Good evening, and thank you for the opportunity to be with you. It is always a pleasure to return to GIBS as I was personally associated with this school after my term of office as a Deputy Governor expired in 2004, and have a great deal of respect for the forums you create for open dialogue.
One would have thought that a return to the South African Reserve Bank would be relatively straightforward because of my previous experience in central banking. However, I found a very different world. Not only had the environment in which central banks operate changed dramatically, the expectations of central banks as reflected in their extended mandates had also changed.
In the early to mid-2000s leading up to the global financial crisis there was an emerging macroeconomic theory and policy consensus. The period was known as the Great Moderation, following an apparent trend towards macroeconomic convergence, particularly in the advanced economies. Business cycles appeared to be more synchronised with lower amplitudes, and low inflation prevailed in most countries.
Some went as far as to suggest that both business cycles and inflation were dead. Inflation targeting had become a broadly accepted monetary policy framework, making price stability the overriding objective. The framework was widely credited for contributing to this benign environment. Mervyn King, Governor of the Bank of England, once remarked that a sign of a successful monetary policy was when it became boring, and indeed it seemed that the world was on its way in that direction.
Well, all I can say is “bring back boring”. At the same time, while many central banks had implicit financial stability mandates, the evolving trend at that time was for the regulation and supervision of the banking and financial sectors to move out of central banks. Notable examples were in the UK and Australia, and serious consideration was also given to moving bank supervision out of the South African Reserve Bank in the early 2000s. The focus of central banks appeared to be narrowing.
Today things have changed quite dramatically. Monetary policy and central banking in general has become anything but boring, and rather than taking a back seat in the economy, central banks are now at the forefront of attempts to stabilise and repair the damaged global economy. The mandates of central banks have been extended in numerous countries, including in our own, bringing with it new and difficult challenges. In my remarks this evening I will focus on the growth and financial stability aspects of broadened mandates in the South African context.
The modern mandate of central banks had focused primarily on price stability, reinforced by the increased adoption of inflation targeting. The simplistic view of inflation targeting is that all a central bank has to do is to conduct policy in such a way as to keep inflation in line with the target.
Other possible objectives of monetary policy are seen to be secondary or not the concern of monetary policy. This relates in particular to economic growth, and by extension to employment. Although the Bank has often been criticised for not being sufficiently concerned about growth, we are also criticised by those who believe that we have gone beyond our mandate by focusing on growth.
In the wake of the global financial crisis, it is clear that growth and employment became foremost concerns of most central banks, particularly in those economies where the recovery was slow and protracted, and where inflation was not perceived to be a threat. The persistence of the extraordinarily low interest rate environment in the advanced economies attests to that. So is this focus on growth a new mandate for central banks and an undermining of the primary inflation objective?
In this respect, inflation targeting has often been misunderstood, or oversimplified. It is important to distinguish between strict and flexible inflation targeting. A strict inflation targeter is one that only focuses on inflation to the exclusion of all other possible objectives. As such, it is a theoretical construct as in reality no central bank acts in this way. A flexible inflation targeter, by contrast, puts some weight on the deviation of output from potential (the output gap), and this weight can change depending on circumstances.
In effect this implies that the objective of monetary policy is to achieve the inflation target with output in line with potential. The focus however is on deviations from potential output, as it is generally believed that monetary policy has limited, if any, long run impact on the growth potential of the economy. This is the domain of other policies. The flexibility is manifested when concern for the state of the economy can result in deciding on a longer time horizon for achieving the inflation target, in order to minimise the negative impacts on output growth.
It also follows that if inflation is in line with its target and if inflation expectations are well anchored, more weight can be given to stabilising or increasing growth that is below potential, without sacrificing much price stability. This does not imply a change in mandate, and is consistent with a flexible inflation targeting framework, such as we have in South Africa.
What does appear to have changed is that the burden of stabilisation and growth stimulus is falling increasingly on central banks, particularly in those countries that no longer have any fiscal space. In many countries, the response to the crisis was initially through both monetary and fiscal policy easing. However, the measures taken did not resolve the crisis and an unanticipated consequence was that, in a many instances, this led to excessive and unsustainable deficits and debt ratios. For example, in 2007, government debt to Gross Domestic Products (GDP) ratios in the US and the UK were 62 per cent and 44 per cent.
