The Economics of Road Pricing An Introduction

1. The Economics of Road PricingAn Introduction Erik T Verhoef VU University Amsterdam e.t.verhoef@vu.nl

2. Road transport… • Undeniable great benefits • But also high cost for the users… • Automobile, fuel, maintenance • But also: time • … and to others in society • Congestion (3.5) • Environment (8.5) • Accident risks (14.5)

3. Attractiveness of pricing: volume and composition Pigou: (nearly) one century ago Excessive social cost Limited acceptability Attractiveness (2): in reality many more “behavioural margins”

4. Acceptability: revenues are central

5. A closer look at capacity • An optimallydesignedandpricedroadnetwork is undercertaintechnicalconditionsexactlyself-financing(Mohring & Harwitz, 1962) • Potential gains from applying: • Efficiency • Acceptability • Transparancyin financing • Transparency in decision making on investments

6. So: Columbus’ Egg? • At least4 “translation issues” fromtheorytopractice • “Certainconditions”: neutraleconomies of scale in infrastructuresupply • Empirical evidence from US seems encouraging, but no thorough study in the BeNeLux (yet) • Onlycongestion component of road pricing • Revenues = capacity cost ≠ investment cost • Indivisibilities: minimum useful capacity (“1 lane”) is binding in periphery • No congestion, so no revenues…

7. Acceptabilty of dynamic pricing • Vickrey (1969) • Considered second behavioural margin: departure time dynamics • Basic insights • It is not total peak demand but its time profile within the peak that creates traffic jams at bottlenecks • That profile arises because travellers trade off two cost components • Travel delays – as considered by Pigou • Schedule delays: cost of inconvenient timing • In dynamic equilibrium: the two balance, with varying importance

8. Example: Coenplein (2002) Capacity Optimal pricing duplicates time-pattern of travel delays and flattens departure pattern to match capacity Notable properties of optimum Total number of cars does not change Arrival moments at work doe not change

9. Search for more acceptable price instruments • Pricing: efficient, effective, but low acceptability • Rewarding (as in “Spitsmijden”: “Peak Avoidance”) • Popular, effective • But (1): Financially unsustainable (rewards!) • But (2): Less efficient (induced or latent demand problem) • Hybrid solutions? • Budget neutral: tradable permits

10. “Spitsmijden” / “Peak Avoidance” • Series of experiments • Road • Public transport • “User paid” instead of “user pays”: • Rewards for avoiding peak travel • Typical characteristics of experiments • Automated (GPS) detection of vehicles or individuals • Participants invited on the basis of observed peak behaviour • Financial incentive of around € 3,- to avoid peak travel

11. Effectiviness: SpitsMijden I (Zoetermeer, 2006, 340 participants) Trips by clock time 50% reduction in peak trips Share With reward Without reward Clock time

12. SpitsMijden in the Train(Abonnements, 2012-2013, 467 participants) 22% reduction in peak trips Post-measurements Pre-measurements With rewards Off-peak Evening peak Morning peak

13. Prospects for tradeable permits… • Acceptability: likely to be higher than for pricing • Efficiency • Likely to be higher than for rewarding (latent demand) • But… will it work?

14. Fri Fri Thu Thu Wed Wed Tue Tue Mon Mon

15. X Fri Fri Thu Thu X Wed Wed Tue Tue Mon Mon

16. Fri Thu Fri Thu Wed Tue Tue Mon Mon

17. € Fri Thu Fri Thu Wed Tue Mon Mon

18. Rewarding 0 + € € Toll – € € € € € – € € € Trade – € + €

19. Design – an example • Behavioural challenges • Avoid undesired speculation • Never more trip-coins in possession than remaining choices • Small transaction fee to avoid manipulation of trip-coin price • Avoid cheating • Automated purchase of trip-coin when needed for a choice made, with a mark-up • Manage transaction costs • Trading with bank • Avoid undesired speculation

20. Lab-in-the-field (U-Smile) • Set-up replicates a permit scheme that is as close as possible to the above case • Virtual / serious gaming environment • No interference with actual mobility behaviour • Pay-off in true money, depending on performance in the game • Parking experiment: parking charge vs parking permit

21. Lab-in-the-field – the looks Trading Parking choice Day tariffs Permit price Tariff Date Permit price My Budget My Budget My Parking Permits My Parking Permits Buy Sell Make your choice Day tariff Permit Parking choice Parking choice Trade Trade

22. Expected equilibrium • Participants received on average 3 permits • 2 in the one week, 4 in the other, to rationalize trading • Participants had full information and could in principle compute the expected equilibrium, which looks like: • All individuals: • Use a permit on days with parking tariffs of €5, €4 and €3 • Pay the tariff on days with €2 and €1 • Equilibrium permit price could be anything between €2 - €3

23. Permit price dynamics

24. Instantaneous rationality (1)

25. Responses Tradeable parking permits could in reality sometimes be a good alternative for paid parking Participating took me little time or effort Scale: 1 Totally Disagree - 5 Totally Agree

26. Responses I found it difficult to determine whether it was best for me to buy, sell, or not trade a parking permit I found it difficult to determine the best parking choice Scale: 1 Totally Disagree - 5 Totally Agree

27. Conclusion • Pricing • Efficient for many reasons • Limited acceptability • Revenue use: capacity costs • Dynamics: smaller price impact • Rewarding • Tradable permits