In the run up to the global financial crisis, European banks significantly increased their lending activities both domestically and outside home markets driven by a procyclical spiral of cheap abundant funding, increasing profitability, and economic growth. European banks not only provided cross-border financing, but became increasingly involved in domestic financial markets via lending activities of their local affiliates.
In the process, European banks became excessively leveraged and reliant on wholesale short-term funding1 making them more susceptible to shocks, which could force them to adjust their operations abruptly and shrink their balance sheets with important ramifications for their overseas lending activities (Figures 1 and 2). Moreover, as banks expanded, the buildup of risks and their potential spillover effects to other parts of the world largely escaped supervisors.
This pre-crisis process of financial expansion and integration has been dramatically put in reverse ever since the US subprime crisis broke out and imbalances and risk underpricing became apparent. Faced with a new reality, policy makers responded by supporting financial markets and initiated an overhaul of supervisory mechanisms and the international regulatory framework. At the same time, banks reacted by boosting capital, slashing trading assets, reducing excessive lending, focusing on core deposits as a funding source, and realigning their business models.
While European banks have made progress and funding conditions are currently more benign, banks remain vulnerable. In October 2012, the IMFadjusted its baseline deleveraging estimates upward from $2.6 to $2.8 trillion by end-2013 (IMF 2012). And according to a recent exercise by the European Banking Authority (EBA 2013), European banks still have a substantial capital shortfall relative to after-tax profits assuming full implementation of the Basel III capital requirements as of 30 June 2012.
Figures 1 and 2.
What causes European banks to deleverage?
The main deleveraging drivers can be categorised into three groups: financial, regulatory, and economic and policy.
- Financial-market conditions: acute deleveraging drivers include tightening interbank and debt-issuance conditions compounded by tightening dollar-funding conditions, deposit outflows, and ongoing financial fragmentation of the European banking system.
Extraordinary central-bank measures have focused on repairing financial markets and averted an extreme funding scenario in 2011. However, markets have not returned to normalcy and remain vulnerable. For example, events such as those surrounding Cyprus could impede banks from pursuing stable funding strategies and raise borrowing costs.
- Regulation: a much-needed stricter international regulatory environment has triggered medium-term deleveraging which aims to put the financial sector on a strong footing.
On the international level, various new requirements were introduced including higher and better loss-absorbency, capital charges for market risk, and liquidity and funding requirements. On the Eurozone-specific level, the European Banking Authority recapitalisation exercise, state-aid programs, and ringfencing practices2 have triggered deleveraging responses.
- Economic and policy conditions: significant economic slowdown in the Eurozone, in part a result of front-loaded austerity programs, depresses bank-asset growth and profitability which undermines the capacity of banks to boost retained earnings;
The negative bank-sovereign-real economy feedback loop that became apparent during the European sovereign-debt crisis has also not yet been broken. European banks are still ‘international in life and national in death’ in the absence of a banking union with common safety nets backed by appropriate pan-European backstops and the inability of the European Stability Mechanism to recapitalise banks directly.
‘Good’ and ‘bad’ deleveraging
Although deleveraging has contributed to tighter global credit conditions, it is still necessary from a prudential perspective to further shrink and strengthen European banks’ balance sheets. This is required to realign their business models and comply with stricter international regulatory requirements which, as they are gradually phased in, aim to induce ‘good’ deleveraging while avoiding a disorderly process. This process will take several more years and deserves ongoing monitoring to address unintended consequences.
However, the deleveraging process has also adversely impacted credit conditions around the world, which can dampen economic growth (‘bad’ deleveraging), particularly in countries where bank credit is the dominant source of funding. This in turn can undermine bank-asset quality which worsens credit conditions further. As a result, credit has contracted in peripheral European countries and has started to weaken more recently in the core. These circumstances have also triggered a steady retreat from non-domestic (lending) activities towards home markets. Since a $24 trillion peak in 2008, European banks have reduced their total foreign claims by over 30% (Figure 3).
Deleveraging can be particularly damaging if it becomes disorderly and the scope to raise capital and shrink non-lending assets is limited. Although conditions are currently more benign due to a range of extraordinary central bank measures that averted an extreme funding scenario, deleveraging risks remain. These risks comprise a flare-up of the underlying causes of the Eurozone crisis which have yet to be fully addressed including incomplete burden sharing frameworks (e.g. fiscal and banking unions), the possibility of renewed financial market tension, potential forbearance in the banking system, and the deteriorating economic outlook in Europe in the context of severe austerity.
The impact on credit conditions in emerging and developing economies
Banks are often the dominant source of funding in emerging and developing economies and international banks in general and European banks in particular play a key role in their financial systems.3 As such, the impact of European bank deleveraging and tighter credit conditions is being directly transmitted to the rest of the world.
