Business and financial cycles of major global economies

Perhaps the most important lesson that central banks and other regulators have learned from the global financial crisis is the need to ensure the stability of the financial system.^[1] The simultaneous peaks of the business and financial cycles have fully manifested the extent of the impact of financial stress on the real economy, i.e. GDP, consumption, investment and unemployment, highlighting the importance of macroprudential policy. This has led to the introduction of a number of new instruments to increase and ensure the resilience of the financial system. The crisis has also motivated central banks to incorporate the financial and banking sectors into their macroeconomic models. In the post-GFC world, business cycle volatility cannot be explained without taking the financial cycle into account. This article deals mainly with the practical aspects associated with the modelling of the business and financial cycles and the use of simple statistical filters. It also presents a comparison of the two cycles for the major global economies – USA, Japan, the UK, Germany and France.

Published in Global Economic Outlook – September 2023 (pdf, 1.6 MB)

---

Financial cycle 101

Monitoring the business cycle is not enough. While economics has recognised the cyclical nature of the real economy since the 1930s (due in part to the painful experience of the Great Depression), the term “financial cycle” has only emerged in the past few decades. Unlike the business cycle, which is commonly simplified as fluctuations in real GDP from its long-term trend (output gap), there is no single measure of the financial cycle. This is mainly due to the fact that financial systems differ across countries. The reIn the literature, the financial cycle and developments in the cycle are most often estimated using the following variables, variables derived from them or a combination thereof:[1]

• The volume of loans to the private non-financial sector – the gap in the ratio of total credit to real GDP (the Basel gap or the credit-to-GDP gap) is a commonly used measure of the financial cycle (see, for example, Minsky, 1982; Drehmann and Juselius, 2014). Other popular indicators include the credit gap, see, for example, Dell’Ariccia et al. (2012), and the debt service ratio, see, for example, Drehmann and Juselius (2012).
• Property prices – most often as the property price gap, for example, Drehmann and Juselius (2014), the price-to-income gap and the price-to-rent gap, for example, Cevik and Naik (2023).
• Equity prices – gaps in stock indices are a significantly less used indicator, mainly due to the amount of noise in time series (for example, Hatzius et al., 2010).

The financial cycle is typically two to four times longer than the business cycle and has a greater amplitude. This observation applies regardless of which of the above variables are used to define the financial cycle (Drehman et al., 2012). Therefore, if we consider the length of business cycle as 4–7 years,^[2] the financial cycle lasts around 8-28 years. In addition, fluctuations of the financial cycle from the long-term trend have been several times larger, especially since the mid-1980s. At that time, the cycles also started to diverge more. Both observations can be attributed to globalisation, be it of the financial or real parts of the economy, and to changes in monetary policy regimes (Borio, 2014). Chart 1 illustrates these stylised facts using the USA, the world’s largest economy, as an example, with its advanced financial sector and strong global impact. The chart shows that the various approaches to measuring the financial cycle (using different variables) yields similar information about its form and length. It also shows that historically the peak of financial cycle indicator is usually followed by a recession. The chart also indicates a peak last year. Analyst forecasts would also have made sense from this perspective. They had initially expected a recession in the USA in mid-2023 and are now expecting one in late 2023 and early 2024.

Historically, peaks in financial cycles have been reliably followed by recessions. A good example is the global financial crisis that was preceded, and ultimately intensified, by a decade-long global financial boom (see Chart 1). However, cyclical recessions are not necessarily preceded by financial stress. For example, the 1996–2001 dot.com bubble was accompanied by the same financial expansion in the United States and Europe for its entire duration (see Chart 1). Recessions or crises^[3] accompanied by a fall in the financial cycle are deeper and longer-lasting. On average, during these recessions, economies record declines in real GDP which are 50% higher than in other recessions (Drehman et al., 2012).  However, the same logic applies to the recovery phase, which is shorter if accompanied by a financial expansion.

Chart 1 – US business and financial cycle

(% deviation from trend)

Source: Federal Reserve Bank of St. Louis (FRED), author’s calculations
Note: The credit gap is the gap in the ratio of total credit to real GDP.

