Bus Rapid Transit and Economic Development: Case Study of The Eugene-Springfield BRT System
Bus Rapid Transit and Economic Development: Case Study of The Eugene-Springfield BRT System
Bus Rapid Transit and Economic Development: Case Study of The Eugene-Springfield BRT System
Abstract
Bus rapid transit (BRT) in the United States is relatively recent. BRT has many promises, one of which is enhancing the economic development prospects of firms locating
along the route. Another is to improve overall metropolitan economic performance.
In this article, we evaluate this issue with respect to one of the nations newest BRT
systems that operates in a metropolitan area without rail transit: Eugene-Springfield,
Oregon. While the metropolitan area lost jobs between 2004 and 2010, jobs grew
within 0.25 miles of BRT stations. Using shift-share analysis, we find that BRT stations
are attractive to jobs in several economic sectors. Planning and policy implications
are offered along with an outline for future research.
Introduction
In this article, we assess the relationship between bus rapid transit (BRT) and the
change in share of jobs in an urban area during the 2000s. Eugene-Springfield,
Oregon, is our case study. The Eugene-Springfield BRT system is well-suited for this
analysis because it has one of the nations newest BRT systems, so we can assess
economic influences in the short-term; its system is reasonably representative of
emerging BRT design; and we were able to acquire employment data, allowing us
conduct spatially-related analysis. Our analysis covers the years 2004 and 2010,
41
which were three years before and after the BRT system was opened. Our article
includes these sections: Fixed-Guideway Transit Systems and Economic Development; Overview of the Eugene-Springfield BRT System; Research Method and Data;
Assessment of Results; and Planning and Policy Implications.
within walking distance of transit stations, wherever they are located (Belzer, Srivastava, and Austin 2011).
There is another aspect of agglomeration economies identified by Chapman and
Noland (2011). Although transit systems can lead to higher-density development
by shifting new jobs and population to station areas, it could lead, instead, to the
redistribution of existing development even in the absence of growth.
In part because of their role in facilitating agglomeration economies, there is a
growing body of research showing that rail-based public transit enhances economic development (see Nelson et al. 2009). These economies are facilitated when
they improve accessibility between people and their destinations (Litman 2009) by
reducing travel time and the risk of failing to arrive at a destination (Weisbrod and
Reno 2009). At the metropolitan scale, adding transit modes in built-up urban areas
increases aggregate economic activity (Graham 2007).
Economic development can be measured in many ways. One is by evaluating how
the market responds to the presence of transportation investments, such as rail
stations. Higher values closer to stations implies market capitalization of economic
benefits, which can occur only when economic activity increases. Only a few studies have shown this with respect to commercial property values (Nelson 1999) and
none for BRT, although one study shows positive residential property value effects
(see Perk and Catal 2009).
Our focus here is whether and the extent to which there is a link between a specific form of transitBRTand job growth. We know from recent work that not
all firms benefit from transit. In their recent study of employment within one-half
mile of transit stations serving 34 rail systems, Belzer, Srivastava and Austin (2011)
found that while jobs increase in the Arts/Entertainment/Recreation sectors, as
well as the Accommodation and Food Services and Health Care and Social Assistance sectors, they fell in the Manufacturing sector. They also found that the Public
Administration sector had the greatest share of jobs found near transit stations.
Several other sectors also concentrated around transit stations, such as Professional, Scientific, and Technical Services and Retail. On the other hand, as a whole,
the station areas experienced declining shares of jobs relative to their regions,
with the exception of jobs in the Utilities, Information, and Arts/Entertainment/
Recreation sectors. Belzer, Srivastava, and Austin (2011) surmised that much of the
metropolitan job growth continues to favor auto-oriented locations. Their study
did not report results for individual systems and, as it was based on data through
2008, came just one year after the Eugene-Springfield BRT opened.
43
Assessment of Results
Table 1 reports our overall assessment of change in employment between 2004
and 2010. We report jobs for areas within 0.25 miles of a station, between 0.25
and 0.50 miles of a station, and the balance of the metropolitan area. Overall, for
the metropolitan area outside the 0.50 mile BRT station areas, jobs fell by about
5 percent or more than 5,000. Jobs stayed about the same between 0.25 and 0.50
miles of station areas but increased by about 10 percent or nearly 3,000 within 0.25
miles of station areas.
