Content uploaded by Heidi Peltier
Author content
All content in this area was uploaded by Heidi Peltier on Feb 19, 2015
Content may be subject to copyright.
PEDESTRIAN AND BICYCLE
INFRASTRUCTURE:
A NATIONAL STUDY OF
EMPLOYMENT IMPACTS
Heidi Garrett-Peltier
Political Economy Research Institute
University of Massachusetts, Amherst
June 2011
PEDESTRIAN AND BICYCLE INFRASTRUCTURE: A NATIONAL STUDY OF EMPLOYMENT IMPACTS / PAGE 1
EXECUTIVE SUMMARY
Pedestrian and bicycling infrastructure such as side-
walks, bike lanes, and trails, can all be used for
transportation, recreation, and fitness. These types of
infrastructure have been shown to create many bene-
fits for their users as well as the rest of the commu-
nity. Some of these benefits are economic, such as
increased revenues and jobs for local businesses,
and some are non-economic benefits such as re-
duced congestion, better air quality, safer travel
routes, and improved health outcomes. While other
studies have examined the economic and non-
economic impacts of the use of walking and cycling
infrastructure, few have analyzed the employment
that results from the design and construction of these
projects. In this study we estimate the employment
impacts of building and refurbishing transportation
infrastructure for cyclists and pedestrians. We ana-
lyze various transportation projects and use state-
specific data to estimate the number of jobs created
within each state where the project is located.
The data for this study were gathered from de-
partments of transportation and public works de-
partments from 11 cities in the United States. Using
PEDESTRIAN AND
BICYCLE
INFRASTRUCTURE:
A NATIONAL STUDY
OF EMPLOYMENT
IMPACTS
HEIDI GARRETT-PELTIER
Political Economy Research Institute
University of Massachusetts, Amherst
June 2011
detailed cost estimates on a variety of projects, we
use an input-output model to study the direct, indi-
rect, and induced employment that is created
through the design, construction, and materials pro-
curement of bicycle, pedestrian, and road infrastruc-
ture. We evaluate 58 separate projects and present
the results by project, by city, and by category. Over-
all we find that bicycling infrastructure creates the
most jobs for a given level of spending: For each $1
million, the cycling projects in this study create a to-
tal of 11.4 jobs within the state where the project is
located. Pedestrian-only projects create an average
of about 10 jobs per $1 million and multi-use trails
create nearly as many, at 9.6 jobs per $1 million.
Infrastructure that combines road construction with
pedestrian and bicycle facilities creates slightly fewer
jobs for the same amount of spending, and road-only
projects create the least, with a total of 7.8 jobs per
$1 million. On average, the 58 projects we studied
create about 9 jobs per $1 million within their own
states. If we add the spill-over employment that is
created in other states through the supply chain, the
employment impact rises by an average of 3 addi-
tional jobs per $1 million.
PEDESTRIAN AND BICYCLE INFRASTRUCTURE: A NATIONAL STUDY OF EMPLOYMENT IMPACTS / PAGE 2
BACKGROUND
This study was undertaken in order to understand
the employment impacts of bicycle and pedestrian
infrastructure. In January 2009 the Political Economy
Research Institute (PERI) published a study analyzing
the needs and job creation effects of public invest-
ments in a wide variety of infrastructure projects,
including energy, water, and transportation.1 How-
ever, the transportation infrastructure we considered
in that study did not specifically include cycling or
walking infrastructure that could be used for com-
muting as well as recreational purposes. In searching
through the literature, we discovered that there were
no studies which specifically addressed the job crea-
tion that results from building infrastructure such as
bike lanes, multi-use trails, and pedestrian facilities.
This study, the first of its kind, was developed to fill
this need.
In this report, we estimate the jobs that are created
in the construction of bicycle and pedestrian facili-
ties. The manufacturing of the materials and equip-
ment, the design of the facilities, and the
construction and installation of each transportation
project can generate a significant number of jobs in
a variety of industries and occupations. Other eco-
nomic impact studies have focused on the use of
trails and other walking and cycling infrastructure,
and the dollars that flow into a community as a result
of this use. While these economic benefits can be
significant, they only represent a part of the picture.
A community will also experience significant em-
ployment benefits resulting from the design and con-
struction of trails, sidewalks, bike lanes, and related
projects.
Pedestrian and bicycle infrastructure offers many
services both to the users of that infrastructure as
well as the community at large. Cyclists, pedestrians,
joggers, and others who use trails, bike lanes, and
walkways to commute to work and school or for rec-
reation and exercise, experience health benefits,
reduced congestion, reduced costs for vehicle main-
tenance and operations, and increased travel safety.
1 Heintz, Pollin, and Garrett-Peltier (2009)
The community benefits from bicycle and pedestrian
infrastructure through increased economic activity,
higher property values,2 and improved environmental
quality. A number of researchers have documented
both these economic and non-economic benefits.
Research conducted by the Rails-to-Trails Conser-
vancy and various state Departments of Transporta-
tion generally draws on user surveys to gauge the
types of users and the revenues attributable to trail
use. For example, in their “Economic Benefits of
Trails and Greenways,” the Rails-to-Trails Conser-
vancy finds that economic benefits include tourism
and recreation-related spending (which is a boon to
businesses and increases local tax revenues), and a
rise in real estate values. Other benefits include
higher quality of life, environmental benefits such as
buffer zones to protect water sources from pollution
run-off, and mitigation of flood damage.3 A 2008
user survey of a multi-use trail in Pennsylvania
showed that over 80 percent of users purchased
“hard goods” such as bikes and cycling equipment in
relation to their use of the trail, and some also pur-
chase “soft goods” such as drinks and snacks at
nearby establishments.4
In some areas, such as the northern Outer Banks of
North Carolina, bicycle facilities partly drive tourism.