The global financial crisis consolidated the role of macroprudential policy as an independent discipline. This policy is tasked with safeguarding the stability of the financial system as a whole. In practice, it has a countercyclical effect on financial variables and thus prevents the build-up of systemic risk. In many countries, macroprudential policy is the responsibility of the central bank (for example, the UK, New Zealand, the Czech Republic). In others, it is that of regulators composed in part of central bank representatives (for example, the USA, the euro area, Japan, Norway). The objectives of macroprudential policy and the use of its instruments can influence monetary policy decision-making and indirectly affect the course of the business cycle and its implications for the economy. This relationship can also be seen in the opposite direction – monetary policies affect the financial cycle and the stability of the financial system. Interest rates directly and significantly affect credit growth, property prices and financial asset prices. Restrictive monetary policy is dampening growth in financial variables and is supporting expansionary growth (see Chart 2). In periods of sustained low monetary policy rates we can observe that banks, for example, lower their lending standards (Maddaloni and Peydró, 2011; Jimenéz et al., 2014). By contrast, the implementation of macroprudential policy instruments may have “spillover effects” on the output of the economy and the transmission of monetary policy at certain stages of the business and financial cycle. However, the effect depends on the specific instrument, its parameters and the situation in the banking sector and sectors of the real economy at the time of its application. For example, the countercyclical capital buffer (CCyB) and banks’ liquidity requirements can have a rather limited effect on economic output (especially in an environment of robust capitalisation, profitability and liquidity of the banking sector), and credit ratio limits (LTV, DTI and DSTI) can have a somewhat stronger effect on the business cycle (Nier and Kang, 2016; Richter et al., 2019). The current and expected position in the financial and business cycle at the central bank’s forecast horizon must thus be taken into account in the central bank’s financial stability decision-making.

Chart 2 – Total real loans to the private sector
(%)

Source: Federal Reserve Bank of St. Louis (FRED)

 

Modelling cycles in practice

How to best model cycles? Burns and Mitchell (1946) presented the first ever formal method of analysing the cyclical behaviour of time series – the break-point method. If we consider its application to the business cycle, the method designates the time between reaching the local minimum (maximum) and local maximum (minimum) of real GDP as the period of expansion (recession). However, the method does not tell us how much the economy deviates from the trend in periods of expansion or recession, making it difficult to use for the purposes of stabilisation policy. In cycle analyses, economics often looks at macroeconomic and financial variables as deviations from the long-term trend.

However, gaps in macroeconomic or financial variables based on theory are not inherently measurable or observable. This is due to their definition as the difference between an observable variable (e.g. GDP, credit volume) and its steady-state level (potential of GDP, credit trend) that is neither measurable nor observable. Univariate statistical filters are an effective solution which combines numerical simplicity and the ability to deliver relatively accurate results. The  Hodrick-Prescott filter (Hodrick and Prescott, 1997) is a widely-used filter in economics and is also used in this article for modelling cycles. In the spirit of economic theory, the Hodrick-Prescott filter assumes that time series can be viewed as the sum of the long-term trend and cyclical components. The trend is determined on the basis of historical values and the “smoothing parameter” λ. Higher values of λ lead to a smoother, even linear trend, while with lower values the trend is closer to the observed values. In the literature, λ = 1,600 is typically used for the business cycle and λ = 2,500–400,000 for a longer financial cycle. However, the simplicity of the method has several shortcomings. End-point bias or the deviation at the end of the series (i.e. the period that matters most) is probably the most serious of these. This is because, in the absence of future values, the filter indicates the position in the cycle unreliably. The use of the filter (at least without a reliable forecast of a given variable) is therefore problematic in real time, which is important for political decision-making. Other often-cited criticisms focus on the univariate nature of the filter (the trend is based on only one variable, unlike, for example, the Kalman filter, which allows trends of multiple variables to be monitored) and the arbitrary choice of the smoothing parameter λ, which is not based on the fundamentals of economic theory.

