This file is also available in Adobe Acrobat PDF format
The Evolution of the International Monetary System
Chapters 4 and 5 discuss the political economy of international money and finance. In chapter 2 we discussed the reasons for the evolution of the international trade system toward institutionalized multilateralism. Similarly, in this chapter we consider how the literature in political economy has tried to explain the evolution of the international monetary system. As for international trade, the international monetary system also became characterized by institutionalized multilateralism by the mid-twentieth century, for a similar combination of material and ideational reasons. In contrast to trade, however, multilateralism in international monetary arrangements peaked in the 1960s and has been in retreat since then. We emphasize the rise of private financial markets and associated political changes in explaining this trend.
Although international monetary arrangements in the nineteenth century were much less politicized at the domestic level than were those for international trade, over the course of the twentieth century this became less true. And at the international level, international monetary arrangements have always been politically controversial. This has much to do with the strong tendencies toward hierarchy in global monetary and financial markets and the asymmetric consequences this has for the distribution of costs and benefits across countries. The conflict between national macroeconomic stabilization policies and international monetary commitments has become more acute with the reemergence of open financial markets since the 1970s. We postpone to chapter 5 a discussion of the consequences of this conflict.
In order to address the question of how to explain international monetary evolution, we must first consider some preliminary issues concerning the balance between adjustment and financing in international monetary organization. After this, we go on to argue that the rise of domestic macroeconomic management in the twentieth century, itself the result of new ideas and of democratization, has had profound effects on this balance and on the international monetary system in general. These effects include the growing politicization of international monetary and financial arrangements compared to the nineteenth century, the gradual undermining of an anchor role for gold, and a shift toward greater exchange rate flexibility and from public to private international finance.
FINANCING AND ADJUSTMENT IN THE BALANCE OF PAYMENTS
The central organizational problem for any international monetary system concerns the mechanism of adjustment to balance-of-payments disequilibria between countries.1 The fundamental issue from the perspective of political economy is that the costs of adjustment to disequilibria are typically distributed asymmetrically across countries and across different groups within countries. International financing mechanisms are a means of sustaining balance-of-payments deficits and thereby avoiding (at least temporarily) such adjustment pressures. Thus, the nature of the international adjustment mechanism(s), and the balance between financing and adjustment, constitutes the core of the politics of international monetary arrangements.
To see this more clearly we need to understand the balance of payments (BoP). This measures a country’s transactions with nonresidents over a given period of time (e.g., monthly, quarterly, or yearly). It categorizes different kinds of cross-border “flow,” and it balances by definition. The sum of the balances on current, capital, and financial accounts, the “overall balance” (line 5 in table 4.1), must equal the change in reserve assets (which consist of a country’s monetary reserves and its credit balance at the International Monetary Fund, or IMF).2 Unless otherwise specified, when referring to a BoP deficit or surplus, we mean the overall balance.
Although one country’s balance-of-payments deficit is matched by foreign surpluses, in practice there is often more pressure on deficit countries to adopt adjustment measures. This is because deficit countries must either “adjust” to eliminate their deficit, or sell assets to foreigners or accumulate liabilities to foreigners to “finance” ongoing deficits. Both strategies, as we will see, are costly. By contrast, surplus countries can sometimes simply accumulate net claims on foreigners (including official reserves).3 The limits to deficit-financing mechanisms vary greatly by country. The sale of assets to foreign residents often hits political limits earlier than economic limits: witness the political controversy in the United States in the 1980s over Japanese investments in the United States and more recently over Chinese investments. Financing deficits by borrowing abroad can also be costly, and these costs often rise as more is borrowed. Some foreign creditors may also impose preconditions on the borrower that are politically or economically costly. Foreign borrowing can also be risky, since the willingness of foreign creditors to extend finance can be volatile. Nevertheless, because borrowing does not normally transfer ownership of national assets to foreigners, deficit countries often rely heavily on this form of financing over long periods of time.
This fact also suggests that adjustment policies sufficient to eliminate deficits are often even more costly for governments. Adjustment measures work primarily by reducing domestic consumption, investment, and growth in the short term. To illustrate, consider South Korea’s BoP from 1996 to early 1999 (table 4.1).
To simplify the discussion, we leave aside the relatively insignificant (at least for Korea) capital account. In the absence of adjustment measures, to avoid running down monetary reserves, any deficit on current account must be financed by a corresponding surplus on financial account. Korea’s current account deficit meant it was spending considerably in excess of its national income through foreign borrowing. Over 1996, Korea achieved this by selling equities to foreigners and by borrowing heavily from international banks. As is now well known, this financing choice proved unsustainable once private foreign creditors drew down or sold their Korean assets. In 1996, before the crisis hit, the Korean economy was growing rapidly, though its exports were not, resulting in a rising current account deficit (5.9% of GDP). The financing problem was greater than this, as Korean firms were investing heavily abroad, while foreign firms had found it difficult to invest in many sectors in Korea (line 3A, table 4.1). For 1996 as a whole, foreign financing included net portfolio (equity and debt) inflows of $15.2 billion, net corporate (nonbank) borrowing of $10.4 billion, and net bank borrowing of $1.8 billion.
When these financial inflows reversed in the wake of the Thai crisis of July 1997, reserves fell quickly and the economy was plunged into a BoP and financial crisis, necessitating rapid, deep, and politically costly adjustment measures in return for IMF assistance. Over 1998, GDP fell by 7%, unemployment rose, and many firms and banks went into bankruptcy. The adjustment costs are apparent in the current account change, from a large deficit to an unprecedented surplus of 9% of GDP in 1998. As table 4.1 shows (lines 1Aa), export receipts continued to stagnate despite the large fall in the value of the currency. The burden of adjustment fell on imports, which fell by one-third from late 1997 to early 1998, indicative of the collapse in domestic consumption and investment. Nevertheless, the large current account surpluses this deep adjustment produced allowed Korea to rebuild its reserves rapidly over 1998–99 (line 6a, table 4.1) and to repay its IMF borrowings early.
The role of official reserves also emerges clearly from this story. These enable countries to finance temporary deficits without more borrowing or adopting costly adjustment measures. Stocks of reserves are accumulated over time, so that past economic performance and policies affect a country’s ability to draw on them in times of difficulty. The accumulation of large stocks of monetary reserves is itself costly in terms of forgone consumption and investment, and after some point governments will prefer to finance BoP deficits by inducing private capital inflows or by borrowing from foreigners. The availability of such private and official finance, and a country’s desire for political autonomy from foreign creditors, will be crucial factors in national decisions about how large monetary reserves need to be. Since the crisis, Korea, along with many other developing countries, decided that its precrisis reserves ($34 billion in June 1997) were insufficient given the demonstrated vulnerability of the country to large and rapid shifts in private capital flows. This has led to an unprecedented accumulation of foreign exchange reserves since the mid-1990s—essentially a form of insurance against future crises. As figure 4.1 shows, developing countries (in marked contrast to developed countries) on average have recently accumulated reserves well in excess of the traditional rule of thumb of three months’ import coverage.
Adjustment in Theory and Practice
Economics textbooks often outline how, in principle, both perfectly fixed and perfectly flexible exchange rate systems4 can facilitate automatic adjustment processes that eliminate BoP imbalances. In both models, the real exchange rate adjusts to remove either external surpluses or deficits.5 Assuming Korea needed a real depreciation of the won (the Korean currency) before the crisis to promote net exports, how might this have occurred?
In the case of a fixed nominal exchange rate and a current account deficit, stagnating exports and rising unemployment in Korea’s export sector could reduce real wages, eventually reducing domestic prices to the point where international competitiveness would be restored.6 A second “semiautomatic” adjustment mechanism may operate if Korean residents sell won for foreign currencies because they expect a won devaluation, obliging Korea’s central bank to sell its foreign exchange reserves to buy won in the foreign exchange markets. This would reduce the domestic money supply, reducing total expenditure and pushing down prices, helping to restore competitiveness.7 These two mechanisms supposedly operated in the ideal world of the gold standard.8 In the real world of the 1990s, prices and wages in Korea were not flexible downward (in part because of its labor unions), so that adjustment was insufficiently rapid, or Korea’s reserves were inadequate.
