VoxEU Column International Finance Monetary Policy

Capital flows, debt and growth: Dilemmas and tradeoffs in the global agenda

The objectives of maximising growth and reducing external imbalances may not be fully compatible in a financially integrated and asymmetric world. This column argues that countries have two choices: they can contain global imbalances and gross financial flows through permanent capital controls, or they can pursue financial integration, managing growing imbalances and external exposures by creating more global safe assets. This implies debt contracts would be either state-contingent, with easy restructuring, or built to be ‘safe’, with a high level of commitment by the issuer.

The global economic agenda aims to promote sustainable growth while strengthening the international monetary and financial system. Underpinning those priorities is one central, although implicit, component: global current account imbalances should be contained. In parallel, stocks of cross-border debt should be flexibly managed.

At face value, this agenda is coherent. Current account imbalances create risks to economic and financial stability. Current account reversals, ‘sudden stops’ and debt crises entail huge economic costs. Because of those strong feedbacks between financial instability and economic growth, it makes sense to prevent crises by limiting imbalances and making debt easily manageable through restructuring.

In the longer run, however, limiting current account imbalances and cross-border debt may not yield maximum growth in a financially integrated and deeply asymmetric world. I therefore question the validity of the dominant policy consensus. Under which conditions are the objectives of maximising growth and reducing external imbalances fully compatible?

The policy implications are significant. In the next decade, countries must decide what kind of financial interaction to establish between their economies. Their choices will determine the shape and structure of the international financial system. One of those choices will be the nature of cross-border financial instruments to be exchanged between countries, in particular debt claims.

A contradiction

As a starting point, look at two well-known characteristics of the current system.

First, growing financial integration. Over the last three decades, gross capital flows have expanded rapidly. This occurred first between advanced economies, and later also included emerging economies. Lately the expansion has been paused, and even reversed. The stock effects are persistent, however. Global external exposures, as measured as the ratio of assets and liabilities to GDP, have risen. In the 1980s they were 70%, in 2011 they were 450% for advanced economies (Borio 2015). According to Obstfeld (2015), Japan's financial openness has doubled since the late 1990s, while that of the US has tripled to about 1.5 times GDP. In newly industrialised Asia (Hong Kong, Korea, Singapore, Taiwan), average openness is about three times GDP. Cross-border stocks of capital are almost certainly larger than at any time in human history (Haldane 2014).

Financial integration also shows up in increased correlations between monetary and financial conditions in different countries. There has been a steady rise in the degree of co-movement in asset prices, particularly long-term interest rates, that are more correlated across countries than short-term rates (Goodhart and Turner 2014). This is due to a growing co-movement in term premia.

Second, asymmetry also defines the current system. Net private saving ratios differ sharply and permanently between countries, as the Feldstein–Horioka correlation has all but disappeared since the start of the 21st century. There are well-known asymmetries in financial development, and therefore differences in ability to generate safe assets, which is especially frustrating for EMEs. Countries that supply safe assets naturally serve as providers of stores of value and foreign exchange reserves. By extension, the same currencies are also used as medium of exchange and units of account. Therefore, their financial conditions tend to ‘dominate’ in global and domestic capital markets. (Caruana 2012).

The conjunction of openness and asymmetry logically produces growing current account imbalances. Saving–investment disequilibria are reflected in current account positions. In addition, as domestic monetary and financial conditions are increasingly determined by global factors, they become disconnected from country-specific economic and financial cycles, often amplifying booms and busts in domestic demand and asset prices. This creates current account imbalances.

Here is the contradiction. The system naturally produces increasing imbalances, but an accepted policy objective is to contain those imbalances, and especially to avoid strong and durable current account surpluses and deficits. Or, to put it differently, countries want to expand gross capital flows but limit net capital flows. They accept the asset price correlations implied by financial integration, but implicitly put a quantitative constraint on the resulting net flows.

Are those objectives mutually consistent? In principle, yes. There is no fundamental reason why the expansion of gross capital flows, it broadly balanced, could not spontaneously generate small net flows. Indeed, this is what happened between the US and Europe in the years preceding the financial crisis. This, however, is highly unlikely to be a permanent feature of an asymmetric world. In an asymmetric world, differences in private saving rates and financial development, if not compensated by policy actions, would result in durable current account imbalances and continuous increases in gross and net international exposures. So there is an intrinsic contradiction in seeking to have limited current account imbalances and deep financial integration while there are differences in saving–investment ratios and financial development across regions.

In trying to resolve this contradiction, countries may create deflationary bias in policymaking. Countries whose financial conditions become easier because of capital opening will run a current account deficit. To compensate, restrictive (most likely, fiscal) policies will be needed to keep the current account in check. This is the standard IMF prescription for countries that must deal with sudden capital inflows. Conversely, countries in which financial conditions become stricter may experience economic contraction as a result. Global GDP will be lower than if no constraints had been imposed on the current account.

