Have you observed how a politician often answers a reporter’s question that he doesn’t want to answer?  Instead of truthfully replying that he doesn’t want to answer the question, he may pretend to answer the question while in fact answering another question that puts him in a more favorable light.

Sunshine Week LogoUnfortunately, Maryland’s State Board of Elections has adopted a similar strategy to avoid fulfilling Public Information Act requests for embarrassing information.   Given the lax enforcement of Maryland’s Public Information Act, this may be a shrewd non-compliance strategy.  But it’s bad for Marylanders, who should have access to the information necessary to hold their election officials accountable for election abuses.

Motive for Secrecy

Consider the handling of election information regarding Maryland’s obscure but highly politically sensitive constitutional convention referendum.  Maryland’s Constitution requires that every twenty years a referendum be placed on the ballot asking Marylanders whether they want to convene a state constitutional convention.  The last such referendum was placed on the ballot on November 2, 2010.  It received a 54.4% ordinary majority (that is, a majority of those voting on it), but the legislature claimed it needed an extraordinary majority, a majority of those voting on any proposition on the ballot.

The majority specified in the Constitution reads “a majority of voters at such election or elections.”  This majority clause is ambiguous, as is the case with many majority clauses in both Maryland law and the law of other states.  But in the overwhelming majority of such cases, the ambiguous clauses are interpreted as ordinary majorities.  This is the case, for example, with the majority in Maryland’s Constitution required to pass a gambling referendum (a “majority of the qualified voters in the state”) and to pass Baltimore bond debt (a “majority of the votes cast at that time and place”).  In popular political speech and in the press, one is lucky if any majority denominator is specified, as this is considered a tedious and self-evident detail.

The majority required to convene a constitutional convention in Maryland was calculated as an ordinary majority from 1851 to 1930, when it received an ordinary majority and was changed to an extraordinary majority.  Changing to the more difficult to attain extraordinary majority allowed the leaders of Maryland’s General Assembly to defeat the referendum, which they opposed, fearing the resulting constitutional convention would lead to reapportionment of legislative districts based on population rather than geography (still the basis of apportionment in the U.S. Senate).  Such reapportionment would have dramatically shifted the balance of power in the General Assembly from rural to urban areas, including the loss of dozens of seats by incumbent legislators and a shift of power to Baltimore and blacks. In 1964, the U.S. Supreme Court forced Maryland to reapportion its state legislative districts based on population.

After the 1930 election, tallying the General Assembly’s newly preferred extraordinary majority involved many practical problems.  Neither the number signing the poll book nor casting ballots was apparently legally acceptable, so the number of ballots counted, a subset of the above, had to be used.   But the election officers hadn’t counted the number of ballots; they had only counted the number voting for particular propositions, which given a secret ballot cannot be translated into a vote count.  Moreover, until the early 20th century, when votes for the constitutional convention referendum and other propositions were consolidated onto the same ballot, calculating the preferred extraordinary majority was a logical impossibility unless ballot secrecy was violated.

So in 1930 a proxy for the number of votes counted was used: the number of votes counted in the race for governor.  This proxy had merit.  It excluded blank ballots, spoiled ballots, and ballots with votes for local and federal but not state propositions.  In an era without calculators, it was also much simpler to calculate than, say, the more accurate tally on a precinct by precinct basis of the highest vote on any proposition.  On the other hand, its novelty was a bit embarrassing.  The first three times Marylanders had voted on the constitutional convention referendum a governor’s race hadn’t been held at the same election.

Surprisingly, this so-called “gubernatorial proxy” for tallying a majority was never codified in Maryland law, which made the 2010 shenanigans feasible.  In 2010, an election administrator at Maryland’s State Board of Elections changed the proxy for votes counted to a mixed proxy based on those hitting the submit button on the new statewide touchscreen voting machines plus the number of counted absentee and provisional ballots using the more old-fashioned optical scan voting machines.  This increased the size of the majority needed to pass the referendum by 7,404 votes over the gubernatorial proxy.

Regardless of the reasonableness of this change, it was made in a politically embarrassing way because it violated core democratic values.  The change was made without any change in law, including statute, regulation, public meeting minutes, public notice, or even merely written guidance.  That is, it was made without due process and by administrative fiat.  Moreover, the change was made  after  the election, thus violating the core democratic principle known as the “veil of ignorance.”  Applied here, the principle entails that election rules shouldn’t be changed after votes have already been cast and outcomes under various vote aggregation rules known.

Given that neither the gubernatorial nor new, mixed proxy, has ever been codified in law, nothing prevents yet another change in the proxy in a future election.

Public Information Act Response

This was a good reason to  document the most recent case  of such electoral shenanigans, so I filed a series of Public Information Act requests with the Maryland State Board of Elections. After months of delay to my most recent request, the Board provided no written correspondence of any discussion or rulemaking concerning the new method of calculating the majority denominator or the new and dubious method of publicly labeling non-votes as “times blank voted” versus, say, the more conventional way of describing non-votes as “drop-off.” Assuming no such documents existed, this part of the response didn’t violate Maryland’s Public Information Act, other than the 30 day rule for replying, which is a common violation in Maryland and one among many that Maryland judges don’t take seriously.

However, I was also interested in a copy of the input parameters entered into Maryland’s election database software to tally the results from raw election data, as well as the resulting output.  My hunch was that the implementation of the new majority denominator involved hiding various controversial tallying practices, such as including Federal election only ballots and ballots with missing pages.

Federal law specified that oversees military and civilian voters must receive ballots with federal propositions on them by a certain deadline. To comply, Maryland sent out more than 3,000 ballots with federal only propositions.  The returned federal only ballots would not have included the state constitutional convention proposition.

Printed absentee and provisional ballots included separate pages, one or more of which might not be submitted for counting without invalidating the ballot.  A missing page could include the proposition on the state constitutional convention.

For output, the Maryland State Board of Elections sent me the data it published on its website but not the actual output of its election database.  For input, it sent me an undated document that included incorrect parameters.  Based on my analysis of the documents, it would have been more accurate, albeit more embarrassing for the Board, to simply acknowledge that it had no documents responsive to my request.

I also requested a copy of the training manual used by the election administrator.  After being told for several months that “The manual you are requesting is proprietary material of the [election database] vendor and we cannot provide you with a copy,” I was eventually given the manual, which I found provided no examples of the type of extraordinary majority used in Maryland.

Recommendations

The public should be able to verify in a timely, newsworthy way that its public officials don’t play politics when tallying election results.  I therefore suggest the following four reforms:

First, Maryland’s General Assembly should codify into law the precise formula for calculating the majority denominator on the constitutionally mandated state constitutional convention referendum.  Prior to an election, Marylanders have a right to know how election results will be tallied.

Second, the General Assembly should eliminate reliance on the ineffective Public Information Act for the public to monitor the activities of the State Board of Elections.  Instead, the Board should be required to proactively disclose on its website all key electoral data, which should include the input and output of Maryland’s election database software, and the training manual the election administrators rely upon to use the database.

Third, the General Assembly should mandate that the State Board of Elections webcast its public meetings and then archive them online.  It is too difficult for Marylanders living remotely and working during the day to attend such meetings.  Moreover, the resulting meeting minutes are both incomplete and delayed beyond the point of newsworthiness.  If Maryland can afford tens of thousands of video cameras to monitor its citizens, surely it can afford a single camera to monitor its State Board of Elections.

Fourth, President Obama’s newly created Presidential Commission on Election Administration should add election transparency problems, such as those in Maryland, to its agenda.

Source: Snider, J.H., Shame On Maryland’s State Board Of Elections, Eye on Annapolis, March 15, 2013.