highway supply, demand & economic efficiecny

8/13/2019 Highway Supply, Demand & Economic Efficiecny

1/24

Economics:Pricing, Demand, andEconomic EfciencyA P R I M E R


2/24

Quality Assurance Statement

The Federal Highway Administration (FHWA) provides high quality

information to serve Government, industry, and the public in a manner

that promotes public understanding. Standards and policies are used to

ensure and maximize the quality, objectivity, utility, and integrity of its

information. FHWA periodically reviews quality issues and adjusts its

programs and processes to ensure continuous quality improvement.


3/24

Contents

The Primer Series and the Purpose of This Volume 2

Introduction 4

Basic Economic Concepts: Highway Supply and Demand 6The Cost of Travel 6The Demand for Highway Travel 7The Supply Side 8Market Equilibrium 10

Pricing and Economic Efciency 11Congestion Externalities and Economic Efciency 11Efcient Pricing 11Variability in Demand and Pricing 12Congestion Pricing and Other Externalities 13Toll-Collection Costs and the Role of Technology 14

Forms of Congestion Pricing 15

Priced Lanes: The Economics of Multi-Class Service 16HOT Lanes 16Benets From Product Differentiation 17Impacts on Capacity and Delay 18

Empirical Estimates of Congestion Tolls 19Reference Inside back cover


4/24

2 | C O N G E S T I O N P R I C I N G

The Primer Series and the Purpose of This Volume

States and local jurisdictions are increasinglydiscussing congestion pricing as a strategy forimproving transportation system performance.In fact, many transportation experts believe thatcongestion pricing offers promising opportunities tocost-effectively reduce trafc congestion, improve

the reliability of highway system performance,and improve the quality of life for residents,many of whom are experiencing intolerable trafccongestion in regions across the country.

Because congestion pricing is still a relativelynew concept in the United States, the FederalHighway Administration (FHWA) is embarkingon an outreach effort to introduce the variousaspects of congestion pricing to decision-makersand transportation professionals. One element ofFHWAs congestion-pricing outreach program isthis Congestion Pricing Primer Series. The aimof the primer series is not to promote congestionpricing or provide an exhaustive discussion of thevarious technical and institutional issues one mightencounter when implementing a particular project;rather, the intent is to provide an overview of thekey elements of congestion pricing, to illustrate themultidisciplinary aspects and skill sets required toanalyze and implement congestion pricing, and to

About This Primer Series

The Congestion Pricing Primer Series is part ofFHWAs outreach efforts to introduce the variousaspects of congestion pricing to decision-makers andtransportation professionals in the United States. Theprimers are intended to lay out the underlying rationalefor congestion pricing and some of the technical issuesassociated with its implementation in a manner that isaccessible to non-specialists in the eld. Titles in thisseries include:

Congestion Pricing Overview.

Non-Toll Pricing.

Technologies That Enable Congestion Pricing.

Technologies That Complement Congestion Pricing.

Transit and Congestion Pricing.

Economics: Pricing, Demand, and Economic

Efciency.

Income-Based Equity Impacts of Congestion

Pricing.


5/24

E C O N O M I C S : P R I C I N G , DE M A N D , A N D E C O N O M I C E F F I C I E N C Y | 3

provide an entry point for practitioners and othersinterested in engaging in the congestion-pricingdialogue.

The concept of tolling and congestion pricing isbased on charging for access and use of our roadwaynetwork. It places responsibility for travel choicessquarely in the hands of the individual traveler,where it can best be decided and managed. The car isoften the most convenient means of transportation;however, with a little encouragement, people maynd it attractive to change their travel habits,whether through the consolidation of trips, car-sharing, by using public transportation, or bysimply traveling at less-congested times. The use ofproven and practical demand-management pricingthat we freely use and apply to every other utilityis needed for transportation.

The application of tolling and road pricingprovides the opportunity to solve transportationproblems without Federal or state funding. It couldmean that further gas tax, sales tax, or motor-vehicle registration fee increases are not necessarynow or in the future. Congestion pricing is not acomplete plan of action. It has to be coordinatedwith other policy measures to maximize success.

This volume describes the underlying economic

rationale for congestion pricing and how it can beused to promote economic efciency. It lays outthe basic theory of travel demand and trafc owand shows how inefcient pricing of the roadnetwork helps create an economic loss to society,as well as the means by which this can be alleviatedthrough pricing. The impact of congestion pricingon highway infrastructure investment and therevenue implications of congestion pricing willbe discussed in a separate volume in this primerseries.


6/24


Introduction

Trafc congestion represents a signicant and grow-ing threat to mobility in the United States. Accord-ing to recent data, the average driver in U.S. urban-ized areas traveling in peak periods experienced 38hours of total delay in 2005 (up from 14 hours in1982), at a total cost of $78 billion in excess traveltime and wasted fuel consumption.

