More on Member/Constituent Communication

April 3, 2012

Yesterday, I discussed the explosion of inbound email to Congress, and some of its possible impacts, particularly the way it alters a congressional office’s representational relationship with citizens and potentially has a nationalizing effect on politics, because of the difficulty of filtering out constituents from non-constituents. I don’t want to make too much of this, however, for a few reasons. First, it would be vastly overstating the case to say that email is nationalizing congressional politics. If it is having an effect — which we don’t really know, after all we’re just theorizing here based on some aggregate statistics and anecdotal evidence — the effect is a marginal one, and almost certainly pales in comparison to other nationalizing forces in congressional politics.

Second, I kind of (purposefully) set up of a strawman yesterday when I said there isn’t a sorting algorithm for constituent email. As most interns on the Hill know (and as several emailed to remind me yesterday!), there are techniques for filtering out non-constituent email. A little background: almost every congressional office uses a Correspondence Management System (CMS) to process constituent communications, the most popular being Intranet Quorum (IQ). The CMS has a database of all district constituents, and allows you to input all aspects of a constituent mailing — date, subject category, opinion, response, etc. — for the purpose of tracking both individual and district-wide communications data.

So one way to filter the emails — which almost all offices try to do now —  is to force constituents to fill out a web form when sending an email to the office that includes their address. That way, you can cross-check it with the CMS and determine whether the sender is a constituent. Of course, there are problems with this. First, it doesn’t work when people skip the web form and directly email staffers by finding out the email address of the LA on their topic of interest. Second, it requires the web form, which may dissuade actual constituents from writing. Third, it can be beaten relatively easily by anyone who really wants to flood Congress with opinion mail; whereas the old postal sorting system could only be beaten by postmarking stuff from each district, a strategic emailer needs only harvest an address from a district.

The end result is that a lot of the representational promise of email from the mid-90’s has not been borne out. At the outset of the email explosion, many people thought it might be a possible source of data for an office, one that could provide a good guide to district opinion. But at this point, it has largely become something that needs to be managed. The low cost of sending an email has resulted not in more useful communication from constituents, but just more work for interns and staff assistants to make sure that every constituent gets a timely response to their inquiry. And so the largest effect of the inbound email explosion has been on the structuring of the congressional offices: more low-level staffers than ever are assigned to correspondence work, but paradoxically the correspondence itself is of less value to the office.

Now, some will say this was always true of the postal mail. That its value to the office was entirely negative, it only created work in making sure that constituents got timely responses, and that the ideal point for any office was to receive zero constituent mail. Undoubtedly, in many cases that was (and is) true. If you are a solid pro-choice Representative, the only thing that pro-life mail does is generate staff work and resource costs in logging the CMS, printing out the form response for “pro-life constituent position inquiry,” and paying the postage to mail the response.

But I can also clearly remember instances of the mail being used to measure the balance of intense district opinion over hot issues. I worked for a conservative Democrat in 1998, who was very torn on impeachment. We tracked every district letter we received for three weeks leading up to the vote, and the fact that they were 70/30 against impeachment made a strong impression on the staff and my boss.And that’s something that I don’t think is possible with the email. Or at least not as possible as it used to be. And so ffices are left in an odd situation: many of them like to “see what the mail says” when they are thinking about hot legislative decisions, but the mail (and the phone calls in the age of cell phones and free long distance) is less representative of the constituency than it used to be.

But leave all this aside, because what I want to talk about today is the other side of the email explosion — the email and other electronic communications sent out of the congressional offices. I don’t think there’s any question that this side of the equation is not only fundamentally changing how congressional offices think about their constituent communication strategies, but also having a marginal effect on the nationalization of congressional politics, by offering Members the capacity to reach a national constituency that did not previously exist. And in the right circumstances, many Members might find strong incentives  to seek such national constituency. Now, we don’t want to go to far with this thinking; Members still get elected in districts by local constituents, and that’s always going to create an overwhelming incentive to focus on the geographic district first.

