François Groepe: The changing role of central banks
Address by Mr François Groepe, Deputy Governor of the South African Reserve Bank, at the University of the Free State, Bloemfontein, 12 August 2016.
Good morning - and thank you for the opportunity to address you.
The past few years have been extremely challenging, both domestically and internationally. The aftershocks of the global financial crisis of 2008-09 have persisted. The advanced economies are still struggling to recover on a sustained basis, while the emerging markets, initially the main engine for the recovery, have fallen out of favour in the past four years, as the Chinese economy has slowed and brought commodity prices down with it.
Low growth is not the only issue. We are seeing increasingly widespread discontent about rising income and wealth disparities in many countries, where the fruits of growth have not been equitably shared and have been aggravated by persistently low growth.
The initial responses to the financial crisis involved both monetary and fiscal policy loosening. But fiscal space was eroded very soon as debt ratios rose to unsustainable levels. This placed a near impossible burden on monetary policy, which persists to this day.
Referring to the post-crisis focus on central banks, Mohamed El-Erian appropriately titled his most recent book The only game in town.1 Today, I would like to highlight the role that central banks can play in the economy and, perhaps more importantly, point out the limits to what central banks can do. It is important to understand these limits because, I would argue, the biggest risk to central bank independence is the possible backlash from being unable to deliver on unreasonable expectations. Central bank mandates have expanded - perhaps appropriately so - but there are limits to what monetary policy was designed to achieve. Central banks cannot be, and should not be regarded as, "the only game in town".
The role of central banks has changed over time. The first central bank was established in Sweden in 1668 and had a chequered early history, from providing commercial banking services to financing a war against Denmark. Most of the early central banks were in fact initially private banks, and for this reason, a number of them (including the South African Reserve Bank) have private shareholders to this day. Over time, however, these banks evolved into official institutions, with various roles.
It is not my intention to give a historical account of the evolution of the role of central banks through the ages. Rather, I will fast-forward to the modern period and highlight the current thinking about central banks and their appropriate mandates and policies. Important to note is that the role of central banks is still evolving, so what is conventional wisdom today could well be supplanted tomorrow. Although central banks are no longer in the business of financing wars, the cost of maintaining peace and stability during periods of crisis is perhaps far higher!
Prior to the financial crisis, there were a number of generally accepted views about the role of central banks. Besides their exclusive responsibility for printing and issuing banknotes and coins, the almost unquestionable view was that a principal role of central banks should be to maintain price stability. While this opinion was generally accepted, the choice of framework was not. Before the 1990s, the focus was on intermediate targets for monetary policy, for instance money supply targets. More recently, the trend has shifted towards targeting inflation directly, generally through using an official interest rate as the policy tool. The inflation-targeting framework evolved, with the objective of monetary policy clearly specified and quantified. It is worth pointing out that the choice of framework is constrained by the nature of the exchange rate regime: in a fixed exchange rate environment, the focus is on maintaining the peg, and monetary policy is determined by the policies of the currency to which it is pegged.
It may be rightfully asked why price stability should be elevated to a policy objective. There are a number of reasons why we want to pursue stable prices. First, it makes planning investment for the future so much easier. The more unpredictable future prices are, the higher the risk premium that investors are likely to price into their decisions. Although price stability is not in itself a generator of economic growth, it is an important prerequisite for providing a sound macroeconomic basis within which growth can be achieved. Second, it is generally the case that it is the poor who are most vulnerable to the ravages of high inflation, as they are the least able to hedge against this. The only people who benefit from high inflation are those who have fixed-interest nominal debt, as the real value of the debt declines during inflationary periods. However, under these conditions, lenders are reluctant to lend at fixed rates and demand higher interest rates to be compensated for higher expected inflation. Therefore, high-inflation countries tend to have high nominal interest rates.
