Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 214
Direct and indirect effects. In the rest of this chapter, we refer to direct and indirect economic
impacts of highway projects. Direct effects impact businesses immediately within the project
area, while indirect effects are experienced by businesses outside of the project area. For
example, if a store is adjacent to a widened highway, a direct effect of the project could be
increased business due to greater vehicle traffic past the store location. If that increased business
represents a shift of business from a competitor a few miles away, the loss in business at the
competing store would be an indirect effect.
During and after construction. The distributive impacts of highway projects can be divided
into two phases--the construction disruption and the postconstruction period. During
construction, access to businesses in the vicinity of the project can be disrupted. Sometimes
businesses fear that they will lose customers because of poor access, loss of parking spaces,
noise, or other disamenities associated with nearby road or highway construction. The disruption
to business activity can be temporary, if the lost customers return after construction is complete,
or permanent, if customers form loyalties with substitute stores or suppliers during the
construction period. Importantly, there is some disagreement about how much businesses are
harmed during a road construction project. Often, transportation agencies take measures to
minimize or mitigate the effects of the loss of access, parking, or noise. Also, the negative effects
on nearby firms will be influenced by the loyalty of their customers and the nature of their
competition. Hence different firms in a construction zone could be affected differently by the
project.
After construction, a completed road or highway project can change the pattern of accessibility
and thus influence the competitiveness of different businesses. One might imagine that the
spatial influence of a road project after construction will be roughly the reverse of the impact
during construction. The businesses near the road improvement will benefit from improved
access, and might be able to lure customers from more distant competitors. Yet while this
example helps illustrate the different distributive impacts of the construction and post-
construction time periods, there are many reasons why the impacts during and after construction
will not be mirror images. Environmental justice analysis should consider the spatial pattern of
the positive and negative impacts of highway projects before and after construction, and how
those patterns affect minority and low-income communities.
METHODS
Table 9-1 summarizes the methods we present in this chapter.
Method 1. Map and GIS assessment
The simplest method for assessing the environmental justice impacts of highways related to
economic development is to map businesses around the project. This could be as straightforward
as walking or driving the corridor (i.e., a so-called "windshield survey") to assess which
businesses might be affected by a construction project. More involved analyses can be carried
out by looking at business locations that are depicted on a map. Typically, such a map would be
developed using geographic information systems (GIS). A similar technique could be used for
219
OCR for page 215
the after-construction period to identify which firms might benefit from improved access and
which might be negatively affected by the loss of business to those firms.
Table 9-1.
Summary of methods for analyzing economic development
Assessment Appropriate Use Data Expertise
Method level uses when needs required
1. Map and GIS Screening Project/corridor Effects are expected to Low Geographic
assessment be low and more information
resource-intensive system
techniques are not (GIS)
available
2. Surveys or Screening/ Project/corridor/ Project or policy is Medium Group
focus groups detailed system controversial and high interaction
level of interaction with and
affected individuals is facilitation
required
3. Gravity Detailed Project Changes are expected High Accessibility
models in accessibility in many modeling
directions and over a
wide area
When to use. This technique is among the least resource intensive and simplest to apply. It is
appropriate either where more resource-intensive techniques are not available to an agency or for
projects with relatively small impacts for which more rigorous (and hence expensive) analyses
would not be commensurate with the scope of the project.
Analysis. For each business identified in the project, you can assess whether the impact of
construction will result in a loss of customers using rules of thumb (e.g., reasonable walking or
driving distances) or expert judgment (see the description of focus groups in Method 2). This
direct effect cannot be cleanly separated from indirect effects because to some extent the loss of
customers will depend on available substitute business locations in the same market area. Hence,
for each business in the construction zone, you may also want to use data in a GIS format to
assess possible competing businesses. This could be a simple listing of such firms, or one could
assess the relative characteristics of the competing businesses and the businesses in the
construction zone to assess the likelihood of shifts in a customer base.
Data needs, assumptions, and limitations. The key element of this method is to document
firms that are likely to be affected either during or after construction because of their proximity
to the project. These firms will experience the direct effects of the project. For indirect effects,
one could catalogue firms in surrounding market areas. The area of interest might be established
based on rules of thumb, as noted earlier, or on more detailed assessments of market areas and
the locations for nearby competitors of particular firms near the project.
220
OCR for page 216
Results and their presentation. One example of this technique would be to walk or drive the
project corridor and catalogue each business by location, type of business, and assessment of the
expected impact during and after construction. A hypothetical example of the survey results is
shown in Table 9-2.
Table 9-2.
