Unconventional Monetary Policy. The hypotheses presented will then be investigated one by one. For each of the hypotheses we will make plots of relevant data, and based on the empirical findings, we will try to draw related conclusions. These conclusions will be followed by a discussion where we put emphasis on potential research errors and market factors that could influence the results before we move on to the next hypothesis. After all the hypotheses related to one focus area are investigated, we will round of the section with a short summary and sub-topic conclusion, before we move on to the next area of focus.
To try to isolate the effects of the NIRP, we will for each of the indicator analyses look for common responses/tendencies for the four NIRP economies. Further, we will compare the indicator developments of the NIRP economies to the corresponding evolution of the indicators in a control, non-NIRP, economy. We have chosen to use the United Kingdom as our control economy, as it has faced similar economic challenges as the four NIRP
economies, but the Bank of England has never implemented negative key policy interest rates. The effect of NIRP will be highlighted if the responses of our four NIRP economies deviate from the development in the control economy. The isolation of NIRP effects is a huge research challenge as the economic indicators are influenced by a series of factors.
However, by looking at general common tendencies, followed by a deviation-comparison with a non-NIRP economy, we could potentially uncover general effects caused by NIRP.
4.4 Data collection
We have used a range of sources to collect the data needed to perform the analysis. For the collection of monetary policy related data, we have primarily been using the four economies’
central banks’ databases. The other data has been assembled through various databases including OECD Data, OECD Stat, Bloomberg, and Datastream. For a complete list of databases used, please see the following section. As all data included in the analysis was already gathered and available in various databases, we did not have to collect data ourselves.
There exists a risk related to the use of secondary sources in research papers, such as this thesis, as we do not know how the data has been collected and handled. However, the databases we have used are all large and well-known, and we, therefore, rely on the data
retrieved from these sources. The specific reliability considerations are also provided in the following section.
4.5 Data sources
Presented below is an exhaustive list of databases used for collection of indicator time-series, including a short description of the databases and an evaluation of the data source.
OECD’s databases; OECD Stat and OECD Data: These are OECD’s online platforms for statistics and data research, created by the OECD Statistics Directorate. All data is available to the public. OECD is the Organisation for Economic Co-operation and Development, an intergovernmental organisation that provides a platform for exchanging policy experience and cooperating to find sustainable solutions to common problems, and is considered a highly reliable source.
Datastream: The database is provided by Thomson Reuters, and contains current and historical financial and economic data. The data is both collected from Thomson Reuters themselves and gathered from other external sources. The external sources are all well-known institutes such as official statistics bureaus and central banks, and we thereby evaluate the data collected to be highly reliable.
Bloomberg: The database provides current and historical news, data, analyses, and videos related to economics, business and financial information. The company also provides financial software tools through their Bloomberg Terminals. The data is both collected from Bloomberg LT themselves, and through external sources. All external sources are well-known and controlled by Bloomberg, and the data is thereby highly reliable.
Trading Economics: The company provides an online database, with limited data available to the public, and full data access may be purchased. The database provides current and historic information about a range of financial variables and economic indicators, and the database is built using official sources, no third-party data is included. The database is a secondary source, but the data provided is checked regularly for inconsistencies and other faults, and hence, we evaluate the data collected as reliable.
Eurostat: The European Union’s statistical database collects and provides statistics for European countries, both for members and non-members of the European Union. The database is built up of data collected and processed by the data provider. Eurostat is the database of the European Union’s statistical bureau and is therefore considered highly reliable.
IMF data: IMF works to ensure stability of the international monetary system, and as a part of the work towards this objective, they publish time series data like exchange rates, IMF lending, GDP and other indicators for the IMF member countries. The IMF database is considered a solid database providing reliable data.
Central banks' databases: The European Central Bank, Sveriges Riksbank, Danmarks Nationalbank and Swiss National Bank provide direct current and historic data on the key interest rates, and other monetary policy data, for their respective countries. The central banks are a primary source and are considered highly trustworthy.
4.6 Data types
We have based our analysis on observation of a range of different macroeconomic variables related to interest rates, saving and borrowing, investment, production, output, business cycles, bankruptcies, and labour. These groups of variables are the most crucial to answer the two sets of problem statement sub-question sets, as they can reveal tendencies related to the economy. By looking at several variables at the same time and comparing NIRP
tendencies to the trends in the control economy, we are, to a larger degree, able to evaluate the NIRP effects. In turn, we should be able to confirm or reject our hypotheses.