By 2012 they have almost doubled and expected to reach 103 per cent and 87 per cent respectively. In the euro area, the debt ratios of Greece and Ireland increased from 105 per cent and 25 per cent, to 157 per cent and 122 per cent over the same period. Of these four countries, only the US is currently experiencing positive growth. The costs of servicing this debt has also increased significantly, particularly in the peripheral Eurozone countries, as doubts about their sustainability persist.
This rapid expansion has resulted in fiscal consolidation or retrenchment becoming the order of the day in many, mainly advanced, countries, even where growth has not fully recovered. In short the global crisis has continued to mutate, and we now face the challenge of systemic banking weakness in a number of countries, for instance Spain, sovereign debt and possible default in parts of the Eurozone, and high and rising unemployment. The unemployment rate in the Eurozone has risen to 10,9 per cent, with Spanish unemployment at 24,1 per cent.
Fiscal austerity has reinforced the slow growth, and consequently increased the burden on monetary policy. Although the mandate may be the same, the shared responsibility is no longer there, and more is expected from central banks. At times these expectations may be beyond what monetary policy can reasonably be expected to deliver.
In this context we can consider whether the conduct of monetary policy in South Africa been much different in the post crisis period. Some have argued that the current accommodative stance of monetary policy is part of a new mandate, and a deviation from inflation targeting. We would argue that the Bank’s actions have been consistent with the flexible inflation targeting framework. Since the crisis, the economy has been growing at below what we estimate potential output to be (around 3,5 per cent), and the absolute output gap, that is the difference between the level of output and the potential level of output, coincidentally also at around 3,5 per cent, has been persistently negative.
It will require a period of above-potential growth to close this gap, unless there has been a destruction of capacity which would reduce the size of the gap. A negative output gap also implies less pressure on inflation. Although inflation had moved outside the target, the considered view of the Monetary Policy Committee (MPC) was that inflation would return to within the target, and, given the subdued state of the economy, it was felt inappropriate to speed up this process through a tighter monetary policy stance. This view was also reinforced by the absence of excess demand-side pressures and relatively well-anchored inflation expectations.
In the 2006-08 period, by contrast, when inflation exceeded the target, growth was in excess of the Bank’s estimates of potential, with measured growth at between 5 and 6 per cent per annum, with very high rates of growth in credit extension and household consumption expenditure. Under such circumstances, a tighter monetary policy stance was justifiable or appropriate.
It is also important to note that recent central bank interventions have not always been about stimulating growth directly through low interest rates. In the context of the crisis, and again more recently when fiscal policy has been constrained, the focus has at times been on helping dysfunctional parts of the financial markets to work better. This is recognition of the central place that financial markets have in the efficient workings of the real economy. These policies have aimed at preventing a negative feedback loop from financial sector stress to the real economy, and have resulted in a number of significant forms of unconventional monetary policy actions, particularly in countries where interest rates had reached their zero bound.
There are a number of illustrations of this. Quantitative easing (or credit easing as it was referred to in the US) was initially aimed primarily at providing liquidity to specific segments of the financial markets that were no longer functioning efficiently. During the crisis the loss of confidence in banks and some financial instruments, and the lack of trust between banks, disrupted some segments of the markets and the interbank markets.
Central banks responded by becoming the counterparties and bought unconventional assets on a significant scale. Between 2007 and 2011, central bank balance sheets in the advanced economies increased from around 10 per cent of GDP to in excess of 20 per cent of GDP, a total of almost US$8 trillion. More recently the US Fed initiated its Maturity Extension Programme (more popularly referred to as Operation Twist), whereby the Fed sold short term Treasury securities and bought longer term securities, in an effort to bring down long term interest rates.
