Central banks and the sterilisation of capital flows in the first era of globalisation
Taming the global financial cycle: Central banks and the sterilisation of capital flows in the first era of globalisation
Countries wish to reap the benefits of financial integration while shielding themselves from the vagaries of international financial markets. Can they have it both ways? A large body of work acknowledges the constraints of a trilemma, pointing out that a fixed-exchange rate regime and capital account openness lead countries to give up monetary autonomy (Obstfeld and Taylor 2004, Klein and Shambaugh 2015). The 1870s–1914 era is often taken as the paradigmatic case for such constraints, with central banks changing their discount rate in function of international pressures only (Eichengreen 1992, Obstfeld and Taylor 2004). Yet, beyond the focus on interest rates, there is a lack of quantitative information on what central banks actually did during this period, and especially how they adjusted their portfolio in response to international shocks.
In the spirit of the recent literature looking at the influence of US interest rates on the global financial cycle and foreign monetary policy (Rey 2016, Bruno and Shin 2015, Jordà et al. 2019), we examine the response of foreign central banks to an interest rate shock from the Bank of England, at the time the leader of global financial markets (Bazot et al. 2019). More precisely, we study the reaction of central banks’ balance sheets, exchange rates, and interest rates to an exogenous increase in the Bank of England rate. Our identification strategy allows us to study simultaneously the degree of monetary autonomy (the response of the domestic rate) and the means employed by central banks to achieve such autonomy (sterilisation, foreign exchange interventions, floating exchange rates, or imperfect convertibility of notes into gold – akin to capital controls). Our study is based on a new dataset of detailed and standardised monthly balance sheets of all central banks in the world from 1891 to 1913.
How central banks offset the effects of international shocks (sterilisation)
We use local projections to simulate the response of national central banks’ discount rates and balance sheet items to a 1% shock to the Bank of England discount rate. Local projections allow us to estimate the responses to shocks without relying on a predefined model. For our benchmark estimations we follow the previous literature, which considers the Bank of England discount rate as exogenous to the policy decisions of other central banks (Obstfeld and Taylor 2004, Jordà et al. 2019). We present robustness checks where we use alternative measures of exogenous changes in the interest rate of the Bank of England, based on a narrative approach. For our sample of monthly variables for 21 countries before 1913, we cannot use monthly data on economic activity. We thus focus on the responses of interest rates, exchange rates, and central bank balance sheets. We can nevertheless control for British monthly activity – for which data are available – as a proxy of the international business cycle. Local projections are particularly well suited for panel data since it is straightforward to include country-fixed effects in the estimations of the impulse response functions and to account for group heterogeneity through state-dependent estimations.
We distinguish four groups of countries: (1) core countries on the gold standard, (2) peripheral countries on the gold standard, (3) countries with a floating exchange-rate, and (4) countries on gold but without a central bank (only the US). As an example, Figure 1 shows how core countries on gold reacted to a shock on the Bank of England discount rate. They increased their interest rate only by a small magnitude – the interest rate pass-through is much lower than unity (approximately 24%). The exchange-rate depreciation is rather small (depreciation by 0.08%) and comes back to parity after two months. What were the balance sheet effects? As core countries offered convertibility (of bank notes into gold), the international portfolio declines quickly and substantially – 1.8% after one month. Yet core countries dilute the impact of this reserve drain by expanding – in fact, by over-expanding – domestic credit (a policy that is called sterilisation or neutralisation, see Mundell 1963, Eichengreen 1992).
Figure 1 The reaction of central banks in gold standard core countries to an English discount rate shock of 100 basis points in the first eight months
Sources: Figure 1 of Bazot et al. (2019). Units: Percentage change compared to month t = -1 (positive exchange-rate response in lower right panel indicates depreciation).
Notes: “international” and “domestic” denote the international and domestic portfolios of central banks, “rate” is the discount rate of central banks and “x” the exchange rate on London.
‘Sterilisation’ is defined as a central bank’s expansion of credit to the domestic economy that offsets the drain of international assets caused by an increase in the international interest rate. The path of the domestic money supply is thus isolated from international shocks, and the exchange rate is stabilised around parity. In practice, the expansion of liquidity by the central bank was a response to the demand of commercial banks that faced a higher money market rate following the rise of the Bank of England rate (as international markets were integrated). Instead of increasing its domestic rate by a similar magnitude (a policy known as the ‘rules of the game’), the domestic central bank expanded credit but maintained the lower domestic interest rate.
