London merchant banks, the central European panic and the Sterling Crisis of 1931

[CrackSmokeRepublican]

London merchant banks, the central European panic and the Sterling Crisis of 1931



Olivier Accominotti

Abstract

The central European panic of the spring 1931 is often presented as a cause of the sterling crisis
of September. But what was the precise transmission channel? This paper proposes to explore how
financial troubles on the continent affected Britain's financial system and currency. The freezing of
central European assets was at the origin of a liquidity shock for London merchant banks because
of their activity as acceptors/guarantors of commercial bills on account of German merchants. New
balance sheet data are used to quantify the shock on various institutions. I then explore 1) how the
various financial institutions reacted to the shock and 2) how the liquidity crisis transformed into a
sterling crisis. The paper provides evidence that international contagion was crucial in transmitting
the global financial crisis of 1931.


1 Introduction

Among the numerous speculative attacks that punctuated the decade of the Great Depression, the
sterling crisis of September 1931 occupies a central place. The pound's collapse was an important mile-
stone in the progress of interwar financial instability because it hit the heart of the international monetary
system. Sterling was at that time a major international currency1 and its devaluation had far-reaching
consequences. The crisis in Britain was followed by speculative attacks in other European countries and
eventually led to the collapse of the gold exchange standard. In the United States, the Federal Reserve
reacted to spreading exchange troubles by tightening its monetary policy, a move that contributed to
banking instability and a deepening of the depression.
The sterling attack also remains of particular interest to economists because it took place in the midst
of a more widespread, international financial crisis. The spring and summer of 1931 were marked by a
wave of banking panics and exchange difficulties in central Europe. Austria was the first victim, soon
followed by Hungary and Germany. Contemporary observers of the year 1931 were depicting a global
crash that was spreading contagiously from country to country.2 The episode therefore appears as a case
study for international crisis transmission.
Explanations for the sterling crisis have alternatively emphasized the pound's overvaluation (Moggrdige, 1972),3 fiscal imbalances (Williamson, 1984, 1992) or the dramatic unemployment rate (Eichengreen and Jeanne, 2000).4 Although they all certainly contributed, these factors cannot really account for
the timing of the speculative attack. Whether overvalued or not, the pound's parity was maintained for
more than six years after Britain returned to gold in 1925. The budget deficit had been deteriorating,5
and the unemployment rate rising since as early as 1929. This is two years before the pound's final

1Eichengreen and Flandreau (2008).
2Such a view appears clearly in Gates W. Mc Garrah's account of the year 1931: The tidal wave of uncertainty and
fear which endangered several national currencies and some banking systems, originated in Austria, swept quickly on to
Hungary and Germany, and, after devastating these areas,
owed onward to Great Britain and the Scandinavian countries,
sweeping down their currencies, and then, backlashing into the United States, carried with its unusual demands upon the
American gold supply and credit system. No such widespread effects, which soon extended to Japan also, could have
occurred except for the already existing essential unity of international finance and monetary relationships, which ignores
political and geographical frontiers.", Bank for International Settlement, 1932, Second Annual Report, p. 10.
3In a famous pamphlet, Keynes (1925) already criticized Britain's stabilization, arguing that the parity retained would
necessitate painful adjustments in terms of deflation and unemployment. However, Matthews (1985) and James (2001) have
challenged the view that the pound was much overvalued.
4Eichengreen and Jeanne (2000) argue that a second-generation model of balance-of-payment crisis does well in explaining
the sterling attack.
5According to The League of Nations, the budget registered a surplus of 7.9 million pounds for 1928/1929, and deficits
of 25 million and 34.5 million for the years 1929/1930 and 1930/1931 (Statistical Year-Book of the League of Nations,
1930/1931 and 1931/1932)

2 Collapse

Given the sequence of events, financial contagion seems to be a better suspect for understanding
the dynamics of the crisis. However, the literature is still divided on this question. On the one hand,
many authors have presented the banking debacle in central Europe as the ultimate cause of the sterling
crisis (Einzig, 1932, Morton, 1943, Williams, 1963, Sayers, 1976, James, 2001). On this account, the
German panic would have directly affected British banks and impaired the pound's position. Yet little
quantitative evidence has been advanced so far for supporting this claim and the transmission mechanism
remains unclear. The extent of London banks' financial losses in Germany are mostly undocumented,
while the link between the banks' troubles and the currency problems has never been formally specified.
On the other hand, there also exists a view that the international financial system was relatively immune
to contagion during the interwar years.6 In an illustration of this argument, Billings and Capie (2008)
have recently cast doubt on the hypothesis that London's City was impacted by global factors during
the Depression.7 Relying on archival material documenting the British joint-stock banks' balance sheets
and investments in Germany, and on little published information on the merchant banks, the authors
propose to assess the impact of the German crisis on the British banking system. Billings and Capie
focus on banking stability over the medium run and conclude that this stability was not altered during
the Depression. The authors therefore oppose Harold James' thesis that the crisis in central Europe had
left London banks under fire. According to them, no `real financial crisis' " hit Britain in 1931. They
conclude that the routes for contagion were limited, and that the system was not at risk".8
In this paper, I adopt an opposite viewpoint and argue that financial contagion was a direct cause
of the sterling crisis of 1931. Relying on new data on London banks' balance sheets collected in various
archival records, I report evidence that financial troubles in central Europe translated into a liquidity
crisis in the British banking system. The approach in this paper differs from previous works on several
aspects. First, I primarily focus on the London merchant banks, which, as opposed to the clearing banks,
were the most affected by the central European events.9 Second, based on information on the institutional

