Per-day insurance and cognitive dissonance

The Italian insurer Sara offers a product called Sarafree. It is a form of car insurance that has to be activated (by sms) by the driver the day before he or she wants coverage, ie before the car will be used. A nice idea for those that seldom drive.

It has a feature which appeals to people trying to influence mobility behaviour: it forces people to re-enter a kind of decision process, instead of sticking with a decision they made earlier. The phenomenon is a form of cognitive dissonance, I was told yesterday, and it entails the way people look at past decisions.

For example, if two otherwise equal persons with the same mobility 'demand' make different decisions (say, one decides to buy a car and the other decides not to), and you ask the car-owner about his or her experiences, he or she might say: well, I sure am glad I bought that car. The buses are just too crowded and smelly for me. The non-car owner might say: I enjoy not having to be attentive to the traffic all the time, I love to prepare things for work while on the move, etcetera.

Of course, there is some truth to both statements. But it is unlikely to be the whole truth about how they really feel, part of it is justification for a decision taken in the past. It is a process thought to make people feel a little happier about themselves, and about the decisions they made. Also, if we constantly had to re-enter all kinds of decision processes in general, normal life would grind to a halt. Partly, that is where things like habits come from.

The point is that the car owner, having once made his decision to own one, will not easily re-enter the decision process leading to using the car yes or no, because he or she has paid a lot of fixed costs for it, and will no longer consciously weigh alternative options.

Some critics of the PAYD idea make that same point. They say: it's not only a financial issue. If PAYD is introduced and people will be charged or credited according to mileage once a year, their behaviour is not likely to change much, because yearly billing does not constitute a 'trigger' to enter the decision process what type of transport to use anew every time a trip is planned.

Sara's product seems to remedy that. The need to remember to insure your car every time you plan to use it looks like a perfect 'trigger'.

All in all, even when contemplating market-based measures or products to influence mobility behaviour, economical 'marginal analysis' alone is not enough. Behavioural analysis is needed also.

Information on SaraFree insurance can be obtained from: http://www.sara.it/ProdottoPres.asp?idProd=9

For those of us who, like me, are not proficient in Italian: go to freetranslation.com, copy and paste the aforementioned URL into the section 'free website translator' and read the info in, well, a kind of English.

Christof

The 'but for' and the 'what if'...

As I read through a little more research information and theory surrounding PAYD each day, I cannot help but feel a need for discussion. So far, most of what I found reflects a broad consensus on two things; one: you cannot argue with the rationality of implementing PAYD when viewed from the angle of societal benefits, two: The only things keeping PAYD from being the most common form of vehicle insurance are: implementation cost, either because of the necessary technology or because of odometer reading schemes, and added uncertainty over annual premium income for insurance companies.

It would be nice to read something from people who take a different view. In the past week, I came across two thoughts that may begin to constitute an argument.

I encountered the first when I read Aaron S. Edlin's article: Per Mile Premiums for Auto Insurance (published in: Economics for an imperfect world. Essays in Honor of Joseph E. Stiglitz, Arnold, Greenwald, Kanbur and Nalebuff, The MIT Press, Cambridge, Massachusetts/London, England, 2003). It runs: (...)"If insurance company C is able to reduce the driving of its insureds, although it will save on accident payouts, substantial 'external' savings will be realized by other insurance carriers and their insureds, who will get into fewer accidents with C's insureds. These externalities follow from Vickrey's observation that if two drivers get into an accident, even the safer driver is typically a 'but for' cause of the accident in the sense that had the driver opted for the metro, the accident would not have occurred"(...)

This would mean that insurance companies implementing PAYD would not only face implementation cost and transfer of benefits to society instead of to their own bottom-line, but also a transfer of benefits into the pockets of the competition.

Incidentally, this view of accident externalities is also taken in the article Distance-Based Vehicle Insurance As a TDM Strategy,(Litman, version 1 december 2004, as published on www.vtpi.org, p. 5), where the same line of thought is used to show that "Since about 70% of crashes involve multiple vehicles, each 1.0% mileage reduction should reduce total crash costs by 1.7%.(...)" So the externality effect seems to increase expected societal benefits while worsening the business case for offering payd for an individual insurance company... what a pity!

