=Paper=
{{Paper
|id=None
|storemode=property
|title=Which Data Can Be Useful to Make Decisions on Foreign Exchange
Markets?
|pdfUrl=https://ceur-ws.org/Vol-1356/paper_104.pdf
|volume=Vol-1356
|dblpUrl=https://dblp.org/rec/conf/icteri/Mesropyan15
}}
==Which Data Can Be Useful to Make Decisions on Foreign Exchange
Markets?==
https://ceur-ws.org/Vol-1356/paper_104.pdf
Which Data Can Be Useful to Make Decisions on Foreign
Exchange Markets?
Karine Mesropyan
Chekhova 41 Rostov-on-Don, Russia 344006
carine@list.ru
Abstract. A communication, settlement of deals, and other services for
participants of foreign exchange markets are mostly served by electronic
infrastructures. Knowledge of the volume change of aggregated data of deals is
useful for all evolved businesses to support their decisions in practice. This
paper investigates whether market data of infrastructures, namely CLS, SWIFT,
and ETFs, can be used as the volume indicators of some FX segments.
Keywords. Foreign Exchange, Data, Time Series, Review, Flow
Key Terms. DecisionMaking, Management
1 Introduction
The largest and the most influential market for the global and national economies,
the foreign exchange market (FX) is opened 24 hours worldwide. According to a
regular semi-annual market research an amount of monetary flows traded on the FX
in all national currencies (global FX volume) is estimated at 5 trillion US dollar
average per day (Bench & Sobrun, 2013). The survey of the Bank for International
Settlement (BIS survey) is an important source of the FX knowledge as it aggregates
semi-annual surveys from FX committees and provides aggregated statistics of the FX
market segments (Fratianni & Pattison, 2001). The monetary amount of trades in
every currency is considered in this research as a segment volume indicator of the FX
market.
The volume dynamics demonstrates variable numbers which affect exchange rates
for national currencies and present volatile level of risk for traders and investors. That
is why a kind of the FX volume indicator is a part of system of key performance
indicators for assets management among FX market participants. This system helps
them in planning and decision making processes such as investment, currency
diversification in saving, and development of transnational networks. Market volume
is also necessary to be evaluated by national market regulators and central banks in
order to have current information on global tendency of their currencies volumes and
exchange rates as a consequence.
The concept of volatility is used on financial markets to measure fluctuations of
exchange rates by their standard deviation during a taken time interval (Schwartz,
Byrne & Colaninno, 2011; Bubak, Kocenda & Zilkes, 2011). As the volatility of the
FX volume is a recognized quantity indicator of the exchange rates dynamics,
financial organizations use it in order to estimate their risks on the FX. They usually
predict in which extent the exchange rates fluctuate between current level and
expiration date.
In order to construct volatility-based market volume indicator, researcher is aimed
to investigate area of electronic statistics which is relevant to the FX volume. Several
types of time interval can be taken to construct the indicator, namely day, month, and
year. The importance of choosing of the time interval for data was emphasized in
previous research (Gill, Perera & Sunner, 2012, p.1): “Over recent years technological
developments and the digitisation of information and activity have generated a vast
array of electronic data, which can potentially be analysed on a daily basis, or even in
real time. Some of these data cover very large numbers of individuals and businesses
– far more than many traditional surveys used by statistical agencies – and have the
potential to be useful for monitoring and measuring aggregate economic conditions.”
The BIS survey cannot provide frequent FX statistics, meanwhile market
participants become also more interested in monthly volume indicators (Cerutti,
Claessens & McGvair, 2012).
Alternative sources of information are investigated in this paper. The FX electronic
communication and settlement infrastructures also aggregate statistics on their
transactions. Continuously Linked Settlement Bank (CLS) and Society for Worldwide
Inter-bank Financial Telecommunication (S.W.I.F.T. or SWIFT) serve financial
organizations by secure settlement of their interests on the FX. Exchange traded funds
(ETFs) also widely play on the FX market as investment companies which provide
efficient and attractive sets of financial instruments in a variety of currencies.
With this research we intend to get a better idea of how the FX market can be
measured by using globally aggregated electronic statistics of CLS, SWIFT, and
ETFs.