The obvious alternative for Korea was simply to float its currency, allowing the nominal exchange rate of the won to depreciate against other major currencies, restoring competitiveness. Economists such as Milton Friedman claimed flexible exchange rates could automatically eliminate external deficits in this way.9 It had the great advantage of bypassing the two mechanisms previously mentioned and could obviate the need for domestic price and wage flexibility altogether. In practice, flexible exchange rates rarely work in this stabilizing manner. In fact, the real value of the won was sustained by continued large capital inflows from abroad before 1997. Even with exchange rate depreciation, total export values may take a long time to recover to close the external deficit.10
Since automatic adjustment is often not forthcoming, governments wishing to undertake active adjustment measures have three main options: trade protection, “expenditure-reducing,” and “expenditure-switching” policies. Higher tariffs or quotas on imports increase the relative price of imported goods, but these measures are generally disparaged by economists. Increased trade protection would also have been incompatible with Korea’s GATT membership and its relations with major countries like the United States and Japan.
Expenditure-reducing policies include restrictive monetary and fiscal policies, but growing capital mobility means that which of them is effective depends on a country’s exchange rate policy. This was outlined in the standard Mundell-Fleming (MF) model of the early 1960s.11 Under fixed exchange rates, monetary policy contraction raises the domestic interest rate, attracting speculative capital from abroad. The central bank must resist currency appreciation by selling domestic for foreign currency, reversing the initial decrease in the money supply and rendering monetary policy powerless to affect prices or output. This result has been termed the “impossible” or “unholy” trinity, since it implies that a government must choose only two of the following three things: a fixed exchange rate, monetary policy autonomy, and capital account openness.12 The power of fiscal policy, however, remains intact and is even enhanced under fixed rates. Fiscal contraction reduces domestic output and interest rates (by reducing government borrowing). Lower interest rates threaten currency depreciation, requiring monetary contraction by the central bank, thus enhancing the effect of fiscal contraction.13
Expenditure-switching policies are simply another term for exchange rate changes. Real currency devaluation “switches” domestic expenditure away from imported goods toward domestic production by altering relative prices. As we discuss later, concerns about policy credibility and costs to domestic consumers often deter governments from devaluing.
One major problem with macroeconomic policymaking is that governments have multiple policy targets but possess limited policy instruments. In practice, a combination of macroeconomic policy adjustment and exchange rate changes is often necessary. For example, if a country is suffering from both high unemployment and an excessively large external deficit, under an open capital account and a pegged exchange rate, fiscal expansion is the only available macroeconomic policy option to reduce unemployment. Fiscal expansion by itself would also worsen the external deficit, so the government would need to combine it with exchange rate devaluation to restore both external and internal equilibrium.14 Further complications and policy dilemmas often arise. For example, unions in sectors sheltered from international trade competition may have little concern for the competitiveness of industries in the traded goods sector. If they dominate national wage-setting and respond to devaluation with increased wage demands, unemployment might coexist with both “cost-push” inflation and an external deficit.15
Note also that when governments have multiple objectives, a “credibility problem” can emerge. It can arise, for example, from a contradiction between an exchange rate commitment and an internal inflation or unemployment objective. Italy in the early 1990s suffered from high unemployment and large external deficits. As noted above, devaluation was part of the textbook solution to this problem (fiscal expansion was made difficult by Italy’s existing high levels of fiscal deficit and public debt). However, the government had made a public commitment not to devalue the lira against other currencies in the European Monetary System (EMS). For this promise to be credible to the private actors in the foreign exchange markets, they would have had to believe that the government was willing to tolerate persistently high levels of unemployment (given the inflexibility of the labor market). Market actors were right to judge that as unemployment grew, the exchange rate commitment became increasingly incredible because the democratically elected government was highly sensitive to unemployment. This is an example of what is often termed a “time inconsistency” problem, in which the credibility of a particular policy commitment declines when policymakers face changed circumstances. The problem arises when a policymaker commits ex ante to following a certain policy rule (such as a given exchange rate peg) in the future, but has an ex post incentive to defect from the rule.16
Note that if the authorities can credibly signal to private market actors that they possess no internal objective and only desire a stable exchange rate, no doubt about their credibility arises. This may approximately describe the situation enjoyed by governments and monetary authorities in the classical era of the gold standard in the late nineteenth century. Since then, many factors have eroded the credibility of exchange rate commitments for most countries. The trend toward democratization, the growth of unions, and new understandings about the power of macroeconomic policy and associated expectations of government have all caused foreign exchange markets to become more skeptical of government promises to maintain “fixed” exchange rates.17 These factors also alter the credibility of exchange rate commitments in an asymmetric way. Countries with low unemployment and current account surpluses, such as Germany enjoyed for many years after 1950, are less vulnerable to collapses of market confidence than countries with high unemployment and external deficits. Developing countries can also exhibit weak taxation bases, fiscal subsidies for basic commodities, and weak social safety nets, making it difficult to undertake policy adjustments in response to changed circumstances, further undermining the credibility of their exchange rate commitments.
International Interdependence and Adjustment
So far, we have considered the adjustment problem from an individual country’s perspective. Since economic policies, especially those of large countries, affect other economies, we also need to consider adjustment from a systemic perspective. This is obviously true when a country undertakes exchange rate adjustment, which creates a potential for international conflict.18 When all governments have domestic policy objectives, the problem can become acute. How far can or should policy coordination go under a fixed exchange rate system? At a minimum, given free movement of capital between two countries A and B, the maintenance of a fixed exchange rate between them requires the same level of real interest rates in both countries, and hence perfectly coordinated monetary policies. This is because if market actors expected currency B to depreciate against A in the future, they would require an interest rate premium over and above what they could obtain by holding assets in currency A, equal to the amount of expected depreciation of currency B over the period (the “interest parity theorem”).
If the world economy only consisted of A and B, a payments imbalance between them raises the question of who will adjust first, the deficit or the creditor country. Since the loss of monetary reserves is more difficult to sustain than is acquiring reserves, deficit countries often come under market pressure to adjust first. If A is the deficit country and the threat of reserve depletion forces it to devalue, this raises the price of imported goods from B and reduces standards of living in A (at least in the nontraded sector). This does not mean that B escapes adjustment costs, since A’s devaluation amounts to a currency revaluation for B, which could hurt its export sector. In a more realistic world in which there are many countries, most of the costs of adjustment may fall on the devaluing country.
In a world economy in which all countries peg their exchange rates, there will be n-1 exchange rates (where n is the number of countries). For the system to be stable, there can only be n-1 exchange rate policies: the nth country must adopt a passive attitude toward its exchange rate. This can create an additional deep asymmetry within a pegged exchange rate system, since the nth country is in the position of being able to focus on internal economic objectives. The classic example is the Bretton Woods exchange rate system in which other countries pegged to the US dollar, allowing the American authorities to concentrate on domestic (and foreign policy) objectives. For many years, this asymmetrical bargain worked reasonably well, providing the basis for the expansion of trade and security for the Western alliance. From 1966, the Johnson administration’s simultaneous pursuit of the Vietnam War abroad and the extension of social welfare at home led to growing fiscal and current account deficits. This prompted exchange market speculation against the dollar and in favor of the German mark, given Germany’s growing current account surplus. Germany’s partial controls on short-term capital inflows and the Bundesbank’s sterilization of the effects of its purchases of foreign currency provided some relief from the inflationary effects of these inflows, but these ultimately proved insufficient to maintain the exchange rate peg.19 Since it was evident that the US government would not substantially reduce the fiscal deficit, from 1968, the German government reluctantly accepted a series of mark revaluations; ultimately, it allowed the mark to float temporarily in 1971 and permanently after March 1973. Currency revaluations were politically costly because of resistance from the important financial and export sectors in West Germany.20 The ability of the United States to shift the costs of adjustment onto others, including the most important payments surplus country, stemmed from the position of the dollar in the international monetary system and America’s relatively low dependence on trade.
