Charlotte’s LYNX Line: A Preliminary Assessment

8/14/2019 Charlottes LYNX Line: A Preliminary Assessment

1/26

Charlottes LYNX Line:

A Preliminary Assessment

DAVID T. HARTGEN, PH.D., P.E.

September 2008

P O L I C Y R E P O R T


2/26

Charlottes LYNX Line:

A Preliminary Assessment

D a v i D T. H a r T g e n , P H .D . , P.e .

S p tm b , 2 0 0 8

3 itoducto d Method

4 i. rdeshp

4 Wht s the bg pctue o Chlotte commutg?

4 How s LYnX deshp dog?

5 Who s usg LYnX?

7 Hs LYnX coced uto des to swtch to tst?

7 Wht bout gsole pces?

8 How does LYnX deshp compe wth oecsts?

9 ii. Costs

9 How much dd LYnX cost to buld?

10 Wht s LYnX costg to opete?

11 Wht e LYnXs totl costs?

12 How much s LYnXs cost pe tp d subsdy pe tp?

12 iii. impcts

12 Hs LYnX ected tfc cogesto?

13 Hs LYnX educed commutg tel?

13 Hs LYnX educed polluto?

14 Hs LYnX ected el estte deelopmet?

18 Hs LYnX ected esdetl deelopmet?

20 iv. Beeft-Cost assessmet

20 How e tspotto beefts d costs detemed?

22 Wht e the beefts d costs o the LYnX le?

24 ackowledgmets

24 about the autho

The views expressed in this report are solely those o the author and do not necessarily

refect those o the sta or board o the John Locke Foundation. For more inormation,

call 919-828-3876 or visit www.JohnLocke.org. 2008 by John Locke Foundation.


3/26

Execuve Suy

Final LYNX construction costs are about $521.9

million, about 130 percent above the initial esti-

mate ($227 million) LYNX operating costs are about $9.22 million/

year, but data are incomplete and likely to be

low. Revenues are averaging about 31 percent

o operating costs. The average are (including

discount ares, transers, and on-are rides) is

about $0.58.

Average weekday ridership initially reached

12,457 (Dec. 2007), about 37 percent higher than

the ocial orecast. However, the initial orecast

did not consider present gasoline prices, which i

included would have produced a higher estimateo initial ridership.

Since December 2007 ridership is up about 23%

(Dec. 2007-July 2008), but CATS (bus) ridership

has risen slightly aster, 24%. LYNX ridership is

about 19 percent o total CATS trac.

About 1/3 o the rise in LYNX ridership is based

on better transit service, and about 2/3 rom

continuing increasing gas prices, regional growth

and economic circumstances.

About 64 percent o trips, but 74 percent o ticket

sales, are or 1-way or round-trip adult tickets,suggesting continuing trial occasional use.

About o the LYNX ridership is shited rom

buses, about is diverted rom cars, is rom

walk, drop-o, etc.

About 19% o the auto driver trac diverted to

LYNX, totaling about 838 weekday trips, is rom

vehicles with South Carolina tags.

LYNXs cost/trip is about $6.90, o which 91% is

subsidized. In other words, taxpayers are paying

$6.30 per LYNX trip, or about 91 percent o the

cost, and the LYNX rider is paying about $0.60,

or about 9 percent o the cost.

LYNX has diverted about 0.079 percent oregional travel, but about 4 percent o peak-hour

travel in the corridor.

Street trac volumes, although incomplete,

show trends that counter the conclusion that

LYNX has visibly aected auto trac in the

corridor.

Impacts on congestion are too small to be seen

in street trac, but are computed to be about

minute in travel-time savings or corridor drivers

remaining on the street system.

LYNX's impact on air quality is about 0.05-0.09percent o regional emissions, too small to be

observable.

The South Boulevard corridors incremental

growth o commercial development, beyond

background growth, is about $50.6 million over

20 years. Its incremental growth o residential

development, beyond background growth,

is about $198.8 million over 20 years. These

estimates, based on observed construction

between 2005 and 2007 and the economic value

o commercial and residential space, are muchlower than the $1.86 billion estimate based on

announcements.

Benet-cost analysis shows total costs at $706.4

million, compared with quantiable benets o

$480.2 million, yielding a benet-cost ratio o

0.68. I just the local costs ($400.4 million) are

considered, the benet-cost ratio is about 1.20.

This means that i the local governments had to

pay the ull cost o the project, they could not

justiy it on benet-cost grounds.

CharlottE S lYNX l iNE : a PrEl imiNarY aSSESSmENt | executive SuMMARY

(LYNX trains at I485/South Boulevard Station)


4/26

CharlottE S lYNX l iNE : a PrEl imiNarY aSSESSmENt | BAckgRound

Ater nearly a decade o planning and con-struction, Charlottes light rail line, termed LYNX,

opened on November 27, 2007. The 9.6-mile, 15-sta-

tion line is located along South Boulevard and serves

the South Corridor (one o ve potential rapid

transit corridors identied by city ocials) rom the

central business district to Carolina Place Mall, near

Interstate 485 (Map 1). Since its opening numerous

articles and observations have appeared, noting the

basic statistics o the line its ridership, parking lot

use, costs o construction, and potential development

impacts. But no comprehensive assessment o the

lines perormance, the source o its ridership, or its

impact has yet been prepared.

This assessment is intended as a preliminary

straightorward overview o LYNXs perormance,

based on about eight months o operation. While not

lengthy, this time period is viewed as long enough todiscern most o the impacts, at least in preliminary

orm. The assessment is based on the inormation

presently available (generally through July 2008,

with some August 2008 statistics). It uses methods

common in transportation perormance assessment.

We hope that this assessment will provide bench-

marks and actual inormation against which later

comparisons can be made. In the years ahead these

impacts will come into ocus more clearly, and the

ndings can then be updated appropriately.

The methods used to prepare this report are

straightorward. Largely, they are based on actual

inormation provided rom published reports,

CATS, and other agencies, on-site observations,

and summaries o real estate databases or Web sites.

Data are then summarized into straightorward

spreadsheets and tables or charts. Where external

statistics are required, or instance air pollution emis-

sion rates, they are drawn rom national sources.

The primary sources o all inormation are listed.

Individuals interested in viewing the supporting

spreadsheets behind the report are invited to contact

the authors.

inducn nd med

Map 1. Mecklenburg County and the South Boulevard Transit Corridor.

(LYNX train outbound, near I485/South Boulevard Station)


5/26

CharlottE S lYNX l iNE : a PrEl imiNarY aSSESSmENt | RideRShip

What is the big picture o commuting in

Charlotte? The Charlotte urbanized area has a population

o 824,000 and about 425,000 resident workers1;

another 100,000+ workers enter Mecklenburg

County daily rom surrounding counties. The

total regional travel (or a slightly larger area) is

about 20 million vehicle-miles daily2.

The ollowing table summarizes the commuting

prole or Charlotte urbanized area resident

workers. About 2.6 percent o the urbanized area

commuters, about 11,092, use public transporta-tion as their primary commuting mode. But

almost twice as many, 17,232, work at home.

The same source also shows that:

Just 2.9 percent o all commuters have no

vehicle available.

The average travel time to work is 24.9

minutes, but transit commuters average 41.6

minutes, compared to 23.7 minutes or auto

drivers and 29.2 minutes or carpoolers. 64% o transit commuters are Arican

American, 9.5% are Latino.

75% o transit commuters earn less than

$25,000.

These ndings generally mirror those o on-board

surveys o CATS riders3 (62% Arican American,

8% Hispanic, 46% income under $20,000). This

data, however, does NOT contain inormation

or LYNX riders, or which an on-board survey

has not yet been released.

How is LYNX ridership doing? In the rst ull month o operation (December

2007), LYNX weekday ridership4 averaged 12,457

unlinked trips5, about 37% above the ocial

orecast. But the prior estimates were prepared

six years ago and do not include the recent run-up

in gasoline prices, which encourages more trac

I redone with current gasoline prices, ridership

estimates would be somewhat higher6.