For the most part, changes in jobs follow similar patterns at three levels of geography, but there are interesting exceptions. Within 0.25 miles of BRT stations, jobs
in the Information, Real Estate, Management, Administrative, Education, Health
Care, Lodging/Food, and other sectors all increased by more than 10 percent, with
Management more than doubling. In contrast, between 0.25 and 0.50 miles from
stations, many of those same sectors lost jobs (Information, Professional, Management, and Administrative), while others grew in both distance-bands (Real Estate,
Finance, Education, and Health Care). Jobs in the Transportation and Arts/Entertainment/Recreation sector increased substantially between 0.25 and 0.50 miles of
BRT stations (160% and 130%, respectively). Retail gained slightly in both distance
bands but fell for the balance of the metropolitan area. A surprise based on other
research is that jobs within 0.25 miles of a BRT station fell in the Arts/Entertainment/Recreation sector and fell slightly in the Public Administration sector; on the
other hand, those sectors gained jobs between 0.25 and 0.50 miles of BRT stations.
Also surprising is that the balance of the metropolitan area did far better in gaining
jobs in Health Care, Lodging/Food, and Public Administration than station areas.
We surmise that the market is sorting jobs based on competition for BRT proximity. It may be that office uses are able to outbid Arts/Entertainment/Recreation for
locations closest to BRT stations (the sector lost nearly 120 jobs within 0.25 miles)
but many of those displaced jobs still located within 0.50 miles of BRT stations (the
sector gained 46 jobs between 0.25 and 0.50 miles).
We cannot say conclusively that there is a cause-and-effect relationship between
BRT locations and increasing concentration of certain kinds of jobs within 0.5 miles
of BRT stations; this will be the subject of future research.
46
3,461
27,737
Public Admin
Total
92
621
2,615
Lodging/Food
826
7,751
Other Services
Arts/Ent/Rec
71
81
Health Care
62
1,015
1,320
72
Education
61
291
Management
Administrative
Professional
54
55
2,366
Real Estate
56
1,285
Finance
52
53
442
1,133
484
1,769
427
Information
4849
51
Retail
Transportation
4445
Wholesale
Manufacturing
42
643
Construction
23
3133
813
475
NAICS Sector
30,582
3,379
717
2,919
707
9,095
1,249
2,042
633
2,221
488
1,447
1,557
517
1,844
269
465
520
513
Jobs within
Jobs within
0.25 Mile of
0.25 Mile of
EmX Station, EmX Station,
2010
2004
Utilities
22
NAICS
Code
10%
-2%
15%
12%
-14%
17%
23%
55%
118%
-6%
10%
13%
37%
7%
4%
-37%
-43%
-19%
8%
Change
in Jobs,
2004
2010
9,072
488
269
1,113
43
920
258
1,514
98
861
177
422
450
52
1,039
584
293
400
91
Jobs between
0.25 and 0.50
Mile of EmX
Station, 2004
9,085
552
294
1,099
99
1,395
303
1,031
75
811
182
524
389
135
1,073
499
174
314
136
Jobs between
0.25 and 0.50
Mile of EmX
Station, 2010
0%
13%
9%
-1%
130%
52%
17%
-32%
-23%
-6%
3%
24%
-14%
160%
3%
-15%
-41%
-22%
49%
Change
in Jobs,
2004
2010
100,031
1,361
4,009
7,445
1,421
9,363
13,983
5,456
1,631
2,751
1,947
2,105
1,550
2,608
14,551
5,313
18,690
5,696
151
94,779
2,084
3,926
8,341
1,526
12,102
15,800
4,441
1,733
2,597
1,516
1,766
1,360
2,260
14,021
4,742
11,685
4,696
183
Jobs
Jobs
Balance
Balance
of Metro
of Metro
Area, 2004 Area, 2010
Table 1. Change in Jobs with Respect to Distance from EmX BRT Stations, 2004 and 2010
-5%
53%
-2%
12%
7%
29%
13%
-19%
6%
-6%
-22%
-16%
-12%
-13%
-4%
-11%
-37%
-18%
21%
Change
in Jobs,
2004
2010
47
Also, this does not necessarily mean that BRT proximity confers a comparative
advantage for selected economic sectors. For this, we turn to shift-share analysis.
