In the aftermath of the global financial crisis of 2007-2009, central banks in a number of advanced economies rapidly expanded their reserve (base) money and resorted to various unconventional monetary policy measures. Quantitative easing (QE), a massive purchase of securities including Treasury bonds, has played a prominent role in this new toolkit. These new practices have been closely watched by policymakers in emerging-market economies, where there is a temptation to follow the example of ‘core’ central banks such as the US Federal Reserve Board (Fed). This should not be surprising because easing measures always look more attractive for an average politician than austerity. However, it has become clear that the room for policy manoeuvre in the ‘periphery’ is much more limited than in the ‘core’ of the world economy. In this commentary, I will try to explain why.
Consequences of financial globalisation
Since the 1980s, the rapid integration of financial markets has had numerous consequences for macroeconomic policy making at the national level. Most contemporary economies remain open de facto to unrestricted capital movement even if they continue some forms of capital account (and, in rare cases, current account) controls de jure.
Sovereignty in monetary policy has been one of the victims of this process. Owners of money balances, no matter whether they are residents or non-residents in a given jurisdiction, can move funds freely across borders seeking the highest rate of return. Even if formal restrictions on cross-border capital movement are still in place in some countries, they can easily be circumvented. Various financial instruments (under- or over-invoicing of trade transactions, transfer prices, etc) can serve this purpose. As result, the consequences of monetary policy decisions of the largest central banks, especially the Fed, go far beyond their formal jurisdictions and determine at least the general monetary policy directions of other central banks. The latter have limited room to ‘lean against wind’, even under flexible exchange rate arrangements (Rey, 2013)
The fact that many central banks need to follow, at least to some extent, the decisions of ‘core’ central banks could serve as an argument in favour of copying unconventional policy measures. However, another phenomenon makes such copying highly problematic: currency substitution.
Many textbook models of monetary policy assume, explicitly or implicitly, a central bank’s monopoly on money emission. That is, they assume that economic agents use only domestic currency. The reality, however, is far from this assumption. Economic agents retain at least some freedom of choice of the means they use for payments and storing their financial wealth. Foreign currencies, especially the US dollar and the Euro, play a dominant role in such a menu even when national law requires the use of domestic currency for some transactions - for example, paying wages and salaries, taxes, or selling goods and services on the domestic market.
Currency substitution is not a new phenomenon. Historically, it was always present in cases of war, high inflation or hyperinflation, inconvertibility of national currency, etc, even if holding foreign currencies was legally banned and subject to criminal prosecution. Think about the experience of former communist countries. However, financial globalisation and liberalisation of the financial industry have increased economic agents’ access to a variety of foreign-currency denominated assets, and decreased the transaction costs of currency substitution.
Globalisation itself has also increased demand for major transaction currencies, such as the US dollar and Euro. Greater exposure to global trade and cross-border financial transactions makes the use of global currencies convenient even in situations where the domestic currency is stable. It is a question of reducing transaction costs and exchange rate risk, easier accounting and business planning within global value chains and easier access to financing. Markets for global currencies are more liquid and offer greater diversity of financial instruments as compared to markets of smaller currencies. One can call this a ‘network externalities’ effect (Eichengreen, 1996).
Thus, even in economies which offer a prudent record in macroeconomic management but are small and open to international trade and financial transactions, a certain degree of currency substitution caused by transaction needs may be unavoidable.
Struggling with limited credibility
Going beyond the benefits of network externalities, insufficient credibility of a domestic currency is another major driver of currency substitution. Credibility problems can result from experience with or expectations of high inflation or hyperinflation, currency crises, confiscatory monetary reforms, defaults on public debt, banking crises, political instability, or the lack of domestic political and intellectual consensus in favor of price stability, central bank independence and prudent macroeconomic policies. Credibility, once lost, cannot be rebuilt immediately. Such a process usually takes years or decades and is costly, in terms of a necessity to pay higher interest rates and risk premiums through a prolonged period of time.