Economics naturally also offers more advanced approaches to capturing the business and financial cycles than statistical filters. A popular method for deriving the business cycle is to calculate potential output using the Cobb-Douglas production function explaining the output of the economy using inputs, specifically productivity, labour and capital (empirical comparison in Box 1) or using structural macroeconomic models. Similar methods can also be applied to estimate the financial cycle. For example, Seidler and Geršl (2012) estimate the steady-state level of the aforementioned credit-to-GDP variables for Central European post-Communist countries using elasticities obtained from a panel of advanced market economies. For most countries, the gap obtained in this way is diametrically different to that obtained using the HP filter – credit expansion does not necessarily mean excessive borrowing (and possible materialisation of risk) but rather convergence to the financial world in the West. Baxa and Žáček (2022) derive the financial cycle from a multidimensional structural model using time series for GDP, credit volumes, property prices and asset prices.

Cycles and central bank decision-making

The position of the economy in the cycle is reflected in the policy-making of some central banks. One such example is the US Fed, whose objectives also include maintaining high employment.^[4] However, in most market economies, price stability tends to be the primary objective for monetary policy decision-makers, and other goals (e.g. stabilising output, i.e. the countercyclical role of monetary policy) are only of secondary importance. The shift in perspective in this direction has been accelerated by the flattening Phillips curve, suggesting that the link between the positive output gap and inflation has weakened over time (e.g. Kuttner and Robinson, 2010; Akerlof et al., 2014). In addition, contemporary macroeconomic theory argues that stabilising output goes hand in hand with price stability – the “divine coincidence” (Blanchard and Galí, 2007). The inflation outlooks at the monetary policy horizon are therefore the primary guide for setting interest rates. However, there are also exceptions. For example, the interest rate cuts introduced with the onset of the Covid-19 pandemic were partly motivated by efforts to prop up a hampered real economy where economic agents really had no idea what such a shock could generate. By contrast, macroprudential policy uses a number of instruments pursuing various intermediate macroprudential policy objectives^[5] but has one primary objective, and that is to achieve overall financial stability. Despite the single institutional framework due to membership of the Basel Committee on Banking Supervision, approaches to the application of specific instruments often differ across countries. This mainly holds true of the countercyclical capital buffer (CCyB) designed to influence the financial cycle. From the time perspective, it is set asymmetrically (i.e. it is increased with a sufficiently long lead time and decreased or released immediately). The aim is to increase the banking sector’s resilience to credit risk materialisation in the recessionary phase of the financial cycle. In practice, during periods of financial expansion exceeding the long-term trend (a positive financial cycle gap), banks are required to create and maintain a capital buffer which would not only be able to sufficiently cover losses from a future financial downturn, but also support smooth lending.

The Basel Committee on Banking Supervision (2010) recommends using the credit-to-GDP gap as a common reference indicator of the position in the cycle for setting the CCyB rate.^[6] Also known as the Basel gap, the indicator is modelled for these purposes using a one-sided HP filter with a smoothing parameter λ = 400,000. The recommended CCyB rate rises as the percentage deviation from the trend rises. The problems associated with this practice are highlighted, for example, by Edge and Maisenzahl (2011). The rest of the article elaborates on the use of the countercyclical capital buffer in the regionally important countries that apply it. More specifically, it offers a view of the extent to which the changes in the CCyB rate correspond to the estimate of the Basel gap constructed according to the common methodology.

The case of Germany

BaFin, which falls under the Federal Ministry of Finance, is responsible for the countercyclical capital buffer in Germany. At the end of June 2019, BaFin announced the introduction and implementation of the CCyB at a rate of 0.25% with effect from 1 July 2020. The decision to introduce the instrument was motivated by a positive Basel gap (BaFin, 2019), which had turned positive for the first time since 2004 in late 2019 H1 (see Chart 3). However, the buffer has not been activated yet, as the decision to introduce the CCyB rate was abolished at the end of March 2020 due to concerns associated with the real and financial developments after the Covid-19 pandemic. This decision was made by BaFin partly because German banks had expected the level of capital to increase by the CCyB rate since the June announcement and were thus prepared for hypothetical losses (BaFin, 2020). However, the gap widened markedly during the pandemic (due to a fall in GDP and simultaneous credit growth) and, along with a record-high drop in property prices (see Chart 3), became one of the reasons for the introduction of a CCyB rate of 0.75% with effect from February 2023 (BaFin, 2022).

Chart 3 – Estimate of Germany’s current position
(deviation from trend in %; right-hand scale in %)

Source: Eurostat, ESRB, author’s calculations
Note: CCyB – countercyclical capital buffer.