This kind of leader-follower relationship, where n-1 countries accept the nth country as the monetary standard-setter, will only work if the leader’s market is dominant in global trade and if its policies provide reasonable stability. Such stability usually means that the center country pursues a relatively low inflation monetary policy. The United States acted in this stabilizing manner from about 1950 until the mid1960s, and West Germany did the same in the EMS until 1990. In both cases, however, these center countries eventually exploited their dominant systemic position to pursue destabilizing policies that shifted the costs of adjustment onto other countries.21
One might conclude that the obvious solution is to abandon pegged exchange rates altogether for floating rates, as West Germany did in 1973. However, for many reasons countries are reluctant to adopt fully floating exchange rates, including high levels of trade openness, commodity export dependence, foreign indebtedness, low internal inflation-fighting credibility, and regional integration objectives.22 Nor do flexible exchange rate policies remove the dilemmas that interdependence creates; they merely alter their form. For example, in a floating rate system, national monetary expansion stimulates demand via lower interest rates and exchange rate depreciation, but it can also worsen the bilateral trade balance of other countries with the devaluing country. This could induce competitive devaluations or protectionist retaliation. The negative foreign impact is lower if national monetary expansion raises the rate of growth, or (in the case of a large country) if it lowers world interest rates. Expansionary fiscal policy, though less powerful for the home country under floating rates, may affect foreign output positively via exchange rate and output effects, and negatively by raising world interest rates. Hence, the demand from small open economies for policy coordination in a floating rate regime could be higher than under fixed rates. The political problem is that large countries, whose policies have most impact on the rest of the world, have fewer incentives to coordinate macroeconomic policies. Others may simply have to bear the costs of adjustment that emanate from large countries’ policies.
INTERNATIONAL FINANCE AND RISK
We noted above that a key determinant of the pressure for adjustment on individual countries is the structure and availability of international finance (often called liquidity). The first line of defense for a deficit country is national monetary reserves, though given the size of today’s foreign exchange markets these are often inadequate in the face of a concerted market attack. Today, monetary reserves are mainly held in the form of highly liquid government debt denominated in only a few currencies, above all the US dollar.23 The emergence of such “key currencies” creates a basic asymmetry in the system between key currency countries and the rest, since it can give the former considerable influence over the supply and price of international finance.
There are three main sources of international finance: private international capital markets, key country external deficits, and public international finance provided by bilateral lenders or by international financial institutions (IFIs). The relative sizes of these sources of finance matters greatly, as does the willingness to lend of those who control them.
Before we discuss in more detail the different sources of international finance, it is helpful to consider the different forms of risk that arise in international borrowing, though note that risk categories can overlap and are often interdependent. For borrowers, maturity and currency risk are often emphasized, though we also emphasize here “reversal risk” and “sovereignty risk.” Maturity risk occurs when the timing of repayments of interest or principal on financial obligations differs from the timing of cash receipts by the borrower. For example, if a borrower uses the receipts from a loan that must be repaid within three months to invest in assets that will not generate positive cash receipts for some years, the borrower is accepting a maturity mismatch risk (“borrowing short and investing long”).24 Banks specialize in managing this kind of risk. Currency risk arises when the borrower borrows in foreign currency to finance investment in assets (or current consumption) that produce cash receipts denominated in another currency (usually domestic currency).
Reversal risk arises when a borrower is dependent on flows of new external finance that can be stopped or reversed, potentially provoking a recession or even a debt crisis. Such stops and reversals may result from policies adopted by foreign creditor governments or multilateral institutions, or from changing attitudes to risk in private financial markets (such as so-called capital strikes and investor panics resulting in “capital flight”). Finally, sovereignty risk occurs when a country obtains finance from an external lender who requires the government to adopt policies it would not otherwise choose (“conditionality”), or when foreign financing reduces national control over important assets, technology, resources, and sources of employment.
From the perspective of lenders, maturity and currency risks also exist and must be managed. For example, investors bear currency risk in the form of the possibility that the exchange rate between the currencies in which the assets are denominated changes relative to the investor’s base currency. Additional sources of risk for lenders/investors include credit risk (the risk of borrower default), market risk (the risk that the market value of the assets in their portfolio are uncertain and can fluctuate), liquidity risk (the risk that it may be difficult to sell an asset in the future for cash without accepting a large discount on its original purchase price), legal risk (the risk that the interpretation or application of legal contracts is uncertain), and political risk (the risk that asset values are affected by changes in a country’s government or its policies). Note that when lenders take measures to reduce some forms of risk, they may increase risk for borrowers, such as when lenders shift currency risk to borrowers, or if they use conditionality to limit political risk. Moreover, attempts to limit some forms of risk may increase other risks, such as when lenders use conditionality to limit the risk of policy change, but their doing so increases the risk of a change in government and overall credit risk.
Private International Finance
Private lenders lend voluntarily to foreign borrowers only if they expect to be repaid with a profit, which is often seen as the reward for assuming one or more of the risks we have identified. Such lenders include banks making loans, bond and equity investors, and foreign direct investors.25 The borrower may be a foreign government or monetary authority or individual firms or banks, though in aggregate this can allow a country to run a current account deficit. From the perspective of borrowers, the maturity risks of these various forms of capital inflow differ considerably. Long-term debt finance (loans or bonds) and FDI can be more stable than short-term debt or equity inflows (i.e., less subject to sudden stops or reversals).26 Debt finance also generally requires fixed contractual repayments at specified intervals, whereas the dividends and profits paid to foreign investors on equity inflows and FDI usually involve no such legal guarantees of size or timing. FDI and foreign purchases of shares in domestic listed companies (equities) usually also require foreign investors to bear the currency risk.
By contrast, until very recently most borrowing in the form of bank loans and bonds by developing countries has been denominated in convertible foreign currencies, so that the borrower has borne the currency risk. Why this is so is a matter of debate, since governments of developed countries usually do not find it difficult to borrow from international lenders in their domestic currency. For example, nonresidents held on average 16% of UK sterling-denominated government debt over the period 1990–2002 (of this, 6% was held by foreign central banks and 10% by others, mainly private investors).27 The comparative historical difficulty that most countries have had in borrowing in their own currency is commonly referred to as the problem of “original sin”—though note that the incidence of own currency borrowing has increased since this became a matter of academic interest.28 The historical phenomenon may be due to investors’ perception of greater currency, political, and liquidity risk in developing countries because of relatively weak domestic institutions, as well as investors’ desire to limit their portfolio diversification by holding assets in a limited number of currencies. It is too soon to tell whether the recent ability of some developing countries to borrow from foreign investors in their own currencies is a temporary phenomenon or whether the permanent redemption of original sin is possible after all.
Whatever its origins, currency risk can be very costly for borrowing countries because depreciation vis-a`-vis the debt denomination currency increases the real external debt burden. At the end of 1997, Korea’s liabilities to international banks maturing within one year totaled $59 billion, almost three times its monetary reserves of $20 billion.29 Much more than the problems of competitiveness discussed earlier, it was the growing reluctance of international banks to refinance this short-term debt that precipitated the crisis of late 1997. As the Korean currency collapsed in December 1997 from 1,100 to (at worst) 1,960 won per dollar, the implied won value of Korea’s external debt soared. As many of the Korean banks and corporations who had borrowed from international banks were now unable to repay, the currency crisis soon became a corporate and banking crisis as well.30 The story in Thailand and Indonesia was very similar.
Bonds were historically important in private international financing in the nineteenth century and have become important again since the 1990s. International securities often have deeper secondary markets than bank debt and so can reduce liquidity risk for investors. But sudden reversals of bond investor sentiment can produce payments crises, such as that which followed the panic in the Russian ruble-denominated government debt (“GKO”) market in 1998. A similar crisis had occurred in Mexico in 1994, after the Mexican government had tried to reassure investors who had previously purchased peso-denominated short-term government bonds (cetes) by issuing US dollar-denominated bonds in their place (tesobonos). Once the rising stock of teso bonos had become larger than Mexican dollar reserves, investors panicked anyway, leading to a severe currency and debt crisis. Mexico’s assumption of the currency risk had increased the credit risk for investors. The same was true in the more recent case of Argentina, which issued large amounts of dollar-denominated public bonds in the 1990s. The credit risk for investors initially appeared to be limited because of Argentina’s currency board system and “hard” peg to the US dollar. When rising indebtedness and recession eventually led to currency crisis, peso depreciation resulted in Argentinean default. Hence, in lending to developing countries there can be a trade-off for creditors between currency risk and credit risk.
