Speech

Special Remarks by John Lipsky, First Deputy Managing Director, International Monetary Fund, at the Third IMF Roundtable of Sovereign Asset and Reserve Managers
Washington, D.C., January 24, 2011

It is a pleasure to welcome you today to the third IMF Roundtable of Sovereign Asset and Reserve Managers.

When this forum met last in February 2009, the financial crisis was near its most threatening moment. Today, with the crisis receding, it is appropriate to draw lessons, and look forward over a changed landscape. As in our past meetings, I am sure that we will be able to draw some valuable conclusions. We have scheduled a discussion at the end of the Forum addressing our future work agenda in the reserve management area. Your views and suggestions are very important to all of us.

Like many institutions engaged in global financial markets, the Fund currently is participating in the intensive international effort underway to build a more robust post-crisis framework. For our part, we have created new lending facilities to help reinforce the global financial safety net, and are working to enhance both our bilateral and multilateral surveillance to strengthen crisis prevention. The Fund also is very much involved—together with the FSB and BIS—in efforts to make the financial sector more resilient.

Moreover, we are engaged with our membership directly—and via the G20—in an examination of the International Monetary System, including the critical and extremely complex issue of reserve adequacy. In this context, it would be most useful if this Roundtable could review experiences during the crisis in the area of reserve management, and share your views on what worked well and what could have been done better, both from the perspective of individual countries and, more broadly, for the global economy.

To start the discussions, I would like to ask a few questions, some that will attempt to benefit from the wisdom of hindsight, and some that are more forward looking. First, were reserves managed during the crisis in a way that complicated or aided crisis management? Second, what level of reserves would have been adequate? In other words, what lessons can we draw from the crisis about the likely demands on reserves? Third, what are the implications of the current low yield environment for reserve management? Finally, what is the scope for, and potential benefit of, holding reserves denominated in the currencies of rapidly growing emerging market economies?

Reserve Management and Procyclicality

According to IMF staff analysis there is evidence that reserve managers’ investment behavior during the crisis was procyclical. In particular, when the crisis struck, deposits from commercial banks were withdrawn, euro zone covered bonds were sold, and U.S. agency debt was sold or put back. Some of my colleagues have estimated that central bank reserve managers withdrew over US$500 billion from the global banking sector during the crisis. This withdrawal coincided with the collapse of the interbank funding market, so that banks were forced to rely largely on funding from a few reserve currency issuing central banks.

If banks had sufficient high quality collateral they could have used this for purposes of accessing discount facilities to offset withdrawals. However, funding needs were so large that many central banks had no choice but to widen the pool of eligible collateral—and to resort to other extraordinary measures—just to keep their banking systems functioning. Such large movements pose risks of a sharp re-pricing of liquidity of individual banks, and this is precisely what occurred during the period of the crisis-led withdrawals. This is not to state that there was a direct cause and effect, but clearly the experience points to a potential concern.

Procyclicality may also raise concerns from the perspective of the investing country. Perceptions of increasing credit risk reduce the price of tradable instruments—thus creating potential losses that heighten the sense of crisis and instability. As highlighted during the last Roundtable, high quality longer-duration instruments performed better during periods of systemic stress. Such instruments benefit from a flight to quality, usually resulting in declining yields for favored securities. Moreover, such instruments protected the balance sheets of central banks from the sharp declines in short-term interest rates that were associated with the onset of the crisis and that still prevail today.

A key issue therefore is how procyclical behavior can be minimized. Has the experience of the past several years already altered investor behavior, or is more formal action in order? Is an adjustment needed in the methodologies used by reserve managers to assess such risks? Would it be helpful to highlight lessons of the crisis in the IMF’s Guidelines for Foreign Exchange Reserve Management, so that all our member countries take notice? Or is there a deeper conflict of interest between safeguarding financial stability and protecting central banks’ investments that raises fundamental questions about appropriate asset classes for reserves and about incentives for reserve managers? My Fund colleagues and I are looking forward to your observations and advice.