To some extent, these trends reect the vibrancyof our Nations cities, which have seen signicanteconomic growth over the past 3 decades. They alsoreect growing imbalances between travel growthand the development of new highway capacity, asthe former has exceeded the latter for many years.However, in an important way, trafc congestionalso results from the inefcient pricing of highwayuse, leading drivers to travel more than what is eco-nomically desirable, particularly during the heightof morning and evening rush hours.

The fundamental economic problem in highwaycongestion is that highway users consider the coststhat they bear themselves (such as fuel costs andtravel time), but not the additional delay that theyimpose on others when using a congested facility.Economists refer to these latter costs as externalcosts or externalities, because they are not borneby the person whose use of the highway createsthem. The goal of congestion pricing is to makedrivers internalize these costs by directly chargingthem for use of the highway in proportion to theexternal costs that they cause for society. In doingso, the economic inefciencies associated with un-derpriced and overutilized highways can be signi-cantly reduced.

Congestion pricing is becoming increasinglyviewed as an important strategy for improving theoperational performance of the highway system. Bymoving from a system based on xed or at chargesto highway users (e.g., vehicle fees and fuel taxes) toa more rened system of charges that can vary bythe time and location of travel, price signals can beused to more efciently and effectively allocatescarce capacity on our Nations highways.


7/24



8/24


This section introduces some of the basic economicprinciples that provide a foundation for understand-ing the economic rationale for congestion pricing,discussed in the following sections. Key conceptsinclude:

The explicit and implicit costs associated withhighway travel; The demand for using highway facilities;

Basic relationships of traf c volumes and traf cow; and

The interaction of supply and demand to deter -mine trafc volumes and highway user costs.

THE COST OF TRAVEL

We live in a world of scarcity, meaning that resourc-es are not unlimited. As a result, when we consumea good or service, we forego the opportunity to con-sume something else. Economists refer to this as anopportunity cost, and it is perhaps the most fun-damental concept in all of economics. In an impor-tant sense, the price that people pay for consump-tion is this opportunity cost.

For most goods and services, the price that con-sumers pay is simply the out-of-pocket monetary costto the user; the foregone opportunity would be tospend that money on something else. However, econ-omists have also recognized that consumption re-

quires time as well as money, which is itself in limitedsupply. This insight has been useful in analyzing thedemand for time-saving devices (e.g., microwave ov-ens) and recreational services, among many others.

For highway travel, the time spent in transit isone of the most critical components of the price

that users face. Time spent in travel is time thatcould be devoted to other pursuits, such as earningincome or engaging in leisure activities. The value ofthese other pursuits represents the opportunity costof travel time.

Safety is another implicit cost associated withhighway travel. When people choose to travel, theyalso take on the risk of property damage, injuries, oreven death due to crashes. These risks can be valuedbased on the willingness of people to pay for reduc-ing the risk of such adverse outcomes.

Other types of costs borne by highway users dorepresent actual, direct costs. Vehicle operatingcosts include such items as fuel, oil, tires, mainte-nance, and depreciation. Travel-related taxes andtolls represent another direct cost to users. In thecase of cash or coins paid at toll booths, these canliterally be out-of-pocket costs.

The preceding discussion of user costs is basedon the predominant form of highway use, namelyindividuals driving or riding in privately owned pas-senger vehicles. Similar considerations hold forcommercial vehicles, with two key exceptions. First,the value of time for drivers is a direct cost ratherthan an implicit one, because the drivers are beingpaid for their services. Second, the user cost fortruck transportation also includes the time value ofthe cargo while in transit.

Basic Economic Concepts:Highway Supply and Demand


9/24


THE DEMAND FOR HIGHWAY TRAVEL

The demand for highway transportation representsthe value that consumers place on traveling in aparticular time, manner, and place, as measured by

their willingness to pay for a trip. Some trips willbe valued very highly, whereas others will be valuedmuch less so. This relationship between the cost oftravel and the level of demand for travel is com-monly depicted as the travel demand curve (seeExhibit 1).

The travel demand curve slopes downward, re-ecting a basic economic truth: As the price of agood or service falls, the quantity that will be de-manded increases, holding other factors constant.The demand for travel is no different: When theprice of travel is high (in the generalized user-costsense described above), fewer people will be willingto make fewer trips; when that price falls, there willbe more people willing to make more trips.

The demand curve is characterized by two im-portant qualities: its level and its shape. The level ofdemand (i.e., the position of the demand curve) isaffected by a number of factors. For example, eachtrip has an origin and a destination. The more peo-ple there are at a particular origin and the moreactivities (e.g., shopping or employment) there areat a particular destination, the more will routes be-tween the origin and destination be in demand fortravel. As income levels rise, the willingness to payfor travel also increases, shifting the demand curveoutward. Demand levels can also vary signicantly(and importantly for the discussion here) by time ofday, due to the simple fact that people prefer tosleep at night and be active during the day, leadingto higher levels of demand for travel in the morningand early evening and lower levels of demand dur-ing mid-day and overnight hours. Finally, subjectivequalities such as comfort and convenience can af-fect the level of demand.