A 2003 economic impact analysis of a bicycle trail
system in this area focused on economic benefits
such as tourist spending on food, lodging, and enter-
tainment.5 Data were gathered through user surveys
and bicycle traffic counts to estimate the amount of
money that tourists spent during a visit, the total
number of tourists, and the proportion of tourists for
whom bicycling was an important reason for the visit.
The researchers found that, annually, approximately
68,000 tourists visited the area at least partly to cy-
cle. This led to an estimate that $60 million in tour-
ism spending and multiplier effects came to the area
2 For example, see Karendeniz (2008) for the relationship between
home prices and trail proximity or Cortright (2009) for the impacts of
“walkability” on home values in U.S. cities.
3 Trails and Greenways Clearinghouse (2004)
4 Knoch and Tomes (2008)
5 Lawrie et al (2006)
PEDESTRIAN AND BICYCLE INFRASTRUCTURE: A NATIONAL STUDY OF EMPLOYMENT IMPACTS / PAGE 3
in relation to the bikeways, and supported approxi-
mately 1,400 jobs.
According to a nationwide survey of over 1,000
households in rural, suburban, and urban areas, cy-
cling and walking facilities are important to a strong
majority of people.6 The Bureau of Transportation
Statistics, in its October 2009 survey, found that 66
percent of people said it was “very important” to
them to have sidewalks, paths, or other safe walking
routes to work or school. In addition, 37 percent of
people said it was “very important” and 33 percent
said it was “somewhat important” to have bike lanes
or paths to work and school. When asked about the
importance of having pedestrian-friendly streets or
boulevards in their downtown or central business
district, 60 percent of respondents said it was “very
important”. When the sample was restricted to re-
spondents in metropolitan areas, these percentages
were even higher.
The above evidence shows that there is clearly public
support for bicycle and pedestrian facilities, and that
both users of these facilities as well as the rest of
the community can experience benefits. As noted by
the Transportation Research Board of the National
Academy of Sciences:
Transportation planning and policy efforts at
all levels of government aim to increase lev-
els of walking and bicycling. To make the
best use of limited transportation funds there
is a critical need for better information about
two important considerations relating to bi-
cycle facilities. The first of these is the cost of
different bicycle investment options. The
second is the value of the effects such in-
vestments have on bicycle use and mode
share, including the resulting environmental,
economic, public health, and social benefits.7
6 Bureau of Transportation Statistics (2009)
7 National Cooperative Highway Research Program, “Guidelines for
Analysis of Investments in Bicycle Facilities,” Transportation Research
Board of the National Academies, Report 552, 2006.
DATA AND METHODOLOGY
In order to estimate the employment impacts of vari-
ous pedestrian and bicycle infrastructure projects, we
start by following the methodology outlined by the
Transportation Research Board (TRB). In its 2006
report, “Guidelines for Analysis of Investments in Bi-
cycle Facilities,” the TRB examines three categories
of bicycle facilities: on-street facilities such as bike
lanes and shared streets; off-street facilities such as
trails adjacent to roads or converted rail trails; and
bicycle equipment such as signs, signals, and park-
ing. The data for the TRB report were gathered from
various sources including transportation profession-
als, a literature review, and industry information from
completed projects and bid prices. Among the cost
data collected by the TRB are detailed capital costs
for construction of bicycle facilities, including such
line items as clearing and grubbing, pavement re-
moval, crushed stone, concrete pavement, and ther-
moplastic pavement markings.
Following the guidelines established by the TRB, we
gathered detailed price data on various components
of design and construction of cycling, walking, and
road infrastructure, including paving materials, sign-
age, structures (such as bridges), equipment such as
bollards and bike racks, and services such as engi-
neering and traffic maintenance. We partnered with
America Bikes to gather transportation project data
from a variety of small and large cities nationwide.
Together we contacted city planning departments
and personnel in Departments of Transportation. We
compiled data on bid prices and costs for completed
projects including bike lanes, sidewalks, multi-use
paths, other bicycle and pedestrian improvements,
as well as construction and resurfacing of roads that
did not include bicycle or pedestrian components.
The cost data were very detailed, generally including
dozens or sometimes hundreds of line items per pro-
ject, including specific dollar amounts for each con-
struction project input.
This report includes data on a total of 58 projects
from 11 cities nationwide. In total, we contacted
transportation officials in 90 cities. Of these, 55
responded with a willingness to contribute to this
PEDESTRIAN AND BICYCLE INFRASTRUCTURE: A NATIONAL STUDY OF EMPLOYMENT IMPACTS / PAGE 4
research project. The data requirements for this rig-
orous project-by-project analysis were substantial,
and this ultimately limited the number of cities that
were able to provide sufficiently detailed cost data
over multiple project types. Twenty were able to send
in some data, and we selected the 11 cities that sent
the most complete data.8 The cities included in this
analysis are:
Anchorage, Alaska
Austin, Texas
Baltimore, Maryland
Bloomington, Indiana
Concord, New Hampshire
Eugene, Oregon
Houston, Texas
Lexington, Kentucky
Madison, Wisconsin
Santa Cruz, California
Seattle, Washington
We analyzed three to six projects in each of these
cities. For most projects we had cost estimates from
multiple sources (for example, an estimate from a
city engineer as well as multiple bids from contrac-
tors) and in those cases we used the average of the
cost estimates for each project.
While we followed the TRB report’s methodology in
collecting and assembling cost data, our analysis dif-
fers from the TRB report in three significant ways.
First, this study focuses only on the capital costs of
building transportation infrastructure, and does not
include the ongoing maintenance and use of bicycle
facilities as does the TRB report. Secondly, we evalu-
ate not only bicycle infrastructure but also pedestrian
and road infrastructure. And thirdly, we extend the
TRB methodology of cost assessment by estimating
employment impacts. The TRB analysis does not in-
clude job creation. We now turn to the methodology
for developing our employment estimates.
