High speed is an important determinant of accidents for speeders as well as for other motorists. This paper develops a framework for analyzing instruments that encourage drivers to internalize the full consequences of their behavior with respect to choice of speed using Pay-As-You-Speed (PAYS) insurance, possibly as an extension of Pay-As-You-Drive (PAYD) insurance. We demonstrate how the combination of a Pigovian taxation scheme and PAYS can be designed in a setting involving two principals (the state and an insurance company) that affect the incentives of commuters to choose between driving and other modes of transport and for those that use the car mode to drive carefully. While the government is assumed to maximize overall social efficiency and therefore wants to implement marginal cost pricing, insurance companies do actuarial pricing, i.e. average cost pricing within risk classes that are homogeneous to the degree that the insurers have information about actual behavior. PAYS insurance improves the insurance industry's possibility to differentiate premiums according to behavior and therefore to target risk classes in a better way than today. Moreover, since our framework is designed to accomplish differentiation by self-selection, compulsory regulation is not necessary, although there may be reason for the government to facilitate the implementation of the new technology.

Contract relevant information asymmetries are known to cause inefficiencies in markets. The information asymmetry is largest in the beginning of the customer insurer relationship but reduces over time; the longer a policyholder stays with the insurer the more the insurer learns about the policyholder’s risk. Two important characteristics of the market studied here imply that the information asymmetry may not be reduced for all policyholders. First, insurers do not have access to traffic violations, which are predictors of risk since policyholders with traffic violations are more likely to report a claim. Second, the insurers do not share information, such as previous claims, which means that the policyholder can flee a poor claim record by switching insurer. Hence, there may be a selection of high risk customers who switch insurer more often, such that the information asymmetry in this group is never reduced. To test this, we compare information asymmetries in two groups of policyholders; new customers who stay with the insurer for a period or less (short term), and long-term customers who stay with the insurer for several periods (loyal). The results indicate that departing policyholders are disproportionately high risks that constitute an adverse selection of risks, while loyal policyholders constitute a propitious (favorable) selection of risks.

Older version: http://swopec.hhs.se/vtiwps/abs/vtiwps2011_002.htm

Risky driving behavior is regarded as being one of the best predictors of traffic accidents. Traffic violations certainly signal risky driving behavior, but the analysis of the linkage of traffic violations, individual and vehicle characteristics and annual mileage has so far been hampered by the difficulty of gaining access to appropriate disaggregate data. The contribution of this paper is that it sets out and explores a rich data set in order to study traffic violations, including both accident involved and accident-free individuals. The data set comprises all insurance policies from Sweden’s largest automobile insurance company covering several years, in total 9.3 million observations, as well as information on fines and convictions for traffic violations. This implies that the methodological issues associated with self-reported violations and only accident involved individuals are disused. The first purpose is to establish the role of age and gender in traffic violations. The second purpose is to make a first attempt to establish whether vehicle owners of status brands are more likely to commit traffic violations. The results support previous findings as well as confirm the association between owners of status brands and traffic violations. The main conclusion is that insurance data provides a viable option when studying behavior, but it also raises new methodological issues.

Older version: http://swopec.hhs.se/vtiwps/abs/vtiwps2011_003.htm

We empirically investigate the effect of policyholders’ private information of risky traffic behavior on automobile insurance coverage and ex post risk. We combine our insurance company’s information with private information data that is not accessible to the insurance company. We show that being unable to reject the null of zero correlation is not necessarily consistent with symmetric information in the automobile insurance market. Our results are twofold: In contrast to much of the previous work we find a positive significant correlation for three groups of policyholders, consistent with the adverse selection prediction. We furthermore find that private information about risky traffic behavior increases ex post risk while it both increases and decreases the demand for extensive insurance. This supports our hypothesis that adverse and propitious is present simultaneously in this market.

By combining Contract Theory and vehicle positioning techniques, insurance companies can replace some of the proxies for risk by actual traffic behavior when pricing the premium. A mechanism design model is used to illustrate that Usage Based Insurance (UBI) can separate risks in terms of driving behavior. This makes it possible to reward safe driving habits since the pricing scheme better reflects the accident risk. The conclusion is that UBI provides an actuarially fair premium for the insuree. It is further an efficient instrument to separate risks for the insurer since it reduces the information asymmetries highlighted in this paper.

Inom EU inträffar varje år 1 300 000 trafikolyckor med kroppsskador som följd. Olyckskostnaderna beräknas uppgå till 160 miljarder euro vilket motsvarar 2 procent av EU-ländernas BNP. Enbart i den svenska trafiken omkommer varje år 400–500 personer och flera tusen skadas. Trafikskador är, som för många andra länder, ett av Sveriges folkhälsoproblem. Samhället avsätter också stora resurser för att minska antalet olyckor och för att begränsa skadorna när olyckan väl är framme. En del av denna resursåtgång avser försäkringsbolagen och deras ansvarstagande för (vissa av) konsekvenserna av trafikolyckor. En central fråga för denna rapport är försäkrings- premiernas utformning och möjligheten att utforma premierna på ett sätt som reducerar olycksrisken. Vi menar att man nu med stöd av modern teknik har möjlighet att koppla premiens storlek till faktiskt beteende. Den snabba utvecklingen av fordonselektronik ger förutsättningar för att observera förar- beteende i form av hastighet i förhållande till gällande hastighetsbestämmelser. Detta öppnar i sin tur dörren för att förmå förarna att sänka hastigheten och därmed minska olycksrisken på ett helt annat sätt än vad som tidigare varit möjligt. I rapporten redovisas en översikt över internationella erfarenheter på området liksom en genomgång av de tekniska förutsättningarna för att använda sådana svarta lådor. Dessutom utvecklas ett förslat till struktur på det tänkta premiesystemet. Där framgår också att försäkringsbolagen kanske inte har tillräckligt intresse av att själva ta till vara detta tekniska paradigmskifte. I stället måste staten spela en aktiv roll, dels för att säkerställa en öppen standard för den tekniska plattformen som ska finnas i bilarna, dels för att lägga grundvalen för de nya spelregler som kommer att behövas. Vi argumenterar för att detta torde vara möjligt att åstadkomma till förhållandevis låga samhällskostnader.