Dealing fairly with trade imbalances in monetary unions

Introduction

Any t-shirt in your wardrobe has travelled more than most people ever will (Blumberg, 2013). The US cotton it is made of may be spun into yarn in Indonesia and sewn in Bangladesh, then shipped to Europe. Institutional arrangements determine where production takes place and who benefits at each step. Subsidies for cotton farmers make cotton production in the United States competitive despite high wages. Indonesia attracts spinning mills because labour is cheap considering education levels and its stable government. Wages in Bangladesh are the lowest worldwide, and weaving companies can count on the government not to enforce labour protections.

This example illustrates that international trade can create harms as well as benefits. To divide benefits and burdens fairly, trade needs a just institutional framework (Rodrik, 2017). Political philosophers have examined some aspects of the trade system, such as trade agreements, international supply chains, and fair trade. ¹ Yet political philosophers have largely overlooked issues of justice related to trade deficits and surpluses. ²

A trade deficit occurs when a country’s imports exceeds its exports. ³ Trade imbalances can be beneficial for deficit as well as surplus countries. A country may have a deficit because it imports what it needs to produce goods and services it will export in the future. The longer a country runs a deficit, however, the worse off will be future generations. To finance trade imbalances, countries have to borrow, saddling future generations with debt and interest payments. ⁴

Shifts in exchange rates can counteract trade imbalances. The currency of the surplus country will be in higher demand than the currency of the deficit country. If both currencies float freely against each other, the surplus country’s currency will appreciate relative to the deficit country’s currency. As a result, goods and services from the surplus country become more expensive for buyers in the deficit country. Goods and services from the deficit country become more affordable in the surplus country. These trends in prices will lead to a reduction in the trade imbalance.

Trade imbalances pose a special problem of justice within monetary unions, such as the Eurozone. Within monetary unions, exchange rates are fixed at par. Members cannot devalue their currency to reduce their trade deficit. Hence they need to find other ways of making their products cheaper for buyers in surplus countries. Members in deficit only have the option to influence the real terms of trade, for instance by reducing taxes or limiting wage growth.

The implications for justice are profound: states running a deficit may not be able to put in place its preferred and otherwise feasible tax and employment policies because of membership in the monetary union, unless other members adjust their policies in a similar direction. The interdependence of economic policies between members gives surplus countries the power to interfere with economic policy in other member states. Policy-interdependence raises deep questions in political philosophy. Do member states have an obligation to respect the policy preferences of other members? And if so, what are the grounds and limits of these obligations?

I start from a republican perspective to answer these questions (Pettit, 2006a, 2010, 2015; Rigstad, 2011; Thomas, 2015). Republicans conceive of political freedom as freedom from domination. An individual is dominated if it is subject to the arbitrary interference of another. Slaves are paradigmatic case of unfree individuals. People living freely under the rule of law in democratically organised states are the paradigm of not dominated individuals.

Republicans have applied the idea of non-domination to international relations. States owe each other the absence of arbitrary interference in other states’ affairs, as well as the absence of the power to do so. Non-domination between states is valuable from a republican perspective to the extent that it contributes to the freedom of individuals. Non-domination among states is not sufficient to protect individuals from arbitrary interference. The state they live in might still arbitrarily interfere in its members affairs. But non-domination among states is required to protect members from arbitrary interference from the outside.

From a republican perspective, democratic states are particularly owed absence of interference. If members of a state organise themselves democratically, members of other states owe them to respect their political choices. States display respect by not interfering arbitrarily with the democratic choices of other state’s members.

The republican idea of non-domination among states is an attractive starting point for investigating trade imbalances in monetary unions. Joining a monetary union creates tight interdependencies between states. Yet member states remain distinct political communities, expected to pursue the democratic preferences of their members. Non-domination is a promising starting point for navigating the balance between interdependency and independence.

Republicanism suggests regulating a monetary union so as to protect the capacities of its members to govern in line with their member’s choices. In section 1 I turn to the tricky questions of what counts as interference in a world of interdependent states, and what makes interference of one state in the affairs of another arbitrary. I do not attempt to settle this question. Instead, I introduce a range of republican conceptions of what makes interference arbitrary, reflecting three different views of domination. I will argue that if a monetary union lacks mechanisms to address trade imbalances, deficit states may be subject to arbitrary interference on each view of domination. The institutional remedies monetary unions ought to adopt vary however.

In section 2, I explain how trade imbalances lead to policy interdependency in monetary unions. If trade imbalances are unchecked, member states can engineer trade surpluses, for instance by lowering taxes or limiting wage growth. Other members can only rebalance trade by adopting similar economic policies. Adopting such policies may well be at odds with the democratic preferences of their citizens.

In section 3, I apply the three conceptions of arbitrary interference developed in section 1 to trade imbalances in monetary unions. I argue that on each conception, some members have the power to interfere with the economic policy of other members. In the absence of checks, this power to interfere is arbitrary.

