Friday, March 15, 2013

Politicians vs. Policy-Seekers: Or, How Parties are Composed of Diverse Actors

"You keep using that word. I do not think it means what you think it means."
- Inigo Montoya.
Background: On Tuesday I wrote a post about the "Hastert Rule" and how the House GOP's recent violations of the "rule" make sense if you think of Congressional parties as election-maximizers. Hans wrote back to stress that “policy outcomes are pretty primary to the true goals of political parties.” Hans goes on to imply that some of us might not really belong to the UCLA school on political parties. Ouch.
At first I thought that this was a minor misunderstanding. Clearly I was talking about legislative parties in general and the House Republicans in particular, while Hans was talking about some other segment of what we call a "party". But, on second reading, it looks like there is a genuine difference in what Hans and I are talking about when we talk about political parties, and it is worth sorting out. At its core, I think Hans and I have different ideas about what parties are and what they (or their components) want.

What is a Party?
But first, some self-promotion. In addition to blogging here, I have done some research on political parties. One strand of work, with Matthew Lebo and other excellent coauthors, focuses on Congressional parties.  We theorize that these parties are election-maximizing organizations who strive to help their members and improve their reputations relative to the competing party. Compared to other research on the topic, we emphasize strategic competition as a source of partisanship and the linkage between legislative behavior and electoral outcomes. We call this the Strategic Party Government framework. My Tuesday post was an expression of this logic.

I have done other work on the party outside of Congress: one article with Jennifer Victor on how lobbyists are loyal partisan donors, and two articles with a couple coauthors on name-trading by political actors. The latter two articles made a big claim that we thought was pretty important: parties are best understood as informal networks of cooperating but autonomous actors. These networks include politicians, formal party organizations (RNC, DNC), party-affiliated interest groups (AFL-CIO, NFIB), and party-friendly media (Human Events, The New Republic). Also in these party networks but not in these studies: partisan donors, lobbyists, campaign consultants, pundits, think tanks. 

The big point of the 2nd line of work is that American political parties are better understood as networks. If one only looks at the formal party structure (RNC, DNC, state parties) they look pretty servile and irrelevant, but in fact the Republican and Democratic teams have extraordinary influence over nominations, campaigns, and governing agendas. 

Well, that explains it. Hans is simply unaware of this cutting edge research and is living in the...what's that you say? We were coauthors? And his other research is on cooperating party elites in presidential nominations and pundits as the source of the ideological glue that builds and binds party coalition? And he has a whole presentation on the concept of parties as networks? I guess we will have to take his concerns seriously. After a quick Star Wars break.

This isn't the party you are looking for.

Parties as Networks: E Pluribus Duum

Let us start with an analogy to organizations that are all-too-familiar to many readers: universities. What is the goal of a university? You might say, "educating young minds." And, yes, there is plenty of educating on college campuses. But it is hard to understand major elements of university life as directly related to education per se. Why, for example, has university spending on administration--people who don't teach--skyrocketed over the last thirty years? Why do universities occasionally pay faculty to not teach? Why do universities provide students with video games, movies, and shuttles to local bars?  In order to answer these questions (which I won't) we would need to develop a more nuanced understanding of the different types of actors who make up a university--administrators, faculty, students, alumni--and how their goals may differ, so that a university is a bundle of cooperating actors with diverse goals operating in a quasi-marketplace.

Now on to parties. We could argue ourselves breathless over what the goal of a Political Party is. Let's skip that. Instead, let me suggest that different actors within a party might have different goals. Media organizations, for example, might want to make money, and find that they make more money appealing to audiences who have a clear ideological and/or partisan bias. For brevity, here's a quick roundup of party network actors and my best guesses as to their goal structure:

Donors: Policy outcomes (actually, there is research on this; I would guess that the party superdonors and fundraisers we are most interested in care about specific policies and/or general outcomes but have come to associate one party with the fate of their policy goals)

Interest Groups: Policy. In part because they are staffed by true believers, but also because their members/donors hold them accountable for shifts in policy.