The responsiveness of the quantity of travel de-manded to changes in the price of travel is mea-sured by travel demand elasticity. Mathematically, itis simply the percentage change in quantity de-manded divided by the percentage change in price.

Intuitively, elasticity represents the shape of the de-mand curve. If the quantity demanded changes sig-nicantly in response to small changes in price, de-mand is said to be relatively elastic; thus, the demandcurve is fairly at. Conversely, if demand changesonly slightly in response to large changes in price,demand is said to be relatively inelastic; thus, thedemand curve will be relatively steep. At the ex-tremes, demand can be said to be perfectly elastic(i.e., any change in price results in an innite changein quantity demanded) or perfectly inelastic (i.e.,any change in price results in no change in quantitydemanded).

The elasticity of demand also depends on a num-ber of factors. Perhaps most important is the time-frame being considered: Demand is typically lesselastic in the short run than in the long run. Whenthe price of travel changes signicantly (as it hasrecently with large runups in fuel prices), travelersinitially have relatively few opportunities for ad-

justing their behavior. They may decide not to makesome trips or to change their mode of travel towork, but their housing and employment locations,key determinants of the level of travel, are likely toremain xed initially. In the long run, however, ev-erything is variable. People may choose to movecloser to their work or take jobs closer to home.Commercial real estate development patterns may

Traffic Volume

C o s t o

f H i g h w a y

T r a v e

l

Exhibit 1. The travel demand curve.


10/24


also respond to reduce the distance between con-sumers and activity centers. As a result, the long-term impact of an increase in travel costs on thevolume of highway may be much higher than theshort-term impact.

The elasticity of demand is also affected by thequality and availability of close substitutes. For ex-ample, if two companies make very similar prod-ucts, then consumers are likely to readily switchfrom one product to the other in signicant num-bers if the price of one of the products changes,resulting in high-demand elasticity for each prod-uct. Conversely, if there are no good substitutesfor a good or service, then consumers might sim-ply be faced with a choice between paying a high-er price or going without, in which case demand islikely to be inelastic.

For highway travel, public transit use can oftenserve as a reasonable alternative to highway use. Ingeneral, the higher the quality of that service (suchas frequency and convenience), the better a substi-tute it will be to driving and the more elastic high-way travel demand will be. Telecommuting can alsoserve as a substitute for highway commuting duringpeak periods; the more readily it is offered by

employers, and the better it is at substituting fortime spent in an ofce (e.g., through high-qualitytelecommunications), the more highway travelwill respond to changes in the cost of travel.

THE SUPPLY SIDE

In determining market outcomes, supply is thecounterpart to demand. For most goods and ser-vices, supply is simply related to production: howmany toys will be fabricated, how many tennis les-sons will be taught, etc. For highway travel, how-ever, the relationship is more complicated. Thehighway infrastructure may be built and main-tained by a public agency or private entity, butconsumers supply the vehicles (and their persons)to travel on that infrastructure. In order to under-stand the supply side of highway travel, it is rsthelpful to review some basic principles of trafcow and how highways become congested.

Trafc engineers typically characterize trafcow as a relationship between travel speeds, trafcvolumes, and trafc density (e.g., number of vehi-cles occupying a given space on the road). Exhibit 2shows the general shape of these relationships.


11/24


Volume (vehicles/mile)

SpeedVolume

S p e e

d ( m i l e s /

h o u r )

Density (vehicles/mile)

VolumeDensity

V o l u m e

( v e

h i c l e s

/ h o u r )

Density (vehicles/mile)

SpeedDensity

S p e e

d ( m i l e s /

h o u r )

Exhibit 2. Fundamental traffic flow relationships.

When trafc volumes are very low, vehicleshave minimal impact on one another, and theirtravel speeds are limited only by trafc-controldevices and the geometry of the road. As trafcvolumes increase, however, trafc density in-creases, and the freedom for vehicles to maneu-ver is more constricted. As a result, travel speedsbegin to decline, relatively slightly at rst, butfalling signicantly as trafc volumes approachthe maximum capacity (service ow rate) on thefacility. As trafc density continues to increasebeyond this saturation point, the speedvolumerelationship actually bends backward, as trafcow breaks down and fewer vehicles are able toget through.

The decline in travel speeds as trafc volumesapproach roadway capacity, of course, is what weall know as congestion delay. The important impli-cation of this is that there will be a relationshipbetween highway-user costs and trafc volumeson a particular road. At lower volumes, user costswill be relatively constant with respect to volume.As trafc volumes increase, however, user costswill eventually begin to rise at an increasing rate;the point at which this occurs depends on the ca-pacity of the road (see Exhibit 3). This relation-

ship is sometimes referred to as a generalized usercost curve.

