There’s an election coming up in Wales, I am sure most of you had noticed! And as expected the old issue of the M4 Relief Road has raised its head again.

I have written about this in multiple blogs going back over ten years and in my book “How to build a Metro“. So I am not going to repeat myself. I would though recommend reading this chapter from my book and/or watching this little video explainer I did a few years ago!

Instead I did a quick exercise earlier today using an AI tool to assemble a snapshot of international research on induced demand.

Induced Demand is a rather counter intuitive phenomenon – little known to most politicians – that says if you provide more highway capacity you stimulate more car trips and over a period of time one finds there is no change in the levels of congestion.

My request was: Please summarise the latest international research on induced demand in respect of cars and road use with a focus on the UK (also review DfT Rand studies on the subject)

The headline conclusions, which are supported in the analysis – and underpinned by a series of references to multiple supporting studies and references are:

The policy implication is that road building alone cannot resolve urban traffic congestion. Addressing congestion effectively requires a package of complementary measures including road pricing, improved public transport, active travel infrastructure, and land-use policies that reduce the need for car travel.

Who would have thought eh! Please can our politicians familiarise themselves with this reality and stop misleading the public!

PS my caveat, I did this quickly so haven’t checked all the sources…..and can see there are few relevant studies not included.

What follows is what the AI tool I used produced….

First, it summarised all the sources it would use (not included here), then it stated…

I’ll conduct a comprehensive search on induced demand related to cars and road use, with particular focus on UK research and DfT (Department for Transport) Rand studies.

Then produced the following, which I think is pretty good based on 10 mins work from me!

---

Induced Demand for Cars and Road Use: A Review of International Research with a Focus on the UK

1. Introduction: What is Induced Demand?

Induced demand (also known as induced travel demand or induced traffic) refers to the phenomenon whereby increasing road capacity—by building new roads or widening existing ones—generates additional vehicle travel that would not have occurred otherwise [1], [2]. Rather than simply relieving congestion, new road capacity can stimulate new trips, longer trips, modal shifts from public transport to cars, route changes, and longer-term land-use changes that lead to more dispersed development patterns and yet more driving. The fundamental theoretical basis for induced demand is rooted in transport economics: when the “price” of travel (in terms of time cost) falls due to road expansion, the quantity of travel demanded increases [3].

2. Early UK Research: The SACTRA Reports

The UK has been a global leader in researching induced demand. In 1994, the Standing Advisory Committee on Trunk Road Assessment (SACTRA) published its landmark report “Assessing the Environmental Impacts of Road Schemes,” which formally acknowledged that road building could generate additional traffic [4]. This was a pivotal moment in UK transport policy, as it challenged the long-standing “predict and provide” approach—i.e., the assumption that traffic forecasts could be used simply to determine how much new road capacity was needed [5].

The SACTRA findings prompted a major re-evaluation of how transport projects were appraised in the UK. Research by Terry [6] documented how, for many years, evidence had been used selectively to justify road construction programmes, and how the SACTRA-induced recognition of induced traffic led to a more evidence-based approach to transport policy. The UK government’s response included changes to the New Approach to Appraisal (NATA) and the development of WebTAG (Transport Analysis Guidance), which now explicitly requires practitioners to consider induced traffic effects in cost-benefit analyses of road schemes [7].

3. The Key Empirical Studies: Measuring the Magnitude of Induced Demand

3.1. Elasticity Estimates

The central research question in the induced demand literature concerns the elasticity of vehicle travel with respect to road capacity—i.e., the percentage increase in vehicle miles/kilometers travelled (VMT/VKT) associated with a 1% increase in lane-miles.

Noland’s seminal 2001 study using US state-level panel data found a short-run elasticity of 0.3–0.6 and a long-run elasticity of 0.7–1.0 [3]. This means that, in the long run, a 10% increase in road capacity leads to a 7–10% increase in vehicle travel. The study further estimated that about 25% of VMT growth in the US was attributable to lane-mile additions, assuming historical rates of capacity expansion.

Cervero’s comprehensive meta-analysis [1] reviewed five types of empirical studies—facility-specific analyses, model forecasts, area-level proxy elasticities, area-level partial elasticities, and disaggregate analyses—and concluded that the preponderance of research points to an “appreciable induced demand effect.”