The more contentious issue is the impact that monetary policy can have on economic growth and employment. It is generally accepted that monetary policy can play a role in stabilising the business cycle. The concept of 'the potential rate of output growth' is central to this view. This concept refers to the maximum growth rate which an economy can sustain without generating inflationary pressures and consistent with employment at the natural rate. Potential output is determined by structural factors in the economy, such as the availability of capital and labour, infrastructure, technology, skills levels, and trade policies. Changes to potential output are determined by changes in these factors. It follows then that if growth is above potential, there will be inflationary pressures and the role of monetary policy will be to moderate these pressures by reducing demand pressures to be consistent with price stability. We do this through raising interest rates.
The converse also holds true. When the economy is growing at levels below potential, demand is weak, as are inflation pressures, and monetary policy has room to stimulate the economy by reducing interest rates. In other words, under normal conditions, monetary policy should be conducted contra-cyclically, by boosting growth in a downturn and cooling down an overheating economy during a cyclical upturn.
The central point here is that monetary policy has little role in determining potential output. Its impact on output is cyclical. The challenge is that potential output is not observed but estimated. Although there are a number of techniques with varying levels of sophistication, these estimates are subject to a great deal of uncertainty and often vary quite considerably from one to another. This means that, while we can observe how current output changes, it is difficult to be certain at any point in time whether the potential itself has changed. This is critically important for policy purposes. For example: if cyclical growth falls below potential, then countercyclical monetary and/or fiscal policies are appropriate, but what if the current observed growth slowdown is also a reflection of a decline in potential output? This could mean that countercyclical policies would not necessarily be effective.
This issue is at the heart of the current debates about the persistence of low growth in the global economy. Some have argued that below-potential growth is a cyclical phenomenon that can be resolved through contra-cyclical macroeconomic policies by stimulating demand. However, others have warned that we are experiencing a period of secular stagnation, associated with a prolonged period of low growth and declining potential output. If this is indeed the case, it follows that the effectiveness of monetary policy in stimulating the economy will be limited. What is required under such circumstances is structural change in the economy, in order to stimulate supply.
Monetary policy may help to prevent deflation by underpinning demand to some extent, but it is unlikely to be the primary source of growth.
Recently, many countries have revised down their estimates of potential output growth. However, it is unclear whether these are short-term estimates or whether a prolonged period of secular stagnation is with us. Domestically, we at the South African Reserve Bank have also successively revised down our estimates of potential output, from around 3,5 per cent in 2010 to the current levels of around 1,4 per cent. We see these as an estimate of short-term potential output; our longer-term estimate is higher, when, for example, the electricity supply constraint is expected to be relieved. But the longer-term estimates remain lower than previous ones. This constrains the ability of monetary policy to stimulate cyclical growth, as the output gap - which is the difference between actual output and potential output - is not as wide as would be the case with a higher potential output. Thus, if the growth problem is structural in nature, the scope for successful monetary stimulus is limited.
When the global financial crisis struck, it was viewed as a cyclical downturn, and there was almost universal agreement on the need for stimulatory monetary and fiscal policies. While fiscal space was quickly eroded, monetary policies in the advanced economies remain highly expansionary and 'unconventional'. It used to be assumed that once interest rates reached the zero lower bound,2 then monetary policy would become ineffective.3 But the post-crisis period saw attempts at monetary stimulus through quantitative easing (or QE) which took various forms but in essence entailed central banks buying government or corporate bonds, thereby injecting liquidity into the system. While the US ceased its QE programme in 2014, the Bank of Japan and the European Central Bank have maintained theirs while the Bank of England recently resumed its QE programme in the wake of the UK's decision to leave the European Union. Interest rates in the advanced economies remain extremely low and in some instances in fact negative. In early 2015, both the US and the UK were expected to be the first of the advanced economies to begin raising interest rates. The US tightened policy in December 2015 but it has remained on hold since then, while the UK has recently reduced rates further. Unfortunately, this highly expansionary monetary policy environment has not generated the growth that was hoped for, but it can be argued that it prevented the immediate situation from getting worse.
