Denmark Policy Uncertainty Index
Download Data                    Download Annotated Index
We are pleased to host two new EPU indices for Denmark developed by U. Michael Bergman and Camilla Hu Worm in "Economic Policy Uncertainty and Consumer Perceptions: The Danish Case."
Their indices reflect frequency counts of articles published in five Danish nation-wide newspapers (Berlingske, BT, Ekstrabladet, Politiken and Weekendavisen) from January 1991 until June 2021. They follow the index construction methods in "Measuring Economic Policy Uncertainty" by Scott R. Baker, Nicholas Bloom and Steven J. Davis.
Denmark is a small open economy and heavily exposed to the world economy and Danish newspapers include many articles reporting on economic issues not necessarily related to the economic political debate in the domestic economy. They therefore include Danish-specific words capturing geographic location (Denmark) to the list of keywords. In addition, they add relevant political terms reflecting the Danish context. Thus, they construct two indices, one that includes only the keywords suggested in the paper "Measuring Economic Policy Uncertainty" by Scott R. Baker, Nicholas Bloom and Steven J. Davis but with the addition of the geographic location; and one index where an extended list of policy-relevant keywords is used.
To construct the indices, Bergman and Worm first obtain a monthly count of newspaper articles that contain at least one keyword in each of four categories: (a) uncertainty (usikkerhed, usikker, usikkert, uro and urolighed), (b) economy (økonomi, økonomisk and økonomiske), (c) policy (Nationalbanken, centralbank, finansministeriet, regering, regulering, lovgivning, Folketinget, Christiansborg and underskud), and (d) geographic location (Danmark). After obtaining the raw EPU article counts, they scale by the number of articles published in the same newspaper and month. They then standardize each newspaper's scaled frequency counts to have a unit standard deviation from January 1991 to December 2009, and then compute the simple average across newspapers by month. Lastly, they normalize the resulting index to have a mean of 100 from January 1991 to December 2009. This first index is labelled EPU_BBD in the excel data file.
To construct their alternative index, they extend the policy-related terms with relevant political terms in a Danish context (finanspolitik, budgetunderskud, ledighed, arbejdsløshed, folketingsval, reform, skat and skattereform). This alternative index is labelled EPU_extended in the excel data file and is shown in their annotated chart.