As we have noted, portfolio equity inflows or FDI both allow the borrowing country to avoid currency risk and a substantial amount of cash-flow/credit risk (since there are usually no fixed payments liabilities attached to such borrowing).31 The relative stability of FDI may derive from the illiquidity of the assets compared to relatively small holdings of equity or debt securities, and the possibility that FDI investors take a longer term view compared to bond investors. In the event of currency depreciation, the value of local FDI or equity stakes may even increase if the assets are in the traded goods sector.
These advantages of FDI and (to a lesser extent) portfolio equity inflows have often been outweighed for developing countries by the perceived disadvantages of foreign ownership and control. In Latin America in the 1970s, and Korea in the 1990s, bank debt as a form of deficit financing was generally preferred because it did not entail loss of control over important corporate assets, technology, resources, and sources of employment (see chapter 6). The vulnerability this borrowing created was thus in part the result of these countries’ restrictive policies toward inward equity flows and FDI, stemming from their desire to prioritize the reduction of perceived sovereignty risk.32
Key Country External Deficits
To the extent that key currency countries run large overall payments deficits and allow other countries to accumulate their debt as monetary reserve assets, this can provide another source of international finance. Countries that accumulate key currency reserves can use these reserves as a source of emergency finance in the future if their payments position deteriorates, or to pay off existing debts. Triffin pointed out that overall US payments deficits became the most important source of new international liquidity in the international monetary system from the late 1950s.33 He argued that if the United States were to undertake adjustment measures to reduce its external deficit, this would prompt offsetting policy adjustments by other countries and result in a global recession.
There has been debate over the stability of this form of international finance, since center country deficits can become excessive, resulting in global inflation and exchange rate instability. Foreign exchange reserves are subject to substantial currency risk, as holders of sterling reserves discovered in the early 1930s and holders of dollar reserves have discovered since 1971. When particular countries accumulate large amounts of such reserves predominantly in one currency (usually US dollars), attempts to reduce such currency risk through diversification may backfire if they prompt others to sell dollars. China and other large holders of dollar reserves are in this situation today, as were Japan and Germany in the 1980s. In addition, there is some reversal risk inherent in holdings of dollar reserves, since the center country could conceivably freeze existing foreign asset holdings to render them unusable or even repudiate its outstanding debts.
The strong incentives for other countries to hold key currency monetary reserves can also be a means by which the key currency country can finance its own current overall payments deficits. Again, the best-known example is the United States since the 1960s. Other countries’ willingness to hold additional US government liabilities (Treasury bills and bonds) as monetary reserves allowed the United States to run continuously large overall payments deficits with apparently little consequence other than rising external liabilities. Although interest is paid to holders of Treasury debt securities, the real return on such debt has historically been very low, suggesting that this source of finance for the United States has been unusually cheap and highly profitable.34 President de Gaulle of France termed this America’s “exorbitant privilege,” although his analysis played down the important role of US payments deficits as a source of finance to the system as a whole.
Public International Finance
For the many developing countries that receive low net private capital inflows, outright grants (aid), debt forgiveness, and new loans from foreign governments or international organizations can be important sources of international finance.35 Occasionally, in the late nineteenth century, central banks in the major countries provided short-term emergency loans to their foreign counterparts, such as when the French and Russian central banks provided emergency loans to the Bank of England in the 1890 and 1907 crises.36 Balance-of-payments loans were also made by the League of Nations in the 1920s.37 However, as we argue later, the idea and the bulk of the practice of public international finance is closely connected with the rise of macroeconomic stabilization and welfare objectives in the mid-twentieth century.
In all cases, public international finance is premised on some form of market failure. Today, the IMF is the main international financial institution (IFI) that provides short-term international finance to countries suffering payments problems. The IMF’s role is often justified by the instability of private sector finance. It was originally seen as a means of maintaining the pegged exchange rate system in the face of potentially unstable market speculation. One aspect of this was the policy surveillance function of the IMF, intended to ensure that national economic policies were compatible with systemic exchange rate stability. Another aspect was the IMF’s role as provider of short-term public international finance to countries suffering temporary balance-of-payments deficits. Keynes saw this public international finance as a means of avoiding costly deflationary adjustment policies of the kind that occurred in the early 1930s. The justification for long-term public international financial assistance, through international institutions like the World Bank and bilateral aid programs, has also been based on the assumption that private long-term financing will be insufficient to promote economic development in the poorest countries.
In practice, IFI lending, especially in the case of the IMF, has become associated with restrictive policies rather than the maintenance of high levels of domestic demand and employment. This is because of the macroeconomic policy requirements for borrowing countries that are standard preconditions of such loans, commonly referred to as policy conditionality. One likely reason for such policy conditionality is that the IMF’s resources only consist of the limited pooled contributions of member states, and the major countries have had strong incentives to ensure that borrowers repay the Fund. In addition, there is evidence that IMF staff recruitment and promotion has been biased in recent years toward individuals with non-Keynesian, neoclassical views.38 We return to this issue later, but it is clear that the governance of the IFIs is of central importance in determining the policy conditionality faced by sovereign borrowers. This can dramatically affect the level of sovereignty risk for borrowing countries and the preferred balance between international financing and adjustment. There is also some reversal risk in borrowing from the IFIs (e.g., in the event of borrower noncompliance with core policy conditionalities), and currency risk (since the IFIs lend in hard currencies).
NATIONAL STABILIZATION POLICY AND INTERNATIONAL MONETARY ORGANIZATION
We have argued that the balance between adjustment and financing in international monetary organization is greatly complicated when countries adopt domestic stabilization objectives. In this section, we argue that the evolution of international monetary organization over the past century can be seen as an uneven process of adaptation to a dramatic shift in domestic monetary organization and economic policy. We postpone the question of “feedback” from the international monetary and financial system to national monetary organization and policy to the next chapter.
The Development of National Money
National monetary systems of the kind assumed in standard modern macroeconomics are a relatively recent historical phenomenon, though the gap between theory and practice has always been considerable.39 The growing centralization of national monetary organization is strongly associated with the development of the modern state, though only in the twentieth century was the “nationalization” of paper money generally established. The institutions associated with national monetary organization are also of recent vintage. By 1900, only 18 countries had central banks, and all but two (Sweden’s and England’s) were less than 100 years old.40 The rise of paper (“fiduciary”) money issued by central banks eventually laid the institutional foundations for national monetary management, but this also took some time (and the acceptability of such money initially depended on it being backed by reserves of gold or silver).
The possibilities of “monetary policy” were also limited by intellectual constraints that were only gradually overcome in the early part of the twentieth century. The Bank of England’s suspension of gold convertibility from 1797 until 1821 mobilized the defenders of orthodox finance, including Ricardo and the “Currency School,” who saw in the emergence of a domestic paper money standard the road to financial ruin.41 The restoration of the fixed gold convertibility of sterling was seen as a crucial constraint against inflation. Another was the legal restriction of the quantity of Bank of England note issuance, notably in the Bank Act of 1844. But politics pushed in the other direction: the 1844 act also consolidated the emerging monopoly position of Bank of England notes.42 The legal and political dependence of central banks on governments generally rendered them subject to periodic pressure to provide cheap deficit financing, particularly during national emergencies or when their charters were up for renewal. The issuance of new notes, the production cost of which was minimal, allowed the state or central bank to expropriate real private resources (“seigniorage”).43
The emerging national monetary hierarchy, with central bank notes becoming “as good as gold,” was reinforced by restrictions on the circulation of foreign currencies and the suspension of convertibility for foreign coins at national mints. Commercial banks increasingly held their reserves at the central bank. This enabled the central bank to act as a “lender of last resort” (LLR) when panic spread through the banking system, although in most countries such LLR responsibilities were accepted only after the 1930s.44
The International Gold Standard
The pre-1914 international gold standard was not a product of collective international design of the kind achieved at Bretton Woods in 1944. There was a clear but gradual trend toward the gold standard in Europe from 1873, led by Germany, the Netherlands, and Scandinavia.45 Britain’s adherence to a gold standard entrenched a growing international role for sterling, with other countries using bills issued by London finance houses to finance international commercial transactions. As both foreign private and public actors accumulated sterling assets for financial purposes, this benefited the City of London, which naturally favored the gold standard.
