We estimate the income and fuel price elasticities of household vehicle kilometres travelled (VKT) and car ownership. To model the VKT, we apply a discrete-continuous model on registrAy micro panel data, covering all Swedish households from 1999 to 2018. We model two joint choices: car ownership and VKT conditional on car ownership, where the elasticity of VKT from these two choices are obtained by using the Two-Part model. We account for unobserved household effects using a correlated random effects specification and take household adjustments into account by including lagged values of fuel prices and income. Our preferred model yields a long-run income elasticity of 0.45 for private VKT, where close to two-thirds of the effect comes from the income elasticity of car ownership. The long-run fuel price elasticity of private VKT is −1.03, with the response in VKT among car owners being larger than the response in car ownership.  

Underhåll av det svenska järnvägsnätet beskrivs ofta som eftersatt och i behov av ökad finansiering. Ett komplement till ökad finansiering är att minska kostnaderna för underhåll genom effektiviseringar. I denna studie redovisas resultaten av ekonometriska analyser av ett datamaterial som omfattar uppgifter om järnvägsnätet och samhällets kostnader för järnvägsunderhåll under åren 2011–2023. Analysen pekar, för det första, på att det finns en effektiviseringspotential i att förändra kontraktens sammansättning. Genom att flytta bandelar mellan kontrakt och i förlängningen slå ihop eller dela upp kontrakt kan kostnaderna minska. En andra analys avser förekomsten av kostnadsskillnader mellan utförandeentreprenader och totalentreprenader. Eftersom avtalen har olika incitamentsstruktur och riskfördelning kan utrymmet för innovationer och därmed slutkostnader skilja sig åt. Resultaten pekar på att det kan förekomma sådana skillnader men en viss osäkerhet i resultaten gör det svårt att dra säkra slutsatser. En tredje fråga som analyseras avser nivån på den tonnagereglering som kompenserar entreprenörer för trafikökningar under avtalstiden. En jämförelse mellan underlaget för nu gällande tonnagereglering och de indikationer som ges av den nya analysen talar för att ersättningarna idag överkompenserar entreprenörerna. Sammantaget bidrar studien med empirisk kunskap som kan stödja Trafikverkets tillgångsförvaltning mot ett effektivare underhåll.

It has proven difficult to reduce carbon emissions from the transport sector; in fact, emissions from this sector are still increasing worldwide. Reducing emissions by reducing road transport is challenging; therefore, a transition to a vehicle fleet with low or zero emissions seems essential. Many new cars in OECD countries are sold to firms as fringe benefit cars (sometimes called company cars in the literature). The generous taxation of such cars has been shown to have negative welfare effects because it increases the consumption of cars. However, it is sometimes justified since it speeds up the transition of the car fleet to lower-emission vehicles. The purpose of this paper is to analyze how fringe benefit cars impact carbon emissions, fuel type, weight, size, engine power, and market value of new cars. We apply micro register data including all adult Swedes and their cars, spanning the years 1999 to 2020. By using a matching model that combines Exact matching and Mahalanobis distance matching, the fuel consumption of the fringe benefit car is compared to the hypothetical new private car that the employee receiving the fringe benefit would have otherwise purchased. We find that new fringe benefit cars tend to be larger, heavier, and more powerful than the hypothetical new private cars that fringe benefit car recipients would have otherwise purchased, However, we also find that new fringe benefit cars sold in 2019–2020 consumed 1.2 L less fuel per 100 km compared to hypothetical new private cars, a decrease of 20 percent. The lower fuel consumption of the fringe benefit cars in these years results from a higher share of electric vehicles among them. We also find that the likelihood of the fringe benefit car being an alternative-fuelled vehicle is 6 percentage points higher than if it was bought as a private car.