The second thought was suggested to me by a workshop participant whom I met in The Hague last week. He said: What if PAYD reduces insurance cost for those who drive little, and what if those who drive little don't drive more because of budgetary constraints (which is often assumed)? Why wouldn't those who drive little spend their new-found savings by driving more, not less? Wouldn't these extra kilometres at least partly offset the reductions of those who drive a lot?

It set me thinking. If so, that might reduce the societal benefits of payd considerably...

Well, the proof of the pudding is in the eating. It would be nice to test some of the predictions made by payd advocates in practice. And since several projects are in progress already (by, amongst others: Progressive Insurance in Texas, Norwich Union in the UK, and Polis Direct in the Netherlands), time will tell.

In the meantime, suggestions for sources on alternative views on payd are more than welcome! I also want to thank those that mailed me already!

Christof

Front page news

Last saturday's issue of 'de Telegraaf' (The Netherlands' biggest newspaper) ran the main headline: 13.000 doden door smog (13.000 deaths through smog). It said that every year, in the Netherlands, up to 13.000 people die prematurely (if that's the term to use here...) because of the continuous exposure to harmful substances in the air. In all of Europe, over 300.000 people die prematurely for the same reason. This is primarily attributable to the inhalation of particles, from exhaust fumes and other combustion processes. Average life expectancy is down by some nine months Europe-wide, according to research done for the project Clean air for Europe. Also, total damages in economical terms would amount to a staggering 80 billion Euros per year, Europe-wide.

The European Commission is due to publish further research results this week.

Christof

Quotes

"The manner in which [auto insurance] premiums are computed and paid fails miserably to bring home to the automobile user the costs he imposes in a manner that will appropiately influence his decisions" (William Vickrey, economist and 1996 Nobel Prize laureate)

"Under a per-mile premium system, the basic unit of [risk] exposure would shift from the car-year to the car-mile" (Aaron S. Edlin, economist at the University of Berkeley, Ca.)

"The very word variabilisation is mentioned in the Van Dale (the leading dutch definiton dictionary, CDW) only in conjunction with the words car costs, car taxation and gasoline price level" (Joop Goos, director of the dutch traffic safety institute (rough translation))

Christof

The University of the Free

In earlier contributions to this blog, I mentioned my plan to attend a workshop on how to prepare a final research paper. It was offered by the Vrije Universiteit Amsterdam (Free University of Amsterdam), and it was to happen at february 22nd. With some effort, I managed to get that day off, and I called to Amsterdam february 15th. There were only four places still available in that workshop, a guy from Vrije Universiteit told me, but if I were to show up thursday morning (17th) 10am at the desk of the student service center and paid in cash for participation, all would be well. It was not possible to make an account transfer, so I decided to get up early, pay for buses and trains and make the 2,5 hr journey from my home to the desk I was supposed to pay at. Behind it I found a nice lady. She looked in her computer, told me that the workshop was annulled, that she could not tell me why but supposed it was because of lack of participants, and that there was no alternative on offer. I realised that I'd taken a risk, and lost.

Well, you cannot win 'em all, can you...?

Christof

Workshop in The Hague

On February 17th, a workshop was organised in the Dutch Insurer's Association's building. It was organised by TNO, the dutch institute for technical-scientific research, and it revolved around PAYD insurance. Todd Litman, the chairman of the Victoria Transport Policy Institute and the world's best known researcher on (and proponent of) PAYD insurance, was there, as well as representatives of every major dutch motor insurance company (including PAYD pioneers Polis Direct), representatives from the dutch motorists' association, from the transport ministry, and from the Waarborgfonds, the dutch motor insurance guarantee fund.

For me, it was a unique chance to hear those in the know on the subject, to follow discussions, to see presentations and to shake hands and trade cards. It was very inspirational, although the 'golden' lead to exactly that piece of information I want to find out with my research still has to show up...

Nevertheless, it is very good for one's understanding of certain research to discuss it with others that have also read it. And it's very clarifying also to talk about it with people with diverse professional interests in developments on the subject. I want to thank TNO for inviting me, and all those present there for their kind supply of information!