The case of France

The HCSF (Haut Conseil de stabilité financière) is responsible for setting the CCyB rate in France. Its members include the Minister of Finance and the Governor of the Banque de France. The implementation of the countercyclical capital buffer has been very similar to that in Germany. Owing to a positive Basel gap and its outlook, a CCyB rate of 0.25% entered into force in July 2019. This was accompanied by the announcement of a doubling of the rate with effect from April 2020. However, the buffer was released in 2020 Q1 due to the pandemic. The latest HCSF decisions are interesting, though – a CCyB rate of 0.5% applicable since April 2023 and a 1% rate to come into effect in January 2024. The Basel gap is negative in both cases, something not mentioned by the Council in its press releases. The decision was motivated primarily by strong credit growth in the second half of 2022 and total private sector debt  (HCSF, 2022).

Chart 4 – Estimate of France’s current position
(deviation from trend in %; right-hand scale in %)

Source: Eurostat, ESRB, author’s calculations
Note: CCyB – countercyclical capital buffer.

The case of the United Kingdom

In the UK, the countercyclical capital buffer rate is set by the FPC (Financial Policy Comittee), made up of representatives of the Bank of England and external experts. The UK regulator does not use the Basel gap to set the rate (FPC, 2016),^[7] as illustrated in Chart 5. The initial introduction of the CCyB rate, announced in July 2016, was cancelled due to Brexit and related concerns about financial stress. As in Germany and France, a buffer rate of 1% was later introduced, and was released during the Covid-19 pandemic. In the current situation, when the Basel gap is negative, a relatively high rate of 2% applies to British banks compared to neighbouring countries. BoE approaches the setting of CCyB pragmatically and takes into account also the results of stress tests.^[8]

Chart 5 – Estimate of UK’s current position
(deviation from trend in %; right-hand scale in %)

Source: Eurostat, ESRB, author’s calculations
Note: CCyB – countercyclical capital buffer.

The case of Japan

Unlike in the advanced economies of the global West, in Japan’s case, there are no clear trends in financial variables. In Japan, for example, the evolution of real property prices is unique, as they have not reached the level observed at the start of the “lost decade” in more than 30 years. With a bit of exaggeration, we can say that the theoretical financial cycle is longer than the time series themselves. Despite this, a financial gap indicator pointing to a positive position like the credit-to-GDP gap (see Chart 6) can be found in the Financial System Reports published by the Bank of Japan (BoJ, 2023).

Chart 6 – Estimate of Japan’s current position
(deviation from trend in %; right-hand scale in %)

Source: Federal Reserve Bank of St. Louis, author’s calculations
Note: CCyB – countercyclical capital buffer, Japan has not applied the CCyB rate yet.

Conclusion

The global financial crisis has brought about a definitive paradigm shift in the perception of the financial sector. Financial variables displaying cyclical behaviour have a marked effect on the real economy. The most important lesson still is that recessions accompanied by financial failures are deeper and longer-lasting. Macroprudential policy, which reduces the build-up of financial stress, was incorporated into the toolkit of central banks (or institutions established for this purpose) also to ensure that countries eliminate as much as possible any losses arising from the suboptimal functioning of the financial system.

The Hodrick-Prescott filter is used as a simple and relatively informative tool for estimating the long-term trend of economic variables. End-point bias and misleading real-time estimates remain a problem. As recognised by the Basel Committee, national regulators rely more on other indicators or more comprehensive approaches when setting the countercyclical capital buffer (see also e.g. the CNB) rather than the standardised representation of the Basel gap (Basel Committee on Banking Supervision, 2022) to enable the CCyB to fulfil its role in mitigating the impacts of the adverse phase of the financial cycle more effectively. Of the three European countries under review applying the CCyB rate, Germany is the only country whose rate is qualitatively consistent with the Basel gap as originally defined. However, this can be expected. As the European Systemic Risk Board (ESRB) points out, financial cycles are relatively different across countries, so the common reference indicator may not always meet financial stability needs (ESRB, 2014; ESRB, 2022).

The Hodrick-Prescott filter is unreliable for evaluating the current state of the financial cycle in selected countries or its outlook. In all three cases of European economies, the effective CCyB rate is currently at historical highs despite low filter-derived values of the financial gap. Systemic risk and the potential losses for which regulators are preparing the banking systems in their jurisdictions cannot be perceived through a univariate filter.