Since December 2007, LYNX average weekday

ridership has risen rom 12,457 to about 16,8957

about 35 percent.

However, as a share o trac, LYNXs share hashovered between 16.5 and 20.3 percent, and ha

averaged about 19 percent o all CATS ridership

This share has not signicantly increased since

December 2007, and seems to be leveling o.

Further, Saturday-Sunday-Holiday traic ap

pears to have sotened somewhat. In the initia

ull month o service (December 2007), abou

30 percent o the total LYNX ridership was Sat

urday/Sunday/Holiday trac. However by July

2008 this portion had allen to 15 percent. By July

Mode to Work Number oWorkers

Percent

Drive Alone 333,744 78.6

Carpool 51,626 12.2

Public Transportation 11,092 2.6

Walk 5,413 1.3

Bike 5,705 0.2

Taxi/Motorcycle/Other 1.2

Work at home 17,232 4.1

Total, Workers 16+ 424,812 100

Charlotte Urbanized Area Commuting, 2005

i. rdesp

L YNX S hare of All C ATS Trips

19.519.019.120.3

16.5

19.519.216.6

0.0

5.0

10.0

15.0

20.0

25.0

Ave rage Daily Rides,LYNX

12,871

10897

7,460

4613

0

5,000

10,000

15,000

20,000

Ave Weekday Rides,LYNX

Ave Saturday Rides LYNX

Ave Sunday Rides LYNX


6/26

2008 average Sunday ridership had allen 39 per-

cent to just 4,613, and Saturday trac had allen

15 percent to 10,897. Clearly a signicant portion

o the initial ridership was recreational and try-

out trac, which has subsided somewhat. August2008 weekend trac rebounded due to estivals

and sports events, and some pick-up o weekend

trac is likely in the all o 2008. Nevertheless,

LYNX usage seems to be evolving into primarily

weekday trac8, at about 19 percent o CATS

ridership.

LYNX ticket sales and revenues rom ares are

averaging about $250,000 per month, or just

over $3 million per year. This puts its weighted

average are (revenue divided by ridership) at

about $0.58 and its arebox ratio (percent o

expenses covered by ares) at about 30 percent.

LYNX ticket sales also indicate that about o

the LYNX ridership is rom transers or did not

purchase a pre-ride ticket.

Who is using LYNX? No on-board survey o LYNX riders has yet been

released. This would be needed to determine the

demographics o LYNX riders, their originating

modes, and locations. Thereore, the assessment

must initially ocus on general trends and sup-

porting inormation.

In the rst ull month o operation (December

2007) total reported LYNX ridership was 356,000

rides (one-way unlinked trips). But ridership on

the remainder o the CATS system declined by

347,706 riders, almost the same as the LYNX

increase.

Details o ridership trends in the ollowing graphic

show that about o the initial average weekday

LYNX trac was ormer bus riders.


LYNX R ide s by D a y o f We ek

249140

371690

43588

64355

44760

23065

0

100000

200000

300000

400000

500000

Dec 07 July 08

Sun/Hols

Saturdays

Weekdays

(LYNX train crossing 7th St, downtown Charlotte)

LYNX Weighted Average Fares

0

200000

400000

600000

800000

1000000

1200000

Jan 08 Feb 08 March 08 Total 1s t Q

Total Sales, $

Reported Total Trips$ 0.58

$ 0.64$ 0.54$ 0.57

CATS L ong Term Rider Trends

438,340356,000

2,305,5021,864,080

1,867,162

1,508,080

1,577,430 1,855,786

0

500000

1000000

1500000

2000000

2500000

l i

LYNX

Total

Total- LYNX

(The I485/South Boulevard bus transer station, at the end o theLYNX line)


7/26


Even discounting historic declines in ridership

between October and December, this suggests

thata considerable portion o the initial ridership was

prior bus riders who shited to LYNX9.

Further evidence o considerable shiting rom

buses comes rom detailed route data showinglarge ridership drops on CATS routes that paral-

lel the new LYNX line, and sharp increases on

eeder routes going to LYNX stations. Paral-

lel routes dropped about 3,300 weekday trips

between November 2007 and December 2007,

while eeder routes increased about the same

amount. Then, both groups experienced growing

trac during 200810.

About 64 percent o trips are adult (1-way or

round trip), 6 percent senior (1-way or roundtrip), 11 percent youth (1-way or round trip),

and 14 percent 7-day. However, 76 percent o

ticket sales (not trips) are or 1-way or round trip

adult tickets. This data, based on ticket sales or

Jan-Mar 200811, imply considerable trial and

occasional travel.

Average Weekday Ridership, CATS Routes Near LYNX

Nov 07 Dec 07 Jan 08 Feb 08 Mar 08 Apr 08 May 08 June 08 July 08Pct Ch,Dec-July

Feeder 3,079 5,367 5,236 5,665 5,751 6,197 6,609 7,054 7,476 39.3

Parallel 8,738 5,440 5,447 5,614 5,536 5,820 6,249 6,209 6,357 16.9

Both 11,816 10,807 10,683 11,279 11,287 12,017 12,858 13,262 13,833 28.0

CATS B us R oute Rider Trends

,, ,

, ,,

, ,,

,

,,,,,,

,

,

,,,,,,,,

,

02,0004,0006,0008,000

10,00012,00014,00016,000

- - - - - - - --

Both

Feeder

Parallel

LYNX opens

(LYNX customers using ticket vending machine)

Average Weekday Ridership

0

20,000

40,000

60,000

80,000

100,000

AveWeekday

Rides, LYNX

AverageWeekdayRidership,Bus Service

LYNX Ticket S ales by Type

1%

3%

64%

4%

11 %

12%

2%

3%

7-Day

1-Day

1-Way Adult

1-Way Sr

1-Way Youth

RT Adult

RT Sr

RT Youth

Based on Ticket Sales for Jan 08-March 08. Doesnot include Transf ers or Off -Site Sales.

P e r ce nt o f L Y NX T ri ps by T i ck e t T y pe

14%

4%

47%

3%

8%

17%

3%4%

7-Day

1-Day

1-Way Adult

1-Way Sr

1-Way Youth

RTA dult

RT Sr

RTYouth

Based on Trips Implied by Ticket Sales for Jan 08-March

08. Does not include Transfers or Off-Site Sales.


8/26


9/26

However, average weekday LYNX traic has

increased about 35.6 percent, about the same as

gasoline price rises through July 200817.

This suggests that almost 1/3 o the growth in

weekday LYNX trac (36 percent 23 percent)is attributable to the improved service that LYNX

oers relative to other bus service, and about 2/3 o

the growth is attributable to underlying pressures

o rising uel prices, increasing infation, economic

activity, and bus service (revenue vehicle-miles).

Without this background pressure, LYNX average

weekday ridership would likely be about 12-13%

higher now than in December 2007.

Recently (August-September 2008) gasoline pric

es declined somewhat then rebounded sharply as

Hurricane Ike approached. Ridership statistics

or August 2008 show that weekday trac was

slightly lower than July 2008, but weekendtrac was higher. This suggests that i gasoline

prices continue to moderate, some sotening o

both bus and LYNX trac is likely. Indeed, ac

cording to CATS, August 2008 average weekday

ridership or the system as a whole was 83,758

down about 7.5 percent rom about 90,539 rides

in July 2008.

How does LYNX ridership compare withorecasts? Forecasts o transit ridership are typically made

only or average weekday trafc, and do not include

weekend or holiday trac. They are based on

computer models that use 20-year orecasts o

population and employment, costs and travel times

or typical trips, auto availability, incomes, captive

and choice markets, and congestion on roads18.

