Curbing Corporate Lobbyists: Easier Than You Think

AP Photo/Patrick Semansky)

A sign discourages lobbyists from entering a hall that leads to the Maryland state senate chambers in Annapolis, Maryland, Wednesday, January 8, 2014, on the first day of the 2014 legislative session. 

American democracy is overwhelmed with corporate lobbying. At every stage of the process, businesses maintain armies of lobbyists, lawyers, and other advocates to weigh in aggressively on all economic and regulatory policy decisions. To put it simply, narrow business interests are wielding disproportionate influence in Washington.

Part of the problem is that congressional offices and their policy staffs, under pressure from federal belt-tightening, are stretched too thin. For decades, more and more organized interests have been coming to Washington, hiring more lobbyists and pressing more demands on Congress. For decades, congressional staffers have found themselves increasingly overwhelmed by this flood of demands. As Congress has failed to invest in its own policymaking capacity, the ability of offices to handle this ever-expanding flood has worsened.

We are now stuck in a reinforcing cycle of disappointment: Voters overwhelmingly see a political system where their “vote does not matter because of the influence that wealthy individuals and big corporations have on the electoral process.” They overwhelmingly see a government of widespread corruption. And they unequivocally disapprove of Congress.

For a long time, reform advocates have argued without success that the solution to this perceived “corruption” is campaign-finance reform. While we support reforms that would encourage small donors and provide partial public funding, such changes only go so far. Much of the power of corporate lobbyists comes from their ability to deluge overwhelmed congressional offices with information. Staffers often don’t even know who is funding their bosses’ campaigns. They just want to sound like they know what they’re talking about on complex policy issues, and lobbyists are more than happy to help. This is not so much corruption as it is an imbalanced information overload.

It’s time for a new approach, one that tackles this problem of information. We propose a simple solution that leverages technology to make the lobbying process more transparent and accountable, to improve congressional decision-making, and most of all, make the political system more representative. We call the system “Post-Map-Ask.”

Here’s how it would work: Lobbying advocacy papers are posted on a public website. Semantic text analysis software maps the arguments into groups, providing an overview of the various positions and claims. And then policymakers ask for underrepresented views, to make sure they hear everything they need to hear before deciding on a policy. Unlike other reform proposals that require passage in a gridlocked Congress, Post-Map-Ask could happen today as a voluntary system—if we build it in a way that congressional offices find so useful that they demand lobbyists start using it.

 

Lobbyists spend much of their time going to Capitol Hill to meet with congressional offices, trying to get them to introduce legislation, co-sponsor legislation, oppose legislation, amend legislation, make public statements, and pester agencies, among other things. Much of this work is steady and persistent asking—making a clear and compelling case for why whatever that lobbyist wants is both good policy and good politics.

This subjects policy staffers in Congress to endless competing demands on their attention. These staffers tend to be young and inexperienced, and have few tools to help them figure out how to prioritize. Since turnover rates tend to be high in congressional offices, it is very easy for Hill aides to get overwhelmed.

In this chaotic environment, staffers tend to revert to reactive mode. They take the meetings and phone calls that come to them, responding to the lobbyists who show up asking for their time. After all, it is easy to fall into the habit of thinking that if somebody had a concern they would register it—since so many people do, in fact, register their concerns.

As a result, when it comes time to prepare the briefing binders and to make policy recommendations—whether to co-sponsor a bill, whether to sign onto a “Dear Colleague” letter, or how to vote on an amendment or a bill—the staffer has only her correspondences and conversations to rely on. Sure, she can (and often will) reach out to other colleagues on the same committee or in the same state delegation. But they, too, will generally rely on what came their way. Few staffers have time to organize communications by issue any more coherently than by searching their email inboxes.

Given the competition for limited attention, those who can speak loudest and most often are in a strong position. They can overwhelm congressional offices by constantly showing up.

The big problem is that this information stream is not balanced. The $2.6 billion in reported annual corporate lobbying spending is now more than the $2 billion combined (fiscal 2015) budget for the entire Senate ($860 million) and the entire House ($1.18 billion).

Perhaps even more significantly, the types of organized interests that we might expect to provide a countervailing force to business—labor unions, groups representing diffuse public interests like consumers or taxpayers—spend only $1 for every $34 that businesses spend on lobbying. Of the 100 organizations that spend the most on lobbying annually, consistently 95 represent the business sector.

 

But the chaos of the process is not inevitable, and technology offers a solution. In fact, the system would be relatively simple to improve. The Post-Map-Ask model would work like this:

In step one—the “post” stage—the Library of Congress would establish a portal for each bill introduced, whereby interested parties could upload advocacy material, which the Library of Congress would then host and catalog. Advocacy materials (relevant one-pagers, reports, op-eds, letters, etc.) would be attached to specific bills. Bills with multiple sections and sub-sections could have advocacy organized by specific sections and sub-sections.

One might worry that such a system will produce even more information overload. But this overload is already happening. We cannot stop it. We can only hope to manage it and make it more transparent. That’s where mapping comes in.

In the second, “map” step, an advanced computer-assisted text analysis would provide simple plots of where interest groups fall across multiple dimensions. Such plots would also show how groups’ positions relate to existing proposals. And perhaps most importantly, they would show which groups’ and players’ positions were missing.