In theory, there was no room for autonomous monetary policy in the gold standard. The so-called rules of the game (or more accurately, behavioral norms) included the commitment to buy and sell national currency for gold without limit at a fixed price and to allow the free cross-border flow of gold. In practice, central banks diverged in various ways from these norms. The Bank of England relied heavily on the manipulation of its discount rate because of its slim gold reserves; sometimes it even took account of the domestic “state of trade” in setting its level.46 In France and Germany, central banks used various devices to limit gold convertibility so as to achieve a degree of national monetary autonomy.47
Nevertheless, before 1914 there was little pressure for outright inflation in the major countries, allowing their central banks to maintain effective independence in practice. Crucially, in Britain, the external commitment to fixed gold convertibility had priority. The limited appreciation of how interest rates affected the real economy and the limited political influence of Britain’s working classes also helped to insulate the central bank from pressure that might have compromised this commitment. The political dominance of the financial bourgeoisie and landed interests, which held a substantial proportion of their financial assets in fixed-income domestic and foreign bonds, reinforced the political preference for low inflation and fixed exchange rates.48 By contrast, in some peripheral countries in Europe and the Americas, persistent fiscal deficits and political instability produced inflation and undermined attempts to peg to gold.49
World War I brought similar levels of political and monetary instability to the heart of the European-dominated international monetary system. The gold standard was suspended in the main belligerent countries at the outbreak of the war. After 1918, the widening of the electoral franchise in Britain and other countries and rising political demands from the working classes everywhere undermined the domestic political foundations of the gold standard. However, the postwar monetary turmoil in Europe ensured that political and economic elites looked back to the prewar gold standard as a beacon of economic and political stability. This set the stage for growing political conflict over national and international monetary arrangements in Europe.
A fundamental problem for all the major European countries was the enormous stock of public debt caused by heavy wartime reliance on government borrowing.50 To make matters worse, in some countries most of this debt was short term. Reducing the level of national debt to sustainable levels required either more tax increases or refinancing the debt at longer maturities and lower interest rates.51 The distributive implications of these alternatives provoked intense social conflict, hampering the process of monetary and currency stabilization. Debtors (including the state and corporate sector) had an interest in inflation. The working classes, which had made enormous sacrifices during the war and which had few assets, also demanded increased government expenditure on social welfare. Creditors, including savers and the financial sector in particular, had most to lose from inflation and lobbied strongly for a return to the gold standard, as well as cuts in public expenditures and taxes.
The years of inflationary chaos in the early 1920s strengthened the postwar creditor backlash and bolstered the ideological appeal of the gold standard. This facilitated a general movement back onto the gold standard, the most notable example being Britain’s decision to return to gold in 1925.52 More direct international pressure, particularly from the United States and Britain, was also important in some cases, such as the international stabilization schemes sponsored by the League of Nations in countries like Austria and Hungary. These schemes included new central banking statutes designed to promote monetary orthodoxy.53 The 1924 Dawes loan, which enabled Germany to return to the gold standard, also included a proviso for the Reichsbank to be formally independent of the government. For a few short years this restored gold standard appeared to be working reasonably well, but by the early 1930s, it lay in ruins. Britain’s departure from gold in the crisis of late summer 1931 marked the end of the system, though it was not until 1933 and 1936 that the United States and France, respectively, left gold. Why did this restoration experiment fail?
One argument is that the gold standard was ultimately incompatible with postwar governments’ new short-term domestic macroeconomic objectives. These objectives lowered the credibility of the commitment to a fixed exchange rate and raised the cost of maintaining it.54 In this view, wider enfranchisement and growing political influence of the working classes increased demands on governments to make priorities of domestic output and employment objectives rather than the exchange rate and low inflation. Eventually, domestic politics demanded a choice in favor of these incipient “Keynesian” domestic policy objectives, and the gold standard was abandoned.
The claim about the impact of the widening of the electoral franchise is more consistent with the British case than with other countries’ experiences. A number of European countries had adopted nearly universal male suffrage before the era of the gold standard, with apparently limited cost in terms of monetary credibility and stability (in 1848 in France, and 1871 in Germany). Conversely, restricted franchises in Russia, Austria-Hungary, and Italy coincided with high national debts and long-term interest rates and exchange rate instability.55 Still, Eichengreen and Simmons are correct to suggest that in many countries leftist political parties were more likely to gain political power after the war than before.
Since Eichengreen and Simmons are also correct that financial markets judged many governments’ gold standard commitments as lacking credibility in the interwar years, what was the source of the problem? Although financial markets in the 1920s had no modern economic models to deduce a policy contradiction between employment and exchange rate objectives (the Mundell-Fleming model was developed in the 1960s), it is possible that they possessed inductive knowledge of this contradiction. Another possibility is that financial markets were simply concerned that governments would be tempted to inflate their way out of their fiscal and debt overhang problems (as some had tried in the early 1920s). More research is required to decide between these two explanations, since both are consistent with the fact that financial markets reacted negatively to evidence of fiscal and monetary laxity and departures from gold standard rules.
Both arguments are compatible with the general claim that the process of democratization, reflected in the growing importance of mass politics, was of central importance in undermining the credibility of the commitment to a “hard” exchange rate peg. Central bank independence (CBI) from government could sometimes partially offset this trend: countries like France with highly independent central banks made a more credible commitment to gold despite mounting unemployment, in contrast to countries with more subordinate central banks, such as Britain, Sweden, and Japan.56 In the years that preceded he Keynesian revolution in economic theory, the absence of a clear intellectual alternative to the gold standard reinforced elite attachment to a system that eventually produced outcomes dysfunctional from most points of view.57 From 1932, British policymakers began very tentatively to experiment with cheap money policies that promoted higher levels of employment.
These arguments about the importance of domestic political change in the major European countries in undermining the gold standard are at odds with the standard argument about the absence of hegemony as the prime cause of interwar monetary instability. In this view, British decline and a newly preponderant America’s unwillingness to assume the burden of leadership undermined the international gold standard. As Kindleberger famously put it:
The world economic system was unstable unless some country stabilized it, as Britain had done in the nineteenth century and up to 1913. In 1929, the British couldn’t and the United States wouldn’t.58
This argument has some merits in focusing attention on the shortcomings of international monetary cooperation in the late 1920s and early 1930s, but it has been criticized on many other grounds. It underplays the attempts by Britain and the United States to restore the international gold standard in the 1920s, the role of policy mistakes in Britain and elsewhere, and the destabilizing effects of the war and postwar settlement. It also ignores the domestic political conflict that arose from the financial and economic legacy of the Great War, and the rising importance of socialist politics in undermining the commitment to gold.59
A New International Monetary System: Bretton Woods
The Bretton Woods agreement of 1944 was a true watershed, representing an attempt to found a new kind of international monetary order on the explicit acceptance of new domestic macroeconomic objectives.60 The establishment of two new international institutions, the IMF and World Bank, introduced a greater element of collective management of international money and finance than in the past. Even so, elements of historical continuity included a pegged exchange rate system, a monetary anchor role for gold, and a reserve role for both gold and major key currencies.
