RDP 2004-03: Fear of Sudden Stops: Lessons from Australia and Chile 4. Building Country-trust and Currency-trust: Lessons from Australia

4.1 Overview

This section describes how Australia has, in the period since the federation of the colonies in 1901, gradually developed currency-trust and country-trust. Section 4.2 outlines the development of currency-trust through a clean inflation history, and the exchange rate not being unduly influenced by the government. It also highlights the roles that the emergence of a government bond market and later currency derivatives market have played in transferring currency risk. Section 4.3 then considers the development of Australia's country-trust through a history free of defaults and the development of financial institutions.

In 1901 Australia was small, with just 3.8 million people, and had only been settled by Europeans for a little over 110 years. But thanks to large increases in wool production and the discoveries of gold after 1850, Australia was already relatively rich.^[17] Governments had little stabilising role in the new country with government expenditure just 10 per cent of GDP.

Australia has always been relatively open, given its distance from trading partners, and subject to large external shocks as a result of its dependence on commodity exports. At Federation, exports were around 20 per cent of GDP. The main exports were wool (42 per cent), minerals (26 per cent), wheat (6 per cent) and meat (7 per cent). On numerous occasions the terms of trade, shown in Figure 11, has doubled or halved, transmitting large shocks to the real economy. For example, the terms of trade doubled in the early 1920s leading to sustained high growth in the early part of the decade. But the subsequent reversal in commodity prices in the late 1920s is credited with triggering the Great Depression in Australia. The sensitivity to external shocks is also evident with the Korean war induced commodity price boom – notably in wool (with a five-fold price increase from May 1949 to March 1951) and metals. The surge in demand precipitated a dramatic pick-up in inflation with the resulting restrictive policies combining with a sharp reversal in the terms of trade to precipitate a short-lived recession.

As seen in Figure 12 Australia has tended to have a large demand for foreign financing with the current account often averaging around 4 per cent of GDP. However, over the course of the century, but mainly in the latter half, Australia has become somewhat less susceptible to external real shocks through a diversification in production and exports, and of its trading partners. Commodities still constitute approximately 60 per cent of exports, but the breadth of commodities contrasts with the earlier dependence on wool and gold. Agriculture has declined from being one-quarter of total output to around 3 per cent. Mining declined in importance after the 19th century gold rushes, from 10 per cent to 2 per cent of GDP by 1950, but other mineral discoveries has since seen it grow to around 5 per cent. Manufacturing grew strongly over the first half of the century so that by around 1950 its share in GDP was similar to other developed economies. There has been a gradual shift in economic relations away from Europe, and the UK in particular, toward the US and, more recently, Asia. At Federation, 70 per cent of Australia's trade was with the UK, now it is around 5 per cent. Meanwhile trade with Asia has increased from around 10 per cent of total just after World War II (WWII), to over half now.

4.2 Development of Currency-trust

4.2.1 Inflation policies and outcomes

Inflation performance clearly plays a large role in developing currency-trust. If a country has a history of expropriating wealth from foreign investors through inflation they are much less likely to trust the currency. Controlling inflation has generally been a fairly high priority of Australian government policy. The result, as seen in Figure 13, has been mostly moderate inflation, averaging 4 per cent over the century, and rarely greater than 10 per cent. Indeed, in the 102 years since Federation, Australian inflation has only exceeded 20 per cent in 1 year. In contrast, inflation in Chile has been greater than 20 per cent in approximately half the years in this period.

Various policies have been used to control inflation with centralised wages, a pegged exchange rate, restrictive fiscal and monetary policies, and even tariff policy, all playing a role at some time. This commitment goes some way to explaining currency-trust in Australia.

The monetary regime in Australia over its first 100 years can be roughly divided into quarters. For the first three quarters Australia had a fixed exchange rate, though for the first quarter this was a consequence of the banking system rather than official policy. In the middle part of the century, and particularly in the third quarter, while the exchange rate was fixed, some independence of monetary policy was possible due to the existence of capital controls. In the final quarter of the century the exchange rate became more flexible and consequently monetary policy more independent.

The government's need for funds during World War I (WWI) led the Treasury (Finance Ministry), who at the time were responsible for currency issuance, to quadruple the money supply over the course of the war. A consequence of this short burst of money growth was that the powerful and conservative banking sector became vehemently opposed to excessive monetary expansion. Indeed it was the private banks who imposed parity of the Australian pound with sterling up to 1931, believing it was fundamental to a sound banking system. A balance of payments crisis in 1929–1930 resulted in the peg breaking. The Commonwealth Bank, the government-owned trading bank and precursor to the RBA, took control of the exchange rate and set the official exchange rate at A£1.25.

