Last Friday, Washington’s House of Representatives took up the Northwest Progressive Institute’s legislation to clean up our ballots and remove a barrier to voting in the Evergreen State by abolishing what former initiative promoter Tim Eyman calls “advisory votes,” but which are really push polls or anti-tax messages.
The passage of Senate Bill 5082 is an important victory for voting justice that’s supported by thousands of Washingtonians, editorial boards from every corner of the state, and dozens of organizations that work with voters.
In the immediate aftermath of SB 5082’s approval in the House, Tim Eyman had nothing to say — at least not publicly. Eyman did not react to the vote on his Facebook page or his mailing list or his website until Monday morning when he shared a meme (specifically, the Darth Sidious Unlimited Power meme) and accused Democrats of wanting “unlimited power,” in line with the meme.
The passage of SB 5082 doesn’t, however, give Democrats new powers or change Washington’s plan of government in any way. What it does do is replace the taxpayer-funded right wing messaging that had been appearing on Washingtonians’ ballots for years with truthful, useful fiscal information that is always available online and referenced in the printed voter’s pamphlet.
Yesterday, Eyman responded at greater length.
In a revealing rant, he not only criticized Democratic legislators for passing Senate Bill 5082, but offered what he called a “belated defense” of his push polls, in a tacit acknowledgment by Eyman that people increasingly see them for what they really are: pieces of propaganda that simply do not belong on anyone’s ballot.
Wrote Eyman:
Finally, let me offer a belated defense of the wording of the short description for a Tax Advisory Vote — it is 100% accurate:
“The legislature imposed, without a vote of the people, …”
Absolutely correct. That’s exactly what happened that caused a Tax Advisory Vote.
” … [identification of tax and description of increase] …”
This portion is written by the Attorney General.
” … costing (most up-to-date ten-year cost projection)] …”
The cost of the tax increase is determined by the Governor’s budget office.
“… for government spending.”
Once the money starts flowing in from a tax increase, the Legislature is within its power (and 99% of the time does) divert the money to whatever pet project or programs they want.
Famously/infamously, the voters passed a tobacco tax increase to fund certain government programs. But within a couple of months, the Legislature declared “an emergency” and they and Democrat Governor Gary Locke diverted that money toward other things.
So the wording in a Tax Advisory Vote (” … for government spending.”) is the only thing the voter knows for sure will happen to the money from a tax increase.
This isn’t “loaded” language — its wording reflects exactly how tax increases and Olympia works.
To our team’s knowledge, this is the first time that Tim Eyman has devoted this many words to a defense of the wording of his push polls, at least in one of his email missives. It figures that he would claim the wording is “100% accurate,” but it’s interesting that he felt the need to elaborate and comment on each part of the formula that he created many years ago for these push polls at this juncture.
We published our first breakdown of the wording of Eyman’s push polls a decade ago. Today, I’m once again going to offer a detailed critique of Eyman’s push polls as well as an explanation of how legitimate public opinion research is conducted. I’ll show you why Eyman’s push polls are defective by design and thus incapable of serving as mechanisms for feedback in which people can express an opinion.
Let’s start with a discussion of how public opinion research differs from elections, and define a few terms that will appear repeatedly later in my analysis.
Elections & public opinion research: Two complementary, different aspects of a functioning democracy
An election is an event in which the people make decisions about who should represent them or what laws their society should have (or both). The ballot is the mechanism by which the qualified electors of the society record their votes. It is the way in which a voter participates in the election, in other words.
Our team at NPI considers the ballot a sacred space. Our research strongly suggests Washingtonians of all different political orientations share this view.
Research by political scientists has found that the design, contents, and complexity of a ballot are of great importance. Unnecessarily complex ballots can lead to unwanted consequences, as Patrick Schoettmer, a Seattle University professor, has explained to the Legislature in testimony for our legislation:
“We find that when ballots are unnecessarily complex, that discourages participation. Bowler, Donovan, & Happ find that ballot complexity increases informational costs, which increases voter fatigue. Fatigue has lots of consequences, including roll-off (which means voters stop voting on issues) inattention (voters engage randomly instead of deliberately) and dropout (they decide not to vote at all).”
