Dating Slovenia’s business cycle
Part 3 moves from forecasting GDP to identifying Slovenia’s business cycle. The business cycle is the movement of economic activity around its longer-term trend. This is not the same as asking whether GDP increased or decreased in a particular quarter. An economy can grow and still be below trend, or slow down while remaining above trend. Filter-based cycle dating therefore gives a more nuanced picture of macroeconomic phases.
This analysis compares three filters: Hodrick-Prescott, Baxter-King and Corbae-Ouliaris. Each separates trend and cycle differently. The aim is not to claim that one filter gives the true cycle, but to identify the episodes that are robust across methods.
The Hodrick-Prescott filter classifies Slovenia as being in expansion for 70 quarters and contraction for 51 quarters. The early part of the sample contains alternating phases, but the first large negative phase runs from 1999Q3 to 2003Q2. A stronger positive phase appears from 2005Q2 to 2008Q2, before the global financial crisis produces a contraction from 2008Q3 to 2009Q2. A short recovery follows in 2009Q3–2011Q1, but the economy then returns below trend from 2011Q2 to 2013Q2. The later expansion from 2013Q3 to 2016Q2 marks a more sustained recovery, followed by another above-trend period from 2017Q1 to 2019Q3. The COVID-19 contraction is dated from 2019Q4 to 2020Q2, followed by recovery from 2020Q3 to 2021Q4. The recent years are more mixed: the HP filter places Slovenia below trend from 2023Q3 to 2025Q1, then above trend from 2025Q2 to 2026Q1.
The Baxter-King filter gives a smoother but shorter view because it loses observations at the beginning and end of the sample. It identifies 49 expansion quarters and 48 contraction quarters. It confirms the long negative phase from 1999Q4 to 2003Q3, the pre-crisis expansion from 2005Q4 to 2008Q1, the global-financial-crisis contraction from 2008Q2 to 2009Q3, the recovery from 2009Q4 to 2011Q3, and the renewed weakness from 2011Q4 to 2013Q1. It also captures expansion from 2013Q2 to 2015Q3, later shorter phases, the pandemic contraction from 2019Q3 to 2020Q3 and the recovery from 2020Q4 to 2021Q4. Because of endpoint loss, it cannot fully evaluate the latest years.
The Corbae-Ouliaris filter records 62 expansion quarters and 59 contraction quarters, giving a balanced view of the cycle. It identifies a long below-trend phase from 1999Q3 to 2005Q3, expansion from 2005Q4 to 2008Q2, contraction from 2008Q3 to 2009Q3, expansion from 2009Q4 to 2011Q3, and contraction from 2011Q4 to 2013Q1. It then shows a long expansion from 2013Q2 to 2017Q4, a short contraction in 2018Q1–2018Q2, a renewed expansion from 2018Q3 to 2019Q2, and the pandemic contraction from 2019Q3 to 2020Q3. In the latest years it alternates more frequently: expansion in 2023, contraction in 2024, expansion in 2025Q1–2025Q3, and contraction again in 2025Q4–2026Q1.
The three filters agree on the main episodes. The global financial crisis and the subsequent European adjustment period are central to Slovenia’s cycle. All three filters identify a pre-2008 above-trend phase, a crisis contraction, renewed weakness in the early 2010s, a stronger recovery after 2013, the COVID-19 contraction and the post-pandemic rebound. The main uncertainty concerns the endpoint. HP is more favourable at the very end, while Corbae-Ouliaris turns negative again in 2025Q4–2026Q1. Baxter-King cannot resolve this because of endpoint loss.
Table 4. Business cycles dating for Slovenia time series
For interpretation, the HP filter is the most convenient main reference because it covers the full sample and is easy to communicate. However, it should not be used mechanically, especially near the end of the sample. Baxter-King and Corbae-Ouliaris are valuable robustness checks. The safest conclusion is that Slovenia has moved from the post-pandemic rebound into a more mature and less clearly defined cyclical phase. The level of GDP is high, but the precise position relative to trend is method-dependent in the latest quarters.
Methodological appendix to Part 3
Business-cycle dating starts by separating GDP into a trend component and a cyclical component. If the cyclical component is above zero, GDP is interpreted as above trend. If it is below zero, GDP is interpreted as below trend. These positive and negative phases are then described as expansions and contractions.
The Hodrick-Prescott filter estimates a smooth trend and treats deviations from that trend as the cycle. The Baxter-King filter is a band-pass filter designed to isolate movements at business-cycle frequencies. The Corbae-Ouliaris filter is another method for extracting cycles from non-stationary data. No filter observes the true cycle directly. Each imposes assumptions. Robust interpretation should therefore focus on episodes that appear across methods, while treating endpoint turning points with caution.