I rapporten redovisas beräkningar av hur kostnader för vidmakthållande åtgärder – drift, underhåll och reinvesteringar – påverkas av den tågtrafik som bedrivs. Beräkningarna genomförs med en bokföringsmetod och en ekonometrisk metod och baseras på ett datamaterial som omfattar huvuddelen av det järnvägsnät som förvaltas av Trafikverket. Bokföringsmetodens rörliga genomsnittskostnad är betydligt högre än den ekonometriska metodens marginalkostnad. Anledningen är att den rörliga genomsnittskostnaden sjunker med ökad trafik, vilket per definition innebär en lägre marginalkostnad.

Resultaten från den ekonometriska metoden beror på vilken estimator som används; Fixed effects innebär en betydligt lägre marginalkostnad jämfört med Random Effects. För att stärka förståelsen av resultatskillnaden används en instrumentvariabel-estimator som kan hantera snedvridningar orsakade av endogenitetsproblem. Den pekar på att tågtrafikens marginalkostnad enligt Random Effects resultaten inte har en betydande snedvridning. Bedömningen är att sannolikheten är lägre att Random Effects överskattar marginalkostnaden med en viss summa jämfört med sannolikheten att Fixed Effects underskattar marginalkostnaden med samma summa. Likväl går det inte att utesluta att Random Effects-resultaten är snedvridna. Mer forskning behövs för att ytterligare klargöra dessa resultatskillnader.

I rapporten redovisas också analyser och bedömningar av möjligheten att använda sig av banavgifter för att styra användningen av tillgänglig kapacitet. Två kategorier av kapacitetskostnader kan ligga till grund för en sådan styrning: (i) knapphetskostnader som avser kostnader för att tågoperatörer inte får tåglägen som ligger nära de önskemål som lämnas i början av tidtabelläggningsprocessen, och (ii) kostnader för trängsel på spåren som innebär att primära tågförseningar fortplantar sig och också drabbar efterföljande tåg. Framställningen har identifierat metoder som skulle göra det möjligt att mäta dessa kapacitetskostnader. Genom att införa avgifter som är proportionella mot kapacitetsutnyttjandet skulle samhällsnyttan av järnvägens infrastruktur öka.

Trafikverkets kostnader för järnvägsunderhåll uppgick till 7,6 miljarder kronor underår 2023 och förväntas satsa 210 miljarder kronor den kommande tolvårsperioden. Trots dessa högabelopp beskrivs ofta underhållet som eftersatt och i behov av ökad finansiering. Ett komplement tillökad finansiering är att minska kostnaderna för underhåll genom effektiviseringar. I denna studieundersöks om det finns en effektiviseringspotential i att förändra kontraktområdenas storlek för attåtnjuta ytterligare stordriftsfördelar.

Company cars, juridically owned cars possessed and used privately by employees, are common in most OECD countries. The cost of possessing and using these cars is generally lower than if the car was bought privately, which leads to welfare losses from increased car possession and expenditure. However, the previous literature has taken the productivity of the employee receiving a company car as given. We believe that there are two ways company cars can impact productivity. First, the lower cost of company cars than privately owned cars also implies a lower commuting cost. This may enable the employee to accept an employer further away, that may be a more productive match. Second, the employee would pay more in income tax if they received the monetary value of the company car as an increased wage rather than a company car. This decrease in the marginal tax rate may induce the employee to take on a more demanding position where they are more productive. 

To estimate the effect of company car possession on productivity (taken as gross wage earnings) and tax revenues, we use fixed effects models applied to Swedish registry micro panel data from 2005 to 2020. The rich dataset contains socio-demographic variables for all adult Swedish individuals, where we can observe the residential location and the firm location for all workers. We only consider employed individuals, yielding 54 million observations. We take gross wage earnings to include cash wage earnings, social service contributions, and potential car costs if the employee is given a company car. Tax revenues include labour tax on wage earnings, the fringe benefit value from the company car, and social service contributions.  

Car use and possession is an increasingly debated topic in our society, as it is one of the major contributors to carbon dioxide emissions in the European Union. This thesis aims to empirically study this question using a large registry database on the entire adult Swedish population and the car fleet from 1999 to 2020.