— Election administration researcher Patrick L. Schoettmer, PhD, of Seattle University, testifying in support of Senate Bill 5182 on January 20th, 2021
Putting anything on the ballot that is not a candidate election or a real ballot measure is a violation of a sacred space. It can lead to confusion and cynicism and hinder voting. That is why legitimate public opinion research is not done through the ballot, but separately, through survey instruments. The ballot just isn’t an appropriate place for polling, advertising, or campaign-style messaging.
The goal of legitimate public opinion research is to find out what people think about a topic. The topic could be an issue, a current event, or somebody’s idea, or… an upcoming election! Credible electoral polling is especially popular and interesting to many people because it provides data to help people discern which potential outcomes could be more probable than others.
Legitimate public opinion research also helps elected officials take the pulse of the people in between elections. Elected officials can be more responsive to their constituents’ needs and wishes if they have access to credible polling.
The key to legitimate public opinion research: Neutral questions asked of representative samples
It’s possible to attempt to survey an entire universe of people (like the adult population or qualified electors of a given jurisdiction), but it’s expensive, which is why pretty much every firm that does public opinion research relies on samples.
For a poll to return sound data, the respondents in the sample must collectively be a representative group. They must look like the universe they’re from, in other words. And the questions the respondents are asked must be neutrally worded. Asking loaded questions or failing to recruit a representative sample will skew the results of the survey instrument, yielding worthless data.
Anyone or any organization that’s genuinely interested in what people think can carry out legitimate public opinion research. It isn’t necessary for the commissioning entity to be “nonpartisan” — subjective organizations such as NPI are perfectly capable of undertaking objective research, and do so regularly.
NPI is a leader in public opinion research in the Pacific Northwest, with a strong track record of excellence in polling that dates back a decade.
Neutrality is nonnegotiable
Neutrality in question wording is not a luxury, it’s a necessity.
You can’t find out what people think if you’ve told them what to think first. This is a mantra that the NPI staff and I have been repeating to Republican lawmakers, right wing media commentators, and Tim Eyman fans a lot recently.
It is telling that in their attempts to amend our legislation to keep Eyman’s push polls, Republican legislators did not propose replacement wording that would remotely qualify as neutral. The furthest that Republican legislators wanted to go was engaging in a little bit of tinkering around the edges.
However, when something is defective by design, tinkering is not a solution.
If neutrality is lacking, a poll can easily become a push poll
Push polls got their name because they attempt to influence public opinion rather than measuring it. They look like polls, yet they are not legitimate public opinion research surveys. They improperly push people towards a certain point of view, frequently with misinformation or disinformation. William Safire characterizes them in his indispensable Political Dictionary as a form of dirty trick:
A push poll is a nefarious telemarketing technique designed to spread negative information about an opposition candidate. During the South Carolina primary of 2000, a caller from the George W. Bush campaign asked 300 potential voters:
John McCain calls the campaign finance system corrupt, but as chairman of the Senate Commerce Committee, he raises money and travels on the private jets of corporations with legislative proposals before his committee. In view of this, are you much more likely to vote for him… or much more likely to vote against him?
A push poll is not a legitimate public opinion survey because its purpose is not to obtain an opinion but to influence it, which qualifies the device as a dirty trick.
— From the Dirty Tricks entry of Safire’s Political Dictionary, 2008 edition (Oxford University Press | New York, New York)
Legitimate public opinion research follows a core set of principles for achieving neutrality, principles that are discussed in David Harris’ book The Complete Guide To Writing Questionnaires (which is worth a read if you’re interested in polling):
- First Guideline: Do not introduce ideas or opinions that will influence responses.
- Second Guideline: Make sure that no answer choice is more loaded than any other.
- Third Guideline: Make clear that either a positive or a negative answer is equally acceptable.
