incorporated into the interest and exchange rates. Therefore,
only surprise or unanticipated announcements and results
will lead to significant volatility in exchange rates. Galati
and Ho (Fama 1970), found that unscheduled announcements have a larger impact on exchange rates than scheduled
announcements. The rationale is that agents have more time
to prepare for an eventuality before it occurs if the
announcement is scheduled. Other research by Ederington
and Lee (Cosset and De La Rianderie 1985) revealed that the
system can take less than 40 seconds to react to news, which
signifies a robust semi-strong form efficient market.
A surprise announcement does not allow for such
preparation time and hence exchange rates are likely to react
to this. It can be assumed that the market will react significantly to unpredicted outcomes, especially seeing as information up to when the event takes place is already factored
into the market. The EU referendum of 23 June can be
related to such an event, seeing as the polls indicated that the
majority intended to vote to remain. This research intends to
test the semi-strong form efficiency, where the political event
results will portray the announcement. As the GBP is a
relatively stable currency and one of the strongest globally, it
is expected that it would be less prone to shocks following
political events or announcements.
Frankel and Wei (Fair 2003) employ a new technique for
obtaining de facto exchange rate regimes for different
countries. In their research, they criticize past research on
exchange rates, which was hinged on a major currency in
which they defined as their currency value of interest. The
currency, which has most commonly been used, is the US
dollar. As much as this may be applicable for currencies,
which are stabilized by the authorities, it is not appropriate to
anchor all currencies of interest to the dollar. They further
explain that the use of currency anchoring can be used with
practicality, such as using the Euro as the base currency for
studies conducted in central Europe, or the US dollar for
Latin American research. There is also the challenge with
anchoring a single currency to emerging economy currencies. A better practice that could be put in place is to estimate
endogenously, whether the currency is tied to a single currency or a basket of currencies. This research, though, does
not explore the endogenous estimation; it does consider the
better approach of using a basket of currencies, rather than
relying on a single currency for obtaining the value of the
currencies of interest. This is revisited in the methodology.
2.1 New Information on Asset Prices
There have been many studies, which analyze the effect that
new information has on various financial assets, though there
is still a considerable amount of research which focused on
the stock markets rather than the currency markets. An
element, which has a major effect on the currencies is foreign
trade as it can significantly influence a currency’s movement
mainly via the economy’s balance of payments.
Calin (2015) investigated the reaction of different currencies to foreign trade announcements using an econometric event study methodology, finding that trade news had
a significant influence on the currencies, which were
observed. Though trade announcements are not necessarily
changing in trade amounts or volumes, they generally imply
that levels of trade will be affected. This notion will be
reviewed in relation to the EU referendum result
announcement in the macroeconomic implication section.
Currency movements, as described above are generally
dictated by macroeconomic shifts. It is with this notion that a
considerable amount of literature has focused on analyzing
the effect, which macroeconomic announcements have had
on currencies. Daily frequencies have been utilized in
research, which aims to quantify currency reactions to economic phenomenon. Other research by Fair (Frankel 2007)
used intra-daily data to show the impact that macroeconomic
events have on currencies and other markets. In the interest
of this research, it should be noted that the British Pound
was one of the affected currencies. Research conducted by
Anderson et al. (2003) analyzed the effect that macroeconomic announcements have on currency exchange rates.
They found that from the currencies they observed, good
news influenced the currency changes less than bad news
did. This would suggest that the currency market is more
susceptible to over-reacting to bad news than it is to good
news.
2.2 Literature on the Model
There are two elements of empirical literature on exchange
rate reactions to announcements or news that can be applied;
(1) the difference between actual and expected values following announcements and (2) the modelling of news or
announcements as time-series developments within the
observed timeline. This research employs variations of the
first element seeing as unanticipated results proved to be a
significant determinant of exchange rate volatility (Frenkel
1981). Unlike Fenkel’s research, this study uses daily data
rather than monthly data, which possibly failed to capture
the precise event time and significant effect, leading to
weaker results. The decision to use daily data works best
with this event as the window is quite narrow. This will
allow the models to capture the event moment and from that,
more significant results can be obtained. This daily data will
then be applied to two models from which we can observe
the abnormal returns for the event.
The abnormal returns model is obtained via two different
models: the constant mean approach and the market model.
