How not to manage capital flows: the IMF guide for developing countries

Daniela Gabor, guest blogger

In 2010, the development community sighed in collective relief as the IMF reconsidered its long-standing rejection of capital controls. Development agendas, it was hoped, would hence be pursued without the well-known disruptions caused by large and volatile capital inflows. And since foreign crises now come through capital rather than trade flows, developing countries could draw on the IMF’s expertise to avoid global financial volatility and contain sudden-stops.

Once details of the new view emerged, optimism faded. The IMF insisted that capital controls should only be used as a last-resort measure, after countries followed this macroeconomic policy sequence: (1) allow exchange rates to appreciate to equilibrium, (2) lower policy interest rates if no threats of overheating (3) use sterilised interventions in currency markets to increase foreign reserves to adequate precautionary levels, and (4) tighten fiscal policy. But this “code of conduct,” Brazil’s finance minister Guido Mantega argued, would hinder the ability of developing countries to carve their own regulatory path, including the regular use of capital controls.

Setting aside such legitimate concerns for policy sovereignty, it is important to ask what, if any, are the merits of the “macroeconomic policy first” strategy. The sequencing strategy runs into difficulties from its first step: do nothing if the exchange rate is undervalued since capital inflows will bring it to equilibrium. However, measuring misalignment is notoriously unreliable because equilibrium theories cannot capture complex currency-trading strategies. For example, Brazil’s currency market nearly tripled between 2007 and 2010, driven by a higher share of derivatives (from 7% to 38%), increasingly traded by non-resident investors (from 33% to 66%) taking short-term positions. Such trends reflect carry-trade activity. Carry-traders borrow in ‘cheap’ currencies to invest in asset markets of countries with high interest rates, as is the case in most developing countries. In contrast, equilibrium approaches assume that carry-trades cannot generate profits and so they generate conflicting measures of misalignment that are poor guide to policy. Furthermore, the IMF’s insistence on equilibrium silences a well-known development lesson: currency undervaluation is necessary for successful long-term growth (a la China).

Interest rate arbitrage further complicates the second step: lower interest rates to preempt capital inflows. But when capital inflows overheat the economy, lower rates provide additional stimulus. The ensuing inflationary pressures will force the central bank to increase interest rates and may perversely prompt commercial banks to actively intermediate capital inflows. For instance, before Lehman’s collapse, most foreign-owned banks in Eastern Europe responded to tighter monetary policy by borrowing abroad from parent banks and lending domestically in foreign currency (often in “exotic” currencies as the Japanese yen or the Swiss franc) at lower interest rates, a destabilizing practice familiar to many East Asian countries. As a result, interest rate manipulation often entailed trade-offs between price stability and financial stability that cannot be easily resolved without capital controls.

Not yet though, the IMF argues. Central banks should next intervene directly in currency markets, with two qualifications: (a) be guided by precautionary rather than mercantilist intentions and (b) “neutralize at all costs” the domestic liquidity injected through foreign-currency purchases. But the distinction between “precautionary” and “mercantilist” is unclear because we lack accepted theories of optimal reserve levels or uncontroversial rules of thumb.

The “neutralize at all costs” advice reflects the IMF’s pervasive concerns that unless central banks sterilize their interventions in currency market, private banks will find themselves with excess reserves to lend out, triggering a credit boom and inflation. This is a credit market story where banks passively accommodate the sterilization strategy. But the story changes if we treat banks as active intermediaries of capital inflows. Banks can obtain carry-trade returns through ‘sterilization games’: they borrow abroad, exchange that in currency markets and place the domestic currency in central bank’s sterilization vehicles. Sterilizations generate further capital inflows; for many, this is the story of the East Asian crisis and a permanent feature of liquidity management in Eastern Europe.

The sterilization strategy matters in more fundamental ways where non-resident carry-traders dominate currency markets, as they increasingly do in developing countries. Non-residents require domestic liquidity to hold domestic assets and domestic banks typically provide that liquidity. Thus the sterilization strategy has direct consequences for banks’ ability to act as counterparty to non-resident carry trades. A “no sterilizations” policy, successfully implemented by various African central banks to manage large aid inflows, instead gives commercial banks ample liquidity to satisfy non-resident demand. But following IMF’s sterilization advice is equally ineffective in containing engagement with non-residents because commercial banks individually choose how much liquidity to place in sterilization instruments and how much to lend to non-residents.

Yet central banks can design sterilization tactics to deter carry-traders. The Malaysian central bank issued long-term debt to non-banking institutions, explicitly to avoid commercial banks. But there is a political economy angle to consider here. Central banks often encourage sterilization games because capital inflows appreciate the currency and make imports cheaper. The central bank achieves its inflation target, but at the expense of financial stability. The IMF’s sterilization advice legitimizes this narrow pursuit of price stability, whereas outright capital controls would force central banks to consider how its liquidity management affects the relationship between commercial banks and non-residents. For example, Peru’s capital controls included a fee on non-resident purchases of central bank debt and reserve requirements on non-resident deposits.

The last step deserves little attention because fiscal adjustments rarely deter short-term capital inflows. In fact governments often chose fiscal tightening to achieve the opposite – “convince” foreign investors to return after a sudden stop (think Spain, Greece or Portugal).

On close examination, the concerns towards the IMF’s code of conduct are justified. At best the code is ineffective, at worst it perversely increases exposure to hot money inflows. If the IMF insists on downplaying capital controls as a regular policy tool, it must discuss the challenges that commercial banks and non-residents that intermediate capital inflows pose for its policy advice. Otherwise the “code of conduct” should be used as a template of how not to manage capital accounts.

This is a short version of the paper Managing Capital Accounts in Emerging Markets: Lessons from the Global Financial Crisis, published in the Journal of Development Studies 48(6): 714-731, June 2012.

Daniela Gabor is a Senior Lecturer at the University of West England. She is the author of “The Road to Financialization in Central and Eastern Europe: The Early Policies and Politics of Stabilizing Transition” forthcoming in the Review of Political Economy.

Triple Crisis welcomes your comments. Please share your thoughts below.

One Response to “How not to manage capital flows: the IMF guide for developing countries”

  1. Gday there, just became alert to your blog through Google, and found that it’s truly informative. I’m going to watch out for brussels. I will be grateful if you continue this in future.

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>