The impact of political announcements on expectations concerning the starting date of the EMU : a microeconometric approach to the detection of event-dependent answering patterns in business surveys

This paper examines German business survey data to uncover the influence of political news on expectations concerning the starting date of the European Monetary Union (EMU). In this survey the participants were asked to indicate whether they expect a punctual or a delayed start of the EMU. Alternatively, they could also tick a don't know category. It is shown that political news actually influence answering patterns. While good news have a negative impact on the probability to cross don't know, they do not influence the probability to expect a delayed EMU start. Bad news, such as the heated debate about the convergence criteria at the Amsterdam EU summit, have a positive impact on the probability to expect a delayed start. The empirical results show that survey results generally have to be carefully interpreted if news about the topic which is investigated in the survey occur. If answering patterns are actually influenced by such news, the usual way to interprete survey responses - treating them as if they were sent back at the very same day - can lead to severe misunderstandings. In this paper a simple microeconometric technique is suggested which makes it possible to detect if survey responses are affected by news.