Poland Policy Uncertainty Index

Download Data                    Download Annotated Index

We are pleased to host two new Economic Policy Uncertainty (EPU) indices for Poland, developed by Jedrzej Bialkowski, Rafal Klepka and Anna Slawik in Local and External Economic Policy Uncertainty and Their Impact on Exchange Rates in Emerging Markets: Evidence from Poland. They follow the index construction method in "Measuring Economic Policy Uncertainty" by Scott R. Baker, Nicholas Bloom and Steven J. Davis.

Their EPU Poland Short index is based on news articles published by five Polish newspapers over an eight-year period from July 2018 to December 2024. Their EPU Poland Long index uses articles from three Polish newspapers and covers the period from March 2003 to December 2024). Their selection of news sources aims to:

• Ensure significant national coverage,
• Cover diversified topic profiles, formats, and editorial lines, and
• Represent diversified ownership, to encompass a wide range of viewpoints

Accordingly, their indexes draw on one tabloid, "Fakt," which has the highest sales among daily Polish newspapers; the influential general newspapers "Gazeta Wyborcza," "Rzeczpospolita," and "Dziennik Gazeta Prawna"; and the leading daily economics-oriented newspaper in Poland, "Parkiet."

They obtain newspaper articles from Factiva. The start and end dates for individual newspapers are determined solely by data availability. Table 1 presents more information about each newspaper, and Table 2 presents their keyword lists. A newspaper article is classified as an EPU article if it contains at least one word from each of the three categories. Additionally, articles were automatically filtered using geographical restrictions within Factiva's interface, excluding regions that do not include Poland.

Raw EPU articles are scaled by the number of articles published in the same newspaper and month. Each newspaper's scaled frequency counts are then standardized to have a unit standard deviation from April 2003 to July 2019 for the Long index and from August 2018 to July 2020 for the Short index. The authors then compute an arithmetic average across newspapers by month of the scaled, standardized EPU counts. Finally, the series is multiplicatively normalized to have a mean of 100 from April 2003 to August 2019 for the Long index and from August 2018 to July 2020 for the Short index.