In particular, because we know where the jobs were located throughout the study
area in 2004 and 2010, we can compare shifts in share of jobs before and after the
introduction of the EmX.
Shift/share analysis is used to decompose employment changes in local areas. The
analysis identifies industries that have a comparative advantage in the local area. In
our case, we use the Eugene-Springfield metropolitan areas non-farming, forestry,
fishing, or mining jobs and apply shift-share analysis to determine the nature of
employment change with respect to being with 0.25 miles and between 0.25 and
0.50 miles of BRT stations in 2004 and 2010.
Shift-share analysis assigns the change or shift in the share or concentration of jobs
with respect to the region, other economic sectors, and the local area. The region
can be any level of geography and is often the nation or the state. In our case, where
we want to see whether there are intra-metropolitan shifts in the share of jobs by
sector, our region is the metropolitan area itself. The local area is often a city or
county or even state, but it can be any geographic unit that is smaller than the
region. Our local areas are the station areas within 0.25 miles and between 0.25 and
0.50 miles of the nearest BRT station; we call this the BRT Station Area. As shifts in
the share of jobs may vary by sector over time because of changes in economic sector mixes (there are now more high-tech jobs in the Eugene-Springfield metropolitan area than jobs in forestry), there is also an industry mix adjustment that we
call the Sector Mix. Using notations by the Carnegie Mellon Center for Economic
Development (no date), the shift-share formula is:
SSi = MAi + SMi + BRTi
Where,
SSi = Shift-Share
MAi = Metropolitan Area share
SMi = Sector Mix
BRTi = BRT Station Area shift
The Metropolitan Area (MA) share measures by how much total employment in
a BRT station area changed because of change in the metropolitan area economy
during the period of analysis. If metropolitan area employment grew by 10 percent
48
during the analysis period, then employment in the BRT station area would have
also grown by 10 percent if there is no BRT effect. The Sector Mix (SM) identifies
fast-growing or slow-growing economic sectors in a BRT station area based on the
metropolitan area growth rates for the individual economic sectors. For instance,
a BRT station area with an above-average share of the metropolitan areas highgrowth sectors would have grown faster than a BRT station area with a high share
of low-growth sectors. The BRT station area shift, also called the competitive
effect, is the most relevant component; it identifies a BRT station areas leading
and lagging sectors. The competitive effect compares a BRT station areas growth
rate in a given economic sector with the growth rate for that same sector at the
metropolitan area. A leading sector is one where that sectors BRT station area
growth rate is greater than its metropolitan area growth rate. A lagging sector is
one where the sectors BRT station area growth rate is less than its metropolitan
area growth rate.3
The equations for each component of the shift-share analysis are:
MA = (iBRT station areat-1 MAt /MAt-1)
SM
BRT = [iBRT station areat-1 (iBRT station areat /iBRT station areat-1 iMAt /
t-1
iMA )]
Where:
BRT station areat-1 = number of jobs in the BRT station area sector (i) at the
beginning of the analysis period (t-1)
BRT station areat = number of jobs in the BRT station area in sector (i) at the
end of the analysis period (t)
MAt-1 = total number of jobs in the metropolitan area at the beginning of the
analysis period (t-1)
MAt = total number of jobs in the metropolitan area at the end of the analysis
period (t)
MAt-1 = number of jobs in the metropolitan area in sector (i) at the beginning
of the analysis period (t-1)
MAt = number of jobs in the metropolitan area in sector (i) at the end of the
analysis period (t)
49
Table 2 reports only the BRT station area shift results for the areas within 0.25 miles,
between 0.25 and 0.50 miles, and within 0.50 miles of BRT stations. Figure 1 illustrates the BRT share for all the first two station area distances (0.25 and between
0.25 and 0.50 miles). The stacked bars in this figure allow us to see the individual
and combined effects of distance from BRT stations by economic sector.