Much IPE literature initially emphasized the importance of US hegemony in the construction of the Bretton Woods system.61 Although US leadership was indeed crucial to the outcome, hegemony theory does not explain the details of the Bretton Woods system. “Social” democracy and associated new Keynesian thinking about policy combined to ensure that many governments in Europe saw high-employment policies as an essential ingredient of postwar political reconstruction.62 National demand-management policies and in many cases state-directed industrial policy were seen as the means to achieve these goals. In developing countries, similar policies were pursued for nationalist and developmental reasons even in the absence of democracy. Almost everywhere, national fiat money systems were entrenched and central banks placed under government control. In the Bretton Woods agreement it was accepted that governments would and should use national stabilization policy to manage domestic output, employment, and price objectives. Keynesian ideas were not fully victorious, particularly in the United States, but classical monetary orthodoxy had collapsed almost everywhere, and financial interests were often seriously weakened by the Great Depression and war.63
The Great Depression had delegitimized the gold standard because of its association with deflation and mass unemployment. Keynes had famously described the gold standard as a “barbarous relic” and declared its incompatibility with active demand-management policy. Yet the United States, which by the 1940s held most of the world’s gold reserves, insisted that gold be retained alongside convertible currencies as a monetary reserve asset and as an anchor for the exchange rate system. The Bretton Woods conference therefore agreed to adopt a more flexible “gold-exchange standard.” Since domestic monetary policy autonomy was prioritized, capital controls were allowed (IMF Article VI.3).64 All currencies would be pegged to gold or to the US dollar (IMF Article IV.1.a), but no country would be required to defend an exchange rate that produced a “fundamental disequilibrium” in the balance of payments (IMF Article IV.5). This meant that pegs could be adjusted in such circumstances, though in principle adjustments required consultation with other countries through the IMF to prevent competitive devaluation. The concept of fundamental disequilibrium was left undefined, but the general understanding was that persistent overall payments imbalances would require exchange rate adjustment. These and other ambiguities in the Bretton Woods system played a constructive role in facilitating agreement and in providing for subsequent flexibility.65
The postwar international monetary system soon evolved into a gold-dollar system, whereby the United States maintained dollar convertibility into gold at the fixed (1934) price of $35 per ounce. Other countries could freely hold dollar reserves and could, if they wished, sell these reserves for gold at the official price, either in the private gold market, or by presenting them for conversion at the “gold window” of the US Treasury. The emerging Cold War also resolved the decades-long problem of persistently large US payments surpluses by facilitating US acquiescence to large European devaluations over 1948–49 and costly American aid and troop deployments in Europe and Asia. These helped to produce the large US payments deficits from the early 1950s that later became associated with the Bretton Woods system and which provided a large proportion of the new liquidity in the postwar international monetary system.
The room allowed in the Bretton Woods agreement for domestic macroeconomic activism and the potentially adjustable character of the exchange rate pegs arguably reduced their credibility. As a result, timely changes to currency pegs were encouraged. Furthermore, the IMF was authorized to use its pool of member contributions to make short term loans to members who needed to defend their exchange rate. National quotas were to be comprised of 25% gold and 75% national currency, but countries could borrow up to 125% of this quota (with successive “tranches” having more conditions attached), repayable within three to five years. From the Keynesian perspective, a temporary deterioration in a country’s payments position should be financed to counteract the “deflationary bias” in the system that derived from the special vulnerability of deficit countries. As Article 1(v) of the Bretton Woods agreement stated, a key objective was
To give confidence to members by making the Fund’s resources available to them under adequate safeguards, thus providing them with an opportunity to correct maladjustments in their balance of payments without resorting to measures destructive of national or international prosperity.
By late 1947 it had become clear that the IMF’s total resources were wholly inadequate to achieve this objective, but the Cold War intervened and the United States provided bilateral financial aid to western Europe in the form of the Marshall Plan. IMF quotas have been increased periodically since 1944, but the Fund’s resources have not kept pace with the growth of the world economy. Later, as capital mobility and pressures on the exchange rate system increased, the major industrial countries supplemented these limited facilities in the 1960s with the Gold Pool (1961), swap facilities (1962), the General Arrangements to Borrow (GAB) (1962), and SDR allocations (from 1969).66 These new financing arrangements largely benefited the industrial countries. The shortage of IMF liquidity ensured that in practice most adjustment pressure remained on deficit countries, especially those in the developing world.
The Cold War had another effect on the balance between financing and adjustment. Although IFI lending and associated policy conditionality was meant to be based on politically neutral criteria, the rise of the Cold War seriously strained the credibility of this neutrality for both the IMF and the World Bank.67 This was compounded by the fact that voting power in the Executive Boards of the Fund and Bank were weighted according to country contributions, with the United States possessing the largest (and wielding effective veto power over important decisions). After the withdrawal of the USSR and mainland China from the Bretton Woods institutions, the United States and its European allies dominated both IFIs. Other countries aligned with the dominant Western powers tended to get better financing deals and very weak enforcement of policy conditionality.68
FROM BRETTON WOODS TO GLOBAL FINANCIAL INTEGRATION
By the early 1960s, the Bretton Woods system had evolved in directions unforeseen in 1944 and was in serious difficulty. It limped on until the early 1970s only through US unilateral actions and America’s ability to obtain the support of most of its political and military allies. The growing political demands for domestic stabilization policies contributed to the further marginalization of gold in the international monetary system and the emergence of a “dollar standard.” Another major long-term development was the reemergence of private international financial markets in the 1960s. Together, these processes undermined the pegged exchange rate system and increasingly marginalized the institutions of public international finance established at Bretton Woods.
From Gold-Exchange Standard to Dollar Standard
It was already clear by the late 1950s that the gold-exchange standard was in difficulty. In the late 1950s, US gold losses accelerated as large overall US payments deficits were financed by the accumulation of dollar reserves by central banks in Europe and elsewhere. At the same time, steady wartime and postwar inflation had eroded the real price of gold (given its fixed nominal price of $35 per ounce), making it less profitable to mine and encouraging excess private demand. The growing shortage of gold’s supply relative to dollars led to growing speculation against the dollar.69
Triffin argued that these developments showed that the gold-exchange standard was “inherent unstable” and was bound to collapse.70 The sustained expansion of international trade, he argued, increasingly depended on new liquidity provided by US payments deficits, but the growing relative shortage of gold compared to dollars would undermine the fixed-price relationship between the dollar and gold. If the United States tried to maintain the dollar price of gold by reducing its overall payments deficits, a global liquidity crisis and recession would result. Fearing the latter outcome, Triffin argued for replacing both gold and key currencies with a new international fiduciary money that could be created and managed by the IMF (reviving Keynes’s “bancor” proposal of the early 1940s).
Triffin’s diagnosis of the weaknesses of the Bretton Woods system was perceptive, but his political economy analysis was less strong. First, to expect the United States to reduce its external deficit was naive, given the growing domestic pressure for welfare spending and job creation in the United States and the large contribution of FDI, military, and aid outflows to America’s external deficits. Second, Triffin’s supranational solution was utopian and at odds with domestic political considerations in the United States and elsewhere. In reality, there were only three plausible solutions to the gold-dollar problem. One solution, eventually adopted, was to break the link with gold permanently and to move to a dollar standard.71 A second solution was to maintain the gold-dollar standard by periodically revaluing gold against all currencies in the system, including the dollar.72 A third, which was adopted after the collapse of the short-lived dollar standard, was to allow the major currencies to float against each other.
Politics initially strongly favored the dollar standard over the second and third solutions. The gold revaluation proposal was politically unappealing for a number of reasons. First, it overlooked that continuous postwar inflation in the major countries rendered an anchor role for gold increasingly untenable: periodic gold revaluations would have been tantamount to accommodating this inflation, undermining the core rationale for a commodity price anchor. Second, the United States was increasingly unwilling to accept the constraint on its own macroeconomic policy flexibility that gold convertibility implied.73 Third, loyal allies of the United States like West Germany and Japan were willing to hold dollar reserves even if they were not convertible into gold. Only France under de Gaulle, chafing under American preponderance, felt sufficiently autonomous of Washington both to withdraw from NATO’s military structure and to sell its dollar reserves. Vocal French demands for a return to a “politically neutral” gold standard also hardened US political opinion against gold revaluation, already resisted out of concern that it would favor the “gold bugs” (speculators) and unsavory gold-producing countries like the USSR and South Africa.