Example of local access assessment
Expected impact during Expected impact after
Location Business construction project completion
22 Main Gas station One entrance lane obstructed No change in access
26 Main Grocery Loss of 20 parking spaces New right turn lane into parking lot
30 Main Hair salon Entrance lane narrowed Five new parking spaces
34 Main Drugstore Increase in traffic in opposing lane New dedicated left turn lane into
at parking lot entrance parking lot
The same type of analysis could be used to understand impacts after construction. For each
business in an area of improved accessibility, you can assess whether that business will likely
draw business away from competitors. This can again be based on rules of thumb or expert
judgment.
Assessment. This method is the simplest and least resource intensive of those described. As
such, it is appropriate in several circumstances. Agencies with limited capabilities can conduct a
GIS-based analysis quite easily. This method is also suited to small projects and particularly to
those where the agency expects few or no environmental justice issues. Agencies can use this
technique as a screening method to verify hunches that projects will have limited environmental
justice implications or to illuminate possible issues that will require further analysis. The
disadvantage of a GIS-based assessment is that the results can be subjective if, for example,
agencies do not carefully articulate the criteria that led to particular conclusions regarding
choosing competing businesses or project impact areas. Agencies should take care to apply
systematic criteria when making such judgments so that the outcome of the assessment can be
clearly linked to the assumptions and methods used.
Method 2. Surveys or focus groups
Understanding attitudes and reactions of parties affected by transportation projects is important
in assessing environmental justice. Surveys or focus groups are a useful tool to acquire
information about the characteristics of the affected parties, as well as their expectations
regarding the project. These can include expected business losses due to construction disruption,
expected benefits due to improved accessibility after the project is complete, and perceived
changes in competitiveness due to the altered pattern of accessibility. The quality and
appropriateness of surveys and focus groups can vary in different contexts. Occasionally,
analysts have been suspicious of these methods, fearing that they allow users to overstate
negative impacts. Yet agencies have sometimes under-appreciated the value of surveys or focus
221
OCR for page 217
groups both as a means of outreach and as an analytical tool. In some instances, the perceptions
of persons and businesses affected by a project might be more accurate than analytical methods,
and surveys and focus groups are an excellent way to get information that otherwise might not be
available.
When to use. Survey or focus groups are especially useful (1) when input from affected parties
is vital for understanding the distributive character of economic impacts, (2) when impacts are
confined to small geographic areas and analytical techniques based on data for larger geographic
regions will not capture the spatial character of the impacts, (3) when data on impacts are not
readily available from other means, and (4) when agencies will benefit from direct interaction
with the public, as in the case of controversial projects where perceptions are important or when
an agency wishes to facilitate communication with parties affected by the project.
Analysis. Surveys can be used to capture the general attitudes of a wide range of parties. The
results can be analyzed to match groups that benefit from, or are disadvantaged by, transportation
projects to geographic locations. Survey sample sizes should be large enough that summaries by
geographic areas, income, or race will have sufficient within-category sample sizes. That will
often require over-sampling within specific geographic areas, income groups, or minority or
ethnic groups.
Data needs, assumptions, and limitations. Surveys can be designed to elicit two kinds of
information. The first includes general characteristics of the parties in question, such as type of
business, number of employees, logistical and other inventory arrangements, and mode of travel
of a firm's suppliers, clientele, and employees. The second includes the expectations of the
affected parties regarding losses and gains during and after construction of the project. Firms can
be asked to estimate how much business will be lost due to construction disruptions, such as
closed lanes, noise, and lack of parking space. The gains are mainly in the form of expected
increases in business volume once the project is completed. After completion the project may
also adversely affect some firms because of shifts in economic activity to locations better served
by the new transportation infrastructure. Hence, some firms might be asked to estimate these
negative impacts.
Surveys are typically conducted either by mail or phone. Various books give detailed advice on
how to conduct a survey using either method. One popular reference book is Mail and Internet
Surveys: The Tailored Design Method (Dillman 1999). These books offer suggestions on matters
such as phrasing questions in a neutral manner, techniques to increase response rates, and follow-
up methods for subjects who do not initially respond to the survey. This literature is useful
whether an agency develops its own survey or contracts with an outside firm to create it. There
are many firms that provide survey research services. Some have a national client base, while
others specialize in particular regions or metropolitan areas.
Internet-based surveys have recently been used in some settings and are also described in
Dillman (1999). A common risk in Internet-based surveys, however, is that the sample of
respondents will not be representative of an underlying population of interest. This can occur
both because persons or firms with Internet access do not represent a random sample of all
persons or firms and because those parties who are particularly interested in a topic are the most
222
OCR for page 218
likely to respond to a survey. While response bias is a potential issue in any survey, it is often
regarded as more problematic in settings, like some crude Internet-based surveys, where
nonrespondents are not contacted to prompt a response. For environmental justice studies, the
fact that low-income populations might not have good Internet access could limit the scope for
Internet-based surveys. For this reason, we cannot recommend them.