In the first paper, I present estimates of the income and fuel price elasticities of vehicle kilometres travelled by private cars. The empirical model allows the elasticities to vary with income and municipality type. The results show that the absolute value of the fuel price elasticity and the income elasticity is largest in the middle of the income distribution and that the absolute value of the fuel price elasticity is larger outside large cities.

In the second paper, I allow for the impact of income and fuel price on car possession by using a discrete-continuous model, the Two-Part model. The main findings are that income primarily impacts car possession and has a relatively smaller impact on vehicle kilometres travelled conditional on car ownership. In contrast, the reverse is found for the fuel price. Furthermore, the results show that the estimated elasticities are larger than a model that does not explicitly account for car possession.

In the third paper, the impact of company car taxation on the choice of new cars is studied, by comparing new company cars and new private cars. The empirical model shows that new company cars consume less fuel per kilometre than new private cars, but are also larger, more expensive, and more powerful. Evaluating the impact of company cars using a welfare model yields a negative welfare effect, where the foregone tax revenues exceed potential welfare gains from lower carbon dioxide emissions and fuel costs.

In the fourth paper, the effect of company car taxation on gross wage earnings, including the price of the company car, is studied. The results from the empirical model show that there is a substantial increase in gross wage earnings when receiving a company car, but also an increase in commuting distance. This may be explained by the fact that the decrease in taxable income from the favourable company car taxation and lower commuting costs induces the employee to increase their labour supply, take on a more demanding position or choose a workplace further away where they have a better job match.

Bilinnehav och användning är en aktuell fråga då personbilar står för en stor andel av koldioxidutsläppen i EU. Därmed är det viktigt för beslutsfattare att veta hur det är möjligt att påverka dessa. Den här avhandlingen ämnar att empiriskt studera dessa frågor med hjälp av en stor registerdatabas som täcker hela den svenska befolkningen och bilflottan från 1999 till 2020.

I det första kapitlet presenterar jag skattade inkomst- och drivmedelspriselasticiteter för fordonskilometrar körda med en privatägd personbil. Den empiriska modellen tillåter att dessa elasticiteter varierar beroende på inkomstnivå och kommuntyp. Resultaten från modellen visar att det absoluta värdet av drivmedelpriselasticiteten och inkomstelasticiteten är större i mitten av inkomstfördelningen samt att det absoluta värdet av drivmedelspriselasticiteten är större utanför storstäder.

I det andra kapitlet tillåtes inkomst och drivmedelspriset påverka bilinnehav genom att använda en variant av en discrete-continuous modell som kallas The Two-Part model. Huvudresultaten från appliceringen av modellen är att inkomst primärt påverkar bilinnehav, medan drivmedelspriset påverkar främst körsträckor för de som redan äger en bil. Utöver detta är de skattade elasticiteterna större än i en modell där bilinnehavet inte är explicit modellerat.

I det tredje kapitlet studeras effekten av förmånsbilsbeskattning på valet av nya bilar genom att jämföra nya förmånsbilar med nya privatägda bilar. Den empiriska modellen finner att nya förmånsbilar är bränslesnålare än nya privatägda bilar, men att de också är dyrare och har en kraftfullare motor. Utöver den empiriska modellen appliceras också en teoretisk modell som beräknar den sociala välfärden av förmånsbilsbeskattningen. Modellen visar att de förgångna skatteintäkterna är större än värderingen av de minskade koldioxidutsläppen och bränslekostnaderna.

I det fjärde kapitlet studeras hur förmånsbilsbeskattningen påverkar bruttolöner, vilket inkluderar värdet av förmånsbilen. Resultaten från den empiriska modellen visar en betydlig ökning i bruttolönen när en anställd ges tillgång till en förmånsbil men också att pendlingsavståndet ökar. En möjlig förklaring är att minskningen av den skattbara inkomsten från förmånsbilsbeskattningen och en lägre pendlingskostnad får den anställde att ta en mer krävande position, öka sitt arbetsutbud, eller välja en arbetsplats där den har en bättre match för sin kompetens.

