Two recent examples illustrate the inherent faultiness of relying on polls to dictate legislation. The first involves arguably the single most disagreed upon economic policy in history: The minimum wage.

According to a recent Reason-Rupe poll, nearly three-quarters of Americans favor raising the minimum wage from the current $7.25 per hour to $10.10. On its face, that result appears closed to any reasonable debate. Yet asking a disinterested group of people whether, in general, low-income workers should make more money is as informative as asking people whether they themselves should make more money — the answer is obviously yes.

Such a poll ignores the fact that in economics there are no isolated decisions; there are only tradeoffs. And when presented with the tradeoffs of raising the minimum wage — fewer jobs for low-skilled adults and teenagers — majority support flips to majority opposition, with 57 percent opposing the policy and only 38 percent supporting it.

In other words, it depends on how you ask. The minimum wage isn’t the only issue where legislating-via-polling is as likely to subvert the will of the people as it is to further it. Consider short-term “payday” loans. According to a poll by the activist Pew Charitable Trusts, 72 percent of respondents believed payday loans should be more heavily regulated. But a Harris Interactive poll that focused exclusively on the views of payday loan recipients found fully 95 percent of them felt it should be their choice — not the government’s — to use payday loans, with an overwhelming majority being either satisfied or very satisfied with their experiences.

The point in these examples isn’t that the voting public is too dumb to decide matters for themselves, but that people are by nature susceptible to supporting cozy-sounding policies when not informed of any offsetting consequences. Polls can be informative of broad trends among the general public, but they should not be used as a tool to dictate legislation.

Ultimately, the root of the problem is not that people are voting on issues about which they are not fully informed — it’s that people are voting on issues on which they should never be voting at all. The solution requires returning the role of government to a more clearly defined and limited scope, like that enumerated by the 19th century political philosopher John Stuart Mill: “The only purpose for which [government] power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others.”

Applying this philosophy to the above examples makes clear that the government is currently out of bounds. The wages a person is paid, or the conditions that attach to a loan, are private matters of concern only to the consenting adults engaged in such mutually beneficial exchange. They do not harm individuals, and should thus not be subject to government control.

Restricting the role of government as such will no doubt be difficult to achieve. However, it is the most effective way to insulate ourselves from the systematic voter ignorance that leads to harmful policies — and prevent politicians from doing the same.

James Stacey Taylor is an associate professor of philosophy at the College of New Jersey.