On the Uncertainty Caused by the Referendum on Brexit

Abstract

In this paper we investigate “the” uncertainty caused by the 2016 Brexit referendum. In particular, we analyse whether the referendum in itself had a noticeable impact on expectations/behaviour of market participants. To investigate this, we analyse two survey-based indicators and a financial variable, namely the consumer confidence index, the economic policy uncertainty index and the GBP/Euro exchange rate. In the first step we estimate the law of motion of these variables using a state-space model in the time domain. In the second step, we transfer these results into the frequency domain. We find that certain indicators changed very soon after the referendum whilst other indicators reacted to the referendum by changing their medium and long-term behaviour. For those variables it is clear that the short-term reaction to any shock is fairly limited leading to the wrong conclusion that the referendum did not have any impact on them. In fact, the impact will only be seen much later than the original shock. In the opposite case, the wrong conclusion is that the reaction to the referendum is only visible in the short term, but not in the long-run. Therefore, we highlight that the dynamics caused by the referendum are of complex nature which may yet have to materialise. That implies that negative consequences of the referendum alone (never mind the actual Brexit) will only become visible well after the referendum by which time they may not be associated with the referendum anymore. However, we also show that there are short term consequences of the referendum.