Brief Notes On Twin Crises

Why do banking crisis and currency crisis so often coincide?
The effects of a banking crisis can be summarised as a “financial distress resulting in the erosion of most or all of aggregate banking system capital” (Bordo et al, 2001). A currency crisis (also known as a balance of payments crisis) can generally be summarised as “a forced change in parity, abandonment of a pegged exchange rate, or an international rescue” (Bordo et al, 2001) of the currency. Both have detrimental effects to the economy, which are exacerbated when they occur together. The dual occurrence of a banking crisis and a currency crisis is deemed a ‘twin crisis’.

To understand the cause of twin crises it is important to study their history. Under Bretton Woods currency crises were common, but banking crises were very rare. The exogenous control of the exchange rate removed the conflicting policy decision between currency control and domestic policy for the central bank

However, the departure from Bretton Woods reintroduced this trade off for central banks. Kaminsky and Reinhart (1999) provide data to suggest that this resulted in an equally high number of currency crises (per year), in combination with a large increase in the number of banking crises. More importantly, the number of twin crises increased.

The literature concerning these crises does not conform to a single opinion regarding their causes. On the whole, there are two conflicting schools of thought. Firstly, there are those who consider financial crises to occur as a result of poor fundamentals in the economy. This is known as the fundamentals school. Secondly, there are those who consider financial crises the result of investor speculation. This is known as the self-fulfillment school.

Kaminsky and Reinhart (1999), who themselves subscribe to the fundamentals school, provide an overview of the potential causation of twin crises. Firstly, they present Stoker’s (1994) view that currency crises cause banking crises. Stoker’s model is based around a country with a fixed exchange rate. Under such a system any shock to the exchange rate is dealt with using central bank reserves. Therefore, a negative external shock will decrease reserves and restrict the credit available in the economy. This can lead to bankruptcies, etc, which will undermine the central bank’s asset base and cause a banking crisis.

Secondly, they present Velasco’s (1987) model that banking crises cause currency crises. Velasco argues that an overcommitment to the domestic economy by the central bank in a time of crisis, for example a large increase in the monetary base, will lead to an increasingly worthless currency. This is an example of poor fundamentals in the economy.

Finally, there are those who consider banking and currency crises to have similar causes. Kaminsky and Reinhart (1999) use the example of the “exchange-rate-based inflation stablization plan…of Mexico in 1987”. Such programs are often fueled by domestic banks borrowing on foreign financial markets. If financial markets deem this borrowing unsustainable, there is likely to be a speculative attack on the currency. This will trigger a currency crises and a simultaneous banking crisis, as asset markets crash. This example feeds into the self-fulfillment school.

Bibliography

• Reinhart, Carmen M, 2009, “This time is different: eight centuries of financial folly”, Princeton, Princeton University Press
• Reinhart and Kaminsky, 1999, “The Twin Crises: The Causes of Banking and Balance-of-Payments Problems”, American Economic Review, vol. 89, issue 1999, pp. 473-500
• Bordo, M.D., B. Eichengreen, D. Klingebiel, and M.S. Martinez-Peria, 2001, Is the Crisis Problem Growing More Severe?, Economic Policy, vol. 16

‘In the very long run, it is difficult not to be an optimist in the standard of living debate: the years 1750-1850 allowed Britain to escape Malthusian constraints, and to make the transition to modern economic growth’

Hans-Joachim Voth, Living Standards And The Urban Environment, The Cambridge Economic History of Modern Britain, Chapter 10, p.293-294

The Standard of Living 1760-1850

This a table showing the different indicators available to assess the standard of living of the pre-modern worker. Below are some notes I made regarding the historiography of the standards of living debate.

The Marxist view:
Marx wrote in The Communist Manifesto that The Industrial Revolution made workers the slave of the bourgeois state. Marxists tend to hold the view that agricultural workers in the 18th century were far better off than factory workers in the 19th century.

Hobsbawm was a Marxist Historian. His argument has focused around real wage data being inconclusive. He argues that if unemployment was high then this would offset rising real wages in terms of the standard of living. Furthermore, food consumption seems to have declined during The Industrial Revolution. Evidence suggests that the number of cattle and sheep slaughtered failed to keep pace with population growth and so there was a decline in food consumption. Howbsawm argues that this cannot constitute a rise in the standard of living.

The Neoliberal/Classical Liberal view:
The Neoliberal view is one that counters the Marxist view. Hartwell was a proponent of the Neoliberal approach to the standard of living debate. He countered Hobsbawm’s arguments in the following way. Firstly, it would have taken a substantial rise in unemployment to offset the rise in incomes and wages (based on Deane and Cole’s figures) that occurred during the Industrial Revolution. Secondly, Hobsbawm’s argument does not take into account the emergence of new markets for food and simply considers the ones that pre-existed prior to The Industrial Revolution. In this sense is fundamentally ignores the changes that occurred during The Industrial Revolution.