Campaign Consultants: Winning elections, and profit maximization. At some point they probably cared about issues when they aligned with a party, but by the time they go pro it's all about winning elections. Winners get more contracts, and eventually maybe a book deal.

Party Committees: Winning Elections. And, also, Winning Elections. On recent tours of the DNC and RNC I saw lots of space and staff devoted to fundraising, candidate training, media relations, and even redistricting. Number of sage policy scholars developing nuanced solutions to the nation's problems? Zero.

Lobbyists: Winning Elections and Profit Maximization. Like campaign consultants, they probably affiliated with a party based on issues, ideology, or family background, but at this point in their lives professional lobbyists have a career interest in their party's success.

Pundits: I defer to Hans, but perhaps Policy and Profit. No readers, no job. 

Think Tanks: Prestige. Like universities, they don't make profits, but they do care about their reputation and their donations. Prestige is based, in part, on generating policy ideas that interest the media and politicians, and thus policy change is an indirect goal.

This brings us to politicians, the subset of the party which was the topic of my Tuesday post. What do they want? As individuals, I think most of them get into this game to change the world. Running for Congress is a risky proposition, and once one wins the job its a hard life: six day workweeks, late hours, splitting time between DC and "home", and constant telemarketing. And for all the complaining one hears about Congressional pay, many take a pay cut for the job. You have to be full of policy passion and/or career ambition to throw these dice.

But once they get these jobs, they act like they want to keep them. And most of the time, most of them act like getting reelected is the most important thing to them, since whatever else they want (policy change, unlimited power) they must get reelected to obtain it. And their party leaders emphasize this point: from early on, the message is legislate less, campaign more.

Legislative Parties

Now we come to the critical question: what do legislative parties want? To be clear: by legislative parties I mean all the members of each party in each chamber of a legislature, with elected leaders and formal party rules. 

My view--the SPG argument--is that parties have a simple purpose: winning elections. Legislative parties are not full service providers for legislators; parties don't help legislators run for higher office, find dates, choose restaurants, or enact policies that harm the collective electoral interests of the legislative party.  Sure, it is often in the electoral interests of legislative parties to successfully enact policies, but these legislative successes are a means to the end of winning elections.

In a limited sense, the relationship between legislative parties and their members is like that between corporations and their stockholders. The corporation exists to make money. Period. Stockholders take the money and advance their utility in a million splendid ways: photo opportunities, redemption, cartoons. Similarly, parties exist to advance their common collective interests of legislators: power.

Musical Interlude and Comparison to Hans

This one seems to fit my point:

In his kicking-me-out-of-school post, Hans writes as if political party as a single unit with a single goal (policy) constrained by the obligation to get elected.  I don't think he means this discussed above, we share a view that parties are not monolithic organizations. But it does suggest a view of parties in which there are lot of different actors with homogeneous world views:

If tax rates increase we get the CLAW!*

I am suggesting a more complex view of party networks, in which actors vary in the extent to which they prioritize policy outcomes, electoral victory, and personal profit.  These varying motives complicate their cooperation. The 2012 Republican Presidential nomination is an example. It was pretty clear from the outset that Romney was as good as it was going to get--that is, he was the most electable candidate acceptable to the major factions of the GOP. And yet, the machinations of policy-oriented activists and donors kept other candidates alive, fanned by conservative media with a greater interest in ratings than forming a united front.

Comparing Politicians' Motives

Hans suggests that parties sometimes promote policies that are not in their electoral interests, e.g. health care. This is tricky, because a) there are several dimensions to a party's "electoral interests," and b) the ideal basis for estimating electoral interests is a comparison to a fantasy world. First, electoral interest includes maintaining the support of the policy-demanding elements of the party network, especially donors and interest groups. So parties may promote proposals that are not generally popular but are important to their base voters, e.g. the Ryan budget cutting tax rates for the highest income taxpayers.