In section 4, I present institutional options for dealing with trade imbalances in monetary unions. One option is to reduce trade imbalances by penalising members for running a trade surplus or deficit. Promising proposals follow a version of the Keynes’ plan for an international clearing union. Another option is to reform the way trade deficits are financed. This option aims at neutralising the ability of surplus countries to interfere with deficit countries. A third option is to delegate authority for harmonising economic policies to a Eurozone institution. This option still allows for interference but removes its arbitrary character. Which option protects members best against domination? I will argue that the answer hinges on what makes interference between members arbitrary.

1. Non-domination in trade relationships

What do members of monetary unions owe each other in resolving policy interdependencies due to trade? In the literature on trade justice, different normative standards have been discussed. Aaron James has argued that the gains from trade should be shared equally between trading partners (James, 2012). Mathias Risse and Gabriel Wollner have argued that any non-exploitative division is acceptable (Risse and Wollner, 2014). Neither of these standards speaks to the problem at hand, viz. how to resolve policy interdependencies due to trade relationships. My focus is not on sharing gains of trade, but on how to resolve policy interdependencies.

Two states may not each be able to pursue their preferred economic policies because they are interdependent. How should this interdependency be resolved? I develop a republican response to this question. ⁵ States owe each other the absence of arbitrary interference in other states’ affairs, as well as the absence of the power to do so (Pettit, 2006a, 2010, 2015; Rigstad, 2011; Thomas, 2015). In other words, states owe each other non-domination (Arnold and Harris, 2017; Blunt, 2015; Lovett, 2012; Pettit, 2006b, 2012). Non-domination of states is attractive because it is required to protect the members of these states from arbitrary interference from other states (as well as the threat of such interference). ⁶ Domination of a state by outsiders limits the freedom of its members, by limiting the capacity of the state to confirm to the democratic preferences by its citizens.

This is a particularly attractive ideal in monetary unions such as the Eurozone. Eurozone members are part of the European Union which is a political as well as an economic union. Nonetheless, the European Union is a union of states whose self-determination it seeks to protect. Peoples that govern themselves democratically deserve respect (Ronzoni, 2017). A monetary union of democratic states should be setup to enable its members to display respect. States display respect by protecting the capacities of democracies to govern in line with their citizens’ preferences.

Non-domination matters for resolving many kinds of policy interdependencies. Trade is but one practice that creates such interdependencies. Another example is tax competition due to integrated capital markets (Dietsch, 2015). By lowering tax rates, states may lure taxpayers into their jurisdiction. On pain of seeing their tax base dwindle, other states have to follow suit. As a result, their preferred tax regime may become unfeasible.

There is a common pattern underlying policy interdependencies in tax and trade. Both free trade and free movement of capital put states in scenarios that are only partially cooperative. States have an incentive to cooperate insofar as there are some scenarios each wants to avoid. But policy preferences differ between states. These differences pit states against each other. Yet not every policy preference has the same prospect to prevail. Rather, power resides with the state aiming at the tax rate most attractive to individual taxpayers and corporations. The most attractive tax rate is not zero, because taxpayers also have an interest that public goods are produced that benefit them. However, given the range of options of tax rates and public expenditure currently available, taxpayers empirically often prefer lower tax rates, other things equal. As a result, if a state lowers taxes, other states come under pressure to follow suit. By contrast, states preferring lower taxes come under no pressure to increase taxes if other states do.

A state is free in domain D if its decisions in D are not dominated by another state or agent (Blunt, 2015; Lovett, 2012; Pettit, 2006b, 2012). A state is not dominated in domain D if no other state or agent has the power to interfere arbitrarily with the state’s decision in domain D. While arbitrary interference may occur in any domain, domination matters much more in some. I focus on domination in key areas of a state’s economic policies, including tax and wage policies.

The republican approach to freedom has two notable features. First, domination may occur without actual interference. What matters is power to interfere. The exercise of that power is not needed. Second, not every interference leads to domination. Rather, only interference that occurs arbitrarily qualifies as domination.

These two features make the republican approach appealing to analyse relationships between states. Sovereign states exercise supreme authority in a territory. The sovereignty of a state is compromised not only when others actually interfere, but also if they reserve the power to do so.

In line with the second feature, the nature of the interference matters. It is one thing if the EU commission asks a member state for payments in line with agreed rules of membership. It is quite another if some member blackmails another to make such payments. Both the Commission and the extortionary member may threaten to interfere if payments are not forthcoming. But only the arbitrary character of the latter undermine freedom. ⁷

What makes interference arbitrary? I distinguish three competing conceptions. Running a trade surplus in a monetary union qualifies as arbitrary interference on all three conceptions, as I show in the next section. But which conception you endorse affects which reforms are effective remedies against domination, as I show in section 4.

The three accounts of arbitrariness are Proceduralism, Control Substantivism, and Interest Substantivism. ⁸

According to Proceduralism, power to interfere is arbitrary if it is not reliably constrained by rules or procedures (Lovett, 2012). Any set of rules or procedures, regardless of their content, renders the power to interfere non-arbitrary.

Substantivism differs from Proceduralism in that the rules or procedures have to meet some further requirement. According to Control Substantivism, A’s power to interfere with B is arbitrary if B lacks control over A’s exercise of that power (Pettit, 2012). B’s control needs not amount to the ability to veto the exercise of power by A. For instance, citizens can exercise control by participating in a democratic process to shape rules and procedures. Even citizens overruled by the majority count as free on this conception.