In practical terms, insufficient currency credibility manifests in limited demand for domestic money balances and high level of spontaneous dollarisation or euroisation: a preference by economic agents to keep their financial assets in US Dollars or Euros rather than in domestic currency. A complete statistical picture of spontaneous dollarisation/euroisation cannot be obtained, because the substantial share of Dollar/Euro cash remains outside statistical record. Nevertheless, Figure 1, which presents shares of foreign-exchange denominated liabilities in total liabilities of individual banking sectors, suggests that it is a serious issue in many economies.
Currency substitution is not limited to the liability side. Economic agents in many emerging-market economies face problems borrowing long-term in local currency at a reasonable price (the so-called original sin problem – see Eichengreen and Hausmann, 1999). As result, they prefer to borrow in US Dollars, Euros, Swiss Francs and other ‘core’ currencies. Financial globalisation facilitates such borrowing at relatively low costs. However, this leads to high foreign-currency exposure, which increases economy’s vulnerability to exchange-rate depreciation and sudden stop in capital flows.
Core and periphery
Based on the above analysis one can distinguish between two categories of monetary jurisdictions: 'core' and 'periphery'. The first refers to global transaction currencies – US Dollar, Euro, Japanese yen, British pound, and Swiss Franc. Perhaps the Chinese Yuan can also join this group at some point but as for now it does not play a role as a global currency and is not fully convertible. All other currencies belong to the periphery because they only play a local, ie national, role - and even there they face competition from global currencies.
Obviously, the group of peripheral currencies is not homogenous. Individual currencies differ between themselves in terms of credibility record and stability and, therefore, resilience to various shocks. For example, this group includes the currencies of some advanced economies (Canada, Australia, New Zealand, Nordic countries, Korea, Singapore) which, although they do not play a global role, are not challenged by currency substitution for credibility reasons. Some of those countries do record high level of dollarisation for transaction reasons or because they host international financial centres. However, as mentioned before, most emerging-market economies do face a credibility challenge.
Consequences of the global financial crisis: core
The difference between core and periphery was best seen in the aftermath of the 2008-2009 global financial crisis. This banking crisis in the US, EU, Japan and a few other advanced economies triggered a deep financial disintermediation. Moreover, the policy response to the crisis, including tighter financial regulation in the post-crisis period, further deepened this trend (Dabrowski, 2016). As result, the money multiplier in ‘core’ monetary areas collapsed, most spectacularly in the case of the US dollar (approximately, by a factor of three between 2007 and 2013 – see Figure 3).
At the same time, demand for broad money balances in the core (as measured by monetisation level – see Figure 4) increased. Part of this phenomenon could be explained by a massive capital flight from the periphery to the core (in search for safe havens) and flight from peripheral currencies.
To compensate for the declining money multiplier and respond to increasing demand for broad money balances, central banks in the core expanded rapidly the supply of reserve (base) money as illustrated in Figure 5. Because they ran out of traditional ammunition quickly (in the case of the Fed, the federal fund rate reached levels close to zero already in December 2008) they decided to resort to unconventional measures, including subsequent rounds of QE. This allowed them to avoid repeating the deflationary experience of the early 1930s.
Consequences of the global financial crisis: periphery
At the periphery, the monetary consequences of the shock generated by the Lehmann Brothers bankruptcy in September 2008 looked very different. As seen in Figure 6, emerging markets experienced either capital outflows (CIS, MENA) or a reduction in capital inflows (other regions except SSA). This created liquidity problems in their banking and corporate sectors (dependence on short-term foreign borrowing) even if they were fundamentally solvent. A large-scale monetary policy response sterilise capital outflows was impossible because of simultaneous pressure on the balance of payments and domestic currencies. Reacting to actual or expected depreciation of national currencies, economic agents start to flee from them, increasing that pressure (see Figure 7 on cumulative currency depreciations between July 2008 and February 2009).
In turn, currency depreciation, if it was substantial, led to a deterioration of balance sheets of governments, corporations and households, which had had net debt exposures in foreign currencies. In many instances, it also increased inflationary pressures coming from exchange-rate pass-through.
