Summary of 'The Economics of European Integration' by Baldwin and Wyplosz. Chapters of the summary do not correspond to chapters in the book. Chapter 1 to 6 corresponds to chapter 13 to 18 of the book. At the bottom an attachment with tables and figures is linked as pdf

Chapter 1: Essential Macroeconomic Tools

Preliminaries: output and prices

It is the most commonly used measure of economic activity; Gross Domestic Product (GDP). It is often called output, as it captures the total production of an economy, all the goods and services that are available for sale. Although sales and production are not exactly equal – for example some production may be stored as inventory – that difference will be ignored. The GDP simply adds up the value, measured in euros, sterling, francs, etc., of all sales in a country over a given period of time, usually one year, sometimes one quarter. An important aspect of GCP is that it also measures all income earned during the same period. Incomes are equal to sales, because any commercial exchange means that one person spends money and another person receives that money.

A frequent use of GDP is to measure economic growth. GDP normally increases, but for two very different reasons:

• More goods and services are produced;

• Some goods and services become more expensive, while some may become cheaper. This changes the value of the output.

There is a distinction between nominal GDP and real GDP. The previous definition, the result of adding up the value of all sales, is the way nominal GDP is measured. Real GDP is computed by estimating the average increase in prices, which is inflation. Real GDP tends to grow secularly, because more capital is put in place and because innovations make us more productive. But that does not mean that actual GDP always follows its underlying growth trend.

The output gap are uneven periods when actual GDP exceeds trend GDP – these are boom years – followed by periods when the output gap is negative – these are slowdown years. These fluctuations along the trend curve are called business cycles and they are accompanied by sensitive changes in employment, firm profitability and, more generally, individual incomes.

The aggregate demand and supply diagram

In contrast to microeconomics, where demand and supply refer to a particular good’s price and output, in macroeconomics it deals with the average price level and total output, i.e. GDP. To prevent any confusion, it is called aggregate demand (AD) and aggregate supply (AS).

The AS curve is upward sloping. Producers of goods and services keep an eye on their costs, including wages. When growth weakens – the output gap declines – markets shrink and firms reduce employment. Fearful of rising unemployment, workers accept wage moderation. This means that production costs grow less fast and firms slow down their own prices. In the opposite situation, in boom periods, unemployment recedes and workers push for higher wages. Higher costs invariably translate into higher prices. This explains why the AS schedule is upward sloping.

The reason why the AD curve is downward sloping is that higher prices erode the purchasing power of large segments of the population and lead the central bank to adopt a contractionary policy. It also makes domestic goods and services more expensive and therefore less able to withstand international competition on both domestic and foreign markets. More explanation about the downward slope of AD can be found in the next section.

Tighter labour markets result in wage result increases and firms raise prices to cope with higher production costs. If we assume that we started at point A in a situation where the output gap was zero – so that GDP was on trend – at point B GDP is above trend. Obviously, by the very definition of trend, the economy cannot remain indefinitely at point B.

The long run (1): neutrality of money

A key principle of macroeconomics is that, in the long run, money is neutral. This principle asserts that the money stock does not affect real variables such as growth, unemployment, wealth, productivity or competitiveness. Instead, increases in the nominal money supply are absorbed by a proportional increase in the price level. More generally, the evolution of real variables (quantities such as real GDP, employment, exports) is independent of the evolution of nominal variables (values expressed in units of the domestic currency, such as prices, the exchange rate, wages), and conversely.

Money is provided by the banking system under control of the central bank. It is held by the public at large, which needs it to carry out daily transactions. But the public is not interested in the number of zeros that figure on banknotes, it is interested in the purchasing power of money. Suppose that the price level doubles up overnight. The public will need twice as much money to hold the exact same purchasing power. Once money and the price level have both doubled up, we are back to the previous situation in real terms. Nothing has really changed, except that everything is worth twice as much in terms of the currency, which is twice more abundant. This is what money neutrality means.

Suppose that the central bank lets the money stock double up. At the initial price level, the public holds twice as much money as it wishes, so it will get rid of it. The way to get rid of money is to spend it. But if spending increases that strongly, production cannot respond, at least not fully. With demand outstripping the supply of goods and services, prices rise. Eventually, they will have doubled up as well. Because prices do usually jump, this is a slow process. This is why the money neutrality principle is not valid in the short run.

In the long run, we expect real GDP to be unaffected by the increase in money and therefore to return to trend. Put differently, in the long run, we expect the output gap to be zero and the real GDP to be back on trend. This means that, over time, the economy moves from point B to point C in graph 1, where prices have increased in proportion to the money supply. The long run AS schedule captures this result: the inflation spiral that sees wages pushing prices and prices pushing wages must eventually come to an end and the economy must return to its trend growth path.

The long term (2): purchasing power parity

Purchasing power parity (PPP) asserts that the rate of change of the nominal exchange rate between two countries is equal to the difference between the inflation rates in these two countries, the inflation differential:

Exchange rate appreciation = foreign inflation rate – domestic inflation rate
=
Inflation differential

Over the short run, there is no visible link between the exchange rate, money growth and inflation. Indeed, we see that the nominal exchange rate is very volatile, even more than the money stock, while price differentials are much more stable.

An alternative definition of PPP leads us to define the real exchange rate. The nominal exchange rate is the one that is widely discussed and reported in newspapers. We will define it as the price in foreign currency of one unit of the domestic currency, for instance 1.3 dollars for 1 euro or 2 dollars for 1 pound. One function of the exchange rate is to translate the price of goods from one currency to another. When it changes, it effects the country’s international competitiveness. The real exchange rate is designed to provide a measure of competitiveness. It is defined as the ratio of domestic to foreign good prices. But for the comparison to be meaningful, prices must be expressed in the same currency. Consider the real exchange rate of the euro in terms of dollars. The real exchange rate compares the dollar price of a basket of goods made in Europe to the dollar price P* of a basket of goods made in the USA. To that effect, we need to convert into dollars the basket price of domestic goods in euros P.

When the real exchange rate increases – appreciates – domestic goods become more expensive relative to foreign goods and our competitiveness declines. The real exchange rate appreciates when:

• The nominal exchange rate E appreciates;

• The domestic prices P rises faster than foreign prices P*, so that P/P* increases.

The exchange rate E and prices are nominal variables. The PPP and money neutrality principles assert that whatever happens to them does not affect the real exchange rate. Thus PPP implies that the real exchange rate is constant. Year-to-year movements in the nominal exchange rate are mirrored in similar but subdued movements in the real exchange rate. While it is not constant from year to year, the real exchange moves around a reasonable level. The stable long run level is sometimes called the equilibrium real exchange rate. Then the real exchange rate is above equilibrium, it is said to be overvalued, and it is undervalued in the opposite case. Over- and undervaluations are instances of misalignments.

The short run: an IS-LM interpretation

The IS-LM diagram shows how the goods and money market interact in the short run. A key assumption is that the price level is constant. The IS curve is the goods market and the LM curve is the money market.

When the interest rate rises, it discourages borrowing and the spending that loans are meant to finance. This reduces demand for goods and services. Equilibrium in this market requires that supply – and therefore the output gap – adjust to demand and therefore decline when the interest rate rises. This is represented by the downward-sloping IS curve.

The money market brings together the central bank and the country’s financial institutions, chiefly banks. A key function of banks is to grant loans to their customers. Thus banks are in charge of the supply of money. On the demand side, we find the public at large: individuals, firms and the government. The public holds money because it is a means of payment. Its drawback is that it brings no interest. This means that when the interest rate rises, it becomes less appealing to hold money. This is one reason why the curve D, which represents the demand for money, is downwards sloping.

If M is the nominal value of the money stock, we can represent the real money stock as M/P: an increase in the price level P reduces the purchasing power of the existing nominal stock.

At any rate, when banks make loans, they routinely raise their cash holdings. For that purpose, they go to the money market – also called interbank or open market – where they borrow from each other at the interbank interest rate. When, in the aggregate, loans have increased, collectively all banks are short in cash. This scarcity translates into a rising interbank rate. Obviously, when loan activity slows down and there is ample liquidity, the interbank rate will decline.

The interest rate depends on how much cash the central bank is willing to provide. This can be seen in graph 2, where we start from point A. Then imagine that the demand for money increases, for example because an economic expansion is under way and more transactions have to be financed. The demand schedule shifts rightward from D to D’. The central bank may decide not to let the money supply expand. In this case, its behaviour is captured by the vertical supply line S₁. As the public demands more money, banks consider granting more loans and try to find more liquidity in the money market. Since the central bank does not provide the required extra liquidity, the interest rate rises, which discourages would-be borrowers. In the end, we move from point A to point B. Another possibility is that the central bank wants to keep the interest rate unchanged. In this case, the supply curve is the horizontal line S₂ and we move to point C as the central bank provides banks with all the cash that they need.
Except for the case corresponding to point C, we see that an increase in output, which raises money demand by the public, results in higher interest rate.

We now ask what happens when the price rises. We consider that the IS curve, which describes the real side of the economy, remains unaffected. On the other hand, the price increase reduces the purchasing power of the existing money stock. What happens next depends on the central bank reaction. Unless the central bank fully compensates the effect of the price increase, the real money supply is reduced.

There are two broad reasons why demand may vary:

• Changing conditions in the goods market;

• Changes in monetary policy.

The open economy and the interest rate parity condition

If the exchange rate remains unchanged, a higher price level results in a real appreciation and a loss of external competitiveness, which should reduce world demand (and therefore flatten the AD curve).

Borrowing at home and investing abroad means initially converting the borrowed amounts into the foreign currency and, when the arrangements mature, moving funds in the opposite direction. If, in the mean time, the exchange rate changes, the profitability of the operation is altered. If, for instance, the domestic currency appreciates, the value of the foreign investment will decline – this is called a capital loss. If instead the domestic currency depreciates, the investor will benefit from a capital gain. International financial markets are in equilibrium when capital flows of this kind are unnecessary because the returns on domestic and foreign assets are equalized, taking into account likely capital gains or losses. This property of international financial markets is called the interest rate parity condition.
Note that this reasoning assumes that capital can move freely from one country to another. In countries that impose some capital controls, there is no reason for the interest rate parity condition to hold.

Monetary policy and the exchange rate regime

Central banks may, if they wish so, intervene in the foreign exchange markets. When they do, they buy or sell their own currencies against other currencies, chiefly the US dollar and the euro. To that effect, central banks hold sizeable amounts of foreign currencies, called foreign exchange reserves. When it adopts a fixed exchange regime, a central bank commits to keep the exchange rate within the declared band of fluctuation. In order to honour its commitment, the central bank must stand ready to intervene in whatever amount is necessary. If its currency weakens, the central bank buys it back, selling some of its foreign exchange reserves. If the currency strengthens, the central bank sells it and accumulates more foreign exchange reserves.