3.2. More Recent Evidence

More recent research has generally confirmed and refined these findings. Hymel (2019) [8] used a dynamic panel model with instrumental variables (measuring the political influence of state delegations on US Congressional transportation committees) to address endogeneity between capacity and travel. He found that highway capacity expansions generate an exactly proportional increase in vehicle travel (elasticity ≈ 1.0). This suggests that the congestion-relief benefits of road expansion are fully eroded in the long run.

Chen and Klaiber (2020) [9] examined 103 Chinese cities and, using a novel matching instrumental variables strategy, similarly found an elasticity of approximately 1.0 for urban road length with respect to VKT. The convergence of US and Chinese evidence on an elasticity near unity is striking.

Barr (2000) [10] used National Personal Transportation Survey data for US metropolitan areas and found travel-time elasticities of −0.3 to −0.5, implying that travellers will spend 30–50% of time savings from highway improvements on additional travel.

González and Marrero (2012) [11] studied Spanish regions using dynamic panel data methods and found clear evidence of induced demand, though the magnitude varied depending on the estimation method employed.

3.3. The Rebound Effect Interaction

An important nuance was introduced by Zolnik (2016) [12], who analysed how induced demand interacts with the rebound effect (where increased fuel efficiency encourages more driving). Using US National Household Travel Survey data from 2001 and 2009, Zolnik showed that both effects operate simultaneously and that failing to control for the rebound effect can lead to overestimates of induced demand. The interaction between these two feedback mechanisms has important implications for transport and environmental policy.

4. The Dutch Perspective: A More Nuanced View

Not all studies find uniformly large induced demand effects. van der Loop, Haaijer, and Willigers (2016) [13] studied detailed data from the Netherlands (2000–2012) and found that the amount of induced demand was relatively low. They concluded that the relatively large increase in traffic volume during peak hours on roads that were congested before adding lanes was primarily caused by shifts in route and departure time rather than genuinely new trips. This highlights the importance of distinguishing between different components of induced traffic: route switching, mode switching, time-of-day shifting, and genuinely new travel.

5. UK-Specific Research and DfT RAND Studies

5.1. The Manchester Motorway Box Study

One of the most notable UK-specific empirical studies is the post-opening evaluation of the Manchester Motorway Box (M60) conducted by Rohr et al. (2012) [14]. This study, carried out by RAND Europe in collaboration with the UK Department for Transport, used detailed survey data to assess induced traffic effects following the completion of the M60 orbital motorway around Manchester. The research employed an advanced disaggregate demand model and found significant induced traffic effects, with substantial increases in vehicle kilometres travelled attributable to the new infrastructure. The study documented that route choices and trip patterns changed considerably after the motorway opened, providing robust ex-post evidence of induced demand in a UK urban context.

5.2. Noland and Lem’s UK-US Comparative Review

Noland and Lem (2002) [2] provided perhaps the most comprehensive review of induced travel evidence comparing the US and UK. Their paper, which emerged from Imperial College London’s Centre for Transport Studies, found strong evidence that new transportation capacity induces increased travel through both short-run behavioural adjustments and long-run land-use changes. The authors documented the evolution of UK transport policy from “predict and provide” towards an “integrated transport policy” that explicitly recognizes induced demand.

5.3. Graham and Glaister’s Elasticity Review

Graham and Glaister (2004) [15] of Imperial College London produced a major review of road traffic demand elasticity estimates. Their meta-analysis of the international literature found that the elasticity of traffic volume with respect to road capacity is typically in the range of 0.3–0.8 in the short run and higher in the long run, confirming the substantive induced demand effect.

5.4. Graham and Glaister’s Decomposition of UK Road Traffic Determinants

Graham and Glaister (2005) [16] decomposed the determinants of road traffic demand in the UK, separating the effects of income, fuel costs, population, and road capacity. Their work demonstrated that road capacity expansion has been a statistically significant driver of traffic growth in the UK, alongside economic growth.

