A public lecture by Lesetja Kganyago Governor of the South African Reserve Bank Stellenbosch University (virtual)
This is a year of anniversaries for the South African Reserve Bank (SARB). The institution itself has reached 100 years. The rand has turned 60. We are commemorating 25 years of central bank independence, and also 21 years of inflation-targeting.
Inflation-targeting is the youngest of these achievements, but it may also be the biggest success.
Turning 100 is always a special anniversary, but for 73 of those years the SARB was embedded in an undemocratic state. Similarly, we are proud of the rand, but we have not always done a good job defending its value. Inflation has averaged 8% over the past 60 years and 9.4% before the introduction of inflation-targeting, leading the currency to lose most of its buying power. Even central bank independence coincided with problematic policy choices in the early years before we adopted inflation-targeting, as I will discuss later in this speech.
By contrast, over 21 years of inflation-targeting we have largely succeeded in delivering on our objective, despite many challenges. Lower inflation has benefitted the economy generally, and households of all income levels in particular. As lenders require less compensation for inflation, interest rates have come down. Furthermore, with expectations anchored inside the inflation target range, businesses no longer raise prices as soon as the exchange rate weakens. This credibility has helped us cut interest rates to record lows during the COVID-19 crisis – and we have been able to keep rates low during the recovery. This stands in stark contrast to some of our peers, where higher inflation has forced rate hikes, even though the pandemic is ongoing.
With the stimulus of low interest rates, and the security of well-anchored inflation expectations, monetary policy stands out today as South Africa’s main source of macroeconomic resilience. This is an area where we as a country have made careful policy choices, and then stuck with them, to our long-term benefit. As we confront our other economic challenges – such as our mass unemployment rates, persistent electricity shortages, record debt levels and, most recently, the wreckage of riots – we should reflect on our inflation-targeting record and take inspiration from it.
In this speech, I will discuss reasons for this success, before turning to the future of inflation-targeting in South Africa.
Four reasons for the success of inflation-targeting
The first factor behind our inflation-targeting success has been realism. Warren Buffet once commented that ‘when a manager with a reputation for brilliance tackles a business with a reputation for bad economics, it is the reputation of the business that remains intact’. Much the same is true of central banks. There are many things that people would like a central bank to do. But even a great central bank will fail at an unrealistic task. Inflation-targeting has asked a lot of the SARB and South Africans generally, but not more than any of us could deliver.
For a start, inflation-targeting is more flexible than critics claim. It doesn’t require the central bank to cancel out every price shock using interest rates, or to have perfect forecasts. Shocks are inevitable, and so are forecast mistakes. What central banks need to do is convince people that they will do what it takes to steer inflation back to target, over a realistic time frame. This is achievable and has repeatedly been achieved, with the key result that the expectations South Africans hold of future inflation have consistently been falling.
The same cannot be said for other objectives. In the 1970s and 1980s, many central banks tried targeting money supplies, but these turned out to be much less predictable than expected, and also less relevant for people’s lives – no ordinary person cared about the growth rates of base money or M3. But they did care about inflation.
Similarly, central banks have also tried targeting exchange rates – an objective that in many cases, including our own, has ended with billions of dollars thrown away for no tangible gain. We have avoided exchange rate adventurism throughout the inflation-targeting era – a policy that has served us well.
Unemployment is another area which could have turned good policy into bad. South Africa is a country where, even in the middle of the greatest boom in a century, unemployment never fell below 20%. Looking at labour utilisation – the share of working-age people who have jobs – there is a huge gap between us and our peer countries, which cannot be explained by variables such as growth or interest rates. We have spatial patterns, entrenched in our landscapes, that put large distances between jobs and workers. Our labour legislation protects insiders to the detriment of outsiders. Our education system absorbs a disproportionally large share of national resources yet delivers some of the worst outcomes anywhere, including in much poorer countries.
The chances of monetary policy overcoming all these structural constraints are slim to none. In this situation, an employment mandate for the SARB would have played out in one of two ways. It might have been a vague aspiration, adding nothing to our existing concern for the state of the real economy. Or it would have been an excuse to make policy mistakes.
The easiest way to destroy price stability in South Africa would be to insist on low interest rates because of unemployment. Our labour market is so dysfunctional, this excuse would rule out ever raising rates – a policy that would leave us in the worst-case scenario of high unemployment and high inflation. That excuse might have impressed people, and won us sympathy, but it would have been profoundly irresponsible. The SARB has not faced a tragic dilemma where there is no right answer – and we must have either jobs or inflation. We have faced a necessary choice, to do the good we can do, understanding the limits of our powers.