Chapter 2: The Exchange Rate of Europe

The impossible trinity principle

Monetary policy is lost as an autonomous instrument under a fixed exchange rate regime, while it is fully available when the exchange rate is free to float. The underlying logic behind this result is central to the European integration process. Exchange rate policy, i.e. the choice of an exchange rate regime, is simply the same thing as monetary policy. Choosing one fully determines the other.

The sentence ‘monetary policy is lost’ can be misleading for two reasons:

• A most, the loss only matters in the short run since monetary policy is neutral in the long run;

• The real long term implication of the loss of the monetary instrument is that the inflation rate is no longer established by domestic authorities. This may be highly desirable for inflation-prone countries and it explains many features of the EMU.

One way to escape having to choose between exchange rate stability and monetary policy autonomy is to restrict capital movements. Many of the new EU members only abandoned capital controls upon accession.

Bringing previous results together, we reach a conclusion that is fundamental to grasp the logic behind Europe’s monetary integration. It comes under the code name of impossible trinity; only two of the three following features can be simultaneously in place:

• Full capital mobility

• Autonomous monetary policy

• Fixed exchange rates

Capital controls block the interest rate parity principle. This combination was widespread under the Bretton Woods system – although some countries such as Germany or Switzerland did not enforce capital controls. It also characterized Europe’s early monetary integration efforts that led to the ERM. Most developing countries follow this strategy. What happens when one tries to violate the impossible trinity? The answer is simple: a currency crisis. Sooner or later a speculative attack wipes out the fixed exchange rate arrangement.

Choices

Trivial as it may sound, the choice of an exchange rate regime only matters if the nominal exchange rate has real effects. If monetary policies were fully neutral, even in the short run, it would have no effect other than determining the rate of inflation. PPP would always hold, the real exchange rate would remain constant and the behaviour of the nominal exchange rate would be irrelevant for any practical purpose. Thus we care about the exchange rate regime only to the extent that the real exchange rate is affected by monetary policy, i.e. that money is not neutral in the short run and that the short run is long enough to matter.

From the previous chapter, we already know that money is not neutral in the short run, and that the short run is not that short; in fact, it extends over several years. This is why exchange rate regimes matter. To understand more precisely how, we need to have a good view of why money is not neutral. Non-neutrality arises because prices and wages move slowly; we say that they are ‘sticky’.

Since the euro was introduced, the real exchange rate has been trendless; this is what PPP predicts for the long run. In the short run, from month to month, the movements of the real exchange rate closely mirror those of the nominal rate. This is evidence that, in the short run, prices move far too little to have any impact on the real exchange rate. This also means that nominal exchange rate movements have real effects over months and quarters, up to two or three years, or more, it also means that, over such a horizon, the exchange rate regime matters.

Exchange rate regimes come in all sorts of shapes and forms. Except when the exchange rate is freely floating, all other regimes require choosing a foreign currency to peg on.

Sorts of exchange rate regimes:

• Freely floating: The monetary authorities decline any responsibility for the exchange rate. The rate is freely determined by the markets and can fluctuate by any amount at any moment. Most developed countries let their exchange rates float freely. The main reason for adopting this regime is that it fully preserves the ability to conduct an autonomous monetary policy and to put the central bank in charge of inflation. Another advantage is that it largely protects the economy from foreign demand disturbances.

• Managed floating: This is sometimes called dirty floating and it is a halfway house between a free float and a peg. Central banks that adopt this strategy but their own currency when they consider it too weak, and sell it when they see it as too strong, but they refrain from pursuing any particular exchange rate target. In the end, the authorities are not making any explicit commitment but they are occasionally present on foreign exchange markets with the aim of smoothing short term movements or keeping the exchange rate within limited margins. The margins may be explicit or implicit, fixed or variable.

• Target zones: Target zones imply the choice of wide range within which the exchange rate is allowed to move vis-à-vis a chosen anchor, for example the dollar or the euro. This leaves some room to manoeuvre for both monetary and fiscal policy. The wider the band of fluctuation, the more room available, but also the closer the regime to a free float or to a managed float. Practically, the central bank must intervene – and loose policy independence – when the exchange rate moves towards the edges of the target zone, but it can also intervene at any time it wishes, even if the exchange rate is well within its band of fluctuation.

• Crawling pegs: In a crawling peg regime the authorities declare a central parity and a band of fluctuation around it. The characteristic of this regime is that the central parity and the associated maximum and lower levels are allowed to slide regularly: they crawl. The rate of crawl is sometimes preannounced, sometimes not. The difference between a crawling peg and a target zone is not clear cut, since both involve an acceptable range – margins considered narrow enough to qualify as a pegged arrangement are typically less than ±5 percent around the official parity.

• Fixed and adjustable: The authorities declare an official parity vis-à-vis another currency, chiefly the US dollar or the euro, sometimes vis-à-vis a basket of several currencies. The arrangement normally specifies margins of fluctuations around the central parity. The band of fluctuation allows the central bank not to intervene continuously, and therefore preserves a limited role for monetary policy. It is understood that the central parity may be infrequently changed, a procedure called realignment. The realignment option is useful to face serious disturbances – the logic is akin to the insulation property of floating rates. It is also needed when domestic inflation durably exceeds that of the anchor currency, which eroded external competitiveness. This option also provides some monetary policy autonomy, mainly the ability for the inflation rate to differ from that in the anchor currency country.

• Currency boards: Currency boards are a tight version of fixed exchange rate regimes. Under a pegged regime, monetary policy has to be dedicated wholly to the exchange rate target but, as we saw, the possibility to devalue or revalue and the existence of margins of fluctuations introduce some degree of flexibility. Currency boards are designed to remove this flexibility. In order to ensure that monetary policy is entirely dedicated to support the declared parity, with no margin of fluctuation, the central bank may only issue domestic money when it acquires foreign exchange reserves. If it spends its foreign exchange reserves, the central bank must retire its own currency from circulation and the money supply shrinks.

• Dollarisation / euroisation and currency unions: Yet a stricter regime is to fix the exchange rate irrevocably, which means adopting a foreign currency, hence the term ‘dollarization’ or ‘euroisation.’ Without a domestic currency, there obviously can be no monetary policy whatsoever. This regime is typically adopted by small countries with very weak political institutions. A related case it the adoption by several countries of the same currency, as in a monetary union.

The wide variety of existing exchange rate regimes indicates that there is no universally best solution and that a country may find it advantageous to choose one regime at some point in time and another regime at another point in time. Many considerations come into play:

• Retaining monetary policy autonomy;

• Insulating the economy from foreign disturbances. Exchange rate movements tend to offset these disturbances, but they do so by effecting external competitiveness;

• The impact of exchange rate volatility.

The case of flexible exchange rates rests on three main considerations:

• Flexible exchange rates provide the fast way to adjust relative domestic and foreign costs and prices;

• Exchange rate changes are unavoidable enmeshed with politics. Politics rarely mix harmoniously with economics, so removing the exchange rate from the realm of politics is desirable;

• It is difficult to renounce the convenience of monetary policy autonomy in each and every circumstance. Too often, governments instruct their central banks to act in a way that lead to bruising currency crisis.

In the opposite camp, the case for fixed exchange rates emphasizes the tendency of exchange markets to misbehave as well as cases where exchange rate policy is useful:

• The exchange markets are driven by a short term financial logic, where information about the future is essential but highly imperfect. The result is fads, rumours and herd behaviour, which occasionally provoke panic.

• Even if exchange market gyrations do not result in panic, they provoke large fluctuations. For international traders and investors, these fluctuations are a source of uncertainty which can hurt trade and foreign direct investment.

• Harnessing monetary policy to an exchange rate target introduces discipline since foreign exchange markets are likely to immediately sanction inflationary policy by launching speculative attacks.

• In case of serious shocks, parity realignments are always possible when the exchange rate is explicitly fixed but adjustable. Adopting a parity is not a permanent commitment; it only provides an anchor for monetary policy.

The two corners

The 1990s was a decade of violent currency crises. Europe’s ERM was hit in 1992-93, Latin America flowed in 1995-99, then it was Southeast Asia’s turn in 1997-98 and Russia in 1998. These countries were operating one or another form of a peg, but countries like Hong Kong and Argentina, both with a currency board, escaped the apparently contagious wave. This had made the ‘two-corner’ view popular, according to which the only safe regimes are the extremes ones, free floating or hard pegs, such as currency boards, monetary unions or dollarization. Hard pegs were seen as impregnable, because the central banks have no opportunity to gain to market pressure, even if they wish to do so. Freely floating rates are presumed to be immune from speculative attacks, simple because there is no peg to attack; like soft pillows, they absorb any blow.

The two-corner view holds that, when capital is freely mobile, soft pegs try to combine irreconcilable objectives and are predestined to fail, possibly badly. It is yet another implication of the impossible trinity principle. In a globalized world of full capital mobility, either monetary policy is completely free or it is entirely committed to uphold the chosen peg. The intermediate regimes, called ‘soft pegs’, may be seen as a reasonable compromise between fear of floating and fear of fixing, but they run against the impossible trinity principle. As the world went through a wave of capital liberalization over the 1990s, the two-corner view predicted that the middle ground of soft pegs would hollow out.

In Europe the two-corner view is reasonably well vindicated: fewer fixed-but-adjustable regimes, more hard pegs with the monetary union and some currency boards (Estonia, Lithuania) and some countries (Sweden, the UK) letting their exchanges float freely. Elsewhere, however, there is little support for this view. Freely floating rates, in particular, have made limited headway.

What Europe did

The gold standard remains a fundamental reference because it had a very nice property: it automatically restored a country’s external balance. This property, which got lost when we adopted paper money, is known as Hume’s price-specie mechanism.

The neutrality principle is represented in the upper left panel of the graph above. The upward sloping schedule describes the proportionality between the money stock M and the price level P. In the same panel, we add a horizontal line meant to capture PPP. When all prices are defined in terms of gold, the exchange rate is fixed and simply equal to unity (E=1). Imagine that the price of domestic goods P rises while the price P* of foreign goods remains unchanged. The domestic economy becomes less competitive and must eventually run a current account deficit. The horizontal line corresponds to the price level P at which exports equal imports and the current account is in equilibrium. Above this line, the current account is in deficit, and it is in surplus below the line. Point E represents the external equilibrium where the money stock M is consistent with price level P.