Interest Substantivism differs from Control Substantivism in that the rules or procedures have to track B’s interests (Pettit, 1997). Democratic control is neither necessary nor sufficient to avoid arbitrariness.

I do not take a stand here on the best conception of arbitrariness. Each of the three conceptions captures something important. In the next section, I show that given any of these conceptions, trade imbalances in monetary unions can lead to domination.

In sum, institutional arrangements in monetary unions allow for domination under two conditions. First, if member states have the power to interfere with other member states. Second, if the interference is arbitrary. According to Proceduralism, arbitrariness consists in lack of any constraints. According to Substantivism, the content of the constraints matters. Control Substantivism requires control by the state interfered with. Interest Substantivism requires that constraints track the interests of the state interfered with.

2. The problem with trade imbalances in monetary unions

How do trade imbalances come about? When are they harmful to members of monetary unions? Answering these questions is important to understand whether trade imbalances lead to domination.

I focus on surpluses and deficits with other members of the monetary union. In monetary unions, dealing fairly with trade imbalances poses a special problem of justice. In the absence of a monetary union, states running a trade deficit can make autonomous monetary policy decisions. ⁹ If a deficit country devalues its currency, it makes their exports more attractive. Moreover, it makes imports less attractive. These two effects lead to a reduction in the trade deficit. ¹⁰

But members of monetary unions such as the Eurozone share a common currency. Thus, members cannot use monetary policy to reduce trade imbalances with other members. The trade balance between member states depends on the terms of trade with other member states. In unions with tariffs and low transport costs, what matters are the prices of goods and services produced at home and abroad. Relative prices reflect differences in productivity. But they also reflect economic policy, both domestic and in other member states. Tax rates and wage policy influence relative prices, and thus the terms of trade between member states in monetary unions. The political and economic decisions of members of monetary unions are interdependent.

I call member states exporting more to other member states than they import from other member states surplus countries. Net importers from other members in a monetary union are deficit countries. In the Eurozone, the Netherlands, Germany, and Belgium are surplus countries. Greece and France are deficit countries (Eurostat, 2018).

Trade imbalances can be beneficial for surplus as well as for deficit countries (Caballero et al., 2008; Cooper, 2007). Trade imbalances can be a pension plan for surplus countries, and an investment programme for deficit countries. Ageing populations, such as Germany’s, can prepare by exporting more now. Exports can finance buying shares in companies abroad, providing income when an old workforce will produce less. Deficit countries enjoy access to capital to develop their economy. Such trade imbalances are sustainable. Trade imbalances are sustainable if the mechanism generating them produces the means for deficit countries of servicing the resulting financial obligations (Blanchard and Milesi-Ferretti, 2012; Obstfeld and Rogoff, 2009). In the beneficial case, deficits countries get capital. They can use the returns to service the resulting financial obligations.

I focus here on cases where trade imbalances are unsustainable. ¹¹ Trade imbalances are unsustainable states lack the means of servicing the resulting obligations. In principle and within limits, obligations incurred through trade imbalances can be paid for through other means, such as taxation. Setting this option aside, trade imbalances are unsustainable if deficit countries consume rather than invest surplus imports. Members of monetary unions can create unsustainable trade imbalances by pursuing an export-led growth strategy. Export-led growth strategies aim to grow domestic production by exporting goods and services to other countries.

How does an export-led growth strategy lead to unsustainable trade imbalances? Consider Surplus, a member of a monetary union. Surplus’ economy grows, yet wages are stagnant. Moreover, Surplus’ government prioritises reducing public debt over funding long-term public investment projects. How does Surplus influence the other members of the union? Lower wages tend to push down the prices for the goods produced in Surplus. Lower prices make goods from Surplus more attractive abroad, leading to an increase in exports. But people in Surplus can afford less. As a result, demand in Surplus both for domestic goods and for imports from other member states is feeble. Since Surplus’ government does not spend much either, demand tanks. In effect, other member states buy more goods from Surplus than they otherwise would, and export less to Surplus. If members are equally productive, this dynamic is likely to lead to a trade surplus in Surplus. Elsewhere in the union a corresponding deficit emerges. If deficit states consume surplus imports, they do not build up the means of servicing the deficit. Surplus’ export-led trade surplus is unsustainable.

Trade imbalances are dangerous both for deficit and for surplus countries. Deficit countries need to finance imports, by incurring debt or other financial liabilities. Many countries are highly indebted already, including many members of the Eurozone. High debt increases the risk that firms, individuals, or states go bankrupt (Turner, 2015). The sovereign debt crisis in the Eurozone shows how grave the risks are (Frieden and Walter, 2017; Pisani-Ferry, 2014). The trade balance is a better predictor of whether a Eurozone member required bailout funds than the budget balance or the level of public debt (Bénassy-Quéré, 2017). The trade balance of all Eurozone members who received bailouts was negative before the crisis. By contrast, the budget balance and debt position of countries who received bailouts varied widely.

