LSE.jpg
VoxEU Column Economic history International trade

Liquidity in government bonds – from the British Empire to the Eurozone

Between 1870 and 1914, 68 countries – both sovereign and British colonies – used the London Stock Exchange to issue bonds. This column argues that bond prices and spreads in this period show that the colonies’ semi-sovereignty lowered credit risk at the price of higher illiquidity risk, and further worsened liquidity by attracting investors that rarely traded. Parallels between Eurozone and colonial bonds suggest that the pricing of liquidity and credit in government bond markets is an institutional phenomenon.

Eurozone bond prices are cursed. Whatever happens, observers are unhappy. As pre- spreads against the German Bund narrowed before the European debt crisis, pundits suspected that markets underpriced credit risk due to perceptions of an implicit no-default guarantee. As post-crisis spreads skyrocketed, they suspected that credit risk was back—and perhaps too much so. Not only were credit fundamentals now priced, but they were also amplified and transmitted across borders due to concerns about liquidity (Bai et al. 2012, Pelizzon et al. 2013) or outright panic (De Grauwe and Ji 2012).

This raises two fundamental questions. What respective roles do credit and liquidity play in the pricing of government bonds? What bearing do institutional and political arrangements have on the subject?

Recent research on Eurozone spreads is not conclusive on these questions. Studies focusing on the pre- or post-crisis period disagree on the importance of credit and liquidity for spreads (Schwartz 2014, and references therein). The lack of historical background is a limitation, because the Eurozone has had only seven tranquil years, for a dozen Eurozone countries at best. Also, Eurozone bonds are typically traded in different exchanges, and denominated in different currencies, which complicates comparison.

The lack of a control group makes it difficult to identify the importance of the institutional framework for bond pricing. As we have seen, default is not an option for 'semi-sovereigns'. Greece cannot default in the same way as, say, Argentina. Within the Eurozone, there is political pressure to roll over and restructure the debt of ailing countries, and the ECB has discouraged the idea of default.

In a recent paper, we study the pricing of government debt in a setup that provides both insightful institutional parallels, and an empirically much friendlier environment (Chavaz and Flandreau 2015). During what some economists call the 'first globalisation' era (1870-1914), 68 countries raised funds in sterling on the London Stock Exchange (LSE). With good reason: London was home to the most astute sovereign underwriters, and an ample supply of capital. Countries on the gold standard would not have to worry about exchange rate risk when they borrowed in sterling. The London market had many liquid instruments. Among them, the British government’s quasi-perpetual 'Consol' reigned over the largest and most liquid secondary bond market worldwide.

One unique feature of the London government bond market of the time was that it was split roughly equally between truly sovereign countries – like Argentina or Greece – and colonies of the British Empire, such as Canada or Jamaica. As with pre-crisis Eurozone members, the fiscal sovereignty of colonies at that time is best described as ambiguous. By law, colonies ran their own budget with either a quasi-full autonomy (in the case of self-governing dominions such as Canada and the Australian colonies), or a limited one (for Crown Colonies such as Jamaica). In reality there was a widely-held perception that London would prevent default, maybe by doing 'whatever it takes'. Bond prospectuses never explicitly stated this as policy but, as British parliamentarian (and future prime minister) Benjamin Disraeli emphasised after the 1857 Sepoy mutiny in India, it would be very difficult not to bail out, in some way, a colony that threatened to default.

The consequence at the time it was that credit risk was hardly priced for colonies. This was later confirmed by econometricians (Accominotti et al. 2011). Despite this, colonial spreads over the British Consol averaged about 1%, or 100 basis points. They also varied substantially across time, and between colonies.

When investigating liquidity, a convenient feature of LSE quotations is that they provided precise information on the range in which transactions were expected to take place for any given security – not a true bid-ask spread, but a kind of measure of liquidity. We discovered that this range is a solid proxy for measuring bid-ask spreads at that time.