Optimal Reserve Levels After the Crisis

This brings me to today’s second topic, the adequacy of reserves. In particular, were reserve levels adequate during the crisis, and what lessons can be drawn from recent experience? As we know, reserve buffers can play an important role in reducing the risk of crises and lessening their impact when they occur. Adequate reserves can provide the authorities with breathing space to adjust macroeconomic policies and strengthen investor confidence that a country can meet its external obligations. We have been reviewing this topic and recalibrating our framework for assessing reserve adequacy. My colleagues will present our preliminary findings on this issue.

Three measures of reserves adequacy have been used traditionally: the ratio of (i) official reserves to imports, (ii) reserves to broad money, and (iii) reserves to short-term external debt. The experience of the 1990s highlighted the importance of import coverage for those countries that lack private capital market access, as well as the critical role of short-term external debt for countries with significant private market access. In particular, the widely cited Greenspan-Guidotti rule, which stipulated as adequate 100 percent coverage of short-term external debt, emerged as the most commonly used rule of thumb for emerging market countries.

During the last Roundtable we discussed whether there were reasons to adjust or augment these indicators. We focused on three aspects. First, whether there can also be an “internal drain” on reserves reflecting capital flight by residents. For example, residents may not be willing to roll over government debt, or a bank run could occur, driven by residents’ deposit withdrawals. Of special concern are maturity mismatches in the banking system in combination with fixed exchange rate regimes or mismatches denominated in foreign currency, regardless of the exchange rate regime.

Second—in addition to short-term debt—other forms of external drains could materialize in a crisis. Loss of reserves could be associated with derivative positions normally used to hedge domestic portfolio positions, or if non-residents have less tolerance for ratings downgrades than residents. Third, an ample reserve cushion provides some leeway to pursue macroeconomic stimulus during a crisis and thereby reduces the risks of an outright economic contraction.

These issues are particularly relevant, given the notable sovereign debt strains in several European sovereign debt markets. Because the countries involved are part of a currency union, they are unable to undertake adjustment via exchange rate shifts. Should such countries build higher reserve cushions and more robust debt structures in order to make themselves less vulnerable to changes in market sentiment?

We look forward to learning about the concrete lessons you have drawn from the crisis, as well as your reactions to our analysis of reserve adequacy. We also look forward to hearing your views on whether recent actions, including the creation of the Fund’s new crisis prevention facilities and the augmentation of our resources—as well as the development of regional pools of reserve funds—reduce the need for precautionary holding of reserves.

Reserve Management in a Low-Yield Environment

This brings me to the third topic: how best to operate in the current low interest rate environment. As you are well aware, the 1-year US T-bill rate currently stands at 0.3 percent, the lowest in decades. The last time rates were below 1 percent was in 1954. Similarly, low interest rates can be observed in other traditional reserve countries.

Of course, there has been a significant uptick in rates since the “save the date” for this Roundtable was sent two and a half months ago. In particular, yields on the 10-year U.S. Treasury note moved from 2.4 percent to a high of 3.5 percent in early January. Once again, the last time yields were as low as they were last October was in 1954. However, the move from 2.4 to 3.5 percent at that time took two years, rather than two months.

Low interest rates can have a significant impact on central bank income. Most central bank liabilities carry zero interest rates, while a significant share of their cost is fixed. As a result, the income position is highly sensitive to interest rates, with the main risk coming from low rather than high interest rates. This is particularly the case at present, when there is a large gap between advanced and emerging country rates, implying that reserves are invested at low rates and that sterilization costs in the context of intervention in foreign exchange market are high.

Moreover, tight income positions can put pressure on central bank independence. Beyond firmly establishing independence within each country’s legal framework, a solution to this potential problem would be to seek adequate central bank capital and ensure that arrangements are in place to recapitalize central banks more or less automatically.

Should central banks seek to generate additional profits from investing reserves in higher-yielding assets? This path carries risks. Additional profits require additional investment risk, including credit risk. Such an approach can heighten procyclicality. It is for good reasons that central banks are not guided by profit considerations in the execution of fundamental policies. At the same time, central banks also have the responsibility of acting as good stewards for the resources allocated to them. As we discussed at earlier Roundtables, we are concerned that central banks often invest at excessively short maturities simply to avoid reporting accounting losses. This ignores the benefit to central banks of investing in longer term fixed rate instruments.

