Challenges to South African monetary policy in a world of volatile capital flows
7 May 2012
Thank you for the invitation to address you in your beautiful city of Zurich, and it is a privilege to be with you this evening.
We meet at a time of turbulent change and challenge, and it is crucial that we, all of us together, exercise reason and choice as we try to see through the noise and forge a path forward that rebuilds confidence and trust. What we must remember at all times is that behind these billions of dollars/euros/pounds/rands that roll off the tongue so easily these days are ordinary men and women whose very survival hangs in the balance.
Eurozone unemployment now stands at 10,9 per cent, and is expected to continue rising. Spain, which is now formally in recession, has an unemployment rate at around 25 per cent and rising as it enters a severe austerity regime. So in essence we have a three-legged crisis that continues to mutate – a banking crisis, a sovereign debt crisis and a crisis of unemployment, particularly among the youth of the world.
The crisis has also seen changes made to the explicit mandates of central banks, which are now expected to deliver both monetary policy and financial stability, with regulation of banks and other financial institutions part of these responsibilities. No doubt this will become an increasingly heated debate once the global economy has stabilised.
The abillionormally low interest rates and quantitative easing in the advanced economies bear testimony to the persistence of advanced economy fragility. In addition, the world economy came to the edge of a precipice last year when the crisis in the Eurozone intensified and the very survival of the fixed currency system was threatened, an outcome that would have had severe consequences for the global economy.
Fortunately, the steps taken by the member countries, in tandem with other global institutions such as the International Monetary Fund (IMF), brought the world back from the brink. However we can be under no illusion that the crisis is over and that things are back to normal. The interventions helped to buy some time, and it is critical that appropriate measures be taken in the meantime to ensure that this space is used productively, to avoid a future recurrence.
At the same time the problem of global imbalances has persisted, and although there are different views of the relationship between the global imbalances and the crisis, there is no doubt that the imbalances contributed to the low interest rate environment. More recently there have been some positive indications that these imbalances may be correcting, with both China and the United States (US) reducing their respective current account surpluses and deficits.
However, we cannot assume that short term developments necessarily indicate a long term trend, but it is essential that some resolution of the global imbalances and the global financial crisis is found. We have seen too often that the will for coordinated action is there at the height of the crisis when the implications of inaction are very clear.
Unfortunately the urgency often subsides when the worst of the crisis is over: the will to take difficult decisions dissipates, the focus on cooperative solutions diminishes, with countries giving greater emphasis to their own concerns, without much regard to the spill over effects of their actions.
The low global interest rate environment sparked a search for yield that was given further impetus by quantitative easing which increased the quantity of global liquidity.
Capital flows to emerging markets, which had dried up in the wake of the crisis, resumed, with implications for emerging economies financial and foreign exchange markets. It also complicated macroeconomic management in general and monetary policy in particular, a point I will return to later on.
But as this audience is no doubt well aware, it is not only emerging markets that have been affected in this way. Switzerland, an advanced economy, has been a recipient of what are regarded as excessive capital inflows and consequent currency appreciation, resulting in a significant change in the exchange rate intervention strategy. However, the difference is that Switzerland is regarded as a safe haven country, and therefore attracts capital at times of global risk aversion, whereas the opposite is true in the case of South Africa.
The nature of capital flows to emerging markets also changed. As shown in a 2011 IMF report1, portfolio flows began to predominate as more fund managers in advanced economies turned to emerging markets for higher yields, while cross-border bank lending declined, in part because of widespread deleveraging in the face of stricter regulatory requirements. This predominance of portfolio flows is not new to South Africa, but we have seen a change in their composition.
In the years prior to the financial crisis, portfolio flows to South Africa were primarily into the equity market. During the period 2003 to mid-2008, non-residents purchased a cumulative net amount of R216 billion of South African equities, indicative of a positive growth outlook and a well-developed and accessible equity and foreign exchange market. Although South Africa has a well-developed bond market, particularly by emerging market standards, net bond purchases were a more modest R28 billion.
An important difference between these two forms of capital flows was that the former generally were unhedged and involved actual inflows, with implications for the exchange rate. Equity flows have also generally been of a long holding duration or time horizon.
By contrast, bonds flows are often speculative, when a view is taken of future interest rate movements. Very often, and particularly in the pre-crisis period, these flows were funded domestically, and therefore involved no foreign exchange flows, but merely a change of ownership. In the wake of the crisis this pattern changed somewhat, as the global ‘real money’ investors, such as pension funds, turned to emerging market bond markets as a longer term investment destination.