Now consider point A, where the stock of gold money brings about a relative high price level and, therefore, a current account deficit. The country sends more gold abroad to pay for its imports than it receives for its exports. The stock of gold money declines. The balance of payment deteriorates as the money increases (because P rises). Point A in both panels describes a situation of external deficit, which corresponds to money stock M’. The price level declines and the deficit reduces. This process will not stop until point E is reached, the equilibrium.

Now take a look at the lower panel in the graph. When the domestic interest rate is below the rate i* prevailing abroad, it pays to borrow gold at home where interest is low and to ship it abroad for lending at the higher interest rate. The financial account is in equilibrium when the domestic interest rate is the same as it is abroad. Above this line, the financial account is in surplus; below it is in deficit. The overall deficit means that gold is flowing out. As the money supply shrinks, the price level declines and the interest rate rises. The capital flow route is very fast while the trade route is slower.

The automatic return to external balance is the key result of Hume’s mechanism. It meant that the gold standard was inherently stable. All markets work towards eliminating the external imbalance and there is no need for the government to intervene. Note also that there is no monetary policy since the stock of gold money is determined endogenously and there is paper money.

The unhappy inter-war period

The gold exchange market was suspended in 1914 as hostilities disrupted gold shipping. Post-war policy makers were committed to return to the gold standard as soon as practical, but at which exchange rate? Different European countries adopted different strategies, which ended up tearing them apart, economically and politically.

Hoping to retain its traditional leadership in international monetary matters, the UK decided to return the sterling to gold standard at its pre-war parity, ‘to look the dollar in the face’. Since 1914 prices had increased much more in the UK than in the USA, and returning sterling to its pre-war value resulted in overvaluation. With hard peg in place, the only solution was to bring prices back down through deflation; a lengthy and painful process. As the impossible trinity principle suggests, monetary policy autonomy was lost. The result was poor growth, a weak current account, and the erosion of trust in sterling, once considered ‘as good as gold’. The City of London lost ground to New York’s Wall Street. The Bank of England withdrew from the gold standard in 1931, sterling promptly lost 30 per cent of its value with respect to gold and the dollar.

France officially returned to the gold standard in 1928 but, in contrast to the pound, its exchange rate was now undervalued. Over the next few years, France ran surpluses in its balance of payments and the Banque of France accumulated large reserves. When the Great Depression hit, France escaped relatively unscathed. Trouble started when sterling’s 1931 devaluation was followed by many others and France lost its competitiveness. When, under duress, the USA too abandoned the gold standard in 1933 and the dollar was sharply devalued by 40 percent, France and the other countries remaining on the gold standard formed the Gold Bloc to protect their now overvalued currencies. The Great Depression belatedly hit France, which faced speculative attacks, as had the UK ten years earlier.

Germany never considered returning to its pre-war level. Its domestic public debt was huge and massive war reparation had been imposed. As in France, Germany’s post-war inflation was high, but in 1922 it slipped out of control. The result was one of history’s most violent hyperinflations. A new Deutschmark was established in 1924 as part of a successful anti-inflation programme. Like the franc, the mark became overvalued when more and more countries devalued their own currencies. Germany first suspended its debt and then started to move away from a free trade system. Exchange controls were established. As the depression deepened, the Nazis combined public spending with wage and price increases. This further dented external competitiveness deepened the trade deficit. Germany completely bypassed y the foreign exchange market by working out bilateral barter agreements with one country after another. Germany respected the impossible trinity principle, in an extreme way, by severing market-based relationships. Once they had returned to the gold standard’s hard peg, which rules out capital controls, France and the UK would have had to give up monetary autonomy. When the Great Depression hit, the urge to use monetary policy became too strong. The impossible trinity principle was violated and the result was the end of the fixed exchange rate system.

The ensuing round of tit-for-tat depreciations, which was be called beggar-thy-neighbour policies, led nowhere but started to disrupt trade. Protectionist measures soon followed and trade exchanges went into a tail-spin, aggravating the depression. Political instability followed, leading to war.

For Europe, a couple of lessons have been learnt from this traumatic period:

• Freely floating exchange rates result in misalignments that breed trade barriers and eventually undermine prosperity.

• The management of exchange rate parities cannot be left to each country’s discretion.

Even before the end of the Second World War, the USA and the UK started to plan the Bretton Woods conference. The aim was to establish an international monetary system. Gold remained the ultimate source of value, but the only currency directly tied to gold was the dollar. All the other currencies were defined in terms of the dollar. Exchange rates were ‘fixed but adjustable’ to avoid both unreasonable adherence to an outdated parity (over- or undervaluation) and an inter-war-type free-for-all. The system was a collective undertaking with the IMF both supervising compliance and providing emergency assistance. Capital controls were not outlawed and most countries made abundant use of them. This was compatible with the impossible trinity.

The Snake

For Europe the Bretton Woods system had provided a ready-made solution to the exchange rate question. Capital controls allowed some degree of monetary policy autonomy, which eventually was misused. By the late 1960s, inflation started to rise in a number of countries, including the USA. The centre of the system, the US dollar, gradually became overvalued. The relaxation of capital controls in the USA and several countries once again resulted in a violation of the impossible trinity principle. The Bretton Woods system came under strain when the USA could no longer guarantee the gold value of the dollar. The demise of the system came in two steps. First, in 1971, the USA ‘suspended’ the dollar’s gold convertibility. Then, in 1973, the ‘fixed but adjustable’ principle was officially abandoned; each country now was free to choose its exchange rate regime. This effectively ended the Bretton Woods era.

Concerned with the inter-war spectre of over- and undervaluations, the continental countries of Europe promptly resolved to limit exchange rate movements among themselves. The first response was the ‘European Snake’, a regional stepped-down version of the Bretton Woods system designed to limit intra-European exchange rate fluctuations. But the Snake was a very loose arrangement. It did not deal with the impossible trinity principle; capital controls were often in place but they were not tight and increasingly evaded, and there was no restriction on national monetary policies. When inflation rose in the wake of the first oil shock of 1973-74, the central banks reacted differently. Some succeeded in keeping inflation in check, whereas others did not. Maintaining exchange rate fixity under such conditions was hopeless and, indeed, several countries had to leave the Snake arrangement.

In spite of its eventual failure, the Snake brought about two innovations that shaped subsequent integration efforts. First, it embodied the determination to keep intra-European rates fixed, irrespective of what happened elsewhere in the world. Second, the gold standard and the Bretton Woods system were gone, ant with them external reference values for European currencies.

The European Monetary System

The heart of the European Monetary System (EMS) is the Exchange Rate Mechanism (ERM), a system of jointly managed fixed and adjustable exchange rates backed by mutual support. When the euro was launched in 1999, the countries that gave up their national currencies left the ERM. A new ERM was then designed and called ERM-2. It now mainly serves as an entry point into the monetary union. During the first ten years, as national monetary policies remained autonomous, inflation rates diverged markedly. With fixed nominal exchanges rates, the result was chronic misalignments. The implicit rule was to observe inflation rates since the previous realignments and change the parities according to PPP. This process was a little too transparent. Exchange markets could easily foresee the next realignment and speculate accordingly. As a result, most parity adjustments occurred in the midst of serious market turmoil, calling into question the sustainability of the ERM. The answer was an informal application of the impossible trinity principle. ERM member countries vowed their inflation differentials. Germany, the largest country with the lowest rate of inflation, became the example to follow. Its central bank, the Bundesbank, gradually emerged as the centre of the system.

Capital controls, in place in most countries since 1945, were formally dismantled by 1990 as part of the Single Act. With speculative flows now unfettered, the ERM was doomed. Yet, remembering the inter-war instability, most European countries were deeply attached to intra-European exchange rate stability. The informal solution was Bundesbank leadership. Replacing the Bundesbank with a common central bank would allow the other central banks to recover some influence over what had already became de facto a common monetary policy. Initially and understandably reluctant, the German government decided to back the project, mostly on political grounds

The Maastricht Council decided upon the replacement of the European Community with the European Union, and included a precise schedule to establish the monetary union. On 4 January 1999, the exchange rates of 11 countries were ‘irrevocably’ frozen.

Chapter 3: Theory of the Optimum Currency Area

The logic of the optimum currency area theory is symbolized in the graph below.

Since the usefulness of a currency grows with the size of the area over which it is being used, its marginal benefit is positive. Yet, it is declining as the area expands, because the extra benefit of adding one more country to an already large currency area is smaller than if the initial area were small. If the marginal benefit is always positive, then would the world be the optimal currency area? It would be if there were no costs. What can these costs be? As a currency area grows larger, it becomes more diverse – in standards of living, for instance. If more diversity means more costs when sharing a common currency, the marginal costs are positive and rising with the size of the area. There are many ways in which diversity matters – some are economic, some are political. Diversity is costly because a common currency requires a single central bank, and a single monetary authority is unable to react to each and every local particularity. The optimum currency area (OCA) theory takes the benefits as obvious and aims at identifying these costs more precisely. The basic idea is that diversity translates into asymmetric shocks and that the exchange rate is very useful for dealing with these shocks.

Adverse shocks

Imagine that the world demand for a country’s exports declines, because tastes change, or because cheaper alternatives are developed elsewhere. This opens up a hole in the trade balance. To re-establish its external balance, the country needs to make its exports cheaper. A depreciation will do the trick if the country has its own currency. If, however, the country is part of a wider currency area, there is no alternative to lowering prices.

The outcome is more painful if the exchange rate is fixed and prices are rigid. In that case, the economy moves to point C, in the graph below, where the output decline is even deeper. At the unchanged real exchange rate λ, domestic producers continue to supply the output corresponding to point A, but point C represents the new, lower, demand. The distance AC represents unsold goods. Obviously, domestic firms will not accumulate unsold goods forever. Something has to give and production will fall. The recession generates incentives to gradually cut prices, eventually bringing the economy to point B. But this is likely to be the outcome of a painful and protracted process. The example illustrates why exchange rate fixity, when combined with sticky prices, makes an already bad situation worse. In a monetary union, instead of a simple once-and-for-all change in the nominal exchange rate, a real exchange rate adjustment can only come from changes in prices and wages. If prices and wages are sticky, the adjustment can take time, creating hardship along the way.

Asymmetric shocks

So far, we have thought of a country in isolation to set the stage for studying the key insight of the OCA theory: diversity means that different countries face different shocks. The simplest case is a currency area with two member countries. If countries A and B are hit by the same adverse shock, we know from the previous section that both have to undergo a real depreciation vis-à-vis the rest of the world. If they are similar enough, to a first approximation, there is no need for their bilateral (nominal and real) exchange rate vis-à-vis the rest of the world to change. The situation is very different, however, in the presence of an asymmetric shock. Assume, for instance, that country A is hit by an adverse shock, but not country B. Country A must now undergo a real deprecation vis-à-vis both country B and the rest of the world.

