VoxEU Column Financial Markets International Finance

The two faces of cross-border banking flows: An investigation into the links between global risk, arms-length funding, and internal capital markets

Cross-border funding between banks collapsed following the bankruptcy of Lehman Brothers, but the withdrawal of funding was not uniform across countries. This column argues that the composition of cross-border bank-to-bank funding can help to explain why. Interbank funding between unrelated banks is particularly vulnerable to global shocks, but intragroup funding between related banks can act as a stabilising force, particularly for advanced economies with a high share of global parent banks. Policymakers should look at disaggregated cross-border bank-to-bank flows, as doing otherwise could result in a misleading assessment of financial stability risks.

Following the collapse of Lehman Brothers in September 2008, global risk spiked and the world witnessed a collapse in cross-border funding between banks. On closer inspection, however, not all countries’ banking systems experienced a withdrawal of cross-border finance. In fact, a number actually enjoyed an inflow of funding from banks overseas (Figure 1).

Figure 1 Cross-border bank-to-bank flows following the collapse of Lehman Brothers

One possible explanation for this finding is that the composition of banking systems’ cross-border funding varies across countries. In particular, cross-border funding between banks can be decomposed into two distinctive forms of funding:

  • Arms-length (interbank) funding that takes place between unrelated banks.
  • Related (intragroup) funding that takes place between global banks and their foreign affiliates.

In Figure 2 we display the mix of interbank and intragroup funding across countries. Due to data confidentiality, we cannot report the identity of each country, but the mixture of alternative funding structures is evident. Some banking systems hold comparatively little cross-border funding from related banks while others largely avoid the wholesale interbank market.

Figure 2 Intragroup funding across countries

If the two funding types respond differently to fluctuations in global risk it could help explain why some countries experienced larger withdrawals of funding than others and, thus, provide information relevant for financial stability – both at home and abroad.

It has been documented that both forms of funding are at risk of withdrawal during periods of heightened global risk (Cetorelli and Goldberg 2011) and recent economic theory predicts that the two forms of funding should behave in the same way when global risk is high or rising (Bruno and Shin 2014).

Yet, the two forms of funding also have key differences, which may mean they behave differently during an economic crisis. For example, within an internal capital market, a global parent bank has the power to shift liquidity from one part of its group to another. Additionally, a bank lending internally has more information about its counterparties’ overall riskiness, relative to banks lending at arms-length, which could influence their lending behaviour during risky episodes. It is therefore possible that some countries’ banking systems could be more insulated from heightened global risk than others, depending on their mix of interbank and intragroup funding and the share of intragroup funding held by global parent banks relative to foreign affiliates.

In new research (Reinhardt and Riddiough 2014), we empirically study the behaviour of interbank and intragroup funding in relation to swings in global risk and find that the split has statistical, theoretical, and economic importance. Using international banking statistics from the Bank for International Settlements (BIS), we find that a period of high and rising global risk – such as that witnessed following the collapse of Lehman Brothers – results in markedly different behaviour in ensuing interbank and intragroup flows.

A first look at interbank and intragroup funding

The left-hand side panel of Figure 3 shows that interbank funding fell on average across our sample of BIS reporters by almost 30% between September 2008 and the end of 2009. Yet, in contrast, intragroup funding increased in the immediate aftermath of the collapse of Lehman Brothers and was stable for the remainder of the crisis period.

The contrasting behaviour of interbank and intragroup flows is not limited, however, to the recent global financial crisis. To see this, in the right-hand side panel of Figure 3, we present the distributional relationship across time between cross-border bank-to-bank funding and the VIX index.

We find that on average, between 1998 and 2011, interbank funding contracted by 2% during quarters when the VIX index was at an elevated level (upper-25th percentile), while during the same quarters intragroup funding expanded by over 2%. In the quarters when the VIX index was particularly low (lower-25th percentile), both intragroup and interbank funding expanded by approximately 4%.

Figure 3 Global risk and cross-border bank flows

Statistical and economic significance

In our formal econometric panel analysis, we examine the extent to which global risk (which we proxy using the VIX index) and other theoretically motivated factors can explain the quarterly percentage change in interbank and intragroup funding between 1998Q1 and 2011Q4. Confirming our preliminary findings, we find interbank and intragroup funding react in opposing directions to fluctuations in global risk.

In contrast to economic theory, intragroup funding, which makes up around half of all cross-border funding between banks, rises when global risk increases, and remains stable during periods of heightened global risk. On the other hand, interbank funding, which we find to be less volatile on average than intragroup funding, displays the opposite behaviour and is withdrawn during periods of elevated global risk, with emerging economies particularly vulnerable.

In fact, the simple split between interbank and intragroup funding can help explain the non-uniform withdrawal of funding across national banking systems during the financial crisis. In particular, we find that considering each country’s mix of interbank and intragroup funding alone can explain up to 45% of the change in cross-border funding to banks following the collapse of Lehman Brothers.