Some asymmetries will disappear in the long run. Saving rates may converge and so will financial development, allowing for a more equal distribution of safe assets around the world. That will take time, however, because it depends primarily on financial liberalisation in the largest emerging countries that are potential producers of safe assets. China's experience over the last year shows the perils of trying to shortcut this process.

For the coming decades, therefore, we must choose between two corner solutions. Either containing global imbalances and gross financial flows – which would imply some kind of permanent capital controls - or, alternatively, accommodate a continuous increase in gross and net capital flows and an accumulation of cross-border claims. Both options are internally consistent. Staying in the middle, with deep financial integration and limits to current account imbalances, may compromise the achievement of strong and sustainable growth. It may create permanent volatility.

The status of debt

The choice has many policy implications. This column explores one: the status of debt. This is it at the core of current debate in both Europe and the G20.

Debt is central to the workings of the international financial system. While equity flows to and from EMEs have been important, it is difficult to imagine any progress in financial integration that would not mainly be based on debt. As Obstfeld (2015) states:

"The extreme ratios of external asset stocks to GDP that some countries display are not feasible except on the basis of extensive two-way trade in debt or debt-like instruments, including derivatives".

Debt flows may also be a potent source of instability. Sudden stops are frequently triggered by the perception that a country's external debt has become excessive.

So what is the proper choice of international regime for debt, so that it reduces the risk of balance-of-payment crises with abrupt adjustments in the current account? There are two opposing views. One would allow for easy debt reduction and restructuring, and the other would strengthen the status of debt as a safe asset. At some point, we must choose. This choice will have significant consequences for the architecture of the global financial system.

In the first view, debt should be state-contingent, and the possibility of restructuring should be provided ex-ante in debt contracts. Most of the financial instabilities that debt generates come from the design of debt contracts. When they are too rigid, they are excessively and discontinuously sensitive to negative shocks, which ultimately creates doubts on debt sustainability.

A related argument asserts that default on debt is a natural feature of the market mechanism, not something to be avoided at all costs. The possibility of default leads to better pricing of risk. It limits over-lending created by moral hazard. It aligns incentives of debtors and creditors to seek early debt relief early. It provides clarity and predictability in resolution processes. In pure financial terms, therefore, state-contingent debt is attractive, because it aids efficient allocation of capital and of risk.

That approach has a long history. It inspired the ill-fated Sovereign Debt Reduction Mechanism (SDRM) and the generalisation of Collective Action Clauses (CACs) in sovereign debt. It also partially drives the current G2O agenda, as well as IMF work on debt sustainability.

The alternative view emphasises the fundamental incompleteness of global capital markets. In a truly global financial system, domestic and foreign assets should be perfect substitutes for all asset classes. This is obviously not the case. Because only a few countries can issue safe assets, cross-border flows, especially between different currencies, mainly involve risky assets. This is a major difference from domestic flows, where safe claims can be exchanged, or serve as collateral. It is also the reason why, in aggregate, cross-border flows are more volatile than domestic ones. There is abundant evidence of a close link between the evolution of cross-border capital flows and risk appetite (CGFS 2011).

In this view, robust and resilient cross-border intermediation needs ‘safe’ cross-border flows. This has consequences in the form of debt that sovereigns can issue.

The two views differ over the information-sensitivity of debt. Debt can be considered safe if its value stays the same in most different states of the world. This implies the value is not affected by new information. Most liquidity crises are triggered by the sudden appearance of information asymmetries. Truly safe debt cannot create liquidity problems because it excludes, almost by construction, any information asymmetries (Holmstrom 2015). For that reason, also, debt is the best collateral. Cross-border intermediation based on safe debt would reduce the volatility of capital flows, and strengthen the resilience of global capital markets. It would support higher current account deficits and international capital stocks.

Safety depends on the underlying fiscal position of the sovereign issuer, but also, as is often overlooked, on its level of commitment. That's where debt structure matters. This is the fundamental choice facing the international community.

There is a discontinuity in the payoff attached to a debt contract. It is fixed in most states of the world, but suddenly drops when the threshold for sustainability is reached. Different forms of contract treat that discontinuity differently. State-contingent debt is designed to be fully information sensitive from the start, and so eliminates the discontinuity. It is therefore protected from the disruption when a previously safe debt is suddenly perceived as risky. Permanent changes in its expected payoff, however, do not allow it to act as a store of value. Contingent debt may prevent the type of market breakdown that creates sudden stops, but it would perpetuate – indeed, reinforce – the fundamental segmentation that currently exists between domestic (relatively safe) and global (relatively risky) capital markets.

On the contrary, rigid debt contracts that embed a high level of commitment would push the threshold for information sensitivity further away, protecting the value of debt in most states of the world with the tail risk that the commitment ultimately has to be breached.

