National Journal Ranks Your Congresscritters, But Is This Exercise Actually Useful?

National Journal, as is its wont, has released its latest round of rankings in an effort to determine who are the most liberal and most conservative members of Congress. Congratulations to those senators and representatives who have won these honors. I'd list them, but they are a range of fairly uninteresting people (Idaho Republican Sen. James Risch) or multiple people tied for first. The "most liberal representative" is actually a seven-way tie for the top spot.

Here's something even crazier. National Journal would have you believe that Vermont Sen. Bernie Sanders (I) -- a self-described socialist who often seems to be one of the few senators who authentically wants to not gut Social Security in some way -- is the 37th most liberal senator. That's just wildly wrong! For Pete's sake, the man wrote a piece for The Huffington Post titled "What Can We Learn From Denmark?" I think that if a United States senator openly advocates for Scandinavian-style social democracy, you are obligated to give him at least two kajillion "liberal points."

So it's not for the first time that I've wondered, What is the use of these rankings, actually? Based on the "news" generated by the story -- and yes, those scare-quotes are intentional -- not much:

Welcome to today's Congress, which in 2013 was more polarized than any Congress since National Journal began calculating its ratings in 1982.

For the fourth straight year, no Senate Democrat was more conservative than a Senate Republican -- and no Senate Republican was more liberal than a Senate Democrat. In the House, only two Democrats were more conservative than a Republican -- and only two Republicans were more liberal than a Democrat. The ideological overlap between the parties in the House was less than in any previous index.

The ideological sorting of the House and Senate by party, which has been going on for more than three decades, is virtually complete. Contrast the lack of ideological overlap with 1982, when 58 senators and 344 House members had voting records that put them between the most liberal Republican and the most conservative Democrat; or 1994, when 34 senators and 252 House members occupied the same territory.

This isn't really news to anyone. For the past four years, the Republican Party has been eating its own in an effort to achieve some Platonic ideal of ideological purity. Thomas Mann and Norm Ornstein explained all of this pretty plainly back in 2012. It's been well established that the most dangerous thing a Republican officeholder can do is "work across the aisle" (Democratic and independent voters are way more forgiving of this), and so that's why we don't have a Sen. Dick Lugar anymore.

But that's the old news! Now, a new conservative group called For America wants to "topple" much of the GOP's leadership, including people like Senate Minority Leader Mitch McConnell -- who is arguably the most effective conservative ideologue in Congress. (There is no equivalent desire for bloodletting on the left.)

So, yes, the trend of polarization is probably going to get worse, but the way National Journal frames it makes it sound like partisan polarization is a disease. It's not; it's a symptom of a disease called "partisan warfare."

Texas political scientist Sean Theriault offers a useful primer on the difference between partisan polarization and partisan warfare, but here are two sentences that sum it all up rather well: "The difference between my senators is that when John Cornyn shows up for a meeting with fellow senators, he brings a pad of paper and pencil and tries to figure out how to solve problems. Ted Cruz, on the other hand, brings a battle plan." Cornyn, by the way, is one of For America's targets. He's targeted because he threatens to "solve problems."

But National Journal has compiled all of this data, right? Sure. Here's what the magazine did:

For the 2013 ratings, National Journal examined all of the roll-call votes in the first session of the 113th Congress -- 641 in the House and 291 in the Senate -- and identified the ones that show ideological distinctions between members. Many votes did not make the cut -- those that involve noncontroversial issues or that fall along regional lines, for instance. In the end, 117 votes in the Senate and 111 votes in the House were selected and were categorized as economic, foreign, or social. ...

Lists were downloaded from the House and Senate websites showing how all the members voted on the selected votes. The votes in each issue area were then subjected to a principal-components analysis, a statistical procedure designed to determine the degree to which each vote resembled other votes in the same category (the same members tending to vote together).

The analysis also revealed which yea votes correlated with which nay votes within each issue area (members voting yea on certain issues tended to vote nay on others). The yea and nay positions on each roll call were then identified as conservative or liberal.

