DATE: July 30, 2018
TO: Boston Region Metropolitan Planning Organization’s Disparate Impact and Disproportionate Burden Working Group
FROM: Betsy Harvey, Transportation Equity Program Manager
RE: Summary of Third Working Group Meeting
This memo summarizes the third meeting of the Boston Region Metropolitan Planning Organization’s (MPO) Disparate Impact and Disproportionate Burden (DI/DB) Policy working group.
Date: July 17, 2018
Location: Northeastern Crossing, 1175 Tremont Street, Roxbury, MA 02120
Time: 5:30 PM–7:40 PM
The following stakeholders were in attendance:
The following MPO staff members were in attendance:
The following members of the public were in attendance:
The meeting began with introductions. B. Harvey then reviewed the goals of the meeting: for stakeholders to reach a consensus on a recommendation for the threshold that the MPO could use to identify disparate impacts and disproportionate burdens in the program of projects in the MPO’s LRTP. B. Harvey then gave a recap of the input received at the June 5, 2018, working group meeting and the June 26, 2018, public workshop about impacts related to the MPO’s transportation investments that could be assessed for disparate impacts and disproportionate burdens.
During the first 50 minutes, B. Harvey moderated a discussion among stakeholders about the benefits and drawbacks of various methods for identifying disparate impacts and disproportionate burdens in the LRTP program of projects. B. Harvey explained two overarching approaches that MPOs and public transit agencies use to identify disparate impacts and disproportionate burdens.
The stakeholders discussed the DI/DB policies of two transit agencies and two MPOs that represent examples of agencies that use those two approaches. TriMet, the transit service provider for the Portland, Oregon, metropolitan region and the Metropolitan Planning Commission (MTC), the MPO for the San Francisco metropolitan region, define disparate impacts and disproportionate burdens as those impacts that are statistically significant. The Massachusetts Bay Transportation Authority (MBTA) and the San Diego Association of Governments (SANDAG), the MPO for the San Diego metropolitan region, use a threshold to define disparate impacts and disproportionate burdens.
B. Harvey then led a discussion on the potential benefits and drawbacks of using the travel demand model’s margin of error as the method to identify disparate or disproportionate impacts. Potential benefits include limiting the possibility of disparate impacts and disproportionate burdens accumulating over time. Potential drawbacks include the possibility of the MPO committing resources to impacts that are less disparate, making it more difficult for the MPO to address its other goals.
J. Rowe mentioned that the MPO staff would screen out impacts that do not exceed the margin of error regardless of the threshold that the stakeholders recommend, so that the MPO could be confident that disparate impacts would only be identified for those impacts that would be projected to occur by 2040 if the projects in the LRTP are built.
B. Harvey then led a discussion about the potential benefits and drawbacks of defining an impact as disparate or disproportionate if it exceeds a given threshold. Potential benefits include allowing the MPO to focus resources on impacts that are most disparate, as well as addressing its other goals. Potential drawbacks include the possibility of disparate impacts and disproportionate burdens accumulating over time that do not exceed the threshold but that do exceed the margin of error.
T. Teich voiced concerns about setting a seemingly arbitrary threshold and the lack of data upon which to base a decision about choosing a particular threshold. M. Ebuña proposed that impacts that exceed the margin of error could trigger an analysis by the MPO and that the MPO could build in a graduated threshold over time. J. Gillooly said that, in the future, the MPO could apply the DI/DB policy to different potential programs of projects to identify those programs that would cause the fewest disparate impacts and disproportionate burdens.
B. Harvey then introduced a third option, which would identify disparate impacts and disproportionate burdens in a two-step process.
If the impact exceeds the margin of error and also exceeds the higher threshold, a potential disparate impact or disproportionate burden would be indicated.
If the impact exceeds the margin of error and does not exceed the threshold, the MPO would continue to track this impact through the development of the next LRTP. If the impact continues to exceed the margin of error but does not exceed the threshold for a certain period of time, this would constitute a disparate impact or disproportionate burden.
J. Rowe worked through an example of how a disparate impact would be identified using this approach. Stakeholders then discussed the potential benefits and drawbacks of this approach. Potential benefits include limiting the possibility of disparate impacts and disproportionate burdens accumulating over time, while allowing the MPO to focus on addressing impacts that are more disparate and also allowing the MPO to pursue its other goals. Potential drawbacks include the possibility that impacts that exceed the margin of error and do not exceed the threshold will take more time to address.