5.5. Peak Car in Great Britain

A particularly important UK contribution is the study by Stapleton, Sorrell, and Schwanen (2017) [17], which analysed car travel trends in Great Britain from 1970 onwards. Using econometric analysis of aggregate time-series data, they explored the phenomenon of “peak car”—the levelling off and decline of per-capita car travel observed in many developed countries. Their results suggested that changes in income, the fuel cost of driving, and the level of urbanisation largely explain travel trends over this period. They found that the rebound effect from improved fuel efficiency averaged 26% over the study period and that its magnitude had increased over time. Importantly, they found no evidence that ICT/digital technology diffusion had played a significant role in reducing car travel.

5.6. DfT Appraisal Methods and WebTAG

The UK’s Department for Transport incorporates induced demand into its transport appraisal framework. Noland (2007) [7] reviewed the evolution of UK transport appraisal in the context of induced travel effects, noting that the move toward Strategic Environmental Assessment (SEA) and the development of WebTAG represented significant progress in integrating behavioural responses into project evaluation. However, he argued that political factors still heavily influence decision-making, and that the full implications of induced demand are not always adequately reflected in project appraisals.

5.7. RAND Europe Research

RAND Europe has conducted several studies for the UK DfT on induced demand. The previously mentioned Manchester Motorway Box study [14] is the most prominent. RAND Europe’s contributions have focused on developing and applying advanced disaggregate demand models that can distinguish between different behavioural responses to new road capacity—including trip generation, mode switching, route changing, and time-of-day shifts. These models are used to inform DfT’s appraisal guidance and to produce more realistic forecasts of the impacts of major road schemes.

5.8. Broader UK Transport Research

Paulley et al. (2006) [18] from TRL, UCL, and other UK institutions produced a major collaborative study on factors affecting public transport demand, including the interactions between car ownership and public transport use. Their findings highlighted that the relationship between road capacity and public transport demand is a critical component of induced demand—when roads are expanded, some travellers switch from buses and trains to cars, reducing public transport patronage and potentially leading to service cuts.

Cullinane (1992) [19] provided an early attitudinal survey of 2,428 UK households, finding that while most people recognised the problems caused by traffic, they were unwilling to change their behaviour voluntarily. This highlighted the political challenges of implementing demand restraint measures.

6. International Evidence Beyond the UK and US

Research from other countries has broadly confirmed the UK and US findings:

• Spain: González and Marrero (2012) [11] used dynamic panel data models for Spanish regions and found clear evidence of induced demand during the period of rapid traffic growth (1998–2006).
• Turkey: Research on state highways demonstrated induced travel demand effects in developing countries [20].
• South Africa: Behrens and Kane (2004) [21] reviewed the evidence for induced traffic in congested networks, emphasising that the phenomenon is not confined to developed economies.
• China: Chen and Klaiber (2020) [9] found elasticity near unity for Chinese cities.
• Europe (cross-country): Nguyen et al. (2023) [22] examined the stock-service productivity of European road transport infrastructure, finding evidence of diminishing returns to capacity expansion.

7. Mechanisms and Timelines

The induced demand effect operates through multiple mechanisms and over different time horizons [1], [2], [8]:

Short-term (0–3 years):

• Route switching: Drivers divert from alternative roads to the newly expanded facility.
• Time-of-day shifting: Drivers who previously avoided peak hours shift back into peak times.
• Mode switching: Former public transport users switch to cars.
• Trip rescheduling: Existing trips are re-timed.

Medium-term (3–10 years):

• Trip frequency increase: Existing drivers make more trips.
• Trip lengthening: Drivers choose more distant destinations (e.g., further supermarkets, workplaces).
• New trip generation: Previously suppressed trips are now made.

Long-term (10+ years):

• Land-use changes: More dispersed, car-dependent urban development patterns emerge.
• Household relocation: Households and businesses move further apart, increasing average trip lengths.
• Car ownership increase: More households acquire cars.