In making this choice, we have equipped ourselves to respond forcefully when there is a genuine cyclical downturn, beyond the structural labour market problems that have long blighted this economy. The fact is, we have lost about 1.5 million jobs from the COVID-19 pandemic. And even though inflation is now back in the middle of our target range, we still have interest rates at record lows. If this were solely about inflation, we would have raised rates already.
So never let anyone tell you that the SARB only cares about inflation and ignores jobs. Rather, bear in mind that we have this power only because we put price stability first. We have anchored inflation expectations. We did not commit to an impossible mission.
The second secret of the success of inflation-targeting is that it is simple. When we started debating it in the 1990s, some people complained it was too simple. At the time, the SARB was using what was called an ‘eclectic approach’, on the grounds that lots of different things mattered and had to be taken into account. The problem was that this made policy complex and hard to understand, both for the public and for the policymakers themselves.
By adopting inflation-targeting, we made a clean break from the era of opaque policy and confusion. Of course, real-world problems are complicated and nuanced, but without clear objectives and techniques to resolve trade-offs, policymakers easily get lost. The sad truth is: if everything is a priority, nothing is a priority. As I have learnt from my years in policymaking: while many people like to indulge in complexity – favouring obscure jargon, dense documentation and complicated plans – the smartest actors prize simplicity and clarity. Inflation-targeting helped give us strategic focus.
The third, and crucial, factor behind our inflation-targeting success has been accountability. Our goal is clear and public, and the evidence for evaluating our performance has been readily available in the form of inflation data independently produced by Statistics South Africa (Stats SA). Objectively, targeted inflation has averaged 5.8% since 2000, which is within the 3–6% target range. It has been 4.5% over the past five years, exactly in the middle of our target range.
Of course, it takes some expertise to assess monetary policy properly, but we have not been short of scrutiny: from markets, academics, analysts, international institutions and others. Consistent with inflation-targeting practice, and with our democratic dispensation, we have also been highly transparent, publishing our data and regularly engaging stakeholders to explain our thinking and answer critics. We certainly haven’t won everyone over, but, on the whole, the expert verdict is that we have implemented the framework effectively. We can feel secure in that judgement because we have robust accountability procedures.
To some extent, the SARB’s success with inflation-targeting also comes down to a fourth factor, which is good luck. Unlike some other policy tasks, such as education, monetary policy can be delivered from a single head-office and maybe a few regional offices. Furthermore, thanks to factors such as our constitutional independence, we have been able to avoid State Capture, unlike many other organs of government. This means we have stayed focused on serving the public, instead of diverting resources to patronage networks.
We must also acknowledge a helpful global environment. In pursuing low inflation, we have been able to draw on a wealth of global knowledge. We have benefitted from disinflationary forces, including globalisation. We’ve even had very low global rates, in fact the lowest interest rates ever recorded.
That said, we should not overstate the impact of a benign global environment. Looking around at the nine nearest countries in our region, there are five that use the rand or, in Botswana’s case, a currency peg where the rand plays a big role. Their average inflation rate for the past five years has been 4.5%. The remaining countries have had inflation rates, for that period, of 9%, 11.5%, 23.9% and 164.5%. Plainly, there is still something to be said for domestic policy, and inflation cannot be explained only by global factors.
Indeed, to be honest with ourselves, we should go further and acknowledge that, from a global perspective, South Africa has actually been a relatively high-inflation country. While labour market rigidities and high administered prices have contributed to this high inflation, the inflation-targeting regime itself could have helped us much more.
Towards a better inflation target
The original target range was meant to shift down from 3–6% to 3–5% by 2004, and then further down to 2–4%, following a clear glide path. But then, after losing our nerve in 2001 when the rand depreciated during the Argentine crisis, the potentially inflationary consequences led us to put off moving lower than 3–6%. This was a major policy mistake, unfortunately, because it entrenched higher inflation and higher inflation expectations. The target range of 3–6% was eventually interpreted as a target point of 5.99%, resulting in nominal interest rates several percentage points higher than they might have been.
I feel we can say with confidence that if the inflation target had been revised down as planned, we would have had lower inflation subsequently. Every other emerging market that started targeting inflation around the time that we did has revised its target lower at least once. This helps explain why we have remained a relatively high-inflation country. Looking at other middle-income countries with working monetary policies, their average inflation rate for the past five years has been just over 3%. Our average of 4.7% ranks us last in this group. This comparison could be made less unflattering if we included poorer countries and high-inflation outliers, but we should be honest, benchmark ourselves against our true peers, and admit we have been underperforming.
