FX Survey 2012: Methodology
The Euromoney Foreign Exchange survey is the most comprehensive quantitative and qualitative annual study available on the FX markets. The FX market is an unregulated OTC market and there are no reliable, aggregated, global statistics made available against which to benchmark the survey outside the BIS studies. The survey also excludes a number of categories of market participant, which means that the total volume reported by the survey is not and not intended to be an accurate reflection of total global foreign exchange activity. Euromoney aims to capture client price-taking activity only. However, given the geographical and participant-type spread represented by the survey, Euromoney believes that the survey provides an accurate proxy for trends in the major areas of activity polled and accurately discerns the relative performance of the banks ranked, particularly over periods of two or more years.
The 2012 survey launches Thursday 26th January and closes Friday 2nd March - and represents the previous calendar years trading (2011). All voting is carried out electronically via a secure, dedicated data collection website. Respondents can amend votes prior to the closing deadline by re-entering the survey form via the link.
Euromoney splits out accounts and volumes by geography, institution type and total account size. They are defined as follows:
Geography: The location of where the voter is physically located NOT the domicile or head office of the institution that the voter represents
Institution type: Bank (non market-making volumes only accepted)/Corporate (non-financial)/Corporate related financial service (funding vehicle)/Electronic FX trading platform/Retail aggregator/Hedge fund/High frequency trading firm/Private equity firm/Insurance company/Private Investor (high-net worth individual)/Real money institutional investor
Total Account Size: The total amount traded by that institution over the survey period with all market making counterparties.
The survey falls into two major sections.
PART 1 – Quantitative
This section is concerned with quantitative measures of overall market activity and banks’ relative markets shares. The key rankings are:
Overall ranking by market share: Respondents (users of FX services) are asked to give us the volumes transacted with their providers of FX services (in practice this will mean their total FX volume for the year for up to 20 counterparties). Respondents are asked to separate out the volumes transacted with each provider into:
Swaps (single leg)
Options (notional value)
Each provider’s volume aggregated across all respondents (for example, the total volume of transactions attributed to XYZ Bank across all responses) is then expressed as a percentage of the total transaction volume represented by all valid responses. Respondents are also asked to give a % of their total FX volume that is executed electronically with each bank. The e-trading market share is then calculated by totaling the % volumes given for each bank across all the responses.
There are the following restrictions/definitions:
1. Responses which do not include an annual total volume number for the respondent organization cannot count towards any of the rankings in the survey, including qualitative rankings, and are discarded.
2. Responses that do not satisfy the following minimum FX turnover threshold for qualification cannot count towards any of the rankings in the survey, including qualitative rankings, and are discarded:
$20mn in turnover for respondents from emerging market economies
$50mn in turnover for respondents from G10 economies
3. Captive votes and votes representing internal transaction flows are excluded in the survey. This includes votes cast by:
Private banks that are part of banking groups with a market-making FX business
Internal bank treasuries
Internal FX and non-FX trading desks
4. Volume must be split by:
Non-swap volume: incorporating spot & forward outrights
Swap volume: incorporating swap/forward roll volumes and only counting one leg of each swap transaction
Options volume: Vanilla options volumes on a notional basis (when the market share tables are calculated we will use a delta equivalent reduction on the notional volumes. Banks will be asked to give a delta figure for corporates and FI’s before the survey starts. Euromoney will average these figures to give a uniform delta value for the notional volumes for Corporates and FI’s)
5. Cross-currency swaps of duration greater than 1-year and rates products as a whole are not part of this volume
6. Swaps are single-leg counted. Outrights such as long-dated forwards are included in the survey regardless of maturity.
Additionally, votes are not included in the survey where:
FX service providers cast a vote for themselves.
Euromoney does not receive confirmation from the respondent of their identity.
An individual respondent submits more than one ballot and we cannot resolve the duplication.
A respondents volume is duplicating that of a colleagues from the same institution
PART 2 - Qualitative
This section is concerned with qualitative rankings of various aspects of the FX providers’ services.
In each category, for example, G10 Currency Research, the ranking is calculated as the arithmetic mean of all the voted ratings (from 1 -7 where 1= very poor 7 = excellent) given to a bank for that category, with any responses that vote 1 in every voted category or which vote 7 in every voted category excluded. To qualify in these categories, banks need to receive 5% of the total individual vote count in that category. Respondents who only respond with a volume figure for one bank in the counterparty section (i.e single-banked) will not have their ratings included in the qualitative rankings.
In 2012 we received 15,423 (13,039 in 2011) valid votes to the survey with a total volume of $208.6 trillion ($177.6 trillion in 2011).