The Technology of Corporate Accountability
Last week a Chilean commission of environment, health, economy, mining and agriculture ministers voted to approve a massive coal mining project on Riesco Island in the far south of the country. Supporters of the project, which will construct five mines and a shipping port, say it will create needed jobs and reduce Chile’s dependence on foreign oil. Opponents warn of a 360% increase in Chile’s carbon emissions and environmental harm to one of the world’s great nature reserves.
Greenpeace Chile activists protesting the proposed coal plant on Riesco Island
Yet for good government proponents there is another, major detail in the story: one of the largest shareholders behind the project is none other than President Sebastián Piñera. When the commission’s decision went public, the company’s stock price rose immediately, earning President Piñera US$ 5 million in a single day of trading.
The Riesco Island coal mining project is only the latest of the many conflicts of interest that define the Piñera presidency and the entire Chilean political class. The following presentation by Miguel Paz, associate director of El Mostrador, reveals the extreme consolidation of wealth and power in just a few Chilean families. In fact, the 15 wealthiest Chilean families account for nearly 20% of the entire country’s GDP, making Chile one of the most unequal countries in the world despite its skyscrapers and modern infrastructure.
Visit just about any Latin American country and you’ll soon hear about the five or ten families that have always been in charge. In Guatemala, journalist Juan Luis Font calls them “the ten families,” but admits that there are in fact many more. In Chile just four families control 47% of the assets on the Santiago Stock Exchange. Nicaraguan economist Francisco Mayorga claims that his country is virtually governed by just 12 families. In Mexico the 39 wealthiest families make up 13.5% of the country’s substantial GDP.
Transparency and good governance advocates in Latin America are less concerned by the consolidation of wealth as its frequently illicit influence on regulation and public policy. In response, a new generation of tech-savvy transparency activists are now building Internet platforms to reveal the often hidden connections between powerful individuals and the political institutions that should represent all citizens.
Forced Transparency: Corporate Image on Wikipedia
In hindsight, the first corporate transparency platform might have been none other than Wikipedia, which has proven a challenge to public relations firms that try to downplay exposed corruption by spending millions on marketing campaigns that paint corporations in the most flattering light possible.
Last year two researchers from Public Relations Society of America published the results of a four-year study which tracked the Wikipedia pages of 10 Fortune 500 companies (Wal-Mart, Exxon Mobil, General Motors, Ford, General Electric, Chevron, ConocoPhillips, Citigroup, AIG, and IBM). The researchers found that over the four years the tone of the pages became more critical, and the content focused more aggressively on related corruption scandals. It is worth emphasizing that, for each of the ten major corporations, the associated Wikipedia page is always among the top three search results on Google.
The unstated advice of the Public Relations Society of America to Fortune 500 corporations was clear: keep an eye on your Wikipedia pages and employ contributors to ensure that the page reflects each corporation’s best interests.
Crowdsourcing Corporate Accountability
The initial excitement around Wikipedia and crowdsourcing inspired some corporate accountability organizations to experiment with wiki software and volunteer contributions. In 2007 the Center for Corporate Policy, the Corporate Research Project, and CorpWatch came together to launch Crocodyl.org. The aim of the website, in its own words, is to:
stimulate collaborative research among NGOs, journalists, activists, whistleblowers and academics from both the global South and North in order to develop publicly-available profiles of the world’s most powerful corporations. The result is an evolving compendium of critical research, posted to the public domain as an aid to anyone working to hold corporations increasingly accountable.
The website has dozens of detailed profiles on major corporations, categorized by issue and industry. There is also a research library of related investigative journalism. Searching through its voluminous content, we discover, for example, that the India-based Jindal Steel & Power Ltd won a bid to develop an iron ore reserve in Bolivia, but in its own country the mining corporation was accused of using so much water in its industrial operations that it left several small towns without drinking water.
Unfortunately, the website has only been updated a few dozen times in the past two years — and mostly by the same small handful of contributors. Several of the articles, such as the profile for Wells Fargo, are filled with spam links, which casts doubt on the vetting process and validity of the content.
Another endeavor into crowdsourced corporate accountability is SourceWatch, published by the Center for Media and Democracy. The wiki-based website — also inspired by Wikipedia — was created in January 2003 and has accumulated over 55,000 articles about “the people, companies, and entities attempting to shape public opinion.” Unlike the now defunct Crocodyl.org, a glance at SourceWatch’s “recent changes” page reveals the the site is still very much active today, though the majority of content still comes from a handful of contributors. Whereas Crocodyl.org aimed to reveal corporate malfeasance, SourceWatch is more focused on public relations, government lobbying, and the manipulation of public opinion.