This peg held for 36 years, when there was a small depreciation. One year earlier, in 1966, the Australian currency was decimalised. The peg was switched to the US dollar in 1971 and then to a trade-weighted basket of currencies in 1974. From 1976 it became a more flexible crawling peg subject to daily adjustment. Monetary policy was then directed by the publication of projections for M3. In 1985 this was replaced by a flexible ‘checklist approach’ to the formulation of monetary policy.^[18]

In 1983 sharp capital inflows precipitated the float of the Australian dollar. While the RBA has intervened in the foreign exchange market since then, it has done so to minimise overshooting rather than target particular values of the exchange rate. In fact, the Australian dollar is considered among the most cleanly floating currencies. Importantly for the development of currency-trust, the clean float means that investors can be confident that the authorities will not deliberately pass currency risk onto foreign investors. Indeed the exogeneity of currency risk is demonstrated by the close correlation the Australian dollar has had with commodity prices over the floating era (Chen and Rogoff 2003). The RBA adopted an inflation target, of 2–3 per cent on average over the course of the cycle, in 1993.

Australia had a centralised wages system for much of the past century. Wage setting was done by an independent arbitration court but wages policy has nonetheless often focused on controlling inflation outcomes and boosting economic performance. Wages were indexed in several episodes, though indexation never contributed to high inflation becoming entrenched as it has in some other countries. Indeed, the concern that indexation could lead to high inflation led to the modification or abolition of indexation on several occasions. For example, following the depreciation in 1985 indexation was discounted by 2 percentage points to avoid a price-wage spiral.

The result of this commitment to controlling inflation over the first half of the century was moderate, though volatile, inflation. Short bursts of inflation following the end of WWI and during WWII did not last more than a few years.

The first episode of significant inflationary pressure followed WWII. Pent-up demand led to inflation accelerating to just under 10 per cent by the end of the decade. Increased commodity export earnings from the Korean war boom, along with untimely policy changes, precipitated a sharp spike in inflation in 1950–1951, peaking at 25 per cent.^[19] But high inflation was not sustained so that this episode if anything demonstrated a commitment to moderate inflation aiding the development of currency-trust. The collapse in the terms of trade and export volumes, combined with tightened fiscal policy in late 1951, contributed to a short recession in 1953 with inflation returning to single-digit rates.

The 1970s oil shocks further increased already rising inflation. Pagan (1987) describes the contractionary monetary and fiscal policies of the 1970s as an ‘inflation first’ focus. It was thought growth could only resume once inflation was controlled. However, inflation remained stuck around 8–10 per cent. The attempt to reduce inflation in the 1980s with wages policy, the Prices and Incomes Accord, was somewhat successful according to Hughes (1997). Though as Carmichael (1990) notes, it was probably less so than hoped. Inflation, while less than 10 per cent, was still seen to be too high. After the large fall in inflation in the early 1990s recession, policy-makers took the opportunity to lock in low inflation with the RBA adopting an inflation target in 1993. Since then the inflation rate has averaged around 2½ per cent, the middle of the inflation target.

4.2.2 The government debt market

Development of currency-trust also requires that foreign investors can take on local-currency risk on terms with which they are comfortable. Typically, this will mean taking on the simplest form of local-currency risk – that is, absent of default risk and other forms of risk. If sovereign risk is low, the government debt market can fill this role and so play a crucial role in inducing foreigners to hold the local currency. In this section we outline the development of the Australian government bond market in order to draw out lessons on establishing a liquid debt market in which foreigners will comfortably participate.

The issuance of government debt began in earnest in the last quarter of the 19th century as the colonial governments undertook greater spending on public infrastructure. Mauro, Sussman and Yafeh (2000) show that from 1875 to 1905 the value of Australian government bonds trading in London increased more than five-fold. Their share of total government bonds in that market increased from 1.4 per cent to 5.6 per cent. By Federation, the Australian states already had outstanding debt equal to their combined GDP. The Commonwealth Government did not issue debt until 1,911, and apart from substantial issuance in the two World Wars, the Commonwealth did not have big increases in indebtedness, as seen in Figure 14. The states remained responsible for the majority of Australian government debt.

At the turn of the century, just 15 per cent of the states' outstanding debt had been issued into the domestic market, with the remainder issued in London (in sterling). The colonies had easy access to the London capital markets and were able to borrow on favourable terms with low interest rates and long maturities.^[20] These favourable terms, and the small supply of domestic savings, led the governments to favour issuing debt in London. New South Wales (NSW) and the Commonwealth made the first major forays into the New York market between 1925–1928, when access to the London market was restricted. Thereafter, abstracting from periods when one or the other market was closed, both sources of international funds were used.

