Debate about Hungary’s euro accession has been revived in domestic policy circles in recent months. I have also made my contribution here (click here for the pdf version), but for the sake of non-Hungarian speakers the main message can be summarised as follows. There are indeed substantial risks involved in giving up monetary autonomy and a flexible exchange rate. Without these important adjustment channels, accommodating asymmetric business cycle shocks or restoring external balance can be much harder and more painful. The alternative to a nominal devaluation in weakening the real exchange rate is the deflationary process of “internal devaluation” (suppressing domestic prices and wages) which raises the real burden of debts and comes with weak demand and high unemployment. Another concern is that less developed economies (ones with a higher equilibrium rate of growth) might require a higher equilibrium real interest rate which, together with their Balassa-Samuelson style inflation surplus and the nominal interest rate convergence within a monetary union, might cause overheating and credit bubbles in these countries, leading to a loss of competitiveness, to external indebtedness and threatening financial stability. A third problem has to do with the incomplete design of the institutional setup of the euro, to the extent that member states in effect issue government debt in a currency over which they do not have control, and which makes them subject to default risk and the possibility of self-fulfilling confidence crises . This also limits the room for national fiscal policies to be accommodative in times of recession. In summary, the euro can amplify the costs of economic policy mistakes and its less flexible framework makes crisis management more difficult – so the risks are potentially large. But…
… but in that op-ed I argued that these risks are either reasonably small for Hungary, or can be managed and mitigated by appropriate economic policy. As one of the world’s most open economies, Hungary is deeply integrated with the euro area, having reasonably synchronised business cycles which makes the probability of asymmetric shocks low. Should they happen despite this, a flexible fiscal policy can step in for the lack of monetary adjustment. By behaving responsibly in good times (perhaps facilitated by credible rules or fiscal watchdogs), the government budget should have room for expansion in a crisis – just as it could cool down an overheating economy. Another remedy can be enhanced macroprudential regulation which can stop the build-up of financial imbalances and credit bubbles. Third, keeping real wage growth in line with productivity could prevent competitiveness loss, ballooning current account deficits and the subsequent need for painful internal devaluation. In addition to prudent fiscal and incomes policies a more flexible labour market can also be of help in this (although it requires much harder structural reforms). In other words, adopting the euro (or any fixed exchange rate regime) requires more disciplined economic policy, but with them it can also bring more stability and can be a success story. All the more so, since these policy prescriptions are in the interest of a healthy economy irrespective of euro accession.
Whether it is worth joining the euro for a new country, however, does not only depend on its own policies but also on how well the euro zone functions. What I wanted to elaborate further on in this post, has more to do with the institutional setup of the euro rather than domestic policy prescriptions. In the above article I highlighted two areas: one is that the ECB has to act as lender of last resort also for national governments (to rule out the possibility of self-fulfilling creditor runs and mitigate default risk), and the other is that surplus countries should bear a larger burden of the balance-of-payments adjustment process, otherwise the monetary union will have a bias towards weak demand and deflation (as pointed out already by Keynes). Start with the latter one (the first will be addressed in another post).
Large current account surpluses in a country reflect excess savings over investments, which savings flow to deficit countries. As long as interest rates and exchange rates can adjust, these extra savings do not lead to falling demand, since a decrease in global interest rates and a depreciation in the surplus country can keep output at its potential. If the associated cross-border capital flows do not finance productive investments in the deficit countries (which can easily happen as low interest rates fuel risk-taking), then after some point the current account imbalances become unsustainable and the balance-of-payments has to adjust. In theory, with flexible exchange rates a large enough real depreciation in the deficit country (expenditure switching) should be able to reduce the trade deficit without having to constrain domestic spending much (expenditure changing). (Of course, this becomes more problematic when there is foreign currency debt involved, or when capital inflows stop very suddenly after a credit bubble bursts, but let us abstract from this now). In summary, both the build-up and reversal of current account imbalances can be accommodated by interest rate and exchange rate adjustments without severely harming demand. (Have a look at this summary by Olivier Blanchard from the IMF).
This already shows us that problems can arise when either interest or exchange rate adjustment, or both, is constrained. In a liquidity trap, due to the zero lower bound, interest rates are already higher than they should be, and they cannot decrease further to stimulate spending which could soak up the extra savings. External demand can substitute for the lack of domestic spending by basically exporting the savings to abroad in the form of trade surpluses and capital outflows. However, the trading partners of such country will necessarily have to run a trade deficit to absorb those savings, of which they already have too much themselves – and without the possibility to lower interest rates this only exacerbates their liquidity trap. Some of their already weak spending will be now diverted towards foreign goods, further depressing domestic demand and raising unemployment. In other words, during a liquidity trap building up large current account surpluses in a country reduce output in other countries.
