There has been a discernible slowdown in international trade and capital flows over the past three years. Are developed economies becoming more structurally autonomous?
Residents of the United States, please read this important information before proceeding
While the cyclical headwinds to cross-border activities are expected to dissipate as the recovery progresses, the potential impact of any near-term spillovers could well end up being more muted than historically observed.
The end of financial integration?
The progressively integrated nature of the international financial system has been a striking feature of the global economy over the last three decades. Developed economies, in particular the US and euro area, saw their external assets relative to GDP increase almost fourfold between 1990 and 2007, with emerging markets showing a similar yet more muted change.1 This trend was abruptly halted in 2008, with international capital flows and trade volumes falling sharply from the nadir of the global financial crisis. (Figure 1)
The recovery since the so-called Great Trade Collapse has been feeble, to say the least: global external assets as a share of GDP declined by 20% through 2013,2 while world trade – in a marked break from pre-crisis norms – is now growing at a more subdued pace relative to world GDP. (Figure 2) These recent trends in capital flows have prompted some commentators to suggest that economies are becoming more self-sufficient, suggesting an era of “deglobalisation” akin to the interwar years of the early 20th century. While comparisons to such an extreme episode of economic history – during which countries effectively behaved like closed economies – are somewhat misleading, one cannot deny the fact that there has been a noticeable break in the trend of financial globalisation since the onset of the Great Recession.
At the heart of the issue lies the global financial system, and more specifically, the banking sector. In part, the slowdown in cross-border activity is a backlash stemming from the crisis, as both national and multinational bodies look set to tighten their regulatory frameworks in an attempt to avoid repeating the mistakes that led to such egregious credit excesses.
However, the move towards a more resilient banking sector may have led to a fundamental change in the business models of large financial institutions, which have begun to scale back their external funding requirements in light of the added costs to cross-border flows.
As such, it could well be the case that the establishment of a safe and sound global financial system may be achieved at the expense of the pace of further financial globalisation, which has natural implications for international trade and global economic growth. Before discussing this trade-off in greater detail, it is important to take a look at the broader evidence demonstrating the contraction in global capital flows and highlight some of key drivers for these changes.
Slow to flow
For many of the world’s economies, capital flows show the same long-run pattern of increasing over time, falling sharply during the crisis and subsequently remaining significantly below previous levels. (Figure 3) The developments in aggregate are certainly dramatic: international capital inflows are now only 1.6% of global GDP, approximately ten times less than the peak of 16% in 2007.3 However, most of the retrenchment has been led by developed economies, in particular the euro area and the UK, which together can account for around two-thirds of the decline in inflows to the developed world since 2007.4 Inflows to the US have also diminished gradually, while flows into Japan have actually increased.
There is an alternative measure of capturing the extent to which financial globalisation has evolved, highlighting that for the UK economy, the stock of cross-border financial exposure has fallen by 23% since its peak.5 (Figure 4) Interestingly, the gap between the actual and trend levels bears a striking resemblance to those charts that show the loss of “potential” output in major economies since the crisis. This highlights the important link between the financial system and economic growth.
Developing on the above, economists often analyse the co-movement between investment and the level of savings to assess the degree to which an economy is financially globalised. In theory, investment in a more open economy can be funded by borrowing abroad, as opposed to merely being financed by the savings of domestic individuals, companies or governments. Therefore, more open economies tend to have a lower correlation between domestic investments and savings.6 One of the key benefits of this is that it allows for an economy to access external finance and smooth consumption requirements, which is especially useful at times when the domestic banking sector is faced with a shortage of liquidity. However, once again, this measure shows that financial globalisation across advanced economies has sharply reversed over the last six years. (Figure 5)
Among these various indicators, it is evident that capital and trade flows remain below their pre-crisis levels. However, it is important to stress the financial nature of the discussion thus far. Other measures of globalisation, based around information and migration, have been steadily progressing in recent years.7 This, coupled with the broader recovery in the global economy, makes the depressed levels of global financial integration even more perplexing.