The Effect of the EU Referendum on the GBP: Evidence From Brexit
229
only surprise or unanticipated announcements and results
will lead to significant volatility in exchange rates. Galati
and Ho (Fama 1970), found that unscheduled announcements have a larger impact on exchange rates than scheduled
announcements. The rationale is that agents have more time
to prepare for an eventuality before it occurs if the
announcement is scheduled. Other research by Ederington
and Lee (Cosset and De La Rianderie 1985) revealed that the
system can take less than 40 seconds to react to news, which
signifies a robust semi-strong form efficient market.
A surprise announcement does not allow for such
preparation time and hence exchange rates are likely to react
to this. It can be assumed that the market will react significantly to unpredicted outcomes, especially seeing as information up to when the event takes place is already factored
into the market. The EU referendum of 23 June can be
related to such an event, seeing as the polls indicated that the
majority intended to vote to remain. This research intends to
test the semi-strong form efficiency, where the political event
results will portray the announcement. As the GBP is a
relatively stable currency and one of the strongest globally, it
is expected that it would be less prone to shocks following
political events or announcements.
Frankel and Wei (Fair 2003) employ a new technique for
obtaining de facto exchange rate regimes for different
countries. In their research, they criticize past research on
exchange rates, which was hinged on a major currency in
which they defined as their currency value of interest. The
currency, which has most commonly been used, is the US
dollar. As much as this may be applicable for currencies,
which are stabilized by the authorities, it is not appropriate to
anchor all currencies of interest to the dollar. They further
explain that the use of currency anchoring can be used with
practicality, such as using the Euro as the base currency for
studies conducted in central Europe, or the US dollar for
Latin American research. There is also the challenge with
anchoring a single currency to emerging economy currencies. A better practice that could be put in place is to estimate
endogenously, whether the currency is tied to a single currency or a basket of currencies. This research, though, does
not explore the endogenous estimation; it does consider the
better approach of using a basket of currencies, rather than
relying on a single currency for obtaining the value of the
currencies of interest. This is revisited in the methodology.
2.1 New Information on Asset Prices
There have been many studies, which analyze the effect that
new information has on various financial assets, though there
is still a considerable amount of research which focused on
the stock markets rather than the currency markets. An
element, which has a major effect on the currencies is foreign
trade as it can significantly influence a currency’s movement
mainly via the economy’s balance of payments.
Calin (2015) investigated the reaction of different currencies to foreign trade announcements using an econometric event study methodology, finding that trade news had
a significant influence on the currencies, which were
observed. Though trade announcements are not necessarily
changing in trade amounts or volumes, they generally imply
that levels of trade will be affected. This notion will be
reviewed in relation to the EU referendum result
announcement in the macroeconomic implication section.
Currency movements, as described above are generally
dictated by macroeconomic shifts. It is with this notion that a
considerable amount of literature has focused on analyzing
the effect, which macroeconomic announcements have had
on currencies. Daily frequencies have been utilized in
research, which aims to quantify currency reactions to economic phenomenon. Other research by Fair (Frankel 2007)
used intra-daily data to show the impact that macroeconomic
events have on currencies and other markets. In the interest
of this research, it should be noted that the British Pound
was one of the affected currencies. Research conducted by
Anderson et al. (2003) analyzed the effect that macroeconomic announcements have on currency exchange rates.
They found that from the currencies they observed, good
news influenced the currency changes less than bad news
did. This would suggest that the currency market is more
susceptible to over-reacting to bad news than it is to good
news.
2.2 Literature on the Model
There are two elements of empirical literature on exchange
rate reactions to announcements or news that can be applied;
(1) the difference between actual and expected values following announcements and (2) the modelling of news or
announcements as time-series developments within the
observed timeline. This research employs variations of the
first element seeing as unanticipated results proved to be a
significant determinant of exchange rate volatility (Frenkel
1981). Unlike Fenkel’s research, this study uses daily data
rather than monthly data, which possibly failed to capture
the precise event time and significant effect, leading to
weaker results. The decision to use daily data works best
with this event as the window is quite narrow. This will
allow the models to capture the event moment and from that,
more significant results can be obtained. This daily data will
then be applied to two models from which we can observe
the abnormal returns for the event.
The abnormal returns model is obtained via two different
models: the constant mean approach and the market model.
The Effect of the EU Referendum on the GBP: Evidence From Brexit
229