Similarly, fixed nominal exchange rates in a monetary union can obstruct the balance-of-payment adjustment process by making expenditure switching, i.e. a change in the relative prices between countries (aka real exchange rate), harder. If the surplus country does not tolerate much higher wage inflation, then a real depreciation in deficit countries is only possible through depressing domestic prices and wages (internal devaluation), which is usually very painful and involves high unemployment and a rise in the real burden of debts. There is a limit to how far this can go, so the bulk of the adjustment will have to be through expenditure changing, i.e. through some combination of the surplus country spending more and the deficit country saving more. Unfortunately, the nature of the problem (debtors being cut off from financing) usually entails that this adjustment is highly asymmetric, with most of the burden being on the deficit country to cut spending in order to balance its books, without being able to rely on some increase in external demand provided by higher spending in the surplus country. This asymmetric process therefore becomes highly recessionary and deflationary, leading to rising unemployment in the deficit country. And this is precisely why Keynes had also proposed a more symmetric BoP adjustment process when they were debating how to design the post-war international monetary system in Bretton Woods. Monetary easing across the currency union could, of course, facilitate this process but if interest rates are up against the zero lower bound (or if the deficit country is too small relative to the whole union, and the rest is unwilling to put up with higher inflation), then the arguments from the previous paragraph prevail again.
Given these considerations, it is not hard to see why persistent German current account surpluses are viewed as a problem in the context of the euro zone, a monetary union in a liquidity trap. Several commentators were expressing this in detail – for that I would guide the interested reader to The Economist (1, 2, 3, 4, 5), Ben Bernanke, Barry Eichengreen, the Financial Times’s Martin Wolf (1, 2, 3, 4), Gavyn Davies (1, 2) or Simon Wren-Lewis (1, 2, 3, 4, 5, 6). They are pushing Germany not only to tolerate higher wage inflation, but also to spend more, e.g. on infrastructure investment. And this takes us back to the prerequisites of a well-functioning euro among which should feature a mechanism which limits excessive current account imbalances on the one hand, and prescribes a more symmetric adjustment mechanism on the other. The European Commission’s Macro Imbalance Procedure is certainly a step in the right direction.
This also puts into new light the narrative about the euro crisis. Europe’s balance-of-payments crisis is usually told as a story about an overheating periphery where too low interest rates fueled a credit bubble (think of housing markets in Spain and Ireland), drove up unit labor costs relative to Germany and the ensuing loss in competitiveness resulted in successive current account deficits which went hand in hand with external indebtedness. The takeaway from these stories is that in the absence of autonomous monetary policy macroprudential regulation is very important in preventing credit bubbles, and that it is crucial to keep wage growth in line with productivity, otherwise it will be very hard to regain competitiveness without the availability of nominal devaluation. While main aspects of this story are true, it often ignores the role German savings played. It was capital outflows from Germany to periphery countries which financed their borrowing, as loosely regulated German banks invested the large German savings in the Spanish property bubble or risky Greek government bonds – instead of investing them at home, or saving less in the first place.
Of course, cross-border capital flows are just the other side of current account imbalances, it is an accounting identity. It is generally hard to say which one is driving which. But there is some evidence which suggests that it was not (just) irresponsible southerners letting their competitiveness deteriorate and going to Germany for loans to finance the ensuing current account deficits, but that an independent increase in German savings were also responsible for fuelling easy borrowing conditions and credit bubbles in the South. These arguments are laid out in a paper by Charles Wyplosz, and also here or here. The Hartz reforms in Germany constituted such a policy mix which in effect redistributed income from workers (who consume the most) to firm owners (who save more), while fiscal policy was also quite restrained. As pointed out here, the increase in Southern unit labor costs relative to Germany was in line with inflation differentials, i.e. their real wages remained in line with productivity. It was rather Germany where stable nominal ULCs meant that real wages are lagging behind productivity and the share of labour from income is declining at the benefit of profits. Viewed against this benchmark, German wage restraint is not so much a sign of prudence, but rather the reason why the average German household is still relatively asset-poor despite Germany being Europe’s strongest economy. This redistribution also explains the increase in German savings.
Also, as argued by Erik Jones, the fact that interest rates in the periphery declined suggests that it was a credit supply shock (flowing in from Germany) rather than a credit demand shock (required by worsening competitiveness and current account deficits), which fuelled excessive borrowing in the South. Within the euro area current account deficits can be financed not only with private capital inflows (which affect interest rates) but also through TARGET2 balances within the Eurosystem (analogously running down FX-reserves). If, instead German banks channelling German savings, a Southern bank provides the loan (created out of thin air) to pay for imports from Germany, then the payment is recorded in the TARGET2 system. But with the exception of Spain, all periphery countries were actually improving their TARGET2 balances in the run-up to the crisis. All this suggests that it was also (or rather?) German savings trying to “force themselves” on the periphery, and not (just) Southerners “pulling in” German loans. The resulting easy credit conditions constituted such a positive demand shock for these countries which can explain the pattern of deteriorating current accounts. As Wyplosz shows in the context of a simple Dornbusch model, observed correlations between real exchange rates, current accounts and output are consistent with a domestic demand shock rather than with some exogenous loss of competitiveness, which was often a central element to previous narratives. Therefore, it seems that competitiveness per se was less important in driving the imbalances as the underlying spending/saving decisions.
To sum up, both in liquidity traps and in a monetary union (especially if a monetary union is in a liquidity trap) there are powerful arguments for why surplus countries should spend more. In addition to highlighting the need for enhanced macroprudential regulation in capital inflow receiving countries, the above story also shows that monitoring competitiveness measures and keeping real wages in line with productivity might not be as important as limiting excess savings, and prescribing a more symmetric adjustment of imbalances.