The path to a monetary policy decision

Monetary policy in the coronavirus crisis

To the report's start page
Monetary policy in the coronavirus crisis

Increased real-time data collection and new methods

Published: 7 April 2022

Statistics and various indicators that the Riksbank normally relies upon to assess economic developments are published once a month or once a quarter, and are subject to a certain time lag. The Swedish National Accounts are one example of statistics that are published with a relatively long time lag. Publication is usually around 60 days after the end of the quarter, which means that the first ordinary GDP outcome for the first quarter is published at the end of May. The Riksbank therefore uses more up-to-date indicator information, when available, in its forecasting models to gain an idea of how, for instance, GDP growth looks at present.[24] See Andersson and den Reijer (2015) for a description of how so-called nowcasting models use information from indicators that are available prior to the outcome of the forecast variable. When the economy ended up in a free fall in March 2020, much of this data was too backward-looking and failed to provide sufficient support for making economic policy decisions. There was a substantial need for real-time data, to be able to assess the current situation and developments in the near future. To gain an oversight of how the pandemic affected the Swedish economy, the Riksbank quickly expanded its collection and compilation of high-frequency data, known real-time indicators.[25] See Ewertzh, Falk, Hesselman, Hull, Löf, Stigland and Tibblin (2020). Data was collected from other authorities, from various Internet websites and in collaboration with various commercial data suppliers that have previously collected and compiled real-time indicators. It was also crucial, when making decisions, to be able to analyse developments in the financial markets, and new types of detailed financial data became important sources of information, such as securities borrowing data and lending.[26] An example of analysis using more detailed data is Frohm, Grip, Hansson and Wollert (2020).

The Riksbank normally carries out company interviews three times per year to obtain in-depth information on how the companies view the current economic situation and how they intend to act on the basis of the present conditions.[27] See The Riksbank’s Business Survey. At the beginning of the crisis, the Riksbank intensified its contacts with both companies and actors in financial markets to be able to better understand and follow developments. The Riksbank had ongoing discussions with banks and other market participants and switched to conducting telephone interviews every two weeks with a large number of bigger Swedish companies to capture their specific experiences of the pandemic. This information became an important complement to other real-time data collection. International cooperation and exchanges of experience with other central banks also took place more often in 2020.

The crisis resulting from the pandemic is a clear example of the fact that analytical methods that are largely based on historical relationships are not so relevant when dealing with a new type of economic shock. It has been very important to have access to high-frequency and industry-specific data when economic activity has been characterised by sharp shifts and required frequent monetary policy decisions. The Riksbank has been able to follow developments in the spread of infection, patterns of movements in the population, credit provision to companies, sector-breakdown data for consumption, notice of redundancies, bankruptcies and much more on a weekly or daily basis. This has enabled the Riksbank to have early information on trends and at the same time to analyse the covariation between different sectors, which has been important for assessing the current economic situation and the near future. High-frequency data also poses challenges. The underlying data is often less processed and more flawed, and because data is often missing for long periods of time, there are no proven practices to filter out seasonal patterns and other white noise. Moreover, in a small economy such as Sweden, economic activity fluctuates particularly much on a high frequency. The Riksbank's ability to collect and interpret data and to make an assessment of the situation and draw reasonable conclusions in a short time has therefore been important during the crisis.