Once we assembled the detailed cost data on our 58
projects, we used an input-output model to estimate
8 By “complete” we mean that the data for each project contained
very detailed descriptions and costs for the project inputs, and that
multiple project types were provided by the city, allowing us to study
the variation between projects within a city.
the employment effects of these projects. The input-
output (I-O) model allows us to assess the economy-
wide impacts of various activities. In addition to the
direct jobs that are created in the engineering and
construction firms involved in infrastructure projects,
jobs are created in the supply chain of these indus-
tries, which we call indirect jobs. These indirect jobs
are in industries such as cement manufacturing, sign
manufacturing, and trucking. Furthermore, as workers
in the direct and indirect industries spend their earn-
ings, they create demand in industries such as food
services and retail establishments, which we call the
induced effects. The I-O model captures not only the
direct employment and output effects of an activity,
but also the indirect and induced effects, and there-
fore provides a more complete picture of the impacts
resulting from infrastructure spending. Table 1 con-
tains a list of the direct and indirect industries that
experience the greatest job creation as a result of
building bicycle, pedestrian, and road infrastructure.
The model we used for this research is IMPLAN ver-
sion 3, an I-O model built primarily from U.S. Bureau
of Economic Analysis (BEA) data along with additional
data sources. The BEA, through its Economic Census
as well as other surveys, collects data from millions of
businesses nationwide which it compiles into input-
output accounts that show supply linkages between
approximately 500 industries, as well as demand
relationships between consumers (individuals, busi-
nesses, and governments) and these industries. We
have used the IMPLAN model for past research pro-
jects including studies of clean energy investments,
environmental regulation, and state taxes,9 and our
employment estimates have been shown to be con-
sistent and accurate as demonstrated most fully
through the large-scale statistical research we con-
ducted for the U.S. Department of Energy.
For this analysis, we used the IMPLAN I-O model
with 2008 data (the most recent available at the
time the analysis was performed). For each of our 11
9 See, for example: Pollin, Heintz and Garrett-Peltier (2009); Heintz,
Garrett-Peltier, and Zipperer (2011); Thompson and Garrett-Peltier
(2010)
PEDESTRIAN AND BICYCLE INFRASTRUCTURE: A NATIONAL STUDY OF EMPLOYMENT IMPACTS / PAGE 5
locations, we used the data specific to that state. In
order to be able to use the I-O model, we first had to
assign industry codes to each of the projects’ cost
categories. The data provided to us were very de-
tailed, and enabled us to identify the type of product
or material for each item in the construction project.
Once we determined which industry would manufac-
ture or provide each item in the project, we assigned
an industry code to that item. For example, we as-
signed individual industry codes to materials such as
hot-mix asphalt and thermoplastic pavement mark-
ings. We first categorized each cost according to the
North American Industrial Classification System
(NAICS), an industrial coding system developed by
the Census Bureau and used by Federal statistical
agencies. We then used the NAICS-IMPLAN concor-
dance to model the project within IMPLAN. Thus for
each of our 58 projects, we constructed very detailed
industry purchasing patterns and then used the
model and data specific to that state to estimate the
employment impacts of those purchases.
Using the I-O model, we estimated the direct and
indirect employment effects. In order to compare
effects between different areas and projects, we use
a standard spending amount of $1 million. Thus in
reporting our employment impacts we show the
number of full-time equivalent jobs that are created
for each $1 million of spending on any given project.
To estimate the induced effects we used state-
specific data on imports and exports to generate
state-specific multipliers. The induced effect esti-
mates the employment and output that result when
workers in the direct and indirect industries spend
their earnings on items such as food, clothing, and
healthcare. In previous work, we found that the in-
duced effect was equal to 40 percent of the com-
bined direct and indirect effects at the national
level.10 At the local (city or state) level, however, the
induced effect will be lower than the national induced
effect, since workers spend their earnings on goods
which are imported not only from overseas but also
from out-of-state. We adjusted the induced effects
10 For example, see the discussion in “Green Prosperity” by Pollin,
Wicks-Lim, and Garrett-Peltier, available at www.peri.umass.edu
downwards by using the ratio of local (state) supply to
local (state) demand for each of our data sets. The
state-specific induced effects in this study range from
a low of 28 percent (Alaska) to a high of 38 percent
(Maryland). On average, the induced effects at the
state level were about three-quarters the national
induced effects, or about 31 percent of the combined
direct plus indirect employment. As discussed below,
the indirect effect is also lower at the state level than
the national level, something that is explicitly cap-
tured in the model since we use state-specific data.
Finally, in order to eliminate any variation in the data
that results strictly from regional price differences,
we converted all of the project data into shares of
the total project cost. So, for example, rather than
inputting the number of dollars that were spent on
asphalt for project X, we inputted the percentage of
the total project cost that was attributable to asphalt.
In this way we can compare projects whose budgets
are of different magnitudes as well as comparing
results across cities. As long as the composition of a
certain type of project (in terms of materials, equip-
ment, and services) is similar across cities, any re-
gional price differences will not affect the results of
the analysis.
RESULTS AND DISCUSSION
Employment Impacts
In the tables below, we show the employment im-
pacts for various types of projects. In all cases, we
present the level of job creation, in terms of full-time-
equivalent positions, that results from spending the
same amount, $1 million, on any given project. By
using a consistent spending amount such as this, we
can more readily see the differences in job creation
that are attributable either to the type of project or to
the city. Below we will discuss the sources of this
variation.
In Table 2, we present the average employment im-
pacts for different types of projects. We analyzed a
total of 58 projects in 11 cities. We first estimated
the employment impacts of each individual project,
PEDESTRIAN AND BICYCLE INFRASTRUCTURE: A NATIONAL STUDY OF EMPLOYMENT IMPACTS / PAGE 6
then grouped similar project types. In the table we
see that for all projects, the average level of job crea-
tion is 4.7 direct jobs, 2.1 indirect jobs, and 2.1 in-
duced jobs, for a total of about 9 jobs per $1 million
spending. It is important to keep in mind that these
estimates are averages of the specific projects, and
that they reflect only the jobs created within the
state in which the project is undertaken. Below we
will discuss how these estimates differ from em-
ployment impacts at the national level using the na-
tional data set.
