Jump to ContentJump to Main Navigation
Pension FundsRetirement-Income Security and Capital Markets: An International Perspective$

E. Philip Davis

Print publication date: 1998

Print ISBN-13: 9780198293040

Published to Oxford Scholarship Online: March 2012

DOI: 10.1093/acprof:oso/9780198293040.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 26 February 2017

(p.288) Appendix 2. Pension Funds and the European Exchange Rate Crisis 1992–1993

(p.288) Appendix 2. Pension Funds and the European Exchange Rate Crisis 1992–1993

Source:
Pension Funds
Publisher:
Oxford University Press

Introduction

Having exhibited quite remarkable stability since 1987, when the previous realignment occurred, the European Exchange Rate Mechanism (ERM) began to suffer tensions in September 1992, which lasted until the end of July 1993. The result was the suspension of ERM membership of the UK and Italy, the abandonment of informal pegs to the Ecu by the non-EU Scandinavian currencies, and the broadening of the permitted fluctuation band of the remaining currencies vis-à-vis the Deutschmark, other than the Dutch guilder, to ±15%—little different from a regime of free floating, except that the mechanisms of the ERM had been preserved and the possibility of a future return to narrow band was not excluded. A subject of considerable controversy—and relevance to the analysis of Chapter 9—is whether the portfolio strategies of pension funds and other institutional investors made a major contribution to the breakdown of the ERM, and whether such a breakdown was unjustified by the fundamentals. If these could be proven, it would be a priori evidence that free international investment of pension funds can cause major macroeconomic problems. But it would not necessarily show that one country, by restricting international investment of its own pension funds, could avoid such difficulties. This is because many of the countries affected—notably France and Italy— have vestigial pension-fund sectors, but were none the less severely affected by the crisis.

(1) The ERM Crisis

The detailed events of the ERM crisis have been extensively recounted elsewhere (see e.g. BIS (1993), Group of Ten (1993), and IMF (1993a)). Suffice to say here that exchange-rate tensions arose in September 1992, in the run-up to the French referendum on the Maastricht Treaty, culminating in the suspension of membership of the UK and Italian currencies, and severe attacks on the French franc and Swedish kronor. In the late autumn the Swedish and Norwegian currencies were forced to abandon their links to the Ecu, and over the period from September to June the Spanish peseta, Irish punt, and Portuguese escudo had to realign against the remaining (p.289) ERM currencies. Last, renewed and sustained tensions arose in June 1993 for the remaining currencies other than the Dutch guilder—that is, the French franc, the Belgian franc, and the Danish krone—as well as the Irish, Spanish, and Portuguese units, leading at the end of July to a widening of the permitted fluctuation margins from 2.25% to 15% in each direction. This diffused the speculative pressure on the system, but also led initially to a depreciation of the former narrow band currencies against the Deutschmark—a depreciation that had been wholly reversed by the end of 1993.

The ‘standard’ macroeconomic explanation for the crises (see e.g. BIS (1993)) highlights a number of causal factors. The success of the ERM in reducing inflation in the long-established member countries, and the drive towards EMU, with its stringent fiscal and inflationary convergence criteria, made membership of the ERM increasingly attractive to others, such as Italy, the UK, which joined in 1990, and the Scandinavian countries, which sought to gain the benefits of membership by pegging to the Ecu. But the drive to EMU also made countries increasingly unwilling to accept realignments, that had been a common feature of the ERM prior to 1987, owing to the associated loss of credibility. Some commentators have argued that this entailed gradual losses of competitiveness for a number of countries, evidenced by balance-of-payments deficits in countries such as the UK, and which, because of the increasing inflexibility of the system, could not easily be resolved.

These tensions were worsened by the expansionary consequences of German reunification for the German economy. In a flexible system, these might have led to a revaluation of the Deutschmark to diffuse inflationary pressures in Germany. Instead, the Germans were forced to use high interest rates as a counter-inflation policy, which, given their status as the anchor of the system, were transmitted to all the other ERM participants. High interest rates were particularly unwelcome to countries that were entering recession and whose private sectors suffered from a high debt burden at floating rates, such as the UK; whose public sectors were partly financed at floating rates, and where the control of the fiscal deficit was in doubt, such as Italy; or which had major banking problems that high rates would aggravate, such as the Scandinavian countries (Davis 1992).

