Publication - The financial cycle and recession risk - BIS Quarterly Review - December 2018 - by Claudio Borio, Mathias Drehmann and Dora Xia
Financial cycle booms can end in crises and, even if they do not, they tend to weaken growth. Given their slow build-up, do they convey information about recession risk? We compare the predictive performance of different financial cycle proxies with that of the term spread - a popular recession indicator. In contrast to much of the literature, our analysis covers a large sample of advanced and emerging market economies. We find that, in general, financial cycle measures provide valuable information and tend to outperform the term spread.1
JEL classification: C33, E37, E44.
Once financial cycles peak, the real economy typically suffers. This is most evident around financial crises, which tend to follow exuberant credit and asset price growth, ie financial cycle booms. Crises in turn tend to usher in deep recessions, as falling asset prices, high debt burdens and balance sheet repair drag down growth.
This suggests that the financial cycle could be helpful for gauging recession risk, in particular as booms are drawn out and exhibit systematic patterns. Given the tight link between crises and recessions, this seems to be implied by the large body of work on the leading indicator properties of financial booms for banking crises (eg Borio and Drehmann (2009), Schularick and Taylor (2012) and Detken et al (2014)). Some studies have also documented that credit booms weaken output in the medium run (eg Mian et al (2017) and Lombardi et al (2017)). And some recent work has begun to study the impact of financial conditions on risks to growth (eg Adrian et al (2018)).2
But research exploring how financial expansions affect recession risk, ie the likelihood that a recession will develop in the near future - say, one to three years ahead - is scant and predominantly focused on the United States.3 Assessing recession risk has a long tradition. Probably the most popular variable in this context is the term spread (eg Estrella and Mishkin (1998) and Rudebusch and Williams (2009)). In particular, an inverted yield curve - long-term bond yields below short-term interest rates - is seen as among the best signals of impeding recessions, if not the best.
In this special feature, we examine the ability of financial cycle proxies to convey information about recession risk. We follow the literature very closely to better benchmark our analysis against the corresponding work on the term spread. In contrast to much of the extant analysis, we look at a large sample of advanced and emerging market economies (EMEs).4
For advanced economies, we find that financial cycle proxies provide valuable information for a horizon of up to three years, outperforming the term spread. The evidence for EMEs mirrors that for advanced economies, although data limitations prevent out-of-sample tests for this group.
The rest of the article is structured as follows. In the first section, we briefly introduce the notion of the financial cycle and document how the nature of the business cycle, and its link with the financial cycle, have changed in the past 50 years. In the second, we explain our methodology. In the third, we evaluate the performance of financial cycle proxies and compare it with that of the term spread based on full-sample information, ie ex post. In the fourth, we consider out-of-sample exercises, seeking to mimic the information policymakers have when assessing risks in real time, ie ex ante.
A look at the data
The term "financial cycle" refers to the self-reinforcing interactions between perceptions of value and risk, risk-taking, and financing constraints (Borio (2014)). Typically, rapid increases in credit drive up property and asset prices, which in turn increase collateral values and thus the amount of credit the private sector can obtain until, at some point, the process goes into reverse. This mutually reinforcing interaction between financing constraints and perceptions of value and risks has historically tended to cause serious macroeconomic dislocations.
The financial cycle can be approximated in different ways. Empirical research suggests that, especially if one is interested in episodes that have proven more damaging for economic activity, a promising strategy is to capture it through medium-term fluctuations in credit and property prices. This can be done either in terms of individual series or, preferably, of their combination. In this special feature, we rely on a "composite" financial cycle proxy similar to that in Drehmann et al (2012). In addition, as an alternative, we also look at the debt service ratio, defined as interest payments plus amortisation divided by GDP.5 Drehmann et al (2018) find a strong link between debt accumulation and subsequent debt service (ie interest payments plus amortisation), which in turn has a large negative effect on growth.6
Previous research has identified two important features of the financial cycle.7 First, financial cycle peaks tend to coincide with banking crises or considerable financial stress. This is not surprising. During expansions, the self-reinforcing interaction between financing constraints, asset prices and risk-taking can overstretch balance sheets, making them more fragile and sowing the seeds of the subsequent financial contraction. This, in turn, can drag down the economy and put further stress on the financial system.
Second, having grown in amplitude over the past 40 years or so,8 the financial cycle can be much longer than the business cycle. Business cycles as traditionally measured tend to last up to eight years, and financial cycles around 15 to 20 years since the early 1980s. The difference in length means that a financial cycle can span more than one business cycle. As a result, while financial cycle peaks tend to usher in recessions, not all recessions will be preceded by financial cycle peaks.
A first look at the relationship between the composite financial cycle proxy and recessions in the United States and the United Kingdom since the early 1980s illustrates these points (Graph 1). Financial cycle booms took place ahead of recessions in the early 1990s and the late 2000s. At the same time, the shallow recession in the early 2000s in the United States did not coincide with a financial cycle peak: while the economy slowed and equity prices tanked, the financial expansion continued as measured by credit and property prices, only to reverse a few years later, triggering the Great Recession. By contrast, in the United Kingdom, no recession took place in the early 2000s, so that the two recessions coincided with the two financial cycle peaks.
Why has the amplitude of financial cycles grown since the early 1980s, raising their importance for economic activity? The reasons are not yet fully understood, but arguably changes in policy regimes may be partly responsible.
Three such changes deserve particular attention. First, financial markets were liberalised starting around that time. Without sufficient prudential safeguards, this change likely allowed greater scope for the self-reinforcing interactions at the heart of the financial cycle to play out. Second, starting roughly at the same time, inflation-focused monetary regimes became the norm. And the evolving thinking of central banks led them to gradually downplay the role of monetary and credit aggregates. This meant that central banks had little reason to tighten policy if inflation remained low, even as financial imbalances built up. Finally, from the 1990s on, the entry of China and former Communist countries into the world economy, alongside the international integration of product markets and technological advances, boosted global supply and strengthened competitive pressures. Coupled with greater central bank credibility, this arguably made it more likely that inflationary pressures would remain muted even as expansions gathered pace. It also meant that financial booms could build up further and that a turn in the financial cycle, rather than rising inflation and the consequent monetary tightening, might trigger an economic downturn.9
Switzerland BISBCHBB
Website:www.bis.org