If a reinvigorated gold exchange standard was politically unappealing, the flexible exchange rate proposal also enjoyed very little support in the late 1960s except among a few maverick economists (we discuss this further in the next chapter). This left the dollar standard as the only alternative that was acceptable to the United States and its major allies. The “Two-Tier” agreement of 1968 ended official attempts to maintain the fixed price of gold in the private gold market, though the $35 per ounce price was retained for transactions between central banks. This was the first formal step toward the demonetization of gold in the international monetary system, though it did little to relieve the pressure on the dollar and continuing losses of US gold reserves. In August 1971, President Nixon unilaterally decided to close the gold window and in December that year obtained the agreement of other major countries to a formal dollar standard, in which other currencies were pegged to a gold-inconvertible US dollar. As we discuss in chapter 5, the dollar standard proved even more unstable than the gold-dollar standard, but the position of key currencies as reserve assets in the international monetary system has become ever more entrenched since then.
The Reemergence of Financial Integration
The second major change in the international monetary system over the course of the 1960s and 1970s was the reemergence of private international capital markets. There is a large literature describing the dramatic increase in short-term capital mobility and longer-term international financial flows since the 1960s that need not be repeated here.74 One indication of the importance of international capital flows is the very high level of daily global turnover in foreign exchange markets, which reached $1,900 billion in April 2004, compared to $620 billion in 1989.75 Another indication is the increase in the stock of financial assets owned by foreigners. According to McKinsey, by 2005, “foreigners [held] 12 percent of US equities, 25 percent of US corporate bonds, and 44 percent of Treasury securities, up from 4 percent, 1 percent and 20 percent, respectively, in 1975.”76 Although there is general agreement about the rapidity of the growth in financial integration, there is debate and some skepticism concerning its absolute level.77 As figure 4.2 shows, although most regions have seen an increase in levels of capital account openness since 1970, financial openness is higher in developed countries and trends vary considerably by region.
Given the potentially transformative implications of greater financial openness, we must first ask what explains this broad trend and the varying cross-regional and cross-country patterns we observe. Here, we discuss international, domestic, and ideational explanations. We focus on the removal of barriers to short-term or portfolio capital mobility, especially capital controls, but keep in mind that financial liberalization more broadly also encompasses the liberalization of domestic financial markets, including the removal of controls on financial activities and on entry into financial businesses, including by foreign financial firms.
The most general explanation is that international competition between states promotes financial liberalization over time.78 As we saw in chapter 3, realist accounts of trade policy have claimed that interstate competition tends to promote either unilateral protection or cautious, reciprocity-based trade liberalization. What is different about finance, realists have argued, is that unilateral liberalization improves the competitive position of national financial sectors and can attract multinational financial firms.79 Hence, relatively closed national financial systems such as generally prevailed in the 1950s and 1960s are seen politically unsustainable over time. Evidence for this claim is provided by the evident competition between global financial centers like London and New York, and within regions like Europe and East Asia, which do appear to have promoted unilateral removal of capital controls and of constraints on international financial business and entry by financial MNCs. The “Eurodollar” markets emerged in the 1960s because of the British authorities’ willingness to allow London banks to provide wholesale dollar-based financial services to foreign residents, even though capital controls on sterling transactions remained in place until the late 1970s. The United States responded at the end of 1973 by removing the remaining capital controls that inhibited Wall Street’s international financial role and encouraged American banks to conduct international business in London. A related argument is that such competition can spiral out of control, causing a race to the bottom in financial regulation, taxation, and financial transparency as ever more jurisdictions compete for international financial business. Concerns have focused particularly on “offshore financial centers” (OFCs), which offer low taxes, light financial regulation, and substantial anonymity for individuals, companies and financial institutions.80
However, not all governments choose to compete for international financial business. The removal of capital controls has also taken place very unevenly, both over time and across different countries.81 At the least, we need to explain why the United States and a few other countries (such as Switzerland, Canada, and West Germany) removed capital controls much earlier than others. Helleiner’s argument is that the relative political influence of financial interests within these countries helps to explain their decisions, which takes us into domestic politics. Another, compatible explanation is that the IT revolution substantially reduced the costs of international financial transactions, providing increased incentives for governments to engage in competitive liberalization. The related dramatic fall in transactions costs has also unleashed successive waves of innovation in financial markets, which has spurred their growth and further raised the costs of financial protectionism. As we will see in chapter 6, such technological explanations have been applied to growing capital mobility in general.
Others retain an international explanatory focus by arguing that the major Western countries, above all the United States, have actively promoted financial liberalization abroad to enhance their structural power within the global system.82 Although this argument also has realist origins, it interprets the general trend in more coercive terms. Potential benefits for the United States include greater business opportunities for highly competitive US financial firms, as well as the possible benefits that the United States as a whole may gain from the international use of its currency, and from financial inflows from other countries, including for the purpose of obtaining cheap finance for its fiscal and payments deficits.
There is evidence that the United States and the EU have promoted greater financial openness abroad in recent years through trade and investment negotiations (notably in the WTO financial services agreement of 1997, but increasingly through bilateral deals). Some also argue that the major countries have used the IMF to promote capital account liberalization abroad.83 But this overlooks the other side of the argument, especially that financial liberalization might benefit indigenous firms and that growing US dependence on foreign finance might diminish rather than enhance America’s global leverage.84 Moreover, international pressures to converge on open capital account policies were not always successful or especially strong. A recent independent review of IMF policy notes that although the IMF did try to persuade governments to liberalize, it lacks the legal authority to require capital account liberalization (a legacy of the Bretton Woods articles of agreement). Generally, it suggests that domestic factors were usually more important in governments’ liberalization decisions.85
Simmons and Elkins show that capital account liberalization has clustered both temporally and spatially, suggesting that domestic factors alone are insufficient to explain this pattern.86 One explanation for this is the competitive liberalization process discussed above. Another channel Simmons and Elkins identify is that countries may observe peer group policies for purposes of gaining information in a situation of uncertainty (i.e., a learning effect), an argument for which they find some support. The IFIs may have played a role in promoting this cross-border diffusion of policy trends.
A final international explanation of capital account liberalization is offered by Haggard and Maxfield, who argue that financial crises induce capital-poor developing countries to open their capital accounts to signal that the risk of future closure should not deter potential investors.87 It is doubtful, however, that any such signals are very credible to investors, since the cost of reintroducing capital controls in the future during a crisis may not be high. Indeed, figure 4.2 suggests that East Asia on average tightened rather than liberalized capital controls after the regional crisis of 1997–98.
The comparative political economy literature suggests a range of domestic interest group and institutional factors that help explain the complex pattern of liberalization. Interest-based approaches begin by specifying group preferences, typically using either the Stolper-Samuelson or specific factors models of trade theory. In the former case, in advanced economies capital as the abundant factor gains and labor loses from financial openness (since capital will be exported). In developing economies, labor gains from capital importation and domestic capital loses. Building on this model, Quinn and Inclán predict different labor preferences according to skill levels and test hypotheses about the circumstances under which left-of-center and right-of-center political parties will support or oppose financial openness.88
In contrast, Frieden uses the specific factors model to predict that financial industries will gain from financial openness and those with sector-specific assets lose because they will pay higher borrowing costs when capital flows abroad.89 In developing countries, industries with sector-specific assets win because they will pay lower borrowing costs. MNCs with internationally diversified asset portfolios will gain because of their ability to borrow at low cost and from their ability to make intrafirm financial transfers. Owners of liquid financial assets in developed countries will gain while those in developing countries will lose as interest rates rise in the former and fall in the latter. These predictions are consistent with some aspects of conventional wisdom, such as the apparently broad support for financial openness among firms and workers engaged in the financial sector of major countries since the 1970s, and among the multinational corporate sector in general.