I should note that there are potential negative effects from very low or indeed negative interest rates, not least of which is that they create problems for defined-benefit pension funds which have to provide cover for ageing populations while many of these funds face the risk of underfunding. The amount of government debt trading at negative yields has increased significantly since the beginning of the year. Pension funds are therefore forced to search for higher yields and, almost by definition, they move their funds to more risky investment categories and jurisdictions. This is a potential time bomb waiting to explode, far more serious than the concerns over the potential inflation risks of ultra-loose monetary policy. In fact, inflation is not currently a concern in the advanced economies, and in some instances the concern is that inflation is too low.
The situation in emerging markets and developing economies is more complicated, particularly in those that are currently facing rising inflation and slowing or low growth. These inflation pressures have typically been driven by weakening exchange rates in response to declining commodity prices - and not by excess demand where monetary policy is most effective. This is the case in South Africa at the moment. Our economic growth outlook is weak following the contraction in the first quarter of this year. The South African Reserve Bank forecasts 0 per cent growth this year and below-trend growth of just 1,1 per cent and 1,5 per cent in 2017 and 2018, respectively. At the same time, inflation is outside our target range of 3 - 6 per cent and is expected to remain outside the range until the third quarter of 2017. However, demand pressures are weak while inflation is driven primarily by supply-side factors, including drought-induced food price increases, a relatively depreciated currency, and administered prices.
None of this means that monetary policy should not be used under such circumstances. We know there is nothing we can do about the impact or first-round effects of these supply-side shocks. We try to look through these effects and focus on the second-round effects, where higher prices feed into higher generalised inflation and higher wage settlements. If we ignore these pressures, inflation expectations are likely to rise. And if wage negotiators and price setters believe that inflation will accelerate, they will adjust wages and prices accordingly, and in so doing bring about a self-fulfilling prophecy.
This scenario of high inflation and low growth creates a dilemma for monetary policy. The tightening of monetary policy could have a negative impact on the already weak economy, but failure to act could cause inflation expectations to become unhinged and ultimately undermine inflation. It is a difficult balancing act. Since January 2014, we have been in a gradual tightening cycle, with a cumulative increase of 200 basis points since then, the most recent a 25 basis point increase in March. The moderate nature of the tightening cycle is due to our concern over the growth outlook; the tightening would have been more aggressive had the economy been more buoyant. Furthermore, real interest rates remain relatively low, implying an accommodative monetary policy stance despite the tightening.
But central banking is not only about monetary policy. Since the global financial crisis, the mandates of central banks have expanded, particularly with respect to financial stability. Prior to the crisis, many central banks had an implicit mandate, or no mandate at all, in this regard. In South Africa, for example, although the South African Reserve Bank had explicit responsibility for regulating and supervising individual banks (micro-prudential policy), it did not have an explicit mandate to ensure the stability of the financial system as a whole. By way of an example, excessive price increases or bubbles in asset markets, including the housing market - particularly if financed by excessive leverage or borrowing against these assets - could pose a risk to the financial system at large should the prices of these assets collapse.
Prior to the global financial crisis, conventional wisdom dictated that there was not much that central banks could do to cool down overheating asset markets. Indeed, many - including Federal Reserve Bank Chairman Alan Greenspan - argued that there was no reason why central banks were better placed to recognise asset price bubbles - and that the best they could do was to clean up after things had already fallen apart. The predominant view at the time was that things were unlikely to fall apart.
There were, of course, dissenting voices, notably at the Bank for International Settlements, who argued that central banks should not have a narrow focus on inflation only and should rather lean against this excessive leverage with higher interest rates, focusing on the financial cycle, which is usually longer than the business cycle. This would have implied higher interest rates than those prevailing before the crisis, despite the low inflation environment.