But trade deficits harm even short of bankruptcy. Deficit countries are liable to pay interest on debt incurred to finance the deficit. Interest payments reduce the budget to pursue economic outcomes preferred by their citizens. If trade deficits are financed by public debt, they reduce the ability of the state to distribute resources. If trade deficits are financed by private debt, the taxable income of individuals and companies diminishes. As a result, tax revenues decrease.

Trade imbalances can be harmful for surplus countries as well. If deficit countries default, the export-led growth model collapses. But there are downsides for deficit countries even short of default. Debt is not the only way of financing trade deficits (OECD, 2017). Financial obligations also take the form of equity in companies in deficit countries. In contrast to debt, the returns from equity vary with economic performance. Trade deficits can depress returns on equity investments, because high debt reduces investment. Deficit countries will also consume less to meet financial obligations, weakening demand for imports from surplus countries.

What can deficit countries do to reduce trade deficits? They could impose tariffs on imports or restrict foreign lending and investment. But these cures will likely be worse than the disease. Reducing barriers to trade is a key goal of economic unions. Other members would likely match new tariffs, undoing their effect on the terms of trade. Members of the World Trade Organisation (WTO) cannot use tariffs to address trade imbalances with specific states. According to the most-favoured-nation-clause, states cannot impose tariffs on specific states. Eurozone members cannot use tariffs or restrict the movement of capital, because this would flout the rules of the single market.

The only option left to deficit countries is to copy the economic policies of Surplus. For instance, deficit countries can improve the terms of trade by slashing taxes or slowing wage growth. But copying Surplus puts all at a disadvantage because Surplus’ strategy relies on strong demand abroad. If deficit countries equally restrain domestic demand, total demand in the union dwindles further. Copying Surplus reduces trade imbalances at the risk of leading the union into recession.

Some commentators see Germany as behaving like Surplus within the Eurozone. Whether this is a fair way of describing Germany’s position is controversial. Some have argued that Germany is on track to close its surplus with other Eurozone members (Di Carlo, 2018). Others point to alternative causes of Germany’s trade surplus (Bundesfinanzministerium, 2017).

But it is irrelevant whether any member actually pursues an export-led growth strategy. What matters is that trade imbalances may come about in the way outlined. Domination persists even if members do not exercise their power of creating unsustainable trade imbalances. The current arrangements in monetary unions do not prevent members from acting like Surplus.

In sum, in monetary unions lack protections against export-led growth strategies leading to trade imbalances. Export-led growth strategies can result in unsustainable trade deficits. Unsustainable deficits reduce the ability of deficit countries to pursue economic policies in line with the democratic preferences of their citizens.

3. Domination in monetary unions

I have argued that by running a trade surplus, member states create trade deficits elsewhere. Trade deficits can threaten the ability of members to put in place their preferred economic policies. Does running a surplus amount to interference with the economic policies of deficit countries? In this section, I argue that it sometimes does. As long as some members have the power to run an export-led growth strategy, they have the power to interfere. Yet the power to interfere is not sufficient for domination. Interference only amounts to domination if it is arbitrary. I discuss whether surplus countries interfere with deficit countries arbitrarily. I argue that they do. Thus running a trade surplus sometimes amounts to domination.

Do surplus countries interfere with deficit country’s economic choices? It may appear that it does not. A trade imbalance is the result of economic transactions between people and companies across borders. In each case, a party from a surplus and a deficit country each have consented to the transaction. No single agent controls the transactions which, in total, constitute a trade imbalance. Rather, some person or company in the deficit country can veto each transaction contributing to the deficit. If none of the transactions is controlled by any single agent, it may seem that nobody controls the resulting trade deficit either. But interference requires control. Thus it would seem that surplus countries do not interfere with the economic choices of deficit countries.

Indeed, no single agent controls any of the transactions contributing to trade deficits. But contrary to the reasoning in the previous paragraph, it does not follow that no single agent controls the emergence of the deficit. In fact, surplus states are well-placed to control trade imbalances. Governments can control wage growth in the public sector. The public sector employs more than a third of the workforce in many countries. Wage levels in the public sector have moreover knock-on effects on private wages. Governments control other factors influencing wages in the private sector as well. This includes minimum wages and labour costs other than wages. For instance, Germany slowed wage growth in the last decade of the last century. In the same period, the German trade surplus with the rest of the Eurozone increased. Wage differences between trading partners influence trade imbalances, because wages influence prices. Relative prices, in turn, drive import and export volumes. Differences in taxes are another important factor. Higher taxes tend to increase imports.

Hence we can grant that for each individual trade, parties in deficit as well as surplus countries must consent. Still people and companies need to make these decisions in circumstances they may not condone qua citizens. The circumstances can be set by surplus countries running export-led growth strategies. These strategies put pressure on the policy choices of deficit countries. In the extreme, deficit countries cannot put in place otherwise feasible economic policies (Bénassy-Quéré, 2017: 202). Both surplus and deficit countries can rebalance trade by influencing wage levels. But if deficit countries lower wages, this will decrease demand, and thus harm deficit as well as surplus countries.