Forecasts o LYNX perormance go back to the

late 1980s. The earlier orecasts o LYNX trac

were prepared with simplied models and then


Monthly R ides

1,508,080

1,867,162

356,000 438,340

0

500,000

1,000,000

1,500,000

2,000,000

Dec07 Jul08

LYNX

Total - LYNX

Increase in Transit and Gas Prices

23.1 23.8

34.0 35.6

0.0

10.0

20.0

30.0

40.0

LYNX Total - LYNX Midgrade

Gas Price

AAA

Ave

Weekday

Rides, LYNX

Percent Increase, Dec 07 - July 08

(Inside the LYNX train, inbound, near Archdale)

(CATS downtown Transportation Center, serving bus and LYNX trafc

LYNX Traffic And Cost Forecasts

5535

15500

12880

18100

17650

1810017900

25760

21100

14000

91009100

16895

16140

16239

12689

1480714246

11930

12457

6691

40.5

227

348.2

370.8385.9

398.7

426.9

462.7

521.9

331.1

0

5000

10000

15000

20000

25000

30000

- l- - - - - - - r- - - - F - r- r- - -l-

i

0

100

200

300

400

500

600

Est.(2005) 2025

Rides/day

Est.OpeningDayRides

Actual Ridership

Est.Cost$M

Opening Day

Full Funding

Final DesignPrelimEng


10/26

were later rened with better models. The early

orecasts put the estimated average weekday

ridership at 5,535/day, and the construction cost

at just $40.5 million.

More extensive studies in the late 1990s raised the

2025 weekday ridership estimate to 14,000 rid-

ers, and the ocial cost estimate to about $227

million. As the project neared ederal approval,

cost estimates and ridership orecasts increased

to about $371 million and 25,700 rides. But ater

approval, more careul assessments lowered the

ridership estimates to 18,100 (rst year 9,100),

but cost estimates continued to rise. The latest

orecasts were prepared in 2001-2002, or the

initial (opening year) and year 2025.19

As noted above, these orecasts were prepared

prior to the surge o gasoline prices. I revised to

account or that, they would undoubtedly yield

higher orecasts20. But they also assumed higher

downtown employment, higher parking rates, and

a ull 4-line system; i corrected, these assumptions

would probably yield a lower orecast.

Summary

Consolidating these observations, the ollowingchart summarizes LYNX ridership by source.

Initially, about 52 percent o trac was shited

rom other transit routes, and about 17.4 percent

was drive-and-park; now, about 47.8 percent o

trac is shited rom prior transit routes, and

about 26.7 percent is rom drive-and-park. The

remainder o the trac is rom walk-up, drop-o,

outbound transers, and other sources.

LYNX ridership is averaging about 19 percent o

total CATS system ridership. LYNX ridership has

grown about 23 percent since December 2007,

about the same as non-LYNX bus ridership.

About 2/3 o the growth in average weekday

ridership is likely attributable to rising gasoline

prices, economic conditions, and more transit

service. This means that ridership is likely to

be vulnerable to declining gasoline prices or a

strengthening economy.

LYNX average weekday ridership is higher than

initially orecast, but those orecasts did not ore-

see the recent sharp rise in gasoline prices.

CharlottE S lYNX l iNE : a PrEl imiNarY aSSESSmENt | coStS

LYNX Ave Weekday R ides by Pr imary

A c c e s s

21694510

6507

80773781

4308

0

5000

10000

15000

20000

Dec 07 July 08

Walk, Drop-off,

OtherBus Transfer

Drive and Park

12,457

16,895

ii. Css

How much did LYNX cost to build?

When the LYNX South Boulevard corridor linewas initially discussed, the projected construction

cost was $40.5 million, in 1987 dollars.

Ocial projected costs rose to $227 million in

July 1998. This was the estimate used to put the

project to the voters in the all o 1998. As part o

the initial estimate, ederal unds were assumed

to cover hal o the construction costs, and state

and local unds would be responsible or the

remaining hal.

Several outside observers at the time, notablyWendell Cox21, concluded that these costs were

low. Estimated nal total costs near $500 mil-

lion; this turned out to be within 5 percent o the

actual value.

Over the nine years that it took to plan and build

the LYNX line, construction costs increased sub-

stantially. These increases have been attributed

to infation, changes in line length (actually a

cost reduction), rising construction costs, exten-


11/26

10

sions o opening year, better engineering details,

and inclusion o additional items. Funding rom

various sources changed. The ederal share was

initially 50 percent, but in 2004, the Federal Tran-

sit Administration capped ederal unding or the

LYNX line at $199 million. Similarly, the State

unding was also limited to $107 million in 2005.

Cost increases ater these dates were born by lo-

cal government. The nal ocial estimate was

$462.7 million, not including o-budget projects

(sidewalks, intersections, utilities, etc.) unded by

the City. These total about $59.15 million22.

Final cost gures have not been released, but they

are expected to total about $521.85 million, based

on published costs or various work.

So the projects construction costs grew rom

$40.5 million to $521.85 million in a span o 19

years, rom concept to reality. Over the nine-year

approved portion o the study (1998-2007), cost

increased rom $227 million to $522 million

about 130 percent. This is somewhat higher a cos

overrun than the average, 104 percent, identied

by Flyvbjerg23 or major transit projects around

the world.

What is LYNX costing to operate? The LYNX operating expense estimate or Cal

endar 2008 is $9.226 million. This estimate

uses actual expenditures or ve months (Janu

ary-May), partial expenditures or June-July, and

estimated expenditures or August-December. In

spite o these caveats, this is the best source or

estimates o operating costs24.


LYNX Construction Cost History

Date oEstimate

FederalFunds, $M

State Funds,$M

Local Funds,$M

Other CityFunds, $M

Total, $M

Jun-89 20.25 10.12 10.13 40.5

Jul-98 113.5 56.75 56.75 227

Nov-00 167 82 82 331

Nov-01 174 87 87 348

Nov-02 185 93 93 371

Nov-03 193 97 97 387

Nov-04 199 100 100 399

Nov-05 199 107 121 427

Apr-07 199 107 156.7

Jul-08 199 107 156.7 59.15 $ 521.85

LYNX Construction Cost T rends

0

100

200

300

400

500

600

Local-City

Local

State

Federal

10

10

20

$521.9

$462.7

$227

LYNX 2008

OperatingEx pens es

Benefits

17%

Fuel

5%

Insurance

21%

Maintenance

3%

Salary

47%

Other

7%

Taxes

0%Administration

0%

$ 9 .23 M Total


12/26

11

The largest portion o the operating costs or

the LYNX system unds the work orce behind

the operations. About 64 percent o the budget

is allotted or the salaries and benets o LYNX

employees.

Insurance ($1.948 million) is the second-largest

item in the operating expense listing, even ex-

ceeding benets and maintenance.

Some o the operating cost estimates appear to

be on the low side (and some o the data is pro-

jected), so these expenditures could be subject

to change and are likely to increase. This is par-

ticularly true or administrative and maintenance

costs. FTA justication or the projects was based

on operating costs averaging about $17 million

per year.

What are LYNXs total costs? Although transit lines oten have longer lietimes

than 20 years, this is a prudent estimate since

many components would probably need to be

replaced in that timerame. This is also the ore-cast period or LYNX ridership. Twenty years is

also the usual lietime assumed or major road

improvements. To compare construction and

operating costs we use 20 years as the eective

lietime, understanding that some components

may wear out aster and some slower.

LYNXs annualoperating and capital costs pres-

ently total about $32.3 million, in 2008 dollars.

Capital costs are $23.1 million, about 2/3 o the

total.