History shows that these dissenting voices were correct; current conventional wisdom prescribes that financial stability should be an explicit focus of central bank policy. There are disagreements, however, about where this responsibility should lie and what instruments should be used. Some argue that it should be part of monetary policy and that interest rate settings should take the financial cycle into consideration. Others argue that monetary policy should remain focused on inflation and that the task of financial stability should be given to a separate committee, although it is not self-evident whether this separate committee should be housed within the central bank or outside the central bank in a separate institution, as is the case in Sweden.
In South Africa, we have adopted a position similar to that of the Bank of England, where the responsibility for financial stability is given to the central bank overseen by a financial stability committee. This committee considers the stability of the financial system more broadly, rather than regulating and supervising individual banks. The latter remains the job of the Bank Supervision Department at the South African Reserve Bank, which is transforming into a Prudential Authority in terms of the Financial Sector Regulation Bill (FSR Bill) currently before Parliament.
The promulgation of the FSR Bill will make the Bank's responsibility for financial stability explicit. The Bill sets the provisions that will give effect to the Bank's financial stability mandate. The Bank will be responsible for protecting and enhancing financial stability, and for maintaining or restoring financial stability if a systemic event has occurred or is imminent. In monitoring the strengths and weaknesses of the financial system, as well as any risks to financial stability, the Bank will be expected to advise the financial sector regulators and other agencies of the steps required to mitigate risks to financial stability. The FSR Bill also requires the Bank to publish its assessment of the stability of the financial system at least every six months in the Financial Stability Review, which is to be submitted to the Minister of Finance and the Financial Stability Oversight Committee.
Once the FSR Bill is promulgated, the Prudential Authority will regulate individual banks and insurance companies while the Financial Services Board, previously responsible for overseeing the entire insurance sector and other non-banks, will assume responsibility for market conduct within the banking and wider financial sector. So while monetary policy, macro-prudential policy and micro-prudential policy are all separate functions of the Bank, their coordination is facilitated by them being housed within the same institution, with some degree of overlapping membership.
But we could ask whether the central bank has a wider role to play in the economy, particularly within emerging markets. We are often asked why the South African Reserve Bank does not get involved with direct lending or development finance, be it to industry or the agricultural sector, and whether we should not be more involved in promoting financial inclusion.
Theoretically, there is no reason why central banks cannot perform wider functions. Generally, apart from the functions I have outlined above, central banks have the important role of being responsible for issuing currency. Furthermore, in South Africa, the banking sector plays a prominent role in the national payment system, which facilitates the settlement of millions of transactions every day; the total value of transactions settled in the 12 months to March through the Bank's SAMOS4 system amounted to R123,9 trillion. We take for granted simple things like credit card or other card transactions that are settled immediately. This requires a significant infrastructure, and the South African Reserve Bank ensures that the rules are applied. At a regional level, we have played an important role in extending this infrastructure and expertise to a number of countries in sub-Saharan Africa, which in turn helps to facilitate intra-regional trade and promotes financial inclusion by helping communities access financial services.
In some countries, central banks play a prominent role in the allocation of credit to various sectors. For example, the Nigerian central bank plays an important role in directing credit to the agricultural sector, and in South Korea the central bank decides how much credit banks should extend to small and medium enterprises. Whether such functions should indeed be performed by a central bank is often dependent on the nature of the financial system, its level of sophistication, and the existence of other institutions that are focused on extending credit to other sectors.
In South Africa, we have a number of public sector institutions that are focused on extending credit to different sectors. These include the Industrial Development Corporation, the Development Bank of Southern Africa, the Land Bank, and a number of institutions that specialise in lending to small businesses. These functions could theoretically be performed by the central bank as a separate function, but there is no clear advantage to this - unless there were scarce skills in the economy and/or a lack of resources that would make a good operational argument for this. In many developing economies, there may be a lack of skills, and combining these diverse functions into one institution may be helpful from the perspective of achieving economies of scale.