Despite these circumstances, the markets might still have remained calm if there had been no doubt about political will regarding convergence to EMU; and indeed, until the spring of 1992, the markets seemed to believe that these tensions would be resolved by adjustment of domestic prices and wages, despite all the historical evidence of the difficulty of inducing shifts in real exchange rates by this route. But the Danish referendum, which went against Maastricht, and the adverse opinion polls for the following French vote, caused increasing doubts about convergence over the summer of 1992. These culminated in the speculative attacks seen in September 1992 in the (p.290) run-up to the French referendum, as the markets assumed that a ‘no’ vote would lead to an immediate realignment or even the break-up of the system. The countries which fell victim, either immediately (the UK and Italy) or in the wake of the initial wave of pressure (the Scandinavian countries), were those which suffered fundamental difficulties in maintaining a position in the ERM, and in defending it with high interest rates for a prolonged period, as outlined above.

Once the UK and Italy had devalued, as major exporting nations both within and outside the EU, attention shifted increasingly to the difficulties caused for trade competitiveness in those countries remaining in the system, even though their fundamentals were not adverse in the ways outlined above, and they had accordingly resisted initial speculative pressure. The slide of the EU into recession, heightened unemployment, and the slow pace with which German interest rates were reduced, compounded this pressure. It culminated in the pressure on the remaining narrow-band countries in the early summer of 1993, and the widening of the bands.

(2) Financial Market Issues

In seeking explanations of the crisis from a financial market’s point of view, it is important first to note that the success of the ERM had been built at times when a number of the larger participants had exchange controls, thus limiting speculative pressures (though clearly not eliminating them, as repeated crises for the French franc in the early 1980s showed) (see Gros (1992)). The disadvantages of such controls—for example, in terms of higher risk premia on domestic assets (Cody 1989), and corresponding restricted access to international capital markets—made them unattractive (as well as being contrary to the EU Single Market). But there is a cost, as noted in Chapter 9. It is widely acknowledged that, in the absence of such controls, the need in a fixed-rate regime for identical monetary policies, for similar inflation performance (ensuring alignment of real exchange rates), and for similar cyclical performance per se, becomes more urgent. It also puts greater weight on intervention and the level of interest rates as means of counteracting speculative pressures.

Second, the overall volume of transaction in the foreign-exchange market had risen rapidly over the 1980s and early 1990s, tripling between 1986 and 1992 to reach $l,000bn., hence growing at a rate far beyond the growth rate in Central Banks’ foreign-exchange reserves, which in 1992 totalled around $500bn.1 (although note that the ERM included rules for limited sharing of reserves during periods of speculative pressure). Besides the traditional (p.291) operations of banks, which could take (limited) positions against currencies, a number of other components of the foreign-exchange market were highlighted by the crises. First, there is the development of specialized, and unregulated, hedge funds, which would seek by leveraged investments to profit from adjustments in exchange rates, and could exert strong pressure on currencies. Second, there is the increasing sophistication of corporate treasury operations, enabling non-financial firms to fund themselves in the cheapest markets and cover themselves by use of swaps, to hedge future earnings against currency shifts, and to take open positions in their own right. And, third, there is the internationalization of institutions’ portfolios outlined above.

Internationalization means pension funds would inevitably be affected by exchange-rate turbulence; and the resources available to pension funds and life insurers far exceed national foreign-exchange reserves (in August 1992 the French reserves were $28bn., British $40bn., Italian $20bn., and Swedish $20bn.). Funds’ increasing willingness to turn over investments and use derivatives would increase their potential leverage. And reasons have already been presented for funds to be exceptionally sensitive to any losses that could make the fund managers perform badly relative to the rest of the market, thus encouraging adoption of similar strategies. But why should institutions be particularly singled out for making the ERM vulnerable in 1992–3?