In Figure 4 we plot the actual and predicted percentage loss in total funding following the collapse of Lehman Brothers (i.e. between 2008Q3 and 2009Q2). The predictions are based on (i) each country’s mix of interbank and intragroup funding and (ii) our regression coefficients on global risk. Given our regression results, countries with smaller predicted falls in funding have a higher share of intragroup funding in total bank-to-bank funding.

A clear upward trend emerges – countries with higher shares of intragroup funding experienced smaller outflows of cross-border funding from banks abroad. The magnitudes are of economic significance. A banking systems with a 20/80 split of intragroup to interbank funding, for example, would be predicted to experience a 30% drop in funding when global risk is elevated to a magnitude comparable to that witnessed during the financial crisis.

In contrast, a country with an 80/20 split of intragroup to interbank funding (the US has similar proportions) would be expected to experience only a small withdrawal of funding in the same conditions. In fact, the US, at the epicentre of the global financial crisis, experienced a relatively mild 5% loss of cross-border bank-to-bank funding in the aftermath of Lehman’s collapse.

To emphasise the theoretical significance of this result, had we assumed that interbank and intragroup funding behaved symmetrically, we would have predicted a uniform 15% loss in funding across every banking system and, in doing so, would explain none of the cross-sectional spread in funding changes during the crisis.

Figure 4 Funding loss during the global financial crisis

Parent and foreign affiliate banks

We investigate intragroup funding further by decomposing flows between:

  • funding to global parent banks from their foreign affiliates;
  • funding to foreign affiliates from their global parent bank and other related affiliates.

We find that increased intragroup funding during episodes of heightened risk is principally driven by global parent banks, headquartered in advanced economies, receiving funding from their foreign affiliates.

The result supports the view expressed by Kohn (2008) that global banks may respond to an economic shock by using foreign affiliates as a source of liquidity, limiting liquidity pressures at home.

The result also supports a recent finding by Hoggarth et al. (2013), who show that gross intragroup lending by foreign branches resident in the UK increased strongly following the run on the British bank Northern Rock.

We do not find, however, any evidence of significantly reduced intragroup funding to foreign affiliates in either advanced or emerging economies during periods of heightened global risk. In fact, we find that foreign affiliates resident in emerging economies experience an increase in intragroup funding, when the average profitability of banks in the local economy is low and is indicative of the beneficial role financial globalisation can play for emerging economies with resident foreign banks.

In Figure 5 we present a visual depiction of these results for the global financial crisis. Specifically, we plot median cumulative changes in aggregate (interbank plus intragroup) cross-border bank-to-bank funding following the collapse of Lehman Brothers, conditional on:

  • the banking systems’ share of intragroup funding; and
  • the proportion of intragroup funding held by resident parent banks.

By the end of 2009, banking systems funded with a relatively high share of arms-length interbank funding had experienced, on average, a 20% drop in funding, while the fall in funding was, on average, less than 8% for banking systems with a high share of intragroup funding. Moreover, banking systems with a high share of intragroup funding held predominately by parent banks experienced almost no loss in cross-border bank-to-bank funding during the global financial crisis – amplifying the contrasting behaviour in interbank and intragroup funding in relation to fluctuations in global risk.

Figure 5 Intragroup funding and global parent banks


The spike in global risk which accompanied the collapse of Lehman Brothers in September 2008 was followed by a collapse in cross-border banking flows. These events prompted academics and policymakers to focus their attention on the behaviour and determinants of this economically important form of finance. In fact, the Committee on International Economic Policy Reform (2012) concluded that, ‘effective regulation of cross-border banking is essential for domestic and global financial stability’.

Our results call for policymakers and academics to monitor the decomposition of cross-border funding between banks, as the contrasting behaviour of interbank and intragroup funding – split between parent and foreign affiliates banks – in response to fluctuations in global risk, has implications for banking system financial stability.


Bruno, V and H S Shin (2014), “Cross-Border Banking and Global Liquidity”, Working Paper, Princeton University.

Cetorelli, N and L S Goldberg (2011), “Global Banks and International Shock Transmission: Evidence from the Crisis”, IMF Economic Review, 59: 41–76.

Committee on International Economic Policy and Reform (2012), “Banks and Cross-Border Capital Flows: Policy Challenges and Regulatory Responses”, Brookings, Washington DC.

Hoggarth, G, J Hooley and Y Korniyenko (2013), “Which Way do Foreign Branches Sway? Evidence from the Recent UK Domestic Credit Cycle”, Bank of England Financial Stability Paper 22, June.

Kohn, D L (2008), “Global Economic Integration and Decoupling”, Speech presented at the International Research Forum on Monetary Policy, Frankfurt, Germany, 26 June.

Reinhardt, D and S J Riddiough (2014), “The Two Faces of Cross-Border Banking Flows: an investigation into the links between global risk, arms-length funding and internal capital markets”, Bank of England Working Paper 498.

2,835 Reads