Hartwell then furthered his argument by adding that imports of luxury goods (tea, tobacco, sugar) rose. There was a rise in savings banks during the first quarter of the 19th century. The end of the Napoleonic Wars reduced prices as agriculture returned to normal and imports started to flow again.

The modern view:
By the 1980s the economists had made the debate more empirical (for better or worse) and it became one of optimism (much like Hartwell had originally argued). Lindert and Williamson proposed the view that living standards were stagnant during the initial Industrialisation period (1760-1820). But once this was completed living standards rose rapidly until the 1850s. Their evidence for this was slow wage growth during the periods immediately before, during and after the Napoleonic Wars, but rapid growth from 1819-1850 (when wages increase 110 for all workers). Thus, The Industrial Revolution brought about a combined increase in output and dramatic improvements in living standards. Wrigley and Schofield furthered this using evidence from increases in life expectancy.

Yet this changed once more when the revisionists (Crafts and Harley) triumphed over the traditionalist (Deane and Cole) in the growth figures debate in the 1990s. The debate became more pessimistic. The substantial downward revision in growth figures (and further work by Horrel and Humphries) suggested that family incomes may have failed to keep pace with prices in the late 18th century due to the diminishing impact of women’s earnings in the overall household income, following the decline of the protoindustries. They also pointed out the statistical errors of Lindert and Williamson’s initial figures (small and unrepresentative sample size).

Feinstein (1998) and Clark (2001) made the outlook even more pessimistic when revising and expanding the statistical accuracy of wage data. The results were the conclusion that food prices fell less than expected following the conclusion of the Napoleonic Wars. This lead to a decrease in the real wage, relative to earlier estimates. Furthermore, unemployment remained high, and despite improvements in agriculture, underemployment was still a factor.

The debate has now settled. There is a general view that there was little to no rise in living standards for the initial industrialisation period (1770-1815), but a more rise in the standard of living from 1815-1850, although the extent to which is debatable. A general consensus in the behaviour of wages has allowed the debate to move past a reliance on wage data. There are now questions of the existence of an urban premium associated with proletarianisation, the differencing experience between the sexes (and ages), and the potential drop in life expectancy when living in a city.

‘Step-by-step, these rude, industrious families – with husband and wife working side by side, exploiting their children, the capital of the poor man – drifted away from the norms of traditional society of peasants and artisans to form its proletariat.’

De Vries, Jan. The Industrious Revolution, 2008, Chapter 3, p.100

Agricultural Output Prior To And During The Industrial Revolution

A graph representing the different views on the growth on English agriculture during the 18th and 19th centuries.

The axis should be labelled:

Y axis- Indices of agricultural output 1700-1850, 1700 value = 1 for all indices
X axis- Time

Economics and Egalitarianism

Economists are known for being rather ignorant when it comes to equality. I’m regularly told in my lectures that welfare and happiness can’t be measured and as economists we’re only concerned with an efficient outcome. I have had it drilled into me that economists should leave decisions regarding welfare to politicians (think of that what you will). I don’t agree with this, but I see it creeping into my thoughts more and more. I hope that as my work load increases, I won’t become as dehumanised as some of my peers.

‘making an empirical regularity the foundation, rather than an implication of economic theory, is always dangerous’

William Pool, “Is Inflation Too Low” the Cato Journal, vol. 18, no.3, 1999, p.456

Sample Populations

I’m currently doing some work on The Industrial Revolution. It should be noted that all the following analysis is strictly in a European context (with a particular focus on Britain).

It is widely regarded that economic growth prior to 1760 was extremely low. Even after the start of The Industrial Revolution growth was modest (the precise figures depend on how pessimistic your outlook is, but it is usually estimated in the region of 0.5%-2.5% GDP growth per year). The general trend is shown on the graph below.

Yet, all the growth figures we have prior to the 20th century are best guesses. The samples used to predict GDP growth are either very small, very niche, or a combination of both.

The first reliable (and I use that word generously) nationwide data collection was carried out in the 1800 English census. This got me thinking. Is it a coincidence that our growth figures predict that growth took off around 1800, the exact same time the quality of the sample populations available improved so drastically? I am not of the opinion that if better economic data was available for the past 1000 years our outlook on growth would be that different. I am just contemplating the idea that there maybe be a correlation between growth estimates increasing so markedly in 1800 and the quality of the data used to estimate it increasing so substantially. If better data had been available prior to 1800 would we have seen a more gradual growth path?

But of course, this is ultimately an issue for Econometricians and Economic Historians (people who are much more qualified than me!).

‘domestic agriculture will largely determine the growth potential of the non-agricultural sector’

Clunies-Ross et al, Development Economics, 2009, p.455

‘Improvements in agriculture are of two kinds: those which increase the productive powers of the land, and those which enable us to obtain its produce with less labour.’

David Ricardo, On the Principles of Political Economy and Taxation, 1817, p. 80