The aim was to give a boost to the ailing domestic housing market, as mortgages are often priced off long-term rates in the US, as well as to the corporate bond market. The Long Term Refinancing Operations (LTRO) of the ECB, through which Euro1,3 trillion of liquidity was injected into the European banking system that was in danger of seizing up, is another example of such measures. The objective was to keep the banking and interbank system operating, and to cushion the real economy from banking sector stress.
Some of these activities were intended to provide additional stimulus to economic growth, but some also had a strong financial stability element. This leads us to the area in which central bank mandates have changed most significantly: that is, responsibility for maintaining financial stability.
Prior to the crisis, financial stability was generally an implied mandate for central banks. The fall-out from the crisis has led to an increasing number of central banks being given explicit financial stability macroprudential mandates. This is distinct from microprudential supervision and regulation of the banking system where the focus of the regulator is limited to individual banking institutions and the banking system.
A broader macroprudential focus would look at the build-up of financial imbalances and the risks posed by the positions taken by leveraged financial institutions to the broader financial sector and to the economy in general. For example, in a low interest rate environment and high rates of growth of credit extension by banks, these funds can be used to finance purchases of assets and result in the build-up of bubbles in the housing or equity markets.
The issue of how central banks should react to asset prices was the subject of much debate prior to the crisis. The dominant view at the time was that financial imbalances and crises should not occur in a low inflation environment, and where they did, central banks were not in a position to recognise or predict them. Nor did they have appropriate tools to prick incipient bubbles. Interest rates would had to have been raised to unacceptably high levels in a low inflation environment in order to deal with the bubbles that had emerged.
The consensus view was that asset prices should be taken into consideration by monetary policy only to the extent they impact on inflation, and that the role of central banks should be confined to cleaning up after the bubble had burst. Unfortunately, as we are all now well aware, five years after some of these bubbles burst central banks are still cleaning up.
But there were some influential voices, notably coming out of the Bank for International Settlements, that argued that the low inflation/low interest rate environment of the 2000s was the cause of the asset price bubbles in the first place. In other words, ironically, the successful attainment of low inflation was leading to longer term financial stability problems. This approach argued in favour of monetary policy leaning against these excesses. The view was that too narrow a focus, or too short a time horizon for policy, blinded policy makers to longer term systemic financial stability risks which could take a long time to evolve.
Since the crisis, a widely accepted perspective has emerged which sees monetary policy and macroprudential policy having different objectives, and different instruments. This approach attempts to avoid a conflict of objectives or trade-offs in the application of the interest rate instrument. Others however contend that it is not feasible to completely separate the two policies, as interest rates affect financial stability, while macroprudential tools affect credit growth and therefore inflation.
While macroprudential oversight has emerged as a distinct policy focus, there is still no unanimity as to the best governance arrangements for such oversight. Some central banks, for example the Bank of England, have established financial stability committees, and this places the mandate for financial stability squarely in the court of central banks. A different approach is expressed by Lars Svensson at the Swedish Rijksbank who argues that as is the case with fiscal policy, financial stability policy is conceptually different from monetary policy, and should be conducted by a completely different institution and different individuals.
More significantly, while it is generally agreed that the interest rate should be the main instrument of monetary policy, the choice and efficacy of financial stability instruments is still very much work in progress. Given the different institutional structures in different countries, the type of instruments will also likely differ from country to country. However there are some common themes in the evolving thinking on this topic. It is generally accepted that excessive credit extension is at the heart of most asset price bubbles, and therefore the instruments that are used should be directed at preventing excessive leverage in the relevant markets.
Charles Goodhart, for example, has argued that because housing markets are at the epicentre of most financial crises, it is sufficient to focus on instruments that prevent excessive build-up of housing related credit. These include setting maximum loan-to-value ratios on property transactions and borrowing restrictions in relation to disposable income.
Other possible macroprudential instruments include reserve requirements for particular types of loans, contracyclical capital buffers for banks and capital surcharges for systemically important banks. In general these are microprudential tools, some of which had previously been applied in many countries as anti-inflation measures and since jettisoned, but are now being reconsidered with a narrower focus. It is still premature to pronounce on the efficacy or success of such policies, and this remains work in progress.