O. Richmond stated that Northeastern Crossing was challenging to get to as a wheelchair user because of construction. She asked what the MPO’s definition of “minority” was for the MPO’s DI/DB policy. B. Harvey replied that the definition is based on race and ethnicity. O. Richmond then asked whether the MPO studies the accessibility of particular building complexes. B. Harvey responded that the MPO does not undertake those kinds of analyses, rather the MPO is concerned about how the MPO’s transportation investments might affect accessibility. M. Ebuña asked how the requirements of the Americans with Disabilities Act (ADA) are incorporated into the MPO’s planning process. B. Harvey responded that when projects are considered for funding, they receive points if they serve people with disabilities. She also stated that the DI/DB policy would apply only to people who identify as minorities and to low-income households—which would include people with disabilities who fall into one or both of those groups.
R. O’Malley asked if the MPO would institute a policy that states that every project that the MPO funds must benefit underserved communities. B. Harvey replied that the DI/DB policy would analyze a collection of projects not individual projects. R. O’Malley then stated that he has not seen much action addressing the transportation concerns of environmental justice (EJ) communities in Malden. He suggested that every project that the MPO funds should prioritize the transportation needs of EJ communities over non-EJ communities because of the systematic underfunding of EJ communities in the past. B. Harvey replied that federal guidance directs MPOs to address only potential disparate impacts and disproportionate burdens that are projected to occur. The DI/DB policy is intended to be a check on the MPO’s planning process that ensures the MPO is appropriately programming projects that serve low-income and minority communities.
After the public comments, J. Rowe stated that there were two decisions for the stakeholders to reach a consensus on: which of the three approaches they would recommend, and what threshold they would like to set, if any.
J. Gillooly asked whether the MPO staff knows what the possible margin of error might be for different metrics. J. Li clarified that the same margin of error, expressed as a percent, would be applied to all of the metrics that are assessed for disparate impacts and disproportionate burdens. B. Harvey then said that in the third approach, the margin of error would be applied to all of the results, and the threshold would be applied only to those that then exceed the margin of error.
H. Perls and M. Ebuña expressed concerns that the stakeholders were being asked to develop a threshold absent of practical guidance that could be used to make the threshold less arbitrary. T. Teich agreed and said that setting a threshold might be a technical task and that the stakeholders would need information that differentiates between one threshold and another. S. Leary concurred.
J. Gillooly proposed organizing impacts that exceed the margin of error into tiers based on the degree of disparity. This could help the MPO prioritize and address the impacts that are projected to be the most disparate first, as well as attain a better understanding of the degree of the potential disparity for any of the impacts for which disparities are projected.
T. Teich said that setting a threshold still would not tell the MPO whether an impact is more disparate or less disparate for minority or low-income populations, as the threshold would depend on the starting numbers and the amount of change. Instead, she suggested that for every impact that exceeds the margin of error the MPO determine why there is a difference and how the MPO could address it. Even if there is the potential of identifying false positives, T. Teich said that it is still important to take a look at those impacts since the LRTP is completed just once every four years. L. Diggins also agreed that the MPO should consider every impact that exceeds the margin of error a disparate impact or disproportionate burden. As part of the MPO’s DI/DB policy, M. Foster suggested that the MPO define the margin of error that would be used.
J. Gillooly suggested that the MPO report on each of the impacts that are assessed for disparate impacts and disproportionate burdens, flagging the ones that are of the most concern. He and T. Teich both suggested that the MPO staff complete this exercise for several different collections of projects that could potentially be programmed in the LRTP, prior to the MPO’s adoption of the LRTP. B. Harvey clarified that this year the MPO must apply the DI/DB policy after the final program of projects is adopted, but that applying it prior to adoption could be an option for future LRTPs. T. Teich asked about what happens if the program of projects shows potential disparate impacts or disproportionate burdens. B. Harvey responded that it would be up to the MPO to decide how to address them. T. Teich suggested that the DI/DB policy applied to the 2019 LRTP be informative only, and that the MPO apply the DI/DB policy to scenarios prior to adoption of the 2023 LRTP and compare those results to the results of the analysis from the 2019 LRTP.
J. Rowe stated that it appeared that most stakeholders prefer the first approach, defining any impact as a disparate impact or disproportionate burden that exceeds the model’s margin of error. She also stated that there seemed to be agreement that MPO staff report on the projected impacts of all metrics assessed for disparate impacts and disproportionate burdens in the next LRTP, Destination 2040, to help inform the MPO’s decision-making about addressing and identifying disparate impacts and disproportionate burdens. She then asked the stakeholders to vote on whether they would recommend that approach to the MPO for the MPO’s DI/DB policy. All of the ten stakeholders voted to recommend that approach. The meeting adjourned at 7:40 PM.