Over 20 years, LYNX capital and operating costs

total about $706.4 million. The local share o

these costs is $400.4 million.25


LYNX 2008 Operating Costs

Administration 29,406

Benefts 1,560,096

Fuel 489,517

Insurance 1,948,041

Maintenance 252,681

Other 602,374

Salaries 4,339,013

Taxes 4,640

Total 9,225,768

LYNX Operating and Capital Costs

Total 20 yrs Annual Basis

Capital Cost, $M Cap-Federal 199.0 9.950 Total/20 years

Cap-State 107.0 5.350 Total/20 years

Cap-Local-LYNX 156.7 7.835 Total/20 years

Cap-Loc-City 59.2 2.958Total O-Budget/20 years

Total 521.85 26.095

Operating Costs Operating-Local 184.52 9.226CATS Est.2008.Likely to rise

Total Costs (2008$) Total 706.37 32.321

LY N X C on s truction an d O peratin g C os ts , 2 0 Years

199.0

107.0

156.7

59.2

184.52Cap-Federal

Cap-StateCap-Local-LYNX

Cap-Loc-City

Operating-Local

Operating C os ts

Capital Costs $ 521.9

Operating Cost $ 184.5

Total Cost $ 706.4


13/26

1

O this, $521.9 million is the initial capital cost,

consisting o $462.7 million in ederal/state/local

unds, and an additional $59.2 million in local

unds expended or ancillary work on nearby

streets, utilities, sidewalks, etc.

The current estimated operating budget is about

$9.226 million annually or about $184.5 million

over 20 years. This estimate, in 2008 dollars, is

likely to be low since operating costs or transit

systems have been increasing more rapidly than

either ridership or service levels.

How much does LYNX cost per trip?

Costs per trip are summarized below. Using theoperating budget, the cost per trip is about $1.97.

Adding the capital cost, the total (capital and

operating) cost per trip is about $6.9026.

The nominal are, $1.30, is not the average

are paid, since many trips use transers or special

ares (commuter, elderly, student, etc.). Most U.S.

systems have weighted average ares about o

the nominal ares.

For LYNX, the weighted average are is estimated

to be about $0.60/trip. This estimate is based on

2006 CATS data showing a weighted average are

o $0.55/trip, and $0.58/trip or the rst quarte

o 2008.

Using a weighted average are o $0.60/trip, the

operating subsidy is about $1.37, implying a are

box ratio (the portion o costs paid by riders) o

about 31 percent. However, using the total cos

per trip, the total subsidy is actually $6.30 per trip

implying a arebox ratio o about 8.7 percent.

In other words, taxpayers are paying $6.30 per

LYNX trip, about 91 percent o the cost, and

the LYNX rider is paying about $0.60, about 9

percent o the cost.

CharlottE S lYNX l iNE : a PrEl imiNarY aSSESSmENt | iMpActS

LYNX Cos t per Tr ip

0.60

$4.93

$1.97

$-

$1.00

$2.00

$3.00

$4.00

$5.00

$6.00

$7.00

$8.00

Cost per Trip Wtd Avg Fare

Capital Cost/Trip

Operating cost/trip

Total $ 6.90

Operating Subsidy $ 1.37

Op + Capital Subsidy $ 6.30

III. Impacts

Has LYNX aected trafc congestion? LYNX was initially justied primarily on its abil-

ity to reduce the growth o (not reduce) uture

congestion, by providing a choice or auto driv-

ing. So, one way to judge LYNXs eect on trac

is to determine i the growth o congestion has

been reduced, looking at trends in trac counts

in the South Boulevard corridor.

The data come rom counts conducted every two

years by CDOT and NCDOT. Unortunately,

the only counts available or 2008 are those or

Interstate 77, which were conducted in April.

However, sucient count inormation was avail-

able to determine basic pre-LYNX trends.

Interestingly, I-77 trafc volumes have declined

slightly over the past six years, possibly due to

slowing economic growth. Average trafc volumes

in 2003 were 157,286, compared with 153,622 in

2008. The interim 2004 and 2006 data show that this

drop pre-datesLYNXand therefore is not related to

the opening of LYNX27.

Trafc volumes for other parallel arterials show a

slight increase over the six years, but a drop between

2004 and 2006. South Boulevard/U.S. 521, which

runs along the LYNX route, shows a slight decrease

in increase in trafc over four years, but an increase

from 2002 to 2004. These gures indicate that the

road trafc on the major routes parallel to LYNX was

probably on the decline before the LYNX System

was even opened28.

Trafc on cross streets might be expected to increase

with the arrival of LYNX, since those streets would


14/26

1

probably be the primary feeder streets to LYNX

stations. However, the data show a moderate increase

in trafc, before LYNX, suggesting that further

increases might be part of this trend rather than at-

tributable to LYNX.

In short, the trafc volume data, while incomplete,

show trends that counter the conclusion that LYNX

has affected trafc in the corridor.

Has LYNX reduced commuting travel? Another way to view the impacts of LYNX on trafc

congestion is to look at how much commuting the

LYNX line has saved or taken off Charlotte roads,

relative to trafc volumes.

Trafc diversion to LYNX can be determined from

the number oftripsdiverted from auto driving. The

numbers come from counts of vehicles at LYNX

parking lots. These estimates slightly underestimate

impacts since walk-up, drop-off, and some carpool-

ing trafc is not included. Mileage saved is then

determined by calculating the distance from each

station to the central business district (assuming no

LYNX trips to intermediate destinations; if thosetrips were included, the numbers would be lower).

Auto drivers diverting to LYNX save, in total, about

15,944 daily vehicle-miles by using LYNX or about

7,972 vehicle-miles in the morning or afternoon peak

hour. This is about 0.08 percent of regional travel,

0.4 percent of peak hour travel, but 4.2 percent of

corridor peak hour travel. Even within the corridor,

therefore, the effects are small, and are probably not

measurable in trafc ows29.

LYNX also produces small savings in travel timefor auto commuters who can save travel time since

other trafc has been diverted. Using congestion-

delay curves, this effect is estimated to average

about 0.27 minutes per trip for peak-hour drivers in

the corridor.

These estimates do not consider possible conges-

tion increases near stations or on portions of South

Boulevard, or additional trafc that might move in

to take the place of diverted trafc.

Has LYNX reduced air pollution? Estimates o air quality improvements attributable

to LYNX are also derived rom diverted trac

(parked cars at LYNX stations). These counts

are multiplied by mileage saved (distance to the

Charlotte CBD), then by emissions rates, to deter-

mine emission reductions. This measure slightly

underestimates savings, since it does not include

walk-up and drop-o trac, or some bus transers

that might have been prior auto drivers.


South Corridor Trafc Counts* (Number o sites) 2002 2004 2006 2008

I-77 (7 sites) 157,286 157,000 155,571 153,622

Other Parallel Arterials (20 sites) 24,842 25,800 25,053

South Boulevard (5 sites) 23,500 26,090 23,200Cross Streets (12 sites) 18,029 22,075 23,325

LYNX Corridor Traffic Count Trends

, , ,,

,,, ,,,

0

20,000

40,000

60,000

80,000

100,000

120,000

140,000

160,000

180,000

2002 2004 2006 2008Y e ar

l

I-77

Other Parallel Arterials

South Boulevard

Cross Streets

LYNX opens

(Trafc at the I -77 and Clanton Rd area.)


15/26

1

Overall reductions in emissions attributable to

LYNX are:

Emissions savings attributable to LYNX constituteless than 1/10 of 1 percent of major compounds.

Savings of this magnitude would not be measurable

in the regions air quality monitoring programs.

On a cost per ton saved basis, LYNX costs $18,439

to $9,418,000 depending on the compound. This

compares with about $55/ton saved for typical pol-

lution control actions.

Has LYNX aected real estate development? Much has been written about land use in the

South Corridor. The City o Charlotte recently

estimated30 that the South Corridor has had abou

219,512 square eet o commercial development(built or under construction) between 2005 and

200731, and that housing units increased by 1,175

units (built or under construction), together total

ing $219 million in development value. Anothe

408,000 square eet o commercial developmen

and another 6,406 housing units have been

announced with an estimated value o $1.569

billion. The report thus asserts that the South

Corridor has had (or will have) $1.86 billion in

commercial and residential development. Thereport does not say but certainly implies tha

this growth is largely attributable to the LYNX

line and thereore justies its $462 million ex

penditure. The report also notes that assessed

valuation (presumably dened as tax value or

property tax purposes) increased 52 percent in the

corridor since 2000, compared with 42 percen

or the city as a whole. It concludes by estimating

the tax take rom this increase at about $24.1

million annually; this number is not orecast

orward over LYNXs expected lietime. So, theCity report ocuses on units constructed, under

construction, and announced, then converts thi

into taxable values and tax takes rom property

tax rates.

