As announced on 23 August 2011, the Bank has now completed the usual annual update of the sterling ERI weights, and the new set of weights took effect from 26 August 2011.
The sterling effective exchange rate index is a measure of the overall change in the trade-weighted exchange value of sterling, calculated by weighting together bilateral exchange rates. It is designed to measure changes in the price competitiveness of traded goods and services and so the weights reflect trade flows in manufactured goods and services. Using the USA as an example, the weights for the US dollar in the sterling ERI are based on:
- Competition in the UK domestic market from imports from the USA
- Competition between UK exports and US products in the USA
- Competition between UK and US exports in third country markets
The calculation is explained more fully in the article "The new sterling ERI" in the
Winter 2004 Quarterly Bulletin, page 429.
The table
attached shows the current set of weights which came into effect on 26 August 2011. The previous set of weights is also
attached. Countries are included in the calculations if their share of either UK imports or exports on average over the latest three-year period, exceeds 1%. ERI weights for each selected country are based on the latest available full set of world trade data, currently 2009. The January 2005 average index value is set equal to 100.
The broad version of the sterling effective exchange rate index uses exactly the same methodology but has an expanded country set. Countries are included in the broad index if their share of UK imports or exports on average over the latest three-year period, exceeds 0.5%.
To reflect changing trade patterns, the weights and country set are allowed to change over time to give an annually chain-linked index. These weights are updated each year for newly available world trade data, so that weights based on 2009 trade data are now used to calculate the chain-linked ERI from the beginning of 2010 onwards. We have also incorporated revisions to trade data for the years preceding 2009. Adjustments to the weights will likely result in small changes to past values of the indices.
In line with last year’s update, adjustments have been made to the trade data to strip out the effects of missing-trader intra-community (MTIC) fraud. This fraudulent activity involves repeated importing and exporting of the same products, which boosts measured UK trade flows with the other countries involved. However, such activity has little relevance for underlying or ‘economic’ trade and so we use ONS estimates of MTIC fraud going back to 1999 to strip out these flows when calculating the weights. The bulk of MTIC fraud is estimated to have taken place with EU countries, in particular between 2001 and 2006, but a specific country breakdown is unavailable so it has been allocated proportionally across all EU countries according to their trade weight.
The table below illustrates changes in the ERI weights for the most important countries and currency area. The most significant change to the existing weights is for the euro area which has fallen by 1.7pp. The index remains referenced to January 2005=100.
| |
Old Weight (%) |
New Weight (%) |
Revision in 2008 (pp) |
Change over 2008-2009 (pp) |
| Country |
2008 |
2008 |
2009 |
|
|
|
| Euro area |
51.0 |
51.0 |
49.3 |
0.0 |
-1.7 |
| USA |
16.6 |
16.6 |
17.1 |
0.0 |
0.6 |
| China |
7.0 |
7.0 |
7.5 |
0.0 |
0.5 |
| Japan |
4.0 |
4.0 |
3.6 |
0.0 |
-0.4 |
Columns may not sum due to rounding
This method follows the latest IMF methodology, referenced to 2005 and reflecting trade flows in manufactured goods for the period 1999 to 2001. It replaced the previous IMF-based index in April 2005, which used IMF trade flows between 1989 and 1991, with 1990 set to equal 100.
The indices up to mid-August 2011 include Bank Holidays using foreign exchange rates from markets that did not have holidays on these days; from then on there are no observations for UK bank holidays as only rates observed by the Bank's Foreign Exchange Desk are used.