So with that caveat in mind, here are five points:

1. Electronic communications are different than old-school mail in three important ways. First, they have very low marginal costs. Sending franked mail to the district not only incurs a fixed marginal cost per letter, but also comes directly out of the Representative’s MRA in the House; any time you send a sizeable mass mailing to your district, it eats into the budget that could be used for staff or other resources. Electronic communications — be it email, social media, tele-townhalls, web advertisements, etc. — tend to have fixed capital or startup costs, but are then largely free on the margin.

Second, congressional offices are not limited as to who they can contact with electronic communications. Following a federal court action (Coalition to End the Permanent Government v. Marvin T. Runyon, et al., 979 F.2d 219 (D.C.Cir. 1992)), the Rules of the House were amended to restrict Members from sending franked mail outside of their districts. So it’s not even possible to reach a wider-than-district audience with postal mail. Electronic communications, however, are not so limited. Members can build email subscriber lists — many offer such subscriptions immediately upon entering their website — and the use of social media tools like Facebook, Twitter, and YouTube allow Members to broadcast and interact with a potential constituency far wider than their geographic district, if they want.

Finally, electronic communications are fast. This is obvious, but it has important ramifications for how congressional offices choose to use it and how it shapes their communications strategy. Once upon a time, if you wanted to send out a tick-tock on the movement of a policy of interest through the floor, the only outlet was more or less to fax a press release to any newspaper that might listen, which invariably meant the local newspapers in your district. There was no point in trying to send postal mail directly to constituents at that speed. Now, however, you can tick-tock floor activity or other business to subscribed email lists or social media instantly, and if you want to send unsolicited mass communications to non-subscribed citizens, the lack of printing time and postal costs let you move them quickly through the franking commission and out the door.

2. The use of franked mail is at record-lows; the use of electronic communications has skyrocketed. The above is all great in theory, but what’s actually happening on the ground? Well, for one, the total cost franked mail coming out of Congress (adjusted for inflation) is at its lowest point since Congress began reimbursing the Post Office for congressional mail costs in fiscal year 1954. In nominal dollars, franked mail costs are down to $12.8 million in FY2011, from a high of over $113 million in FY1988.

Now, the steep decline in mail costs between the late 80’s and the mid 90’s was due mostly to two reforms: public disclosure of mail costs for individual Members, and direct charging of Members’ budgets for the cost of mail they send, instead of allowing unlimited mailings from a common funding source. But there’s a more telling decrease in mail costs in the last eight years, which is somehwhat masked in the chart. Here are the odd-year mail costs from FY2003 to FY 2011:

FY2003: $19.3 million

FY2005: $17.5 million

FY2007: $17.5 million

FY2009: $16.8 million

FY2011: $12.8 million

That’s a 33% drop in just five cycles. And comes during a period when the price of a stamp (which is a rough measure of postal cost inflation) went from 37 cents to 44 cents, more than a 20% increase.

(As an aside, the pattern persists of more mail being spent in even-numbered fiscal years than in odd-numbered years. While many observers have attributed this to the increase in mail sent prior to the elections, the truth is actually more complicated than that. The single month with the most mail sent during any Congress is almost always the December of the first year, as many Members send out end-of-session newsletters. By quirk of the fiscal calendar, which start in October, that means that the two peak months of franked mailing — December of the odd-year and the more modest increase in the two months before the pre-election cutoff in the even-number year (June and July) — happen to fall in the same fiscal year, distorting the stats. If you go by calendar year, there is almost no difference between election-year and non-election year mail totals, as the two peaks cancel each other out. I will do a full blog post on this someday.)

Well, how about the other side of the coin – what has happened to electronic communications coming out of Congress? We don’t have quite as fine-grained data as we do on the postal side, but we have a good proxy: the volume of “mass communications” (defined by the House as “unsolicited communication of substantially identical content to 500 or more persons in a session of Congress” which includes things like mass unsolicited emails, web or print advertisements, radio spots, newspaper inserts, etc. the House has been tracking this data since FY2009. Basically anything you send unsolicited to a whole bunch of people. The chart below shows the volume of quarterly mass postal mailings in the House from 1997 to 2008, and then the quarterly volume of all mass communications (which include postal mailing) from 2009 to 2011.

