Srdan Tatomir, Iryna Kaminska, Marek Raczko and Gregory Thwaites
How have equity markets responded to news about Brexit? To answer this we split firms into those whose share prices are particularly sensitive to Brexit-related news, and those which are not. The latter group provides a “control sample”, against which to assess the impact of individual pieces of news on the former. The ratio of the two groups’ prices gives a barometer of equity market sentiments around Brexit. So far, this measure points to downward pressure on valuation of companies more exposed to Brexit. The bulk of the fall occurred on the night of the referendum, with little movement afterwards, suggesting little additional “news” from subsequent developments beyond the immediate aftermath.
Brexit-related moves in equity prices provide a measure of investors’ perception of the long run effect of leaving the EU on companies’ profitability….
Equity prices are forward-looking and reflect investors’ perceptions of the future profitability and riskiness of companies’ based on the information they have at any point in time. If Brexit news affects markets’ central expectations of profit streams or changes markets’ perceived uncertainty around it, movements in the equity price immediately after that news will reflect the markets’ assessment of that news. Although such price-based measures cannot distinguish the relative contributions of the two channels.
To gauge the impact of Brexit-related news we adopt a three step approach: first we attempt to isolate Brexit related news events, we then use the events to identify which firms are particularly sensitive and insensitive to Brexit news, and finally we take the ratio of price indices of Brexit-sensitive to non-Brexit sensitive firms in the day of these events to estimate the “size” of the Brexit induced move.
We begin by constructing a database of Brexit-related events …
Our starting criterion was that the sterling dollar exchange rate moved by more than 0.5% over a 30-minute window. Based on that list, we applied our own judgement as to whether there was an obvious Brexit-related event coinciding with that window. For example, sharp moves on referendum night were judged to be Brexit related (in the absence of any other major news that might have moved the exchange rate). Subsequently, we looked for news which might materially influence perceptions of post-Brexit arrangements. The full list of 28 events we use is in the Appendix.
… we then look at how companies’ share prices move in response …
Just looking at the FTSE All-Share index is not the best guide since it contains many multinational firms that produce and sell their products outside the UK/EU and that might be largely unaffected by Brexit. Indeed there are some companies which are part of FTSE All-Share index, which have no operations at all in the UK. On the other hand, some companies which focus on foreign sales and are not sensitive to UK demand might nevertheless be exposed to potential post-Brexit barriers to either exporting or operating abroad. Therefore a simple a priori split based on where companies trade may not be appropriate.
We therefore adopt a different approach which “lets the data speak”: we identify the companies whose share prices appear to react significantly to Brexit related events. We measure firms’ sensitivity to Brexit by analysing events in a simple CAPM framework. We gauge how much the equity prices of firms in the FTSE All-Share react to Brexit-related events over and above what can be explained by general market variation. We pick a statistical threshold to choose which firms are likely to be particularly sensitive and a separate threshold to choose which firms are not particularly sensitive to Brexit-related news. This allows us to have a conservatively selected list of Brexit sensitive and insensitive companies. It is also important to flag that since we drop the firms “in the middle” – these two lists include only a fraction of all FTSE All-Share companies.
We model equity returns in dollars to account for the impact of the exchange rate on firms’ equity prices, particularly on those with foreign revenues. Also, dollar-denominated returns might give us a clearer signal as Brexit-related news could affect both sterling and the euro.
The advantage of this approach is that is doesn’t impose any judgement or prior assumptions about which firms or sectors “should” be Brexit-sensitive. Or any about the potential direction of any Brexit-effect on a given company’s share price.
The composition of the group of Brexit-sensitive firms is shown below, and gives a sense of which sectors investors tend to believe will be more affected by Brexit:
Chart 1 – UK equity indices decomposed by industry
Note: the decompositions are weighted by market value on March 6, 2018. We use the ICB industrial classification.
Source: Thomson Reuters Worldscope and authors’ calculations.
Our group of Brexit-sensitive firms mostly contains financials, which are likely to be highly sensitive to the outcome of negotiations. It also contains consumer-oriented firms and industrials, which might be more exposed to moves in consumer confidence and any change in the UK’s future trading relationship with the EU. There may also be other affected sectors in the economy which are not sufficiently represented by firms listed on the London Stock Exchange.
