Global financial cycles since 1880

Galina Potjagailo and Maik H. Wolters

Global financial cycles: a long-term affair

Today’s financial system is global: credit and several financial asset classes show booms and busts across countries, sometimes with severe repercussions to the global economy. Yet it is debated to what extent common dynamics rather than domestic cycles lie behind financial fluctuations and whether the impact of global drivers is growing. In a recent Staff Working Paper, we observe various global financial cycles going as far back as the 19th century. We find that a volatile global equity price cycle is nowadays the main driver of stock prices across advanced economies. Global cycles in credit and house prices have become larger and longer over the last 30 years, having gained relevance in economies that are more financially open and developed.

The idea of the “global financial cycle”, initially introduced as the co-movement in capital flows, asset prices and in credit growth, has spurred much interest in recent years. Empirical evidence looking at data from recent decades suggests an important role of a common global component in risky asset prices, composite global financial indices or global credit measures in driving domestic financial conditions. But is this really a new or increasingly important phenomenon? Are global forces mainly reflected in risky assets, or do they drive a wider range of financial sectors? There are also opposing views that question the relevance of the global financial cycle or find little evidence for its increasing role over time. The mixture of findings partially reflects that the concept of the global financial cycles itself remains somewhat blurry. Does the global financial cycle primarily reflect co-movement in risky asset prices or a wider range of financial aggregates? And how does the global financial cycle compare to domestic financial cycles, i.e. the prolonged ups and downs in credit observed in many individual countries that constitute an important criterion for macroprudential policy measures (as discussed here recently)? Another caveat is that the rather short series of data typically available to macroeconomists can only tell us so much about the behaviour of potentially extremely long-lasting ups and downs in financial markets.

Historical perspective meets flexible model

In this post, we provide a new detailed perspective of global financial cycles. We measure the joint fluctuations in a number of financial series across advanced economies, including credit and house prices, equity prices and long-term interest rates over a period of 130 years. Using data from the Macrohistory database, we go back as far as the 19th century when in a first era financial globalisation credit and capital flows started to expand markedly for the first time around. We compare historically relevant ups and downs in global finance such as the periods of the Great Depression in the 1930s and the Great Recession following the 2008 financial crisis. Importantly, we feed the data into a dynamic factor model that measures the common components behind financial series across countries in a flexible way (see here and here for overviews on dynamic factor models). Our model allows for various global components behind our financial series, in order to see whether the global financial cycle is best represented by a composite index across financial variables, or instead various sector-specific indices (e.g. credit or equity cycles), or both. Finally, the size of global forces and the response of individual countries to these forces can evolve over time in our model – this is crucial to capture the long historical period in a meaningful way, and it will show us how the role of global financial cycles changed over time.

Global financial co-movement occurs at different cycle lengths

We observe various global financial cycles since the end of the 19th century, presented in Figure 1. An aggregate financial cycle captures co-movement across countries between credit, asset prices and interest rates jointly. This cycle displays both high and medium frequency fluctuations. Once this aggregate co-movement is controlled for, there are also global cycles within individual sectors: an equity price cycle, a credit cycle and a house price cycle. Equity price cycles are highly volatile throughout the 130 years that we analyse. Credit and house price cycles appeared as short-run fluctuations for a long time, but over the last 30 to 40 years their characteristics changed substantially. The global credit cycle and global house price cycle have become much larger in amplitude and more long-lasting, with each boom-bust period taking about 15 years.

Figure 1: Global factors

The global ups and downs in financial variables trace historic events well, such as for instance the bank runs at the beginning of the 20th century, the relatively low financial volatility during the post-War Bretton-Woods years, or the stock market crashes in the 1970s and during the 2001 Dot-Com bubble. Both the Great Depression during the 1930s and the Great Recession after the 2008 financial crisis are reflected in large busts in various of our financial cycles simultaneously, but despite stronger asset price responses the bust in global GDP remained less severe during the Great Recession. Figure 2 shows that, over the recent decades, the aggregate financial cycle compares very closely to the global component in over 900 risky asset prices found in previous research (denoted MAR-factor on the chart). At the same time, our global credit cycle is quite similar to filtered domestic credit-to-GDP measures for various countries that the Bank for International Settlements uses for assessments of macroprudential policy tools, as we show in the paper.

Figure 2: Global financial factor compared to annualised Miranda-Agrippino and Rey (2015) factor

Country characteristics matter for the susceptibility to global cycles

For equity prices, the role of global factors increased strongly over the historical time span for all countries in our analysis. In most economies, global dynamics now constitute the main driver of equity price fluctuations. A high degree of liquidity, fast-moving information and little institutional differences in stock markets seem to make equity return dynamics a truly global phenomenon. On the other hand, the relevance of global cycles for credit and housing is smaller and has not changed over time on average across countries, which seems somewhat surprising given the change in shape we observe for those cycles. However, taking averages masks substantial differences across countries. Figure 3 shows that the role of global forces for credit and housing did increase quite substantially for a sub-group of countries that tend to be more financially open and developed economies including the two largest financial players in our sample period, the United States and United Kingdom. In the paper, we investigate this more formally within a panel regression setting and find that a country’s susceptibility to the global financial cycles tends to increase with financial openness and financial integration, the extent of mortgage-related lending, and the efficiency of stock markets.

Figure 3: Variance explained by global factors

Don’t forget the “global” in credit and housing cycles

While the view of the financial cycle in terms of risky asset prices has been related to the debate on the role of monetary policy in large economies and the propagation of financial conditions across countries, the view of financial cycles in terms of credit and house prices has mostly focused on domestic financial conditions. Our findings suggest that there can be a strong global dimension to credit and house price dynamics as well, with potential gains from internationally coordinating macroprudential policy measures.

Our analysis captures two sides of global financial co-movement and suggests that both are ultimately intertwined: global cycles in equity returns and global cycles across domestic credit markets. These cycles occur over different cycle lengths, including lengths that coincide with business cycle frequencies, and can each be relevant for financial stability in their own right. But they also become jointly relevant since there is joint co-movement across different parts of the financial system. In fact, the impact of global forces can amplify in times when global equity price booms coincide with global credit and house price booms. As such, monitoring global equity markets and credit markets in parallel might help detecting and understanding the build-ups of global financial instabilities early on. Country characteristics related to the degrees of financial openness and financial development can serve as an important criterion to evaluate the need of financial stabilization policies and the role for coordination of policies across countries and sectors.

Galina Potjagailo works in the Bank’s Current Economic Conditions Division and Maik H. Wolters works in the University of Wuerzburg, Kiel Institute, and IMFS at Goethe University Frankfurt .

If you want to get in touch, please email us at or leave a comment below.

Comments will only appear once approved by a moderator, and are only published where a full name is supplied. Bank Underground is a blog for Bank of England staff to share views that challenge – or support – prevailing policy orthodoxies. The views expressed here are those of the authors, and are not necessarily those of the Bank of England, or its policy committees.

One thought on “Global financial cycles since 1880

  1. Very Good Article. Agree with cycle analysis, but suggest you actually go back another 100 years and add to your analysis, key events at top and bottom of cycles, you will find some interesting items. Also, you should add to your cycle analysis both social mood, positive/negative, and pandemic outbreaks, which would be revealing.

    Going forward you can see where we are at in the economic/social mood cycles and where we are headed.

    We use long term cycles in our daily analysis.

Comments are closed.