— David F. Harris: The Complete Guide to Writing Questionnaires, Chapter 9 (I&M Press | Durham, North Carolina)
What Tim Eyman calls “tax advisory votes” are in reality push polls, since the questions are purposely worded to suggest their own answers. Even the answer dichotomy is arranged so as to invite people to respond in the way Eyman wants.
Let’s take a look at each element of Eyman’s push poll formula.
Breaking apart Eyman’s push polls
Above, I’ve included a slide from the deck we first made back in 2019 to make the case for our bill. This deck made a big impression on the State Government Committee in the Washington State Senate: all three of the committee’s Republican members voted for the bill’s 2019 version after hearing the presentation, which just goes to show that there are some Republicans who get it.
The first element is the opening phrase:
“The legislature imposed, without a vote of the people…”
Eyman characterizes this as “absolutely correct.” But it’s not enough for a question to contain wording that is correct. The wording must also be neutral.
If I say, “Tim Eyman, the admitted chair thief…,” I am making an “absolutely correct” statement: Eyman pleaded guilty to a misdemeanor after he stole an office chair from the Lacey Office Depot in 2019. It’s a fact.
But it’s also prejudicial.
Prejudicial statements are fine in commentary, like this blog post, but not in official representations of legislation shown to voters on a ballot, and not in survey questions unless balanced out by a countervailing viewpoint for neutrality.
The statement “The legislature imposed, without a vote of the people” invokes an anti-representative democracy frame, which is totally inappropriate.
It’s actually the Legislature’s responsibility to make decisions on the people’s behalf. It’s what we elect our representatives and senators to do.
The people have reserved to themselves the right of initiative and referendum (the latter being a right Eyman contends has been sabotaged through the use of the emergency clause) but the Legislature remains primarily responsible for legislating. It is normal for budgets to be approved without a vote of the people and it is normal for revenue to be raised without a vote of the people.
Tim Eyman happens to dislike taxes as well as representative democracy, so when the Legislature votes to raise revenue, he gets very, very upset. It’s like a double whammy for him. So that’s why this wording is part of his push poll formula. It is a deliberate choice. Eyman is trying to activate his framing in voters’ heads, rather than giving them an opportunity to “express their opinion.”
The second element is a fill-in-the-blank:
” … [identification of tax and description of increase] …”
This part has to be supplied by the Attorney General’s office.
The third element is a ten year cost estimate:
” … costing (most up-to-date ten-year cost projection)] …”
This is extremely misleading.
Eyman states: “The cost of the tax increase is determined by the Governor’s budget office,” meaning, the Office of Financial Management.
But he fails to admit that the ten-year period was dictated by him to make the numbers bigger. It’s an example of deception through manipulated statistics: Anything sounds more impressive when you take it out over ten years.
Eyman knows that governments don’t budget in decade-long increments and revenue forecasts decline in accuracy and usefulness beyond a couple of bienniums. It is difficult to know what even the short-term future holds.
Eyman likes to bring up that many tax increases are indefinite or open-ended, meaning they don’t have an expiration date.
However, by that logic, there could be a twenty-year projection or a hundred-year one that’s even more useless. Nobody knows what state revenue is going to look like in the long term. Revenue projections shown to the public should not deviate from accepted best practices, but in Eyman’s push polls, they do.
The fourth element is another prejudicial statement:
“… for government spending.”
Whenever the Legislature passes a revenue increase, it’s always for a reason and a purpose. Eyman’s push polls deliberately omit what that reason and purpose is, because he wants to activate a right wing frame about taxes in voters’ minds.
To justify this, Eyman argues:
“Once the money starts flowing in from a tax increase, the Legislature is within its power (and 99% of the time does) divert the money to whatever pet project or programs they want.… So the wording in a Tax Advisory Vote… is the only thing the voter knows for sure will happen to the money from a tax increase.”
Wrong. When the Legislature raises revenue, it specifies where that revenue should go by enacting a law, and voters deserve to know this information. Withholding essential details about what a law says and does is unacceptable.
Eyman’s push polls appeared on general election ballots after the Legislature had adjourned Sine Die for the year. Changes to revenue bills rarely occur in between Sine Die and a general elections since the Legislature is out of session. There was never even a figment of a good excuse for failing to connect the dots.