Note that, when assessing the economic impacts of transportation projects, the survey population
is likely to be firms. This raises additional complications beyond those in household surveys.
Before the survey is sent, the appropriate officer within the firm must be identified. This should
be an individual who has the needed data readily available. For surveys about economic impacts,
this individual will likely be someone high in the management structure of the firm. Thus a pre-
survey contact to explain the importance of the study is important. Response rates will increase if
firms clearly understand how the survey study will benefit them. Even with the best efforts,
response rates for surveys of firms are often lower than response rates for surveys of households.
For firms, response rates of 20 to 30 percent are not unusual. See, for example, the discussion in
Boarnet (1998b) or Kalafatis and Tsogas (1994).
Focus groups provide opportunities for open-ended responses and discussions that are typically
not possible in surveys. The groups are usually small--often not more than six or eight persons.
They are thus typically used when the need for detailed information outweighs the need for
statistical analysis. Focus groups are desirable when agencies are in the exploratory phase; often
the information gathered can be used in later research. For example, it is common to use focus
groups in the preliminary phase when designing a survey.
Results and their presentation. The central purpose of surveys and focus groups is to acquire a
clear understanding of the general attitudes, concerns, and preferences of minority populations
and low-income populations regarding a proposed transportation project. Results of these
methods of interacting with protected populations can be used to help assess whether the project
would have a generally positive or negative effect on the economic well-being of these
populations. The results also can be used to identify changes in the project or measures that
could be taken to mitigate undesirable effects.
Often, the analysis of survey data involves summarizing responses for the population. For
environmental justice analyses, survey responses will typically be summarized by geographic
area, income, or race. This requires that data be collected on locations of survey respondents
(firms or households) and on income or race for individuals or for firm owners, employees, and
clientele.
Assessment. Surveys and focus groups provide detailed information due to the first-hand
knowledge of the persons surveyed or interviewed. Hence, these techniques can improve the
understanding of the distributive impacts of transportation projects, and the results can be used to
design mitigative measures. One must be careful when using survey data, however. Survey
respondents who oppose a project, for example, may exaggerate its negative impacts. In addition,
the gains and losses predicted by respondents will be based on their perceptions, which may be
inaccurate. Care should thus be taken when interpreting the results. Surveys are most reliable
223
OCR for page 219
when respondents are better able to judge impacts than any other group and when they have little
incentive to exaggerate or misstate impacts.
Method 3. Gravity models
In addition to being used as a trip distribution model in the traditional four-step transportation
planning process, gravity models are also widely used for market analysis. These models can be
used to determine such things as the spatial extent of retail market areas, optimal store or public
facility sizes, and optimal store or facility locations. The ability of gravity models to give
estimates of the size of market areas is of particular interest when assessing the economic
impacts of highway projects. Gravity models can be used to understand how changes in
accessibility will change market areas, and hence sales, at particular store locations or in
particular geographic areas. They can give both qualitative assessments of locations that will
likely experience changes in business activity after a transportation improvement and
quantitative estimates of those changes. Good references on gravity models for this sort of
application include Hayes and Fotheringham (1984), Bendavid-Val (1991), and Filipovitch
(1996).
Gravity models have two basic elements, scale and distance, which are the determinants of the
interaction between any pair of geographical areas. For example, densely populated cities (large
scale) tend to generate and attract more trips than sparsely populated cities. Moreover, trips are
more likely to occur between cities that are located closer together (short distance) than between
distant cities. Gravity models can be adapted to assess environmental justice impacts of
highways by analyzing how changes in accessibility (distance impacts) affect the relative
attractiveness of communities or neighborhoods.
When to use. Gravity models are most appropriate for a highway project that creates significant
changes of accessibility in many directions and over a wide area. They can help answer questions
about how a proposed transportation project would affect the ability of businesses operated by or
serving minority populations to be competitive. Agencies with an operational transportation
planning model can make some adjustments to the model to analyze environmental justice
impacts. Also, gravity models can be used as a stand-alone tool. Some technical skills are
required, however, and so this method may not be suitable for agencies with limited resources.
Analysis. With traffic analysis zones (TAZs), census tracts, or census block groups as the units
of observation, gravity models can predict how a highway project will affect the attractiveness of
an area through changes in its relative accessibility. Consider the hypothetical example in Table
9-3. In the table, origin TAZs are shown in each column, and destination TAZs in the rows. For
example, initially, average time for trips that start and end in TAZ A is 5 minutes. Average time
for trips that start in TAZ A and end in TAZ B is 15 minutes.
