Impact of economic and non‐economic factors on gasoline demand: a varying parameter model for Sweden and the UK

• Date: 01 December 2014
• Author: Amin Karimu
• DOI: http://doi.org/10.1111/opec.12035
• Published date: 01 December 2014

Impact of economic and non-economic

factors on gasoline demand: a varying

parameter model for Sweden and the UK

Amin Karimu

Postdoctoral Researcher, Department of Economics, Umeå School of Business and Economics, Umeå

University, Umeå, Sweden. Email: amin.karimu@econ.umu.se

Abstract

This paper investigated the impact of price, income and non-economic factors on gasoline demand

using a structural time series model. The results indicated that non-economic factors did have an

impact on gasoline demand and also one of the largest contributors to changes in gasoline demand in

both countries, especially after the 1990s. The results from the time-varying parameter model indi-

cated that both price and income elasticities were varying over time, but the variationswere insig-

nificant for both Sweden and the UK.The estimated gasoline trend also showed a similar pattern for

the two countries, increasing continuously up to 1990 and taking a downturnthereafter.

1. Introduction

In the past decades both the UK and Sweden haveexperienced an increasing share of trans-

port oil in their total energy mix, irrespective of the increasing campaign and polices tar-

geting the reduction of transport oil in favour of less polluting sources of fuel such as

biofuels, coupled with the emphasis in both countries on energy efficiency as a whole and

fuel efficiency in particular. Besides these similarities, there are remarkabledifferences in

transportation culture between the two countries, which makeit interesting to study, espe-

cially the importance of smart driving choices by consumers for the possible reduction in

road transport gasoline consumption. The choice of the two countries is because I wantto

assess whether there are any behaviouraldifferences in terms of driving culture between a

country in the European union with the largest supply of crude oil and one in the Nordic

country with a large share of CO2emissions from transport sector but without crude oil

production, and with high potentials of decarbonising the sectors with great strides

towards biofuels relative to the other Nordic countries. More importantly, how these dif-

ferences influence the response rate to both price and income changes over time.

JEL Classification: C22, Q41, Q4

445

© 2014 Organization of the Petroleum Exporting Countries. Published by John Wiley & Sons Ltd, 9600 Garsington

Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.

According to the International Energy Agency (IEA, 2010), oil use in the transport

sector in the UK increased from just 66 per cent in 2000 to around 70 per cent in 2009. In

Sweden, oil use in road transport was around 95 per cent of total fuel use in this sector in

2007 (IEA, 2008). Gasoline is still the dominant transport oil used in both Sweden and the

UK (for personal passenger transport), although in recent years its share on total transport

oil demand in these two countries has been gradually declining. The future forecast for

fossil fuel use in the transport sector is expected to increase due to increasing demand in

both countries but much of the increases are likely to be due to increases in the demand for

diesel rather than gasoline, partly due to policies favouring diesel and also due to the effi-

ciency of diesel engines compared with petrol engines.

It is therefore clear that for both countries to achieve significant reductions in their

respective CO2emissions from fossil fuels, more reductions have to come from gasoline,

both in terms of volumes and efficiency.As indicated earlier, gasoline demand is one of the

major contributing sources of aggregate energy demand in these two countries and hence

there is a need for policy to shift some of the demand from this source of fuel to less pol-

luting sources of fuel. For any policy to be effective in reducing demand for gasoline and

the associated CO2emissions there is the need to have knowledge about how consumers

respond to changes in prices, taxes and income.Thus, there is a need for accurate estimates

of both the price and income elasticities for gasoline demand. This is because for any

policy to have the desired impact on gasoline demand and CO2emissions, the responsive-

ness of demand to price changes and incomes changes or both, will be the main determin-

ing factor on the level of success of the policy (assuming there is a political will to

implement the policy).

The response on the supply side is also important for the success of any policy to

reduce gasoline consumption via, e.g. taxes, but the degree of importance depends on the

magnitude of demand elasticity. In this study, the supply side of the market is not consid-

ered: the emphasis is rather on modelling gasoline demand and deriving its elasticity with

respect to price and income in particular and to assess how non-economic factors such as

technology and preference (taste) affect gasoline demand. Given that I study separate

national markets that only account for a small fraction of worlddemand, it is reasonable to

assume that the price is exogenous and hence do not necessarily have to consider the

supply side.

In the literature, various econometric models havebeen implemented over the years in

estimating gasoline demand elasticities and some of the popular models include: static

models, dynamic models, vehicle stock models, vehicle characteristic models, lagged

endogenous models, inverted-v models, and recently cointegration and structural time

series models. These models have produced price and income elasticities for individual

countries and groups of countries using time series, cross-sectional and panel data sets. In

addition, there is a group of models using micro data within countries, thus focusing on

Amin Karimu446

OPEC Energy Review December 2014 © 2014 Organization of the Petroleum Exporting Countries