References

Akerlof, G. A., Blanchard, O., Romer, D., & Stiglitz, J. E. (Eds.). (2014). What have we learned?: Macroeconomic policy after the crisis. MIT Press.

Basel Committee on Banking Supervision (2010). Guidance for national authorities operating the countercyclical capital buffer. Basel, Switzerland.

Basel Committee on Banking Supervision (2022). Buffer usability and cyclicality in the Basel framework. Basel, Switzerland.

BaFin (2019, September 4). General Administrative Act governing the rate for the domestic countercyclical capital buffer under section 10d of the KWG. Bafin.de[online]. Retrieved 1 July 2023 from https://www.bafin.de/dok/12946932

BaFin (2020, April 23). General Administrative Act regarding a decrease in the domestic countercyclical capital buffer rate. Bafin.de[online]. Retrieved 1 July 2023 from https://www.bafin.de/dok/13991372

BaFin (2022, February 16). General Administrative Act regarding a decrease in the domestic countercyclical capital buffer rate. Bafin.de[online]. Retrieved 1 July 2023 from https://www.bafin.de/dok/17394924

Baxa, J., & Žáček, J. (2022). Monetary policy and the financial cycle: International evidence. Czech National Bank, Economic Research Department.

Blanchard, O., & Galí, J. (2007). Real wage rigidities and the New Keynesian model. Journal of money, credit and banking, 39, 35–65.

BoJ (2023). Financial System Report. boj.or.jp[online]. Retrieved 5 July 2023 from https://www.boj.or.jp/en/research/brp/fsr/data/fsr230421a.pdf

Borio, C. (2014). The financial cycle and macroeconomics: What have we learnt? Journal of Banking & Finance, 45, 182–198.

Burns, A. F., & Mitchell, W. C. (1946). Measuring business cycles. National Bureau of Economic Research.

Cevik, M. S., & Naik, S. (2023). Bubble Detective: City-Level Analysis of House Price Cycles. International Monetary Fund.

Dell’Ariccia, G., Igan, D., Laeven, L., Tong, H., Bakker, B., & Vandenbussche, J. (2012). Policies for macrofinancial stability: How to deal with credit booms. IMF Staff Discussion Note, 12(06).

Drehmann, M., Borio, C. E., & Tsatsaronis, K. (2012). Characterising the financial cycle: don’t lose sight of the medium term!

Drehmann, M., & Juselius, M. (2012). Do debt service costs affect macroeconomic and financial stability? BIS Quarterly Review September.

Drehmann, M., & Juselius, M. (2014). Evaluating early warning indicators of banking crises: Satisfying policy requirements. International Journal of Forecasting, 30(3), 759–780.

Edge, R. M., & Meisenzahl, R. (2011). The unreliability of credit-to-GDP ratio gaps in real-time: Implications for countercyclical capital buffers. International Journal of Central Banking, 7(4), 261–298.

ESRB (2014). Recommendation of the European Systemic Risk Board on guidance for setting countercyclical buffer rates. Eur-Lex.Europa.eu. Retrieved 30 August 2023 from https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:32014Y0902(01)

ESRB (2022). Review of the EU Macroprudential Framework for the Banking Sector. Esrb.Europa.eu. Retrieved 30 August 2023 from https://www.esrb.europa.eu/pub/pdf/other/esrb.reviewmacropruframeworkcfa.220331~5d81cb2173.en.pdf?7263115b46a985b4481328afd3f2326dFPC (2016). The Financial Policy Committee’s approach to setting the countercyclical capital buffer. Bankofengland.co.uk[online]. Retrieved 5 July 2023 from https://www.bankofengland.co.uk/-/media/boe/files/statement/2016/the-financial-policy-committees-approach-to-setting-the-countercyclical-capital-buffer.pdf

Hamilton, J. D. (2018). Why you should never use the Hodrick-Prescott filter. Review of Economics and Statistics, 100(5), 831–843.

Hatzius, J., Hooper, P., Mishkin, F. S., Schoenholtz, K. L., & Watson, M. W. (2010). Financial conditions indexes: A fresh look after the financial crisis (No. w16150). National Bureau of Economic Research.