One factor is the existence of convergence plays. The drive to EMU, as long as it was considered credible, led to large potential profits from holding assets in the weaker, higher-yielding currencies. So long as the fixed exchange rate was expected to hold, or even with small realignments prior to EMU, large capital gains could be anticipated as yields on bonds denominated in such currencies converged with German ones. Such so-called convergence plays grew to extremely large volumes, as evidenced by portfolio inflows to countries such as Spain, France, and Italy over 1989–91 (Table A2.1). UK funds built up foreign bond exposures quite considerably over this period, from under 1% of their portfolios in 1986 to 4% in 1991 (Source: WM 1993). The IMF (1993a)) suggests that the total value of such investments for all nationalities and types of investor prior to the crisis was $300bn. Note also that governments sought to encourage such international investment, as a means to reducing the cost of financing fiscal deficits and avoiding monetary financing, as well as improving access of domestic firms to equity finance and improving the competitiveness of their financial centres; the success of such approaches for countries such as France is apparent from the scale of foreign holdings of government bonds (as well as equities) as shown in Table A2.2. And reflecting confidence over convergence, US institutions in the high-yield currencies would often content themselves with hedging against the Deutschmark—i.e. in the most liquid derivatives (p.292)

Table A2.1. Portfolio capital inflows for EU countries (billions of local currency)

Country

1989

1990

1991

1992

Italy

4,750

–337

–7,561

Spain

718.3

417.6

433.3

157.4

France

162.8

188.1

80.6

187.9

Germany

–5

–6.4

37.7

–21.5

UK

–20.9

–10.6

–14.8

–2.1

Source: IMF (1993).

Table A2.2. Foreign holding of shares and bonds end-1992 (%)

Country

Government bonds

Equities

France

33

18

Germany

36

20

UK

13

13

Italy

10

n.a.

Japan

n.a.

6

Source: Plihon (1993).

market (proxy hedging). Not that institutional investors were the only convergence players. In addition, non-financial and financial companies in the high-yield currency countries often sought to fund themselves in Deutschmarks or guilders, and US corporations as well as pension funds carried out proxy hedging. The overall pattern of convergence plays could be seen as a form of overreaction of financial markets to the prospect of EMU, encouraged by forms of herding—what Guttentag and Herring (1984) term ‘disaster myopia’.

Given the scale of the exposures involved, the unwinding of such ‘convergence-based’ exposures, or at least increased hedging, in the wake of the Danish referendum could clearly have been an important component of pressure on the system. Extending the discussion in Chapter 9, Section 4, which highlighted the importance of uncertainty in generating exchange-rate volatility, this reaction within the ERM was likely to be particularly strong since confidence—in a process such as EMU—is rarely measured in terms of gradations (as is the case of most forms of risk). Either there is confidence, or there is not (a characteristic of uncertainty). As noted by Raymond (1990), credibility may be binary in the ERM, either complete or zero. The importance of confidence meant that any stimulus such as a data item, perception of policy conflict, or inconsistency in an economy that (p.293) would lead markets to revise their opinions could have consequences seemingly totally out of line with the scale of the event in question, as it would lead the market to question not merely its current decisions but the processes and assumptions underlying such decisions. Similar effects were apparent before financial crises such as the crash of 1987 and the ldc debt crisis (Davis 1992).

A second feature linked to institutions (albeit also used by banks to hedge their over-the-counter derivative positions) is innovative techniques developed for institutional investors seeking to protect the value of their foreign-currency securities (or of options they have written on their assets)—so-called dynamic hedging. These involved the construction of synthetic put options on a currency by a combination of a short position in one currency and a long position in another, and adjusting the ratio continuously in line with exchange rates, interest rates, and expected volatility. Such instruments could exert increasing pressure on currencies when Central Banks raise their discount rates, contrary to the authorities’ expectations, because they require the short position in the currency in question to be made shorter when the spread between the attacked currency’s interest rate and domestic interest rates rises. In addition, according to the IMF, market illiquidity in the cash and derivatives markets, by making such dynamic hedging strategies less viable, would often lead portfolio managers to shift to 100% hedged positions using futures, which would entail further selling of weak currencies.

(3) Evidence

Evidence on the role of institutions in the crises is fragmentary, partly as a consequence of the lack of data on institutions’ portfolios, particularly at periods of less than a quarter, and the almost total lack of data on institutional participation in the derivatives markets. So one is forced to rely on partial data and on anecdotal evidence.

