In the aftermath of the referendum, the British pound lost more than 10 per cent of its value against the euro. But following a short-term slump in key business indicators, the economic situation in the United Kingdom stabilised once again. This applies both to the recovery of stock prices (Diermeier, 2016) and especially to the relevant economic sentiment indicators. One explanation for this surprising resilience is this: While there remains a high level of insecurity regarding how things will continue to develop, nothing is changing in the trade relations between the UK and the EU for the time being. Furthermore, the depreciation and interest rate reduction are having a stabilising effect. However, aggregated prognoses by Consensus Forecasts for the UK for the year 2017 have decreased since the Brexit referendum, from over 2 per cent to a solid 0.5 per cent with an upward tendency. This basis allows for the calculation of the impact of the growth slowdown by about 1.5 percentage points and the depreciation of the pound by some 10 per cent on German trade with the UK and on Germany’s economic output. An econometric study (cointegration analysis; see Johansen, 1995) was conducted for this purpose, allowing for the calculation of relevant long-term effects based on historical data (see appendix).

The study reveals the following results for the year 2017:

  • A coefficient of –0.62 was calculated for the ex­change rate responsiveness of German exports of goods and services to the UK. On this basis, a depreciation of the pound by 10 per cent against the euro is accompanied by a reduction in German exports to the UK by some 6 per cent (all other things being equal).
  • The effect of the depreciation of the pound on Germany’s imports from the UK is smaller, but still negative. The coefficient lies at –0.35, implying that a 10 per cent depreciation of the British pound is accompanied by a decline in Germany’s imports by 3.5 per cent. This negative effect contradicts the prevalent textbook hypothesis that a currency appreciation tends to have a positive impact on an economy’s imports. In Germany’s case, this can be explained by the strong link to the UK in terms of intermediate input and the high share of imported intermediate inputs used in German exports.
  • For the responsiveness of German exports to the UK in relation to the British GDP, the coefficient lies at 1.78. Thus, with all other things being equal, a decline in the UK’s GDP by 1.5 percentage points is associated with a decline in German exports to the UK by around 2.7 per cent.

The estimated decline of German exports imported into the UK in 2017, resulting from the supposed impact of the Brexit debate on the UK, should cause Germany’s economic output in the coming year to decrease by approximately one-fourth of a percentage point.

Besides this main scenario, two alternative scenarios were examined, showing the differing expectations connected with either a “soft” or a “hard” Brexit (see chart). While the UK’s departure from the EU will not take place until well after 2017, the British government and the EU circles could either take a soft line (willing to compromise) or a hard line (highly conflictual) in their negotiations – in contrast to the current assessment in the main scenario depicted above. Thus, the assumptions were adjusted to account for variations in the decline of the GDP in the UK and the depreciation of the pound. Under these conditions, Germany’s economic output would decrease by around one-seventh of a per cent of GDP in the event of a soft Brexit, and by one-half of a per cent in the case of a hard Brexit (as defined in the table).

Since the model only covers the bilateral trade relationships between Germany and the UK – thus leaving out the indirect effects of the Brexit on other countries as well as the impact of uncertainty on investments and consumption in Germany – these estimates are more likely to reflect the minimum potential impact on the German economy.

Appendix

Within the framework of a time series analysis, an examination was conducted into how trade between Germany and the UK reacts to changes in the real exchange rate (RER) and to economic activity, measured by GDP. As the explanatory variable, the bilateral trade flows for goods and services were taken from the Deutsche Bundesbank’s balance of payments statistics and then converted into real values with the help of the overall price deflator for exports and imports. In doing so, the assumption was made that the price development in German trade with the UK roughly corresponds to the price development in Germany’s overall trade of goods and services. The annual price changes are relatively insignificant, making it seem reasonable to apply the calculated coefficients – at least for a limited time period – to the development of nominal exports from the balance of payments statistics. Seasonally adjusted real GDP data for the UK and in Germany, and the real exchange rate between the pound and the euro (on a consumer price basis), were used as explanatory variables for the bilateral trade flows. All data used came from Eurostat.

In an initial step, the stationarity of the time series was examined. The augmented Dickey-Fuller test (ADF, see Dickey/Fuller, 1975) showed that both the trade flows and the time series for the GDP and the RER are integrated of order one. The Johansen test was used to check for the presence of a long-term relationship in terms of a cointegration of time series. The lag structure was determined by applying the Akaike information criterion (AIC) and the Hannan-Quinn information criterion (HQC). Corresponding to these criteria, a lag length of two was used for the further analysis. The trace statistic confirmed the presence of a cointegration relationship among the applied variables. The econometric estimate shows that all five variables used have some influence on this cointegration relationship, with coefficients that are significantly different from zero. The exact estimated results will be provided upon request.

The analysis of impulse response functions shows that a shock coming from the exchange rate or from GDP of the UK would have a lasting impact on Germany’s exports and that this long-term effect would already set in after two or three quarters. These results can thus be used for an estimate of the Brexit effects on German trade with the UK and on the German GDP for the year 2017. An OLS estimate based on (stationary) rates of change of nominal exports, imports and real GDP reaches similar results with regard to the estimated coefficients and to the impact on Germany’s economic output.