This is the fourth in a series of GUEST POSTS by Matthew Knee, a Ph.D. candidate at Yale University specializing in campaigns and elections, ethnic voting patterns, public opinion, and quantitative and experimental approaches to political science.
Today I will provide an overview of the polling situation in the national battle against the public employee unions (PEU’s). This will not be a cheery post, as the battle is not going all that well on the public opinion front. (It seems the moratorium on martial language in politics has been lifted). Tomorrow I will discuss what reformers can do to turn things around.
As I explained on Monday, most persuadable voters lack set opinions on most issues, but do have various values and considerations floating around in their minds. Answering a polling question is an active task. The polling question and the answers provided activate considerations, which are then weighed and integrated into an answer.
Depending on which poll and what question you look at, the reformers are anywhere from about even to facing a two to one disadvantage. The wording and sampled population tend to determine how the landscape looks, but a number of patterns emerge.
As with many issues, from 40,000 feet fiscal conservatives are doing acceptably well, but the closer in we get, the more people shy away from specific forms of austerity and fall into the demogogues’ traps.
1. Which Public Employees Do We Mean?
It seems that mentioning specific kinds of public employees increases support for PEU’s. Previous Gallup polls show that school teachers, firefighters, and policemen – who are usually cited as examples of PEU members in polls, are among the most trusted professions in the country. In fact, a recent Bloomberg poll shows that people overwhelmingly support these groups unionizing, but are divided on the unionization of janitors and office workers.
Preferring unionization of public safety and high-status occupations over lower-status, commoditized occupations is highly illogical from almost any serious policy point of view, but makes sense if respondents simply hear these questions as “blah blah firemen blah” and the answers as “pro or con.” (This phenomenon happens more than you might think.)
2. Polls On Collective Bargaining
The worst results come from questions on eliminating collective bargaining for public employees. This is a tricky question to ask. First of all, I doubt that most people actually know what collective bargaining is, so they are going to rely on the pollsters’ description of it, or just the fact that the words sound good.
There are other issues as well. Proposing a revocation of a sympathetic group’s privileges is inherently unpopular, and describing collective bargaining as a right is not only false but highly prejudicial.
On the other hand, mentioning the budget, pay levels, strikes, or pensions, in the question or previous questions, might bias a poll against the PEU’s. As always, polls that sample adults will skew a bit more liberal than those composed of registered or likely voters.
Results from polls that display most of the pro-PEU features (Gallup, NYT/CBS) show support for collective bargaining in the 60%-30% range. Bloomberg’s just-released poll shows a 64%-32% pro-collective bargaining split on a mostly-neutrally-worded question (despite following questions about public employees making more money) but this could represent voters supporting “rights” when provided with no context whatsoever. Polls that prime negative feelings towards PEU’s with previous questions and/or mention the budget (Quinnipiac, Rasmussen), show the anti-PEU camp ahead by a few points.
This puts reformers in a tough situation. They ARE trying to take collective bargaining away from PEU’s, which does in fact make them fairly useless. Of course, such moves are intended to weaken PEU’s (as is necessary for budget balancing), but that is not the discussion we are having. As usual, conservatives can win – though only barely – when discussing austerity in the abstract, but once spending cuts are portrayed as harming those who would otherwise benefit from them, the American people get cold feet.
3. Polls On Walker vs. The PEU’s
Although this fight looks to be mostly over legislatively, we still can learn from how people respond to this formulation. Rasmussen, whose samples tend to have a rightward skew and whose previous questions, shows a small advantage for Walker, though it decreased from 10 percentage points to 5 between their two polls. Pew, with an overly-Democratic adults sample (D+10) and a mention of collective bargaining, has the PEU’s up by 11 percentage points. Bloomberg’s poll, asking respondents to choose between statements biased each way, went 49%-46% for the PEU’s, but I greatly disfavor this methodology since it is difficult to come up with equally-biased statements.
Still, again, PEU opponents do alright in the most generalized formulations. It is a good sign that the enormous PEU leads in some polls do not materialize usually in polls that provide little information of their own. That could indicate that people are only retaining moderately pro-PEU views of the conflict, even though more descriptive discussions can provoke highly pro-PEU views. This limits the long-term damage, PEU’s will still be able to bring out these very negative views, with some effort and lots of cash, when an election or recall comes around.
4. Balancing the Budget vs. Union Busting
Here the news is somewhat good. In both the Quinnipiac and NYT/CBS polls, which could not be more different on other results, as well as the Bloomberg poll, more Americans believe these attacks on PEU’s are motivated by balancing budgets than hurting PEU’s. This suggests that some of the opposition to taking privileges from public employees may be fairly shallow, since even when most disagree, more attribute the actions of PEU opponents to noble goals than political gain.
5. Generic Feelings Towards Unions
Since this too is a general question, so PEU opponents fare better here than in other areas. The D+10 NYT/CBS poll have union support at 33%-25% favorable-unfavorable, but the rest show better news. Gallup has Americans disapproving of PEU’s by 47%-46%. Rasmussen has Americans supporting PEU’s by a 50%-44% margin, although 54% of independents oppose them. This has become worse, however, over the course of the conflict, with an earlier poll showing a 45%-45% split.
6. Assorted Policies
Opinion is mixed on other policies. More people generally believe that government employees (without mentioning specific types) are overpaid than underpaid, but never constitute a majority. Rasmussen finds overwhelming support for public employees paying the same for their health care as private sector workers (though that is a very loaded question). Strong pluralities against teachers and public safety workers striking, and 50% support for cutting state payrolls by 10% over 10 years. The NYT/CBS poll shows only slight advantages for allowing public employees (with teachers, police, and firefighters mentioned specifically) to retire and earn pensions early, without even mentioning the size of these pensions, which are the main problem, and with a somewhat liberal sample. The same poll has Americans supporting cuts to pay and benefits of public employees to reduce budget deficits by a 47%-46% margin.
7. Partisan Breakdown
On union issues, polls show that Democrats are usually more united than Republicans, especially in the scenarios in which Dems have big leads. This means that big labor is probably closer to their support ceiling than its opponents. This also demonstrates the need for reformers to inoculate Republicans against pro-PEU framing. There is evidence that knowledge can inoculate people against framing, and that this effect is particularly strong when it is in accordance with one’s overall ideological system. This is to be expected, since groups are more likely to be enthusiastic about issues that affect them directly.
So what can reformers do about this? How can they limit the drop-offs in support from general budget-cutting to forcing PEU’s to make major concessions? I will provide some suggestions tomorrow.