Interestingly, there is a considerable overlap with the UK-focused equity index. 60% of firms that are our measure classifies as sensitive to Brexit news are also UK-focused. In terms of market value, the overlap is nearly 80%. But not all firms affected by Brexit will be UK-focused. Some affected firms, such as airlines, are based in the UK, but are internationally oriented and hence depend on the UK’s agreements with the EU.
… and then compare the movements in share prices between our two groups.
In line with FTSE All-Share index methodology we use individual firms’ stock market capitalisation to form price indices for Brexit-sensitive and not sensitive firms.
Chart 2 – Changes in dollar-denominated UK equity indices since 1 January 2016
Note: data goes up to March 6, 2018.
Source: Thomson Reuters Datastream and authors’ calculations.
Plotting the two indices reveals that the Brexit-sensitive index has underperformed the non-Brexit sensitive one since early 2016. The largest movement occurred on referendum night itself, when Brexit sensitive stocks fell by about 20pp more.
Since our approach relies on a somewhat arbitrary choice of cutoff-value to identify sensitivity, we computed several measures based on different values. These are shown below in the red swathe.
Our summary measure suggests that Brexit has had a downward and fairly persistent impact on the equity prices of Brexit-sensitive firms …
Chart 3 – A measure of equity market perceptions of a Brexit impact
Note: data goes up to March 6, 2018.
Source: Thomson Reuters Datastream and authors’ calculations.
Although there is some disagreement between the magnitude of the different measures, their evolution over time is similar. The swathe turns negative shortly after the announcement of the referendum, and then remains roughly stable for the final two months of the campaign. There is a sharp downward movement on referendum night, as markets, expecting a remain vote, suddenly realised the outcome would be different.
… but the bulk of the impact happened on the night of the referendum
But beyond a few days after the result, market sentiment hasn’t changed that much in either direction- despite a high degree of media coverage and public debate. One explanation is subsequent developments have been difficult to price. Another, not necessarily mutually exclusive, explanation is that subsequent events contained little additional information about the likely shape of post-Brexit arrangements and their impact on company profitability/risk.
The price of Brexit sensitive firms tends to fall in reaction to ‘hard’ Brexit news events, relative to those of non-sensitive firms. We interpret this as market participants, on average, associating the flow of ‘hard’ Brexit news with lower profitability/greater uncertainty and hence marking down valuations.
Of course no identification scheme is perfect. Because ours focusses on the difference between share prices, to the extent that a piece of news affects all firms equally, our methodology will fail to detect it. But we think it is unlikely that i) this effect would operate in the opposite direction to the differential effect we focus on (noting the fall in almost all share prices on referendum night itself) and ii) that it would be large enough to change the sign (or have a material effect on the size).
This, and the fact that we drop firms “in the middle”, means that our measure doesn’t have a one-to-one mapping with share prices, and so cannot be read as a direct numerical estimate of the effect of Brexit news on the value of UK PLC as a whole. But we think it can be a useful barometer of how market sentiment has evolved in response to the cumulative Brexit-related news flow.
Srdan Tatomir, Iryna Kaminska, Marek Raczko all work in the Bank’s Macro Financial Analysis Division and Gregory Thwaites works in the Bank’s Global Analysis Division.
If you want to get in touch, please email us at bankunderground@bankofengland.co.uk or leave a comment below.
Comments will only appear once approved by a moderator, and are only published where a full name is supplied. Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.
True, It’s moved sideways effectively since then but the UK has not left and there may be another move when that occurs. Will it be up or down? Will the UK have to become more deregulated than the EU and more open and friendly to foreign firms or does leaving the large market trump all else?
If the pound depreciates yet equities still
fall can we infer that the non tariff barriers and lack of access are deemed more important factors than currency competitiveness?
CAPM maybe one way to measure systemic sensitivity, but have you considered tail-measures for such rare events as shown in Van Oordt and Zhou’s paper titled Systemic risk under extremely adverse market conditions?