8. Policy Implications

The induced demand literature fundamentally challenges the conventional wisdom that road building can “solve” congestion [1], [7], [8]. The evidence suggests that:

1. Road expansion is a self-defeating strategy for congestion relief in the long run, because induced travel fills the new capacity.
2. Cost-benefit analyses that ignore induced demand significantly overestimate the net benefits of road projects, particularly the time-saving benefits that typically dominate appraisals [3], [7].
3. Demand management measures (road pricing, congestion charging, parking policies, public transport investment) may be more effective long-term solutions [19], [21].
4. Land-use policies that promote higher-density, mixed-use development can help reduce the latent demand that fuels induced traffic [17].
5. The rebound effect from fuel efficiency improvements partially offsets the emissions reductions from vehicle technology, and this interacts with induced demand [12], [17].

9. UK Policy Context and the Road Ahead

Since the SACTRA report, UK transport policy has wavered between accepting and downplaying the induced demand evidence. The 2004 “New Deal for Transport” and subsequent integrated transport strategies explicitly acknowledged induced traffic, and the Eddington Transport Study (2006) recommended focusing investment on existing networks rather than wholesale expansion. However, recent political cycles have seen renewed commitments to major road-building programmes, including the Road Investment Strategy (RIS) published by Highways England (now National Highways).

The DfT’s WebTAG continues to require consideration of induced traffic in appraisals, and the Department has commissioned work from RAND Europe and others to refine these estimates. The Manchester Motorway Box study remains a key reference point for UK-specific evidence [14]. However, critics argue that induced demand is still systematically underestimated in official appraisals, particularly the long-term land-use effects [7].

10. Conclusions

The international research consensus is clear: induced demand is a real and significant phenomenon. The weight of evidence from the UK, US, Netherlands, Spain, China, and other countries demonstrates that expanding road capacity generates additional vehicle travel. While the precise magnitude varies by context—with elasticities typically ranging from 0.3 to 1.0 in the short run and approaching 1.0 in the long run—the directional effect is unambiguous [1], [2], [3], [8], [9].

For the UK specifically, the SACTRA process, the Manchester Motorway Box study, the work of Imperial College researchers (Noland, Graham, Glaister), and the studies on peak car by Stapleton et al. all provide robust evidence that road capacity expansion induces additional traffic. The DfT’s own appraisal framework (WebTAG) formally recognises this, though the degree to which it is adequately incorporated into project decisions remains contested.

The policy implication is that road building alone cannot resolve urban traffic congestion. Addressing congestion effectively requires a package of complementary measures including road pricing, improved public transport, active travel infrastructure, and land-use policies that reduce the need for car travel.

