Oil Prices and Personal Consumption Expenditures: Does the Source of the Shock Matter?
| Date | 01 April 2019 |
| Author | Zeina N. Alsalman,Mohamad B. Karaki |
| DOI | http://doi.org/10.1111/obes.12276 |
| Published date | 01 April 2019 |
250
©2018 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd.
OXFORD BULLETIN OF ECONOMICSAND STATISTICS, 81, 2 (2019) 0305–9049
doi: 10.1111/obes.12276
Oil Prices and Personal Consumption Expenditures:
Does the Source of the Shock Matter?*
Zeina N. Alsalman† and Mohamad B. Karaki‡
†Department of Economics, 332D Elliott Hall, School of Business Administration, Oakland
University, Rochester, MI 48309, USA (e-mail: alsalman@oakland.edu)
‡Department of Economics, Adnan Kassar School of Business, Lebanese American
University, Beirut, Lebanon (e-mail: mkaraki@lau.edu.lb)
Abstract
This paper studies the effect of structural oil shocks on personal consumption expenditures
(PCE). First, we estimate a nonlinear simultaneous equation model, compute impulse re-
sponses by Monte Carlo integration, and conduct a test of the symmetry of the impulse
response functions. We find that aggregate PCE responds asymmetrically to positive and
negative oil-specific demand shocks. Second, we find that aggregate PCE responds nega-
tively to positive oil demand shocks, while adverse oil supply shocks are of limited effect.
Third, we find important heterogeneity in the magnitude, sign and timing of the disaggre-
gate PCE responses to structural shocks in the crude oil market. Our results clearly indicate
that the response of PCE to an unexpected oil price increase depends on the source of the
oil price shock. Our findings are robust to different nonlinear transformations for the real
price of oil.
I. Introduction
Following the stagflations that occurred in the 1970s, most economists viewed oil price
shocks as cost shocks. Yet recently, most policymakers and researchers agree that oil price
shocks primarily operate through demand channels (see Kilian, 2008b, 2014; Herrera,
2018). Work by Edelstein and Kilian (2007, 2009), Baumeister and Kilian (2017), and
Baumeister, Kilian and Zhou (2017) shows that oil prices primarily affect the economy
through changes in spending by domestic households and firms due to changes in dis-
cretionary income. For instance, Baumeister and Kilian (2017) underscore that the re-
duction in oil prices since mid-2014 has increased households’ purchasing power and
consumer spending. How responsive are consumption expenditures to oil price shocks?
Does the composition of consumption spending change following a change in oil prices?
Does the source behind the oil price change matter? Does consumer spending respond
JEL Classification numbers: E21, Q43.
*Weare thankful to Ana Maria Herrera, Lutz Kilian and three other referees for helpful comments and suggestions.
The online appendix is available at https://sites.google.com/site/mohamadbkaraki/Appendix AK.pdf
Oil prices and PCE 251
asymmetrically to positive and negative structural shocks in the crude oil market? Given
that oil prices primarily operate through changes in consumer spending, and because US
consumption expenditure constitutes 70% of US GDP, answers to these questions are
important especially given the large recent fluctuations in oil prices.
This study investigates the effect of oil demand and oil supply shocks on real personal
consumption expenditures (PCE). Previous studies such as Edelstein and Kilian (2009),
Wang (2013), Baumeister and Kilian (2017) and Baumeister et al. (2017) studied the
response of consumption expenditure to an energy price increase that comprises a combi-
nation of demand and supply elements. This paper separates the effect of structural supply
shocks and demand shocks in the crude oil market on PCE, building on Kilian (2009a) and
Kilian and Park (2009).
In theory, many channels can explain the transmission mechanism of oil price shocks
on consumer expenditure. First, an unexpected increase in the real price of oil, all else
equal, reduces discretionary income as gasoline prices rise, which lowers consumption
expenditure.1Second, an unexpected increase in the real price of oil may be associated
with an increase in oil price uncertainty (see, e.g., Bernanke, 1983, Baumeister and Kil-
ian, 2017). Consequently, consumers may delay their spending on irreversible purchases
such as durable goods. The uncertainty channel associated with oil price shocks can also
affect consumption expenditure on non-durable goods, as consumers start to save more
as a precaution for future economic turmoil (see Edelstein and Kilian, 2009). Third, the
operating-cost channel entails that an unexpected increase in the real price of oil will pull
consumers awayfrom durable goods that rely on energy such as motor vehicles (see Hamil-
ton, 1988). Finally, the change in relative prices associated with an unexpected change in
the real price of oil may trigger sectoral reallocations (see Davis and Haltiwanger, 2001;
Herrera and Karaki, 2015). In the presence of frictions in the economy that prevent this
reallocation, sector-specific capital and labour may be unemployed, resulting in further
losses in real income and hence consumption (see Hamilton, 1988; Baumeister and Kil-
ian, 2017). Note that the uncertainty effect, the precautionary savings effect, and frictions
preventing the reallocation of capital and labour in response to changes in the real price
of oil, all may cause asymmetric responses of consumer spending to positive and negative
oil price innovations (see Kilian, 2014).
In this paper, we use a nonlinear simultaneous equation model to study the effect of oil
price innovations on PCE, building on Kilian and Park (2009) and Kilian and Vigfusson
(2011a, 2017). Our nonlinear modelling strategy allows us to nest both symmetric and
asymmetric responses to oil price innovations. We decompose global real oil price inno-
vations into oil supply shocks, aggregate demand shocks, and oil-specific demand shocks.
Our benchmark model specification is based on the real oil price increase measure (see
Mork, 1989). This nonlinear transformation in the real price of oil, which censors for the
monthly decreases in the oil price change, is closely linked to theoretical models that imply
asymmetry in the response of economic activity to oil price increases and decreases. We
also consider an alternative model specification based on the 1-year net oil price increase
measure (see Hamilton, 1996). This nonlinear transformation in the real price of oil is
1see Edelstein and Kilian (2009), Kilian (2014), Baumeister and Kilian (2016), Hamilton (2009, 2013), Baumeister
et al. (2017).
©2018 The Department of Economics, University of Oxford and JohnWiley & Sons Ltd
To continue reading
Request your trial