56
ECB
Monthly Bulletin
December 2014
In this context, a counterfactual exercise
based on such elasticities suggests
that oil prices have accounted for
approximately 0.6 percentage point of
the 0.9 percentage point decline in HICP
inflation excluding energy and food since
the end of 2011 (see Chart C). This mostly
reflects the unwinding of the upward impact
associated with the rise in oil prices up to that
point in time. As the impact in this exercise is
estimated on the basis of the HICP excluding
food and energy, it excludes the direct effects
on the energy component.
A more data-oriented way to gauge the
presence of indirect effects is to focus on
those items of the HICP excluding energy and
food that are more likely to be affected by oil
price developments. From the available set of
items, these are selected by means of simple
regressions of quarter-on-quarter changes in
the individual HICP item on an autoregressive
term and relevant lags of changes in oil prices. Chart D shows the evolution of the HICP
aggregate comprising the items for which significant lagged oil price impacts were found.
3
These
items have a weight of approximately 10% in the HICP excluding energy and food. However,
given their greater amplitude, they have contributed around 25% to the decline in HICP inflation
excluding energy and food over the past two years.
Overall, in addition to the fairly significant and immediately visible direct downward effects of
the recent oil price declines on the energy component of the HICP, it is reasonable to also expect
downward impacts on other HICP components via indirect effects. The precise magnitude and
timing of these effects is generally uncertain. Like direct effects, however, indirect effects on
the annual rate of change in prices should, in principle, be temporary, and related to the period
of adjustment to the change in oil prices. Therefore, they should not influence inflation on a
sustained basis. Nevertheless, it is important that such temporary developments do not feed
into longer-term inflation expectations and do not have a more lasting impact on wage and
price-setting behaviour via second-round effects.
3 From the non-energy industrial goods component, these selected items include (i) Non-durable household goods (056100), (ii)
Materials for the maintenance and repair of the dwelling (043100), (iii) Pets and related products (0934_5), (iv) Spare parts and
accessories for personal transport equipment (072100) and (v) Clothing materials (031100). From the services component, they
include mainly transport-related services, in particular (i) Other purchased transport services (073600), (ii) Passenger transport by
road (073200), (iii) Other services relating to the dwelling n.e.c. (044400), (iv) Passenger transport by sea and inland waterway
(073400), (v) Passenger transport by air (073300), and (vi) Package holidays (096000). These items were chosen on the basis of
their link with oil price movements. A relatively simple (and partial) autoregressive distributed lag model was estimated for
each of the 93 detailed HICP subcomponents i in each of the euro area countries c over the sample period from 2000 to 2014,
):,(
41,,,,
t
t
tci
tci
--
=
, where HICP denotes the quarter-on-quarter rate of change in the seasonally-adjusted HICP
component i in country c at time t, and OIL denotes the quarter-on-quarter rate of change in oil prices (in euro terms) at time t. Owing
to co-movements among commodity prices, it may be that this methodology also captures impacts from other commodity prices such
as food and industrial raw materials.
Chart D Evolution of oil prices, HICP
excluding food and energy and selected
HICP components
(annual percentage changes; monthly data)
-50
-25
0
25
50
75
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
2007 2008 2009 2010 2011 2012 2013 2014
HICP excluding energy and food
selected HICP items
1)
brent crude oil (EUR/barrel; right-hand scale)
Sources: Eurostat and ECB calculations.
1) These HICP components have a high sensitivity to oil price
movements. See footnote 3 for a more detailed explanation.