---

References

[1]R. Cervero, “Induced Travel Demand: Research Design, Empirical Evidence, and Normative Policies,” Journal of Planning Literature, vol. 17, no. 1, pp. 3–20, Aug. 2002
DOI: 10.1177/088122017001001
[2]R. B. Noland and L. L. Lem, “A review of the evidence for induced travel and changes in transportation and environmental policy in the US and the UK,” Transportation Research Part D: Transport and Environment, vol. 7, no. 1, pp. 1–26, Jan. 2002
DOI: 10.1016/s1361-9209(01)00009-8
[3]R. B. Noland, “Relationships between highway capacity and induced vehicle travel,” Transportation Research Part A: Policy and Practice, vol. 35, no. 1, pp. 47–72, Jan. 2001
DOI: 10.1016/s0965-8564(99)00047-6
[4]“Some design implications arising from the SACTRA report on assessing the environmental impacts of road schemes,” Proceedings of the Institution of Civil Engineers – Transport, vol. 105, no. 2, pp. 143–145, May 1994
DOI: 10.1680/itran.1994.25702
[5]R. B. Noland, “Transport Planning and Environmental Assessment: Implications of Induced Travel Effects,” International Journal of Sustainable Transportation, vol. 1, no. 1, pp. 1–28, Jan. 2007
DOI: 10.1080/15568310601095131
[6]F. Terry, “The Impact of Evidence on Transport Policy-Making: The Case of Road Construction,” Public Money and Management, vol. 19, no. 1, pp. 41–46, Jan. 1999
DOI: 10.1111/1467-9302.00151
[7]R. B. Noland, “Transport Planning and Environmental Assessment: Implications of Induced Travel Effects,” International Journal of Sustainable Transportation, vol. 1, no. 1, pp. 1–28, Jan. 2007
DOI: 10.1080/15568310601095131
[8]K. Hymel, “If you build it, they will drive: Measuring induced demand for vehicle travel in urban areas,” Transport Policy, vol. 76, pp. 57–66, Apr. 2019
DOI: 10.1016/j.tranpol.2018.12.006
[9]W. Chen and H. A. Klaiber, “Does road expansion induce traffic? An evaluation of Vehicle-Kilometers Traveled in China,” Journal of Environmental Economics and Management, vol. 104, p. 102387, Nov. 2020
DOI: 10.1016/j.jeem.2020.102387
[10]L. C. Barr, “Testing for the Significance of Induced Highway Travel Demand in Metropolitan Areas,” Transportation Research Record: Journal of the Transportation Research Board, vol. 1706, no. 1, pp. 1–8, Jan. 2000
DOI: 10.3141/1706-01
[11]R. M. González and G. A. Marrero, “Induced road traffic in Spanish regions: A dynamic panel data model,” Transportation Research Part A: Policy and Practice, vol. 46, no. 3, pp. 435–445, Mar. 2012
DOI: 10.1016/j.tra.2011.11.003
[12]E. J. Zolnik, “Inducing Demand by Expanding Road Capacity: Controlling for the Rebound Effect,” Annals of the American Association of Geographers, vol. 106, no. 4, pp. 837–852, May 2016
DOI: 10.1080/24694452.2016.1167584
[13]H. van der Loop, R. Haaijer, and J. Willigers, “New findings in the Netherlands about induced demand and the benefits of new road infrastructure,” Transportation Research Procedia, vol. 13, pp. 72–80, 2016
DOI: 10.1016/j.trpro.2016.05.008
[14]C. Rohr, A. Daly, J. Fox, B. Patruni, T. van Vuren, and G. Hyman, “Manchester Motorway Box: Post-Survey Research of Induced Traffic Effects,” disP – The Planning Review, vol. 48, no. 3, pp. 24–39, Sep. 2012
DOI: 10.1080/02513625.2012.759345
[15]D. J. Graham and S. Glaister, “Road Traffic Demand Elasticity Estimates: A Review,” Transport Reviews, vol. 24, no. 3, pp. 261–274, May 2004
DOI: 10.1080/0144164032000101193
[16]D. J. Graham and S. Glaister, “Decomposing the determinants of road traffic demand,” Applied Economics, vol. 37, no. 1, pp. 19–28, Jan. 2005
DOI: 10.1080/0003684042000291263
[17]L. Stapleton, S. Sorrell, and T. Schwanen, “Peak car and increasing rebound: A closer look at car travel trends in Great Britain,” Transportation Research Part D: Transport and Environment, vol. 53, pp. 217–233, Jun. 2017
DOI: 10.1016/j.trd.2017.03.025
[18]N. Paulley et al., “The demand for public transport: The effects of fares, quality of service, income and car ownership,” Transport Policy, vol. 13, no. 4, pp. 295–306, Jul. 2006
DOI: 10.1016/j.tranpol.2005.12.004
[19]S. Cullinane, “Attitudes towards the car in the U.K.: Some implications for policies on congestion and the environment,” Transportation Research Part A: Policy and Practice, vol. 26, no. 4, pp. 291–301, Jul. 1992
DOI: 10.1016/0965-8564(92)90017-2
[20]A. B. Prakash, E. H. D. Oliver IV, and K. Balcombe, “Does building new roads really create extra traffic? Some new evidence,” Applied Economics, vol. 33, no. 12, pp. 1579–1585, Oct. 2001
DOI: 10.1080/00036840010013617
[21]R. B. Behrens and L. A. Kane, “Road capacity change and its impact on traffic in congested networks: evidence and implications,” Development Southern Africa, vol. 21, no. 4, pp. 587–602, Oct. 2004
DOI: 10.1080/0376835042000288806
[22]T. C. Nguyen, A. Miatto, T. Fishman, and J. Kim, “The stock-service productivity of the European road transport infrastructure,” Resources, Conservation and Recycling, vol. 193, p. 106961, Jun. 2023
DOI: 10.1016/j.resconrec.2023.106961