Ideally, central banks could invest at somewhat longer maturities than at present—this could boost income, but also aid in the development of domestic debt capital markets. Of course, right now may not be the optimal moment to make this adjustment. In order to find the right balance, central banks and debt managers could jointly engage in more active asset and liability management (ALM) of the overall sovereign balance sheet, and coordinate the management of the currency and interest rate exposure of the combined balance sheets. Countries with large reserves can, perhaps, consider shorter maturities in their government debt position and vice versa. But few countries utilize an ALM approach. We would ask you to consider why this is the case, and whether promoting such an approach more universally would be fruitful.

Of course, the alternative is to go it alone and optimize risk solely from the central bank perspective. We are interested in your views on the risk of such an approach, at this juncture, and what time frame should be considered for lengthening duration.

Reserve Managers Holdings of Emerging Market Currencies

Emerging markets are playing an increasingly important role in the world economy. In terms of reserve currencies, however, these countries play a minimal role. In our estimation, approximately 96 percent of reserves are held in the four main reserve currencies constituting the current SDR basket: the U.S. dollar, Japanese yen, euro and British pound. This is understandable from a historic perspective. Until recently, many emerging markets struggled with persistent inflation and external vulnerabilities. However, these countries have made massive strides: growth rates are significantly and hopefully sustainably faster than in industrialized countries, while macroeconomic policies have reduced inflation and fiscal balances.

Looking ahead, expectations of continued strong growth in the emerging market and developing countries imply that in the aggregate they will match the current main reserve economies in economic weight. Our own WEO projections imply that, in terms of GDP, this will occur within the next decade. With regard to trade, the numbers are even more striking: China already is the world’s second largest exporter and emerging markets are projected to match the exports of the major reserve currency economies as soon as the second half of this decade.

In thinking about the evolution of the currency distribution of reserve holdings, it is important to keep several factors in mind. First, one traditional “rule of thumb” is that the currency distribution of an economy’s international reserves should match—in broad terms—the currency basket of an economy’s international imports, and that of its short-term external debt. Second, the prevailing currency distribution of aggregate international reserves is about 65% U.S. dollars (a far cry from an alleged “single currency” system), apparently broadly in line with the “rule of thumb”. Third, as emerging market economies’ financial systems develop (and this may be true for some advanced economies, as well) their exports will feature increasingly as imports, and their currencies, in debt transactions of other countries.

Thus, it seems likely—if not inevitable—that as the role of emerging market currencies becomes more important in trade and debt transactions, they will account for a larger share of international reserves. But under what conditions could EM currencies actually function as reserve currencies? At present, many EM currencies are not convertible, and full convertibility may be some way off.

One interesting issue this analysis suggests is whether intermediate options are feasible for using EM currencies as international reserves. Could EM central banks guarantee convertibility of deposits other central banks hold with them? In fact, this has been observed in some cases. Could other convertibility arrangements be envisaged? Could or should the Fund play a role here? What are participants’ views of synthetic exposures through offshore markets and what about the scope for expanding the SDR basket to include EM currencies and increasing its use as a reserve currency? What is the experience with the Asian Bond funds and regional pooling arrangements?

A related set of questions revolve around the issue of for whom such investments would be wise, and what share of an optimal portfolio they might comprise. In particular, investing in currencies of countries in one’s own region or one that is very similar in risk profile will increase short-term risk as such currencies/markets are more likely to be exposed to the same set of risks. Thus, some bias will likely remain in the currency composition of international reserves, especially for countries with lower cushions to absorb short-term volatility. Here, as well, we are looking forward to a lively and thoughtful discussion, and to benefit from your views on the way forward.

I will conclude by noting that the Roundtable topics look very rich and topical. I look forward to hearing the outcome, which we plan to use as an important input for a report to our Executive Board. My Fund colleagues also would like to know what further issues we should consider highlighting in the IMF’s Guidelines for Foreign Exchange Reserve Management, so as to maximize their operational benefit to all our member countries. Finally, let me take this opportunity to thank you for all your support you have provided us in the context of the cross-country analysis my colleagues have been undertaking in the area of reserves and reserves management. I am certain that the papers we are planning to take to our Executive Board over the next few months will benefit greatly from the inputs you have been providing, including the discussions that will take place today.