Pass-through of external shocks along the pricing chain: A panel estimation approach for the euro area

Campa and Goldberg (2006) use a different approach to assess exchange rate pass-through for
industrialised countries. They use data on inter alia imported inputs from input-output tables
and distribution margins to calibrate the exchange rate pass-through to consumer prices. They
find an exchange rate pass-through of 0.13 to 0.30 for their sample of countries, depending on
the assumption regarding the sensitivity of distribution margins to exchange rate variations
used in the calibration. The weighted average for the euro area amounts to 0.16 to 0.25; this is
somewhat higher than what we find but, again, we only estimate the impact on the HICP
excluding unprocessed food and energy.

Taking up the idea of cross-checking results using input-output tables, Table 5 shows for the
three PPI components and for total consumer prices the share of imported inputs divided by
the total output for each sector (first column).¹⁰ This number is taken from Eurostat input-
output tables of 2000, aggregating the tables for Germany, France, Italy, the Netherlands,
Austria, Finland and Belgium for the euro area. The second column shows the impact of a 1%
exchange rate appreciation after 4 years, according to our estimation results. The share of
imported inputs in total output decreases along the production chain, i.e. it is highest (34%)
for energy production and lowest for final consumption. This is in line with our estimated
impact of the exchange rate pass-through, which is highest for energy producer prices and
diminishes along the production chain, with very similar coefficients to the shares in the
input-output tables. This might reflect the increasing role of distribution margins, as also
suggested in Campa and Goldberg (2006). Note that the total imported inputs do not take into
account the indirect effect via imported inputs in other sectors, which would increase these
numbers somewhat. Therefore, they cannot be interpreted as an upper range, implying a
100% pass-through.

Table 5 Imported input shares versus estimation results of a 1% NEER appreciation

Source: Eurostat and own calculations.

¹⁰ We cannot replicate the results of Campa and Goldberg (2006) because there are no data for mark-
ups and imported inputs for the breakdown of components we have used in our study.

ECBMH

Working Paper Series No 1104

November 2009∣   17

<<   15   16   17   18   19   20   21   >>

More intriguing information