A future Legislature might change the law, but that doesn’t change the fact that the revenue was previously raised for a declared reason and a purpose that was set forth in our body of laws — which is something we can be sure of. Even on occasions when dedicated funding is later “diverted,” the money is still benefiting one or more public goods, and we can see which ones by looking at our RCWs.
For example, here is a law that says where money in the Education Legacy Trust account must go. This is the account that proceeds from our new capital gains tax on the wealthy are deposited into. RCW 83.100.230:
Education legacy trust account.
The education legacy trust account is created in the state treasury. Money in the account may be spent only after appropriation. Expenditures from the account may be used only for support of the common schools, and for expanding access to higher education through funding for new enrollments and financial aid, early learning and child care programs, and other educational improvement efforts.
The law that created our new capital gains tax on the wealthy, Engrossed Substitute Senate Bill 5096, was subjected to an Eyman push poll in 2021.
The push poll did not explain who would pay the capital gains tax, what the rationale for it was, or what the revenue would benefit.
This is the language voters saw:
The legislature imposed, without a vote of the people, a 7% tax on capital gains in excess of $250,000, with exceptions, costing $5,736,000,000 in its first ten years, for government spending.
Senate Republicans have gone around crowing that most people voted “Repealed” in response to this Eyman push poll, waving the figures around as if they meant something. For the record, the figures don’t mean anything. They’re worthless. They cannot inform us how Washingtonians feel about the capital gains tax.
The people were not asked a neutral question and therefore were deprived of an opportunity to give feedback to their elected representatives.
Contrast that with our legitimate public opinion research.
A few months before voters saw that Eyman push poll, we asked this question:
QUESTION: Proponents say that Washington State’s new state capital gains tax on the wealthy will raise about $500 million a year in crucial funding for education in Washington State, including early learning and childcare, and will help balance our upside-down tax code by requiring the wealthiest 8,000 individuals to step up and pay their fair share in dues to our state. Opponents say that the new state capital gains tax on the wealthy is an unconstitutional and illegal income tax that will hurt job creation and put the state at a competitive disadvantage, hurting the whole economy while failing to address regressivity. Both sides agree that the text of the capital gains tax law fully exempts retirement accounts, family farms, and all real estate. Having heard the arguments for and against, do you strongly support, somewhat support, somewhat oppose, or strongly oppose Washington’s new state capital gains tax on the wealthy?
2021 ANSWERS:
- Support: 57%
- Strongly: 39%
- Somewhat: 18%
- Oppose: 40%
- Somewhat: 10%
- Strongly: 30%
- Not sure: 3%
(For methodology information, please read the post.)
See the difference? Our question is balanced — it provides arguments from both sides. In fact, it utilizes Rob McKenna’s arguments against the capital gains tax pretty much word-for-word. It explains who pays and where the money is going.
Here, respondents were given context and informed of the clashing perspectives. They then had an opportunity to give their opinion. Had we pushed them to respond one way or another, we would have deprived them of that opportunity.
There’s one final element of Eyman’s push polls to break down: the answer dichotomy. Eyman left this out of his “belated defense,” but I’m going to discuss it. Eyman’s formula specifies that voters be given two choices:
[ ] Repealed
[ ] Maintained
The formula dictates that the first choice that must be shown to voters is “Repealed” and the second is “Maintained.” That is the inverse of the normal dichotomy that voters expect, which would be either “Yes” or “Approved” shown first, followed by “No” or “Rejected” shown second. Eyman flipped the order to encourage “Repealed” responses by voters, and chose a word that isn’t as strong as “Repealed” to be the other choice — “Maintained.”
Talk about stacking the deck!
With a real ballot measure, voting Yes or Approved passes the law; voting No or Rejected defeats it. With Eyman’s push polls, regardless of how people responded, the law didn’t change. “Maintained” was the only possible outcome one hundred percent of the time and there was no disclaimer stating this anywhere.