The projects listed in Table 2 include a range of
transportation infrastructure. Among them are road-
only projects (such as widening an existing road or
repaving/resurfacing roads that do not have either
bike lanes or sidewalks), road projects that include
pedestrian components such as sidewalks, road in-
frastructure with both pedestrian and bicycle compo-
nents such as bike lanes and signage, projects that
are uniquely pedestrian facilities (such as refurbish-
ing sidewalks or improving pedestrian crossings),
others that are specific to cycling (such as adding or
marking bike lanes), multi-use trails which could ei-
ther be alongside (but separate from) a road or off-
road trails such as converted rail trails, and on-street
facilities that are for both bicycling and walking but
do not include road construction itself (such as refur-
bishing or expanding sidewalks and bike lanes).
For each project category, we list in Table 2 the
number of projects as well as the direct, indirect,
induced, and total employment impacts per $1 mil-
lion spending. We see that the largest category in
terms of number of projects is road construction with
bicycle and pedestrian facilities. In fact, many of the
cities we contacted informed us that the majority of
their road projects now include at least some com-
ponent of biking or walking infrastructure, be it side-
walks, wide shoulders, or designated bike lanes. Out
of our 58 total projects, 13 (or 22%) were roads with
pedestrian and bicycle facilities. The next largest
group was road-only projects, which consisted of 11
total projects or 19% of the total. That was followed
by pedestrian-only projects (10) and then road infra-
structure with pedestrian components (9). We also
collected data on nine multi-use trails in six cities.
Finally, the data included a small number of bike-
only projects (4) and on-street projects that had both
cycling and walking components but no other road
construction elements (2).
We see from the table that the greatest level of job
creation is for infrastructure projects that are specific
to cycling, such as creating or refurbishing bike
lanes. This category results in an average of 6 direct
jobs per $1 million spending, plus 2.4 indirect jobs
and 3 induced jobs for a total of 11.4 jobs created
for each $1 million spent on bicycling infrastructure.
The lowest level of job creation is for road-only pro-
jects such as repaving or widening roads. This type of
infrastructure creates 4 direct, 1.8 indirect, and 1.8
induced jobs, for a total of 7.8 jobs per $1 million
spent on road-only infrastructure. The remaining pro-
jects, which consist of various elements of pedes-
trian and/or cycling facilities, range from job creation
levels of 4.2 direct jobs and 8.4 direct, indirect, plus
induced jobs (for on-street biking and walking) to 5.2
direct jobs and 9.9 total jobs including the direct,
indirect, and induced effects (for pedestrian-only in-
frastructure). Thus, on average, these various trans-
portation infrastructure projects create between 8
and 11 total jobs for each $1 million spent. The job
creation effects are higher for bicycle-only and pe-
destrian-only facilities and are lowest for road-only
facilities. Below we will discuss reasons for these
differences.
Next we turn our attention to specific project catego-
ries in each of the 11 cities from which we gathered
data. As we see from the city tables, the job creation
effects of projects in some cities are quite different
from the national average. We saw above that pe-
destrian and bicycle infrastructure creates, on aver-
age, more jobs for a given level of spending than
road-only projects. We also saw that, on average,
bicycle-only and pedestrian-only infrastructure create
the most jobs, followed by off-street multi-use
(bike/ped) trails. When we evaluate the impacts
within each city, however, we sometimes find that
these relationships change. For example, in Anchor-
age, Alaska, the city’s one road-only project actually
created more jobs than projects that included pedes-
trian and/or bicycle components. In Bloomington,
PEDESTRIAN AND BICYCLE INFRASTRUCTURE: A NATIONAL STUDY OF EMPLOYMENT IMPACTS / PAGE 7
Indiana, we find that the city’s one road-only project
created slightly more jobs per given amount of
spending than either of the city’s two road projects
that contain bicycle and pedestrian facilities. How-
ever, we also see that Bloomington’s two trail pro-
jects and one pedestrian-only facility each generated
more employment than any of the three projects that
included a road element, and that the employment
impacts of all six projects in this city had a relatively
narrow range, from about 7.3 to 9.0 jobs per $1 mil-
lion. Below we discuss the reasons for the overall
differences in the numbers of jobs created.
Out of the 11 cities we studied, seven cities were
able to provide data on projects that were road-
specific and did not contain any pedestrian or bicy-
cling facilities. Out of these seven cities, we found
that in only one city, Anchorage, Alaska, the road-only
project actually created more employment than all
other transportation projects. In the other six cities,
cycling and/or walking infrastructure created more
jobs per $1 million than road-only infrastructure. As
explained in more detail below, the cycling and walk-
ing infrastructure projects analyzed for this study
generally create more jobs than road infrastructure
because of their relative labor intensity and lower
leakages (purchases made out-of-state). In the case
of Anchorage, the road project (resurfacing an exist-
ing road) was relatively labor-intensive and the mate-
rials were almost completely sourced from in-state
suppliers. The bicycle and pedestrian facilities, on
the other hand, involved more goods imported from
out-of-state, such as some lighting fixtures and alu-
minum products.
Of the 58 projects studied, the lowest job total job
creation was 5 total jobs per $1 million, for road-only
infrastructure in Santa Cruz, California. The highest
was over 14 total jobs per $1 million, for cycling in-
frastructure in Baltimore, Maryland. The median level
of total job creation for our 58 projects was about 9
jobs per $1 million.