With asymmetric shock, monetary union membership becomes seriously constraining. What happens then? The situation is examined in the graph below. The vertical axis measures each country’s real exchange rate vis-à-vis the rest of the world: EP/P* and EP_B/P*, where P_A and P_B are the price indices in country A and country B, respectively, P* the price level in the rest of the world, and E is the common currency’s exchange rate, initially equal to E₀. Points A in both panels represents the initially balanced situation, with the same real exchange rate λ₀ in both countries: λ₀ = E₀/P_A/P* = E₀/P_B/P*. Prices are assumed to be sticky – otherwise, the exchange rate regime does not matter.

The common central bank must make a choice on behalf of country A and country B. If it cares only about country A, it depreciates the common exchange rate to E₁. With sticky prices, both countries must share the same real exchange rate λ₁. The graph shows that this is not good for country B, which now faces a situation of potentially inflationary excess demand (represented by the distance B’B’’). If the central bank favours country B, it will keep the common exchange rate unchanged. Both countries retain the initial real exchange rate λ₀, and stay at the initial point A. This suits country B well, which does not face any disturbance, but it means excess supply for country A (represented by the distance A’A). Clearly, in the presence of an asymmetric shock, what suits one country hurts the other.

If the union’s common external exchange rate floats freely, it will depreciate because of the adverse shock in one part of the area, but not all the way to E₁. It will decline to an intermediate level such as E₂, to which corresponds a real exchange rate λ_2. The outcome is a combination of excess supply in country A and an excess demand in country B. Both countries are in disequilibrium. Disequilibria cannot last forever. Over time, prices are flexible and will do what they are expected to do. Recession and disinflation in country A, boom and inflation in country B: these are the costs of operating a monetary union when an asymmetric shock occurs.

There are many reasons why no two countries react the exact same way to the same shock. Their differing reactions may be due to their different social-economic structures, including labour market regulations and traditions, the relative importance of industrial sectors, the role of the financial and banking sectors, the country’s external indebtedness, the ability to strike agreements between firms, trade unions and the government, and so on.

Another asymmetry concerns the ways monetary policy operates. When a common central bank reacts to a symmetric shock, it is not a foregone conclusion that the effect of its action will be the same throughout the currency union. Differences in the structure of banking and financial markets or in the size of firms – and their ability to borrow – may result in asymmetric effects.

The optimum currency area criteria

There are three classic economic criteria and an additional three which are political. The first criterion looks at a way of minimizing the costs of an asymmetric shock within a currency area. The next two economic criteria take a different approach: they aim at identifying which economic areas are likely to be hit by asymmetric shocks infrequently or moderately enough to be of limited concern. The last three criteria deal with political aspects; they ask whether different countries are likely to help each other when faced with asymmetric shocks.

Labour mobility (Mundell)

The first criterion was proposed by Robert Mundell when he first formulated the notion of an OCA. The idea is that the cost of sharing the same currency would be eliminated if the factors of production, capital and labour were fully mobile across borders. Since it is conventionally assumed that capital is mobile, the real hurdle comes from the lack of labour mobility.

1. The Mundell criterion: Optimum currency areas are those which people easily move within.

The Mundell criterion is symbolized in the graph below

The adversely affected country A undergoes unemployment while non-affected country B faces inflationary pressure. Both problems could be solved by a shift of the production factors (labour and capital) which are idle in country A to country B, where they are in short supply. This reallocation is shown as a shift of both countries’ supply schedules to AS’. This reallocation changes trend GDPs so that the output gap is zero at both equilibrium points C. What is remarkable is that there is no need for prices and wages to change in either country. Once the factors of production have moved, the currency area’s nominal exchange rate E₂ delivers the real exchange rate λ₂ that is best for each country.

A few words of caution are warranted. First, across borders, not only do cultural and linguistic differences restrain migration, but institutional barriers further discourage labour mobility. Changes in legislation may make cross-border labour mobility easier and enlarge the size of optimum currency areas and, indeed, this is part of Europe’s quest for closer integration. Second, the goods produced in country A may differ from those produced in country B. It may take quite some time to retrain workers form country A to produce the goods of country, if at all possible. If the shocks are temporary, it may not be worth the trouble of moving, retraining, etc. Finally, labour needs equipment to be productive. Financial capital can move freely and quickly, unless impeded by exchange controls, but installed physical capital (means of production such as plants and equipment) is not mobile. Creating new production facilities may take months, if not years.

Production diversification (Kenen)

If substantial asymmetric shocks only rarely happen, the overall costs are episodic, while the benefits accrue every day. The Kenen criterion takes a first look at this question by asking what the most likely sources of substantial shocks are. Most of the shocks likely to be permanent are associated with shifts in spending patterns. Such shocks actually occur continuously, but most of them are hardly noticed outside the affected industries. To create a problem for the monetary union, a shock must be large and asymmetric.

The countries most likely to be affected by severe shocks are those that specialize in the production of a narrow range of goods. Conversely, a country that produces a wide range of products will be hardly affected by shocks that concern any particular good, because that good relatively weighs little within the total production. This explains the second criterion for an optimum currency area, initially stated by Kenen: in order to reduce the likelihood of asymmetric shocks, currency area member countries ought to be well diversified and to produce similar goods.

2. Kenen criterion: Countries whose production and exports are widely diversified and of similar structure from an optimum currency area.

Openness (McKinnon)

The next relevant question is whether the exchange rate is at all helpful in the presence of an asymmetric shock. If not, little is lost by giving it up. The distinction between ‘domestic’ and ‘foreign’ goods refers to where the goods are produced. However, many standard goods, such as paper sheets or electric bulbs, are produced in different countries, but they are virtually identical. In that case, trade competition will ensure that their prices are the same everywhere, or nearly so, and therefore largely independent of the exchange rate. The point is that the exchange rate does not affect competitiveness, because competition forces prices to be the same. The third OCA criterion, initially formulated by McKinnon, recognizes that when the economy is small and very open to trade, it has little ability to change the prices of its goods on the international markets. In that case, giving up the exchange rate does not entail much of a loss, at least for moderate shocks.

3. McKinnon criterion: countries which are very open to trade and trade heavily with each other form an optimum currency area.
    

4. Another OCA criterion is called the ‘Transfer Criterion’: Countries that agree to compensate each other for adverse shocks form an optimum currency area.

In practice, a ‘best way’ to deal with a shock rarely exists. The resulting decision of the central bank will probably not be the same across different countries because national preferences are not necessarily homogeneous. Whatever the central bank chooses to do will be controversial and will leave some, possibly all, countries unhappy. At best, there will be resentment, at worst the currency union may not survive. The collective preference, which shapes the policy response, thus intimately depends on domestic politics, and there is no reason for all countries of a certain currency area to share the same balance of political forces. The fifth criterion states that these differences should not be too wide.

5. Homogeneity of preferences criterion: Currency union member countries must share a wide consensus on the way to deal with shocks.

Since none of the previous criteria are likely to be fully satisfied, no currency area is ever optimum. When separate countries contemplate the formation of a currency area, they need to realize that there will be times when there will be disagreements and that these disagreements may follow national lines, especially if the shocks are asymmetric or produce asymmetric effects. For such disagreements to be tolerated, the people that form the currency union must accept that they will be living together and extend their sense of solidarity to the whole union. In short, they must have a shared sense of common destiny that outweighs the nationalist tendencies that would otherwise call for intransigent reactions.

6. Solidarity criterion: When the common monetary policy gives rise to conflicts of national interests, the countries that form a currency area need to accept the costs in the name of a common destiny.

To sum up, there are six criteria of the OCA;

• The Mundell criterion;

• The Kenen criterion;

• The McKinnon criterion;

• The transfer criterion;

• The homogeneity of preferences criterion;

• The solidarity criterion.

Is Europe an optimum currency area?

The OCA theory emphasizes the role of asymmetric shocks, so a natural starting point is to ask whether asymmetric shocks happen often enough, and are large enough, to be of serious concern. Of course, we do not know what the future has in store. Best is to assume that the past can be a guide for the future.

Openness matters in the OCA theory, because in a small open economy most of the produced and consumed goods are traded on international markets. Accordingly, their prices on the local market are largely independent of local conditions and any change in the value of the currency tends to be promptly passed into domestic prices. When this is the case, exchange rate changes fail to affect the country’s competitiveness and are, hence, essentially useless, which is exactly the McKinnon criterion.

Openness is defined as the share of economic activity devoted to international trade. The ratio of exports to GDP measures the proportion of domestic production that is exported. The ratio of imports to GDP measures the proportion of domestic spending that falls on imports. Most European countries are very open, the more so the smaller they are, which explains why the smaller countries have traditionally been the most enthusiastic supporters of the monetary union. This applies to both old and new EU member countries. Another measure of trade openness looks at the importance of bilateral trade links between each country and the centre country.

The EU countries are relatively less open to immigration than similarly developed countries, with Luxembourg being an exception. Moreover, citizens from other EU countries form a small proportion of immigrants. Europeans seem to take little advantage of the single market which allows them to work and settle anywhere in the EU. Low migration by European nationals could be compensated by immigration from outside the EU. If immigrant workers were to move to where job offers exceed supply, some of the costs of a monetary union would be reduced. Even viewed this way, immigration – a big political issue in Europe – is relatively limited in Europe.

How deep is the European sentiment of solidarity? Put differently, how much are citizens willing to give up parts of national sovereignty in the pursuit of common interest? There is no simple, uncontroversial way to measure the willingness of European citizens. However, looking at the breakdown by country, there is a tendency amongst citizens of the new member states to be more willing to rely on joint decision making than the Nordic countries. They may reflect trust in national decision making. Undoubtedly, the poorer countries are also more interested in joint welfare policies because they expect to benefit financial, yet the reluctance is widely shared. With few exceptions, nationalism does not exert a powerful influence.

In the end, most European countries do well on openness and diversification, two of the three classic economic OCA criteria, and fail on the third one; labour mobility. Europe also fails on fiscal transfers, with an unclear verdict on the remaining two political criteria.

Will Europe become an optimum currency area?