6This tradition goes back to Friedman and Schwartz (1963), who have argued that the banking panics in the United
States were the result of the failures of domestic monetary policy rather than international transmission.
7The paper is not to be quoted without the authors' permission.
8Billings and Capie (2008, p.10).
9Billings and Capie recognize that the merchant banks were more seriously hit but but they minimize the impact of
their troubles on the overall banking stability because these were smaller institutions that could benefit from loans from
the joint-stock banks. James (2001, p. 71) already noted the concentration of the banking troubles among the merchant
banks: The position in this regard of the large joint-stock clearing banks was much safer than that of the private bankers,
Schroeders, Lazards or Kleinworts, who had committed themselves heavily to central Europe."

organization of international banking relations, I specify the precise transmission channel through
which the central European panic endangered the liquidity of London banks. Last, I keep track of various
indicators of the liquidity crisis in banks' balance sheets and in daily interbank interest rate data, identify
the actions of the Bank of England to deal with the problem and relate this information to the collapse
of the pound.

London's illiquidity, I argue, was the by-product of the merchant banks' activity as guarantors of
short-term commercial debt on account of German merchants through a specific financial instrument: the
bankers' acceptance. During the credit boom of the late 1920s, the weakly capitalized merchant banks
had guaranteed bills for foreign merchants on an extensive basis because this activity did not necessitate
them to immobilize resources, and therefore allowed them to earn substantial income. At the end of the
1920s, the amounts of the bills they had insured largely exceeded the value of their capital. This was not
a problem in normal times because defaults on the side of merchants remained limited. However, just
as the burst of the housing bubble affected the liquidity of monoline insurers during the recent crisis, an
economic shock provoking substantial defaults among foreign merchants could at any time endanger the
position of London banks. In the summer of 1931, exchange controls in central Europe and the Standstill
agreements, by imposing a freezing on all assets, resulted in the effective default of all borrowers from
this region. Since merchants could not honor their sterling debts anymore, the liability for these debts
fell upon their guarantors in London. The result was a liquidity shock on largely exposed financial institutions.10

In this paper, I first document the extent of the shock on London banks. I then describe the banks'
reactions to the shock. The crisis was followed by a drastic shrinkage of balance sheets. Fears that
some institutions might fail resulted in a run on these banks. London acceptance houses were forced to
liquidate their assets in order to meet their liabilities. They also severely restrained commercial credit
over subsequent years. In the high-frequency, short-term interest rates surged in the interbank market, as
banks started holding cash in anticipation of their liquidity needs. But the Bank of England immediately
put a halt to this situation. In July, the Bank engaged in open market operations so as to stabilize the
interbank rates. For this purpose, it had to print additional money and the limit of the fiduciary issue

10Schnabel and Shin (2004) have emphasized the role of acceptances in transmitting the financial crisis of 1763 from
Amsterdam to Hamburg and Berlin. Note however that the channel they are considering is the reverse from those at play
in 1931. In this episode indeed, it is the Amsterdam acceptance houses that failed in the first place, making all subsequent
discounters of the acceptances, and ultimately, borrowing Berlin merchants, liable to the bills' holders. A more relevant
analogy is with the recent financial crisis, where troubles have been extensively propagated to institutions having sold credit
default swaps on Collateralized Debt (or Mortgage) Obligations.

-----------

The last  
gesture of this paper tests for this proposition. It displays weekly quotations of the French
franc/pound sterling spot and forward exchange rates in Paris during the year 1931.52 The spot exchange
rate fell below the gold export point towards France in the week following 15 July and stabilized thereafter,
showing signs of relief in August. However, forward sterling quotations reveal that the pound's rally was
pure illusion. In fact, the apparent stabilization was probably the consequence of the Bank of England's
active exchange market interventions in August, which were conducted with the support of the Bank
of France's and Federal Reserve's credits.53 They allowed to maintain the spot rate between the gold
points, therefore avoiding further gold outflows. But in the opinion of investors, the fate of the pound
was already sealed since mid-July. The forward franc/sterling rate continued its fall and never came back
into the band, indicating strong devaluation expectations. This diagnosis was also shared by informed
observers such as John Maynard Keynes. In a letter quoted by Cairncross and Eichengreen (2003) and
dated 5 August, Keynes expressed the view that the abandonment of the gold parity was nearly certain"
at that point: when doubts as to the prospects of a currency, such as now exist about sterling, have
come into existence, Keynes wrote, the game is up..."54

This timing is suggestive that the exchange troubles were related to the Bank's efforts to relieve
pressure on the banking system. During the summer of 1931, the Bank of England was confronted to a
conflict of goals, as is typical in modern theoretical models of twin crises: the Bank could either decide
not to intervene at the risk of provoking a wave of banking failures, or it could accommodate the banks
and provide them with liquidity, therefore increasing the likelihood of a balance-of-payment crisis.55 The
Bank eventually opted for banking stability at the expense of the gold standard. When investors realized
that monetary authorities were reacting to the central European shock by open market operations rather
than an increase in the Bank rate, they provoked a run on the pound.

http://w4.stern.nyu.edu/economics/docs/ ... Crisis.pdf


---------

Also worth a look:

London Banks, the German Standstill Agreements, and 'Economic Appeasement' in the 1930s
Neil Forbes
The Economic History Review, New Series, Vol. 40, No. 4 (Nov., 1987), pp. 571-587

http://www.jstor.org/pss/2596394
http://www.jstor.org/pss/2596394