HCSF (2022, December 13). Press Release. Economie.gouv.fr[online]. Retrieved 1 July 2023 from https://www.economie.gouv.fr/files/files/directions_services/hcsf/HCSF_20221213_pressrelease.pdf?v=1671095905

Hodrick, R. J., & Prescott, E. C. (1997). Postwar US business cycles: an empirical investigation. Journal of Money, Credit and Banking, 1–16.

Jiménez, G., Ongena, S., Peydró, J. L., & Saurina, J. (2014). Hazardous times for monetary policy: What do twenty‐three million bank loans say about the effects of monetary policy on credit risk‐taking? Econometrica, 82(2), 463–505.

Kuttner, K., & Robinson, T. (2010). Understanding the flattening Phillips curve. The North American Journal of Economics and Finance, 21(2), 110–125.

Maddaloni, A., & Peydró, J. L. (2011). Bank risk-taking, securitization, supervision, and low interest rates: Evidence from the euro area and the US lending standards. The Review of Financial Studies, 24(6), 2121–2165.

Minsky, H. P. (1982). Can it happen again? Essays on instability and finance, M. E. Sharpe, Armonk.

Nier, E. W., & Kang, H. (2016). Monetary and macroprudential policies – exploring interactions. BIS Paper, (86e).

Richter, B., Schularick, M., & Shim, I. (2019). The costs of macroprudential policy. Journal of International Economics, 118, 263–282.

Seidler, J., & Geršl, A. (2012). Excessive credit growth and countercyclical capital buffers in Basel III: an empirical evidence from Central and East European countries.

Yamada, H. (2012) A note on band-pass filters based on the Hodrick-Prescott filter and the OECD system of Composite Leading Indicators. OECD Journal: Journal of Business Cycle Measurement and Analysis, vol. 2011/2. https://doi.org/10.1787/jbcma-2011-5kg0pb01sbbt.

Keywords

Financial cycle, business cycle, Hodrick-Prescott filter, countercyclical capital buffer

JEL Classification

E32, E58, G28

---

^[1] The activities of the Bank for International Settlements at the turn of the millennium made a fundamental contribution in this area. At that time, special departments were beginning to emerge in a number of central banks, explicitly focusing on assessment of the risks to the financial system.

^[2] A fitting example is the CNB’s composite financial cycle indicator, which aggregates several time series, weighted by the ability to capture the banking sector’s future credit losses. See https://www.cnb.cz/en/financial-stability/thematic-articles-on-financial-stability/An-indicator-of-the-financial-cycle-in-the-Czech-economy/ 

^[3] There are a number of approaches to describing cycles in economics, relating mainly to their length and nature. One of the short-term cycles is the Kitchin cycle, lasting 18–40 months, which represents short-term fluctuations in real output caused by fluctuations in inventories. The causes are difficult to determine. The Juglar cycle is a medium-term cycle lasting 8–10 years associated with investment in fixed capital, alternating periods of increased wear and tear and increased investment. They may also involve alternating commodity generations, agricultural fluctuations, etc. Kuznets cycles (also Schumpeter, Kondratiev waves) are long-term cycles lasting 20–50 years. These can be explained by wars, scientific discoveries, major infrastructural investment, innovation waves, etc.

^[4] In a statistical sense, a recession is defined as a decline in GDP lasting at least two quarters. The situation can be described as a crisis if the fall in GDP lasts more than four quarters.

^[5] See https://www.federalreserve.gov/monetarypolicy/monetary-policy-what-are-its-goals-how-does-it-work.htm

^[6] See https://www.esrb.europa.eu/pub/pdf/recommendations/2013/ESRB_2013_1.en.pdf

^[7] The ESRB (European Systemic Risk Board), an authority supervising the EU financial system, draws attention to the differences in the appropriateness of the indicator across countries and recommends that national regulators take into account more relevant indicators. See https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:32014Y0902(01)

^[8] The FPC’s primary measure for setting the CCyB rate is the ability of domestic banks to absorb potential losses. For example, standards for the provision of loans and stress test results are taken into account.

^[9] See, for example https://www.bankofengland.co.uk/paper/2023/ps/the-financial-policy-committees-approach-to-setting-the-countercyclical-capital-buffer