Many European policy makers strongly believe that stable exchange rates promote trade integration, which enhances fulfilment of the McKinnon criterion. The reasoning is that a common currency reduces the costs of buying and selling goods and services across borders. Not only can firms avoid going through selling and buying different currencies, possibly keeping accounts in different currencies, but they also avoid the risk that the exchange rate will move and erode profits. In addition, using the same currency facilitates comparisons between various products, which should increase competition. All in all, the presumption is that adopting the euro should boost trade within the Eurozone and thus bring it closer to an OCA.

What effects may trade integration have on diversification (the Kenen criterion)? The evidence is open to debate. On one side, it has been argued that trade leads to more specialization as each country or region focuses on its comparative advantage. Trade takes the inter-industry form, whereby exports and imports correspond to different goods. This would go against the diversification criterion and makes the monetary union more costly as time goes by. On the other side, it is argued that, among developed countries, integration leads to intra-industry trade: exports and imports include similar goods. Every country produces the whole range of goods, simply with different brands, offering customers more choice. In the process, trade becomes more diversified.

European labour mobility is low and few expect it to increase dramatically in the near future. An alternative to mobility is flexibility, and the argument runs as follows: European labour markets are noticeably less flexible than their US counterparts. For example, in the US, firms are quite free to fire workers when economic conditions worsen, whereas in Europe firing is costly, because of heavy severance pay and numerous regulations. Europeans frown on US harshness, but the result is that unemployment is generally higher in and longer lasting in Europe. The question is whether the adoption of a common currency will change that. One possibility is that the single currency increases the costs of the ‘European way’ and reduces opposition to measures that aim at flexing the labour markets. If each country had their own currency, workers were advocating using monetary policy and the exchange rate to boost the economy. This is now impossible, at least at national level, and there is no pan-European trade union to date. In addition, with prices set in euros across the Union, there is increasing transparency in goods markets, which should benefit countries where labour markets are more flexible. Thus, it is believed that economic competition will indirectly lead to competition among the welfare programmes and this will shift the trade-off between economic performance and labour protection.

We have reached two important conclusions. First, Europe is not exactly an optimum currency area; it does well on some, but not all of the criteria. Second, it is not just labour mobility that is insufficient, but -more generally- the labour markets that display significant rigidity, especially in the large countries. In these countries, the monetary union may worsen an already painful situation of high unemployment.

Chapter 4: The European Monetary Union

The Maastricht Treaty

In December 1991, the 12 heads of state and governments of the EU gathered in Maastricht to sign a treaty that replaced the European Community (EC) with the European Union (EU). The change of name was symbolic. It was meant to signal that the treaty was not just about economics, but also included political considerations. Two new pillars – foreign and defence policies, justice and internal security - were added to the first, economic pillar. The power of the European Parliament was enhanced and it was also agreed to substitute ‘qualified majority’ for ‘unanimity’ for Council decisions on a number of issues. Many of these ambitious-looking steps were incomplete, which called for the subsequent treaties of Amsterdam, Nice, and Lisbon. The monetary union part of the Treaty, however, was fully worked out. It included an irrevocable decision to adopt a single currency by 1 January 1999. The Treaty described in great detail how the system would work, including the statutes of the European Central Bank (ECB) and the conditions under which monetary union would start.

In order to join the monetary union, a country has to fulfil the following five convergence criteria, which remain applicable to all future candidate countries:

1. The first criterion deals directly with inflation. To be eligible for monetary union membership, a country’s inflation rate should not exceed the average of the three lowest inflation rates achieved by the EU Member States by more than 1.5 percentage points.

2. In order to weed out potential cheaters, a second criterion requires that the long term interest rate should not exceed the average rates observed in the three lowest inflation rate countries by more than two percentage points. The reasoning is shrewd. Long term interest rates mostly reflect markets’ assessment of long term inflation. Achieving a low long term interest rate therefore requires convincing naturally sceptical financial markets that inflation would remain low ‘for ever’.

3. A country must have demonstrated its ability to keep its exchange rate tied to its future monetary union partner currencies. The requirement is therefore that every country must have taken part in the ERM for at least two years without having to devalue its currency.

4. Budget deficits should not exceed three percent of GDP.

5. The maximum lever for public debt is sixty percent of the country’s GDP. However the criterion was couched in prudent terms, calling for a debt to GDP ratio either less than sixty percent or ‘moving in that direction’.

An important aspect of the Maastricht Treaty is that it, for the first time, introduced the idea that a major integration move could leave some countries out. Initially the intention was to protect price stability and not for the inflation wolf to enter to the den. Then things turned out differently. Prime Minister Thatcher’s Britain was firmly opposed to the monetary union. For a while, Thatcher stonewalled the discussions, which went ahead without her. The view in London was that this was a bizarre idea with no future at all, a sort of conventional exercise that no one really intended to bring to fruition. The UK found itself cut off from a major negotiation that would powerfully shape Europe’s future. Partly because of her unwillingness to engage her partners on the issue, Thatcher was dismissed and replaced by John Major. Unable to scuttle the project, the best he could achieve was to obtain an opt-out clause, which stated that the UK, alone, was not bound to join the monetary union. This further confirmed that Europe could move at different speeds. A similar opt-out clause was given to Denmark after a first rejection of the Treaty by Danish voters. De facto, Sweden is treated as Denmark, with the right to decide when to apply for membership to the Eurozone.

The Eurosystem

No matter how daring the founding fathers of the EMU were, they stopped short of merging the national central banks into a single institution, partly for fear of having to dismiss thousands of employees. In fact, national central banks have hardly downsized their staff. The solution was inspired by federal states, like Germany and the USA, where regional central banks coexist with the federal central bank. The newly created ECB coexists with the national central banks, one of which did not even exist prior to 1999 since Luxembourg, long part of a monetary union with Belgium, only established its own central bank to conform to the new arrangement. On the other hand, in many respects the euro was meant to be a continuation of Europe’s most successful currency, the Deutschmark. The structure of the Bundesbank was used as a blueprint for the monetary union. This inspiration is visible not only in the new institution, but also in the policy objective and framework.

The European System of Central Banks (ESCB) is composed of the new, especially created ECB and the national central banks (NCBs) of all EU member states. Since not all EU countries have joined the monetary union, a different term, Eurosystem, has been coined to refer to the ECB and the participating NCGs. The Eurosystem implements the monetary policy of the Eurozone. If needed, it also conducts foreign exchange operations, in agreement with the Finance Ministers of the member countries. If holds and manages the official foreign reserves of the EMU member states. It monitors the payment systems and it is involved in the prudential supervision of credit institutions and the financial system.

The ECB is run by an Executive Board of six members, appointed by the heads of state or governments of the countries that have joined the monetary union, following consultation of the European Parliament and the Governing Council of the ESCB. It comprises the six members of the Executive Board and the governors of the NCGs of monetary union member states. The Governing Council is the key authority deciding on monetary policy. Its decisions are, in principle, taken by majority voting, with each member holding one vote, although it seems to operate by consensus. A transitory body, the General Council, includes the members of the Governing Council and the governors of the NCGs of the countries that have not joined the monetary union. The General Council is essentially fulfilling a liaison role and has no authority. While decisions are taken by the Governing Council, the ECB plays an important role. Its president presides over the meeting of the Governing Council and reports its decisions at press conferences. The ECB prepares the meetings of the Governing Council and implements its decisions. It also gives instruction to the NCBs to carry out the common monetary policy. An important characteristic of the ECB is that its Executive Board members are not representing any country: they are appointed as individuals.

The size of the Governing Council is seen by external observers as a source of difficulty. With 6 + N members, where N is the number of countries that have joined the monetary union, the Council is large. It will get larger as new members join the union by the end of the decade, although some changes will be introduced to ensure that the decision-making committee’s size is, and remains, efficient.

Objectives, instruments, and strategy

The Maastricht Treaty does not give an exact definition of price stability. The Eurosystem has chosen to first interpret it as follows: ‘Price stability is defined as a year-on-year increase in the Harmonized Index of Consumer Prices (HICP) for the Eurozone of below two percent. Price stability is to be maintained over the medium term.’ In 2003, with fears of deflation rising, ‘the Government Council agreed that in the pursuit of price stability it will aim to maintain inflation rates close to two percent over the medium term’. Thus, while many central banks typically announce an admissible range for inflation, the Eurosystem only indicates an imprecise target. Neither does it specify the meaning of the ‘medium term’.

Like most other central banks, the Eurosystem uses the short term interest rate to conduct monetary policy. The reason is that very short term assets – 24 hours or less – are very close to cash. As central banks have a monopoly on the supply of cash, they can control very short-term rates. On the other hand, longer term financial instruments can be supplied by both the public and private sectors, making it nearly impossible for central banks to dominate the market and control the rate. In fact, longer term rates incorporate market expectations of future inflation and future policy actions. These expectations are beyond control of the central bank, and therefore long-term interest rates cannot be steered with any degree of precision. By concentrating on short-term rates, central banks can achieve greater precision. The problem is that central banks control the short maturity, whereas it is the long-term interest rate that effects the economy, because household and firms borrow for relatively long periods, typically form one to twenty years or more.

The Eurosystem focuses on the overnight rate EONIA (European Over Night Index Average); a weighed average of overnight lending transactions in the Eurozone’s interbank market. Control over EONIA is achieved in two ways;

• The Eurosystem created a ceiling and a floor for EONIA by maintaining open lending and deposit facilities at pre-announced interest rates. The marginal lending facility means that banks can always borrow directly from the ECB at the corresponding rate.

• The Eurosystem conducts, usually weekly, auctions at a rate that it chooses.

The first pillar is what the Eurosystem calls ‘economic analysis’. It consists of a broad review of the recent evolution and likely prospects of economic conditions (including growth, employment, prices, exchange rates and foreign conditions). The second pillar, the ‘monetary analysis’, studies the evolution of monetary aggregates and credit which, in line with the neutrality principle, effect inflation in medium to long term.

Officially, the Eurosystem only deals with inflation, but has visibly adjusted its actions when it felt the need to smooth the edges of what is considered to be secondary concerns. Importantly, the Eurosystem does not take any responsibility for the exchange rate, which is freely floating.

Is the Eurosystem’s strategy special? Over the past decade, many central banks have adopted the inflation-targeting strategy. In Europe, this is the case most non-monetary union member central banks. Inflation targeting comprises announcing a target, publishing an inflation forecast at the relevant policy horizon, and adjusting the interest rate according to the differences between the forecast and the target. For example, if the forecast exceeds the target, the presumption that monetary policy is tightened, i.e. that the interest rate is raised.