In reality financial stability is generally a shared responsibility as financial crisis resolution often requires fiscal intervention and the application of public money. This has the potential to create challenges for central bank independence. Even if narrow monetary policy independence remains intact, the boundaries between monetary and macroprudential policies are often blurred.
However, as Jaime Caruana of the BIS argued in his address at the Bank last year, because there will always be strong opposition to central banks resisting asset price accelerations, or taking away the punchbowl when the party gets going, the arguments for independence apply with even greater force, and such independence will be needed not only from political cycles but also from the financial markets.
In South Africa, the implicit financial stability mandate of the Bank was made explicit in the letter from the Minister of Finance to the Governor in February 2010. The Bank’s financial stability committee has been reconstituted and meets in alternate months to the MPC. Furthermore, and in recognition of the shared responsibility for financial stability, a Financial Stability Oversight Committee (FSOC) has been established, jointly chaired by the Governor and the Minister of Finance. The focus of the Bank FSC is on crisis prevention, and while the FSOC does review current conditions, the primary purpose of this committee is crisis management and resolution.
Furthermore, government has decided that regulation and supervision of the financial sector should move to a “twin peaks” approach. This means that all prudential regulation will become the responsibility of the South African Reserve Bank (SARB), while market conduct will be the province of the Financial Services Board. This is a very significant reform and adds to the tasks and responsibilities of the Bank.
Work is on-going to determine the appropriate policy instruments. To date, while counter-cyclical measures have been introduced to address the fallout of the on-going global economic crisis, South Africa’s banking system has emerged largely unscathed. While there is no room for complacency, no policy decisions have had to be taken in respect of financial stability.
One area of potential concern for financial stability is the acceleration in the growth of unsecured lending by the banking sector, although unsecured lending is not restricted to the banking sector only. This form of lending has been growing at rates in excess of 30 per cent and, at face value, anything growing at such elevated rates must be a cause for concern. But this also needs to be seen in context as total loans and advances, of which unsecured lending forms part, is still growing at relatively moderate levels. Both the FSC of the Bank and the Registrar of Banks are keeping a close eye on these developments, and are trying to understand this phenomenon more fully.
Without coming to any definitive conclusion on the issue, a number of points can be raised. Firstly, should we be concerned about this from a financial stability, a microprudential, or a monetary policy perspective? From a monetary policy point of view, we would need to understand the extent to which this lending translates into excessive expenditure with potential inflationary consequences.
At this stage, this does not seem to be the case. While it may well be that unsecured lending may have contributed to the growth rate of 5 per cent in household consumption expenditure, this growth was in line with real income growth and not considered unsustainable or excessive.
In fact, there are signs of a moderation in the rate of growth of household consumption expenditure in the first quarter of 2012, particularly with respect to durable goods. Furthermore there is little evidence that CPI inflation is being driven in any meaningful way by excess demand pressures.
A microprudential approach would focus on the risk profile and key ratios of the individual institutions. This is part of the function of the Registrar of Banks. At this stage, there are no signs of stress in any of the banks or in the banking system as a whole. To the contrary, the ratio of impaired advances to total loans and advances, which had remained stubbornly high for some time, has declined from 5,8 per cent in March 2011 to 4,6 per cent in March 2012.
From a more systemic perspective, the risks to the banking sector at this stage appear to be limited, as unsecured lending (which includes traditional forms of unsecured lending such as overdrafts, credit cards and loans to SME’s), at around 8 per cent, is a relatively small proportion of total lending although this ratio has been rising. If loans to SME’s, overdrafts and credit cards are excluded, the ratio is 4 per cent.
From a financial stability perspective, we also need to consider whether it is likely to lead to asset price bubbles elsewhere in the system. The housing market remains very subdued, with some of the house price indices still reflecting falling prices. Although the equity market reached an all-time high in April before falling back in recent days following increased global risk aversion, it would be difficult to argue that this is a bubble. In any event the link between high rates of growth in unsecured lending and equity or bond price movements is extremely tenuous at best.