In the first paragraph we state the problem. In the second paragraph we study
distinctions and commonality of CLS, SWIFT, and ETFs data regarding to the
segments of the FX market. In the third paragraph we investigate relationships
between some FX segments volume indicators, namely ETFs in developed currencies
and CLS. In the fourth paragraph we discuss methods and our results and in the fifth
paragraph we discuss findings and make conclusions.
2 Foreign Exchange Markets in the Last Decades
Developed in 2000s investment opportunities provide a ground for constant
enlargement of trades in developing currencies (Bryan, 2008). The latter are
reasonably called exotic currencies among the FX practitioners (Tsuyuguchi &
Wooldridge, 2008) because they did not find suitable conditions for stable growth
worldwide. Thus, from beginning of the post Bretton Woods system in condition of
US dollar domination less than 5 % of global trading was made in other local
currencies (Pojarliev, 2005). “The relative insignificance of these currencies in
international markets reminds us of the growing disjuncture between countries and
“their” currencies. Most Indian- and Chinese-related trade and investment is
undertaken in US dollars, with that currency often being used directly without any
formal currency conversation (for example, for the purchase of US bonds).
Alternatively, for those outside India and China looking for a share of their growth
economies, it is possible, using derivatives, to take on exposure to their growth
without the need for actual investment in these countries nor for foreign exchange
conversion to local currencies.” (Bryan, 2008, p. 503).
One could see different environments struggling with implementation of
diversification strategies of exchange, saving, and investment in “3 big currencies”
and domestic currency. Term of “3 big currencies”, namely US dollar, euro, and yen,
has become recognized due to trinity’s domination in the FX structure (Pojarliev,
2005).
Next, after a crisis of 2008 the global FX market has created a fertile ground for
diversification. The post-crisis market conditions have immediately influenced the
exchange in a variety of currency pairs, especially in the currencies of developing
countries (Bryan, 2008): “…with a declining role for the “big 3” currencies in
aggregate, perhaps even the status of any leading national currency being treated as a
proxy global anchor is being challenged. Consistent with this trend, it is apparent that
foreign exchange is itself being treated increasingly as an asset class (a store of value)
as well as a means of exchange, so that investors see intrinsic benefit in holding a
wider range of currencies in a diversified asset portfolio.”
Nowadays the first candidates to leaders on the FX are Chinese renminbi (yuan)
and the Indian rupee which present economies of two members of BRICS (Brasil,
Russia, India, China, and South Africa). By World Bank estimation, BRICS
contributes a quarter to global domestic product that is more than any other group of
developing countries. Although “evolution of the Chinese currency on the FX market
remains slow and runs the risk of failing” (Batten & Szilagyi, 2012, p.2), there is an
expectation of long-term shift in currency markets. As an evidence of this tendency,
New Development Bank (BRICS Development Bank) has been established in 2013
by 5 developing countries as an alternative to International Monetary Fund and World
Bank.
Along with currencies diversification a way of presence at the FX is also important
for market players. As BIS survey reported, some participants of the FX communicate
for trading by using services of brokers but major players replace such supervision by
making over-to-counter operations (OTC) themselves. Such market participants are
usually members of CLS or SWIFT.
CLS bank serves other banks and financial institutions by mitigating a settlement
risk that appears when one party of exchange pays the currency it sold but does not
receive the currency it bought (Fisher & Ranaldo, 2011). This kind of risk is called a
settlement risk. CLS executes exchange operations (CLS instructions) through
provision of its unique payment operation versus payment settlement service. Owing
to its service value the CLS is highly appreciated by international financial
community (Fisher & Ranaldo, 2011). According to CLS strategy, its large
contribution to the developed markets accompanies by absence on the developing
ones.
To serve secure FX transactions, SWIFT plays another role in the industry (Scott &
Zachariadis, 2010).
It communicates financial institutions, corporations and their counterparties by
SWIFT messages. Their customers are financial institutions, fund managers and
brokers, fund managers, settlement members and central settlement systems including
CLS members. SWIFT message (MT300) consists of all information about
transaction on the foreign exchange such as currency pair, monetary amount, type of
trading, and others (SWIFT, 2015).
