Adoption of the Gold Standard and Real Exchange Rates in the Core and Periphery, 1870–1913

• Published date: 01 April 2016
• Author: Andre Varella Mollick
• Date: 01 April 2016
• DOI: http://doi.org/10.1111/infi.12079

Adoption of the Gold Standard

and Real Exchange Rates in the

Core and Periphery, 1870–1913

Andre Varella Mollick

Department of Economics and Finance, University of Texas Rio Grande

Valley.

Abstract

In this paper I estimate the speed of adjustment to shocks to real

effective exchange rates (REERs) during the gold standard years. Adop-

tion of the gold standard by the United States in 1879 resulted in all

four core countrie s (France, Germany, the United Kingd om and the

United States) being fully committed to gold. I use the concept of half-

life (HL) to measure the time it takes for a deviation from purchasing

power parity (PPP) to dissipate by 50%. Relative to the years 1870–

1913, between 1880 and 1913 the half-lives of REERs in core countries

decrease from between 4.4 and 5.2 years to between 3.1 and 3.4 years,

with similar declines across dynamic panels. Combined with evidence

elsewhere that interest rates adjusted quickly, the evidence herein

suggests that adjustment in goods markets was faster following adop-

tion of the gold standard.

I. Introduction

This paper examines the international effects of the adoption of the gold standard as

measured by real effe ctive exchange r ates (REERs). Rat her than dea ling with th e

International Finance 19:1, 2016: pp. 89–107

DOI: 10.1111/infi.12079

© 2016 John Wiley & Sons Ltd

causes of adoption,

1

the focus is on its implicat ions. First, did the speed of REERs’

reversion to their mean cha nge when all the major countries swi tched to the gold

standard? Second, is t here any difference between the deg ree of mean reversion of

currencies whose countries adopted the gold standard and those under other

standards? In other words, di d the currencies of countries on a s ilver standard take

longer to converge to their long-ru n values than those on the gold stand ard? Third,

how consistent are these results with thos e of studies that focus on the abi lity of

economies to conduct an in dependent moneta ry polic y?

The way I investigate these ques tions is through the half -life of REERs. The h alf-

life (HL) measures the ti me it takes for a deviation from purchas ing power parity

(PPP) to dissipate by 50%. Papers on long-horizon data by Froot and Rogoff (1995)

and Rogoff (1996) survey the literature and report a consensus of between three and

five years for the HL of a shock to the real excha nge rate. How do the REERs during

the existence of th e gold standard system compare to th ese values? Viewed as a

fixed-rate regime, the gol d standard should carr y faster adjus tment to shocks. In

this paper, I find supporting eviden ce for this hypothesi s. Using the data set in

Cat~

ao and Solomou (2005) with four core (France, Germany, the United Kingdom

and the United States) and 12 p eriphery countries , I calculate the tendenc y of REERs

to revert to their mean.

This paper examines relat ive PPP under the gold st andard, and thus looks at a

commodity standard in greater detail. Rolnick and Weber (1997) find that, under

fiat standards, the grow th rates of various moneta ry aggregates are more high ly

correlated with i nflation and with each other than they are under commodity

standards. I also take a closer look at the sub-period 1880–1913, which has been

characterized as the ‘era of intensified globalization’by Bordo et al. (2010) in studies

of sudden stops (the large and unexpected fall in a country’snetcapitalinflows).

Studying the ‘trilemma’across regimes, Obstfeld et al. (2005, p. 424) refer to the

classical gold standard as a ‘highly globalized period of mostly fixed rates, unfet -

tered capital mobil ity and, hence, l imited monetar y independe nce’. The adoption of

a new monetary system (such as the convertibility of n otes into gold by 1880 in all

core countries) carries important choices. Velde and Weber (2000) argue that

bimetallism is not de sirable because, a mong steady-states, wel fare under monomet-

allism is higher than under any bimeta llic equilibrium. Domestic political and

economic factors affec ting the adopt ion of new monetar y systems have been studied

by Friedman (1990a, 1990b) and Gramm and Gramm (2004) for the United States

and by Flandreau (1996) for France, among others. However, there are no studies

focusing on the v arious speeds of adjustment to PPP le vels within the gold standard

years.

1

Meissner (2005) exami nes what determined countries to sw itch to the gold standard using a sample

of 19 countries from 1870 to 1913 . He finds that trade channels mat ter, as well as the desire to

decrease borrowing costs on inter national capital market s, and the level of development.

90 Andre Varella Mollick

© 2016 John Wiley & Sons Ltd