Volume

C o s

t o

f T r a v e

l

Exhibit 3. Generalized user costs and traffic volumes.


12/24


Demand

Volume

Cost

V0

P 0

Exhibit 4. Equilibrium user costs and traffic volumes.P = price. V = volume.

MARKET EQUILIBRIUM

When supply and demand are in balance, a mar-ket is said to be in equilibrium. This is often rep-resented as the intersection of a supply curve and

a demand curve, which determines the market-clearing price and quantity (see Exhibit 4). Atthis point, everyone who purchases the good iswilling to (collectively) buy that amount at thatprice, and producers are willing to supply thatquantity at that price. If either the supply or de-mand curves shift, the market price and quantitywill also change.

For highway travel, demand is determined asdescribed above. The supply curve, however, isessentially represented by the generalized costcurve. The intersection of these two curves deter-mines how high trafc volumes will be and whatthe associated average highway-user costs will beat that volume level. When the level of demand islow relative to the capacity of the road, it will beuncongested, and prices will be relatively constanteven as volumes increase (the at part of the usercost curve in Exhibit 4). However, when demandlevels are high and the road is congested, both usercosts and trafc volumes will be higher, potentiallyrising sharply as demand continues to increase.


13/24


CONGESTION EXTERNALITIES

AND ECONOMIC EFFICIENCY

Economics is focused on the allocation of scarce re-sources in the economy. When resources are allo-cated to their highest valued use in the economy,

the outcome is said to be economically efcient . Un-der most circumstances, the setting of prices inmarkets through the interaction of consumers andproducers can help achieve an efcient outcome. Inmany other cases, however, market forces can leadto excessive consumption in a particular sector;highways are a good example of this.

The key to achieving economic efciency in amarket is ensuring that prices reect the opportu-nity cost to society of producing and consuming aparticular good or service. For most goods and ser-

vices, this opportunity cost will be equivalent tothe incremental cost of the resources that are re-quired to produce the good. For some goods, how-ever, one persons consumption can have a negativeimpact on other users (or non-users) of the good, asituation that economists refer to as an external cost or negative externality. If the market price does notreect these external effects, then the opportunitycost to society will be greater than the price thatconsumers face, leading to excess consumptionfrom the point of view of society and thus creating

economic inefciency.From an economic point of view, externalities

are a key problem of congested highways. To seewhy, consider the generalized cost curve shown inExhibit 3. This curve is an average cost curve andreects the rise in travel costs that individual usersface as trafc volumes increase. What it does notreect, however, are the incremental costs that

each successive vehicle that enters the trafc streamimposes on all the others by causing the speed ofthe entire ow of vehicles to decrease as congestionbuilds. The true opportunity cost to society of us-ing highways thus includes both the costs that indi-vidual users face and the congestion externality.

This is sometimes referred to as the marginal sociacost of congestion (see Exhibit 5).

From the point of view of society, the efcienttrafc level would occur at the point where themarginal social cost curve meets the demand curve,shown in Exhibit 5 as V*. At this trafc level, all theusers of the highway value their trips at least asmuch as the incremental cost to society of addingmore users. The equilibrium trafc level, however,is somewhat higher, at V 0. At this point, there are alarge number of drivers who are using the facility,because the value they place on travel is greaterthan the cost that they face (the demand curve isabove the average user-cost curve), but the cost tosociety is greater than the value they receive. Thus,there is an economic efciency loss due to excessivetrafc volumes.

EFFICIENT PRICING

Improving economic efciency under these cir-cumstances requires the elimination of trips thatare valued less than their social cost. One way todo this might be to somehow identify all of thelower valued trips and to ensure that they do notoccur during this demand period on that particu-lar facility, perhaps through prohibition. A muchsimpler means, however, is to adjust the price sig-nals that potential users of the highway facilityreceive. This can be done by imposing a toll on all

Pricing and Economic Efciency


14/24


users of the facility during congested periods cor-responding to the magnitude of the congestion ex-ternality; thus, the price that users face is equiva-lent to the marginal cost to society. At the optimumtoll level (t* in Exhibit 6), lower valued trips willbe shifted to other routes or time periods (or notmade at all), such that the new equilibrium trafcvolume is also the socially optimum level. This iscongestion pricing.

There are several important points to note withregard to efcient pricing. The rst is to emphasizethat the optimal congestion toll is intended to re-duce congestion on the facility but not eliminate it.To see why, consider Exhibit 7. Eliminating conges-tion in this example would require much more sig-nicant reductions in trafc volumes to V 1, justbefore congestion begins to build. Doing so with

pricing would require a toll of t 1, much greater thanthe optimal toll of t*. This would yield an inef-ciently low level of trafc on this facility, becausemany users would be priced based on those whovalue their travel more than it costs society, evenwhen accounting for the external costs that theyimpose on others.

