Algorithmic violence and market-based neighborhood rankings
Download PDF
Mahmoudi, Dillon, Dena Aufseeser, and Alicia Sabatino. 2025. “Uneven development and the anti-politics machine: Algorithmic violence and market-based neighborhood rankings.” Political Geography 116 (January): 103247. doi.org/10.1016/j.polgeo.2024.103247.
Summary
In this article, we examine how algorithms like the Market Value Analysis (MVA) perpetuate racial capitalism, embedding the uneven geographies of wealth and poverty into urban planning under the guise of objectivity. Much like 1930s redlining maps, which codified racial and class hierarchies into spatial forms, the MVA ranks neighborhoods based on market metrics such as foreclosure rates and subsidized housing. By framing investment as a response to “market health,” these algorithms erase the historical processes of dispossession and segregation that produced these conditions, presenting disinvestment as a neutral outcome rather than a deeply political choice.
Algorithms are presented as objective instructions, and in our case, naturalize the processes of ranking neighborhoods. This framing sidesteps the continual political process of deliberation while embedding the power and bias of its creators.
We argue that algorithms act as an “anti-politics machine,” depoliticizing decisions about resource allocation while embedding the biases of their creators. In Baltimore, for example, the MVA prioritizes investment in affluent, predominantly white neighborhoods like those in the “White L” while deprioritizing services in majority Black, disinvested neighborhoods in the “Black Butterfly.” This framework locks in cycles of wealth preservation for certain areas while suppressing property values elsewhere, creating opportunities for speculative real estate profits. By presenting these processes as apolitical and data-driven, the MVA obscures its role in reproducing the systemic inequalities that shape urban space.
Ranking neighborhoods in tiers naturalizes the notion that developers’ investments are required to strengthen transitional neighborhoods, without questioning what caused these neighborhoods to become distressed in the first place.
We situate these dynamics within broader critiques of technocratic urban governance, where GIS tools and algorithmic typologies mask political intent with a veneer of objectivity. Just as HOLC maps grouped homes into racialized categories of risk, the MVA ranks neighborhoods into tiers, signaling developers where to invest while erasing questions about what caused disinvestment in the first place. These rankings do not simply reflect urban inequality—they actively produce and reinforce it, enabling the preservation of wealth in some areas while justifying the abandonment of others.
The MVA effectively compels governments to deprioritize the immediate material needs of vulnerable constituents and to approach communities as entities which must respond to the profit motive, reproducing and magnifying existing uneven geographies.
Our findings reveal that algorithms like the MVA are not neutral tools but critical mechanisms in the production of uneven development. They function as contemporary instruments of redlining, reproducing racialized and classed geographies of the past while presenting them as neutral, technical outcomes. We call for the dismantling of investment-oriented algorithms and a fundamental rethinking of the political forces shaping urban space, challenging the technocratic assumptions that underlie today’s uneven cities.