Nowadays SWIFT possesses a worthy demand in the industry because of its
capacity to operate with high value delivered and relatively lower costs in comparison
with rivals on both types of markets (developed and developing). This stable trend
implies its importance in the industry which provides the SWIFT data potential
contribution to the FX volume measurement. Besides, SWIFT does not limit its
custodians by a kind of currency to trade. CLS, on the contrary, executes operations in
17 currencies which are mostly developed.
Current tendency on the market is an extremely high growth of ETF segment in
both developed and developing economies. Nowadays such indices are traded in a
number of currencies on the FX due to its attractiveness for investors.
3 Use of Global Data of CLS, SWIFT and ETFs
CLS data was usually an adequate way of the FX volume estimation. Its monthly
market review indicates dynamics of trades on the developed part of the FX. For
instance, recently BIS have leveraged CLS information and own data (Bench &
Sobrun, 2013). Owing to a mixed approach in monthly numbers measurement,
estimated this way FX dynamics was able to explain sources of odd jumps and drops
of the market by concrete instruments, which have been described in the BIS survey
in a detailed way. Fig. 1 illustrates this first attempt to measure aggregated FX
volume for all currencies, including the developed and developing ones owing to the
local FX committees’ contribution. It makes clear the necessity of the different
sources of data combination.
Meanwhile, being outside of mutual work of CLS and BIS, SWIFT could pretend
to be considered as a source of data for the FX volume estimation. Its statistics is
usually published only in its annual market review where the FX trends are shortly
described and illustrated by SWIFT service activity during the year.
Resent research (Cook & Soramaki, 2014) shows that SWIFT data (MT300
message type) is correlated with the FX volume for currency pairs of US dollar and
Chinese renminbi (yuan). Authors have found linear relationship between these values
(Fig. 2). Absence of similar research of US dollar and yuan from other data sources
(CLS, for example) makes impossible to conclude which data is more useful for
market analysts by comparing with results of (Cook & Soramaki, 2014) with others.
Fig. 1. Monthly CLS data in estimation of global FX volumes in all type of currencies (Bench
& Sobrun, 2013)
Alternative source of information comes from investment funds or exchange traded
funds which publish their indices volatility. The concept of volatility is used to
indicate uncertainty regarding degree of ETFs’ volume changes (Schwartz, Byrne &
Colaninno, 2011). Although it has been traditionally used for analysis of exchange
rates volatility dynamics (Britten-Jones & Neuberger, 2000), we have found examples
of its application to measure the range of probable change of traded volume by its
dispersion analysis (Melvin & Peiers Melvin, 2003).
Fig. 2. Monthly relationship between SWIFT data and developed currencies FX segment
volume (Cook and Soramaki, 2014, p.27)
As it was reported above nowadays constant growth of the ETF worldwide
accompanies by increasing ETF contribution to the financial markets of developing
economies. The evolution of financial instruments led to use of ETFs which had
provided implementation of extremely successful trading strategies after the global
financial crisis in 2008 (Bryan, 2008, p. 502). As a result, in developed countries, for
instance, in the United States ETFs have contributed 40 % of the financial market
volume (Guedj & Huang, 2009). As for developing countries, ETFs in currencies
have contributed 23 % of the FX market (Fig. 3).
Fig. 3. Yearly ETFs contribution to FX developing currencies segment volume
We have found several studies which are focused on relation between volatility of
volume of one of the FX ETF (FXE) and volume of other market segments (Daigler,
Hibbert & Pavlova, 2014). Other researchers studied the ETF segment statistics much
more widely (Li, Klein, & Zhao, 2012). The common way to construct time series is
ARIMA method but the specific for volatility variables is ARCH method (Le &
Zurbruegg, 2010). Data about ETFs flows is available on website of Currencyshares’
and Powershares’ on-line databases which present three largest global currency ETFs
indices (US dollar, euro, and yen).
4 Methods and Findings
We have analysed the FX structure by using information from the BIS Survey
(2013). According to the survey, traditionally the largest volumes of trading take
place in Europe, USA and Japan. Respectively, this trend is presented by the biggest
volumes of the FX trading in the main currency pairs, namely Euro versus US dollar
(EUR-USD), US dollar versus Yen (USD-JPY), Euro versus Yen (EUR-JPY).