Trends in banking can explain most of this decline
A reversal of international banking flows has primarily driven the retrenchment in global capital flows. This can be observed when the financial accounts of an economy are disaggregated into three broad types of capital: banking flows, portfolio flows (both equity and debt), and foreign direct investment (FDI).8 Up until 2013, global cross-border banking flows were still negative, but they have begun to show tentative signs of picking up this year.9 However, the reversal of these flows since the crisis highlights how banks across the world have been actively reducing their foreign exposures in an attempt to keep capital closer to home. This major contraction in the global banking network has been a crucial driver in the recent trend of financial deglobalisation.
Of course, the magnitude of contraction has varied across countries. Since 2008, the euro area and the UK, two economies heavily reliant on the financial services sector, have seen the foreign claims of their domestic banks fall by almost 40% and 10% respectively.10
In particular, for the UK, banking flows have been substantially more volatile in recent years compared to the rest of the developed world. (Figures 6 and 7) However, it is important to note that the high volatility of banking flows relative to portfolio flows (and FDI) is a common feature across all major economies.
Can we determine what is driving this volatility in banking flows? More recent academic studies have identified the existence of a “global financial cycle” that influences the movement of capital to and from advanced economies, as well as emerging markets11 This cycle is predominantly a function of changes in global risk aversion, US interest rates, and global banking conditions (leverage). For example, a decrease in global risk – as proxied by the implied volatility of the S&P 500 (VIX) – has historically led to an increase in international portfolio flows, in particular banking flows.12 Intuitively, this makes sense; when economic uncertainty is low, banks are more likely to increase leverage, as well as overseas lending. However, a number of commentators have highlighted that in the post-crisis period, this relationship has broken down. Banking flows are becoming significantly less affected by changes in global risk.13 (Figure 8)
Recoveries following a banking crisis
While the list of possible explanations for the decline in cross-border banking flows is long, there are four broad trends worth acknowledging: (i) the deleveraging of individual bank balance sheets; (ii) uncertainty stemming from banks, businesses and households; (iii) an abundance of “easy money”; and finally, (iv) the intense regulatory pressure on the financial system. In retrospect, the emergence of these factors should not be particularly surprising; they are what one would expect given both the severity – and financial nature – of the recent crisis.
“The build-up in cross-border lending was driven primarily by European banks”
Over time, the first three of the above headwinds are likely to wane as the global economic recovery progresses; however, it is the role of the fourth factor – and in particular more restrictive lending standards – that could act as a prolonged drag to future global banking flows.
Europe is primarily driving the deleveraging of global banks
Much evidence suggests that the rapid pre-crisis build-up in cross-border lending was driven primarily by European banks, which was partly financed through wholesale funding – loans from international investors such as money-market mutual funds. Following the onset of the crisis, this source of funding dried up, with the banking system in Europe largely affected. Wholesale funding remains a substantial element of the financing mix for European banks, still standing at over 45% of total liabilities in 2014.14
The subsequent spread of the euro area sovereign debt crisis exacerbated the higher cost of funding. While some of the retrenchment in capital affected domestic Europe – as seen in tighter lending standards and slump in credit growth – cross-border exposures were prioritised in cutbacks.15 The strongest deleveraging occurred in banks within the European periphery, but the process is still ongoing, as it began much later than in the UK and US. Furthermore, the one-off balance sheet adjustment that occurred in the latter has not been an option for European policymakers, in light of the fiscal constraints facing the monetary union.
Outside of Europe, bank deleveraging has either slowed (as in the UK and Japan) or ended (as in the case of the US) since 2011.16 While there has been an uptick in cross-border lending activities over the course of this year, it is almost negligible when compared to pre-crisis levels. This is partly due to some of the other headwinds highlighted in this section, which indicate that it may take some time for such flows to rebound with any dynamism.