A look at the SourceWatch page for Altria Group, formerly Philip Morris, offers a sample of the website’s tone and format. We learn that in 2010 Altria Group spent nearly $2 million on political contributions to federal candidates. The same year they spent $10.4 million on lobbying. A link to OpenSecrets reveals that Altria Group filed ten reports advocating against the Family Smoking Prevention and Tobacco Control Act and 17 reports against the health care reform bill. The article also suggests that in 2005 a Philip Morris employee deleted a statement on the Wikipedia page for Marlboro cigarettes (the deleted information was not detected and reinstated for two and a half years). Most of the information is focused on the United States, but we also learn how Philip Morris circumvented a total ban on cigarette advertising in Saudi Arabia.
From Crowdsourcing to Structured Data
As true in other sectors, the initial excitement around crowdsourcing among corporate accountability advocates soon subsided. A former contributor to Crocodyl.org wrote to me over email, “we didn’t have the thousands of independent contributors, Wikipedia-style, contributing their knowledge.” Instead they depended on a small team of part-time, paid researchers who did their best to come up with a standard format and keep up with the accelerating avalanche of information related to corporate accountability.
Soon the Silicon Valley buzz transitioned from crowdsourcing to open data. Corporate accountability organizations followed suit.
Crocodyle.org fell dormant, but CorpWatch soon partnered with the Sunlight Foundation to develop CrocTail, “an interface for browsing information about several hundred thousand U.S. publicly traded corporations and their foreign subsidiaries.” Rather than depending on volunteer contributors, CrocTail automatically takes information from the U.S. Securities and Exchange Commission and re-structures it so that users can see a map of subsidies of all US publicly traded corporations. Again, if we use Altria Group as an example, we see that it has 44 subsidies — mostly in the United States, but also in Canada and Brazil.
Just as the United States Security and Exchange Commission makes some basic information about corporations available via its (very user unfriendly) EDGAR platform, in the United Kingdom corporate information is available to all citizens through a website called Companies House. Information about all UK-registered corporations is available to the public, but the website charges £1 per document, which is delivered in PDF format, a nightmare for researchers engaged in statistical analysis. At one point, writes James Ball for the Guardian’s Data Blog, the Companies House website even had operating hours, as if it were a brick and mortar office. That soon changed when Rob McKinnon, who single-handedly launched TheyWorkForYou.co.nz, scraped all of the data and made it available (24 hours a day) at “Companies Open House,” which later evolved into Open Corporates. Today Open Corporates, which dubs itself “the open database of the corporate world,” has information on more than 21 million companies in nearly 30 jurisdictions including India, Gibraltar, Malta, and Panama. A few months ago Open Corporate was joined by a new player in the field, Level Business, which has made 2.4m UK company accounts open to everyone, for free, according to the Guardian. Level Business is actively hiring and expanding. Their business model: “make the raw companies data available for free while selling services that make use of the information.”
“Corporations Are People, My Friend”
That now infamous remark by US presidential hopeful Mitt Romney holds some truth when put in the context of how we understand corporate influence on public policy and regulatory enforcement. The open data platforms we have observed so far — CrocTail, OpenCorporates, and Level Business — are all company-centric. They show profiles of corporations and their subsidies, but provide little information about the individuals behind the corporations and their relationships with politicians and political institutions.
Such databases of companies, individuals, and investors have long existed in the private sector — for example, CrunchBase and Blackboard — but it wasn’t until 2009 that the Public Accountability Initiative launched LittleSis, “a free database detailing the connections between powerful people and organizations.” In their own words:
We bring transparency to influential social networks by tracking the key relationships of politicians, business leaders, lobbyists, financiers, and their affiliated institutions. We help answer questions such as:
- Who do the wealthiest Americans donate their money to?
- Where did White House officials work before they were appointed?
- Which lobbyists are married to politicians, and who do they lobby for?
All of this information is public, but scattered. We bring it together in one place. Our data derives from government filings, news articles, and other reputable sources. Some data sets are updated automatically; the rest is filled in by our user community.
The site currently offers profiles of 61,308 people, 18,846 organizations, and 325,177 relationships between them. It takes a hybrid approach; scraping data from government websites, but also depending on crowdsourced contributions based on journalistic citations. According to the website, “making edits on LittleSis is more like adding friends on Facebook than modifying a Wikipedia page.” Users who make inaccurate or malicious edits can lose editing privileges. All edits are currently monitored by paid staff.
Another relationship-mapping platform is Influence Networks, a collaboration between OWNI, ObsWeb, Transparency International and Zeit Online. The open source platform allows journalists and volunteer contributors to add individuals and organizations to the database and then suggest relationships between them that are then vetted by a community of reviewers. The platform was recently featured on the popular Spanish-language blog ALT1040, which pointed to Spanish Culture Minister Ángeles González-Sinde as a clear example of conflict of interest between industry and government. A filmmaker who was formerly the president of the Spanish Academy of Arts and Cinematographic Sciences, González-Sinde has deep ties in the Spanish film industry, which she clearly represented in her proposal to restrict online file sharing. She is also an active member of the powerful and controversial Society for Authors and Editors, the country’s main collecting society.
