Since Bernanke’s 2005 speech (Bernanke 2005 ), it has become commonplace to assert that current-account imbalances were a key factor in stoking the permissive financial conditions that led to subprime lending in the US. The ‘global savings glut’ is what Ben Bernanke called it. This phrase provided a powerful linguistic focal point for thinking about the surge in net external claims on the US on the part of emerging economies. The biggest worries concern the financial stability implications of these large and persistent current-account imbalances.
But maybe the finger is being pointed the wrong way. My recent research suggests that the ‘global banking glut’ may have been more culpable for the crisis than the ‘global savings glut’ (Shin 2011).
What is the ‘global banking glut’?
To introduce the distinction, it is instructive to start with the financial crisis in Europe. What role did current-account imbalances play there? There are some superficial parallels with the US in the run-up to the crisis.
The current-account deficits of Ireland and Spain widened dramatically before the crisis (Figure 1). This despite the fact that Spain and Ireland were paragons of fiscal rectitude – with budget surpluses and low debt ratios (much lower than Germany’s and the Eurozone average in 2006, see Figure 2).
Figure 1. Current account of Ireland and Spain (% of GDP)
Source: IMF International Financial Statistics
Figure 2. Government budget balance and debt-to-GDP ratios of Ireland, Spain and Germany
To push the analogy with the US further, imagine for a moment that the Eurozone is a self-contained miniature model of the global financial system. In this miniature model, Germany plays the role of China, while Spain and Ireland play the US.
According to the analogy, excess savings in Germany find their way to Spain and Ireland where they inflate the property bubbles there. The bubbles subsequently burst, resulting in the socialisation of private debt through bank bailouts and precipitating the sovereign-debt crisis.
However, the further we push the analogy, the stranger it gets. According to the ‘global savings glut’ hypothesis, Chinese savers favour US Treasuries because China lacks deep financial markets that could cater to demands for safe assets. In the miniature model of the savings glut for the Eurozone, Spanish and Irish bank deposits play the role of US Treasuries, since current-account imbalances in the Eurozone were financed through capital flows in the banking sector. To sustain the analogy, we would need to argue somehow that German savers shunned bank deposits in Germany to favour bank deposits in Ireland and Spain. Why would German savers believe that deposits in Spain and Ireland are safer than those in Germany? At this point, the savings glut analogy strains credulity and breaks down.
A more plausible narrative is a banking glut associated with the explosive growth of cross-border lending in Europe, as illustrated by Figure 3 which plots the cross-border domestic currency lending and borrowing by EZ banks.
Figure 3. Cross-border domestic currency assets and liabilities of EZ banks
Source: BIS Locational Banking Statistics, Table 5A
There is a mechanical jump in the two series at the start of 1999 with the launch of the euro, as previously foreign-currency lending and borrowing are reclassified as being in domestic currency (i.e. euros). But from 2002, cross-border bank lending saw explosive growth as the property booms in Ireland and Spain took off and as European banks expanded their operations in central and Eastern Europe.
What drove European banks to do this? By eliminating currency mismatch on banks’ balance sheets, the introduction of the euro enabled banks to draw deposits from surplus countries in their headlong expansion. Meanwhile, the permissive bank-capital rules under Basel II removed any regulatory constraints that stood in the way of the rapid expansion. To be fair, the permissive bank risk-management practices epitomised by Basel II were already widely practised within Europe before the formal introduction, as banks became more adept at circumventing the spirit of the initial 1988 Basel Capital Accord.
Compared to other dimensions of economic integration within the Eurozone, cross-border mergers in the European banking sector remained the exception rather than the rule. Herein lies one of the paradoxes of Eurozone integration. The introduction of the euro meant that "money" (i.e. bank liabilities) was free-flowing across borders, but the asset side remained stubbornly local and immobile. As bubbles were local but money was fluid, the European banking system was vulnerable to massive runs once banks started deleveraging.
Europe’s crisis: A banking crisis first, a sovereign-debt crisis second
The banking glut hypothesis is a better perspective on the current European financial crisis than the savings glut hypothesis. The crisis in Europe is a banking crisis first, and a sovereign-debt crisis second. It carries all the hallmarks of a classic "twin crisis" that combines a banking crisis with an asset-market decline that amplifies banking distress. In the emerging-market twin crises of the 1990s, the banking crisis was intertwined with a currency crisis. In the European crisis of 2011, the twin crisis combines a banking crisis with a sovereign-debt crisis, where the mark-to-market amplification of financial distress worsens the banking crisis.
The banking glut in Europe was part of a global phenomenon, as documented in a recent paper delivered as this year’s Mundell-Fleming lecture at the IMF (Shin 2011). Effectively, European global banks sustained the “shadow banking system” in the US by drawing on dollar funding in the wholesale market to lend to US residents through the purchase of securitised claims on US borrowers, as depicted in Figure 4.
Figure 4. European banks in the US shadow banking system
Although European banks' presence in the domestic US commercial banking sector is small, their impact on overall credit conditions looms much larger through the shadow banking system. The role of European global banks in determining US financial conditions highlights the importance of tracking gross capital flows in influencing credit conditions, as emphasised recently by Borio and Disyatat (2011). In Figure 4, the large gross flows driven by European banks net out, and are not reflected in the current account that tracks only the net flows.
The netting of gross flows shows up in Figure 5, which plots US gross capital flows by category. While official gross flows from current-account surplus countries are large (grey bars), we see that private-sector gross flows are much larger.
Figure 5. Gross capital flows to/from the US
Source: US Bureau of Economic Analysis
The downward-pointing bars before 2008 indicate large outflows of capital from the US through the banking sector, which then re-enter the US through the purchases of non-Treasury securities. The schematic in Figure 4 is useful to make sense of the gross flows. European banks' US branches and subsidiaries drove the gross capital outflows through the banking sector by raising wholesale funding from US money-market funds and then shipping it to headquarters. Remember that foreign banks' branches and subsidiaries in the US are treated as US banks in the balance of payments, as the balance-of-payments accounts are based on residence, not nationality.
European banks: Gross flows and US pre-crisis credit conditions
The gross capital flows into the US in the form of lending by European banks via the shadow banking system will have played a pivotal role in influencing credit conditions in the US in the run-up to the subprime crisis. However, since the Eurozone has a roughly balanced current account while the UK is actually a deficit country, their collective net capital flows vis-à-vis the US do not reflect the influence of their banks in setting overall credit conditions in the US.
The distinction between net and gross flows is a classic theme in international finance, but deserves renewed attention given the new patterns of international capital flows (see, e.g., Borio and Disyatat 2011). Focusing on the current account and the global savings glut obscures the role of gross capital flows and the global banking glut.
Net capital flows are of concern to policymakers, and rightly so. Persistent current-account imbalances hinder the rebalancing of global demand. Current-account imbalances also hold implications for the long-run sustainability of the net external asset position. For the US, however, the current account may be of limited use in gauging overall credit conditions. Rather than the global savings glut, a more plausible culprit for subprime lending in the US is the global banking glut.
As the European banking crisis deepens, the deleveraging of the European