The report errs in its assessment in a number o

basic ways:

First, the report contains summation errors that

undermine its presumed data accuracy.

Second, it treats announcements as i they


Chemical Compound Meck. Co.Total1LYNX

SavingsPercento Co. Cost/Ton Saved*

CO2(Carbon dioxide, tons/ day) 9646.85 7.66 0.079 $ 18,439

CO (Carbon monoxide, tons/ day) 302.92 0.24 0.079 $ 586,180

NOX (Nitrous oxides, tons/ day) 31.6 0.018 0.057 $ 7,848,330

VOX (Volatile organic compounds, tons/ day) 18.4 0.015 0.082 $ 9,418,000

*Based on daily cost o $ 141,250 ($706.4 M/20 Years/250 commuting days)

LYNX Emissions Savings and Cost per Ton Saved

LYN X Air Quality Impact

9646.85 302.92 31.6 18.4

0%

10%

20%

30%

40%

50%

60%

70%

80%

90%

100%

CO2 (tons/ day) CO (tons/ day) NOX (tons/ day) VOC (tons/ day)

r il E i i

Meck CoTotal

LYNXSavings

7.66 .24 .0187 .0156

Congestion Effect of LYNX,

Peak Hour Traffic

7,972

190,040

2,008,882

-

500,000

1,000,000

1,500,000

2,000,000

2,500,000

LYNX S. Corridor Meck Co

l l - l 4.2 % ofCorridor

0.39 % ofCounty


16/26

1CharlottE S lYNX l iNE : a PrEl imiNarY aSSESSmENt | iMpActS

were built-and-occupiedproperties, and thereore

signicantly overstates, by nearly a actor o

4, the actual observable on-the-ground growth.

This is highly signiicant since growth has

slowed sharply in the last year, and some othe announced growth may be delayed or

canceled.

Next, the report does not quantiy the dierence

in growth rates in the corridor relative to the

rest o the region (it incorrectly counts the

corridor growth as part o the city growth).

Further, the report implies that all o the

corridors growth is attributable to LYNX,

when clearly many actors have contributed to

the corridors resurgence.

Further, it treats allo the corridors growthas new to the region, and does not consider that

some growth in the corridor might have come

rom within the region.

Further, the report treats assessed valuation as

the appropriate measure o development, when

in act it is the economic rental value o land,

not its assessed value, that should be the basis

or analysis32.

Further, it does not consider (in its avor) some

growth outside the corridor might have been

infuenced by LYNX, the fip side o assumingthat all growth in the corridor is LYNX-

based.

Finally, it treats taxes as benets. This is inap

propriate since taxes are NOT benets but

transers rom the private sector to the public

sector, and i let in the private sector they

would also have generated economic activity.

In short, the report presents an overly optimistic

view o the South Corridor growth and LYNXs

eect.

Many transportation economists eel that the eco-

nomic activity itsel is not a benet but merely aseparate maniestation o the gain in user benets.

But i they are evaluated, the appropriate way to

measure land use impacts rom transportation

actions is to isolate transportation user benets or

economic activity, not use assessed valuation or

taxes. And o course announcements are not

growth and should not be counted until they actu-

ally appear as in-use property. The separate eect

o LYNX itsel on growth should be isolated, and

the relative growth within the corridor versus the

rest o the region should also be determined. Theusual way o gathering this inormation is through

careul studies o the change in economic activity

as a unction o location to the LYNX line, versus

changes in economic activity elsewhere, and in-

terviews o businesses to determine the relative

importance o LYNX versus other actors.

An important question is the expected length o

benets streams. User benets are typically as-

sumed to fow throughout the projects expected

lie, in this case 20 years. Other benets, suchas land use changes, are oten discounted over

much shorter lietimes, sometimes just one year.

This is because longer-term land use benets are

partially double-counted in user benets, are not

directly attributable to the project, and are subject

to economic orces beyond the project. For the

analysis below, we counted each years increment

o growth as a single one-year benet. This pro-

duces a conservative estimate o land use impacts;

an assumed longer lie o land use benets would

increase the projects benet-cost ratio. Another important point is that growth in the

remaindero the region also needs to be accounted

or and actored out o the corridor growth rates.

The Mecklenburg County economy is growing

everywhere, and the key question here is what

dierence in growth ratesthe LYNX line is making,

relative to the rest o the region.

(Some development along South Boulevard pre-dates the LYNX lineby about 10 years.)


17/26

1 CharlottE S lYNX l iNE : a PrEl imiNarY aSSESSmENt | iMpActS

To study these issues we track commercial activ-

ity rom several sources. One useul source is the

quarterly reports o square ootage o commercial

space actually in use, and average rents or oce,

retail, and industrial commercial space, as main-tained by Karnes Research and reported in the

Charlotte Business Journal33. This source maintains

summaries o activity by sub-region within Meck-

lenburg County, permitting partial extraction o

South Boulevard corridor activity versus the rest

o the county.

Ofce space Karnes Research summarizes oce space by 11

zones, o which ve are within or partially within

the LYNX corridor. For each zone, or part, wecomputed the annual rental income (occupied

square eet times average rent) in the LYNX

corridor and in the rest o Mecklenburg County,

and then computed changes in rental income

over time. This method accounts or occupancy

rates, growth in constructed space, actual use and

rental value on the ground, and infation. It is a

reasonable estimate o the value o oce space

over time.

The LYNX corridor accounts or about 12 percento Mecklenburg Countys $852 million in oce

space annual rental value. However, it has been

growing at a aster rate than the rest o the county.

From 2005 to 2006, as LYNX construction began,

the growth rate o oce space rental income in

the South Boulevard corridor was slightly aster

than the rest-o-county oce growth rate (5.1 ver-

sus 4.7 percent), or a corridor kick o about 0.4

percent. This may be due to pre-2005 construc-

tion activity and greater parcel availability, or

other actors, as well as anticipation o LYNX.

In the next year, 2005-06, the South Corridor

also grew more rapidly in oce space rental

value, 12.6 percent versus 10.5 percent or the

rest o the county, implying a corridor kick o

about 2.1 percent. This seems to be indicative o

a mild rush to build oce space in the corridor

in advance o LYNX, but the rest o the county

was also adding oce space rapidly.

Ater LYNX opened (in late November 2007) the

LYNX corridor oce market continued to grow

slightly more rapidly than the remainder o the

region (9.8 percent versus 9.0 percent), imply

ing a smaller kick o about 0.8 percent, but theoverall rates slowed as the U.S. economy began

to slow.

Converting these dierences in growth rates into

dollar value, the 2005-06 increment is about

$299,00034. Similarly, the increment or 2006

07 is about $1.683 million, and or 2007-08 it is

$734,000. For uture years, we use the last year

increment times 20 years to arrive at an estimate

o $14.671 million, in 2008 dollars (treating each

years increment o growth as a benet or justone year, not the ull project lie). Thereore

the corridor kick or oice space is abou

$17.387 million; even though not all o this can

be directly attributed to LYNX, we optimistically

assume so.

Retail space For analysis purposes, Karnes Research divides the

county into eight zones, o which two are partially

in the LYNX corridor. Counting the CarolinaPlace Mall and nearby space, along with some o

the Ballantyne area space, the corridor contains

(optimistically) about 22 percent o the countys

$614 million o retail annual rental value.