The discussion so far suggests that surplus countries can wield power over deficit states. Surplus countries cannot force people and companies in deficit countries to trade. But they can create circumstances that put pressure on deficit countries to change their economic policies. To be sure, surplus states cannot control how deficit countries respond to deficits. But fiscal, welfare, and wage policies are the remaining areas available to members of monetary unions. If monetary unions do not adopt arrangements to share the burden of reducing trade imbalances, it falls on deficit countries alone (Barry and Tomitova, 2007). As a result, deficit countries may not be able to sustain otherwise feasible economic policies. The power that surplus countries wield over deficit countries thus amounts to interference.

The question remains whether surplus states have the power to interfere arbitrarily. In section 1, I distinguished three conceptions of arbitrary interference. According to Proceduralism, arbitrariness consists in lack of constraints. Monetary unions currently lack institutional arrangements constraining member states in running trade surpluses. Thus interference is arbitrary according to Proceduralism. Both versions of Substantivism further qualify the content of the controls required. Since according to Proceduralism interference is arbitrary, it is arbitrary on both versions of Substantivism too. Hence regardless of the conception of arbitrariness, interference is arbitrary.

Someone might object that surplus states are constrained by the democratic preferences of their citizens. Governments cannot ignore citizens’ preferences on pain of losing the next election. Hence, the argument runs, the exercise of power by surplus states is not arbitrary.

The objection ignores that citizens should respect the sovereignty of citizens of other member states. In monetary unions, institutions must be legitimate on two levels (Bellamy and Weale, 2015). First, governments must be democratically accountable to their citizens. Second, monetary union institutions must respect the democratic sovereignty of each member state. Hence governments cannot appeal the will of their citizens as constraint removing the stain of domination. This constraint operates on the level of accountability within member states. At the level of the monetary union, eligible constraints need to be external to member states.

One might also object that surplus countries do not wield arbitrary power because they are constrained by countries outside the monetary union. A country in surplus within the monetary union may in principle be in deficit with the rest of the world. But this is not currently the case for surplus states in the Eurozone. In general, this scenario is unlikely to unfold in practice. Yet if surplus countries within the union face a deficit outside the union, deficit countries within the union are likely to face an even larger deficit outside the union. In this situation, surplus and deficits countries both benefit from devaluing the union’s currency. The union’s central bank would adjust interest rates to devalue its currency vis à vis the rest of the world.

Another objection starts from the observation that if adoption of certain economic policies amounts to arbitrary interference with other member states, domination between member states is mutual. ¹² After all, it would seem that any member state can adopt those policies. Since it is sufficient for domination that an agent has the power to arbitrarily interfere, member states would mutually dominate each other. But mutual domination seems to be at odds with paradigm cases of domination. After all, in the paradigm case of master and slave the roles are clear and stable. It appears to be a contradiction in terms to have two people being both master and slave to each other.

I grant that trade imbalances in monetary unions allows in principle for the possibility of mutual domination. But that does not pose an objection to the account. Mutual domination is not a contradiction in terms. Recall the definition of domination from section 1: A state is dominated in domain D if another state or agent has the power to interfere arbitrarily with the state’s decision in domain D. There is no contradiction in saying that state B has the power to arbitrarily interfere with state A and vice versa. In fact, the other examples of domination in the economic domain discussed by Pettit seem to allow for mutual domination as well. Consider the example of dumping goods on the world market competing with the dominated state’s products (Pettit, 2015: 54). Two states might well find themselves in a situation where each is dominated by the other in this respect, because each has the power of subsidising the production of goods the other state produces and thereby lowering world market prices for these products. Similarly, two states might both hold significant amounts of the other state’s currency (Pettit, 2015: 54). Each would then have the power to sell off holdings in the other state’s currency.

The paradigm case of the master and the slave misleads us to think that mutual domination is impossible. In Jean Genet’s play Les Bonnes, two maids take turns in subduing the other in elaborate rituals. Relationships of mutual domination are perhaps rare in the personal domain, but there is nothing contradictory about them.

While mutual domination is a possibility in principle, in practice certain members of a monetary union will often find themselves in a much better position to dominate others. ¹³ Consider the Eurozone. Northern member states such as Germany or the Netherlands have long been industrialised. Southern member states such as Greece remain more agricultural. Without the introduction of the Euro, northern currencies would have been in higher demand than southern currencies, corresponding to the higher demand for industrial goods from the north. As a result, a currency issued by Germany alone would command a higher value than the Euro, whereas a currency issued by Greece alone would command a lower value. In effect, the Euro is undervalued from a German perspective, and overvalued from a Greek perspective. As a result, northern members export more to southern members than they would with independent currencies, and vice versa. Moreover, northern countries export more and southern countries less to third countries such as the US or China than they otherwise would. Germany may be able to offset this asymmetry by a generous labour policy and high taxes. But it is hard to see how Germany could have built up a trade deficit with respect to Greece.