Data have been taken from the website of CLS bank, Currencyshares’ and
Powershares’ on-line databases which present three largest global currency ETFs
indices, namely ETF FXE (euro), ETF FXY (yen), and ETF UUP (US dollar).
Data of this research consist of five elements for every month during period from
January 2008 till March 2014, namely:
average number of CLS operations (instructions);
volatility of volume of exchange-traded fund FXE;
volatility of volume of exchange-traded fund FXY;
volatility of volume of exchange-traded fund UUP.
We have studied autocorrelations of these volumes. Table 1 shows that all levels
of tested variable of t-statistics are not statistically significant for CLS. According to
these results, there are significant autocorrelations between the nearest levels of lags
(from month to month) for each time series of ETFs except the time series of CLS.
Table 1. t-statistics of autocorrelation (ACF) and partial autocorrelation functions (PACF)
Lag ACF PACF
ETF ETF ETF ETF ETF ETF
№ CLS CLS
FXE FXY UUP FXE FXY UUP
1 6,59* 4,64* 5,74* 1,73 6,48* 4,56* 5,64* 1,62
2 5,37* 4,30* 4,93* 0,29 0,48 2,42* 1,82 -0,31
3 4,34* 3,14* 4,24* 0,95 -0,03 0,13 0,63 0,97
4 3,15* 2,90* 3,10* 0,69 -0,90 0,56 -0,80 0,01
5 2,19* 2,66* 3,16* -1,13 -0,25 0,61 1,17 -1,32
6 1,28 1,99 3,06* -1,79 -0,52 -0,35 0,74 -1,03
7 0,29 2,87 2,01* 0,31 -0,82 1,63 -1,33 1,13
8 -0,12 1,68 2,16* -0,15 0,47 -0,85 0,64 -0,32
9 -0,62 0,86 1,97 -1,05 -0,45 -1,49 0,47 -0,11
10 -1,17 1,02 1,03 -0,59 -0,58 0,76 -1,21 -0,39
11 -1,05 0,92 1,32 -0,09 0,83 0,33 0,43 -0,6
12 -0,69 0,64 1,39 -1,59 0,75 -0,63 0,91 -1,19
Significant levels are signed by (*) on the base of t-statistics critical values at the
confidence level of 97,5%
Next step should consist of regression models constructing on the base of the time
series by using the results of significant lags’ autocorrelation. As we have not found
out existence of linear relationship between CLS operations from month to month, we
could not estimate regression of the CLS and ETFs volumes.
5 Conclusion
A lack of frequently available data can negatively affect strategic decisions of
businesses. This research has been motivated by industry’s willingness to explain
sources of the FX market volume dynamics in developed and developing currencies.
In this field we found out several results. Owing to international finance
transformations, nowadays currencies of developing countries become more often
used among deals on the FX market than several years ago. This trend had been
appeared in the post-crisis period after 2008. The financial organizations had to
struggle between two options by making choice on the FX markets. They could adapt
to decreasing trends of US dollar domination or they could seize opportunities
relating to currencies of developing economies. In 2013 the BIS survey has concluded
that unpredictable trends on emerging currencies markets attract more attention of
participants to this FX segment volume measurement.
We have also studied what kind of time interval should be taken for the FX volume
indicator. We have found that market participants are interested in the FX volume
indicators to fill absence of monthly data (Cerutti, Claessens & McGvair, 2012, p.2).
Our findings are confirmed by existing statistics source, namely the Triennial Central
Bank Survey of Foreign Exchange and Derivatives Market Activity. It collects only
long-term overall statistics of volume so the survey cannot respond to need of
frequent availability of the FX data without additional market information. That
confirms the opinion that “while official statisticians are increasingly using electronic
data in the production of economic indicators, this is still very much in its infancy.”
(Gill, Perera & Sunner, 2012, p.1).
In our review we have concluded that month could be taken as a time interval to
make decisions on the FX market by constructing volatility-based volume indicator.
Meanwhile, our experimental findings did not provide enough evidence for that.
Next, we have studied the FX data regarding developed and developing currencies.
The FX has a number of participants and nowadays only three infrastructures’
performances can indicate its overall activity’s performance from month to month.
Thus, CLS, SWIFT and ETFs data’ features analysis has helped to shed a light to data
search for the FX volume estimation.