Uncertainty is present in both the supply and demand of capital
While the above focuses on the fall in the supply of bank capital and its impact on cross-border activities, one must also consider headwinds to the other side of the equation. Uncertainty is a key factor driving the weakness of household and business demand for overseas goods or capital. Economic trade models tend to show that in times of uncertainty, firms adopt a “wait-and-see” approach and are more inclined to slow their foreign investment activities relative to domestic inventory orders.17 Households, likewise, respond to uncertainty by reducing their consumption of foreign goods, as seen by the sluggish levels of advanced economy imports in recent years. (Figure 9) The weakness in demand for foreign goods and services would naturally lead to a lower quantity of foreign transactions by domestic banks, thereby explaining some of the weakness in cross-border banking flows.
As we explained above, the relationship between market volatility and bank lending has broken down in recent years. While this could be due to the fact that market volatility is not an accurate proxy for overall economic uncertainty, a more fitting explanation can be found in the latest studies of the bank-lending channel. The main finding highlights that when banks face uncertainty over their deposits – as captured by the greater variance in reserves – they are less likely to increase their lending, even if their balance sheets are healthy.18 More specifically, if it is likely that deposits will not be available in future periods, banks will naturally be less willing to lend against these “flighty” assets. This has strong implications for the effectiveness of quantitative easing (QE); while the policy initially leads to a greater level of reserves in the economy, its main channel of portfolio rebalancing naturally leads to greater movement of reserves from one bank to another. Therefore, central bank asset purchases can in effect inhibit bank lending. However, while further discussion of this is beyond the scope of this article, the main purpose is to highlight how other forms of uncertainty exist when banks choose to supply loans to economic agents.
Unconventional policies in advanced economies
In previous recessions, governments have typically attempted to protect their domestic industries from foreign competition through protectionist policies.19 For example, in the aftermath of the Great Depression, there was an increase in the number of tariffs and import quotas imposed by developed economies. After the financial crisis of 2008, many feared that the trade policy developments of the 1930s would be repeated. But it seems that following the establishment of the G20, such traditional forms of protectionism have been largely avoided.
However, inadvertently, the unconventional policies pursued by developed world policymakers in the aftermath of the 2008 crisis may have given rise to a new wave of “financial” protectionism. The redirection of lending towards domestic uses or buy-local clauses in bailout packages may have (unintentionally) encouraged banks to keep their limited capital within borders, thereby reducing their foreign exposures.20 Furthermore, the exceptionally loose global monetary conditions have partly facilitated a compositional shift in capital flows from bank lending towards portfolio flows, with banks seeking “safer” ways to gain a return on their excess capital.
Global regulation since the crisis
The substantial overhaul of regulation – aimed at reforming the global financial system – may have potentially hindered the return to large banking flows in the post-crisis world. The nuances of the Great Recession demonstrated the risks posed by large global financial institutions that can be too big to rescue, and one of the main themes of reform has been focused on limiting the size of banks, which would therefore, in effect, limit their international operations as well.
A far-reaching global regulatory agenda, under the policy leadership of the G20, has been put in place to strengthen the balance sheets of financial institutions and enhance the resilience of financial markets. Over the course of this year, we have seen many advanced economies introduce macro-prudential policies to help mitigate financial risks related to capital and leverage.21
However, it seems likely that such regulatory reforms have already had a substantial impact on both the behaviour and business models of banks.22 For example, the coordinated global policy involving the increase of minimum capital requirements was aimed at reducing bank leverage,23 which would therefore lead to a decline in global – and domestic – lending. Any such reduction, however, could be proportionally larger in international lending than for domestic lending – given that the risk-weightings on international lending are typically greater. There has been a unanimous tightening of minimum capital requirements across the developed world, as economies seek to strengthen their banking systems and prepare for the implementation of the global Basel III requirements in 2019.
Barring the implementation of such policy, most regulatory proposals have all been inward-looking. It seems that the international regulatory community is losing hope of building a resolution regime that would work across countries and is focusing instead on ring-fencing approaches, where a portion of a company's assets or profits are financially separated, without necessarily being operated as a separate entity. As a result, the banking world is also becoming more uneven. In the absence of any co-ordinated policy, the regulatory drag on cross-border banking flows could remain in effect over the medium-term.