The second point is to note that revenue genera-tion is not the primary purpose of congestion pric-ing, which is intended to better align supply anddemand to overcome the economic inefcienciesimposed by congestion. In reality, of course, anysuch tolling scheme could potentially generate sig-nicant revenues, whose use becomes a signicantpolicy and operational issue. The relationship ofcongestion pricing to user revenues and infrastruc-ture investment will be discussed in a separate vol-ume in this primer series.

For a given facility, the level of the optimal con-gestion toll and the resulting effect on congestionlevels is largely a function of the elasticity of demandfor travel. If demand is relatively inelastic, the opti-mal congestion toll will be high, and the effect onreducing congestion will be relatively low. Converse-ly, if demand is more elastic, then a greater reductionin trafc volumes and congestion can be achieved

with a smaller toll. As was discussed above, a keyfactor in determining the elasticity of demand is thequality and availability of alternatives to highway usein peak periods, such as transit or teleworking. Theimplication of this is that measures to improve suchalternatives can greatly improve the effectiveness ofefcient pricing in reducing congestion.

VARIABILITY IN DEMAND

AND PRICING

The level of the congestion toll is also a function ofthe level of demand. Although this is intuitively ob-vious, it has important implications for the designof congestion-pricing schemes and helps demon-strate its superiority to other forms of user charges.As was noted above, one of the key features of thedemand for highways is that it varies signicantlyby time of day as well as by location. This implies

Cost

VolumeV* V0

P 0

P*

User Cost

Marginal Social CostDemand

Exhibit 5. Marginal social cost and user cost.P = price. V = volume.

VolumeV*

t*

User Cost

Marginal Social CostCost

P*

Demand

Exhibit 6. Optimal congestion pricing.P = price. V = volume. t = toll.


15/24

E C O N O M I C S : P R I C I N G , D E MA N D , A N D E C O N O M I C E F F I C I E N C Y | 13

V* Volume

Cost

P*

V 1

P 1 User Cost

Marginal Social CostDemand

t*

t1

Exhibit 7. Inefficiently high congestion pricing.P = price. V = volume. t = toll.

that the optimal congestion tolls will vary over thecourse of the day on the basis of uctuations in de-mand. Suppose, however, that a at toll (e.g.,xed tolls and fuel taxes that do not discriminatebased on when travel occurs) must be charged thatdoes not vary in this manner. In theory, the at tollcould still be set to match the optimal toll duringcongested, high-demand periods. However, if thesame toll rate applies during uncongested, low-de-mand times (when there are no congestion exter-nalities and thus no optimal toll), trafc volumeswill be inefciently low in those periods. In fact,

because the tolls are the same throughout the day,some of that trafc might even move back into thecongested period. Because of the potential for thisto occur, the optimal toll during the peak periodwould be lower (and less effective at reducing con-gestion) if the tolls cannot vary by time of day.

This example can be extended to multiple de-mand periods. Suppose the congestion toll is struc-tured to be charged at a xed rate when it is col-lected but may be zero (i.e., not collected) duringoff-peak periods. To the extent that demand varies

during the congestion period (such as between theheight of rush hour and shoulder hours adjacent tothe peak), the same considerations would apply.Congestion tolls for the shoulder period would ei-ther be inefciently high or inefciently low. Theresult is that the more variability there is in the con-gestion-pricing scheme, the more effective it can beat reducing congestion and the closer it can come to

achieving an economically efcient out-come.

Similar considerations apply whenlooking at pricing across facilities in dif-ferent locations. Although trafc conges-tion can be quite severe in large urbanareas, there is much less congestion insmaller cities or rural areas. As a result, auser charge (such as fuel taxes) that isapplied uniformly to both congested anduncongested areas will not be nearly aseffective as congestion-based pricing inefciently allocating resources within theeconomy.

Another consideration in limited congestionpricing concerns alternate routes. Suppose there aretwo parallel roads between the same origin and des-tination, both subject to congestion, but only oneof them can be priced. If congestion tolls are onlycharged on the one facility, some of the trips thatare priced off that facility will shift to the unpricedfacility, exacerbating congestion there. Given thepossibility of this to occur, the optimal toll undersuch a situation will be lower, as will the impact onreducing congestion levels. An important implica-tion of this is that congestion pricing will be much

more effective if it is jointly applied to all majorhighway facilities in a congested corridor, ratherthan to only a subset of the alternative routes.

In reality, of course, highway networks are com-prised of a whole series of interconnected segments,supporting both alternative and connecting routes,as well as access to connections with other modes(such as train stations). The Federal HighwayAdministrations (FHWAs) Ofce of Operationshas been conducting research to examine the po-tential impact of congestion pricing when applied

to such real-world transportation networks.

