An Empirical Assessment of the Swedish Bullionist Controversy*

• Author: Nils Herger
• DOI: http://doi.org/10.1111/sjoe.12337
• Published date: 01 July 2020
• Date: 01 July 2020

Scand. J. of Economics 122(3), 911–936, 2020

DOI: 10.1111/sjoe.12337

An Empirical Assessment of the Swedish

Bullionist Controversy*

Nils Herger

Study Center Gerzensee, CH-3115 Gerzensee, Switzerland

nils.herger@szgerzensee.ch

Abstract

In the 18th century, a fierce political debate broke out in Sweden about the causes of an

extraordinary depreciation of the currency. More specifically, the deteriorating value of the

Swedish currency was blamedarbitrarily on monetary causes (e.g., the overissuing of banknotes)

and on non-monetary causes (such as balance-of-payments deficits). This paper provides a

comprehensive empirical assessment of this so-called “Swedish Bullionist Controversy”. The

results of vector autoregressions suggest that increasing amounts of paper money did give rise to

inflation and a depreciation of the exchange rate. Conversely, non-monetary factors were probably

less important for these developments.

Keywords: Bullionism; floating exchange rate; inconvertible banknotes; paper money; Sweden

JEL classification:B12; B17; E42; F31; N13

I. Introduction

“Bullionism” commonly refers to an economic point of view in a famous

controversy about the effects of inconvertible paper money in Great Britain

during the so-called “Bank Restriction Period” (1797–1821). In particular,

after the Bank of England had suspended the gold-convertibility of its

banknotes in 1797 to deal with the major financial disturbances from the

Napoleonic wars, a fierce debate broke out between a bullionist (or anti-

restrictionist) and an anti-bullionist camp as to whether inconvertible paper

money was responsible for the subsequent price instability for goods and

foreign currency (see Viner, 1937, chapters 3 and 4).1According to the

bullionists, the severing of the specie-link of Bank of England notes had

paved the way towards overissuing, which, in turn, resulted in a deterioration

*This paper has benefited from insightful comments by twoanonymous referees. Any remaining

errors are my own.

1In a narrow sense, the controversy dealt with the question as to whether excessive amounts

of banknotes had been issued and, hence, the paper pound had been depreciated relative to the

benchmark of a gold-backed currency. However, insofar as a depreciation manifests itself in a

premium of gold (e.g., the external currency value) over inconvertiblebanknotes, the answer to

this question is closely intertwined with the appropriate level of the exchange rate and prices in

general (Eagly,1963, p. 627; Myhr man, 1976, p. 171).

C

The editors of The Scandinavian Journal of Economics 2018.

912 An empirical assessment of the Swedish Bullionist Controversy

of the value of sterling (see Viner, 1937, pp. 124ff.). Conversely, to

explain price and exchange-rate changes, the anti-bullionists emphasized

the role of non-monetary factors, such as balance-of-payments deficits,

caused by large food imports after bad harvests, or by extraordinary capital

outflows to support allied powers on the European continent. Meanwhile,

the upsurge of banknotes in circulation was interpreted as a symptom

of these developments (Viner, 1937, pp. 136ff.). This controversy matters

even today, as it anticipated important questions in monetary economics

about the impact of fiat money on prices and the determinants of floating

exchange rates. Above all, in rudimentary form, the monetary and the

balance-of-payments approaches to determine the exchange rate are already

encapsulated in the bullionist and anti-bullionist positions, respectively.

Furthermore, to make their case, later monetary debates – above all between

the Keynesians and Monetarists – often revisited the experiences of Britain

around 1800.

Because of the lack of sophisticated econometric tools, the con-

temporaneous empirical assessments of the Bullionist Controversy did not

go beyond subjective interpretations of data figures and tables. However,

even the later empirical work is arguably prone to finding spurious

correlations between the time series of banknotes in circulation, exchange

rates, the price level, and real output (see Officer, 2008, p. 7). It was

not until Officer (2000) when multivariate time-series techniques, in the

form of vector autoregressions (VARs), were introduced to account for

the potential presence of unit roots, cointegrating relationships, and to

deal with the thorny identification issues when uncovering the structural

inter-relationships arising from inconvertible paper money during the Bank

Restriction Period. Nevertheless, Hendrickson (2018a, pp. 227ff.) has

recently argued that even this latest empirical work is flawed, for example,

by confounding Granger causality with economic causality to determine the

VAR structure.2Given these methodological shortcomings, it is perhaps not

surprising that the empirical literature has hitherto been inconclusive in the

sense of reporting evidence in favour of the anti-bullionist position (e.g.,

Nachane and Hatekar, 1995; Officer, 2000), but also the bullionist position

(Hendrickson, 2018a).

This paper focuses on the “Swedish Bullionist Controversy”, which

also dealt with the inter-relationship between inconvertible paper money

and a deteriorating currency value, but predated its British namesake by

2Forthe case of the Keynesians versus Monetarists debate, the point that the temporal correlations

inherent in Granger causality tests are not necessarilyinfor mativeabout the structural connections

between variables has long been made by Tobin (1970). For a more general discussion on the

limited capabilities of Granger-causality tests to uncover structural relationships, see Geweke

(1984).

C

The editors of The Scandinavian Journal of Economics 2018.