On the one hand, as class of major developed currencies has mostly become an
area of CLS business. CLS data can help to measure a volume of trading in currencies
of developed countries. CLS does not include trades in currencies of the BRICS
countries. South Africa is only one exemption in this group of 5 countries as its
currency is considered as a major one and it can be traded by CLS members.
On the other hand, today SWIFT is known as a provider of efficient supply chain
for financial organizations in majority of countries including developing ones. SWIFT
services are available for all exotic currency pairs on both developed and developing
markets. That is why SWIFT membership has become more popular, especially for
banks which were not members of CLS.
Potential role of SWIFT information for the FX volume measurement has not been
acknowledged yet. Meanwhile, SWIFT has already presented its contribution to
economy forecast which was presented by dynamic models for developed (Gill,
Perera & Sunner, 2012), developing, and global economies (Bauwens, Gillain &
Rombouts, 2011). Thus, SWIFT analytics are more concentrated on current trends of
some developing currencies’ internationalization such as Chinese Yuan, RMB (Batten
& Szilagyi, 2012).
Finally, in our research we have stated a question: ‘To which extent does volume
of CLS activities indicate the standard deviation of volume during a month for three
major ETF FX segments in developed currencies?’ We have calculated the ETF
standard deviation on the base of daily volumes in order to aggregate data on the
volatility of ETF volume for each month. We have obtained results which have not
approved a hypothesis that relation between CLS volumes and ETFs volatility-based
estimation of volumes does exist. We were focused on the ETF segment and its three
major representatives. These imperfections have affected our research by its inability
to extrapolate directly our results for the whole ETF segment of the FX. Next stage of
this research could be conducted with more types of ETFs statistics and SWIFT data.
As CLS bank and SWIFT are rapidly evolving competitors in the industry, they
consider promotion of the own business intelligence to the FX volume estimation. It
makes possible to start research projects in this field. In future research a question can
be stated as following: ‘To which extent do SWIFT and CLS activities indicate the
volume of the major FX segments?’ The research objective can be FX volume
indicator constructing. Sources of information for the FX size measurement can come
from the website of CLS bank, SWIFT, and ETFs (Currencyshares, Powershares, and
others) on-line databases.
References
1. Batten, J. A. & Szilagyi, P. G. The Internationalisation of the RMB: New Starts, Jumps and
Tipping Points. SWIFT Institute Working Paper, 2012-001. Retrieved from
http://ssrn.com/abstract=2325340 (2013)
2. Bauwens, L., Gillain, N., Rombouts, J.V.K. Forecasting GDP Growth Through SWIFT
Information. CORE UCL Report. Retrieved from
http://www.swift.com./about_swift/shownews?param_dcr=news.data/en/swift_com/2013/P
R_index.xml (2011)
3. Bench, M. & Sobrun, J. FX Market Trends Before, Between and Beyond Triennial
Surveys. BIS Quarterly Review, December Retrieved from
http://www.bis.org/publ/qtrpdf/r_qt1312f.htm (2013)
4. BIS (Bank for International Settlement). Triennial Central Bank Survey of Foreign
Exchange and Derivatives Market Activity in 2013. Monetary and Economic Department
Working Paper. December. Basel, Switzerland. Retrieved from
http://www.bis.org/publ/rpfx13.htm (2013)
5. Box, G. E. P., Jenkins, G. M., Reinsel, G.C. Time Series Analysis, Forecasting and Control.
Hoboken: Wiley (2008)
6. Britten-Jones, M. & Neuberger, A. Option Prices, Implied Price Processes and Stochastic
Volatility. Journal of Finance, 55, 839 – 866 (2000)
7. Bryan, D. The global Foreign Exchange Market: An Interpretation of the Bank of
International Settlements’ Survey of Foreign Exchange and Derivative Market activity.