Another objection queries the power dynamics between surplus and deficit countries. ¹⁴ I have claimed above that states want to avoid running persistent trade deficits because they will otherwise accrue unsustainable levels of debt. It is for that reason that they are forced to adjust their economic policies in line with the policies on surplus countries. But does this dynamic always hold? Keynes remarked in a memo circulated to the British War Cabinet in 1945: ‘[B]y cunning and kindness, we have persuaded the outside world to lend us upwards of the prodigious total of £3,000 million. The very size of these sterling debts is itself a protection. The old saying holds. Owe your banker £1,000 and you are at his mercy; owe him £1 million and the position is reversed’ (Keynes, 1979: 258). It would be a stretch to claim that northern countries were at the mercy of Greece when the Greek debt crisis hit. In contrast to the UK, the size of Greek debt is small in relation to the size of the economy of the Eurozone. Yet it is plausible that the fact that northern banks were exposed to Greek debt swayed northern member states to agree to several rounds of bailouts. Italy, which also faces a debt crisis, plausibly has substantial leverage based on its debt pile that is much larger in absolute terms than Greece’s.

Generally, the distribution of power to interfere between surplus and deficit countries depends on who the deficit country owes the debt to, and how significant the debt surplus countries hold is relative to the size of their economies. While over-indebtedness always poses a problem for deficit countries, their over-indebtedness sometimes poses a problem for surplus countries as well. If surplus countries are negatively affected by deficit countries defaulting, deficit countries can exercise some pressure on surplus countries to adjust policies so as to reduce trade imbalances.

A final objection starts from the observation that states consent to entering the monetary union. I have discussed above that consent by economic agents to individual transactions which, in the aggregate, constitute a trade imbalance does not amount to consent to domination. But a different objection operates of the level of consent of member states for entering the monetary union. It is well-known that monetary unions across economically varied regions are prone to trade imbalances. While some economists expected swifter convergence of the economies of member states, it was foreseeable for member states that entering the monetary union would create trade imbalances and their associated policy interdependencies. The objection suggests that consent should have some bearing on the question of dealing fairly with trade imbalances in monetary unions.

I concede that consent often makes a normative difference. In particular, many activities that benefit someone still require that person’s consent to be morally permissible. For instance, the patient’s consent is typically required to render medical interventions permissible. But it does not follow that there are no consent-independent considerations that render actions or institutional arrangements unjust. For instance, violating people’s basic human rights is typically wrong even if they consent. My claim is that the requirement to design monetary unions to prevent domination is also consent-independent. This is in line with Pettit’s claim that non-domination in general is not based on consent (Pettit, 1997: 61–63). I agree that states freely entered the European Monetary Union knowing that the rules governing the union allowed for relationships of domination to develop. This is relevant, for instance, when assessing claims to restorative justice. Consent diminishes or even removes any claims deficit countries might otherwise have had to compensation by surplus countries for disadvantages suffered to date. But consent does not diminish the fact that the Eurozone allows for relationships of domination, and that domination is unjust. Hence it does not remove the obligation of member states to reform the rules of the Eurozone.

I conclude that by pursuing an export-led growth strategy, surplus countries can arbitrarily interfere with deficit countries. This is despite the fact that deficit countries sometimes have some power over surplus countries as well by virtue of being indebted to them, if surplus countries fear default by deficit countries. The point remains that running a trade surplus sometimes amounts to domination. Vilifying the behaviour of surplus countries is insufficient to escape domination. Domination persists as long as monetary unions lack institutional protections against arbitrary interference.

4. Institutional remedies for dealing fairly with trade imbalances

This section focuses on how monetary unions can protect members from domination. How can monetary unions prevent surplus countries from arbitrary interference with deficit countries? Institutions can treat the cause by preventing trade imbalances within the monetary union. Or they can treat the symptoms by neutralising the power of surplus countries to interfere. Or else member states can delegate authority over economic policies to a supranational institution.

Treating the cause requires preventing trade imbalances in monetary unions. Monetary unions can incentivise members to rebalance trade by imposing a penalty on running a surplus or deficit. I will return below to a version of Keynes’ proposed international clearings union which pursues this end.

Treating the symptoms aims at checking the power of surplus countries to interfere with deficit countries. Monetary unions can promote alternative ways of financing trade deficits. Debt financing put the burden of rebalancing trade on deficit countries alone. I will return below to alternative proposals to finance trade deficits.

Delegating authority tasks a legitimate supranational institution with rebalancing trade. The institution would set acceptable ranges for economic policies such as tax rates. As part of the Macroeconomic Imbalance Procedure, The EU Commission has begun to watch trade imbalances. I will return below to how the procedure could be strengthened to prevent domination.

A radical way to treat the cause would be not to create monetary unions in the first place. If states have already formed a monetary union, they can be dissolved or broken up into units of economically strongly convergent regions. The Eurozone could be abolished altogether, or split up into a northern and a southern monetary union, until such a time when the economies of both regions have converged (Bird, 2012; Lapavitsas et al., 2011). I leave these options aside here. My focus is on addressing the problem of domination in monetary unions due to trade imbalances short of dissolving them.

One option for treating the cause that leaves monetary unions in place is to incentivise surplus countries to rebalance trade. Keynes made a proposal for an international clearings union to address international imbalances (Hockett, 2006, 2011, 2012; Keynes, 1980). It is part of Keynes’ proposal to rebuild the international economic system after the WWII.