The first sharp increase in Australian government indebtedness was after Federation occurred in WWI. The Commonwealth Government ran large fiscal deficits to cover war costs (total government deficits averaged 12 per cent of GDP during WWI). While the government increased taxes and monetised part of the deficit, it also issued substantial new debt. With the London market all but closed, much of the new debt was issued in the domestic market, and so in domestic currency. Total government debt had increased to almost 125 per cent of GDP by the end of the war. After the war the Commonwealth Government issued more debt abroad, taking advantage of longer maturities and a larger supply of funds. This resulted in an increase in the proportion of Commonwealth debt domiciled abroad.

For much of the Great Depression-era debt markets it was not possible to issue new debt. Rather, most of the increase in the ratio of debt to GDP seen in Figure 14 is the result of the 30 per cent decline in nominal GDP. The Commonwealth Government again issued large amounts of debt in WWII. As in WWI, this was mostly in the domestic market due to the effective closure of international markets. Repeating the post-WWI patterns, the Commonwealth Government turned to the international market for its financing after WWII, leading to a decline in the proportion of Commonwealth debt issued domestically.

What is striking in Figure 14 is that the State governments' move toward domestic issuance was gradual through the century, despite large fiscal shocks. This suggests the domestic bond market was progressively developing with the states taking advantage of the cheaper domestic funds as the market grew large enough.^[21] Indeed, the fact that interest rates on domestic government debt tended to be lower than those on foreign debt suggests that, apart from the benefits of longer maturity available abroad, governments tended to borrow abroad because of the small size of the domestic market. By the late 1970s almost all of the states' outstanding debt was domiciled domestically, in Australian dollars.

In the second half of the century the ratio of government debt to GDP declined fairly steadily. This was largely due to strong economic growth, though smaller government deficits played some part.

It was not until the 1980s that foreigners began to hold Australian dollar denominated debt. Australian government foreign borrowing had always been in foreign currencies. Australian dollar debt (issued in the domestic market) had always been held by domestic residents. In 1980 less than 1 per cent of this domestic debt was held by foreign residents. However, over the late 1980s and early 1990s the proportion of Commonwealth debt held by non-residents rose, even as the government rebalanced its issuance to Australian dollar denominated debt. Many commentators, including Stebbing (1994), have attributed foreigners' decision to hold Australian-dollar debt to the financial deregulation that started in the late 1970s. Deregulation reduced the captive market for government debt and made yields market-determined. The resulting higher yields on Australian bonds were then more attractive to foreign investors. Since the mid 1990s around one-third of Commonwealth debt, all of which is now denominated in Australian dollars, has been held by foreigners.

4.2.3 The development of currency derivative markets

Another key to establishing currency-trust is that domestic agents are able to spread foreign-currency risk to those most able to bear it. This is facilitated by a currency derivative market. The derivative market also provides an additional avenue for foreigners to take on pure local-currency risk. Indeed, foreign investment in the bond market and foreign holdings of derivative exposures require the same willingness to hold domestic-currency risk. In this sense, the capability to develop credible money and bond markets would seem to be a prerequisite for an active derivatives market. As described in Section 2.3.2, there is a very active Australian dollar derivative market. In this section we outline how this market was established.

From 1939 the Commonwealth Bank provided forward cover to domestic residents engaging in international trade. Following the breakdown of Bretton Woods in 1971, growing exchange rate volatility increased the demand for currency hedging. The official market was highly restrictive; it could only be used by domestic residents within seven days of incurring a currency need from international trade. This left residual demand for hedging from capital flows, and for speculation and arbitrage. A foreign-currency futures contract started trading on the Sydney Futures Exchange (SFE) in 1980 after being initially proposed almost a decade earlier. A broker-based non-deliverable hedging market, which brought together two companies with opposite future currency needs, grew in size from the mid 1970s, with turnover ultimately reaching levels similar to the physical market. So the market for currency risk was developing prior to the float of the Australian dollar.

The RBA ceased to provide forward cover around the time of the float of the Australian dollar in December 1983. Domestic trading banks began to provide this service in their own right. The experience gained through the 1970s and the significant knowledge of commodity derivatives helped the market to grow. The existence of a liquid yield curve assisted pricing derivatives of any maturity. Continuing financial deregulation and increased financial sophistication saw the currency derivative markets grow steadily through the 1980s. The growth in the currency derivative markets was also underpinned by the spot market for the Australian dollar being highly liquid from shortly after its float. By 1991 the turnover in foreign exchange swaps involving the Australian dollar exceeded that in the spot market, as seen in Figure 15.^[22] Anecdotal evidence suggests the participation of foreigners followed shortly after their participation in the government debt market.

4.3 Development of Country-trust

4.3.1 Default history

Demonstrating the commitment to repay debt, especially foreign debt, is important in developing country-trust. Australia has a long history of nurturing this trust with no experiences of default by the Federal or State governments. This record was not achieved easily. Notably, in the Great Depression Australia had large debts and was subject to large shocks. Indeed, Australia was one of few countries heavily indebted at the outset of the Depression that did not subsequently default. We outline this experience to demonstrate the commitment required to develop country-trust.