But even if it were the case that the South African Reserve Bank granted loans to the agricultural sector or small businesses, these loans would not be financed by 'printing money'. Rather, such an arrangement would disburse funds raised in the money and capital markets at competitive rates, or it would be funded by grants from government and/or multilateral organisations. There is no reason why the Bank would necessarily have the advantage of assessing the needs of these sectors, as opposed to the specialised institutions that focus on these sectors, or have the necessary expertise to do so.
In developing economies, central banks often have a role in promoting financial inclusion. This is, however, a particularly challenging area of policy formulation and one that means different things to different people. At a superficial level, financial inclusion is often simply seen as having access to a bank account and consequently to the national payment system. Mobile banking has brought banking services to the rural areas in a number of African countries. In South Africa, the low-cost Mzansi accounts have given easy and cheap access to rudimentary banking services to a broader segment of the population.
I would, however, argue that financial inclusion entails much more than this and that, from a political economy perspective, it is a complex matter, particularly due to the broad range of opinions on how best to achieve this policy objective. There is, however, broad consensus that financial inclusion should extend to include access to loans, transactional banking, and risk products and services.5
Financial inclusion should not mean irresponsible or reckless lending to people who cannot afford it. After all, it was the unfettered provision of cheap loans to people who could not afford them, or the so-called sub-prime lending, that largely triggered the global financial crisis.
Typically, banks grant loans on the basis of some security provided by the borrower. Often, this entails fixed property. But in a country like South Africa, with a highly skewed distribution of income and wealth, the ability to provide such security is beyond the reach of a significant proportion of the population. It becomes even more complicated when we consider rural communities and communal forms of land ownership which is not conducive for providing surety for loans.
As the sub-prime crisis in the US has illustrated: while lending to disadvantaged communities may be a noble cause, there are dangers to excessive lending, particularly when such lending is simply financing consumption. In South Africa, we experienced extremely strong growth in unsecured lending between 2009 and 2013, some of which was to lower-income groups previously excluded from formal sector borrowing and some of which may yet come to manifest itself in higher levels of non-performing loans - although it is the view of the South African Reserve Bank that our banks have sufficient capital buffers to absorb any such increases.
Access to unsecured loans has meant that lower-income groups could avoid having to borrow from ruthless informal money-lenders at exorbitant rates of interest. But it has not solved the problem of over-indebtedness of the poor and its social consequences. In numerous instances, large portions of earnings were paid over to lenders (both formal and informal), often in terms of court orders, leaving wage earners with very little to live on.
There were also concerns that this form of lending and its rapid expansion posed a systemic risk to the financial sector and broader financial system. However, unsecured lending was a fairly small portion of total bank lending. Nevertheless, at a micro-prudential level, the Bank Supervision Department of the South African Reserve Bank kept a close eye on these forms of loans - and although one significant player in the field, African Bank, did collapse, its failure cannot be solely attributed to its provision of unsecured lending. The role of the central bank in this instance was not to decide who the beneficiaries of bank loans should be or how much should be lent; the role of the central bank was to ensure that commercial banks in general maintained adequate capital and made appropriate provisioning for risk.
The African Bank episode did illustrate the danger of talking about the need for financial inclusion in a glibly manner. Banks have often been criticised for not extending lending to the poorer segments of society, but when they did, they were criticised in some quarters for "reckless lending" and for facilitating the accumulation of excessive debt. While increasing financial inclusion is a laudable and important goal, a fine balance needs to be struck between protecting consumers from exploitation and extending credit in such a manner that the stability and soundness of the financial system as a whole is not undermined. The market conduct regulators, including the National Credit Regulator and the Financial Services Board, are responsible for preventing reckless lending, and the role of the South African Reserve Bank in the context of promoting financial inclusion is to ensure the stability of banks and the financial system.