We estimate the income and fuel price elasticities of private car vehicle kilometres travelled (VKT) using fixed effects on registry micro panel data covering all Swedish households from 1999 to 2018. Such registry data, covering all individuals and cars in the country, are unique to Nordic countries and are comprehensive enough to allow fine segmentation of the population by both income groups and several municipality types. To address potential endogeneity arising if employees receive a wage compensation for long commutes, we apply the temporal changes in earned income tax credits as an instrumental variable. We find lower income and price elasticities (in absolute value) in the large cities, and larger elasticities in suburbs, other cities and in rural areas. We also find that the elasticities decrease with income, excluding the lowest income quartile, having the lowest elasticities. Specifically, we show theoretically and empirically that because the income elasticity varies considerably along the income distribution, the resulting income elasticity depends heavily on how the estimator assigns weight to different income groups, unless the specification explicitly allows for variation in the impact of income on VKT. Moreover, the impact of an income increase depends on to whom the income increase accrues to. For a uniform income increase, 0.2 is the preferred income elasticity. Our preferred long-run fuel price elasticity is −0.53. The short-run elasticities are lower. These elasticities apply to the full population and not only to car owners or drivers. 

Understanding how households’ car ownership and use respond to changes in income and fuel price is essential for travel demand forecasts and the design of car use policies. The literature estimating the effect of income and fuel prices on car use have almost exclusively studied the decision of owning a car and how much to drive as two separate choices. This may potentially result in the failure to capture the full response of the household or in model misspecification. In this paper, we estimate the income and fuel price elasticities of the joint decision of car ownership and car travel on household registry micro panel data using a discrete-continuous model. 

We use a correlated random effects approach to control for unobservable household heterogeneity. Our discrete-continuous model allows us to estimate the effect of income and fuel price on car ownership, car travel conditional on car ownership, and the total effect on vehicle kilometres travelled (VKT). 

We use a register database covering the Swedish population from the years 1999 to 2018, containing 96.7 million observations. The database includes socio-economic information on all registered adult Swedes, and information about all privately owned cars.

We find an income elasticity of VKT 0.4, half the elasticity comes from the impact of income on car ownership and the other half from the car travel conditional on car ownership. Hence, estimating the income elasticity of VKT on only car owners as earlier studies on micro panel data studies have done can underestimate the elasticity.

The model yields the fuel price elasticity of VKT -0.72 which is larger than has been found by meta-studies, but only slightly larger than the previous micro panel data studies and the previous discrete-continuous studies. We find that fuel prices primarily impact car travel and has a smaller impact on car ownership. 

Given the urgency and costliness of climate abatement policies, it is surprising that company cars, cars that are provided as fringe benefits to employees for private use, have received relatively little attention from policymakers and researchers. The taxation of company car is typically designed such that the employee pays a portion of the car's value as a taxable benefit, which is subject to income tax. However, in most OECD countries, company cars can be possessed and driven at a lower cost than the expenses of a private car. Previous studies have shown that company cars benefits, lead to increased size and power of the car, higher car ownership, and welfare losses. Furthermore, many countries have reduced cost for less-polluting company cars. This may induce larger welfare losses, unless they actually do speed up the renewal of car fleet to less-polluting cars efficiently.

In this paper we study to what extent company car taxation policies impact the choice of new car. to estimate how new company cars differs from new private cars in fuel type, weight, size, and engine power. Furthermore for 2019 and 2020 we also estimate the difference in carbon dioxide emissions and fuel costs between new company car and new private cars, which we use to calculate the welfare effects of company car benefits.

We use Swedish registry data covering the entire adult Swedish population from 1999 to 2020 and apply a matching model to control for selection into company car possession. The model combines exact matching and Mahalanobis distance matching, and matches upon socio-demographic variables, preference for car use, and previous company car possession.

We find that company car benefits lead to a larger share of alternative fuel but the also increases, the size, the weight, and the engine power of the car. The policy leads to welfare losses due to the increased car expenditures exceeds the gain from reducing carbon dioxide emissions and fuel costs.