However, when we examine González-Sinde’s relationships on Influence Networks, all that is revealed is that she is a member of the Society for Authors and Editors. Even her current role of Minister of Culture is omitted. Influence Networks claims to have mapped 1,158 relationships, but without a large influx of meaningful content and context, the site’s automated visualizations of networks leave us with little useful information.
Finally, it is worth noting Sunlight Foundation’s Influence Explorer, a collection of tools and data to help users better understand the relationships between people, organizations, politicians, and industries. For example, the Checking Influence tool gives you detailed information about the lobbying activities and campaign contributions of the companies listed on your bank statement. Two other notable platforms are Subsidy Tracker by Good Jobs First and Follow the Money by the National Institute on Money in State Politics.
Technology and Corporate Accountability in Latin America
So far the use of technology in corporate accountability has been mostly limited to the United States and the European Union, where information about corporations has been made public by respective governments (albeit often in convoluted and poorly formatted ways).
Yet, as emphasized at the beginning of this post, Latin American countries suffer from much more egregious acts of illicit influence by corporations. Journalist Denise Dresser calls Mexico a classic example of Crony Capitalism, or what she dubs “capitalismo de cuates.” She lists countless examples, such as the country’s infamous lack of competition in television and telecommunications. Earlier this year it seemed that Mexican antitrust regulators were finally going to impose a record US$ 1 billion fine on Carlos Slim’s telecommunication monopoly, America Movil, but effective lobbying by Slim’s lawyers has called the judgement into question. As Mexican media shift their attention campaign season, it seems that Slim will likely once again out-maneuver government regulators.
Similarly, Yul Francisco Dorado Mazorra of Corporate Accountability International has described how the global tobacco industry has managed to influence legislation and treaty enforcement in Peru, Guatemala, and Ecuador.
The cases go on and on, but there are few organizations working to document them in any meaningful way. Two of them happen to be based in Chile where legislators are required to submit asset declarations before taking office. Chilean law requires that legislators refrain from voting on bills in which they (or close family members) have conflicts of interest or stand to make personal economic gains, but there is no clearly defined enforcement procedure.
As such, the Smart Citizen Foundation (Fundación Ciudadano Inteligente) is currently developing a web application called “Interests Tracker” that is designed to integrate with Vota Inteligente, their congressional monitoring platform. The Interests Tracker app will take information from the asset declarations of legislators; cross-reference it with information from the commerce registry, land property registry, and public tax returns; and build a database of relationships between politicians, companies, organizations, and industries. Whenever a related bill is discussed in Chile’s Congress the web app will notify Internet users and journalists of potential conflicts of interest that can then be further investigated.
Also based in Chile is Poderopedia, which was cited at the beginning of this piece. Earlier this year they received a $250,000 grant from the Knight Foundation to build an open source platform similar to LittleSis, but focused on Chile and geared toward investigative collaborations with professional journalists. Co-founder Miguel Paz speaks of the project as an important piece of the context-based news movement. Too often, says Paz, investigative journalists duplicate the work of colleagues by repeatedly investigating the relationships between influential actors. Only a small amount of that research makes it into the final story, however, and rarely do journalists share information with others. Paz hopes that Poderopedia becomes a common, shared resource for journalists and others.
Finally, the Mexico-based Project on Organizing, Development, Education, and Research (PODER) is currently seeking a programmer to develop a “Who’s Who Wiki.” According to the project description, the “platform will consist of four components: a database containing information on corporate elites and their companies in Latin America, a network analysis and visualization tool illuminating the corporate social network, a wiki site to coalesce and empower diverse corporate stakeholders, including investors and civil society organizations, and a secure whistle blowing mechanism to inform on corporate malfeasance.”
Corporate Leaking: There’s no turning back
Micah Sifry has eloquently articulated why there is no turning back from the new normal of leaking confidential information. Most analysis and hyperbole around Wikileaks has focused on its role in making accessible leaked documents from governments. But, in fact, much – perhaps most – information in the Wikileaks archive relates to corporate malfeasance. Rumors of forthcoming Wikileaks exposés on major corporations have even inspired a number of columnists — such as Fast Company’s E.E. Boyd — to offer corporations free advice to avoid internal leaking.
The case of corporate malfeasance by Enron demonstrates the importance of leaking and whistleblowing by employees like Sherron Watkins who risk their jobs to reveal the truth. For example, former Wal-Mart executive James Lynn was fired when he revealed labor law violations in Costa Rica (Lynn later lost a lawsuit against Wal-Mart because, according to the court’s ruling, Wal-Mart’s actions did not violate Arkansas state law).