Three additional remarks can be made.

  • There is a close relationship between debt and international liquidity provision.

Safe government debt is the asset in which much of foreign exchange reserves are invested. Thus the public component of ‘global liquidity’ is mainly comprised of debt. A new ‘Triffin dilemma’ ensues (Obstfeld 2015) in which the provision of international liquidity necessitates that reserve countries issue more sovereign debt. This increase in the stock of debt may endanger the fiscal sustainability of reserve issuers, and the safety of their debt. This could lead, in the future, to a shortage of safe assets that would complicate the construction of a resilient and open global financial architecture. Delinking the production of safe assets from specific country risk could therefore be a major contribution to the stability and resilience of the international financial system. Proposals exist for creating such safe assets in Europe, underpinned by a diversified portfolio of sovereign debt (Brunnermeier et al. 2016). In a distant future, safe multi-country debt could be used worldwide as a cross-border collateral, and underpin an extensive network of swaps between central banks (Landau 2013).

  • The status of public debt determines the core–periphery dynamic that may be the source of asymmetries in any financial system, whether regional — as in Europe — or global.

As above, if one or more countries have a monopoly in the issuance of safe debt, they have a dominant influence on overall financial conditions. In periods of stress, a flight to quality naturally generates capital flows from the risky periphery to the safe centre, exacerbating local liquidity problems and balance-of-payments tensions (Brunnermeier et al. 2016). This has been apparent both globally and during the euro crisis. Because of so-called private sector involvements (PSI), a large chunk of (peripheral) government debt instantly loses its status as a safe asset. It is hard to see how a more symmetric monetary system could emerge from a situation where only a couple of countries have debt that would be considered safe.

  • The discussion on debt regimes is not limited to global cross-border flows.

There is a close relationship between this discussion and the debate taking place inside the euro area on debt structure, especially on PSI and the regulatory treatment of sovereign debt held by banks. The core–periphery tensions in the Eurozone largely mirror the tensions that periodically appear in the international monetary system. A failure to reconstruct a euro-wide safe asset would burden the single capital market with a permanent threat of disruption and instablility.

Tradeoffs and choices

International economic policy is all about tradeoffs and choices. Here, the basic choice is between accommodating – or trying to reduce and eliminate – the incompleteness of global capital markets. It would be sound and consistent to institutionalise the framework of capital management, contain current account balances and have state-contingent cross-border debt. Conversely, it would also be consistent to pursue further financial integration and work towards an expansion in the universe of safe assets. In such a universe, the importance of the current account as a policy variable would – and should – be greatly diminished.

Both architectures have their own benefits and costs. Both have different implications on symmetry and the provision of global liquidity. Trying to mix them will sow the seeds of further instability, ultimately risking setbacks in trade and economic liberalisation.


Borio, C. (2015) On the centrality of the current account in international economics Speech to ECB-Central Bank of Turkey conference ‘Balanced and sustainable growth - operationalising the G20 framework’. 28 August.

Brunnermeier, M.K., Langfield, S., Pagano, M., Reis, R., Van Nieuwerburgh, S. and Vayanos, D. (2016) ‘ESBies: Safety in the tranches’, European Systemic Risk Board Working Paper, 21.

Bruno, V. and Shin, H.S. (2015) ‘Capital flows and the risk-taking channel of monetary policy’, Journal of Monetary Economics, 71, pp. 119–132.

Caballero, R.J. and Simsek, A. (2016) ‘A Model of Fickle Capital Flows and Retrenchment: Global Liquidity Creation and Reach for Safety and Yield’, NBER Working Paper, (22751).

Caruana, J. (2012) International monetary policy interactions: Challenges and prospects Speech to the CEMLA-SEACEN conference on ‘The role of central banks in macroeconomic and financial stability: the challenges in an uncertain and volatile world’. 16 November.

Committee on the Global Financial System (2011) ‘Global liquidity – concept, measurement and policy implications’, CGFS Papers, (45).

Haldane, A. (2014) Managing global finance as a system. The Maxwell Fry Annual Global Finance Lecture, Birmingham University. 29 October.

He, D. and McCauley, R.N. (2013) ‘Transmitting global liquidity to East Asia: policy rates, bond yields, currencies and dollar credit’, BIS Working Papers, 431.

Landau, J.-P. (2013) ‘Global Liquidity: Public and Private’, in Global Dimensions of Unconventional Monetary Policy. Jackson Hole: Federal Reserve Bank of Kansas City, pp. 223–259.

Obstfeld, M. (2015) ‘Trilemmas and tradeoffs: living with financial globalisation’, BIS Working Papers, 480.

Turner, P. (2014) ‘The global long-term interest rate, financial risks and policy choices in EMEs’, BIS Working Papers, 441

735 Reads