In the second half of 2008, South Africa experienced net sales of both bonds and equities by non-residents, but by early 2009 the inflows resumed. In 2010 and 2011 net purchases amounted to R205 billion, of which R95 billion was in equities, and R110 billion in bonds. Year to date, non-residents have been net sellers of equities to the value of R4 billion, but net purchases of bonds have totalled R35 billion.
The domestic bond market has therefore become a source of longer-duration capital inflows, subject to less volatility, and represents an important structural shift. The cumulative ownership of all government bonds by non-residents increased to around 29 per cent, from 13 per cent in 2008.
The domestic bond market was given a further boost in April 2012 when Citibank announced the potential inclusion of South Africa in the World Global Bond Index by October this year. Estimates vary with respect to the potential impact of this development. South Africa will have a weight of 0,44 in the index, and some funds already hold South African government bonds.
But analysts expect the increase in non-resident holdings of South African bonds to be in the region of around US$6 billion to US$8 billion. The announcement had a significant impact on the bond market and non-residents have since purchased around R19 billion worth of bonds. This is despite the fact that the three main ratings agencies have revised the outlook on South Africa’s sovereign risk rating from stable to negative.
Notwithstanding the increased predominance of longer-duration flows, the rand remains one of the more volatile currencies and is regarded in the foreign exchange market as a high-beta currency i.e. it tends to react more (in either direction) than most other currencies to changes in sentiment.
For example, in the early days of the crisis the rand depreciated precipitously, but soon recovered in response to the significant capital inflows. The rand traded at around R11,80 to the US dollar at its most recent weakest point in October 2008, recovered to levels of around R6,50 in mid-2010, but weakened to around R8,60 to the US dollar in November 2011, and this morning was trading at around R7,85 to the US dollar.
Part of the underlying volatility arises from the fact that while in general capital flows to emerging markets have increased, these flows have not been unidirectional, and have followed changes in global market sentiment. The continued uncertainties in the global economy, and particularly concerns about the Eurozone, have contributed to periodic bouts of risk aversion, often resulting in a flight to so-called safe havens, despite the fact that the underlying fundamentals in the emerging markets have not changed.
The rand foreign exchange market, being relatively large, open and liquid is often used as a hedge for emerging market risk in general and therefore is highly sensitive to these changes in risk aversion, even if the source of the risk is not domestic.
The high degree of volatility is clearly not helpful to the real economy, as it is not always possible to hedge against this, particularly over longer periods. But it is not only the volatility that is important. A number of countries that have been recipients of increased capital inflows have expressed concern about the consequent overvaluation of their exchange rates, adversely affecting the countries’ trade competitiveness both relative to the advanced economies and to third-country competitors such as China that are more able to resist these pressures.
There is a concern that persistently high levels of capital inflows could cause prolonged periods of exchange rate misalignment which could damage export and import-competing sectors irreversibly.
This raises the question of whether the current level of the exchange rate in South Africa is appropriate. Unfortunately, the equilibrium or ‘fair value’ real exchange rate is unobservable, and may be very different from what some may regard as a desirable exchange rate. There are various techniques, all with their own shortfalls, that can be used to determine this level, and the results can differ widely.
One problem is that the equilibrium real exchange rate may not be constant and may change as the underlying fundamental determinants change. So any given nominal exchange rate could represent a different level of misalignment, depending on how the fundamentals have changed.
To give an indication of these difficulties, the IMF in its 2009 Article IV report on South Africa, gave a range of overvaluation of the rand of between 6-8 per cent, 12 per cent, 16 per cent, or 25 per cent if the surge in public investment was included. This tells us that at that time the rand was either moderately overvalued or highly overvalued. So any estimation procedure should be treated with caution.
A recent study undertaken at the South African Reserve Bank (SARB), and published on our website, indicates that the trade-weighted rand was more or less at its equilibrium value in the final quarter of 2011, but is currently overvalued by around 5 per cent. However, these estimates, as indicated, cannot be taken as hard and fast estimates, and do not necessarily represent the official view of the SARB. At this stage we do concur that the rand may be moderately overvalued, but we would not want to put a precise number to this. But given the high degree of nominal exchange rate volatility in the past, there have been times, particularly in 2010, when the rand appeared to be more significantly overvalued.
From a monetary policy perspective, the exchange rate outlook is critical. Ideally we would like to have a stable and predictable exchange rate that does not pose a risk to the inflation outlook. The exchange rate is in fact one of the important determinants of inflation, with an estimated pass-through of around 0,2. However this number cannot be used mechanistically and is subject to a number of important qualifications.