Table 2. Shift-Share of Analysis of Job Change with Respect to
Distance from BRT Stations, 2004 and 2010
BRT Shift
0.25 Mile
BRT Shift
0.25-0.50 Mile
BRT Shift
0.50 Mile
Utilities
(38)
30
(8)
Construction
(8)
(14)
(22)
NAICS Sector
Manufacturing
(41)
(8)
(50)
Wholesale Trade
(103)
(10)
(113)
Retail Trade
118
59
177
69
87
156
Information
361
(86)
276
187
110
298
111
31
143
(7)
(7)
281
(43)
238
Administrative/Support/Waste Management/
Remediation Svcs
846
(341)
504
Educational Services
95
10
104
(615)
242
(373)
Arts/Entertainment/Recreation
(134)
55
(79)
26
(132)
(106)
91
23
114
Public Administration
(542)
(1)
(543)
Total
698
12
710
50
Services, and Accommodation and Food Service. In many instances, the positive
shift into the 0.25 mile band was greater than the negative shift out of the 0.25
0.50 mile band. While these are sectors that Belzer, Srivastava, and Austin (2011)
expect to be attracted to station areas generally, the fact that their positive shift is
so large toward the closer band suggests that, at least for BRT, the location advantage may not reach out as far as for rail modes.
There is also the reverse situation in which there is a negative shift in the closest
band but a positive one in the 0.250.50 mile band. This is the case with Health
Care and Social Assistance in which the shift away from the closer band was the
largest of all shifts, while the shift to the 0.250.50 mile band was the largest there.
Part of this may be explained by a major medical facility that opened in the late
2000s outside the BRT station areas.
Then there is the interesting case of Public Administration, which had the second
largest shift away from the closest distance band and there does not appear to any
offsetting shift in the 0.250.50 band. The explanation is likely severe local government budget cuts during the late 2000s that resulted in hundreds of jobs being cut
that were near BRT stations.
There are two other observations. First, of the combined shift in jobs toward BRT
station areas of 710 jobs, only 12 are in the 0.250.50 distance band. Thus, essentially, the entire overall shift in jobs favoring BRT station areas occurred within 0.25
miles of them. Second, the BRT system may have a resiliency effect. Where the
Eugene-Springfield metropolitan area as a whole lost jobs between 2004 and 2010,
jobs were actually added within 0.25 miles of BRTs stations.
52
2. In cities where the real estate market is not already strong, an active transit
agency with a TOD program and/or active community development organization is critical.
3. Real estate developers and owners view permanence as an important factor
for building around a BRT system. A key advantage of rail is that once the
investment has been made, the real estate industry can usually rely on its
permanence over the many decades it takes to maximize profits from highdensity investments at or near those stations. However, even in the cities with
a relatively low level of infrastructure, BRT may be viewed as permanent when
there is a clear long-term commitment by the transit agency. In the case of
EmX, this commitment includes substantial capital investment in providing separated lanes for exclusive BRT use and light-rail-like transit stations.
4. The transit corridor must be amenable to high-density development, so the
route needs to assure this opportunity. Corridors placed in areas without
major employment or housing destinations are not likely to attract development, regardless of mode.
5. Providing financial incentives for TODs at BRT stations does not appear to
be as important for attracting developer interest. Developers are much more
interested in an expedited permitting or rezoning process, as time is a critical
factor in making development projects financially viable.
One implication is that BRT may provide for many more opportunities for smaller
metropolitan areas to serve numerous job sectors. We note that an urbanized population of about one million appears to be the smallest capable of supporting light
rail, with Salt Lake City being an example. Light-rail-like benefits may be achieved
only in smaller metropolitan areas through BRT. Moreover, within metropolitan
areas that have light or heavy rail, costs may prohibit their expansion. BRT could
be the next-generation solution to increase multimodal options. In either case, the
BRT results for Eugene-Springfields EmX may provide metropolitan planning organizations with a rationale for investing in BRT for economic development reasons,
especially in situations where rail does not pencil.