Discussion
There are three main reasons why employment im-
pacts could differ between types of projects and be-
tween cities. These are:
Labor intensity
Leakages (spending on goods from out-of-
state)
Wage differences
Labor intensity refers to the ratio of labor to capital
(materials, plant, and equipment). In labor-intensive
industries such as construction and engineering,
more of the total dollars spent go to wages and sala-
ries. In more capital-intensive industries such as ce-
ment manufacturing, relatively fewer dollars are
spent on salaries and more are spent on materials
and equipment, in comparison to labor-intensive in-
dustries. Thus in the projects we studied, the infra-
structure with higher labor intensity of production will
create more jobs for a given level of spending. This is
the primary reason why pedestrian-only and bicycle-
only infrastructure create more jobs than road-only
projects. For the former types, a greater portion of
the spending is used to employ construction workers
and engineers, both labor-intensive industries. In the
latter, a greater proportion of the total spending is
used for materials such as asphalt and stone prod-
ucts. Thus, for example, a bike path which requires a
great deal of planning and design will generate more
jobs for a given level of spending than a road project
which requires a greater proportion of heavily
mechanized construction equipment and relatively
less planning and design. Engineering and related
services are labor-intensive items, thus projects
whose budgets have a higher percentage of these
services will create more jobs.
The other reason for variation in the employment
impacts presented here is leakages. When pur-
chases of materials are made, some of these mate-
rials are supplied by in-state businesses, creating
jobs within the city and state. However, there is
some amount of “leakage”, or flow of dollars out of
the state, resulting from purchases of goods that
come from other states or countries. When a higher
percentage of goods can be provided by in-state
PEDESTRIAN AND BICYCLE INFRASTRUCTURE: A NATIONAL STUDY OF EMPLOYMENT IMPACTS / PAGE 8
Comparison of state employment impacts to
national employment impacts
Number of
projects
Direct jobs per
$1 million
Indirect jobs per
$1 million
Induced jobs
per $1 million
Total jobs per
$1 million
Average of all projects using state data 58 4.69 2.12 2.15 8.96
Average of all projects using national data 58 4.53 3.93 3.38 11.84
% above state effects -3% 86% 57% 32%
suppliers, the leakages are lower and the total in-
state employment effect is higher. For example,
some of the materials needed to build a road include
asphalt, stone, and iron manhole covers. If a road is
being built in California and all of these products can
be bought from companies in California, then jobs
will be created in the state. If, however, some of
these products need to be purchased from suppliers
in another state, then some jobs will be created in
that state and fewer jobs will be created in California.
These out-of-state purchases are considered “leak-
ages” and reduce the in-state employment impact.
The leakages appear as lower indirect effects and
lower induced effects. A city such as Anchorage,
Alaska, which has to source some of its project in-
puts from other states, will have lower indirect and
induced job creation within the state. As mentioned
above, even within a city such as Anchorage, there
can be differences between projects in the extent to
which goods are sourced from in-state or out-of-state
suppliers. For the sake of comparison, after using
state-specific data to estimate our 58 separate pro-
jects, we also estimated the same projects using
IMPLAN v3 with the 2008 U.S. national data set.
When we use the national data, jobs that are created
through interstate trade are captured -- there are no
leakages from purchases made from other states.
Thus in our example of the road-building project in
California, the national estimate would capture both
the jobs created in California as well as the jobs cre-
ated in Arizona, Oregon, or any other state which
supplies goods for the road building project in Cali-
fornia. However when we use the California data set,
we estimate only the jobs created in California. This
is an accurate estimate of the in-state job creation
but understates the full job-creation effect of the pro-
ject. The employment effects of using state data in
comparison to national data are presented here:
As we see from the table, the direct jobs are nearly
identical when using the national data set versus
averaging the results of the state data. However,
once we estimate the indirect and induced effects,
we see a large difference. The national employment
impact for indirect jobs is nearly twice as high (close
to four jobs using the national data, compared to just
over two jobs using the average of the state results).
This difference captures the out-of-state leakages. At
the national level, the only leakages are out of the
country, while interstate trading creates jobs. At the
state level, interstate trading creates jobs in other
states, and therefore is not captured in the employ-
ment impacts of the state being studied.
The overall estimated employment effects of the pro-
jects studied here would therefore be higher if we
counted indirect and induced employment creation
in other states. As we see in the table above, nearly
two additional indirect jobs are created in other
states for each $1 million spent on these types of
projects. Furthermore, because the induced effects
also suffer from leakages, more than one additional
job is created out of state through the induced effect
for each $1 million spent on the projects studied
here. If we added the jobs that are created in other
states, both indirectly and through the induced ef-
fect, the 58 projects studied here would create an
average of 3 additional jobs, or 32 percent greater
employment creation, for each $1 million spent on
transportation infrastructure.
The third reason why projects can differ in their job-
creation potential is wage differences. It is beyond
the scope of this study to evaluate differences in pay
between various industries and cities. Therefore we
cannot conclude whether or not wage differences
play a role in explaining the variation in employment
impacts among the projects presented here.
PEDESTRIAN AND BICYCLE INFRASTRUCTURE: A NATIONAL STUDY OF EMPLOYMENT IMPACTS / PAGE 9
One other point that deserves mention here is the
difference between the employment estimates from
the 58 transportation projects in this study and the
results we obtained in our national study of infra-
structure investments published in January 2009.11
In the earlier study, we found that nationally, road
infrastructure construction created about 19,000
jobs per $1 billion investment (or 19 jobs per $1 mil-
lion, since the model is linear). In the 58 projects
analyzed in this report, on average nine jobs are cre-
ated per $1 million, or 12 jobs if we incorporate the
job creation from out-of-state purchases. In the 2009
study, we used national data from 2006, while in this
study we use state data from 2008. This may ac-
count for some of the variation in these estimates.
However the main reason for this difference stems
from the level of detail at which we analyzed the
transportation infrastructure investments. In the ear-
lier study, our estimate was derived from an em-
ployment multiplier which included all types of
infrastructure construction, whereas in this study we
collected very detailed costs on materials, design,
construction, and other services. The resulting em-
ployment estimates are therefore much more spe-
cific to the particular projects we studied and vary
from our earlier, more general national estimate.