By analyzing the words and clusters of words in dozens to hundreds of documents, computer-assisted text analysis can help congressional decision-makers and outside observers sift through the mountains of information, discern the main themes and concerns, and distill the key debates and preferences.

That brings us to the final stage of the process—the “ask” step. At this point, it may often become clear that one side has dominated the process, and that other perspectives are missing. Here, we see a role for congressional committees asking questions so that they make sure that all sides get a chance to make their case. An online transparency system makes it easy to map the perspectives that are already at the table, and to thus figure out who has not yet been represented.

We envision this as a voluntary system for two reasons. One, it means we could get started building it immediately, if resources were available. Two, it means that it can be flexible and adaptable, like a lean start-up, rather than constrained by rigid statutory definitions. A voluntary system poses no First Amendment challenges.

The system only works, of course, if lobbyists upload their policy papers, and if staffers consult the site. For this reason, design is key. The system must be well-designed, and useful for staffers. This will require a significant up-front investment, with considerable user testing.

Ultimately, lobbyists will use the system if staffers ask them to do so. And staffers will ask them to use the system if it makes their jobs easier, which a well-designed system will do.

While lobbyists may resist at first, we believe that most lobbyists will come to see this as useful. After all, the majority of lobbyists can confidently claim that they are simply making the best argument for what their clients want, and that they have nothing to hide or be ashamed of.

Ultimately, this system might look something like the amicus curiae brief system that the Supreme Court uses to solicit external opinions. Anybody with an opinion is free to share it with the Court. The Court has no obligation to use these opinions, but judges frequently draw on them in their rulings. Amicus briefs enjoy a high level of prestige, and are seen by many as a valuable way of participating in judicial decision-making.

 

A leading Post-Map-Ask system benefit is that it makes the legislative process much more transparent. If it works well, we will have a better understanding of who advocated for what, and why.

This transparency will produce more accountability. Lobbyists can tell staffers many things, and unless the busy staffer aggressively fact-checks what she is told, she does not know whether she is getting a fair assessment of an issue.

But the typical busy staffer does not have time to do this fact-checking. So she must rely on her own intuition and whatever other communications come her way, or on whomever she happens to know who can provide an independent assessment.

We fully acknowledge that public policy analysis is complicated, and many public policies involve difficult trade-offs between competing public policy values. We are not so naïve as to fall prey to technocratic utopian visions of pure scientific administration and fully rational scientific debate. However, we also believe that advocacy as it is currently practiced involves much sloppy and unsubstantiated analysis, as well as considerable factual omission. Our hope is that by bringing all lobbying advocacy into the open and subjecting it to a kind of public review, it will weed out the most misleading policy papers and analyses. 

Over time, the post-map-ask model might help produce a PolitiFact-style fact checking mechanism for lobbying and advocacy. Perhaps a rating system might even develop around individual lobbyists and organizations, and those who are found to be distorting or misrepresenting the truth would lose credibility.

From the perspective of a Hill staffer, this would make life considerably more efficient. Rather than having to frantically search through email or cull through an overflowing stack of papers and folders, staffers could just go to the Post-Map-Ask website to learn who is arguing what on a particular bill or policy, and whom to contact should she have any questions.

Obviously, this can never fully replace the work that staffers should do in thinking through complex policy issues. But it should save them considerable time in organizing the resources they need, freeing them up to do more of their own independent analysis.

Since congressional offices suffer from a lack of resources, this website could be a very cost-effective way for congressional offices to save lots of time and do more with less. Using this website would be an easy way for staffers to get smart in a few hours on issues that in the past might have required them days to survey. Our current lobbying system is distressingly unequal in the representation it affords to competing interests. Business is increasingly and disproportionately well represented. Diffuse interests are poorly represented. The system is also far too chaotic, and this chaos only serves to reinforce the corporate advantage. It also reinforces the disconnect that a growing number of voters feel with Washington—a place where they have no idea what’s going on other than that the place is overrun by special interests and lobbyists, and that there’s no way for them to meaningfully participate.

One way to improve public engagement is to modernize our outdated lobbying and legislative information system. The system would also be open to the public. Interested citizens could see where advocacy has been one-sided, and organize their own lobbying organizations to represent underrepresented causes.

The Post-Map-Ask system could also create opportunities for citizens to register their support for particular policy positions by voting them up or down, a system we could model after Reddit, the popular social networking news site. We could also simply create a space for citizens to register their preferences for or against a bill, or even vote for or against specific sub-sections, as POPVOX.com does.

More ambitiously, we could also allow a space where a select group of citizens could deliberate on legislation of specific proposals based on the materials provided, providing a kind of “citizens review” feature. Such a system would offer representative groups of citizens numerous opportunities to engage in the legislative process.

Certainly, frustration and anger at our dysfunctional democracy is reaching a boiling point. And admittedly, we are not advocating radical change or revolution as others are. Instead, we offer a simple and straightforward way to leverage technology to improve representative democracy, a system that could expand citizen participation, improve accountability, and limit the power of corporate lobbying. Best of all, it’s something we can start building right now.

You may also like

Advertisement