The Eurosystem has resisted this approach, along with the US Federal Reserve and the Bank of Japan. One reason is that the Eurosystem wants to claim the Bundesbank heritage, but the Bundesbank did not target inflation; it targeted money growth, which explains the second pillar. On the other hand, the Eurosystem’s strategy resembles inflation targeting: there is an implicit target and inflation forecast is published twice a year.

Both the ECB and the NCBs are strictly protected from political influence. Before joining the Eurozone, each country must adapt the statues of its NCB to match a number of common requirements. In particular, the EU Treaty explicitly rules out any interference by national or European authority in the deliberations of the Eurosystem. In addition, to guarantee their personal independence, the members of the Executive Board are appointed for a long period (eight years) and cannot be reappointed, which reduces the opportunity for pressures. Similar conditions apply to the NCB governors, although they differ slightly from one country to another, but their mandates must be for a minimum of five years. No central bank official can be removed from office unless he or she becomes incapacitated or is found guilty of serious misconduct, with the Court of Justice of the European Communities competent to settle disputes. Finally, the ECB is financially independent. It has its own budget, independently from the EU. Its accounts are not audited by the European Court of Auditors, which monitors the European Commission, but by independent external auditors.

Democratic accountability is typically exercised in two ways: reporting and transparency. Formally, the Eurosystem operates under control of the European Parliament. Its statutes require that an annual report be sent to the Parliament, as well as to the Council and the Commission. This report is debated by the parliament. In addition, the Parliament may request that the President of the ECB and the other members of the Executive Board testify to the Parliament’s Economic and Monetary Affairs Committee. Transparency contributes powerfully to accountability. By revealing the contents of its deliberations, a central bank conveys the rationale and difficulties of its decisions to the public. Many central banks report on individual votes, which is a clear way of indicating how certain policy makers feel about their collective decisions. The Eurosystem is nearly alone in doing none of that. It considers that revealing individual votes could be interpreted in a nationalistic manner that in fact does not correspond with the thinking of members of the Governing Council who are duty-bound to look only at the Eurozone as a whole.

The first decade

When the euro was launched in January 1999, inflation was very low, partly because all member countries had been working hard at meeting the Maastricht entry criteria. Soon thereafter, oil prices rose three-fold in 2000. An oil shock both means more inflation and less growth, a classic dilemma that all central banks fear. Simultaneously, stock markets fell worldwide, the end of a long lasting financial bubble, fed by unrealistic expectations of the impact of the information technology revolution. Within months, the US economy went into recession, and Europe’s economy also slowed down. Then, the terrorist attacks of 9/11 shook the world economy. A mellower period followed until oil prices rose again to record levels and a massive financial crisis erupted in the USA in mid-2007.

The result was an inflation rate almost always above the two percent ceiling. It would be wrong to conclude that the Eurosystem has failed to deliver price stability. Until late 2007, it delivered an average inflation rate close enough to two percent for comfort. In fact, no member country enjoyed such a long period of low inflation since the Second World War. Growth, on the other hand, had been slow, which generated much criticism, including from member governments. The overall Eurozone’s growth rate is low, because some of the largest members (Germany, Italy and France) have achieved only slighter better than 0; a disappointing performance.

The Eurosystem has faced another vexing issue. Just when the euro was launched in early 1999, the dollar started to rise vis-à-vis all major currencies, including the euro and, to a lesser extent, the pound sterling. Given that the US dollar has long been the world’s standard, the general interpretation was that the euro was weak. This left the impression that the Eurosystem was unable to deliver the strong currency that had been predicted upon its price-stability commitment, following the PPP logic.

From the start, the Eurosystem clearly announced that it would not take responsibility for the exchange rate. Its view is that the euro is a free floating currency. Noting that capital movements are completely free, it follows the logic of the impossible trinity and refuses to take responsibility for the exchange rate.

A key implication of the OCA theory is that joining the monetary union implies the acceptance that member countries will occasionally have to bear some costs.

Over the first decade of monetary union, lasting inflation differentials occurred in several countries; inflation were lower than average in Germany, Finland and France, and higher than average in Ireland, Spain, Portugal, the Netherlands and Italy. Why? The potential explanations are:

• The Balassa-Samuelson effect;

• Wrong initial conversion rates;

• Autonomous wage and price pressure;

• Policy mistakes;

• Asymmetric shocks.

Different rates of inflation have another, more subtle effect. The single monetary policy implies a common nominal interest rate throughout the Eurozone. Yet, monetary policy does not affect the economy through the nominal interest rate. What matters is the real interest rate, i.e. the nominal rate less expected inflation. Take the case of a country where inflation is, and is expected to remain, higher than average. Its real interest rate is lower than average, which means that monetary policy is comparatively expansionary. Conversely, a country with low growth and low inflation faces a relatively high real interest rate, which exerts a contractionary effect. Is the single monetary policy therefore systematically destabilizing? It all depends on the sources of the inflation differential. If higher inflation is driven by productivity gains, e.g. the Balassa-Samuelson effect, the real exchange rate must appreciate through wage and price increases. In that case, the low real interest rate encourages investment, which is needed to implement the productivity gains, and its expansionary impact speeds up the rise of prices to their new equilibrium level. If higher inflation is due to national policy mistakes or unwarranted wage increases, possibly encouraged by a cyclical expansion, the low real interest rate adds fuel to an already overheated economy and monetary policy appears indeed to be destabilizing. How worrisome is this effect? One view is that there is no risk that things will get out of hand. Hume’s mechanism ensures that excesses will eventually be corrected. Another view, however, is that the correction may be painful. As noted above, the pain may one day lead a country caught in the process to reconsider its continuing monetary union membership. There is nothing that the Eurosystem can do about it. The only policy prescription is to avoid falling into the trap of inflation and, if that happens, to use a contractionary fiscal policy to try to contain the expansionary pressure.

Chapter 5: The Stability and Growth Pact

Fiscal policy in the monetary union

Some important differences between monetary and fiscal policy imply that the two instruments are not as easily substitutable as suggested by the IS-LM analysis. In particular, fiscal policy is more difficult to activate and less reliable than monetary policy. A common problem with both instruments is that they affect spending largely through expectations. For monetary policy the central bank can only control very short term interest rates while private spending is financed through long term borrowing. For fiscal policy, changes in spending and/or taxes impact the budget balance, which immediately raises the question on financing the public debt.

Fiscal policy faces a major additional drawback: it is very slow to implement. Establishing the budget is a long and complicated process. First, the government must agree on the budget, with lots of heavy-handed negotiations among ministers. The budget must then be approved by the parliament; a time-consuming and highly political process. The delay may even be such, that when fiscal policy finally effects the economy, the problem which was meant to be solved, has disappeared. In principle, macroeconomic policies are meant to be countercyclical, i.e. to slow down a booming economy or speed up a sagging economy. Fiscal policy has occasionally been found to be pro-cyclical: an expansionary action designed to deal with a recession hits the economy when it has already recovered. If this is the case, it actually speeds up the economy while it is already desirable to slow it down.

Another way of looking at fiscal policy is that the government borrows and pays back on behalf of its citizens. During a slowdown, the government opens up a budget deficit that is financed through public borrowing. In an upswing, the government runs a budget surplus which allows it to pay back its debt. A government that borrows to reduce taxes now and raises taxes later to pay back its debt is, in effect, lending to its citizens now and making them pay back later.

When a country faces an adverse asymmetric shock, its government can borrow from countries that are not affected by the shock. This is the equivalent of a transfer: instead of receiving a loan or a grant from other Eruozone governments or from Brussels, the adversely affected country’s government borrows on international private markets. This way, fiscal policy makes up for the absence of ‘federal’ transfers in a monetary union.

Automatic stabilizers and discretionary policy actions

Fiscal policy has one important advantage: it tends to be spontaneously countercyclical. When the economy slows down, individual incomes are disappointingly low, corporate profits decline and spending is rather weak. This all means that tax collection declines: income taxes, profit taxes, VAT, etc. are less than they would be in normal conditions. At the same time, spendings on unemployment benefits and other subsidies rises. All in all, the budget worsens and fiscal policy is automatically expansionary. These various effects are called the automatic stabilizers of fiscal policy.

The automatic stabilizers just happen. Discretionary fiscal policy, on the contrary, requires explicit decisions to change taxes or spending. Such decisions are slowly be made and implemented. In some countries, this is why the budget law sets aside some funds, which can be quickly mobilized by the government if discretionary action is needed, even then, the amounts are small and their use is often politically controversial.

The difference between the evolution of the actual and cyclically adjusted budget balances is the footprint of the automatic stabilizers. The cyclically adjusted budget balance is reliable gauge of the stance of fiscal policy since it separates discretionary government actions from the cyclical effects of the automatic stabilizers. An improvement indicates that the government tightens fiscal policy whereas an expansionary fiscal policy worsens the cyclically adjusted budget balance.

Fiscal policy externalities

Fiscal policy actions by one country may spill over to other countries. Such spillovers, called externalities, mean that one country’s fiscal policy actions can help or hurt other countries. In principle, all concerned countries could agree on each other’s fiscal policy to achieve a situation that befits them all. This is what policy coordination is about. On the one hand, the setting up of a monetary union strengthens the case of fiscal policy coordination as it promotes economic integration. On the other hand, fiscal policy coordination requires binding agreements on who does what and when. Such detailed arrangements would limit each country’s sovereignty, precisely at a time when the fiscal policy instrument assumes greater importance.

When two monetary member countries undergo synchronized cycles, for example both suffer from a recession, each government will want to adopt an expansionary fiscal policy, but to what extent? If each government ignored the other’s action, their combined action may be too strong as each economy pulls the other one from the recession – an effect of the Keynesian multiplier. If, instead, each government relies on the other to do most of the work, too little might be done. Next, consider the case when the cycles are a-synchronized. An expansionary fiscal policy in the country undergoing a slowdown stands to boost spending in the already booming country. Conversely, a contractionary fiscal policy move in the booming country stands to deepen the recession in the other country. The risk, in this case, is too much policy action.

Europe is fully integrated in the world’s financial markets so any one country’s borrowing is unlikely to make much of an impression on world and European interest rates. On the other hand, heavy borrowing may elicit capital inflows. This could result in an appreciation of the euro, which would hurt the area’s competitiveness and cut into growth. Borrowing costs thus represent another channel for spillovers.