We hope our work stimulates more research in this area. In the case of EugeneSpringfield, we find that the job growth occurred near BRT stations over a short
period of time where otherwise the metropolitan area lost jobs as a whole. Further
research is needed to determine cause-and-effect relationships between BRT stations and employment change, whether there is variation among economic sectors, whether employment shifts occur in the short term as well as the long term,
53
and the extent to which local economic benefits improve with respect to BRT,
among others. There is also the question of whether and to what extent BRT affects
residential location patterns.
It would also be important to know whether BRT technologies have different economic development and residential location outcomes. Most light rail systems, for
instance, use the same system design and mechanical technologies. In contrast,
BRT systems can vary widely based on rail, station/platform, carriage, signalization,
and other features. Success with EmXs BRT flavor may not be replicated with other
BRT flavors.
We hope this article serves as a starting point for advancing discussion on BRT as a
viable economic development tool.
Endnotes
Compiled from http://www.ltd.org/search/showresult.html?versionthread=45a4b8
3927fba5cb751c741bf4ac81e3.
1
Acknowledgments
The authors acknowledge that support for research leading to this article was
provided by the U.S. Department of Housing and Urban Development through its
Sustainable Communities program and the National Institute for Transportation
and Communities.
54
References
Anas, A., R. Arnott, and K. A. Small. 1998. Urban spatial structure. Journal of Economic Literature 36(3): 14261464.
Belzer, Dena, Sujata Srivastava, and Mason Austin. 2011. Transit and Regional Economic Development. Oakland, CA: Center for Transit-Oriented Development.
Boarnet, Marlon. 1997. Highways and economic productivity: Interpreting recent
evidence. Journal of Planning Literature 11(4): 476486.
Bogart, William T. (1998). The Economics of Cities and Suburbs. Upper Saddle River,
NJ: Prentice Hall.
Carnegie Mellon Center for Economic Development. Undated. Methods of
regional analysis: Shift-share. Pittsburgh, PA: H. John Heinz School of Public
Policy and Management, Carnegie Mellon University. Accessed from http://
www.andrew.cmu.edu/user/jp87/URED/readings/Shift_Share.pdf.
Cervero, R., S. Murphy, C. Ferrell, N. Goguts, Y. Tsai, G. B. Arrington, J. Boroski,
J. Smith-Heimer, R. Golem, P. Peninger, E. Nakajima, E. Chui, R. Dunphy, M.
Myers, S. McKay, and N. Witsenstein. 2004. Transit-Oriented Development in
the United States: Experiences, Challenges, and Prospects. Washington, DC:
Transportation Research Board.
Ciccone, Antonio, and Robert E. Hall. 1996. Productivity and the density of economic activity. American Economic Review 86: 5470.
Diaz, R. B., M. Chang, G. Darido, M. Chang, E. Kim, D. Schneck, M. Hardy, J. Bunch,
M. Baltes, D. Hinebaugh, L. Wnuk, F. Silver, and S. Zimmerman. 2004. Characteristics of Bus Rapid Transit for Decision-Making. Washington, DC: Federal
Transit Administration, U.S. Department of Transportation.
Glaeser, Edward. 2011. Triumph of the City How Our Greatest Invention Makes Us
Richer, Smarter, Greener, Healthier, and Happier. New York: Penguin Books.
Graham, D. J. 2007. Agglomeration, productivity and transport investment. Journal
of Transport Economics and Policy (JTEP) 41(3), September: 317343. www.
ingentaconnect.com/content/lse/jtep/2007/00000041/00000003/art00003.
Summarized in OECD/ITF Discussion Paper 2007-11, www.internationaltransportforum.org/jtrc/DiscussionPapers/DiscussionPaper11.pdf.
55
56
Bruce Appleyard (bappleyard@cfaconsultants.com) is a Principal of the planning, research and urban design firm CFA Consultants. He has more than 20 years
of experience on the intersection between transportation, land use, urban design,
and environmental quality in support of a diverse range of sustainability and livability objectives.
57