During the period April 2005 to June 2006

retail space rental income in the South Corridor

increased at a much slower rate, 3.8 percent, than

the rest o the county, 9.8 percent. This 6.0-per

Year-to-Year P ct Changes in Office R ental Value

5.1

12.6

9.8

4.7

10.5

9.0

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

Mar 05-June 06 July 06-June 07 July 07-June 08

LYNX Corridor

Rest of Meck Co.


18/26

1

cent drag may be due to LYNX construction,

lack o availability o appropriate sites, or weaker

demand within the corridor.

However in the next year, 2006-07, growth in thecorridor actually exceeded growth in the rest o

the county, 15.8 percent versus 12.9 percent, im-

plying a corridor kick o about 2.9 percent. This

is likely due to a combination o actors, including

LYNX construction, the opening o more space,

and some eect o rising rental prices.

In the last year, 2007-08, the corridors retail

rental value actually contracted slightly, about

-0.5 percent, while the remainder o the region

continued to expand at a very low rate, just 1.5percent.

Converting these dierences in growth rates into

dollar values, the estimated corridor kick or

2005-06 is anegative$6.685 million, since the cor-

ridor grew slower than the region. For 2006-07,

the increment is $3.341 million, and or 2007-08,

the increment is also negative, - $2.745 million.

Going orward, it is likely that the corridor will

continue to show a small advantage or retailgrowth. Assuming a small 0.75% dierential in

the per-year growth rate (but only or a single

year o each increment), based on comparisons

with oce and industrial trends, the total is about

$20.167 million in incremental growth o retail

rental value.

Combining these statistics, about $14.078 million in

added value to retail space can be attributed to cor-

ridor growth, compared with the rest o the county.

Industrial space Industrial space is summarized in the Karnes

reports by eight zones, two o which are partially

in the South Corridor.

From April 2005 to August 2006, industrial space

rental income in the South Corridor increased

considerably more slowly than in the rest o the

county. The dierence, 6.5 percent, may be due

to conversion o industrial space in the corridor

or other uses, LYNX construction, space avail-

ability or quality, or other actors.

From 2006 to 2007, the corridor grew more

rapidly, but not as ast as the remainder o the

region. However, the dierential, 1.9 percentage

points, narrowed considerably.


"C orridor Kick" Value of Office R ent, $K

299

1,683734

14,671

-

2,000

4,000

6,000

8,000

10,000

12,000

14,000

16,000

2005-06 2006-07 2007-08 2008-2025

Total $ 17.387 m over 20

Year-to-Year Perce nt Changes in Re tai l Space Annual Rental Value

3.8

15.8

9.8

12.9

1.5

-0.5

-2.0

0.0

2.0

4.0

6.0

8.0

10.0

12.0

14.0

16.0

18.0

Mar 05-June 06 July 06-June 07 July 07-June 08

LYNX Corridor

Rest of Meck. Co.

"C orridor Kick" in Value of Retail Space, $K

(6,685)

3,341

(2,745)

20,167

(10,000)

(5,000)

-

5,000

10,000

15,000

20,000

25,000

2005-06 2006-07 2007-08 2008-2025

Total $ 14.078 m over 20 years

(Parcel under development along the LYNX line, near Arrowood)


19/26

1 CharlottE S lYNX l iNE : a PrEl imiNarY aSSESSmENt | lMeS And coMMunitY-BASed cARe

Between 2007 and 2008, the corridors growth o

industrial space value was aster than the remain-

der o the county, 6.7 percent versus 4.2 percent.

So, in the nal year, the corridor exhibited a

corridor kick o about 2.5 percent. Converting these statistics into dollar values,

the LYNX corridor shows a negative $2.003

million dierence in the rst year, a negative

$0.541 million dierence in the second year,

but a $870,000 kick in the third year. Over 20

years, the increment is about $15.755 million

(once again, assuming just one year o benet

or each years incremental growth).

Has LYNX aected residential development? LYNXs eect on residential housing activity is

analyzed in a similar ashion. We determine, rst,

the magnitude o the incrementalgrowth in hous-ing units o various types in the corridor versus

the rest o the region, then determine the annual

economic rent or those units. This permits

estimation o the corridors incremental growth

and economic activity.

We retrieved the residential parcel data by zip

code rom the Residential Building and Land

Summary report in the Charlotte-Mecklenburg

Integrated Data Store. There are 40 zip codes

assigned to Mecklenburg County, 29 o which

are currently in use. O these 29, 21 are ully

outside the South Corridor; one is ully withinthe South Corridor; and seven are partially within

the South Corridor. We also reviewed South

Corridor TOD35 Redevelopment data rom the

City o Charlotte36. We then combined both data

sets to assess the impact o the LYNX system on

residential development in the South Corridor

and the rest o Mecklenburg County.

During the period 2003-2006, Mecklenburg

County added more than 82,500 residential units

about 20,600 units annually. O these, almos1,100 (5.2%) are in the South Corridor. (2007

data was incomplete and so was not used.)

Ater a 16% decline rom 2003-04, the growth rate

o new residential units has been increasing in

the rest o the County, with over 11% more unit

added in 2006 than 2005. In the South Corridor

growth in newresidential units declined in both

the 2003-04 and 2004-05 time periods, beore

showing a signicant rise rom 2005 to 2006.

The current value o the units can be determined

rom the total parcel values and is refected belowTotal value in the South Corridor is a small raction

o the total value o new residential units across

the County, but the valuepernew residential uni

is higher in the South Corridor (24% higher, on

average). While this could perhaps be part o the

corridor kick, it could also be a result o diering

sizes o new units (larger units command larger

prices) and diering land values (more accessible

parcels command higher rents).

"C orridor Kick" in Value of Industrial Space, $K

870

17399

-2003 -541

(5,000)

-

5,000

10,000

15,000

20,000

2005-06 2006-07 2007-08 2008-2025

Total $15.755 m over 20 years

Mecklenburg County New Residential Units

21,45418,063 18,384 20,454

1,152

1,098 8211,201

-

5,000

10,000

15,000

20,000

25,000

2003 2004 2005 2006

Year BuiltRest of County South Corridor

Year-to-Year Perc ent Changes in Industrial Space Rental Value

3.4

8.1

6.7

10.0 9.9

4.2

0.0

2.0

4.0

6.0

8.0

10.0

12.0

Ma r 0 5-J une 06 Ju ly 0 6-Ju ne 0 7 J ul y 0 7-Ju ne 0 8

LYNX Corridor

Rest of Meck. Co.


20/26


21/26

0 CharlottE S lYNX l iNE : a PrEl imiNarY aSSESSmENt | Benefit-coSt ASSeSSMent

Category 2003-06 2006-07 2007-08 20 More Years TotalOfce Space 0.299 1.683 0.734 14.671 17.387

Retail Space - 6.685 3.341 -2.745 20.167 14.078

Industrial -1.961 -0.580 0.871 17.425 15.755

Residential 18.117 6.286 174.432 198.835Total 9.770 10.730 225.555 246.055

iV. Benef-Cs assessen

How are transportation benefts and costsdetermined? Classical transportation economics determines

benets fowing to userso new services (savings

in travel time, savings in operating costs, andsavings in accidents), and to non-users (reduced

air pollution or congestion, better land use pat-

terns, reduced environmental impacts, etc.). Most

analyses do not compute the impacts o more or

wider choices or minor secondary impacts such as

reduced oil dependence. Some economists (but

not all) consider land use changes to be second-

ary benets since they are generally derived rom

user benets (travelers switching to new routes or

modes to save time). Specialized benets, such as

savings in parking costs, are sometimes oset byreductions in revenues to parking companies.

Costs are typically computed as total costs o

construction and operation over the projects lie-

time, or through one complete lie cycle (initial

construction through maintenance and repairs,

ending just beore the second reconstruction).

Taxes are not considered as either benets or

costs, since they are transers rom the private to

the public sector.

Future benets and costs are oten brought back

to present value (to account or the time value o

money), although this may not be done i most

costs are in current dollars. Infation is generally

not considered in cost estimation.