The only clue that Eyman’s push polls were fake referenda was in their headings, which contained the word “advisory.” How many people incorrectly inferred from the answer options that the push polls were binding, when they actually weren’t?
Eyman’s push polls constituted fakery in the extreme, a con against the voters.
But not anymore, as Eyman said in boldface in his email. At last, the Eyman push polls are hitting the dustbin once Governor Jay Inslee signs Senate Bill 5082.
Eyman’s push poll scheme was also rigid and prescriptive with respect to what had to go into the voter’s pamphlet. There were no pro and con statements facilitated by the Secretary of State and valuable additional context was not offered there either — only selected info that Eyman wanted people to see.
That’s a lot of loaded language, both on the ballot and in the pamphlet.
Eyman’s denial of this truth is, as far as I’m concerned, an admission of guilt.
Voters win with SB 5082; right wing operatives lose
Judging by a commentary that KTTH published yesterday from often grumpy talk show host Jason Rantz, Eyman is not the only one who’s upset.
Right wing Republicans are taking the elimination of Eyman’s push polls pretty hard. They’re already out of power in Washington, having not won a gubernatorial election since 1980, a U.S. Senate election since 1994, a state House legislative majority since 1996, or a a state Senate legislative majority since 2016.
Consequently, they have become increasingly fond of Eyman’s push polls. These anti-tax ads, dressed up to look like ballot measures, were costing them nothing since taxpayers have been footing the multi-million dollar bill every year.
But now we’ve put an end to this totally inappropriate in-kind contribution.
Only one Republican broke with Eyman and the Washington State Republican Party to vote for Senate Bill 5082 in 2023: State Senator Brad Hawkins.
However, as I mentioned above, is Hawkins is not the only Republican in Washington who supports our legislation. He has company!
In our polling, we discovered that many Republican voters favor abolishing Eyman’s push polls because doing so saves millions of dollars and eliminates something that definitely isn’t a core function of government.
Hawkins’ former colleague Hans Zeiger had this to say four years ago when he voted yea on our legislation the first time it came up in the Senate:
Thank you, Mr President. Rising in support of this bill… and although I add the caveat that there will be probably a number of people on this side of the aisle who oppose the bill.
But I’ll tell you why I support it. Advisory votes are like polls after the Legislature has taken a vote on a tax issue, and here’s why I’m okay with removing these kinds of votes from the ballot after we’ve had now over a decade of experience with these kinds of votes.
I get calls every election from people who are perplexed by advisory votes since they have no effect on the outcome of a tax issue after the Legislature has already acted.
Polls certainly have their place in our policy process, but I believe that items that show up on the ballot should actually enable voters to make a difference in the outcome of an issue. It is the referendum and the initiative enshrined in our state Constitution that is the mechanism by which citizens can uphold or overturn an act of the Legislature on a tax issue, or any other issue. That is a process that actually fulfills a legislative function.
Indeed, the people of this state are just as much a part of the legislative branch as we are. So, I hope that we’ll find new ways for people to give the Legislature feedback on any number of issues that we cast votes on by contacting us directly.
Face to face relationships matter greatly in our democratic process and the main way that I think that we’re going to rebuild trust in this process is by finding new ways for citizens to get personally involved. But for now, I think that… I don’t see that advisory votes fill that function, and so asking for a yes vote on the bill.
Distinguished Republicans like former Secretary of State Sam Reed are also supportive. Here are Secretary Reed’s observations from a few years ago:
The issue is about “advisory votes” on measures with fees or tax increases. Since they are non-binding and meaningless, they are a total waste of money. Fiscally conservative Republicans object to such wasteful spending. It cost hundreds of thousands of dollars for the Secretary of State’s Office to put those measures on the ballot and in the voters’ pamphlet. They are typical grandstanding by Eyman at the taxpayers’ expense.
Not to mention the millions of dollars counties have incurred in added ballot design, printing, and tabulation costs every year for a decade!
We are deeply appreciative to these Republicans for making our legislation bipartisan. SB 5082 is good for voters and good for democracy. We are looking forward to its addition to the Revised Code of Washington later this year.