Heavy public borrowing by one country is a sign of fiscal indiscipline that could trouble the international financial markets. If markets believe that one country’s public debt is unsustainable, they could view the whole Eurozone with suspicion. The result would be sizable capital outflows and euro weakness. This is another potential source of spillover. There is still another potential spillover. If a government accumulates such a debt that it can no longer service it, it must default. The experience with such default is that the immediate reaction is a massive capital outflow, a collapse of the exchange rate and of the stock markets, and prolonged crisis complete with a deep recession and skyrocketing unemployment. Too bad for the delinquent country, but being part of a monetary union changes things radically. It is now the common exchange rate which is the object of market reactions. The spillover can extend further to stock markets throughout the whole monetary union. A further fear is that the mere threat of one country’s default would so concern all other member governments that they would feel obliged to bail out the nearly bankrupt government. This last risk has been clearly identified in the Maastricht Treaty, which includes no-bailout clause. The clause states that no official credit can be extended to a distressed member government. In spite of the no-bailout clause, it remains that, in the midst of an emergency, some arrangement could still be found to bail out a bankrupt government. It is feared that a sovereign default would badly affect the Eurozone and undermine its credibility, with seriously adverse effects on the euro.

Principles

The theory of fiscal federalism asks how, in one country, fiscal responsibilities should be assigned between the various levels (national, regional, municipal) of government. It can be transposed to Europe’s case, even though Europe is not a federation, by asking which task should remain in national hands and which ones should be shared responsibility, i.e. delegated to Brussels.

As mentioned before, spillovers lead to inefficient outcomes when each country is free to act as it wishes. One solution is coordination, which preserves sovereignty but calls for repeated and often piecemeal negotiations, with no guarantee of success. Another solution is to give up sovereignty, partly or completely, and delegate a task to a supranational institution. In Europe, some important tasks have already been delegated to the European Commission and to the Eurosystem.

Heterogeneity of preferences and information asymmetries imply that it would be inefficient to share competence at supranational level. Much of the criticism levelled at Brussels’ concern cases where either heterogeneity or information asymmetries are important: deciding on the appropriate size of cheese or the way to brew beer is best left to national governments, or even local authorities, no matter how important the externalities or increasing returns to scale.

It should be clear by now that in most cases the four arguments for and against centralization at EU level are unlikely to lead to clear-cut conclusions, and the warning about the quality of government further complicates the issue. In each case, one has to weigh the various arguments and trade off the pros and cons. This is often mission impossible, hence another question: where should the burden of proof lie? The EU has taken the view that the burden of proof lies with those who argue in favour of sharing sovereign tasks. This is the principle of subsidiarity.

It is crucial to separate two aspects of fiscal policy. The first aspect is structural, that is, mainly microeconomic. It concerns the size of the budget, what public money is spent on, how taxes are raised, i.e. who pays what, and redistribution designed to reduce inequalities or to provide incentives to particular individuals or groups. The second aspect is macroeconomic. This is the income stabilization role of fiscal policy, the idea that it can be used as a countercyclical instrument. Here, we focus on the macroeconomic stabilization component of fiscal policy, ignoring the structural aspects.

There are a number of spillovers: income flows, borrowing costs and the risk of difficulties in financing runaway deficits, possibly leading to debt default. Some of these spillovers can have serious effects across the Eurozone. In addition, some countries have or established political institutions that are conductive to fiscal discipline so it may be in their own best interest to use Brussels as an external agent of restraint. On the other hand, it is difficult to detect a scale economy. These externalities call for some limits on national fiscal policies, and such limits can take various forms, ranging from coordination and peer pressure to mandatory limits on deficits and debts.

Macroeconomic heterogeneity occurs in the presence of asymmetric shocks. A common fiscal policy, on top of a common monetary policy, would leave each country without any countercyclical macroeconomic tool. Heterogeneity can also be the consequence of differences of opinions regarding the effectiveness of the instrument. Finally, national political processes are another source of heterogeneity. In some countries, the government has quite some leeway to adapt the budget to changing economic conditions, whereas in others the process is cumbersome and politically difficult. Whether and how to use fiscal policy is particular; juncture is part of a complex political game, which makes national politics highly idiosyncratic.

The Stability and Growth Pact

The Stability and Growth Pact (SGP) was adopted in 1997 and was meant to be strictly enforced. However, because fiscal policy remains a national competence, final word had to be given to ECOFIN, the council of Finance Ministers of the Eurozone, acting on proposals from the Commission. The Commission has assumed the responsibility of ‘tough cop’, but ECOFIN has been loath to make decisions that would strongly antagonize its members, especially the Finance Ministers from the large countries. In November 2003, France and Germany were to be sanctioned. Under pressure from the French and German Finance Ministers, ECOFIN recanted and put the SGP ‘in abeyance’. The Commission took ECOFIN to the Court of Justice of the European Communities for violation of the Pact, an unprecedented action. A new version of the SGP was adopted in June 2005. The SGP has some difficulties. The benefits from coordination are limited, the collective need for discipline is high, but collectively enforced discipline clashes with sovereignty. This conflict is unavoidable. Discipline cannot be enforces without the threat of sanctions. The revised SGP does not solve the logical conflict that lies at its heart, as it keeps the principle of sanctions; instead it seeks to avoid a situation where sanctions have to be applied.

The SGP consists of four elements;

1. A definition of what constitutes an ‘excessive deficit’;

2. A preventive arm, designed to encourage governments to avoid excessive deficits;

3. A corrective arm, which prescribes how governments should react to a breach of the deficit limit;

4. Sanctions.

The SGP applies to all EU member countries, but only the Euro zone countries are subjected to the corrective arm.

• Excessive deficits: The SGP considers that deficits are excessive when they are above three percent of GDP. In order to leave room for the automatic stabilizers to play their role, the Pact also stipulates that participants in the monetary union commit themselves to a medium term budgetary stance ‘close to balance or in surplus’. The medium term is understood to represent about three years. The Pact defines exceptional circumstances when its provisions are automatically suspended. A deficit in excess of three percent is considered exceptional if the country’s GDP declines by at least two percent in the year in question. The SGP also identifies an intermediate situation, when the real GDP declines by less than two percent, but by more than 0.75 percent. In that case, if the country can demonstrate that its recession is exceptional in terms of its abruptness or in relation to past output trends, the situation can also be deemed exceptional.

• The preventive arm: The SGP can exert counter-pressure in the form of peer pressure, called mutual surveillance. The preventive arm is designed to submit Finance Ministers to a collective discussion of each country’s fiscal policy in the hope that this will be enough to deliver budgetary discipline. Prevention is meant to prevent the need for correction. Formally, each Euro zone government submits a Stability Programme early each year. The document present the government’s budget forecast for the current year and the next three years. The Commission examines each programme and submits its individual assessments to ECOFIN. ECOFIN than delivers an opinion, adopted by qualified majority.

• Corrective arm: When a country does not meet the requirements of the SGP, ECOFIN applies gradually increasing peer pressure. The process starts with an ‘early warning’ and recommendations. The next step is the excessive deficit procedure (EDP). The procedure is triggered by an ECOFIN opinion stating that a country’s budget is in excessive deficit. ECOFIN issues recommendations – following a recommendation from the Commission – which the country must follow. The country must present a set of prompt corrective measures soon. What follows is a set of evaluations of the measures taken by the delinquent country and of the budget outcome. If the country remains in excessive deficit, sanctions are to be imposed by ECOFIN.

• Sanctions: The sanction takes the form of a non-remunerated deposit at the Commission. The deposit starts at 0.2 percent of GDP and rises by 0.1 percent of the excess deficit up to a maximum of 0.5 percent of GDP. Deposits are imposed each year until the excessive deficit is corrected. If the excess is not corrected within two years, the deposit is converted into a fine.

Several aspects of the SGP are noteworthy. First, formally it does not remove fiscal policy sovereignty. Governments are in full control. Second, the intent is clearly pre-emptive since there is a lengthy procedure between the time a deficit is deemed excessive and the time when a deposit is imposed with two more years before the deposit is transformed into a fine. Third, while a fine is politically a nuclear bombshell, the declaration that a country is in violation of the Pact is a more conventional bombshell, meant to elicit prompt corrective action. Finally, all decisions are in the hands of the Council, a highly political body that can exploit many of the ‘ifs’ included by the Pact.

SGP critics make two objections. They say maybe Europe’s poor economic performance since 1999 is partly due to the SGP, which has prevented a more dynamic countercyclical use of fiscal policies. In addition, SGP critics observe that if the budget is in surplus in normal years, it means small deficits and bad years and big surpluses in good years. On average, therefore, the budget will be in surplus. This may sound good, but what does it mean for the public debt? Surpluses, year in and year out, imply that the public debt will be on a declining trend. But a question remains: how far will the debt decline? As long as the budget remains average in surplus, the debt will decline, so one day it will be zero. After that the debt will become negative, which means that the government will start lending to the private sector, at home or abroad. Put differently, the government will raise taxes to make loans, competing with private banks. Critics argue this makes little sense.

The SGP is meant to serve two main useful purposes: to counteract the deficit bias and to reduce the odds of a debt default within the monetary union which could result in highly painful spillover effects. The economic rationale is weak and the political conditions of its implementations are bound to be contentious.

If the SGP is to have any influence on governments subject to deficit bias, it must be backed by credible sanctions. If sanctions are imposed, the condition under which they are triggered must be clear and uncontroversial. The problem is that computing the cyclically adjusted budget balance is more art than science.

Is there a better measure than fiscal discipline? Yes, there is, and it is rooted in good economics. Why do we care about fiscal discipline in the first place? Because lack of discipline eventually leads to a debt default. The natural implication is that the SGP should target debt-to-GDP ratio.

Imposing fiscal discipline from outside has the obvious advantage of protecting governments from domestic interest groups. On the other hand, using Brussels as a scapegoat may be good politics in the short term but, if invoked too often, it can undermine general support for European integration. Yet, leaving the final decision in the hands of ECOFIN has a serious drawback. Finance Ministers are, by definition, politicians. As such, they make elaborate calculations involving tactical considerations often far away from the principles discussed here.

Chapter 6: The Financial Market

Capital markets

At a very general level, the capital market performs three main functions:

1. It transforms maturity;

2. If performs intermediation;

3. It deals with inherent risk.

Insurance companies are also considered to be financial institutions. Part of their activity is to provide insurance, which, strictly speaking, is not a financial service. Yet, in order to face potentially high payments, they accumulate large reserves, which they want to manage in order to obtain returns as highly as possible. In effect, they take ‘deposits’ – the insurance premia paid by their customers – that they use to ‘make loans’ as they invest in financial assets.

The bond and stock markets represent the other component of the financial system. Like banks, they are designed to collect savings and lend them back to borrowers, with the crucial difference that the end users – lenders and borrowers – ‘meet’ each other on the market. Bonds are debts issued by firms and governments for a set maturity at an explicit interest rate, which can be indexed and therefore be variable. Stocks (also called shares) are ownership titles to firms: they have no maturity, since they last as long as the firm and its returns are determined by the firm’s performance.