The impact of the project is depicted in Figures 9-1 and 9-2. Before a new highway is built,
traffic from TAZ A to C must go through B. The construction of the highway causes disruption
in TAZs A and C, thereby increasing travel time both within the TAZs and between them.
However, after the highway is opened, travel time between A and C (bypassing B) is reduced by
more than half. Because less traffic needs to go through TAZ B, the travel time between A and B
224
OCR for page 220
also decreases. However, the travel time between A and C becomes less than between A and B.
The question is then, how does the new highway affect the relative attractiveness of each zone?
The production-constrained gravity model can be written as:
Oi w a b
j dij
Tij =
w a b
d
j ij
j
where
Tij = the shopping spending of zone i residents in zone j
Oi = total shopping spending of zone i residents
wj = the number of retail stores (or total retail square footage) in zone j
dij = the distance between zone i and j
The total amount of retail sales in zone j, Rj is given by
R j = Tij
i
Table 9-3.
Travel times between TAZs
Initial travel times (minutes)
Origin TAZ
Destination TAZ A B C
A 5 15 25
B 15 5 10
C 25 10 5
Travel times during the construction (minutes)
Origin TAZ
Destination TAZ A B C
A 10 15 30
B 15 5 10
C 30 10 10
Travel times after the opening of the highway (minutes)
Origin TAZ
Destination TAZ A B C
A 5 12 10
B 12 5 8
C 10 8 5
The parameter a is expected to be positive, as the larger the number of stores in a zone, the more
attractive it will be to shoppers. In contrast, b is expected to be negative because the more distant
225
OCR for page 221
the zone, the less likely shoppers will travel to shop there. The above model can be estimated
with survey data on shopping travel patterns. In past studies, the ranges of a and b were found to
be between 0.5 and 2.0, and -0.5 and -2.0, respectively.
Although how the values of the parameters are determined is beyond the scope of this
guidebook, some insights into their relative values may be provided here. For example, the
values of the parameters may reflect characteristics of trip makers and road networks.
Inaccessible locations, such as those within congested areas, may be associated with larger dij
because even short distances can provide disincentives for making a trip. Also, destination
characteristics and trip purpose can be reflected by the parameter wj. For instance, grocery stores
attract trips with higher frequency than furniture stores. In this case, zones with stores that attract
trips with lower frequency may be associated with a smaller wj.
A
C
15
10
B
Figure 9-1. Initial network travel times between TAZs
A
10 C
12
8
B
Figure 9-2. Network travel times between TAZs
after the opening of the highway
226
OCR for page 222
The forecasted link travel times during and after construction can be input into the gravity model.
The results will be a new pattern of trip distribution between zones. If we assume that each zone
generates the same amount of traffic, the travel time between zones will determine the amount of
traffic between them. The initial distribution will therefore be such that TAZ B will be the most
attractive because it is most accessible to both A and C. During the construction, the distribution
pattern will not change much because even though travel times in TAZs A and C increase, the
relative travel times do not change overall. However, after the highway is opened, travel times
from both TAZs A and B to C will decrease, thereby improving the overall attractiveness of TAZ
C. TAZ B, on the other hand, will become less attractive relative to C. With this pattern of
changes in trip distribution, we can assume that businesses in TAZ C will benefit from the
highway opening, but those in TAZ B will likely suffer. According to these results, we can create
a map that reflects the distribution of benefits as shown in Figure 9-3.
A
Neutral C
Positive
Negative
B
Figure 9-3. Distribution of economic impacts after construction
This map can then be overlaid on a map that represents ethnic and income groups to analyze the
environmental justice impacts of the new highway. It is also possible to quantify the gain and
loss at different locations in monetary terms. A gravity model can be used to allocate retail-
shopping spending to various neighborhoods based on existing residential locations and
accessibility changes due to the new highway project. For example, if the population and average
income in the three TAZs above are known, they can be used to estimate retail-shopping
spending by residents of each TAZ using the gravity model described below.
To quantify the impact of a highway project on retail sales, you can compute the retail sales in
each TAZ based on current network travel times. This is used as a basis for comparison. The new
network travel times can then be used to predict retail sales after the project is completed. The
change in retail sales reflects gains and losses in each TAZ. This information can be used to
construct a map of distributive impacts. The overlaying technique, as discussed earlier, can then
be used for further analysis to determine how distribution impacts may differentially affect
protected populations.
Data needs, assumptions, and limitations. The key information for this method is how a
highway project affects accessibility. This information will be used to determine changes in link
227