The flow-of-funds accounts of the UK and Dutch pension-fund sectors are of particular interest in the present context, as they show the activities of the largest international investors among European pension funds. Other EU countries’ sectors, as shown in Chapter 9, tend not to hold significant quantities of foreign assets, and/or are themselves extremely small (US funds would be expected to play a major role, but flow data on foreign asset holdings are not available). The data do not show major shifts out of sterling, or repatriation of Dutch foreign assets, as might have been expected (Tables A2.3 and A2.4). Note, however, that desire to retain portfolio balance might lead funds to shift between foreign markets rather than repatriating funds. The data for inflows to German bond markets over 1992 do show quite sizeable inflows; in the second half of the year, total inflows (p.294)

Table A2.3. UK institutions: flow of funds (£bn.)

Asset

1992 Q1

1992 Q2

1992 Q3

1992 Q4

Pension funds

Domestic shares

–0.2

–0.2

0.5

–0.2

Domestic bonds

–0.4

1.2

0.2

–0.7

Foreign shares

–0.3

0.5

–0.2

0.1

Foreign bonds

–0.2

–0.4

–0.1

0.8

All institutions

Domestic shares

0.3

1.5

1.2

0.5

Domestic bonds

0.4

5.6

4.8

4.1

Foreign shares

0.1

0.7

–2.1

–0.8

Foreign bonds

1.0

–1.2

1.1

2.5

Table A2.4. Dutch pension funds: flow of funds (billions of guilders)

Asset

1992 Q1

1992 Q2

1992 Q3

1992 Q4

Private pension funds

Domestic shares

1.0

1.2

0.2

0.7

Domestic bonds

1.1

1.7

1.0

4.0

Foreign shares

2.2

–0.2

–0.9

4.4

Foreign bonds

2.1

1.3

0.5

0.0

Life insurance and pension funds

Domestic shares

3.5

1.3

0.1

1.5

Domestic bonds

2.9

4.3

3.3

6.4

Foreign shares

3.3

0.2

0.4

5.6

Foreign bonds

2.6

1.5

0.4

1.2

were DM120bn. ($75bn.), compared with DM13bn. ($8bn.) in the first half; flows into DM bonds came notably from UK investors, consistent with the hypothesis of portfolio shifts by UK pension funds between foreign markets, albeit also from France, Switzerland, and Japan. But it cannot be proven that pension funds were particularly active.

Data for other sectors show much larger shifts. As recorded by BIS (1993), the banking sector in the UK carried out net exports of domestic currency of $11bn. in the third quarter of 1992, and French banks of $24bn., which were largely taken up by international banks located outside these countries and sold on the foreign-exchange markets, as illustrated by the foreign banks’ net foreign currency positions, which deteriorated by $5bn. to net liabilities for sterling and $21bn. for the franc. There were also large net flows in the banking sector into the Deutschmark and other safe-haven currencies, with a $32bn. increase in the net Deutschmark asset position of banks located outside Germany and the German banking sector (p.295) importing $21bn., for example. But again, as noted by the BIS, the data are ambiguous, as they do not necessarily reflect the banks’ own position taking, but were probably mainly the counterpart of forward transactions resulting from customers’ sales of the currencies in question—in other words, not inconsistent with the pre-eminence of non-banks during the crisis.

Data for the derivatives markets show record levels of turnover for 1992, notably for options and futures traded on exchanges, which saw a 35% increase. Interest-rate futures and currency options were in particular demand. The hedging needs of investors were the main reason for this increase, although difficulties in the over-the-counter markets—largely because of interest-rate and exchange-rate volatility that made pricing difficult, but also partly because of heightened credit risk—compounded the effect on exchange-traded instruments. Meanwhile, the IMF estimates that currency sales from dynamic hedging were 10–20% of sales during the crisis for sterling in September.

As regards anecdotal evidence, the author spoke to a number of UK pension-fund managers in the wake of the crisis. Such discussions did reveal a willingness to use forwards to hedge exposures against the risk, for example, of an ERM realignment. Some suggested a process whereby hedges would be put on when a currency was in the middle of the band, with a view to closing them out when it reached the edge and realizing a profit. But, once there, they realized a realignment was possible and hedges were retained. It was suggested that such leads and lags could put intense pressure on currencies. Most, however, suggested that pressure from funds was not particularly significant in the crises of 1992 and 1993 and cited money funds and corporate treasurers as more active. Some funds reportedly even ‘helped the Central Banks to hold the ERM together’, by repatriating funds (see Table A2.3). Such declarations may risk being self-serving, but they do leave the burden of proof on those wishing to establish a destabilizing influence.