Term deposits are issued when you withdraw money before it has reached maturity and have to pay a penalty for that. On the one hand, term deposits offer more attractive interest, which grows with the maturity. The bank thus encourages its customers to choose longer term deposits, because the bank will re-lend deposits to another customer, who will be ready to pay a higher interest for loans of longer maturity. Time has value, and the market sets its price.

Bonds and shares are risky: if the issuer goes bankrupt, they are probably worth nothing, at best a fraction of their face value. Market balance demand and supply by setting the risk premium. Put differently, the markets put a price tag on risk and all assets fit nicely on the same risk-return schedule shown in the figure below. The risk-return schedule reveals the price of risk, and financial markets allow everyone to trade off a higher return for more risk, or the converse, depending on personal preferences.

The existence of different currencies is a barrier and the creation of a single currency removes this particular barrier to competition.

One response to scale economies is the emergence of large financial institutions and markets. Another response, which goes hand in hand with the first, is the building of networks. When a financial firm receives funds from a saver, it needs to re-lend these funds as soon as possible since ‘time is money’. With some luck, it will find a borrower with matching needs and preferences amongst its customers, but more often not. The solution is to re-lend the saver’s money to another financial firm which may have spotted a borrower or identified another financial firm which may have spotted a borrower, etc. This is why financial markets operate as networks. Indeed, money passes quickly from firm to firm until it finds house – suitable borrower – somewhere in the network, and quite possibly in a very different corner of the world. The larger the network, the better it works.

A fundamental characteristic of financial activities is that the borrower always knows more about his own riskiness than the lender. This information asymmetry carries profound implications. Borrowers may intentionally attempt to conceal some damning information for the sake of obtaining a badly needed loan.

Microeconomics of capital market integration

Until the 1986 Single European Act and the 1988 directive that ruled out all remaining restrictions on capital movements among EU residents, EU capital markets were not very integrated. Although the free movement of capital is in the Treaty of Rome, the Treaty provided several large loopholes that EU members eagerly exploited. The basic problem was that, until recently, EU nations just did not believe that unrestricted capital mobility was a good idea.

The main goal of EU capital market liberalization prior to the 1980s was to facilitate real business activities. For example, national policies should not hinder a company based in one member state from setting up business in another member state. This so-called right of establishment covered international transfers of capital that may be necessary to set up business. Likewise, national policies were not supposed to hinder the repatriation of profits or wages among member states to the extent that such hindrances act as restrictions of the free movement of goods and workers. Article 56 of the Maastricht Treaty banned all national restrictions on the movement of capital except those required for law enforcement and national security reasons.

The economic consequences of capital market integration are symbolized in the graph below. We start by focusing on Home, ignoring capital mobility. The MPK curve in the diagram shows how the ‘marginal product of capital’ declines as the total amount of capital employed increases. If the capital stock in Home is given by K₀, then the equilibrium marginal product of capital in Home will be r₀, assuming that the capital market is competitive. The idea is that firms competing for Home’s capital supply force the ‘price’ of capital, r, up to the point where the price they pay for capital just equals its marginal product. By the usual logic of competition, the outcome is that the competitive firms pay r₀ and all capital is employed. Following the same competitive logic as for Home, we see that the foreign return to capital will be r₀* since this is the MPK* where all of the capital is employed in Foreign.

As the diagram is drawn, capital earns a higher return in Home than it does in Foreign. If we now allow international capital flows and, for the sake of simplicity, assume that such flows are costless, it is clear that capital will leave Foreign and move to Home in search of a higher reward. Such capital flows raise the level of capital employed in Home and lower it in Foreign, thus narrowing the gap between r₀ and r₀*. Indeed, under our assumption that capital flows are costless, capital moves from Foreign to Home until the returns are equalized. This occurs at point A, where the two MPK curves intersect. The resulting capital flow and the common reward is r’. Notice that capital movement has raised the return in sending nation and lowered it in the receiving nation.

The graph below shows that the area under the MPK curve gives the total output of Home. The total earnings of Home capital is just the equilibrium reward, r₀, times the amount of capital, K₀. And, since we are assuming that capital and labour are the only two factors of production, labour receives all the output which is not paid to capital. Graphically, this means that capital’s income is the grey rectangle shown in the diagram, while labour’s income is the blue area between the MPK curve and the r₀ line.

The ‘native’ capital owners in Home lose, since their reward has fallen from r₀ to r’. The amount of loss is measured by the rectangle A in the graph below. Home labour increases its earnings by area A plus the triangle B. Thus the total economic impact on Home citizens is positive and equal to triangle B.

Correspondingly, Foreign output drops by D + E, while the capital remaining in Foreign sees its reward rises from r₀* to r’. The size of this gain is shown by rectangle F, which is the change in r times the amount of capital left in Foreign after the integration (this is illustrated by point A). Foreign labour sees its earnings drop by D + F. Combining all these losses and gains, the Foreign factors of production that remain in Foreign, lose overall by an amount measured by triangle D. However, if we count the welfare of Foreign factor owners, including the capital that is now working in Home, the conclusion is reversed. Total gains to Foreign capital are C + D + F, while the loss to Foreign labour is D + F. Foreign gains from the capital outflow by an amount equal to triangle C. In short, while capital flows create winners and losers in both nations, collectively both nations gain from the movement of capital. The deep reason for this has to do with efficiency. Without capital mobility, the allocation of productive factors was inefficient. The result that both countries benefit from capital market integration is the basis for the single market and the Commission directive. It underpins the view that, by encouraging capital mobility further, adoption of the euro will raise economic efficiency and welfare.

When the perfect market assumption is removed, the result no longer holds: capital market integration may or may not be beneficial; it all depends on the details of the deviations from perfection and on a host of other features. Does it mean that capital market integration is a bad idea? Probably not. The presumption is that, as long as competition is strong enough, integration is beneficial.

A single financial market first means more competition as national currencies that used to act as non-tariff barriers are eliminated. Rents are associated with dominating positions should disappear and the need to retain and attract new customers should push financial institutions to constantly improve their performance. A unified financial market should also allow a better exploitation of scale economies, with the emergence of large financial institutions and markets.

One possible negative aspect of the euro, however, is that the potential for diversification shrinks. Before the advent of the euro, a Belgian saver could diversify her portfolio by acquiring German, Italian and other European assets. Now these assets are less diverse as they all share the same currency and as cyclical conditions become more homogeneous.

Financial institutions and markets

The banking industry is special in three respects. First, banks are naturally fragile, and bank failures can be systemic. As a consequence, banks are highly regulated and supervised. Second, the information asymmetry problem is acute since banks earn profits primarily from their lending activities. Banking does not conform to the perfect market model. One implication is that long term relationships are important as they provide banks with track records of their customers and help to build up confidence, breaking somewhat the information asymmetry. The downside is that well-established customers have little incentive to quit their banks. While this attachment alleviates the information asymmetry problem, it also reduces competition.

Banks started to develop at the local level, which allowed them to know their customers reasonably well, thus minimizing information asymmetries. Scale economies next led to a process of growth and mergers as banks sought to become ever bigger, but so far this process has generally taken place within national boundaries. The number of banks has declined in the Eurozone recently, and a large part of this decline is explained by mergers and acquisitions. Mergers and acquisitions in the big countries are predominantly conducted internally. There exists further evidence that the euro has not led banks to operate at the Eurozone level. For instance, loans by Eurozone-based banks to customers located elsewhere in the area remain a small and unchanged proportion of total bank loans.

Several reasons have been advanced to explain the continuing apparent parochialism of banks in contrast with other industries. First, local regulations still differ. This has long been recognized and has led to a succession of harmonization efforts. In addition, while in theory a ‘single banking market’ is now in place, several non-regulatory hurdles remain. They operate like non-tariff barriers and are used by national authorities to protect home-grown banks, and in effect, stifle competition. Second, for several centuries, banks have developed along diverse lines, leading to different traditions in banking. Third, tax treatment of savings differ from country to country. Finally, protectionism is suspected. The fact that mergers and acquisitions in the large countries are predominantly within-borders may simply be a size effect, but it may also be the result of protectionism.

The evidence so far is that, through mergers and acquisitions, banks have been consolidating at the national level. First, their strategy seems to be to reach a size that is large enough to enable them to engage in foreign purchases. Meanwhile, as banks merge at national level, concentration increases, which may result in less, not more, competition. Thus, in contrast with the effects expected form the Single European Act it could be that the early impact of the monetary union is to reduce competition.

While banks do not consolidate at pan-European level, they could still offer services across borders. They can open new branches and move close to their customers to circumvent the information asymmetry.

Increasingly, individuals buy shares from collective funds designed to offer good risk-return trade-offs through extensive diversification. Yet, for all the hype about globalization, it is striking that stock markets are characterized by a strong home bias: borrowers and savers alike tend to deal mostly on domestic markets and to hold domestic assets. One reason for the home bias is information asymmetry: investors know more about domestic firms. This is unlikely to change. Another reason is currency risk. This obstacle to capital mobility has been eliminated within the Eurozone, so we would expect less of a home bias. Is it happening? Apparently, yes. Assets held by Eurozone-based investing funds that report pursuing a Europe-wide strategy has sharply increased following the adoption of the euro in 1999. Another piece of evidence is provided by the evolution of stock exchanges, the marketplace where shares are traded. Because stock exchanges display strong scale economies, the perception is that the Eurozone is evolving in the direction of one major centre and a handful of secondary exchanges.

In every country, financial markets are regulated and the financial institutions are closely supervised. The presence of scale economies implies that a few large firms eventually dominate the market. The tendency for competition to become monopolistic challenges the perfect competition assumption. It also means that financial markets are vulnerable to difficulties suffered by one or two of these important players. This vulnerability is sharpened by the two other characteristics. The network feature means that all large financial institutions are continuously dealing with each other, and routinely borrowing and lending huge amounts among each other. If one of these institutions fails, all the others may be pulled down. Failures tend to be systemic. The presence of information asymmetries means that all financial firms routinely take risks. Every asset represents the right to receive payments in the future, be it 24 hours or 15 years. It is trivial to observe that the future is unknown, but this feature has deep implications for financial markets. Today’s value of an asset represents the best collective judgement by financial market participant of the likely payments that the asset holder may expect to receive upon maturity. But one thing is sure: the future will differ from today’s expectations. The asset may yield better returns than expected, but can also be revealed as catastrophic and its value can deeply deteriorate.