Finally, the impacts studied in this report are specific
to the design and construction of roads, bicycle, and
pedestrian facilities. They do not consider the ongo-
ing maintenance and use of these facilities. As men-
tioned above, other studies have estimated the
economic benefits and non-economic impacts of the
use of bicycling and pedestrian facilities, including
revenues and jobs for local bike shops and other
businesses. In addition to the use impacts, there is
also employment associated with maintenance of
these facilities, such as grounds-keeping. In short,
the employment effects of bicycle and pedestrian
infrastructure presented in this study represent only
one portion of the total impacts.
11 Heintz, Pollin, and Garrett-Peltier (2009)
CONCLUSION
The U.S. is currently experiencing high unemploy-
ment, unsustainable use of carbon-based energy,
and a national obesity epidemic. All three of these
problems can be partly addressed through increased
walking and cycling. Providing pedestrian and cycling
infrastructure for the purposes of commuting, rec-
reation, and fitness, is arguably more important than
ever before. In addition, this study finds that design-
ing and building this infrastructure can also address
the problem of unemployment, by creating jobs for
engineers, construction workers, and workers who
produce the asphalt, signs, and other construction
materials.
We collected data from departments of transporta-
tion and public works departments in 11 cities na-
tionwide and evaluated 58 separate projects. These
projects ranged from road construction and rehabili-
tation, to building new multi-use trails and widening
roads to include bike lanes and sidewalks. Using an
input-output model with state-specific data, we esti-
mated the employment impacts of each project and
presented the results by project, by city, and by type.
We found that on average, these various transporta-
tion infrastructure projects create 9 in-state jobs for
each $1 million of spending and an additional 3 jobs
if we include out-of-state effects. In addition, we
found that the highest level of job creation was for
bicycle-only infrastructure such as building or refur-
bishing bike lanes. These projects created up to 11.4
jobs per $1 million when we consider only in-state
effects. This was followed by pedestrian-only infra-
structure (such as sidewalks and pedestrian cross-
ings) and multi-use trails, which created close to 10
jobs for each $1 million spent on the project. These
findings suggest that when confronted with a deci-
sion of whether or not to include pedestrian and/or
bicycle facilities in transportation infrastructure pro-
jects, planning officials should do so, not only be-
cause of the environmental, safety, and health
benefits but also because these projects can create
local jobs.
PEDESTRIAN AND BICYCLE INFRASTRUCTURE: A NATIONAL STUDY OF EMPLOYMENT IMPACTS / PAGE 10
TABLES
Table 1. Top 20 industries: direct and indirect job creation from bicycle, pedestrian, and road infrastructure
Construction of other new nonresidential structures
Cut stone and stone product manufacturing
Concrete product manufacturing (not including ready-mix concrete or concrete pipes)
Ready-mix concrete manufacturing
Greenhouse, nursery, and floriculture production
Architectural, engineering, and related services
Asphalt paving mixture and block manufacturing
Other support services (includes traffic maintenance)
Concrete pipe, brick, and block manufacturing
Sign manufacturing
Plastics product manufacturing (other than pipes, bottles, packaging materials)
Wholesale trade businesses
Transport by truck
Employment services
Food services and drinking places
Services to buildings and dwellings
Management of companies and enterprises
Real estate establishments
Maintenance and repair construction of nonresidential structures
Accounting, tax preparation, bookkeeping, and payroll services
PEDESTRIAN AND BICYCLE INFRASTRUCTURE: A NATIONAL STUDY OF EMPLOYMENT IMPACTS / PAGE 11
Table 2: National Average Employment Impacts by Project Type
Project type
Road
Bicycle
Pedestrian
Off-street trail
Number
of projects
Direct jobs
per $1
million
Indirect
jobs per
$1 million
Induced
jobs
per $1
million
Total jobs
per $1
million
Total, all projects 58
4.69
2.12 2.15
8.96
Bicycle infrastructure only • 4
6.00
2.40 3.01
11.41
Off-street multi-use trails • 9
5.09
2.21 2.27
9.57
On-street bicycle and pedestrian facilities (without
road construction) • • 2
4.20
2.20 2.02
8.42
Pedestrian infrastructure only • 10
5.18
2.33 2.40
9.91
Road infrastructure with bicycle and pedestrian
facilities • • • 13
4.32
2.21 2.00
8.53
Road infrastructure with pedestrian
facilities • • 9
4.58
1.82 2.01
8.42
Road infrastructure only (no bike or pedestrian
components) • 11
4.06
1.86 1.83
7.75
Employment Impacts by City
In the tables below, we present the employment impacts of various categories of transportation infrastructure by
city. Each line in the table represents a distinct project. For most projects, multiple cost estimates were aver-
aged in order to estimate the employment impact, as described in the “Methodology” section of this report.
Rather than using specific site or street names for these projects, we simply list the type of project (for example,
“Road Infrastructure with Pedestrian Facilities”) and list an A, B, or C after the category name if more than one
project of this type is listed in a city.