A similar story was repeated in 2014-2016, when several emerging-market economies were hit by the gradual tightening of US monetary policy, appreciating dollar, and declining oil and other commodity prices (Dabrowski, 2015). Apart from deteriorating trade and current account balances, some of them had to deal with large-scale capital outflow, exchange-rate pressures, and declining trust in national currencies. This has been visible, in particular, in the countries of the former Soviet Union, Nigeria, Argentina, Venezuela and some other Latin American and African economies.
Lessons for monetary policy making
Even in the core, the marginal effectiveness of unconventional measures seems to decline over time. Furthermore, their continuation involves numerous risks such as building new financial bubbles (Feldstein, 2016; Claeys and Leandro, 2016), compromising central banks' independence (Dabrowski, 2016) and perhaps further financial disintermediation.
Applying those measures in the periphery has never looked like a feasible option for two reasons. First, central banks in emerging-market economies have not faced the problem of zero-level bound that justifies using unconventional measures. Second, the fundamental fragility of many peripheral currencies would make such measures extremely risky. Even in periods of relative calm on global financial markets (like that of the early and mid-2000s) peripheral central banks must look carefully at fluctuations in demand for their currencies, avoid risky policies, build buffers for rainy days and try to fix past credibility problems. In times of market turmoil, even if triggered somewhere else (like in the case of 2008-2009 crisis), they must conduct prudent and transparent policies.
In many instances, they can and should follow the easing cycle of core central banks (as said before, their room to ‘lean against wind’ is limited). Nevertheless, they should refrain from decisions and measures which can be considered by financial markets and economic agents as compromising their independence and price stability mandate (or other monetary policy rules like exchange rate peg)., They should also avoid measures that could be seen to constitute quasi-fiscal activities, the monetising of fiscal deficits, or bailing out individual banks or companies. All decisions taken should be fully transparent and promptly communicated to the general public.
Claeys, G. and Leandro, A. (2016): The European Central Bank’s quantitative easing programme: limits and risks, Bruegel Policy Contribution, No. 2016/02, February 15, https://bruegel.org/2016/02/the-european-central-banks-quantitative-easing-programme-limits-and-risks/
Dabrowski, M. (2015): The impact of the oil-price shock on net oil exporters, Bruegel Blog, November 24, 2015, https://bruegel.org/2015/11/the-impact-of-the-oil-price-shock-on-net-oil-exporters/
Dabrowski, M. (2016): Interaction between monetary policy and bank regulation. In-depth analysis, CASE Network Studies and Analyses, No. 480, February, http://www.case-research.eu/sites/default/files/publications/S%26A%20480.pdf
Eichengreen, B. (1996): Globalizing Capital: A History of the International Monetary System, Princeton University Press.
Eichengreen, B. and Hausmann, R. (1999): Exchange rates and financial fragility, in: Proceedings, Federal Reserve Bank of Kansas City: pp. 329–368, http://www.kc.frb.org/publicat/sympos/1999/S99eich.pdf
Feldstein, M. (2016): The Fed's Unconventional Monetary Policy: Why Danger Lies Ahead, Foreign Affairs, April 18, https://www.foreignaffairs.com/articles/2016-04-18/feds-unconventional-monetary-policy
Rey, H. (2013): Dilemma not Trilemma: The Global Financial Cycle and Monetary Policy Independence, http://www.helenerey.eu/AjaxRequestHandler.ashx?Function=GetSecuredDOC&DOCUrl=App_Data/helenerey_eu/Published-Papers_en-GB/_Documents_2015-16/147802013_67186463733_jacksonholedraftweb.pdf
 This is a revised and reedited version of my presentation delivered at the 5th Research Conference on ‘Economic and Financial Cycle Spillovers: Reconsidering Domestic and Cross-Border Channels and Policy Responses’ organized by the National Bank of the Republic of Macedonia in Skopje, April 7-8, 2016. The original presentation was titled ‘Limits to Unconventional Monetary Policy: Core and Periphery’.
I would like to thank Guntram Wolff and Zsolt Darvas for their comments to the earlier version of this article. Obviously I accept the sole responsibility for its content and presented opinions and conclusions.
 As of May 2016 the ECB and Bank of Japan continue QE. The Fed and Bank of England stopped further assets purchases but large stocks of assets purchased in the past QEs remain in their balance sheets.