These include the fact that the impact of an exchange rate change on the Consumer Price Index (CPI) is not necessarily linear, so small changes may have a relatively small impact, but larger changes induce proportionately larger responses. A lot also depends on whether the change is seen to be permanent or temporary. If the latter, the impact on inflation may be minimal. Furthermore, some prices are impacted quicker than others.
For example, the petrol price is set monthly according to a given formula, on the basis of changes in the exchange rate and changes in US dollar product price, so the impact on CPI as well as on consumers will be felt very quickly. Finally, we often observe that the response to an exchange rate change is not symmetrical: prices often react quicker to a depreciation than to an appreciation.
The volatility of the exchange rate and the uncertainties relating to the longer term outlook creates enormous challenges for monetary policy. The Bank’s forecasting model assumes a stable real effective exchange rate over the forecast period, which allows the nominal exchange rate to adjust to the inflation differential between South Africa and its main trading partners.
The MPC has to decide whether the prevailing level of the real exchange rate is an appropriate assumption: whether to set it at a different level for the forecast period; or whether to leave it unchanged and communicate a change in the risk to the inflation forecast. This is particularly challenging when after a long period of relative stability, the exchange rate experiences a step change, and the MPC has to assess whether this is temporary or a new longer term level.
Given that the exchange rate is buffeted by a multiplicity of volatile forces, ranging from changing global risk perceptions, commodity price changes, and domestic considerations to name a few, forecasting the exchange rate is subject to a high degree of uncertainty and changes in the real exchange rate assumption are at times the source of change in the inflation forecast of the Bank.
There is an incorrect perception that the Reserve Bank attempts to keep the exchange rate strong in order to help with inflation. There is no doubt that an appreciating currency would have a moderating impact on inflation. But in order to sustain this moderating impact, the exchange rate would have to have a continuously appreciating trend, which is clearly undesirable and not feasible.
On the other hand, there are those who argue for an undervalued exchange rate in order to increase the country’s international competitiveness. Such gains are likely to be short-lived because higher inflation would erode this advantage. More importantly, it is unclear how such an outcome would be achieved. This suggests that a more desirable outcome is relative stability of the exchange rate, which would then reduce the risk to the inflation outcome and at the same time provide a higher degree of certainty to the real sector of the economy.
I should note that while we would like to see a reduction in exchange rate volatility, the advantages of exchange rate flexibility should not be forgotten. When faced with shocks, an economy has to adjust, and generally exchange rate adjustment can help with this process. We need only look at the current crisis in the Eurozone to observe the painful adjustment that countries such as Greece have to endure because of the absence of the exchange rate instrument.
So the question is, can something be done to reduce the level and volatility of the rand exchange rate? Unfortunately we have learned that there are no easy answers.
One proposed solution is to lower interest rates further, on the assumption that this will cause the currency to depreciate. There are a number of problems with this. Firstly, capital flows and the exchange rate are not always ultimately determined by interest rate differentials. There have been times when lower rates, to the extent that they may have signaled higher future growth, resulted in significant equity inflows which dominated the negative incentive to interest-sensitive capital flows. There is therefore a degree of uncertainty as to how the exchange rate is likely to respond to changes in interest rate differentials.
We also have to be concerned about the impact of lower interest rates on inflation. The current level of the repurchase rate is at its lowest in over 30 years, but of greater significance is the real level of the policy rate which, with inflation expectations hovering around the upper level of the inflation target range, is in fact slightly negative.
Inflation is currently at the top end of the target range, but we expect it to return to within the 3 - 6 per cent range by the end of 2012 on a sustainable basis, whereafter the rate is expected to remain close to the upper end of the range for some time. The current monetary policy stance is accommodative because of the persistence of the negative output gap.
However, the expected inflation trajectory suggests that there is limited, if any, room for further monetary accommodation at this stage. It is important to bear in mind, however, that this view is conditional on no significant adverse changes in the other factors that the MPC takes into consideration when determining the stance of monetary policy. These include the domestic growth outlook and the continuing global fragility.
At the same time, long term interest rates, which are also important for capital flows, are not determined directly by monetary policy. These rates are affected indirectly through the impact of monetary policy on inflation expectations. A lower policy rate could in fact cause long term rates to increase. But other factors, in particular fiscal policy, are likely to have a more direct impact on long term interest rates.
Another possible option is direct intervention. Although the Bank’s stated policy has been to allow exchange rate flexibility and not to attempt to achieve a predetermined exchange rate, activity in the foreign exchange market has been fairly sizable in line with our stated objective to increase the level of foreign exchange reserves. Since 2006 the gross gold and foreign exchange reserves increased from about US$20 billion in 2005 to over US$50 billion currently.