Anchorage, Alaska
Transportation infrastructure category
Road
Bicycle
Pedestrian
Off-street
trail
Direct jobs
per $1
million
Indirect
jobs per
$1
million
Induced
jobs per
$1 million
Total jobs
per $1
million
Total jobs
(avg. by
type)
Pedestrian infrastructure only • 5.6 1.9 2.07 9.57
Road infrastructure with bicycle and pedestrian
facilities • • • 3.9 1.3 1.44 6.64
Road infrastructure with pedestrian facilities – a • • 5.5 1.6 1.96 9.06
Road infrastructure with pedestrian facilities – b • • 5.7 1.8 2.07 9.57
Road infrastructure with pedestrian facilities – c • • 5.2 1.6 1.88 8.68
9.1
Road infrastructure only • 7.2 1.9 2.51 11.61
Average all projects 5.52 1.68 1.99 9.19
PEDESTRIAN AND BICYCLE INFRASTRUCTURE: A NATIONAL STUDY OF EMPLOYMENT IMPACTS / PAGE 12
Austin, Texas
Transportation infrastructure category
Road
Bicycle
Pedestrian
Off-street
trail
Direct jobs
per $1
million
Indirect
jobs per
$1
million
Induced
jobs per
$1 million
Total jobs
per $1
million
Total jobs
(avg. by
type)
Bicycle infrastructure only • 5.9 2.4 2.73 11.03
Off-street multi-use trails – a • 5.9 2.4 2.73 11.03
Off-street multi-use trails - b • 5.8 2.6 2.76 11.16 11.1
Road infrastructure with bicycle and pedestrian
facilities • • • 6.2 2.8 2.96 11.96
Road infrastructure with pedestrian facilities – a • • 5.3 2.3 2.5 10.1
Road infrastructure with pedestrian facilities – b • • 3.1 1.9 1.64 6.64 8.37
Average all projects 5.37 2.4 2.55 10.32
Baltimore, Maryland
Transportation infrastructure category
Road
Bicycle
Pedestrian
Off-street
trail
Direct jobs
per $1
million
Indirect
jobs per
$1
million
Induced
jobs per
$1 million
Total jobs
per $1
million
Total jobs
(avg. by
type)
Bicycle infrastructure only – a • 7.9 2.5 3.95 14.35
Bicycle infrastructure only – b • 6.1 2.4 3.23 11.73 13.04
Pedestrian infrastructure only • 6 2.2 3.1 11.3
Road infrastructure with pedestrian facilities – a • • 3.8 1.5 2 7.4
Road infrastructure with pedestrian facilities – b • • 3.4 1.5 1.9 6.8 7.1
Average all projects 5.44 2.02 2.84 10.32
Bloomington, Indiana
Transportation infrastructure
category
Road
Bicycle
Pedestrian
Off-street
trail
Direct
jobs per
$1 million
Indirect
jobs per
$1 million
Induced
jobs per
$1 million
Total jobs
per $1
million
Total jobs
(avg. by
type)
Off-street multi-use trails – a • 5 1.9 2.12 9.02
Off-street multi-use trails – b • 4.8 1.9 2.05 8.75 8.89
Pedestrian infrastructure only • 4.4 2.2 2.02 8.62
Road infrastructure with bicycle and pedestrian
facilities – a • • 3.8 1.7 1.69 7.19
Road infrastructure with bicycle and pedestrian
facilities – b • • 3.9 1.7 1.72 7.32
7.25
Road infrastructure only • 4.6 1.6 1.9 8.1
Average all projects 4.42 1.83 1.92 8.17
PEDESTRIAN AND BICYCLE INFRASTRUCTURE: A NATIONAL STUDY OF EMPLOYMENT IMPACTS / PAGE 13
Concord, New Hampshire
Transportation infrastructure category
Road
Bicycle
Pedestrian
Off-street
trail
Direct jobs
per $1
million
Indirect
jobs per
$1
million
Induced
jobs per
$1 million
Total jobs
per $1
million
Total jobs
(avg. by
type)
Pedestrian infrastructure only – a • 6.7 1.9 2.71 11.31
Pedestrian infrastructure only – b • 5.4 1.8 2.27 9.47 10.39
Road infrastructure only – a • 4.8 2 2.14 8.94
Road infrastructure only – b • 3.3 2.1 1.7 7.1
Road infrastructure only – c • 4.3 2 1.98 8.28
8.11
Average all projects 4.9 1.96 2.16 9.02
Eugene, Oregon
Transportation infrastructure category
Road
Bicycle
Pedestrian
Off-street
trail
Direct jobs
per $1
million
Indirect
jobs per
$1
million
Induced
jobs per
$1 million
Total jobs
per $1
million
Total jobs
(avg. by
type)
Pedestrian infrastructure only • 4.9 2.8 2.42 10.12
Road infrastructure with bicycle and pedestrian
facilities – a • • • 3.7 2.2 1.85 7.75
Road infrastructure with bicycle and pedestrian
facilities – b • • • 4.6 3 2.38 9.98
Road infrastructure with bicycle and pedestrian
facilities – c • • • 5 2.4 2.32 9.72
9.15
Road infrastructure with pedestrian facilities • • 4.5 2.2 2.1 8.8
Road infrastructure only • 3.4 1.8 1.63 6.83
Average all projects 4.35 2.4 2.12 8.87
Houston, Texas
Transportation infrastructure category
Road
Bicycle
Pedestrian
Off-street
trail
Direct jobs
per $1
million
Indirect
jobs per
$1
million
Induced
jobs per
$1 million
Total jobs
per $1
million
Total jobs
(avg. by
type)
Off-street multi-use trails • 3.7 2.4 1.83 7.93
Road infrastructure with bicycle and pedestrian
facilities – a • • • 4.2 2.3 1.95 8.45
Road infrastructure with bicycle and pedestrian
facilities – b • • • 3.9 2.4 1.89 8.19
8.32
Average all projects 3.94 2.36 1.89 8.19
Lexington, Kentucky
Transportation infrastructure category
Road
Bicycle
Pedestrian
Off-street
trail
Direct jobs
per $1
million
Indirect
jobs per
$1 mil-
lion
Induced
jobs per $1
million
Total jobs
per $1
million
Off-street multi-use trails • 5.1 2 2.12 9.22
On-street bicycle and pedestrian facilities • • 4.9 2.2 2.12 9.22
Road infrastructure with bicycle and
pedestrian facilities • • • 4.3 1.9 1.86 8.06
Average all projects 4.77 2.03 2.03 8.83