Core versus periphery – floating versus fixed exchange rates
Four results stand out from our study of the different responses across groups. First, central banks in the gold standard did not follow the ‘rules of the game’; they did not raise their interest rates by the same order of magnitude as the Bank of England. ‘Sterilising’ the effects of international shocks on the domestic money supply was the norm. Second, while central banks in core countries (Austria-Hungary, Belgium, France, Germany, and the Netherlands) relied exclusively on sterilisation to offset short-term international shocks, the central banks on the periphery of the gold standard also used restrictions on gold convertibility (i.e. capital controls) to minimise reserve losses. In our estimations, restrictions on gold convertibility are observed because the gold stock does not move after a change of the rate of the Bank of England. Such a strategy allowed central banks in the periphery to operate with wider exchange-rate bands, without suspending officially their adherence to the gold standard. Accordingly, the response of the central bank’s discount rate was also lower in these countries compared to core countries on the gold standard. Third, none of the mechanisms described above was observed by central banks in countries off gold. In floating countries, the exchange rate fully absorbed the international shock (in line with the predictions of the trilemma; see Obstfeld and Taylor 2004).
The cost of not having a central bank
Last but not least, we study the response of money market interest rates and gold held by the Treasury in the US, a major country without a central bank at the time (Figure 2). The response of the US money market interest rate to a change in the English rate was three times as large as the average response of rates in countries with a central bank. Consistent with a strong and rapid response of interest rates, the exchange rate between New York and London adjusted more quickly than in countries with a central bank. The US enjoyed much less autonomy and – as suggested by Davis et al. (2009) – lacked a central bank that could have sterilised the effects of shocks on the domestic money supply.
Figure 2 The reaction of the US monetary system to an English discount rate shock of 100 basis points in the first eight months
Sources: Figure 4 of Bazot et al. (2019). Units: Percentage change compared to month t = -1 (positive exchange-rate response in lower right panel indicates depreciation).
Notes: “international” denotes the gold reserves of the U.S. Treasury, “rate” is the call money rate in New York and “x” the exchange rate in New York on London. Alternative measure based on Green (2018).
Taken together, our results show in a novel way why central banks matter. They also highlight the different mechanisms – combining sterilisation and convertibility restrictions – that were used by central banks during the first wave of globalisation to mitigate the potentially adverse effects of short-term international shocks. The immediate response of foreign exchange rates to an increase in the Bank of England interest rate confirms that the integration of global financial markets during this period was very high. Our findings also confirm the textbook trilemma (Obstfeld and Taylor 2004), as a floating exchange rate gave full autonomy to domestic monetary policy. Questions remain about whether the second era of financial globalisation, since the 1990s, is different in this respect (Rey 2016, Obstfeld et al. 2019). The most important conclusion of this study is that the era of the first globalisation was not a period of total submission of countries to the fluctuations of international financial markets. The balance sheets of central banks stood as a buffer between the domestic economy and the global financial cycle.
Bazot, G, E Monnet, and M Morys (2019), “Taming the Global Financial Cycle: Central Banks and the Sterilization of Capital Flows in the First Era of Globalization (1891-1913)”, CEPR Discussion Paper no. 13895.
Bruno, V, and H S Shin (2015), “Capital flows and the risk-taking channel of monetary policy”, Journal of Monetary Economics, 71, 119-132.
Davis, J H, C Hanes, and P W Rhode (2009), “Harvests and business cycles in nineteenth-century America”, The Quarterly Journal of Economics, 124 (4), 1675-1727.
Eichengreen, B J (1992), Golden fetters: the gold standard and the Great Depression, 1919-1939, Oxford University Press, New York.
Green, G (2018), “Monetary policy spillovers in the first age of financial globalisation: a narrative VAR approach 1884–1913”, Bank of England Working Paper no. 718.
Jordà, Ò, M Schularick, and A M Taylor (2019), “The effects of quasi-random monetary experiments”, Journal of Monetary Economics.
Jordà, Ò, M Schularick, A M Taylor, and F Ward (2019), “Global Financial Cycles and Risk Premiums”, IMF Economic Review, 67 (1), 109-150.
Klein, M W, and J C Shambaugh (2015), “Rounding the corners of the policy trilemma: sources of monetary policy autonomy”, American Economic Journal: Macroeconomics, 7 (4), 33-66.
Mundell, R A (1963), “Capital mobility and stabilization policy under fixed and flexible exchange rates”, Canadian Journal of Economics and Political Science, 29 (4), 475-485.
Obstfeld, M, and A M Taylor (2004), Global capital markets: integration, crisis, and growth, Cambridge University Press.
Obstfeld, M, J D Ostry, and M S Qureshi (2019), “A tie that binds: Revisiting the trilemma in emerging market economies”, Review of Economics and Statistics, 101 (2), 279-293.
Rey, H (2016), “International channels of transmission of monetary policy and the Mundellian trilemma”, IMF Economic Review, 64 (1), 6-35.