In another case of corporate corruption reported to Wikileaks, a former employee of EMC Corp was allegedly fired when she reported that EMC’s Mexico City office hired the spouse of a government official responsible for approving contracts at the Mexican Institute of Social Security (IMSS), and frequently invited both to dine.
Perhaps most notoriously, Wachovia employee Martin Woods was intimidated and harassed when he discovered — and later exposed — that his employer was helping launder billions of dollars for Mexican drug cartels.
It is worth emphasizing that in all three cases, corporate misconduct in Latin America was reported by employees based outside of the region. A culture of whistleblowing and internal corporate accountability still has yet to take hold in Latin America. There is a strong culture of corporate loyalty and yielding to authorities. Still, the increasing popularity of Wikileaks will likely attract a new generation of tech-savvy, idealistic leakers in the near future.
Small (Puzzle) Pieces Loosely Joined
A broad survey of the use of technology in corporate accountability reveals that most of the necessary pieces of the puzzle have been developed, or are in active development, but they do not effectively share data and link to each other’s contributions.
In 2002 David Weinberger wrote an (at the time) influential book called Small PIeces Loosely Joined, which offered a vision for the so-called Web 2.0 ecosystem. Web applications would offer small, highly focused services that share data with one another. Twitter, for example, didn’t need to offer its users photo or video uploads because dozens of other companies could do it better, and tap into Twitter’s user base which ultimately benefitted both companies.
Within the realm of corporate accountability, however, most online projects still try to be everything to everyone rather than focusing on particular niche analysis that adds specific value to the overall corporate transparency ecosystem. For example, why doesn’t LittleSis link to CrocTail’s subsidy information to automatically generate relationships between corporations and their subsidies? Why is PODER developing its own leaking platform when they can act as a local conduit and analysis for WikiLeaks, OpenLeaks, Cryptome, WSJ SafeHouse, and the many other leaking platforms already in operation? Why is Influence Networks trying to build its own database of people, institutions, and companies rather than offering its visualization tool to LittleSis, which has already documented 325,179 relationships (compared to Influence Networks’ humble 1,158 relationships)? Why don’t OpenCorporates profiles link to related information from LittleSis, SourceWatch, and Crocodyle.org? One can even imagine a collaboration with Sourcemap so that users can better understand the related corporate accountability issues behind individual products — such as an Ikea bed — and not just corporations. Increased sharing of information will increase the utility of all the websites surveyed above.
Beth Noveck, formerly the United States Deputy Chief Technology Officer (2009-2011), has returned to New York Law School and the Do Tank, a “democracy design workshop.” One of the major projects of Do Tank is ORGPedia, which aims to provide unique identifiers for all legal entities in the United States in order to facilitate “the ‘mashing up’ of disparate data sets about the ownership, structure, performance and regulatory compliance of organizations.”
Noveck explains how it is nearly impossible to compare information for a single company across different government agencies: “Currently, one federal agency might refer to a company as ABC Inc. while another uses ABC Corp.” Her vision is to “help solve this problem by mandating open, universal identifiers here rather than exacerbate it by creating yet another IT system with yet another set of disparate naming conventions.”
The Alfred P. Sloan Foundation — which funded the Encylopedia of Life, a collection of individual web pages for every known biological species — has provided Noveck with initial funding to attempt the same ambitious project for corporations and organizations, though they tend to evolve at a far faster pace than their biological counterparts. You can read more about future plans for ORGPedia at Noveck’s personal blog, and explore an experimental demo site that pulls in corporate data from various federal agencies
Even if Noveck and her colleagues do eventually achieve the world’s most complete encyclopedia of organizations and corporations, it won’t necessarily increase corporate transparency and accountability unless there is meaningful collaboration with investigative journalists and civil society organizations. In many cases, litigation is the only mechanism to resolve corporate malfeasance. ORGPedia could take a tactic from the collaboration between Congreso Visible, a parliamentary monitoring group, and the El Espectador newspaper in Colombia. Each time a congressional representative is mentioned in an article on the El Espectador website, a pop-up balloon shows related information from the Congreso Visible website.
In the end, Wikipedia will likely remain the world’s most read resource on corporate information, providing a challenge to both the corporations themselves and the activists that frequently aim to tar their reputations. Websites like LittleSis, ORGPedia, and OpenCorporates can form partnerships with Wikipedia to automate the cross-publication of relevant information — as well as its contextualization.
Now, more than ever before, corporations have lost their ability to control the narratives around their activities. In fact, so much information is currently available, the challenge now is to aggregate and make sense of it.