Conclusion: near and long-term applications
Subdued cross-border banking flows – substantially lower relative to GDP – have been a significant feature of the global economy over the past six years and have only modestly picked up in 2014. Going forward, we have highlighted the factors that could lead to this trend continuing over the near-term. There are a number of implications, not only for policymakers, but also markets.24 However, for the purpose of simplicity, we identify two key take-away points from this analysis:
Spillover effects are likely to be muted in the near-term
Given that cross-border banking exposure continues to remain systemically weak across the world, it is likely that the propagation mechanisms that transmit spillovers, both positive and negative, are weak.
On the one hand, this can be viewed with some optimism for the select group of economies that seem to be outperforming, since any contagion effects stemming from the weakness of those more vulnerable are likely to be limited in the near-term. This is of particular importance for the US, where exports, at around 10% of GDP, are less relevant for domestic growth in comparison to most other advanced economies. (Figure 10)
On the flip side, those economies that were seeking a helping hand from a booming US economy may be left disappointed. Central bankers pursuing a loose monetary stance – and looking for a devalued currency to provide an external boost to the economy – may not observe any immediate gains via this channel in the absence of any material pick-up in import demand from advanced economies. (Figure 9) Long-term solutions aimed at fixing the domestic impairments should be a first-order priority for these economies.
Regulate today, develop tomorrow
In a long-term context, increased regulations that reduce bank flows today, but ensure a more robust financial system in the future, could well end up supporting higher levels of financial globalisation in the future. The global banking network will remain an important source of capital flows and financing; major financial institutions will continue to be active across borders, more so if they are better capitalised and more resilient as they emerge from the crisis. But the international community should be working together to underpin the system of global capital flows. A more financially integrated economy and expanding global economy can co-exist if all players act in a cohesive manner.
1 Source: IMF World Economic Outlook (WEO) as of 30 Nov 2014.
2 Source: Institute of International Finance (IIF) as of 01 Dec 2014.
3 Source: Speech by Kristin Forbes, Bank of England Monetary Policy Committee, on 18 Nov 2014.
4 Source: IIF as of 01 Dec 2014.
5 Foreign assets plus liabilities as a share of GDP, source Forbes (2014).
6 Source: IIF as of 01 Dec 2014.
7 Source: The DHL Global Connectedness Index as cited in “Signs of life”, The Economist, published on 15 Nov 2014.
8 As in Forbes (2014), we associate the “other” component of the IMF IFS data set as “banking” flows. This is largely accurate; however, the data series also includes transactions in currency and deposits, loans, and trade credits.
9 Source: IIF and Bank of International Settlements (BIS), figures include latest data for Q2 2014 as of 08 Dec 2014.
10 Source ibid.
11 This is the pioneering research of Hélène Rey (2013) that has attracted much attention over the past year.
12 A number of studies highlight the role of risk aversion; see Rey (2013) for a comprehensive literature review.
13 Forbes (2014) presents a more robust analysis.
14 Source: IIF as of 01 Dec 2014.The relative figure for the US banking sector is 10%.
15 “Global Liquidity and Drivers of Cross-Border Bank Flows”, Cerutti et al. (2014).
16 Source: IIF as of 01 Dec 2014.
17 “Trade and Uncertainty”, Novy and Taylor (2014).
18 For full details see “QE and the Bank Lending Channel in the United Kingdom”, Butt et al. (2014).
19 “The mystery of the missing world trade growth after the global financial crisis”, Economic Review, Riksbank (2014).
20 Cerutti et al. (2014) argue that it was less unintentional than thought. For example, the “Buy American” clause in the American Recovery and Reinvestment Act of 2009 prohibited the use of stimulus funds for construction, repair or maintenance of a public building or public work unless all of the project’s iron, steel and manufactured goods used have been made in the United States.
21 This includes the likes of the UK, New Zealand, Sweden, Norway and more recently Australia as of 10 Dec 2014.
22 See IMF, Global Financial Stability Report, October 2014.
23 And importantly did not directly favour domestic over foreign lending, source Forbes (2014).
24 For more causes and implications see Forbes (2014), IIF (2014).