointment. The consensus was that state appointments (which we have in Colorado) were better, because judges often have to rule on cases involving local governments and would still "owe" something to the parties in such cases, something much less likely to be the case with a state appointment. The same reasoning applies to elections officials.
Keep in mind that the total statewide number of elections officials is actually rather small, especially in the "off season" and that election administration doesn't very naturally fit county boundaries these days now that Baker v. Carr imposed a one man, one vote requirement on states that broke the connection between state legislative districts and county boundaries. The clerk and recorder's office is typically a small percentage of total county employment (on the order of 4% or less), and the elections division is typically a small fraction of the total office's employment, which is dominanted by the DMV and the real estate recording division.
Denver's Election Commission employs 21 permanent and temporary employees (including the executive director) who report to three election commissioners. The Commission is constrained by all of the City's elected officials and by the D.A. and the Secretary of State. Denver makes up about 1/8th of the population of the state of Colorado. Thus, it is fair to estimate that a single statewide office for elections would require about 168 employees, plus some additional employees transferred from the elections division of the Secretary of State.
Similarly, the Secretary of State's office is a small percentage of total state employment (the entire Department including the business division, notary services and bingo supervision has a $28 million budget out of a $17,025 million state budget, about 0.2%, its share coming entirely from user's fees; there are about 111 employees in the office (FTE) about 0.5% of the total state employment in Colorado), and the elections division is a small fraction of the office's total employment, about 18 people. National elections division FTE figures are available here.
It is fair to guess that fewer than 200 employees statewide could carry out out electoral functions currently carried about by local clerk and recorder's officers, election commissions and the Colorado Secretary of State, for a comparable cost to current arrangements, which is a tiny share of state and local budgets. (For comparison purposes, the State Treasurer's office in Colorado has 23 FTE employees, and the Attorney General's office has 132, while the State Auditor, who is appointed by the legislature, has 52; about a quarter of state executive branch employees are in the Department of Corrections and about a quarter are in the Department of Human Services).
Interestingly, Colorado's Constitution does not require that a county clerk and recorder administer local elections, although it does require that counties have an elected clerk and recorder. A state statute turning over all election administration duties to state government would probably be constitutional. Some of the Secretary of State's election administration duties, however, are constitutional in nature. Some of the Secretary of State's constitutional electoral duties, like transmitting election results to the General Assembly, however, could be reduced to ceremonial duties by statute, and the administration of the campaign finance system, which is the Secretary of State's responsibility, could be severed from the other obligations of the position, like enacting regulations under the statutory election laws. Colorado already has a constitution that provides for state legislative redistricting to be conducted by a bipartisan commission, and if Amendment 41 passes, as it is likely to, there will also be an independent ethics board for the state. Thus, there is precedent for electoral duties to exist outside the Secretary of State's office.
Early Judicial Review
I can understand court reluctance to get involved in elections before they take place, a problem than (3) seeks to address. Few elections are close and many problems that can be foreseen don't turn out to be very important in real life, so most cases end up being dismissed as moot. But, the negative effects of a pre-election lawsuit are much milder than those of a post-election lawsuit, so there is much to be said for this proposal as well.