Reserves are generally regarded as a form of insurance against capital reversals, and the process of accumulation is also seen to be a means to moderate the volatility of the exchange rate. However, despite a higher level of reserves, and at times substantial foreign exchange purchases, the rand has remained volatile, particularly during times of global turbulence. With average daily turnover in the domestic foreign exchange market of around US$16 billion, trying to lean against currency moves would require extremely large amounts of intervention.
The costs of sterilising these purchases are not trivial and place a burden on taxpayers. These costs arise because of the difference in the cost of absorbing the money market liquidity created by foreign exchange purchases, currently at around 5,5 per cent, and the negligible return we earn on the these reserves. Government has assisted by directly purchasing foreign exchange and placing those on deposit with the central bank. This avoids the sterilisation costs, but is at the expense of other expenditures.
An option favoured by some countries, for example Brazil, is to impose controls on capital inflows. While the jury is still out on the effectiveness of these policies, it is difficult to generalise solutions for countries with different market structures and features. At different points in time the exchange rate is driven by different types of flows, and as the complexity of the foreign exchange market increases, the ability to stem gross flows effectively also declines.
A complicating factor is that a significant proportion of trades take place externally. According to the 2010 BIS triennial survey, approximately 60 per cent of foreign currency transactions involving the rand are conducted offshore.
I should emphasise that capital is important for a savings-constrained country like South Africa, and there are enormous advantages to being part of the global community. Capital flows allow for consumption smoothing over time, it allows for the financing of inter-temporal current account deficits when large investment projects are undertaken, and certain types of flows, particularly direct investment, lead to skills or human capital transfers as well.
The problem that many countries face is one of excessive and excessively volatile portfolio flows, which respond to the vagaries of global risk aversion. Managing these flows are difficult, particularly because interventions could have unintended consequences of providing disincentives to the desirable and much needed capital flows. Similarly, exchange rate flexibility has its advantages, but it is the extreme volatility, brought about by the excessive capital flows, that create problems for macroeconomic management.
Resisting the impact of large-scale capital flows is not easy, given the size of these flows relative to the size of the domestic foreign exchange market. Small emerging markets are therefore innocent bystanders in the world of global imbalances, low interest rates and abundant liquidity. This is not to say that the policy choices of the advanced economies were inappropriate. The world would most likely have been a worse place now had it not been for these policies.
However, the burden of the adjustment is not borne evenly. Allowing the exchange rate to adjust to global pressures provides some cushion in the event of capital flow volatility, but this is only partial. Smaller economies, particularly those with relatively well-developed and open financial markets, bear a disproportionate share of the burden of advanced economy spillovers.
Some see the solution as tightened currency and trade controls. This can only lead to an intensification of trade and currency wars, which is not in the longer term interests of the global economy. Sound domestic policies can help to attract more long-term capital, but that on its own is not enough. Too often the discussion relating to how the world returns to sustainable growth is only about the large economies. Greater consideration must be given to the implications of policy choices on smaller economies.
There are no easy or quick answers. The resolution of these issues is very much work in progress and is core to the deliberations of, among other bodies, the G20, the Financial Stability Board and the Bank for International Settlements. South Africa is an active participant in such policy discussions. The crisis has taught us that there no holy cows when it comes to policy choices, particularly given the expanded mandates of central banks which now includes financial stability, and the highly uncertain political and economic environment.
I have focused this evening on the challenges arising from the impact of the volatile global environment on capital flows and the exchange rate. As for how South Africa is weathering the global storm more generally, without us being complacent, we should remember that since the start of the global financial crisis in 2007/8 South Africa has not had a systemic banking crisis and has not had to assist any bank, nor provide liquidity to the system.
However the economy did suffer a short recession through the trade channel, at the cost of around one million jobs. The recovery has been relatively moderate but looks to be sustained. Economic growth was 3,1 per cent in 2011, and is expected to be at a similar level in 2012, but dependent to a significant degree on global developments. Inflation averaged 5,0 per cent in 2011, and is expected to be slightly above the upper band of 6 per cent for 2012, returning to within the inflation target on a sustained basis by the end of this year.
Government has embarked on a significant infrastructure programme that, as it is implemented, should improve electricity supplies, rail, port and road efficiencies and capacity, improve water supplies and generally support both employment and all round economic growth and efficiencies. But sustainable foreign capital will be required to supplement domestic financing sources. South Africa, an integral part of the African continent, has many challenges but is part of the emerging market story that is shaping the significant changes in world dynamics currently under way.
Thank you for the opportunity to speak to you today, and for your attention.
Issued by: South African Reserve Bank
7 May 2012
[ Top ]