Global Society, 22 (4), October, 491 – 505 (2008)
8. Bubak, V., Kocenda, E., Zilkes, F. Volatility Transmission in Emerging European Foreign
Exchange Markets. Journal of Banking and Finance, 35, 2829 -2841 (2011)
9. Cerutti, E., Claessens, S., McGvair, P. Systemic Risk in Global Banking: What Available
Data Can Tell Us and What More Data are Needed? BIS Working Papers, 12. Retrieved
from http://www.bis.org/publ/work376.htm (2012)
10. Cook, S. & Soramaki, K. FX MT300 Correlation Analysis. Mapping Financial Network
Analysis (FNA) Report, March (2014)
11. CLS bank. Historical Data from Official Website: http://www.cls-group.com/
MarketInsight/Pages/ReportArchive.aspx (2013)
12. Guedj, I. & Huang, J. Are ETFs Replacing Index Mutual Funds? AFA 2009 San-Francisco
Meetings Paper. Retrieved from http://ssrn.com/abstract=1108728 (2009)
13. CurrencyShares Euro Trust (Symbol: FXE). Historical Data from Official Website:
http://currencyshares.com/products/overview.rails?symbol=FXE
14. CurrencyShares Japanese Yen Trust (Symbol: FXY). Historical Data from Official
Website: http://currencyshares.com/products/navs.rails?symbol=FXY
15. Daigler, R.T., Hibbert, A.M., Pavlova, I. Examining the Return-Volatility Relation for
Foreign Exchange: Evidence From Euro VIX. The Journal of Futures Markets, 34 (1), 74 –
92 (2014)
16. Fisher, A.M. & Ranaldo, A. Does FOMC News Increase Global FX Trading? Journal of
Banking and Finance, 35, 2965 – 2973 (2011)
17. Fratianni, M. & Pattison, J. Review Essay: The Bank of International Settlements: An
Assessment of its Role in International Monetary and Financial Policy Coordination. Open
Economies Review, 12, 197 – 222 (2001)
18. Gill, T., Perera, D., Sunner, D. Electronic Indicators of Economic Activity. Australian
Reserve Bank Bulletin, June (2012)
19. Le, V. & Zurbruegg, R. The Role of Trading Volume in Volatility Forecasting.
International Financial Markets, Institutions, and Money, 20, 533 – 555 (2010)
20. Li, M., Klein, D., Zhao, X. Empirical Studies of ETF Intraday Trading. Financial Services
Review, 21, 149 – 176 (2012)
21. Melvin, M. & Peiers Melvin, B. The Global Transmission of Volatility in the Foreign
Exchange Market. Review of Economics and Statistics, 85, 670 – 679 (2003)
22. Pojarliev, M. Performance of Currency Trading Strategies in Developed and Emerging
Markets: Some Striking Differences. Financial Markets and Portfolio Management, 19 (3),
297 – 311 (2005)
23. PowerShares DB US Dollar Index Bullish Fund (Symbol: UUP). Historical Data from
Official Website (click on ‘historical_navs_uup’): https://www.invesco.com/
portal/site/us/financial-professional/etfs/product-detail?productId=uup
24. Qian, B. & Rasheed, K. Foreign Exchange Market Prediction with Multiple Classifiers.
Journal of Forecasting, 29, 271 – 284 (2010)
25. Sarno, L. & Taylor M. The Microstructure of the Foreign-Exchange Market: A Selective
Survey of the Literature. Princeton Studies in International economics, 89 (2001)
26. Schwartz, R.A., Byrne, J.A., Colaninno, A. Volatility. Risk and Uncertainty in Financial
Markets. Springer Science+Business Media (2011)
27. Scott, S. V. & Zachariadis, M. A Historical Analysis of Core Financial Services
Infrastructure: Society for Worldwide Interbank Financial Telecommunications (SWIFT).
Information Systems and Innovation Group, London School of Economics and Political
Science, London, UK. Working paper series, 182 (2010)
28. SWIFT Creating Confidence in a Changing World. Annual Review. Retrieved from
http://www.swift.com/assets/swift_com/documents/about_swift/2013_SWIFT_
Annual_Review.pdf (2013)
29. SWIFT Harnessing Timely Data for Better FX Decisions. Information Paper. January.
Retrieved from http://www.swift.com/assets
/swift_com/documents/products_services/Harnessing_Timely_Data_For_Better_FX_Decisi
ons.pdf (2015)
30. Tsuyuguchi, Y. & Wooldridge, P.D. The Evolution of Trading Activity in Asian Foreign
Exchange Markets. Emerging Markets Review, 9, 231 – 246 (2008)