CONGESTION PRICING AND

OTHER EXTERNALITIES

Highway travel creates other external costs for so-ciety, in addition to congestion. Perhaps the mostsignicant of these other impacts is the damage


16/24


caused to the environment and human health byhighway vehicle emissions. In a broad sense, themarginal social cost of highway use would includethese costs. Market and regulatory strategies fordealing with environmental externalities also exist,such as taxes on fuel or regulations on vehicle-emis-sions control systems and fuel content, and suchstrategies are the most effective way of dealing withthese particular externalities. However, to the ex-tent that current mechanisms do not fully capturethese environmental externalties, congestion pric-ing can help reduce their magnitude by reducing

excessive highway use (at least during congestedtravel periods).

TOLL-COLLECTION COSTS AND

THE ROLE OF TECHNOLOGY

The preceding discussion of the economic princi-ples underlying congestion pricing does not address

the cost of actually collecting such tolls from users.Indeed, although William Vickrey articulated thebasic theory of congestion pricing in his NobelPrize-winning work in the 1950s, its potential forwidespread application was limited by the availablemethods of collecting tolls. With manned toll-booths and the trafc-ow restrictions that theyimpose on roadways, it was a case of the cure beingworse than the disease.

Advancements in tolling technology over thepast two decades, however, have made variabletolling in urban cores and on limited-access high-

ways a viable option, and future developments inglobal-positioning satellite (GPS) and dedicatedshort-range communications technologies maymake it feasible to adopt congestion pricing on aneven wider basis. Two other volumes in this primerseries, Technologies That Enable Congestion Pricinand Technologies That Complement Congestion Pricing, describe these technologies in greater detail.


17/24


Forms of Congestion Pricing

The discussion above describes the classic econom-ic case for congestion pricing, based on mitigatingcongestion externalities. However, as used today,the term congestion pricing has come to be appliedmore broadly, to any projects involving tolls or oth-er charges for road use that vary by the level of de-

mand. The degree of variation in such applicationscan range from simple policies that apply differentprices in peak and off-peak periods (with no toll atall being sometimes applied in the off-peak) to ratesthat change within peak periods, often targeted atmaintaining a specic level of service. Variablecharges may be applied on the basis of a pre-setschedule (updated periodically) or dynamically(changing as often as every 36 minutes) in re-sponse to real-time changes in demand.

Congestion-pricing projects that involve tollsmay be categorized according to their extent of ap-plication, as follows:

Priced Lanes: Pricing is applied on a limited num-ber of lanes of a roadway, leaving other travel

lanes on the facility unpriced. The priced facility(commonly referred to as express lanes, managedlanes, or high-occupancy toll [HOT] lanes) istypically separated from the other lanes, with lim-ited access points. Examples of priced lanes in theUnited States include SR-91 in Orange County,CA; I-15 in San Diego; I-394 in Minneapolis; SR-167 in Seattle; and I-10 and US-290 in Houston.

Priced Roadways: Congestion pricing is appliedon all lanes of a roadway facility. This is fre-quently done on existing toll facilities that con-vert their xed toll regime to a system that variesby the time of travel. Examples include bridgesin Lee County, Florida; the Dulles Greenway inNorthern Virginia; and freeways in Singapore.

Zone-Based or System-Wide Pricing: Pricing isapplied on all roadways within a specied zone.Pricing can take the form of cordon pricing, inwhich vehicles are charged as they cross a cor-don line into the priced zone, or broader basedtolling that is applied to trips within the zone as

well. Projects of this type that have been imple-mented to date have typically focused on con-gested central business districts, such as those inLondon, Stockholm, and Singapore; however,future applications may include larger districts,perhaps encompassing entire urban areas.

The Congestion Pricing Overview primer includesmore information on the practical application ofthese different approaches.

Variable Pricing in Other Sectors

Although a relatively new concept to highway transportation, chargingdifferent prices in peak and off-peak demand periods is a well-establishedpractice in many other sectors and industries. This is especially commonin industries in which supply (e.g., the number of available hotel rooms orthe capacity of the water delivery system) does not vary signicantly indifferent demand periods. By charging higher rates during high demandperiods, proprietors are able to better allocate demand to optimize theutilization of the available capacity.

Examples of such common practices include higher rates for lodgingand other amenities in tourist areas during the high season, discounts

for afternoon showings at movie theaters, and evening and weekenddiscounts for telephone use. More recently, other industries have movedin this direction, including professional sports teams (which have beguncharging more for tickets to more desirable games, reecting long-stand-ing practices in the aftermarket for tickets) and electric utilities (in whichadvanced electronic meters now allow usage at different times of day tobe recorded).