PEDESTRIAN AND BICYCLE INFRASTRUCTURE: A NATIONAL STUDY OF EMPLOYMENT IMPACTS / PAGE 14
Madison, Wisconsin
Transportation infrastructure category
Road
Bicycle
Pedestrian
Off-street
trail
Direct jobs
per $1
million
Indirect
jobs per
$1
million
Induced
jobs per
$1 million
Total jobs
per $1
million
Total jobs
(avg. by
type)
Off-street multi-use trails • 4.2 2.1 1.91 8.21
Pedestrian infrastructure only – a • 5.5 2.6 2.46 10.56
Pedestrian infrastructure only – b • 3.6 2.1 1.73 7.43 8.99
Road infrastructure with bicycle and pedestrian
facilities • • • 4.4 2.5 2.09 8.99
Road infrastructure with pedestrian facilities • • 4.7 2 2.03 8.73
Road infrastructure only • 3.9 1.7 1.7 7.3
Average all projects 4.38 2.17 1.99 8.54
Santa Cruz, California
Transportation infrastructure category
Road
Bicycle
Pedestrian
Off-street
trail
Direct jobs
per $1
million
Indirect
jobs per
$1
million
Induced
jobs per
$1 million
Total jobs
per $1
million
Total jobs
(avg. by
type)
Bicycle infrastructure only • 4.1 2.3 2.14 8.54
On-street bicycle and pedestrian facilities • • 3.5 2.2 1.91 7.61
Pedestrian infrastructure only – a • 5.6 2.9 2.85 11.35
Pedestrian infrastructure only – b • 4.1 2.9 2.34 9.34 10.35
Road infrastructure only – a • 2.2 1.5 1.24 4.94
Road infrastructure only – b • 2.3 1.6 1.31 5.21 5.07
Average all projects 3.63 2.23 1.97 7.83
Seattle, Washington
Transportation infrastructure category
Road
Bicycle
Pedestrian
Off-street
trail
Direct jobs
per $1
million
Indirect
jobs per
$1
million
Induced
jobs per
$1 million
Total jobs
per $1
million
Total jobs
(avg. by
type)
Off-street multi-use trails – a • 6.2 2.6 2.69 11.49
Off-street multi-use trails – b • 5.1 2 2.17 9.27 10.38
Road infrastructure with bicycle and pedestrian
facilities – a • • • 3.9 2 1.8 7.7
Road infrastructure with bicycle and pedestrian
facilities – b • • • 4.3 2.5 2.08 8.88
8.29
Road infrastructure only – a • 4.8 2.5 2.23 9.53
Road infrastructure only – b • 3.9 1.8 1.74 7.44 8.49
Average all projects 4.7 2.23 2.12 9.05
PEDESTRIAN AND BICYCLE INFRASTRUCTURE: A NATIONAL STUDY OF EMPLOYMENT IMPACTS / PAGE 15
BIBLIOGRAPHY
Bureau of Transportation Statistics, October 2009 Omnibus Household Survey, available at
http://www.bts.gov/programs/omnibus_surveys/household_survey/2009/october/pdf/entire.pdf
Cortright, Joe, August 2009, “Walking the Walk: How Walkability Raises Home Values in U.S. Cities,” Impresa, Inc. for CEO for
Cities.
Heintz, James, Robert Pollin, and Heidi Garrett-Peltier, January 2009, “How Infrastructure Investments Support the U.S. Economy:
Employment, Productivity, and Growth,” Political Economy Research Institute.
Heintz, James, Heidi Garrett-Peltier, Ben Zipperer, February 2011, “New Jobs — Cleaner Air: Employment Effects Under Planned
Changes to the EPA’s Air Pollution Rules,” Political Economy Research Institute.
Karedeniz, Duygu, 2008, “The Impact of the Little Miami Scenic Trail on Single Family Residential Property Values,” University of
Cincinnati Master’s Thesis.
Knoch, Carl, and Patricia A. Tomes, “Perkiomen Trail 2008 User Survey and Economic Impact Analysis,” Rails-to-Trails Conser-
vancy, December 2008.
Lawrie, Judson J., Thomas P. Norman, Mary Meletiou, and Sarah W. O’Brien, “Bikeways to Prosperity: Assessing the Economic
Impact of Bicycle Facilities,” North Carolina Department of Transportation and Institute for Transportation Research and Education,
TR News 242, January-February 2006.
Pollin, Robert, James Heintz and Heidi Garrett-Peltier, June 2009 “The Economic Benefits of Investing in Clean Energy: How the
Economic Stimulus Program and New Legislation Can Boost U.S. Economic Growth and Employment,” Political Economy Research
Institute.
National Cooperative Highway Research Program, “Guidelines for Analysis of Investments in Bicycle Facilities,” Transportation
Research Board of the National Academies, Report 552, 2006.
Thompson, Jeff and Heidi Garrett-Peltier, April 2010, “Generating Jobs through State Employer Tax Credits: Is there a Better Way?
(Revised),” Political Economy Research Institute.
Tomes, Patricia A. and Carl Knoch, “Trail User Surveys and Economic Impact: A Comparison of Trail User Expenditures,” Rails-to-
Trails Conservancy, March 2009.
Trails and Greenways Clearinghouse, 2004, “Economic Benefits of Trails and Greenways,” Rails to Trails Conservancy.
Data Provided By
Anchorage, Alaska: Municipality of Anchorage Public Works Department
Austin, Texas: City of Austin Department of Public Works
Baltimore, Maryland: Baltimore City Department of Transportation
Bloomington, Indiana: Indiana Department of Transportation
Concord, New Hampshire: City of Concord Engineering Division
Eugene, Oregon: City of Eugene Public Works Engineering
Houston, Texas: City of Houston Public Works and Engineering Department
Lexington, Kentucky: Lexington-Fayette Urban County Government
Madison, Wisconsin: City of Madison Public Works Department
Santa Cruz, California: City of Santa Cruz Public Works Department
Seattle, Washington: Seattle Department of Transportation