Some benets and costs such as construction costs

or increased economic activity are sometimes

expanded to account or their ripple eect

through the local economy. This step, sometimes

called multiplier modeling, can be done using

sophisticated economic sotware, or by using

simple actors. This method is not generally

done in transportation impact studies since it is

not generally known where transportation costs

or benets are spent. In this assessment, we did

not include multipliers.

For the present study o the LYNX line, the costs

are essentially known or can be estimated with

reasonable certainty. Some user benets, par

ticularly savings in travel time, operating costs

and accidents by prior car drivers, can also be

estimated reasonably well. Benets rom land use

changes can also be reasonably well estimated

Environmental benets such as air pollution or

CO2reduction can also be estimated but are more

dicult to dollarize. Benets rom increased

mode choices, choices or urban TOD housing,

visually pleasing environment, smart growth,

and the like cannot be estimated easily and are

not quantied.

Local ocials oten point to transportation con

struction itsel as a benet to the region. Thi

benet is sometimes called the construction

benet, and consists o the dollar value o theconstruction, expanded to account or its ripple

through the local economy. However, those ex

penditures are really just re-investments o taxes

taken prior (hence transers), and most are no

even spent locally but are expended on steel

electrical, and other materials produced out o

the region. These expenditures are also shortlived

and dissipate quickly ater construction. Some

analysts take a middle ground and count only

the non-local (ederal and state unds) as a benet


22/26

1CharlottE S lYNX l iNE : a PrEl imiNarY aSSESSmENt | Benefit-coSt ASSeSSMent

9/16/2008

LYNX Beneft/Cost Assessment

Item Category BasisAnnualEstimate2008 M$

20-YearEstimate2008 M$

Costs ConstructionConstruction cost +Additional City cost

26.095 521.9

Operating Costs LYNX Est. AnnualOperating Budget 9.226 184.5

Total Costs 35.321 706.4

Benefts UserAuto Diversion Time Savings,$m

4,510 daily trips at @$ 10/hr and10 minutes saved/trip (probablyhigh)

1.872 37.4

UserTransit Shiter T ime Savings,$ m

8,077 daily trips@$10/hr/ and 5min saved/trip.

1.683 33.7

User Walk-up Time Savings

4,308 daily trips*10/hr and 5 min

saved/trip 0.897 17.9

UserAuto Driver Operating CostSavings

VMT Reduced*0.20/mi vehicleoperating cost

0.797 15.9

UserAuto Driver Accident Cost Sav-ings

VMT Reduced*1.25 atalities/100mm*$5m/lie

0.249 5.0

UserParking Costs Savings orDiverting Wkdy Drivers

2,050 Parked cars*$5.15 avgCBD pkg cost/day

2.639 52.8

UserParking Cost Savings or Week-end Auto Drivers

800 parked cars*5.15 avg parkingcost/day*115 days/year

0.474 9.5

Non-User Travel time savings to non-userdrivers (rom diverted vehs)

4% reduction in corridor PH trafc

(0.27 min. savings)*190,040 non-users*$ 10/hr

0.428 8.6

9.039 180.783

Land UseOfce, retail, industrial value oeconomic rent

Initial Corridor kick + 20 yrs oincremental growth

2.530 50.6

Residential housing, value oeconomic rent

Initial Corridor kick and 20 yearso incr growth

9.941 198.8

Increased CBD Activity (newecon. activity)

500 trips/day *$ 30/trip (incre-ment)

5.475 109.5

Losses to the Parking Industry- Losses rom reduced parkingees, equal to user gains

-3.113 -62.3

14.832 296.7

Environ Reduced energy consumption Included in operating cost savings

Reduced air pollution (CO, NOx,VOC)

0.275 tons/day saved*$55/tonremoved

0.004 0.1

Reduced greenhouse gases(CO

2)

7.66 tons/day saved*$55/tonremoved

0.105 2.6

0.109 2.7

Other Value o Mobility/Choice

Total Benefts $ 23.98 $ 480.18

Beneft-Cost Ratio 0.68 0.68


23/26

Our policy here is to ollow common transporta-

tion planning practice, which is NOT to count the

construction expenditure as a benet since it is

essentially a tax-based transer, is shortlived and

has an unknown local component.

What are the benefts and costs o the LYNXline? The table on page 21 summarizes the computa-

tion o benets and costs or the LYNX line.

Total construction and operating costs are

about $706 million over 20 years in current

dollars. These are likely underestimated, since

operating costs tend to rise aster than infationin most transit systems.

User benets total are approximately $180.8

million over the 20-year lie o the acility.

These include benets rom savings in travel

time, operating costs, accidents, parking, and

time savings to non-using drivers rom reduced

trac congestion. CBD parking cost savings (a

direct benet to drive-and-park users, but a loss

to the parking industry) are about $62.3 million

over 20 years.

Benets rom additional corridor land-use

economic activity, additional (not-duplicated)

downtown CBD activity, along with losses

to the local parking industry, are estimated

at $296.7 million over 20 years, using some-

what optimistic assumptions about continuing

growth in the corridor. But even though these

benets are thought by some economists to

be mere double-counting o user benets, we

include them because they are o considerablelocal interest. I not included, the benet-cost

ratio would be substantially lower.

Environmental benets are estimated at about

$2.7 million over 20 years. These are primarily

rom reduced pollution by diverted auto drivers.

Thus, the total estimated benets o the project

are about $480.2 million, compared with a tota

cost o $706.4 million, o which $400.4 million

is the local cost.

The overall benet-cost ratio or the project is

thereore about 0.68. I only local costs are in-

cluded, the benet-cost ratio is near 1.20. This

means that i the local government had to pay

or the ull project, $706 million, it could not be

justied rom a benet-cost basis. In essence,

what makes the project justiable rom the

local governments perspective is that the local

government pays only 57 percent o costs (state

and ederal governments pay the remainder),

yet the local government reaps all benets.Butrom the perspective o overall benets and

costs, the project has benets that are lower

than costs, and it thereore would have to be

justied on other grounds.

The benet-cost methodology does not quan-

tiy some benets. These include the benet o

dierent land use patterns, reduction in sprawl

(i any), arguments or world class status, ben-

ets o just having choice or uture residents,

etc. This is not to say that these benets do notexist, just that they cannot be easily quantied.

But the benet-cost procedure allows deter-

mination o the remaining value that these

additional benets would have to be to make

the project viable.

It is widely understood that there are many

problems with the benet-cost methodology.

Among the most widely cited are the ailure to

determine geographic disparities between who

gets and who pays; uncertainties o computa-tion o virtually all terms; and uncertainties

about interest and infation i considered. Nev-

ertheless, it remains one o the central ways by

which transportation projects are evaluated.

CharlottE S lYNX l iNE : a PrEl imiNarY aSSESSmENt | Benefit-coSt ASSeSSMent


24/26

CharlottE S lYNX l iNE : a PrEl imiNarY aSSESSmENt | noteS

1 US Census Bureau, American Community Survey,2005, Urbanized Area Data. See www.census.gov.2 A vehicle-mile, dened as one vehicle traveling onemile, is a widely used measure o urban travel.3 Charlotte Area Transit System, Market Research Re-sults 2000-2007, Summary Fall 2007.4 Ridership (rides, riders, trac) is dened as one-wayunlinked trips, not the persons making those trips. Soor instance i a commuter walked to LYNX and thenreturned, that would be 2 unlinked trips. I the commuterused a bus to transer to LYNX and then returned, thatwould be 4 unlinked trips.5 Metropolitan Transit Commission, Charlotte AreaTransit System Ridership Report, July 2008, and earlierreports. Received August 2008.6 The orecast also assumed 4 transit lines, higher down-town parking ees, and about 100,000 workers downtown.

