pass-through on consumer prices takes longer than for the PPI, with most of the effect coming
through after 3 years. The weighted average of the impact multipliers of CPFDPR, CPNEIG
and CPSERV (i.e. CPEX) suggests an impact of around -0.09% after 16 quarters.
Simulations with macro-models (NiGEM and the Oxford Economic Forecast) yield an impact
of 0.2 on consumer prices from a 1% depreciation of the exchange rate, similar to Campa and
Goldberg (2006a), Bailliu and Fujii (2004), Gagnon and Ihrig (2004), Choudhri et al. (2002)
and Choudhri and Hakura (2002) who all find a medium- to long-term exchange rate pass-
through to consumer prices of 0.2 for euro area or industrialised countries9. Meanwhile, the
results of Hüfner and Schroder (2002) and Campa and Gonzalez Minguez (2006) are
substantially lower, with a pass-through of 0.04 and 0.05, respectively. As we estimate the
impact on the HICP excluding unprocessed food and energy, our results should be somewhat
smaller than what was found in the literature, because the impact of exchange rates via euro-
denominated oil prices on the HICP energy is excluded. Indeed, Faruqee (2006) who also
estimates the effect of exchange rates on the HICP excluding unprocessed food and energy,
finds a pass-through of 0.02 after 18 months, which is somewhat lower than our results.
In addition, it has been argued in the literature that the pass-through of exchange rate changes
to consumer prices has become somewhat lower around the 1990s, as central banks
increasingly focussed on stabilising prices. This could also explain our somewhat lower
estimates as the HICP series start only in the 1990s (which is also true for Faruqee (2006)).
For example, Gagnon and Ihrig (2004) find that the long-run pass-through of 20 industrialised
countries was on average around 0.16 from 1971 to the mid-80s for an exchange rate
depreciation, while it was reduced to 0.05 from the mid-80s to 2003, possibly related to an
increased focus on price stability in many central banks. This is also confirmed by Choudri
and Hakura (2002) who find a lower exchange rate pass-through for countries with low
inflationary environment. In addition, increased competition on foreign markets could also
have led to a stronger pricing to the market, reducing thereby the exchange rate pass-through
(see also Bussière and Peltonen (2008)). However, as argued by Campa and Goldberg
(2006b), the exchange rate pass-through could also have increased due to an expansion of
imported inputs used in the production and due to a change in sectoral expenditures on
distribution services.
9 Note that these are the results of Gagnon and Ihrig (2004) for the full sample (1971 to 2003) while
they find a lower pass-through when they use the part of the sample where prices were relatively stable.
For Choudri and Hakura (2002), the estimates relate to low inflation countries.
И ECB
Working Paper Series No 1104
November 2009