This graph should be striking. Mass postal mail volumes follow a familiar pattern of peaking in the last quarter of the first year of each Congress (from the December newsletters) and then again in the period preceding the election, and then drop off in the prohibited period (late 3rd quarter and early 4th quarter of election years) and the lame duck 4th quarter of a Congress as well as the 1st quarter of a new Congress. In the first Congress in which mass communications were tracked — the 111th, 2009-2010, a similar pattern was observed, albeit at a naturally greater scale (since mass communications are inclusive of mass mailings). But then in 2011, in the first session of the 112th Congress, the mass communications simply explode, to something like approaching 10 times the volume of mas communications sent in the first quarter of 2009.

My instincts tell me this isn’t due to radically-increased use of mass faxes or mass newspaper inserts. This is almost certainly electronic communications of the internet age, taking off for real out of congressional offices. I have not examined the data carefully yet at the individual level, but I think there’s good theoretical reasons to believe that a lot of it can be attributed to two things: first, the influx of 90+ new House Members, most of whom have come to politics in the information age. Second, and related, is the effect of the 2010 election, in which social media and electronic communication played a large role in both campaign information dissemination and fundraising strategies. Freshmen Members are arriving in Congress already electronically plugged-in to large networks of constituents and non-constituents through email lists, Facebook accounts, Twitter feeds, and other media. In short, times are changing and young replacement Members are most savvy to it. That would be my guess, at least.

Anyway, with the outbound data trends out of the way, let’s get back to the more speculative talk about the impact this might be having representation:

3. The opportunities for surrogate representation have seemingly incrased. In her excellent APSR article, Jane Mansbridge defines surrogate representation as happening when Members represent constituents outside their district. In the traditional formulation, this often happens around specific issues with dispersed national constituencies: Dennis Kucinich representing anti-war advocates, Barney Frank representing gay rights advocates, and so forth. My sense is that, twenty years ago, very few Members were engaged in such surrogate activities. They simply did not have the resource capacity. Members were (and still are) of course barred from sending franked postal mail outside of their districts. The only way to get a national audience was to get on TV — which usually meant having at least the power of a committee chair, or doing something extraordinarily provocative. And it would have been crazy to suggest spending any significant portion of campaign money on outside-the-district activities.

Today, the entire playing field has been rearranged. Even backbench Members can seek a national followings with relative ease, and at virtually no cost. The Internet, and in particular the social media application like Twitter, Youtube, and Facebook, have zero marginal cost. One can stake out an issue, make a concerted effort to become a national leader on the issue, and have some chance of success, all without expending pretty much any marginal resources. The upside is clear: national leadership on issue means a higher political profile both inside and outside the House, more campaign fundraising opportunities, and (lest we forget) greater opportunity to influence public policy. My sense is that Members are beginning to alter their representational strategies around these facts: connecting themselves to national movements, inserting themselves into national policy debates more often, and modifying their fundraising strategies to more optimistically look for out-of-district money. And the more that Members engage in surrogate representation, the less they engage in traditional district representation.

4. There may be electoral pressure to nationalize representation. But it goes deeper than this. Electoral challengers may be nationalizing their representation, too. Why wouldn’t they? If a Twitter townhall  focused on a national issue or a viral youtube clip can expand your potential fundraising base, get your name in faraway papers, and maybe get you invited onto a cable news show, there’s almost no incentive not to do it. Add on that nationalizing a challenger campaign can create an army of pseudo-activists to target the incumbent and its a no-brainer. And thus Members choosing not to undertake a new media strategy might at a serious disadvantage. And pretty much any new media strategy is inherently a nationalized strategy from a infrastructure perspective.