Right now I am reading a guide to the setup of a scientific research project. I realise that my education so far has lacked a practical and coherent enough approach to the setup of research and that I have trouble applying edifying principles to petty reality. My own fault entirely, needless to say, I just need some extra help with that. I hope to bridge some of the gap with the book I am reading (Methods and techniques, by Baarda and De Goede, Stenfert Kroese, second edition, Houten(NL), 1997) and the remainder by talking to yet more experts. I will meet with a professor, an insurance professional and a traffic psychologist soon. I hope they will provide me with both useful suggestions for literature and suggestions on what 'unknown' bit to carve out and make a 'known' of with my research.

Christof

Been there, done that (the empirical step forward...)

Yesterday, I had a phone-talk with one of my university mentors. I explained to him what I wanted to do, and one thing he said was: well it is good to gather empirical data on the subject and to make a critical synthesis, but that is not enough to base a graduation research project on. You should try to identify a question, issue or problem that has not yet been solved, preferably not even been researched much, gather data on that, complete an empirical research cycle, and substantiate your motives for taking the inductive reasoning step at the end of that cycle.

Of course he is right. Much of what I proposed as research questions so far has been researched already. Maybe not as exhaustive as possible, but rather thoroughly nonetheless.

Therefore I will search my source material for suggestions for further research, I will attend a workshop on preparing a graduation research paper and I am so fortunate as to meet some experts shortly, who I hope can give me further suggestions. And then I can try to select what exactly it is that I really want to find out, and how to go about finding it out. I will be meeting with my mentor March 1st, so (fingers crossed!) I will have my research proposal ready by then...

Christof

Surging media attention: Vicenza

Today, the dutch news program Twee Vandaag featured two items covering environmental issues (partly) linked with mobility. One was an item on oil getting ever dearer, and the problems that this can create in terms of our ability to drive our fossil-fueled cars as much as we like, and another was about a measure taken by the mayor of Vicenza, an Italian city: for four days it was forbidden to drive cars in the city.

Below is a quote, dutch and english (well, I tried to translate)

"Het gemeentebestuur van het 'autominnende' Vicenza in Noord- Italië vatte uit milieu-overwegingen het plan op om gedurende vier dagen al het autoverkeer te verbieden.

Hoe groot de noodzaak is van nieuwe, schone energiebronnen merken we elke dag, in grote delen van Europa. Na het autoverbod in Vicenza is de EU nu van mening dat dit verbod uitgebreid zou moeten worden".

Municipal authorities in the 'car loving' Vicenza, Northern Italy, took up the plan to prohibit al the car traffic for four days, out of environmental considerations.

"In large parts of Europe, we notice every day how big of a necessity it is to use new, clean energy sources. After the prohibition of car-traffic in Vicenza, the European Union is of the opinion that it should be extended". (tweevandaag.nl, broadcast feb. 16th)

Christof

PAYD? PAYP? An idea for the Dutch situation

Today, I thought up a way of implementing PAYD in the Netherlands. It may seem outlandish, and it straddles somewhat from the exact research goal I set myself, but maybe it is worthwhile mentioning it here anyway. I call it: PAYP, Pay As You Pollute. That may sound negative at first, but I stumbled on this thought while thinking: consumer freedom of choice should be enhanced, not limited, by mobility pricing measures, mobility prices should better reflect costs, preferably also those costs which are at present for the greater part external (ie, not reflected in the prices paid for the products purchased).

I closed my eyes and dreamt up a situation in which the effects of mobility were truly reflected in transport cost. I imagined a consumer, buying a vehicle, buying insurance for it, and being asked: how much kilometres do you expect to drive this year? The answer could be: 20.000. Then he or she would be told: ok, the car you just bought runs 10 kilometres on 1 litre of fuel on average, so here's your insurance card, which doubles as a gasoline station credit pass, with credits for 2000 litres of fuel on it. Go drive your car, pay for fuel with your petrol card, and at the end of the insurance year, we will provide clearance with you based on the amount of fuel you will have purchased. If you will have purchased more, we will bill you, if you will have purchased less, we will credit you. Have fun with your new car!