Growth in credit extension by banks to the private sector has been relatively muted. Growth over twelve months in private sector credit extension was 7,3 per cent in April, down from to 9,2 per cent in the previous month. It is significant that total loans and advances to households, which includes unsecured lending, is still only growing at annual rates of around 7 per cent.
There is a potential concern about the possibility of an excessive burden of household debt. Although the ratio of household debt to disposable income is still high, it has been declining. Having peaked at 82,3 per cent in 2008, the ratio declined to 74,6 per cent in the final quarter of 2011. Household debt is nevertheless still rising, but at a slower pace than that of disposable incomes. So at a macro level, there does not appear to be a cause for concern. The concern, however, could be the distribution of these debts. There is the risk that some households are becoming over-leveraged, but this would suggest that the limits set by the National Credit Act are not being adhered to.
A further concern that is often expressed is: what happens when the interest rate cycle turns and many of these new borrowers may find it difficult to service their loans. Some of these loans are on a fixed interest basis with monthly repayment commitments, thus the servicing of the loans should not be materially affected by interest rate increases. There is also evidence that some of the growth in unsecured loans is as a result of loans with a greater value and longer repayment term being extended to existing clients with a proven repayment record.
It is also possible that we are observing a structural change with respect to bank lending. Mortgage credit extension is subdued and it is more difficult than in the past for households to access their mortgage bonds to finance consumption expenditure. So to some extent, this may be an adjustment away from mortgages and once the adjustment has taken place, more sustainable rates of increase should be observed. But because rates charged on unsecured lending are much higher than on mortgages, consumers are getting access to more expensive credit, although this is not evident yet in the data on the cost of servicing household debt which has declined consistently since the end of 2008.
These developments are being closely monitored for signs of unsustainable increases which could have systemic implications for the banking sector, for financial stability in general or for inflation. But as I have illustrated above, the answer is not simple.
In conclusion, the context that central banks operate in has become increasingly complex. The global growth outlook has deteriorated in recent weeks, and coupled with increasingly volatile and risk averse financial markets in response to uncertainties in the Eurozone, central banks will again be expected to play a core role in helping to manage or resolve the crisis.
The scope for fiscal policy to play a meaningful role is limited by the high sovereign debt ratios which are part of the underlying problem. There are also increasing concerns about the sustainability of expanded central bank balance sheets and the abnormally low global interest rates, which are seen by some as already sowing the seeds of the next crisis.
At the same time, the financial stability focus, the tightening of banking regulations and the move towards Basel III is having the effect of increasing deleveraging by banks, particularly in Europe, and raising the cost of capital for banks. These developments are focused on ensuring a more stable financial system in the future, but in the process are having a contractionary impact on growth. The potential for conflict between the different mandates are evident.
In South Africa, we have not had to make any exceptional liquidity provision, and monetary policy has been more accommodative than it would have been in the absence of below-potential growth rates and a persistently negative output gap. Inflation is expected to return to within the target range in the near term, and to remain contained over the forecast period. The banking system is stable despite challenges expected in implementing Basel III, especially over the longer term. However, we vigilantly monitor global developments and the possibility of contagion to our economy.
It is important that, in extending the mandates of central banks, there should be a clear understanding of what central banks can and cannot do, and an appreciation of the possible conflicts between the different objectives. Central banks, in normal times, function in an uncertain environment.
In the prevailing difficult global conditions uncertainty is at an even higher level, and many of the actions taken have no precedence. It is therefore incumbent upon central banks to share information and together learn from our collective experiences. As the lines between the various mandates become increasingly blurred, there is a danger that the burden of expectations could be excessive, and ultimately undermine confidence and credibility in central banks themselves.
These challenges are being considered by central banks and others, and require that all of us better understand the immediate challenges of the mutating, extremely severe global financial crisis. At the same time we need to appreciate that measures required to deal with the crisis may well have unintended consequences for central banks, their mandates, independence, capacity and role in society.
Thank you again for the opportunity to speak to you.
Address by Gill Marcus, Governor of the South African Reserve Bank at the Gordon Institute of Business Science (GIBS) in Johannesburg on the Changing Mandates of Central Banks: The Challenges for Domestic Policy