The global crisis has already produced considerable spillover effects to bank credit conditions in the rest of the world (Figures 4 and 5). Emerging and developing economies began to experience significant tightening of bank credit conditions in 2010 reaching a trough in 2011Q4, particularly in emerging parts of Europe. At that time, all emerging and developing economies across the world indicated their credit standards had been severely affected by European financial turmoil. Especially international funding conditions for emerging and developing economies deteriorated excessively. However, an account of extraordinary central-bank measures, conditions have improved significantly. On the demand side, conditions have continued to be expansionary, but at an increasingly slower pace since 2010.
Figures 4 and 5.
European bank deleveraging and the provision of credit in emerging and developing economies
As part of their pre-crisis expansion outside Europe, European banks regularly extended more new loans outside the Eurozone than within. Indeed, consolidated foreign claims4 of European banks to emerging and developing economies experienced extremely high pre-crisis annual growth rates of over 40%. However, as European credit conditions weakened, European banks disproportionately scaled back their external lending flows.
As another sign of deleveraging pressures, European banks retrenched much more than other international banks and stronger local banks in emerging and developing economies and in some cases these banks filled the gap that European banks left behind.5 However, European banks remain dominant (Figure 6).
The impact of European financial stress on emerging and developing economies credit - Supply or demand?
Figure 7 provides an illustrative exercise which assesses the extent to which financial shocks in Europe were transmitted to emerging and developing economies via European banks’ cross-border and local affiliates activity.6 The exhibit is based on a panel regression model which disentangles supply from demand.7 Supply factors are proxied by the three-month EURIBOR-OIS spread which reflects interbank stress. Demand factors are captured by country-level quarterly GDP growth in emerging and developing economies. The exhibit displays the portions of European foreign bank claims growth that can be explained by supply and demand factors for the average emerging and developing economies country.8 Claims growth is mostly driven by demand for credit in emerging and developing economies, with the exception of a significant slowdown in 2009. Average emerging and developing economies demand explained two percentage points of pre-crisis claims growth and vanished in 2009. Emerging and developing economies’ demand picked up in 2010 but has been falling and only explained about 1-1.5 percentage points. The exhibit clearly shows supply shocks propagated to emerging and developing economies in 2007, 2008, and 2011 when European interbank stress conditions spiked.
Recent research also found that as European banks deleveraged in 2008 and their foreign claims contracted, the domestic credit supply in recipient countries was directly negatively affected, even after taking into account the impact of local demand effects (Aiyar and Jain-Chandra 2012). This suggests many economies may not have been able to fill the gap European banks left behind.
Aiyar, S and S Jain-Chandra (2012), “The Domestic Credit Supply Response to International Bank Deleveraging: Is Asia Different?”, IMF Working Paper 12/258.
Dailami, M and J Adams-Kane (2012), “What Does the Future Hold for the International Banking System?”, World Bank Economic Premise Note 94.
European Banking Authority (2013), “EBA publishes results of the Basel III monitoring exercise as of 30 June 2012”, report.
Feyen, Erik and Inés González del Mazo (2013), “European Bank Deleveraging and Global Credit Conditions Implications of a Multi-Year Process on Long-Term Finance and Beyond”, World Bank Policy Research Working Paper 6388.
International Monetary Fund (IMF) (2012), Global Financial Stability Report, October.
Le Lesle, V (2012), “Bank Debt in Europe: Are Funding Models Broken?”, IMF Working Paper 12/299.
Takáts, E (2010), “Was it Credit Supply? Cross-border Bank Lending to Emerging Market Economies during the Financial Crisis”, Bank for International Settlements mimeo.
1 For a discussion on European bank funding models, also see Le Lesle (2012).
2 Some national regulators have imposed additional regulatory requirements on local affiliates of foreign banks to avoid capital and liquidity from exiting the country. While these measures might safeguard the local financial system, it harms European financial integration and makes international banks more vulnerable to shocks.
3 For example, the share of bank assets owned by foreign banks is on average about 10% in developing countries. However, large variation between countries and regions exist. For example, representing over 50% of total assets, foreign bank ownership is particularly prevalent in sub-Saharan Africa and emerging Europe.
4 Consolidated foreign claims include cross-border claims and claims of foreign affiliates in all currencies. Parent claims on foreign affiliates are netted out.
5 See Dailami and Adams-Kane (2012) for a discussion on the future of the international financial system.
6 Note that foreign claims do not include cross-border flows between the parent bank and its affiliates. Moreover, the foreign claims data are not available in currency-adjusted form because the currency composition of cross-border claims in unknown.
7 The regression is based on country panel data from 200Q1-2012Q2. The dependent variable is the quarterly growth rate of European consolidated foreign claims on EMDEs as provided by the Bank for International Settlements. The independent variables are quarterly changes in EURIBOR-OIS, GDP growth (YoY), and country- and season-fixed effects. The supply and demand factors are highly statistically significant using robust standard errors clustered on the country level. The model’s R-squared is 0.08. For a similar analysis using currency-adjusted locational claims, see Takáts (2010).
8 Since there is substantial heterogeneity between EMDEs, these results should be interpreted as illustrative only.