III: Recommendations

As Part I shows, intentional decisions were made to divide and harm communities of color. And as Part II shows, the overt racism of the early interstate age may no longer exist in the same form, but our standards and measures and models are failing to fix that damage, and often make it worse. This requires those in control to be intentional about undoing the damage and unwinding an approach that is still leading to the same, inequitable outcomes.

Leaders at the federal, state, and local levels have an opportunity and a responsibility to put an end to these injustices, but they will need to intentionally prioritize improving equity with every investment, and examine how the current system fails to do so.

To repair the damage of existing transportation investment and prevent future harm, first and foremost, impacted communities must be centered in the decision making around investment in their community and the vision for their future. We cannot truly rebuild the fabric of these communities without prioritizing those who have been marginalized or disenfranchised by past decisions. Some agencies and practitioners are moving from a public engagement to a co-creation model, which is exactly what is needed. But to ensure that the co-created vision is realized, there are a lot of barriers to knock down. The following recommendations focus on those barriers to improving access to jobs and essential services and ensuring that locals benefit from and are not displaced by investment.

To advance these goals, we propose four broad recommendations, with several specifics under each. (Click any sub-recommendation below to expand its detail, or refer to page 38-44 in the PDF to read them all.)


As Part II shows, data measures that focus exclusively on vehicle travel, particularly vehicle speed, come at the expense of people traveling outside of a car— often children, the mobility impaired, and those that can’t afford a car. This contributes to a system where almost all Americans must own a car to travel, which has acute economic and public health impacts for everyone, but particularly people in communities of color.


We know that there are a range of destinations people need regular access to, like grocery stores and daycare; but because federal and state decision-makers focus so exclusively on the work trip (as traveled by car), these other connections have been to a great extent ignored. Non-work trips are often shorter trips (the kinds that can be taken more easily outside of a car), and therefore some transportation agencies have found them hard to measure and others have even viewed them as less significant, even though they in fact make up the majority of trips taken.

While measuring access to everyday needs by all modes of travel was nearly impossible decades ago, it is no longer expensive or difficult thanks to GIS, the availability of transit and congestion data, and cloud computing. It can be used with measures like level of service or even in place of it. It is past time for transportation agencies to modernize and measure how well their transportation system gets people to the places they need to go by all travel options (beyond private vehicles) like walking, biking, or taking public transportation—modes of transportation Black and Brown Americans are statistically more likely to use.

If USDOT wishes to measure the value of time saved by transportation investments, they should update their guidance to connect it to this far superior approach of measuring access. This approach means measuring actual trips, not just segments of trips, which can more accurately be translated to time saved. Further, any value of time guidance should provide a way to place value on time savings for all trips, not just trips during peak times or work trips, and make clear how land-use changes can be considered as alternatives to transportation investments. Most importantly, there should be greater consideration for the time saved by people in more vulnerable communities, including those that do not have access to a car.

As shown in the report, when transportation investments are made, they often replace housing and businesses. While the benefit of faster travel through the region is quantified, the loss of housing, jobs, tax dollars and economic activity is not. Some economists will argue that this activity will just move around, but this is a dangerous concept, as we have seen. Smaller businesses do not have the capacity to pick up and move anywhere. And in the cases cited in Washington, DC and Atlanta, that economic value did move around—from Black to white communities.

If we are serious about fully evaluating potential investments, we will name and quantify all negative impacts and subtract them from any benefits. USDOT should provide guidance on how to do this, particularly in terms of safety and time savings benefits, and such an analysis should be required in all environmental and civil rights analyses.

As we wrote in Dangerous by Design 2022, the current approach to local reporting on crashes, guided by federal forms, is woefully incomplete and outdated. It often fails to record the race or ethnicity of victims, lumps skateboard users in with wheelchair users, and fails to gather usable data about street design factors—all of which veil the equity impacts of dangerous roadways. Data collection must change at the state and federal level as well. The only national dataset on traffic fatalities (the Fatality Analysis Reporting System, or FARS, which is made up of state-reported data), is currently being released ~15 months after the data is collected, making it difficult to get a clear picture of current road conditions. This is unacceptable and indicates that inequitable traffic deaths (and traffic deaths period) are not a priority. It’s time for USDOT to be transparent about what is causing this delay so their partners in Congress can help them address it.

Safety and access are often discussed in long-range plans, but then get discarded or downplayed when specific criteria are put in place to choose projects within four-year regional or state Transportation Improvement Plans. And they are not fundamental priorities in the design manuals in the same way that vehicle movement and throughput are. Good intentions do not matter if they are not reflected in what is actually chosen for funding and built.

