Internet Explorer 11 is not supported

For optimal browsing, we recommend Chrome, Firefox or Safari browsers.

What’s the Matter With Legislatures?

A lot, says one prominent political scientist. But most of all, they aren’t accountable to anyone.

The Utah House in session.
The Utah House of Representatives in session. (Matthew Thomas Allen/Shutterstock)
American democracy rests on quite a few fragile foundations, but one of the most important is the idea that when voters go to the polls, they more or less know what they are doing. But do they? Scholars and journalists have been arguing about this for the past century.

The legendary pundit Walter Lippmann essentially touched off the debate in the 1920s, when he wrote that “the average voter has been saddled with an impossible task and … he is asked to practice an unattainable ideal.” A few decades later, rigorous research at the University of Michigan came to the same conclusion.

The opposite point of view boasted its own big names. The eminent political scientist V.O. Key Jr. famously declared in the 1960s that “voters are not fools.” He argued that while they might not know the details of what a candidate was proposing, they were perfectly capable of judging the quality of government that they had been getting. Key wrote that “the electorate behaves about as rationally and responsibly as we should expect, given the clarity of the alternatives presented to it.” In the decades after that, the late political writer David Broder (a particular hero of mine) repeatedly asserted that the voters were ahead of the politicians — that they knew more about what they needed than those who served in office did.

And so it went. If there was any consensus view, it was that voters knew a fair amount about what the president was up to and a smaller but still significant amount about what was going on in Congress. But when it came to states and their legislatures, the skeptics appeared to have the upper hand. One of the Michigan studies found that 30 percent of the electorate couldn’t come up with something they liked or disliked about their elected leaders, and that 70 percent couldn’t name anything their legislators had done for their district. Philip Converse, one of the Michigan researchers, proclaimed that when surveyed about state politics, most voters offered “meaningless opinions that vary randomly in direction.”

Now the doubters have been given reinforcement by Steven Rogers, a political scientist at St. Louis University. In a new book called Accountability in State Legislatures, Rogers argues that the reality is worse than even the die-hard skeptics have thought: The voters are uniformly ignorant about state politics, and the result is legislatures whose members essentially answer to nobody. “Voters are not fools,” he concedes, echoing V.O. Key, “but are asked to do too much in a democracy where federal politics is king.”

THIS IS A SERIOUS FLAW IN AMERICAN GOVERNMENT because legislatures are a crucial component of the political system. A typical legislature considers twice as many important bills as Congress does. Legislatures have primary responsibilities in health care, taxation, regulation of guns and now abortion as well. And yet four-fifths of American voters do not know who their elected state legislators are, and 40 percent are not sure which party is in control. In Rogers’ words, “The evidence of electoral accountability in state legislatures is meager at best.” The legislators seem to agree. In one survey, only 15 percent of them said voters know whom to blame when something goes wrong. “How does someone reward or punish those in power,” Rogers asks, “if they rarely know who is in charge?”

The absence of accountability shows up in many different ways. One is the overwhelming number of state legislative elections in which there is little or no competition. At least one-third of the legislators face no primary or general election opposition at all. “A staggering number,” Rogers concludes, “win re-election by just signing up for another term.”

This creates a state of affairs that can legitimately be described as moral hazard, the common-sense idea that if people know they won’t be punished for their misdeeds or mistakes, they won’t make much of an effort to avoid them. Moral hazard is most commonly invoked as a criticism of public policy: If corporations know they will be bailed out in times of jeopardy, or investors know their large bank accounts are protected, they will be tempted to take imprudent risks.
Rogers book cover
(University of Chicago Press)

Rogers creatively but rather plausibly applies the idea to electoral politics. If state legislators know their seats are safe, and if voters have little knowledge of what representatives are doing, what is there to guard against mischief in the legislative chamber? Not very much. “Why,” Rogers asks, “should a legislator work to stimulate the economy or gain his constituents’ approval if a voter’s election day decision is not based on his own performance?”

THE LATE HOUSE SPEAKER TIP O’NEILL was fond of saying that all politics is local, based on the advice of his father, and most people who follow elections seem to accept that piece of wisdom. But when it comes to legislative elections in the 21st century, the wisdom isn’t so wise. “All politics are not local at the state legislative ballot box,” Rogers tells us. “Fatherly advice is not always right.”

Politics can’t very well be local if the voters don’t know what their representatives are doing. And so we get contests that are supposed to turn on the successes and failures of local legislators, but in fact merely reflect the national mood. We get election years like 1974 and 2006, when hundreds of legislative seats changed parties due to resentment against Watergate in one instance and the Iraq War in the other. Or years like 1994 or 2010, when voters turned out Democratic lawmakers at the state level not on the basis of anything the legislature had done but because they had lost confidence in the Democratic administration in Washington. That isn’t the way things were designed to work.

Some political scientists respond with something we might call “the miracle of aggregation.” The individual voter might not be well-informed or might be mistaken, they argue, but the errors and the ignorance balance each other out, and the result is usually a reasonable decision. This is akin to the belief that individual jurors in a criminal case might be prone to mistakes, but the aggregate decision of 12 of them strikes a fair balance. Or more prosaically, the well-documented finding that when hundreds of people are asked to estimate the number of jelly beans in a huge jar, most of the individual guesses will be off, but if you put them all together, you generally arrive at a pretty accurate guess. My only quarrel with this doctrine is that, in politics at least, there doesn’t seem to be any evidence in its favor. When you multiply individual political ignorance by a large number, you don’t get wisdom. What you get is more ignorance.

ONE WAY TO UNDERSTAND OUR CURRENT PREDICAMENT is to think about the state legislative history of the past 70 years or so. When it comes to accountability, the climate has shifted more than once. In the 1950s, most state legislators were accountable — in a way. They tended to be local lawyers, merchants and insurance brokers who answered to a business elite that had backed them. It wasn’t exactly popular democracy, but it wasn’t moral hazard, either.

That changed in the 1970s. Ambitious young activists, mostly Democrats, began running for legislative seats on their own, bypassing the establishment that had chosen their predecessors. When they won, they could do pretty much what they wanted. But they had to be careful to work their districts and cultivate the constituents who had placed them in office.

The 1990s saw a resurgence of party control, based initially on the centralization of campaign funding. In the decades since then, party dominance has increased dramatically, due in large part to gerrymandering but also to the lockstep ideologies of those who wish to seek legislative office. State legislators, Rogers says, are now disproportionately a collection of ideologues and party hacks. One might call this accountability of a sort, but it is not accountability to the voters who keep returning them to office. The voters, contrary to Key and Broder, don’t really know what they are doing.

Rogers offers a few modest proposals for how we might escape this quagmire. He would like to see higher salaries and, as a more drastic measure, an experiment with unicameral legislative bodies. But he is not optimistic. “I do not think my current judgment about state legislatures is wrong,” he says in an outburst of candor, “but I hope it is.”

I hope so, too. In the years since World War II, we have seen enough changes in the makeup of state legislatures to justify a belief that things can change again, possibly for the better. Voters are not, in fact, fools. As we all know, it is perfectly possible to be intelligent and ignorant at the same time. In a democracy, even one as shaky as ours, there are moments when intelligence triumphs over ignorance. We can all hope that one of them is coming.
Alan Ehrenhalt is a contributing editor for Governing. He served for 19 years as executive editor of Governing Magazine. He can be reached at
Special Projects