The IMF (1993a) is more positive in identifying a role for institutions. It suggests that, in order to protect the value of their investments, funds sold their foreign assets, hedged their exposures, and sold the vulnerable currencies short, using their assets as collateral in roughly equal proportions, although outright sales were more common in Italy, a market in which forward cover is hard to obtain. But they also suggest that companies which had arranged ‘convergence’ financing in Deutschmarks undertook massive hedging to cover their exposures, while US corporations and investors that had hedged high-yield currencies with the Deutschmark sought to unwind their hedges. Meanwhile, in the IMF’s view, the hedge funds were less important for their direct leverage than in leading institutions and companies to re-examine their assumptions. Banks were constrained by capital (p.296) adequacy requirements in their open positions, and were perhaps most crucial in arranging the financing for institutions and companies’ strategies (the IMF notes that short-selling, hedging, and liquidation of long positions all require bank finance, whether directly or to a counterparty).

Conclusions

The ERM crises certainly illustrated the power that can be exerted by the international capital markets once they are convinced that a fixed exchange rate is unsustainable, as well as the speed with which such judgements may change. But it is harder to maintain that the markets were wrong in a fundamental sense. Particularly in countries such as the UK, authorities have since acknowledged that the exchange-rate/interest-rate constellation in the ERM was unsustainable in the light of the situation in the domestic economy and could ultimately have led to a ‘debt-deflation’. Similar arguments can be made for other countries which were victims of speculation in 1992. More doubt may be expressed about the market’s judgement over currencies such as the French franc in 1993, where inflation and debt burdens were low, the banking system sound, and a balance-of-payments surplus was maintained (Moutot 1993).

Yet more questionable is whether the events suggest that controls on institutions’ portfolios at a domestic level can help protect a currency, in the absence of exchange controls. It is notable that countries with large institutional sectors were unaffected by the crisis (Netherlands) or at least did not undergo extensive capital outflows from domestic pension funds (UK). Countries affected were often those with small pension-fund sectors and/or controls on their international investments already in place. The crucial importance to pension funds of international investment in reducing risk has been emphasized and illustrated. The broader issue of capital controls for all transactions remains a potential response to exchange-rate instability, but most countries, at least in the EU, have concluded that the benefits of open international capital markets, in terms of cost and efficient allocation of funds, for finance of economic development, budget, and trade deficits are too valuable to be cast aside. Moreover, temporary introduction of exchange controls in a crisis would probably raise the risk premium on assets denominated in the currency concerned for a considerable period, and lead markets to anticipate their introduction in advance during the next crisis, thus aggravating the situation.

Furthermore, it is not clear that the special circumstances of the ERM translate readily to floating rates elsewhere. The potential for a realignment offers a strong focus for speculative pressure, since it entails a possibility of a large, discrete shift in the value of assets. Rates of interest needed to compensate for a small possibility of a realignment a short time in the future (p.297) might destabilize the domestic economy.2 Central banks defending current alignments without such increases in rates risk making large transfers of value to speculators over such periods. Such circumstances are less likely to arise for floating exchange rates, when rates may gradually adjust to perceived disequilibria. Concern in such cases is rather that, because of bubbles or fads, rates will diverge over a long period from fundamentals, overshooting sustainable levels. This raises a different set of issues—in particular the relationship between government policy and such shifts. Such volatility may be of particular concern for ldcs. But the arguments above that restrictions on domestic funds may have little benefit and sizeable costs carries over. The case to be answered is rather whether foreign investors should be restricted. Most countries in the Far East and Latin America have concluded that the benefits of open capital markets exceed the risks.

Notes:

(1) Actual sales of Deutschmarks by Central Banks to protect ERM currencies in the second half of 1992 totalled DM188bn. ($118bn.).

(2) Wyplosz (1988), for example, notes that a 10% depreciation expected in a week’s time requires an interest-rate increase of 520% to offset it.