Second, the fact that legislative parties may promote legislation that correlates with electoral losses (e.g. Obamacare) does not prove that parties don't care about their electoral interests. The appropriate comparison is between the events we observe (votes for a landmark health care reform law) and the events we do not observe: not attempting health care reform in the first place, or trying and then giving up. Democrats clearly paid a price, but having promised reform to their base (including key endorsers), failing to deliver would have sparked a revolt. And if the Democrats had stopped midway--e.g. after Scott Brown won Ted Kennedy's seat--they would have likely have paid a huge cost for trying and disappointed their base.

We might also ask whether legislative parties engage in behavior with electoral payoffs without any prospect of changing policy. As it happens, yes they do!  It was clear early in 2009 that climate change legislation, aka cap and trade, had little hope in the Senate, but then-Speaker Pelosi made passage of a doomed climate change bill a top priority. Second, in 2011 the House Republicans voted to entirely repeal Obamacare, and eventually voted 33 times to repeal all or part of Obamacare. They did this after reading the Constitution aloud in the House chamber, so they had to know there was a Senate and President involved, but they persisted anyway because it was good politics for them.

Or, we might also ask if legislative parties sometimes invest effort in enacting policies that are good for them politically but contrary to the (proclaimed) policy preferences of their members. Famous case from the last decade: the passage of the Medicare prescription drug act of open-ended expansion of an entitlement program without any sort of offset...just hundreds of billions of dollars tacked onto the debt. When Democrats do this it's called socialism. When Republicans do it it's called "buying senior citizens' votes". What it is not called is "conservative."

* and by "the CLAW" I mean a primary challenge by the Club for Growth.


  1. Excellent post, the part about the parties at least. Funny that the public was not included in your list of who a party consists of. Apparently whoever said democracy is of, by, and for the people was smoking something. I think an interesting question is, what role do parties play in a democracy? Sure, it serves the interests of various constituents, but if democracy is intended to be government that is of, by, and for the people, and it’s rather by, and for those constituents, does that tell us that something about what is missing from democracy? I think it does. I think the enormous distance between citizens and the government inevitably disconnects them from the government and leaves a gigantic gap, or void. Special interests fill that gap and create political parties. I’ve written more about that here: The site and some of the ideas are dated at this point, it’s a start.

  2. I should just declare victory. Greg starts out saying parties only care about re-election, and now he has come around to saying half the party cares about policy.

    The next step is to accept that all those diverse goals can be held by the same people, and that someone needs to balance those different goals. It's not just a marketplace of diverse folks. The party is about solving the problem of competing goals, not just saying there are different goals.

    I couldn't say it better than Greg, Seth and I said it back in that piece in BJPS:

    "In the network approach, the defining characteristic of a party is co-operative behaviour, not formal positions. Actors ‘join’ the party when they begin communicating with other members of the network, developing common strategies and co-ordinating action to achieve shared goals."

  3. I've enjoyed this exchange greatly, in part because my entry into the world of party networks came through the study of party-aligned interest groups:

    These groups are classic example of entities with multiple goals. Not only do they want policy success, not only do they want favorable candidates to win office, but they also need their group to stay in business. So sometimes the NRA needs to play up the threat to gun ownership when there is none.

    In regards to legislative parties, isn't one important goal simply the exercise of power, both by leaders and the party collectively? Not only do House Republicans need to maintain their majority, but John Boehner wants to be re-elected as Speaker. (Sometimes that may mean letting the party take actions that actually hurt its electoral reputation -- taking relatively extreme positions that please the more policy-oriented members of his caucus).

    He also wants the majority party to remain the dominant coalition in the House, not some VAWA-type amalgam of Democrats and electorally vulnerable Republicans, i.e. not become John McCormack circa 1962, holding the gavel, but seeing the "conservative coalition" winning many critical votes.