All these votes were then assigned a "weight" that was "based on the degree to which it correlated with other votes in the same issue area," absences and abstentions were reckoned with, and everyone got ranked in order of who is liberal and who is conservative. And somehow Bernie Sanders ends up being a Blue Dog Democrat, for some reason.

So, yes, that's a lot of activity and a lot of data, but it seems to me to be merely sufficient to make a lot of bad assumptions and provide the foundation for a lot of faulty premises. The big "story" about growing polarity is an example. They compiled ideological data and imputed from the data that it explains the widening polarity. The problem is that the widening polarity is driven by factors that are, as Theriault describes, "beyond ideology."

But beyond that, I think that these rankings make a lot of strange assumptions about what it means to "vote liberal" or "vote conservative" and that their reasoning is actually detached from the logic that governs how voting happens.

Let's take a hypothetical example. Imagine, if you will, a very liberal member of the Senate, who proposes that the Senate pass a bill to fund, say, an infrastructure project to the tune of $10 billion. Along comes a less liberal member of his caucus, who says, "I'd like to vote for something like this, but it should be a $5 billion project." The original proponent says, "But it would be more effective if we funded it at my level." His colleague replies, "That may be true, but at five billion, I can get you six GOP votes."

And let's say that something miraculous happens, those six votes are obtained, and the Senate passes the $5 billion infrastructure project. Who voted liberal? Who voted conservative? Do the six conservative senators get credit for making this bill more conservative, or do they get punished for helping to pass a "liberal" bill? Does the original liberal senator get bonus points for wanting a more liberal bill in the first place? Actually, leave that aside -- is a $10 billion infrastructure project more liberal than a $5 billion one?

This is why we describe this as "sausage making": Bills usually end up being a grim mixture of a lot of parliamentary imperatives, all of which add up to something distinct from the ideological directives that drive the players involved.

Back when we had a "Sen. Russ Feingold (D-Wis.)," Feingold often cast votes against the Troubled Asset Relief Program. He didn't vote for it when George W. Bush was president. He voted to end it twice when Barack Obama was president. When was Feingold voting "liberal" and when was he voting "conservative"?

In 2013, Reps. Justin Amash (R-Mich.) and John Conyers (D-Mich.) proposed the appropriately named Amash-Conyers Amendment, which was intended to put the kibosh on the National Security Agency's bulk collection of American telephone data. The amendment failed 217-205. Voting in the affirmative were 94 Republicans and 111 Democrats. Voting against the amendment were 134 Republicans and 83 Democrats. Who voted "liberal," and who voted "conservative"? A majority of the yeses happened to be Democratic. Does that mean Justin Amash cast a "liberal vote" on his own amendment? Because that would be news to him.

Finally, let's talk about something that I've thus far left out of the discussion entirely: the fact that many members of Congress cast many votes that are not rooted in ideological frames like "liberalism" or "conservatism" at all. Rather, they cast votes in alignment with the powerful corporate interests that lobby their offices and fund their campaigns. Many members of Congress -- I'd say "most," actually -- don't wake up in the morning and say, "Time to do some liberal/conservative stuff!" Rather, they wake up ready to do their campaign donors' bidding.

This is a useful thing to remember when, say, West Virginia Sen. Joe Manchin (D) deliberates over what he'll do in response to the fact that 300,000 of his constituents had to go without safe drinking water. He could advocate for strong regulation that puts the fear of God into negligent scofflaws. Or he could have his ear bent by the chemical and coal industries and protect their interests because they've funded his political career. In all likelihood, he'll do something "in the middle." How will those decisions be scored as "liberal" or "conservative"? I don't have the foggiest idea, and I won't trust anyone who says that he does.

That's why when it comes to properly sorting our members of Congress, I'll take a pass on National Journal's rankings and migrate instead to things like the Center for Responsive Politics' "Open Secrets" site, to see what's really driving the members of Congress to vote the way they're voting. That's where the raw power is calculated. And, spoiler alert, you can basically sort the entirety of Congress into "a few honest people" surrounded on all sides by "on the makers" and "on the takers."

There are more things in the House and Senate, National Journal, than are dreamt of in your tidy left-vs.-right paradigm, I'm afraid.

[Would you like to follow me on Twitter? Because why not?]

113th Congress Facts