Keynes designed his proposal for a fixed exchange rate regime. Today, large global currencies float against each other. But Keynes’ proposal remains relevant to monetary unions. We can think of monetary unions as very rigid fixed exchange rate regimes. Member currencies always trade at par with the currencies of other members, because they are identical. I sketch a proposal inspired by Keynes’ plan adapted to the Eurozone. Other versions of the proposal would also achieve the aim of incentivising surplus countries to help rebalancing trade (Whyman, 2015). My aim is only to show that Keynes’ proposal can be adapted for monetary unions. ¹⁵

Adapted to the Eurozone, Keynes’ proposal would require monitoring cross-border transactions between members of the Eurozone. Such transactions could be cleared by the European Central Bank. An imbalance above a certain threshold between a member state and all other member states would trigger a penalty procedure. ¹⁶ The procedure kicks in regardless of whether members run a surplus or deficit. Similar to the MIP, member states in imbalance would receive recommendations for policies to reduce the imbalance. But in addition, surplus countries have to pay a penalty into a Eurozone convergence fund. The penalty needs to be large enough to make the incentive effective. For instance, the penalty could amount to 5 per cent of the excess surplus above a certain threshold, capped at 0.5 per cent of GDP. The convergence fund would finance infrastructure investment in deficit countries. Any profits from these investments would be distributed among member states.

What makes adapting Keynes’ plan for the Eurozone compelling is that it places a burden of adjustment onto surplus countries. As a result, the burdens of adjustment are distributed more symmetrical between deficit and surplus countries. Currently, only deficit countries are under pressure to adjust their economic policies to rebalance trade, on pain of rising costs for servicing their mounting debt. ¹⁷ By imposing a penalty on persistent surplus countries, the proposal creates a counterweight to the burden of adjustment falling on deficit countries. The threat of penalty payments gives surplus countries an incentive to reduce their trade surplus that is symmetrical to the incentive on deficit countries.

It is worth noting that this plan would mimic the introduction of a limited fiscal union. ¹⁸ Fiscal unions solve the problem of regional variation in economic structure in many nation states. For instance, in Germany economic imbalances exist between economically stronger regions such as Bavaria and economically weaker regions such as the Saarland. Regular fiscal transfers between the regions make it possible for these regions to use the same currency. If the penalties paid by persistent surplus members were effectively transferred to deficit members, the adapted Keynes’ proposal would go be a partial simulation of a fiscal union. In the extreme, the penalty payments can be calibrated to offset the amount of new debt deficit countries are required to take out to finance the trade deficit with other member states. The Keynes proposal would then be equivalent to a full fiscal union. The burden of adjustment would be shifted entirely onto surplus countries. Other versions resulting in lower transfers to deficit countries are of course also possible, striking a balance between burdens on deficit and on surplus countries.

Another option is to treat the symptoms by changing the way trade deficits are financed. Two dimensions matter: who finances the deficit, and how the deficit is financed. Who finances the deficit determines who bears the burden if a deficit country defaults (Wiedenbrüg, 2018; Wollner, 2018). Are deficits financed by people or companies in the deficit country, in the surplus country, or in other countries? In the first case, deficit countries borrow from their own citizens to finance the deficit. If the deficit country defaults, the loss falls on their own citizens. In the second case, citizens from surplus countries bear the risk of default. In the third case, citizens from third countries do.

How the deficit is financed matters too. If the deficit is financed with debt, creditors expect a set interest rate, and the repayment of the principal at a set date (Barry and Tomitova, 2007). By contrast, equity-like instruments such as shares in companies yield variable returns. The rigid terms of debt make deficit countries carry the burden of rebalancing trade. By contrast, if a trade deficit is financed by an equity-like instrument, investors share the burden.

So-called GDP-indexed bonds and sovereign contingent convertibles (CoCos) incorporate some of the benefits of equity-like instruments into sovereign debt (Barr et al., 2014; Borensztein and Mauro, 2004). GDP-indexed bonds provide a variable return indexed to the GDP of the issuing country. Sovereign CoCos yield a fixed return like normal bonds but provide a more flexible repayment schedule. Their time to maturity increases in case of pre-defined contingencies, for instance a recession in the issuing country. These innovations ease the burden of servicing debt for deficit countries in difficult times. They thereby reduce interference with economic decision making due to trade deficits.

The introduction of innovations in sovereign debt faces challenges. ¹⁹ If sovereign debt becomes less rigid, this introduces risk and monitoring cost which could raise the cost for raising debt (Holmstrom, 2015). Even if prices do not rise prohibitively, creditors as well as debtors may have reasons to shun these instruments. Debtors may be concerned to signal to the market that they are worried about their own creditworthiness. Big investors such as pension or insurance funds may be required to hold a certain amount of safe assets, for which new instruments may not qualify. Creditors may also be concerned about the shallow secondary markets for new debt instruments, and misreporting of GDP in the case of GDP-indexed bonds (Brooke et al., 2013: 14f.).

These challenges explain why innovation in sovereign debt instruments is slow in the making. But there is some reason to be optimistic that these challenges can be overcome. In the mid-2000s, collection action clauses were successfully introduced. Collective action clauses make the agreement of a supermajority of bondholders to a debt restructuring legally binding for all bondholders, including those who oppose a restructuring. They make it much easier to coordinate bondholders. This innovation faced the same objections that CoCos and GDP-indexed bonds have to contend with. Since 2013, all bonds issues by Eurozone member states with maturities exceeding 1 year include a collective action clause. Settling the feasibility of CoCos and GDP-indexed bonds is beyond the scope of this paper. But the success of collective action clauses shows that if the international community coordinates, challenges to innovations in sovereign debt can in principle be overcome.