The car-owner would like to drive economically in terms of fuel efficiency, and would like to avoid being in traffic jams or in city traffic too much, because those are situations in which fuel economy would be below the average on the basis of which his or her fuel credit balance is calculated. He or she would be able to buy additional fuel credits with the insurer, or simply pay with other means at the pump, when perceived as a better option than to forgo mobility.

The incentive for him or her to stick with consuming fuel credits provided by the insurer would be that the government levies less tax on fuel purchased in that way. Consumer choice would remain free, only the purchase of fuel outside the insurer's credit scheme would be more costly.

Then I opened my eyes. I realised that all of the technical infrastructure needed to enable such a system is already in place: The Netherlands have a sophisticated debit card system that is already being used for executing schemes as outlined above, only that use is made by car-leasing companies that, at present, offer all-you-can-eat service with respect to mileage and/or fuel consumption.

Also, the integration of offering a car lease contract and providing insurance for the leased vehicle is a very common phenomenon in the Netherlands.

So, if car-lease contracts could be fashioned this way, a special version of PAYD, namely PAYP, could be realised with very low additional implementation costs. Lease/Insurance packages of this type could be offered to companies, thus offering them an opportunity to reduce their transport costs in the most cost-effective way possible, that is by giving their employees an incentive to rationalise their mobility behaviour.

Well, pie in the sky this may be. But I think it would be worthwhile to try the idea, not least because the costs of trying it are small.

Christof

Divvying up research questions

I made a first attempt at subdividing my questions. I am not yet happy with the result, but I'll keep working on it. Here's what I've come up with so far:

1. How much environmental value can be created by the introduction of PAYD vehicle insurance in the Netherlands?

This question can be divided in the following subquestions:

- What is the role of (car-)traffic emission, congestion and damage-repair and wreck-dumping of total-loss damaged cars in current total pollution in the Netherlands?
- What is the expected reduction in emission of polluting agents when PAYD is introduced?


2. How much corporate value (ie, business opportunities for individual vehicle insurance companies) can be created by the introduction of PAYD by individual dutch insurance companies?

This question can be divided in the following subquestions:

- What reasons can individual insurance companies have to introduce PAYD?
- What implementation barriers exist?

3. Should the government help create circumstances in which insurance companies and/or consumers stand to gain more from PAYD insurance?

This question can be divided into the following subquestions:

- To what extent can PAYD help achieve public policy goals?
- Is facilitating PAYD a cost-effective governmental policy option?

In my next contribution I will try to further detail these questions. Right now I am in doubt as to keep or ditch the third question.

Christof

More on consumers

In my research, the question on there being a business case for individual insurance companies to have PAYD on offer can also partly be answered by focusing on consumers. Fortunately, Polis Direct insurance company has made an extensive market research report on consumer interest in PAYD available on another website: kilometerpolis.nl.

Now I am primarily busy divvying up my research questions into subquestions, and thinking up ways to answer those one by one. I will make the results available as soon as I can.

Christof

It's the consumer, stupid!

A friend pointed out to me that an important category of stakeholders is conspicuously absent in my writing so far: the consumer, or the citizen.
The very reason mobility is viewed as a good thing is that it enables people to do things they want to do: go to work, visit family and friends, recreate, etcetera. That is one role.

The other role is that it is the same citizen, traveller, consumer, that is faced with problems associated with mobility. For instance, on February 3rd the dutch tv show Zembla showed Austrian research that said Dutch people live one to three years shorter because of small dust particle pollution hanging in the air over Holland (see image).




It also featured an interview with Guido van Woerkom, chief of the ANWB, a dutch motorists association, who essentially said that if mobility is what we want, we should be ready to pay a price, as small as possible of course, in reduced life expectancy. That is material I can use in the introductory section of my research report.

Mobility and its effects begin and end with the individual consumer. It seems like a good idea to give that consumer individual choices and incentives in making his or her mobility decisions.

Christof

What have we got so far?