Too often a statewide or regional view of transportation performance masks the real failures that fall disproportionately on marginalized communities, including low- income neighborhoods and communities of color. Looking at this difference will bring sharp focus on the moments when considerable investments are being made that bring marginal benefits to those with more means while creating huge, often life-threatening, burdens for those with far less. A system that works for the most vulnerable works better for everyone.


The first step to solving any problem is to stop making it worse. And yet, even as state and local decision makers make plans to utilize federal funding to reconnect communities, they also continue to plan costly highway expansion or construction projects that will further harm and divide communities. This must change. In addition, leaders and practitioners must making solving this problem a priority, using every opportunity at their disposal to enhance public health and safety, particularly for the most vulnerable road users.


State DOTs currently operate under the mistaken assumption that addressing congestion will solve every other problem. This approach continues to decimate and displace communities of color and create substantial monetary, health, and social costs while generally inducing more traffic. Yet the cycle continues as states build new highways and expand existing ones no matter the damage. It is not enough to reconnect communities (see next recommendation) if we are disconnecting them at the same time. To rectify past mistakes, decision makers must first stop repeating them.

Federal decision makers have provided a total of $4 billion over five years to repair this damage and reconnect communities (through the Reconnecting Communities Program in the Infrastructure Investment and Jobs Act and Neighborhood Access and Equity grants in the Inflation Reduction Act). These are valuable sources of dedicated funding that can help make a difference, but they are not enough. States are free to use their other, more plentiful federal funding sources to advance these projects. They do not need to wait for a specialized federal program to reconnect communities.

Small changes, like implementing Complete Streets on arterial roadways—the most dangerous roads in the country— can go a long way in knitting communities back together, serving local businesses, improving public health, and allowing all road users to safely access their needs.


State DOTs are often pressured to solve the wrong problem, and it will take leadership at every level to allow them to change course. As we said in Building a Better State DOT in 2019, state DOTs were tasked in the 1950s with the primary job of building highways to move vehicles quickly, a mission that made more sense when their primary job was to build a network of Interstate highways from scratch. But their mission never changed, even as the network was completed. And by applying that approach to the entire transportation system, on and off the highways, we have created a deadly, inconvenient system that is weighted by income, and Black and Brown communities suffer the worst impacts of this approach. It’s time for that to change and to give our agencies new goals, standards, and tools. USDOT should lead the way on all of the recommendations, but state DOTs that prioritize equity do not have to wait for USDOT.


There is a proven method for supporting safe, higher-speed vehicular traffic, one that we employ when designing interstates. We simplify the roadway by separating oncoming traffic, managing traffic entering and exiting the roadway, prohibiting development access (driveways and parking lots), and fully  separating vulnerable travelers (i.e., people walking, biking, and rolling). This gives a driver moving at a higher speed the chance to take in and process a reasonable amount of information, and react accordingly. This is how our interstates are designed.

But for roads or streets that have businesses and homes along them, driveways and parking lots, cross streets and crosswalks, street parking, and people walking, biking, and rolling, this complexity requires everyone to process an immense amount of information and react quickly.

Conflicts arise too quickly to respond at high speed and then drivers are blamed for not being able to do the impossible. Speeds need to come way down in these places to keep these roads and streets safe. A choice must be made between these two approaches on every road. Doing a little of both leads to roadways that are both dangerous and inconvenient. This choice should be clear in all of USDOT’s design and safety guidance.

Black and Brown communities are paying the heaviest price for design guidance and standards that prioritize the speed of private vehicles over the safety of everyone, even when that roadway goes through their communities. This must change. And USDOT needs to provide real leadership in making that change.

Specifically, USDOT should update its value of time guidance so that vehicle speed is no longer conflated with time savings benefits, especially when pedestrian travel time and safety suffers as a result. In addition, FHWA should update their guidance on traffic control devices—like signage, pedestrian crossings and bus-only lanes—as well as road and street design standards to support more pedestrian crossings and encourage slower vehicle speeds in areas with many conflict points and vulnerable road users.15

Many transportation agencies have a mindset that walking and biking are leisure activities, ignoring that these modes are statistically more likely to be necessary for everyday travel in Black and Brown communities as well as for low-income and younger people. As a result, our streets have become more dangerous for all, and traffic fatalities have reached all-time highs. The default approach should include protected sidewalks and paths and designs that slow down vehicles overall so that all road users, particularly the most vulnerable, are protected.