We could therefore create a novel database of more than 80,000 price quotes for the entire universe of government bonds (sovereign and colonial) quoted in London between 1871 and 1910, at monthly frequency. This is the most comprehensive coverage of the market to date. After studying our indicator of liquidity, we matched this dataset with credit risk proxies provided in Flandreau and Zumer (2004), and performed a panel analysis of the role of credit and liquidity in determining bond spreads.

The analysis confirms that sovereign and semi-sovereign (colonial) markets were very different: liquidity was priced in the colonial market only, and credit risk was priced in the sovereign market only. Liquidity explains, on average, 21% of colonial yield spread variations. It explains zero for sovereigns. The dynamics of market liquidity also differed markedly: aggregate sovereign liquidity dried up during episodes of sovereign turmoil, but aggregate colonial liquidity remained unaffected. By contrast, events such as the famous 1878 Glasgow banking crisis had an adverse effect on colonial liquidity.

Our conclusion – backed by institutional analysis, historical evidence and econometric tests – is that colonies’ semi-sovereignty significantly decreased the severity of information asymmetries between issuers and primary market investors, and among secondary market investors. In turn, this resulted in two paradoxes of semi-sovereignty.

  • Because of different information asymmetries, the two foreign debt markets (sovereigns and colonies) were characterised by different underwriting technologies. Sovereigns typically tried to buy the services of prestigious merchant banks to circumvent the pre-commitment problem inherent to sovereign borrowing (Flandreau and Flores 2009). Deep-pocketed bankers not only lent prestige at issuance, but also stood ready to promote liquidity in the secondary market. Given the absence of severe information asymmetries for investment in the colonies, this was a much less attractive business in the case of colonial debts, and merchant banks soon disappeared from colonial issuances. Instead, colonies used a variety of unusual intermediaries such as second-tier commercial banks or broker-underwriters, both with little means or interest in promoting liquidity. This is the first paradox of semi-sovereignty: it lowered credit risk, but at the price of higher illiquidity risk.
  • For related reasons, the two markets had different clienteles. The absence of asymmetries of information made colonial bonds a natural habitat for those in search of securities free from adverse selection, like insurance companies, commercial banks, and uninformed rentiers who were seeking to cash in coupon payments. This is the second paradox of semi-sovereignty: the lack of adverse selection further worsened bond liquidity by attracting investors that rarely traded.

The parallels between Eurozone and colonial bonds are important. They suggest that the pricing of liquidity and credit in government bond markets may be ‘always and everywhere an institutional phenomenon’. Our historical evidence suggests that investigating the microstructures of sovereign debt markets – their logic, operation, and segmentation – might help in interpreting bond price movements. For policymakers, our research may provide an encouragement to investigate which institutional features can explain the survival of the British imperial architecture, while that of Eurozone was deemed fragile enough to allow destructive liquidity-credit feedback loops.

References

Accominotti, O, M Flandreau, and R Rezzik (2011) "The spread of empire: Clio and the measurement of colonial borrowing costs." The Economic History Review 64(2): 385-407.

Bai, J, C Julliard, and K Yuan (2012) "Eurozone sovereign bond crisis: Liquidity or fundamental contagion." Federal Reserve Bank of New York Working Paper.

Chavaz, M and M Flandreau (2015) “High and dry’: the liquidity and credit of colonial and foreign government debt in the London Stock Exchange (1880–1910)” Bank of England Staff Working Paper 555.

De Grauwe, P and Y Ji (2012) “Mispricing of sovereign risk and multiple equilibria in the Eurozone” VoxEU, 23 January.

Flandreau, M, and Z Zumer (2009) The Making of Global Finance 1880-1913. OECD Publishing.

Flandreau, M, and J H Flores (2009) "Bonds and brands: foundations of sovereign debt markets, 1820–1830." The Journal of Economic History 69(3): 646-684.

Pelizzon, L, M G Subrahmanyam, D Tomio, and J Uno (2013) "The microstructure of the European sovereign bond market: A study of the Euro-zone crisis." Mimeo.

Schwarz, K (2014) "Mind the gap: Disentangling credit and liquidity in risk spreads". Mimeo.

1,680 Reads