First, I might have to explain where I got the idea of researching PAYD. I stumbled on the subject when talking to people I thought could be helpful in giving me ideas about a subject. I guess there’s nothing more to say about that, because the choice is highly individual.
Well, the idea is there, and I have been fortunate enough to establish contacts with some people involved in the field. I hope to be able to interview some of them in the near future. At this point I have to make some decisions on how to go about:

- What kind of research do I want to do? Is it prospective, retrospective, experimental, descriptional? Qualitative? Quantitative?
- What theory is available on the subject? I will need theory on how to assess environmental impacts of market based measures targeting mobility. Moreover, I will need data and sources to substantiate every link in the reasoning chain set up in the previous blog contribution. For the question on the creation of corporate value, I will need theory on company decision-making.
- How to go about collecting data? Literature sources? Data from Past, ongoing and future tests? What about the validity, reliability of the data?
- Is there a plan B? What if the data I look for are classified, unreliable or otherwise less than ideal? Is there a way to assess the effect that the imperfection of the data can have on the validity of my conclusions?

First, I want to broaden my sources. I have planned a visit to the University Library next Wednesday, in order to look for books on, most of all, company decision making, preferably seen from an environmental point of view (a field I know relatively little about). And before doing that, I will comb out a dozen links about PAYD collected in my Favorites, and ‘snowball them’ (ie, to go through the lists of links and sources provided on the sites of my favorite URLs, to check them, to go through their list of references, and so forth).

I have also created two spreadsheets: one with all available data on all relevant contacts, and another with all the data on all the sources that could be needed to make an adequate source reference in my report (name of writer, title, publisher, place of publishing, year, that sort of thing).

Both are quite empty now, but I hope they will be filled as the work on my project progresses.

I have even made a few business cards to dole out to potential contacts. Frivolous? Well, it can’t hurt.

The first goal is to put together a research proposal (onderzoeksvoorstel) that is good enough to be approved by the exam committee, that is: a credible proposal that contains the following things: a provisional problem-description, a theoretical framework, a provisional set-up of the report, a ‘plan of attack’ and a time schedule.

The theoretical framework is bothering me most. I really hope I can make progress on that at least when I get to searching books in the library…

Christof

... and off we go!

Today, february 8th, marks the start of my university graduation research project. I am a student of environmental sciences at the University of Utrecht, the Netherlands.

I plan to investigate the value-creating potential of Pay as you drive (PAYD) vehicle insurance, which is a car insurance that is paid for in cents per mile or kilometre instead of a fixed sum per year, as is currently the case with most insurance companies.

The idea is that people who pay their insurance fees per distance traveled will more consciously weigh their options when planning a trip, leaving their car unused more, thus contributing to reductions in traffic congestion, and reductions in transport emissions and in damage-claims, both collision- and liability claims.

Congestion, traffic (exhaust) emissions and damage repair all have considerable environmental impacts, so arguably PAYD insurance is potentially a good way to use a so-called market based instrument to reduce those impacts. But in order to have it available to individual motorists, individual insurance companies will have to have it on offer. Insurance companies will be reluctant to provide the possibility to insure 'the PAYD way' when a good business case is lacking.

My research will be focused on two main questions:
1. How much environmental value can be created by the introduction of PAYD vehicle insurance in the Netherlands?
2. How much corporate value (ie, business opportunities for individual vehicle insurance companies) can be created by the introduction of PAYD by individual dutch insurance companies?

Depending on the outcome of those two questions, the remainder of the research report will be about:
- suggestions to make PAYD a more attractive product for an insurer to offer (currently, hardly any company has it on offer)
- suggestions to make PAYD more value-creating in terms of reducing environmental impacts (under those involved with the product, there seems to be a sort of consensus about two things: 1. PAYD on its own will not revolutionize traffic-related environmental impacts; 2. It might be a sensible complement to other market-based transport regulation measures, such as a 'kilometerheffing', a general tax on mobility levied by the government, an idea that is aired regularly in dutch politics but has failed to pass through parliament as legislation on several occasions).
More generally, the idea is this: a comprehensive variabilisation of the costs of individual mobility, which is a part of Dutch governmental transport policy, is an indispensable part of the transition to a more sustainable mobility.

If the two questions should yield two clear 'zero' answers, the remainder of the research report should focus on suggestions to reduce transport-related emissions in other ways.

I hope to keep myself focused, motivated and productive by the contributions to this blog, at the same time practising my English writing skills.

So, dear reader, whoever you are, do not hesitate to reprimand me (in comments, for example) in case I fail to make good progress at any time!

Christof