VoxEU Column Exchange Rates

How durable is the hard peg of the euro? Lessons from the classical gold standard

The Eurozone crisis has led some to seriously consider the prospect of a breakup of the euro. This column presents evidence from the classical gold standard era (1870-1913) suggesting that even then investors doubted the credibility of emerging market countries sticking to a hard currency peg – with higher premiums on sovereign debt as a result.

At the time of adoption of a single currency for much of Europe, many policymakers believed that exit from the euro would not only be politically difficult, but also undesirable in the sense that the new hard peg would confer greater benefits than costs. But the recent rise in sovereign spreads and the prospect of default among some Eurozone members has given credence to the view that highly-indebted countries such as Greece might leave the euro. This is despite economists having long warned that the economic consequences would be catastrophic (Eichengreen 2007).

Among researchers, the turmoil in European sovereign debt markets has rekindled interest in understanding how market participants perceive the durability of hard pegs and the extent to which the adoption of hard pegs enhances credibility. For example, countries may be able to borrow at lower rates if the adoption of fixed exchange rates confers credibility. Establishing regime credibility may be particularly important for emerging-market countries since their rates for borrowing tend to be higher than those for high-income countries. Lower interest-rate spreads for emerging-market countries can in turn stimulate investment and economic growth (Berg and Borensztein 2000, Schmukler and Serven 2002).

The analysis of yield spreads is central to the debate about exchange-rate regime choice. Proponents of hard pegs argue that credible commitments to fixed exchange rates significantly reduce the premiums that emerging market countries pay in order to borrow in international capital markets. The premium has two components: country risk and currency risk. The country-risk premium represents the risk that a country will default on its debt obligations, while the currency-risk premium represents the compensation that an investor receives for an adverse movement in the exchange rate of a paper-currency bond (Domowitz et al. 1998). The latter component of the risk premium is particularly relevant for the debate over exchange-rate regime choice. Advocates of hard pegs argue that they can reduce the currency risk premium, and if perfectly credible, may even eliminate it altogether. If financial markets do not consider the peg to be perfectly credible, however, then the currency risk premium will remain positive.

Durability tests

Empirically testing the durability of hard pegs should therefore be a straightforward exercise in calculating currency risk, but in practice, it is often complicated by institutional details, such as the ability of countries to default on selected bond issues, or the fact that many emerging market borrowers appear to have limited ability to issue home currency debt. In a recent paper (Mitchener and Weidenmier 2009), we argue that the classical gold standard era (1870-1913) is particularly well suited to testing whether hard pegs are credible. First, the classical gold standard was a global monetary system that many economists consider to be the most credible and widely used hard peg in modern financial history. Second, capital markets were unfettered, and global investors were able to buy and sell sovereign bonds throughout the world without restriction. Third, this earlier period of globalisation precedes the creation of multilateral or extra-national institutions like the IMF and EU, which today, can provide a backstop to sovereign borrowers in crisis periods. The existence of such institutions is yet another reason why it is more difficult to use today’s sovereign bond spreads to make inferences about the durability of hard pegs.

Using a large, new database of hand-collected, weekly sovereign debt prices from the period 1870-1913, we examine the movement in sovereign yields for nine major borrowers (constituting roughly 60% of the world’s outstanding external debt) all of which issued bonds denominated in both paper currency and pounds sterling (or gold) in order to identify the country-risk and currency-risk components of sovereign yield spreads. For seven of the nine countries in our sample, we used long-term paper and gold bonds that traded side-by-side on the same market. The paper and gold bonds were either perpetuities or long-term bonds that had a maturity length greater than 10 years at the time of gold standard adoption. If a country made a completely credible, non-contingent, and permanent commitment to join the gold standard, then the probability of a devaluation of the exchange rate would be zero, and “paper bonds would have been as good as gold” (Obstfeld and Taylor 2003). That is, for a country that credibly committed to the gold standard, the interest-rate differential between a country’s paper currency and sterling bonds would also have been zero. A large spread of paper currency over sterling denominated debt after the introduction of the gold standard, however, would suggest that the commitment to the fixed exchange rate was not seen as a credible monetary regime by financial markets.

Table 1. Measuring the currency risk premium: 10-year windows

Country
Whole Period
Pre-Gold
On Gold
Argentina
1286.63
1465.26
1107.95
Austria
142.00
157.86
126.03
Brazil
72.48
64.01
81.20
India
256.47
282.47
245.46
Italy
89.27
124.19
60.54
Mexico
582.70
709.13
489.38
Russia
820.43
820.02
820.77
US
102.02
95.46
108.57
Chile
581.13
580.11
583.19
Average
437.01
477.61
402.57

 

Our results suggest that joining the gold club did not entirely eliminate the interest-rate differential between a country’s paper currency debt and gold bonds issued on international capital markets. Five years after a country joined the gold standard, the currency risk premium averaged more than 400 basis points. The results hold for a wide variety of borrowers, including colonies like India, European nations such as Russia and Austria, and large Latin American borrowers such as Argentina, Brazil, Chile, and Mexico. The large and persistent currency risk premium implies that markets expected exchange rates to depreciate approximately 20% for our sample of gold standard countries.

We develop a simple model of expected exchange-rate movements, which we use to compare expected devaluations to actual movements in exchange rates. This comparison is possible during our sample period since many countries issued gold, silver, and paper currencies, including Argentina, Austria, India, Mexico, and Russia. Argentina, for example, even passed the Conversion Law of 1899, which stipulated that paper pesos could only be converted into gold pesos at a 120% premium. Our analysis suggests that the premium on paper exchange rates is similar in magnitude to the expected depreciation rate calculated from the paper-gold interest rate differential. While politics perhaps make exit from the euro more difficult than abandonment of the gold standard, the existence of large currency premiums after countries adopted gold suggests that financial markets believed that these hard pegs were not fully credible. As our calculations suggest, investors still considered devaluation and departure from gold a high probability event in emerging markets.

References

Berg, Andrew and Eduardo Borenzstein (2000), “The Pros and Cons of Full Dollarization”, IMF Working Paper 00/50, March.

Domowitz, Ian, Jack Glen, and Ananth Madhavan (1998), “Country and Currency Risk Premia in an Emerging Market”, Journal of Financial and Quantitative Analysis 33:189-216.

Eichengreen, Barry (2007), “The euro: love it or leave it?”, VoxEU.org, 17 November.

Mitchener, Kris James and Marc Weidenmier (2009), “Are Hard Pegs Credible in Emerging Markets? Lessons from the Classical Gold Standard”, NBER Working Paper 15401.

Obstfeld, Maurice and Alan M Taylor (2003), “Sovereign Risk ,Credibility and the Gold Standard, 1870-1913 versus 1925-1931”, Economic Journal, 113:1-35.

Schmuckler, Sergio and Luis Serven (2002), “Pricing Currency Risk under Currency Boards”, Journal of Development Economics, 69(2):367-391.

105 Reads