5. Such trends would be in conflict with the basic electoral logic and Fenno-esque model of constituent relations. Certain things have not changed. The most important, of course, is that only people in the district can vote. But there are other important things too: district offices have to be in the district, franked mail still can only go to the district, and so forth. So the electoral connection, and most of the resources available to maintain it, are still tied squarely to the district. And this means that Members will always be tied, first and foremost, to the district. The largest Fenno constituency that the Member has — the geographic constituency — still rules. But it may not be the largest constituency the Member sees anymore when he looks bak home from Washington. The national constituency may now enter his or her thinking — whether he wants it or not; whether he knows it or not — in a way that fundamentally rearranges the lens through which he sees his district.

This has potential implications. The most important thing that comes to mind is that the Member may greater incentives now than ever to try and shape his district in a more national mold. This would be akin to Mansbridge’s idea of “educating” the constituency under an anticipatory representation model. But it might just be a Member choosing to frame issues in the district in a national way, or choosing to emphasize national over local issues when communicating to the district.

6. But the constituents themselves may be nationalizing. Nationalizing their representational profile, of course, is also potentially dangerous from a Member perspective. As Mayhew writes in The Electoral Connection, Members treat national partisan or ideological swings as acts of god that they can’t control; they instead focus on what they can control, mostly district-related things. To tie one’s fortunes to the national party is to place one’s future in someone else’s hands. But this may dovetail with what is happening to constituents: it’s not crazy to suggest that voters themselves are nationalizing as well. And if that’s the case, then Members may be forced into a national representational context, one that affords them less safety from trends they cannot control.

Now, again, we don’t want to go overboard here. The electronic communications are at best having a marginal effect on nationalization of politics or transformation of Member offices, and the effect is almost certainly indirect if anything: by increasing the capacity of Members to nationalize, it offers a greater strategic menu of options to Members who might want to go that route. But working in concert with other nationalizing forces — the centralization of party power in Congress, the nationalization of fundraising, the breakdown of local and regional media structures, etc. — I think it may be playing a bigger role than is currently appreciated.


3 Responses to More on Member/Constituent Communication

  1. On Writing your Congressman | Matt Glassman on April 3, 2012 at 12:53 pm

    […] [This is part 1 of a two part post; for part 2 click here] […]

  2. John on April 4, 2012 at 9:19 am

    Do any of the congressional staffs use any type of IP address sorting.mechanism? Spams can sitll figure out how tyo beat this but it can
    Give a better pic of actual constituantsccs those otside of the of tr locale areas.

  3. Sam on April 9, 2012 at 12:12 am


    It seems to me that the idea of surrogate representation is certainly possible, but only for a select group of members, even though email may make congressional newsletters much more widely available. Given the nature of the subscription system for those outside the district, a certain amount of publicity–which few House members, in particular possess–is required to convince potential non-geographic, surrogate constituents to even reach the required websites to register for solicited emails. In essence, then, those who do reach the site of the surrogate “representative” in question will already feel represented by the member, thereby removing any further incentive for that member to further tailor his newsletters and policies to meet the demands of those subscribers.

    Even among Senators,who, by nearly all measures, are more well-known, it seems nearly all email newsletters are still targeted at the geographic constituency. That fact is obvious right from the beginning of those emails, as the banners of most email newsletters contains something along the lines of “serving the people of New Mexico” or “California’s Senator.” The text of the emails themselves, moreover, are replete with references to the state; even those of prominent Senators, like Dan Inouye, make sure that the geographic constituency understands that it is the focus of their emails. Similarly, those Senators from large, populous states, where, following Lee and Oppenheimer (1999, ch. 5), we might expect Senators to be focused on national politics rather than local benefits, make sure that plenty of attention is paid to the geographic constituency. One Barbara Boxer email, for example, includes references to California three times in the banner and three subsequent times in the first paragraph of her email, even though Boxer is discussing the Recovery Act.

    This is not to say that surrogate representation is not possible over email–it undoubtedly is, as emails from Senators like Bernie Sanders show. I would just contend that the presence of email is no more likely to encourage further surrogate representation in Congress than before email newsletters became so prominent. It may now be easier for members to reach a national audience, but only if that audience is first willing to come to them.

Leave a Reply

Your email address will not be published. Required fields are marked *