Debt-servicing costs rose in the late 1920s, doubling to almost 30 per cent of export revenue, due to the growing debt and slowing exports. By 1929 New York was no longer lending to Australia and loans from London had a substantially shorter maturity.

Premier Lang of the state of NSW proposed renegotiation of external debt, so that the interest on war debt be halved to the rate charged to Britain by the US, and the abandonment of the gold standard. His plan was rejected by the state premiers, while the proposal of Treasurer (Finance Minister) Theodore was vetoed by the government-owned but semi-independent Commonwealth Bank on the grounds it was inflationary. After substantial deliberation, the ‘Premiers' Plan’ was signed in 1931. It cut government expenditure by 20 per cent, increased taxes and duties, and cut domestic bank interest rates and the interest paid on existing domestic, but crucially not foreign, debt.^[23] The Premiers' Plan was significantly contractionary at the worst possible time but, as Schedvin (1970) notes, such drastic action was felt necessary to avoid default on government debt.

Default was seen to be a real possibility and the largest state, NSW, didn't meet interest payments on overseas debt from April to June 1931. However, the Commonwealth Government and Commonwealth Bank were keen to protect the rating of Australian governments and, because of the centralised structure of government borrowing, made the interest payments on behalf of NSW.^[24] The Commonwealth Government withheld revenue from NSW to compensate for the missed interest payments. Largely because of this episode, Premier Lang was dismissed in 1932 by the state Governor.

4.3.2 Development of institutions

The quality of institutions and stability of the financial system play a key role in establishing country-trust. The structure of many institutions in Australia were adopted from Britain. In particular, Australia has a stable bi-cameral political system and an independent judiciary. This framework and good governance have enabled the development of sound legal and bankruptcy procedures. For example, Australia ranks very high in the Legal Structure and Security of Property Rights component of the Economic Freedom of the World report (Gwartney and Lawson 2003). Indeed, Australia's ranking has steadily increased over the past three decades, to being equal 1st in 2001, while Chile's is much lower (equal 45th). Related to legal structure, transparent and thorough accounting standards also work to instil confidence in foreign investors and so contribute to the development of country trust.

The stability of the financial system is also crucial. In particular, the development of a stable banking system in Australia has assisted in developing country-trust. As we discussed in Section 2.3.2, banks also play an important role in using currency-trust and currency derivatives to smooth external shocks through foreign borrowing.

Schedvin (1970) argues that the banking crisis of the 1890s, in which most banks suspended payments and had to restructure, shaped banks' conservatism thereafter. Despite this crisis, the financial system was highly developed at Federation following the rapid economic growth in the second half of the 19th century. Indeed it was not far behind the US and UK, and well ahead of Latin American countries; bank assets were already 70 per cent of GDP, while those of all financial institutions were 107 per cent of GDP, and M3 was around 60 per cent of GDP. The total financial sector was relatively stable over the 80 years after Federation, with total assets fluctuating between 80 and 100 per cent of GDP. Financial deregulation through the 1980s, discussed by Battellino and Macmillan (1989), led to the rapid expansion of the financial sector with total assets to GDP more than doubling over 20 years.

Footnotes

At the turn of the century, GDP per capita was around 90 per cent of the levels in the US and the UK. Prior to the severe 1890s recession in Australia, GDP per capita was around 40 per cent greater than the US (Maddison 2003). [17]

The checklist consisted of inflation, the nominal exchange rate, interest rates, the balance of payments, monetary aggregates and the general state of the economy. [18]

Contributing policy changes include an increase in the legislated minimum wage, the relaxation of price controls, and the effective depreciation resulting from the devaluation of sterling in September 1949. [19]

The British Colonial Stock Act of 1900 enabled many trust funds to purchase Dominion bonds, giving the Dominions an even greater advantage over many domestic borrowers. Australian governments were able to issue long-dated debt from early on. In 1,913 their fixed-maturity debt had an average maturity of just under 18 years, while 6 per cent of debt had been issued as perpetuities. [20]

An alternative interpretation is offered by Bordo, Meissner and Redish (2003). They argue that the development of the domestic market for government bonds accelerated in those periods in which international markets were closed. [21]

The pick-up in swaps toward the end of the 1990s is exaggerated somewhat by the RBA's use of swaps to conduct open market operations in the face of declining liquidity in government securities markets. [22]

The conversion of domestic government debt to lower interest rates was voluntary but a large propaganda campaign led to 97 per cent of borrowers taking up the new loans with longer maturities and interest rates reduced by 22½ per cent. The 3 per cent of ‘dissenters’ ended up having their debt compulsorily converted. [23]

The Loan Council, formed in 1923, facilitated the coordination of government borrowing. The Council was formalised later in the 1920s with government debt amalgamated and state debt explicitly guaranteed by the Commonwealth. [24]