Bertrall Ross and Doug Spencer have posted this draft on SSRN (forthcoming, Cornell Law Review). Here is the abstract:

Campaigns’ increasing reliance on data-driven canvassing has coincided with a disquieting trend in American politics: a stark gap in voter turnout between the rich and poor. Turnout among the poor has remained low in modern elections despite legal changes that have dramatically decreased the cost of voting. In this Article, we present evidence that the combined availability of voter history data and modern microtargeting strategies have contributed to the rich-poor turnout gap. That is the case despite the promises of big data to lower the transaction costs of voter outreach, as well as additional reforms that have lowered the barriers to voting in other ways. Because the poor are less likely to have voted in prior elections, they are also less likely to appear in the mobilization models employed by data-savvy campaigns.

In this Article, we draw on a novel data set of voter data laws in every state and show that turnout rates among the poor are lower in states that disclose voter history data to campaigns. We also find that after states change their laws to provide voter history to campaigns, these campaigns are far
less likely to contact the poor.

The consequences of this vicious cycle are already known: the unique interests of the poor have been entirely unrepresented in the political process. Such political marginalization and alienation of an entire class from the democratic process is not only a problem for the poor; it poses a systemic threat to political moderation and democratic stability. Politically marginalized and alienated groups may resort to nonpolitical means to effectuate social change and may also become ripe for recruitment by extremist and anti-democratic elements that are latent in every society. Recent incidents of domestic political violence demonstrate that the United States is no exception.

To address this threat of marginalizing the poor from democratic politics, we advance three sets of proposals. First, we argue that states should regulate the information environment of political campaigns. Prohibiting the collection and distribution of voter history data is not practical, but states should lean into their privacy laws to prohibit the matching of voter files with other administrative data sets and should provide voter history data to campaigns independent of any information about individual political preferences. Second, states should create financial incentives for campaigns to expand their mobilization efforts to include a more representative target population that is more inclusive of the poor. Traditional campaign finance voucher and tax rebate programs are likely inadequate on their own. Instead, we propose a series of novel incentive programs that would provide cash grants to campaigns that report the most donors during each reporting period and to parties that generate more turnout than their historical average. Finally, we advance proposals for social media platforms to self-regulate “look-alike” targeting and segmented online political ads that amplify inequalities in mobilization and exacerbate political marginalization.

Political parties and individual campaigns in the United States are currently not mandated by law to promote political equality. The above reforms aim to align the short-term interests of parties and campaigns (winning the next election) with the long-term public interest in preserving a healthy democracy. Constructing a more inclusive political system will benefit everyone who seeks to live in a sustainable representative democracy, not just those who are currently marginalized.