18/24


Priced Lanes:The Economics of Multi-Class Service

The preceding discussion describes the fundamentaleconomic motivation for efcient pricing in itspure form. However, as noted earlier, the termcongestion pricing has come to be applied to a vari-ety of different practices of applying tolls or otherroad-user charges that vary based on time and space.One such form is tolled express lanes, in which afreeway is partitioned into two different classes(priced and general purpose), and tolls arecharged on the priced lanes to ensure that trafcremains free-owing on them. Although similar insome ways to the classical case of congestion pric-ing, the economic justication for offering multipleclasses of service is actually quite different.

HOT LANES

HOT lanes are a special case of tolled expresslanes, in which high-occupancy vehicles (HOV;including carpools, vanpools, and transit vehicles)are allowed to use the special lanes for free, where-as low-occupancy vehicles are required to pay atoll to use the lanes. To date, most HOT lane proj-ects in the United States have been conversionsfrom existing HOV facilities.

The purpose of HOV lanes is to encourage moreefcient use of highway capacity by increasing aver-

age vehicle occupancy, thereby accommodatinghigh levels of passenger travel while reducing totaltrafc volumes. The lanes operate by establishingminimum occupancy thresholds (typically two orthree passengers per vehicle) that are set highenough to keep trafc at low enough levels to en-sure that the lanes remain free-owing, even duringpeak demand periods. However, because these oc-

cupancy minimums must be set at whole numbers,the lanes often wind up with signicant excess ca-pacity available, even as the general-purpose lanesmay be severely congested.

The purpose of converting such facilities to HOTlanes is to make use of the excess capacity in theHOV lanes, while still preserving the incentives forcarpool and transit use (no toll and free ow). Forexample, several HOV-3 facilities in the UnitedStates, such as the reversible express lanes on I-395in Northern Virginia, were initially constructed asexclusive busways; however, because the busesalone did not fully utilize the available capacity,carpools and vanpools were soon allowed to sharethose facilities. HOT lanes simply extend this fur-ther down the vehicle-occupancy scale, while in-

troducing the element of pricing to manage demandmore precisely.Because most toll-paying users of the HOT lanes

are likely to shift from the other lanes, congestionon these lanes will be reduced and travel times willbe improved, whereas existing HOV users will seeno reduction in the quality of the service they re-ceive. The result is a pure gain to highway users.

Although many HOV lanes operate with excesscapacity during peak hours, this is not always thecase. In some places, growing trafc volumes have

resulted in HOV lanes that no longer provide reli-able, uncongested travel to HOVs, even if they didso when rst established. One approach to return-ing such facilities to free-ow conditions is to in-crease the minimum vehicle occupancy standardfor the special lanes, thereby reducing the numberof vehicles that qualify to use them. The excess ca-pacity that is created through such a move can then


19/24


make the lanes a candidate for HOT lanesin somecases, additional capacity for HOV lanes (or forHOV-to-HOT conversions) can be created throughrestriping or other improvements to roadway ge-ometry and operations. By taking these two actionssimultaneously (increasing occupancy standardsand conversion to HOT), a poorly performingHOV facility can be transformed into a highly func-tioning tolled facility, while still allowing users whowould otherwise be forced off the HOV lanes tocontinue using them by paying a toll.

BENEFITS FROM PRODUCT

DIFFERENTIATION

In addition to the capacity utilization benets af-forded by HOV-to-HOT lane conversions, expresslanes can provide highway users with multiple class-es of service quality from which to choose at differ-ent prices. This form of product differentiation, aseconomists refer to it, can produce signicant ben-ets for consumers. The key is that users of a facilitymay vary signicantly in the value that they place ontravel-time savings. For some users, reducing traveltimes for a particular trip may be quite valuable,whereas others may be less sensitive to such reduc-

tions.1

By partitioning the highway into tolled ex-press lanes and free general-purpose lanes, users aregiven the option of selecting the pricequality com-bination that best suits them, rather than beingforced into a one-size-ts-all solution.

It is important to note that the value of timesavings reects to the total value of all passengersin a vehicle, not just the driver. Thus, some of thehighest value trips are likely to be those in busesor other transit vehicles, even if the riders indi-vidually have lower values of time than single oc-

cupant vehicle drivers. In addition, the value oftime savings can vary from day to day, even forthe same user, depending on the time sensitivityand value of the activities being pursued at theusers destination (such as attending an impor-tant family function). This is reected in the ex-periences on express lane projects such as SR-91,where a large number of facility users use the ex-

press lanes on an occasional basis, when theirtrips are most critical.

An important component of the appeal thatexpress lanes can have for highway users is theirreliability. Variable pricing on express lanes (par-ticularly dynamic pricing based on real-time traf-c conditions) can be used to manage demand insuch a way as to virtually guarantee a free-ow-ing trip at all times, not just on average. Reduc-ing the likelihood and severity of unexpected de-lays can be very valuable to users and may be animportant component of their willingness to payfor the premium service afforded by expresslanes. In this way, the introduction of expresslanes can be viewed as a new product entirely (aguaranteed high-speed trip during peak hours),previously unavailable at any price (i.e., any pricebelow that of private helicopter taxi services),


20/24


that enhances consumer welfare. Such servicesdo exist, of course, for the most extremely high-valued trips, such as emergency trips to traumacenters or ferrying top business executives aroundmajor cities.).