Some difficulties emerge in both of these approaches, though neither has been subject to decisive testing to date. First, some of the predictions are not obviously consistent with the evidence. For example, domestic firms in specific sectors in the advanced countries may not have lost from capital openness, since many firms now enjoy access to a much greater pool of capital than before because of widespread liberalization across many countries. Indeed, both theories rely for their predictions on the standard neoclassical assumption that financial openness will promote capital exports from developed countries to capital-poor developing countries. It is not commonly recognized in the political economy literature that this crucial assumption is not supported by the evidence, with the United States acting as a net importer of capital from developing countries for many years.90 Second, it is one thing to attempt to identify those groups who win and lose from financial openness, but this does not mean that such groups will mobilize to lobby policymakers. As for trade policy, institutions matter for financial policy outcomes, including the political parties that Quinn and Inclán consider. However, political parties may represent multiple interest groups who are differently affected by financial opening. More generally, it is not clear how strong and coherent the preferences of interest groups will be, since financial openness raises a range of complex issues. As Frieden notes, financial openness has important implications for the exchange rate, so much may depend on whether specific assets are employed in the traded or non-traded-goods sector (see chapter 5, the section “Financial Integration and Exchange Rate Policies”). Voters possess multiple identities in practice: as savers, as workers or employers, as consumers, and so on, which can result in ambiguous preferences. Banks may have global fund management and international lending divisions that favor financial openness, but their domestic lending departments with corporate clients in the traded goods sector may favor exchange rate stability over capital openness. The majority of the benefits of financial liberalization tend to accrue to particular, often politically influential, groups, including financial firms and MNCs.91 The potential costs of such liberalization, such as greater exchange rate instability, lower macroeconomic policy autonomy, and financial crises tend to be delayed and more widely distributed among societal groups. Given this asymmetric distribution of costs and benefits and the obstacles to collective action among losers, liberalization is likely to win out in the long run.
If policymakers have some autonomy from societal interests in this area, what determines their preferences? One possible answer, addressed below, is economic ideas. Another is that the state itself, often the largest debtor within countries, may favor financial openness if this expands its borrowing capacity and lowers its cost of debt (a generalization of Loriaux’s argument from the previous section). Governments of countries with low domestic savings or underdeveloped capital markets might be particular beneficiaries. Governments facing highly independent central banks may enjoy little influence over monetary policy with or without capital controls, undermining the rationale for retaining them. Central banks themselves might favor capital mobility either because they are open to capture by the financial sector or because they believe it will constrain the government’s propensity to run fiscal deficits.92 These possibilities remain speculative at present.
Finally, what impact has democratization had on capital account policies? It is difficult to believe that there is a simple linear relationship between democracy and capital account policy. The restriction of the franchise in nineteenth-century Britain and the political dominance of the asset-owning elite may have favored the policy of financial openness under the gold standard until 1914. Ruggie’s argument that the “reembedding” of economic liberalism within a social democratic framework made capital controls politically acceptable pushes in the same direction.93 Others find that more recently, democracy has had a positive effect on financial liberalization in developing countries.94 Countries in which voters support social democratic welfare policies have moved in recent years toward greater financial openness (notably most of developed western Europe in a relatively coordinated fashion as part of the 1992 Single Market Programme). This might demonstrate that as these voters have become richer over time they have also come to favor financial openness to maximize returns on savings, or it may simply be that they are relatively poorly informed or weakly mobilized. Evidently, this important area is ripe for further research.
Capital account decisions may be influenced by particular ideas if policymakers have some autonomy from domestic and international interests. In the early post-1945 period, the orthodox (Keynesian) consensus among economists and technocrats was that capital controls were a necessary plank of national macroeconomic stabilization policy. By contrast, most contemporary economists argue that capital controls rarely work for long, create large inefficiencies in the global allocation of capital, and promote corruption.95 This broad ideational shift is associated with the rise of an anti-Keynesian policy consensus since the 1970s that monetary policy should focus mainly (if not entirely) on inflation stabilization and in favor of floating exchange rates.
The question of how much this new orthodoxy influenced policy outcomes is a difficult one. Optimism concerning the net benefits of capital account liberalization probably played a role in a number of decisions by developing countries to liberalize in the early 1990s, though the impact of neoliberal ideas on financial policy varies considerably across countries.96 Those Latin American governments adhering to the new “Washington consensus” concerning the benefits of neoliberal market reform were most prone to ideological conversion to the capital liberalization cause, Mexico’s economist-dominated government being the most notable example.97 But it is difficult to separate the role of ideas from that of pressure from powerful external actors. The US Treasury and the IMF proselytized in favor of financial liberalization in the 1990s, and the former was associated with proposals in the mid-1990s for an amendment to the IMF’s articles of agreement to give the IMF authority to promote capital account liberalization. These proponents played down the costs of financial openness: though they recognized at the conceptual level that premature capital account liberalization could be risky, in practice little attention was given to these concerns before the Asian crises of the late 1990s.98 Although key players in the US Treasury and IMF may have been convinced of the intellectual case in favor of financial openness, in both cases material interests arguably pushed in the same direction. Ideational and material factors also interrelate in the impact of theend of the Cold War, which led to a transition toward market democracies in many central and eastern European countries in the early 1990s.
An example of the role of ideas in influencing material outcomes is the effect of financial theory on the way in which financial markets operate. Finance theorists working within the neoclassical tradition on risk and portfolio optimization laid the foundation for modern financial techniques that revolutionized the financial sector. As Bernstein notes, some of these academic theorists went on to work in the financial markets, providing a direct link between academic ideas and financial practice.99 The rapid growth of the financial sector that this helped to produce increased its political influence, including over the nature and content of financial regulation. This also has made it more difficult for countries with substantial financial sectors to contemplate the reintroduction of controls that inhibit financial innovation and globalization.
In short, it seems clear that both international and domestic factors matter in explaining the uneven patterns of financial liberalization over the past few decades. Ideational and material factors are difficult to separate, though the relative importance of ideas may be greater in some countries than in others, depending among other things on the coherence of policymaking teams.100 Ideas also have been important in reshaping the financial sector in ways that have in turn influenced policy and the shape of financial regulation (see chapter 5).
We have argued in this chapter that the very uneven process of democratization in combination with Keynesian economic ideas promoted demands for national macroeconomic activism by governments and monetary authorities. Over the course of the past century, this process undermined the role of gold in international monetary organization and rendered pegged exchange rate systems increasingly unviable. In the 1970s, the principle of national macroeconomic activism came under challenge from monetarist and new classical economics, but most central banks have retained the Keynesian idea that discretionary monetary management remains a powerful policy tool. Domestic political pressure for monetary activism remains strong in most developed democracies.101
Although one of the components of national monetary activism was the often substantial use of capital controls to limit international financial interdependence, over time this proved unsustainable. The reemergence of private international financial markets has been driven by a range of international and domestic forces, both material and ideational. Combined with the continuing importance of national macroeconomic policy activism, growing financial interdependence has had a powerful impact on the shape of the international monetary organization, in particular by increasing the vulnerability of pegged exchange rate regimes. We consider this and other implications in more detail in the next chapter.
Harold James. International Monetary Cooperation since Bretton Woods. Washington, D.C.: International Monetary Fund; New York: Oxford University Press, 1996. An excellent history of international monetary relations since World War II.
Ngaire Woods. The Globalizers: The IMF, the World Bank, and Their Borrowers. Ithaca, N.Y.: Cornell University Press, 2006. Provides a clear analysis of the two main international financial institutions, their governance, and their relationships with developing countries that borrow from them.
Benjamin J. Cohen. Global Monetary Governance. London: Routledge, 2008. A collection of important essays on the evolution of the international monetary and financial system since the 1960s.
- http://web.pdx.edu/~ito/. Menzie Chinn and Hiro Ito’s “KAOPEN” database, which currently provides capital account openness indices for 181 countries over 1970–2005 and is periodically updated.
- http://www.imf.org/external/data.htm. The IMF’s data and statistics site, which includes a range of important databases including International Financial Statistics, World Economic Outlook, the Joint BIS-IMFOECD–World Bank Statistics on External Debt, the Dissemination Standards Bulletin Board, and the Annual Report on Exchange Arrangements and Exchange Restrictions.
- www.imf.org/ieo. The IMF’s Independent Evaluation Office site, which produces useful independent reports on the IMF’s role in international money and finance.
- http://www.brettonwoodsproject.org. The Bretton Woods Project site, an independent organization that monitors the IMF and World Bank and offers reform proposals.
Return to Book Description
File created: 10/13/2008
Questions and comments to: email@example.com
Princeton University Press