The problem is worse than safety being the third or fourth priority in many places. Some states are prohibiting cities and towns from taking steps to improve safety, as Indiana is considering for Indianapolis, or canceling projects that slow traffic, as Texas did in San Antonio. States should at least get out of the way of these local safety efforts. Preferably, they should be full partners and establish more modern standards themselves. But where state agencies are the obstacle, the Infrastructure Investment and Jobs Act allows cities to adopt and use safer street design guidelines approved by FHWA, even if their state has prohibited localities from doing so. Cities and towns should adopt and implement their own Complete Streets policies and adopt newer safe streets guidance (such as updated guidelines from NACTO or AASHTO), and they can improve equity and reconcile past wrongs by prioritizing investment in communities of color—where people are more likely to be struck and killed.



Things that get labeled as transportation problems are often land-use problems. Because of restrictive zoning, housing gets built on the fringes of a city or town far from jobs, groceries, banks, and other necessities. 1 This results in traffic snarls, makes walking and biking difficult, and drives up transportation costs, especially for lower-income households. The state or local DOT and transit agency is then called on to attempt the impossible task of fixing a problem created by land-use choices. Or new infill development is stopped for fear of traffic at the closest intersection or increased parking demand, considering the needs of only those with the money and ability to drive. Then such development is forced to the fringes where it will cause the problems listed in the previous example.

The transportation implications of these development decisions are predictable but rarely considered, while transportation agencies do not admit that they can’t fix a problem that is caused by the spatial mismatch of this development pattern. Both land-use and transportation agencies need access to tools that allow them to consider the impacts of both decisions on the other and determine which produces the most effective, efficient, and equitable results.


A huge problem with our national approach to transportation is that the models fail to consider the distance of a trip when estimating travel time, even though distance is required to know how much time travel will take. Land use agencies, on the other hand, either don’t know how to or aren’t directed to consider system-wide transportation challenges created by their land-use decisions. Both agencies should use access to everyday destinations (as discussed in recommendation #1 above) in their standards and codes in order to fully understand how transportation and development decisions will impact the other and what the final outcome will do for the public.

Specifically, this means that transportation agencies will review whether a transportation project can fix a travel problem as well as land-use solutions that might more effectively solve it. It also means that local land-use authorities would use multimodal access to destinations instead of localized level-of-service to consider the impact of proposed development decisions on all people in the community, not just the drivers in front of the facility.

The decisions about where to place schools, businesses, and retail centers are too often considered separately from transportation needs, costs, and impacts. For example, a new school building is placed on the outskirts of town on donated land to save the district money. Only when it opens do they realize their students without access to a car and parental chauffeur can’t reach them and they don’t have money for bus service. Transportation agencies (and appropriators) are charged with trying to solve this land-use problem with a limited suite of incredibly costly transportation solutions, like new roads or wider lanes, when the fundamental problem is one of location. These problems can be prevented by considering how well everyone, particularly low-income and communities of color, can access the facility by all modes of travel, during the process of evaluating potential land-use decisions.

Too often we save money on housing by placing it on less expensive land far from the things people need. As a result, these households, who are disproportionately Black and Brown, are saddled with high transportation costs, especially as the average cost of a new vehicle has skyrocketed to almost $50,000. These are costs that are more subject to unexpected spikes, like in 2008, that people cannot plan for or absorb. People must then choose between spending more or walking long distances on roads not built for them, making vulnerable communities more vulnerable. That is too high a price for affordability and should not be labeled as such. The Center for Neighborhood Technology has created a wonderful tool for calculating housing plus transportation costs, and housing agencies at all levels should use it or create a similar methodology for the implementation of their programs.

Pushing development to the fringe because of traffic instead only makes trips longer and requires more driving overall. Requirements like parking minimums, building setbacks, height restrictions, and strict separation of each type of use (e.g., housing separate from retail which is separate from schools and jobs) leads to longer trips, more driving, higher transportation costs, and mandatory auto-dependent lifestyles. These agencies should prioritize forms of development that make it possible for more people to live closer to their everyday needs and support the development of infill and missing middle housing, reducing both transportation and housing costs for more people, but especially within the communities that have been so often left out of transportation, housing, and land use decision-making to date.

Read the rest of Divided by Design

Explore the report’s full content using the sidebar menu at right (or below if on a mobile device), or jump to one of the three parts with the graphics below.

report cover graphic showing a stylized highway cutting through a city.graphic showing a stylized scene of construction of a highway through a city neighborhoodgraphic showing a stylized scene a few blocks away from a highway running through a city neighborhoodgraphic showing a stylized scene of what a neighborhood could look like after tearing a highway down

Don’t miss supplemental maps, videos, and animations in the DC and Atlanta case studies which are not in the hard copy. Download a PDF version of the report.