IMPACTS ON CAPACITY AND DELAY

Although offering multi-class service can be appeal-ing to road users, it may also have some drawbacks.

When a road is partitioned into two facilities, it lim-its the ability of drivers to change lanes and ll gapsthat develop in the trafc stream. Separating theroadways may also require barriers or pavementmarkings that take up space that could be otherwiseused for shoulders or travel lanes. As a result, thecombined capacity of the two separated facilitieswill be lower than the overall capacity that could berealized without the partitioning.

Providing free-ow conditions on the express

lanes also requires tolls set sufciently high to limitthe number of vehicles that might want to switchfrom the free lanes. As a result, trafc and conges-tion levels may be quite high on the general-pur-pose lanes. Because delay functions are quite non-linear in the sense that each additional vehicle ona congested roadway causes more delay than theprevious one, adding more vehicles to the general-purpose lanes causes overall user costs on the com-bined facility to be higher than they would be iftrafc were more evenly allocated.

These capacity and delay-related drawbacks toproviding multiple classes of service on freewaysmake it difcult to make general statements as towhether providing express-lane services will en-hance or detract from overall economic welfare.Under some circumstances (particularly wherethere is signicant variation in the value of time fordifferent users), the gains from product differentia-tion may outweigh the increased congestion costson the general-purpose lanes, whereas in other cas-es the latter effect may dominate.


21/24

E C O N O M I C S : P R I C I N G , D E M A N D , A N D E C O N O M I C E F F I C I E N C Y | 19

Empirical Estimates of Congestion Tolls

Because congestion pricing is still a relatively newapproach for managing congestion, especially in theUnited States, and has yet to be implemented com-prehensively in its pure form, empirical estimatesof the magnitude of optimal congestion tolls arelimited. However, the results from two separate

studies of potential congestion-pricing applicationshave yielded interesting insights into the potentialof congestion pricing.

An analysis of future highway investment andperformance that used FHWAs Highway Econom-ic Requirements System (HERS) examined the po-tential impacts of implementing efcient pricing onall congested roads after 2020. The estimated con-gestion tolls averaged 34 cents per vehicle mile (in2006 dollars) on roads where it was implementedin the analysis, which were predominantly located

in major urbanized areas. Over 50 percent of thevehicle miles subject to congestion pricing in thescenario had estimated toll rates below 25 cents permile, whereas approximately 7 percent had tollrates over $1.00 per mile (see Exhibit 8).

By using a grant from the Value Pricing Pilot

Program, the Puget Sound Regional Council inSeattle recently completed a study (1) of the impactthat congestion-based tolling could have on travelbehavior. The project involved 275 householdswhose vehicles were equipped with GPS-basedtolling meters. Participants were subjected to a per-mile toll schedule on major arterials in the regionthat varied signicantly based on the time of day(with charges up to 50 cents per mile in the week-day afternoon peak) and the location of travel (withhigher rates applied on freeways). The study foundthat households made several modications to theirtravel patterns in response to the congestion tolls,including reductions in the number and length oftrips, choosing alternate routes and travel times,and switching to transit. On the basis of the resultsof the pilot program, the study estimated that afull-scale implementation of congestion pricing onthe areas road network could yield over $28 billionin net benets.

35.0%

30.0%

25.0%

20.0%

15.0%

10.0%

5.0%

0.0%

Exhibit 8. Distribution of toll rates per VMT (20202025).VMT = vehicle miles traveled.

$0.00$0.10

$0.10$0.25

Percent of Tolled VMT

$0.25$0.50

$0.50$1.00

>$2.00$1.00$2.00

Source: Report of the National Surface Transportation Revenue and Policy

Study Commission, Volume III, Paper 6C-04, January 2008.


22/24


23/24


For more information, contact:

Darren Timothy

Ofce of Innovative Program Delivery, HINFederal Highway Administration

U.S. Department of Transportation

1200 New Jersey Avenue, S.E.

Washington, DC 20590

Tel: 202-366- 4051

E-mail: [email protected]

Reference

1. Puget Sound Regional Council. (2008, April). Trafc choices study:Summary report. Retrieved on January 1, 2009, from http://psrc.org/projects/trafcchoices/summaryreport.pdf


24/24

Ofce of Transportation Management

Federal Highway Administration

U.S. Department of Transportation

1200 New Jersey Avenue, S.E.

Washington, DC 20590

Tel: 202-366-6726

www.ops.fhwa.dot.gov/siteindex.htm

November 2008

highway supply, demand & economic efficiecny

Documents