Yet another option is to delegate authority. The European Commission puts some pressure on surplus countries in the context of the Macroeconomic Imbalance Procedure (MIP). The MIP aims to identify, prevent and address the emergence of harmful macroeconomic imbalances (European Commission, 2017). One element in the procedure is to watch the current account balance of members measured as a percentage of GDP. Members are subject to an Excessive Imbalance Procedure if the 3-year moving average is below −4 per cent or above 6 per cent of GDP. As part of this process, the European Commission makes recommendations to reduce trade imbalances. Germany is currently subject to in depth review partly due to its trade surpluses, which have been found to be excessive (European Commission, 2016). The recommendations to Germany from the review include ‘measures to review corporate taxation and the local trade tax’, ‘measures to reduce the high tax wedge for low-wage earners’, and ‘measures to reduce disincentives to work for second earners’. This example shows that the Commission makes recommendations in policy areas such as wage policy and taxation.

The procedure acknowledges that trade surpluses should be addressed as well as trade deficits. But it has three shortcomings. First, the lower threshold for deficits than for surpluses is unjustified. Surplus countries bear at least as much responsibility for resolving trade imbalances as deficit countries. Surplus countries are also better placed to reduce imbalances, because they can do so without decreasing demand.

Second, the MIP watches trade imbalances with the rest of the world, rather than with other members of the Eurozone. Trade imbalances with the rest of the world matter. The ECB can use monetary policy to address these trade imbalances. But trade imbalances within the Eurozone matter too, as I have argued.

Third, the recommendations made in the course of the in-depth review procedure are not binding. The European Commission has the power to make binding recommendations enforceable by sanctions by triggering the Excessive Imbalance Procedure. Yet this procedure has never been used to date, and there is no automaticity to triggering the procedure should thresholds be crossed. In the meantime, deficit countries are under pressure to address unsustainable trade deficits to keep debt in check. The MIP fails to impose pressure on surplus countries to do their bit in rebalancing trade in the Eurozone.

Addressing domination by delegating authority requires addressing these shortcomings of the MIP. The Commission would need authority to force deficit and surplus countries to adjust economic policies to reduce trade imbalances. A first step could be to make the recommendations of the Excessive Imbalance Procedure binding. European institutions would assume authority over policy areas such as wage and tax policy. Taking this approach would be a step towards a federal union.

Do these three proposals stop domination due to trade imbalances? That depends on the conception of arbitrariness. Innovations in how debt is financed mitigate domination on all three conceptions introduced in section 1. Those proposals are agnostic about the conception of arbitrariness because they reduce the power of surplus countries to interfere with deficit countries. But the options discussed under treating the symptoms do not end domination. They only mitigate domination by reducing the power of surplus countries to interfere.

On different conceptions of arbitrariness, the options discussed under treating the cause and delegating authority fare differently. On the Proceduralist conception, power is arbitrary unless it is constrained by rules or procedures. Treating the cause and delegating authority both meet this requirement.

According to Control Substantivism, members need to have control over whether they are interfered with. If the supranational institution is democratically controlled, delegating authority would meet the requirement. But treating the cause would not. The proposal imposes a penalty on members running a surplus. Deficit countries do not have control over whether surplus countries rebalance or accept the penalty.

According to Interest Substantivism, the constraints must track the interests of members. Treating the cause meets this requirement, because penalties give both surplus and deficit countries incentives to rebalance. The option gives greater weight to the interests of the disadvantaged deficit countries. Whether delegating authority meets the requirement depends on whether the supranational institution protects the interests of all members. Without further constraints, delegating authority is insufficient to address domination on this conception.

Conclusion

Forming a monetary union creates challenges of domination due to trade imbalances. Monetary unions such as the Eurozone currently do not protect members against such domination. Yet monetary unions have several options for mitigating domination due to trade imbalances. Either to reduce imbalances by incentivising surplus countries to rebalance trade. Or to reduce the impact of deficits on the economic sovereignty of members by adopting new forms of financing deficits. Or else to delegate some authority for economic decision making to a supranational institution.

Which of these options address domination depends on how we understand arbitrary interference. Only adopting new ways of financing deficits addresses domination regardless. GDP-indexed bonds and sovereign CoCos reduce the power of surplus countries to interfere. Hence how we understand arbitrariness is irrelevant. But we should not rest all our hopes on innovative debt instruments. These instruments reduce domination, but do not stop it.

The Keynes plan of imposing penalty payments on surplus countries gives weight to the interests of deficit countries. It reduces domination by constraining surplus countries in a way that tracks the interests of deficit countries. But it does not give deficit countries control over the economic policy decisions of surplus countries.

Delegating authority restores democratic control over economic policy to deficit countries. Whether it protects the interests of deficit countries depends on the design of the institution. In any case, this option requires to take a big step towards an integrated federal union.