I redone with revised (lower) assumptions, the orecastwould have been lower.7 Ibid. August 2008 average weekday trac was 16,357,3.2 percent lower than July, but August Saturday andSunday trac was higher.8 August 2008 trac (454,524 trips) was 75 percentweekday. Source: MTC, CATS Ridership Report, August2008, received September 16, 2008.9 Metropolitan Transit Commission, Charlotte AreaTransit System Ridership Report, Data July 2007-July2008, Received August 2008.10 Charlotte Area Transit System, Ridership by Routes orRoutes That Connect to LYNX, Data November 2007-

July 2008, Received August 2008.11 Charlotte Area Transit System, Ticket Sales, Memorom O. Kinard to J. Taylor, April 29, 2008.12 Parked vehicles * 2 trips/day * 1.1 riders/vehicle.Source: Charlotte Observer parking lot counts.13 Parking lot observations by The Charlotte Observerand counts by the Hartgen Group, Wed., August 20,2008, 9:20 am-11:20am.14 City o Charlotte CDOT Planning and Design Oce,Central parking report 2006-2007.15 Metropolitan Transit Commission, Charlotte AreaTransit System Ridership Report, July 2007-July 2008,Received August 2008.16 Federal Transit Administration, National Transit Data-base, 2000-2006, available at www.ntdprogram.org17

AAA Carolinas, Monthly Average - Metro Average- North Carolina - Charlotte-Gastonia-Rock Hill (N.C.only), January 2006-August 2008, Received August 2008.18CATS Four-Corridors Major Investment Study, TravelDemand Models Methodology Report, Feb 12, 2003. Thetechnical term or these mode choice models is nestedlogit.19 Federal Transit Administration, New Starts Reports,various years. Available at www.ta.dot.gov.20Raising the gasoline price rom $3 to $4 per gallon in-creases the cost o a 10-mile trip by about $0.50, or about9 percent o the total cost o driving 10 miles, $5.80. Inthe computer models, this would produce about a 1/3

percentage-point increase in the weekday transit share,noticeable in transit ridership but too small to be seen inauto trac.21 Wendell Cox, Analysis o the Charlotte South Boule-vard Corridor LRT, 2000. www.demographia.com.22 South Corridor Inrastructure Program (SCIP),Charlotte Engin. (James Keenan) and then veri-ed/edited. http://www.charmeck.org/Department/City+Engineering/See+Our+Projects/Transportation/home.htm, Received August 2008.23 Flyvbjerg, B. et al, Megaprojects and Risk: Anatomy o Am-bition, Cambridge University Press, 2003. He also notesthathighwayprojects also have cost overruns, averagingabout 40 percent.24 Charlotte Area Transit System, Light Rail FY2008Operating Expenses, Received August 2008.25 Assumes no state or ederal unds will be spent or

operations. Most cities do spend some ederal and stateunds or transit.26 This assumes no increase in either operating costs orridership. I both increase, this cost might be even highersince historically costs have risen aster than ridership.27 http://www.ncdot.org/it/img/DataDistribution/Trac-SurveyMaps/byYear.html?year=Urban , Data Year 2002-2006, Received August 2008. Additional 2008 counts orI-77 were obtained rom NCDOTs Trac MonitoringCenter, Received September 2008.28 http://www.charmeck.org/Departments/Transporta-tion/Trac/Trac+Counts.htm , Data Year 2003-2007,Received August 2008.29 Trac volumes typically fuctuate 5-10 percent rom

day to day, about twice the likely LYNX savings.30 City o Charlotte, South Corridor Economic Develop-ment Update, August 27, 2007. The supporting spread-sheet is dated 8/24/06, a year earlier.31 The report contains an error in the spreadsheet; theactual total is 164,512 square eet.32 For instance, a condo valued at $300,000 might berented or only $750/month, or $9,000 year, i (say) theowner had no mortgage on it. This is the economic rento the property, which must cover carrying costs andprot, and is the basis or comparison with other invest-ments such as stocks, bonds, etc.33Charlotte Business Journal, Commercial Real EstateQuarterly, April 2005 and August 2006-August 2008. (or

the rst quarter o 2005 and the second quarters o 2006,2007, and 2008).34 0.39 percent times $76.68 million, the value o annualoce rent in the South Corridor in 2005.35 Transit Oriented Districts (TODs) are planned small-area developments, usually only mile in radius, aroundtransit stations. They typically have mixed land uses(residential and commercial together), higher densities, avariety o activities, oriented toward the station.36 City o Charlotte, South Corridor Economic Develop-ment Update, August 27, 2007. The supporting spread-sheet is dated 8/24/06, a year earlier. Received July 2008.

Nes


25/26

CharlottE S lYNX l iNE : a PrEl imiNarY aSSESSmENt | AcknowledgeMentS

abu e Jn lcke Fundn

The John Locke Foundation is a nonprot, nonpartisan policy institute based in Raleigh. Its mission isto develop and promote solutions to the states most critical challenges. The Locke Foundation seeks totransorm state and local government through the principles o competition, innovation, personal ree-dom, and personal responsibility in order to strike a better balance between the public sector and privateinstitutions o amily, aith, community, and enterprise.

To pursue these goals, the Locke Foundation operates a number o programs and services to provideinormation and observations to legislators, policymakers, business executives, citizen activists, civic andcommunity leaders, and the news media. These services and programs include the oundations monthlynewspaper, Carolina Journal; its daily news service, CarolinaJournal.com; its weekly e-newsletter, Caro-lina Journal Weekly Report; its quarterly newsletter, The Locke Letter; and regular events, conerences,and research reports on important topics acing state and local governments.The Foundation is a 501(c)(3) public charity, tax-exempt education oundation and is unded solely romvoluntary contributions rom individuals, corporations, and charitable oundations. It was ounded in 1990.For more inormation, visit www.JohnLocke.org.

abu e aus

David T. Hartgen, Ph.D., P.E., is Proessor Emeritus o Transportation Studies at UNC Charlotte andPresident o The Hartgen Group. Pro. Hartgen is widely known in transportation circles. He establishedthe UNC Charlottes Center or Interdisciplinary Transportation Studies in 1989 and now conducts

research in transportation policy. He is the author o about 335 publications on a wide variety o topics intransportation policy and planning, is US Co-Editor o the international academic journal Transportation,and is active in proessional organizations. He is a requent local and national media interviewee. Beorecoming to Charlotte he directed the statistics and analysis unctions o the New York State Department oTransportation and served as a Policy Analyst at the Federal Highway Administration. He holds engi-neering degrees rom Duke University and Northwestern University. He has taught at SUNY Albany,Union College and Syracuse University and lectures widely. His recent studies o congestion reduction,perormance o state highway systems, sprawl, and cost-eective highway investment have attracted widenational attention.

M. Gregory Fields is a retired military ocer, currently enrolled in the PhD program in Urban Re-gional Analysis at UNC Charlotte. A West Point graduate, he has masters degrees rom Webster Uni-

versity in St. Louis and UNC Charlotte and has participated in a number o comparative studies, includ-ing the Fraser Institute study o the Canadian provinces, Reasons recent national study o congestion,Reasons recent study o accessibility and productivity, and global climate change.

Elizabeth San Jose is a researcher with the Hartgen Group, working on studies o transit service,urban congestion and climate change. She holds a BS rom Indiana State University.

Anthony Layzell is a graduate student at UNC Charlotte, Department o Geography and Earth Sci-ence. He holds a BS in Geography rom the University o Nottingham in England.

Matthew Scottholds a BA in Economics rom Davidson College and is a graduate student in econom-ics at UNC Charlotte. His projects with the Hartgen group include transit perormance and climate

change.


26/26

To prejudge other mens notionsbefore we have looked into themis not to show their darkness

but to put out our own eyes.johnlocke (16321704)

Author, two treAtises of GovernmentAn dfundAmentAl constitutions of cArolinA

200 West Morgan St.

Raleigh, NC 27601

V: 919-828-3876

F: 919-821-5117

j h l k

Charlotte’s LYNX Line: A Preliminary Assessment

Documents

Transcript of Charlotte’s LYNX Line: A Preliminary Assessment