There is a view that giving too much discretion to central banks in some of the areas I have highlighted could politicise the central bank unduly. This brings us to the issue of central bank independence. The main argument for independence is that it minimises the politicisation of monetary policy decision-making and avoids what is known as the "political interest rate cycle". This refers to the incentive of politicians to lower interest rates in advance of elections. The argument is that an operationally independent central bank does not have to bow to such pressures.
That is not to say that central banks are not pressurised by political and other elements in society to act in a particular way, and monetary policy in South Africa has not been without its domestic critics and political economy challenges. The monetary policy framework has been criticised, and, not surprisingly, opposition has tended to intensify during the upward phases of the interest rate cycle. But the South African Reserve Bank has been able to resist such pressures, as its operational independence is entrenched in the Constitution. The inflation-targeting framework lends itself well to the separation of goal and instrument independence: generally, central banks do not have goal independence as the target is usually, but not always, set by government, but, in order to prevent the so-called "political interest rate cycle", central banks do have instrument and operational independence to pursue the mandate given to them.
However, the global crisis has created challenges for central banks that could undermine this seemingly comfortable insulation from the political arena. The expanded mandate of financial stability in itself may have implications for their independence, while the other possible objectives - such as financial inclusion and direct finance - imply political decisions being made by unelected officials.
Compared to financial stability decisions, decisions on monetary policy, while not easy, are more straightforward and better understood by the public. They usually involve the use of one tool, namely the interest rate, and there is a clear objective. The financial stability mandate is more complicated, as it is a shared responsibility. It generally involves government in crisis resolution, particularly when public funds are involved, and the policy tools are more directed at particular sectors; it may therefore be more politically sensitive as the distributional impacts are more apparent than in the case of monetary policy. Furthermore, as has been argued in a paper published by the International Monetary Fund, financial stability is difficult to measure but financial stability crises are evident, so policy failures are observable, unlike successes. As has been noted in the IMF paper, "central banks would find it difficult (even ex post facto) to defend potentially unpopular measures, precisely because they succeeded in maintaining financial stability".6 Whichever failures may be perceived on the financial stability front have the potential to undermine monetary policy independence through a general loss of credibility of the central bank.
In conclusion, we are living in challenging times. Central banks are called on to do more and more, and are still called on to provide solutions to the low-growth environment we find ourselves in. But, as I have argued, although central banks play an important role in the economy and society at large, there are limits to what they can do - and these limits are not always well understood. Mohamed El-Erian argues that while we should give central banks due credit, their effectiveness is waning given the limited number of tools available to them. He argues that the world has come to a critical junction, and faces a choice of two roads. One road "involves a restoration of high-inclusive growth that creates jobs, reduces the risk of financial instability, and counters excessive inequality. It is a path that also lowers political tensions, eases corporate governance dysfunction, and holds the hope of defusing some of the world's geopolitical threats. The other road is one of even lower growth, persistently high unemployment, and still worsening inequality. It is a road that involves renewed global financial instability, fuels political extremism, and erodes social cohesion as well as integrity".7
This is a sombre warning - and relevant both domestically and at the global level. It is clear what the preferred road is. Taking it requires political leadership and political will. This is not a responsibility that can be abdicated to central banks.
1 El-Erian, M A (2016). The only game in town. Central banks, instability and avoiding the next collapse. New York: Random House.
2 This is when short-term nominal interest rates are at or near 0 per cent. A liquidity trap is caused and the effectiveness of monetary policy is limited.
3 This is due to the existence of physical money. As interest rates fall below 0 per cent, economic agents may switch from central bank money or electronic cash to physical cash.
4 South African Multiple Option Settlement.
5 In this regard, I prefer the term "financial deepening' over "financial inclusion" for the reasons stated above; the former is a much broader concept.
6 Bayoumi, T et al. (April 2014): Monetary policy in the new normal. International Monetary Fund Staff Discussion Note.
7 El-Erian, M A (2016). The only game in town. Central banks, instability and avoiding the next collapse. New York: Random House.