Only now, with the COVID-19 shock, have we seen what it’s like to have inflation rates nearer 3% and, with that, low interest rates. The last time South Africa had rates this low for this long was in the 1960s. And that wasn’t because the economy was in crisis; the economy was actually booming. The low interest rates in the 1960s were possible because of low inflation. It follows that, if we want to keep interest rates low, the most important thing we can do is to lower the inflation target.
As I have argued previously, a more appropriate target would be a point target of around 3% or 4%, putting us in the same territory as our peers. Given the uncertainties around inflation, it would be useful to bracket the point target with an error range: probably plus or minus 1 percentage point. We will not be able to control inflation precisely all the time, and it is useful to acknowledge that up front. That said, it would be useful to have a clear point target rather than a range, to anchor expectations and to prevent any target drift in future, like the one we suffered when the 3–6% target range morphed into a 5.9% target. I take the view that 3% would be a good point target.
Naturally, a step like this would involve some work. Part of this would be coordination with government. The task of locking in a lower target will be fundamentally easier and cheaper if we get buy-in from administered price setters.
Stats SA reports a measure of inflation excluding administered prices, and this is already in the region of 3% – it has averaged 3.5% so far this year. This shows that much of the economy is already achieving lower inflation. Items like water and electricity need to be priced with lower inflation in mind, and lower prices would in turn help us lock in a lower target, anchoring expectations at permanently lower levels.
But getting these commitments would require a corresponding commitment and timeline for a lower target. The right way to do this would be through an exchange of letters between myself as the SARB Governor and the Minister of Finance, whom the Constitution identifies as the person we should regularly consult with. This process of consultation should also feed into government’s debt management strategy, to maximise the benefits of lower interest rates.
I am well aware that this is a delicate moment for South Africa’s economic recovery, and also that global inflation has picked up. That said, I am even more painfully aware of a South African disease: we are scared of reform; we make every excuse to avoid it; we emphasise any short-term pain and discount any long-term benefits; and then we sit around wondering why our economy is stagnating, why our young people can’t find jobs, and why we are getting steadily poorer relative to the rest of the world.
If indeed the SARB is a credible and effective institution, then you should expect us to deliver the number one thing in central banks’ power to achieve, which is low and stable inflation. You should also not be especially worried about disinflation costs. We have made progress in getting inflation expectations from the top of the 3–6% target range to the 4.5% midpoint, and we achieved this without relying on high interest rates. But 4.5% is not an optimal target – it’s just a logical and balanced interpretation of a target we should have reformed two decades ago.
We should really be benchmarking our performance to the high achievers in our class, not contenting ourselves with doing somewhat worse than average. And temporarily higher global inflation isn’t the main problem; the main problem is the wedge separating us from these higher-performing peers. There is no chance of fixing this unless we start having this conversation in a serious way. This is why I have been making an effort to talk about the target this year. Good policy is about making progress, not just holding the line and fending off crises. And I am deeply committed to good policymaking.
Conclusion
Ladies and gentlemen, in conclusion.
In South Africa today, we are playing for high stakes. There are things going right and there are things going wrong. We could tip into a bad equilibrium – or into a good one. The recent spate of looting gave us a glimpse of an unacceptable future: a prolonged breakdown in law and order, large numbers of violent deaths, extensive destruction of property, and shortages of even basic goods such as fuel, food and medicine.
At the same time, 2021 has brought us good news. Prices for our export commodities have increased strongly. Our terms of trade, which compare the costs of our imports to the prices of our exports, have never been more favourable. Our economic recovery from the COVID-19 shock has proceeded faster than expected, despite disruptions. Although we are learning many horrifying details about State Capture, the process of repairing our institutions is clearly moving forward. There is tangible progress on structural reforms, for instance in the energy sector. South Africans are finally being vaccinated against COVID-19 in large numbers.
As Governor of the SARB, I head a 100-year-old institution. We have endured many difficult episodes during our country’s troubled history. Sometimes we made serious mistakes; at other times we did well. Over the past 21 years, we have delivered on our commitment to control inflation. We are proud of this record, and we want to do better still in future.
In this speech, I have outlined some key components of inflation-targeting that have helped secure its success. The framework is simple. It is realistic. It keeps us accountable. And, as we must acknowledge, it has been bolstered by favourable global conditions, or good luck. We have avoided hundred-point plans, hazy aspirations and no consequences for failure. We are forward-looking and ambitious for the country, which is why we are talking about a lower inflation target, set to bring us in line with our best-performing peers.
I hope that this experience and this attitude will help inform and inspire others – in the private sector, in the public sector, and in civil society – who want South Africa to succeed and who are willing to work hard for that. We don’t have it easy, and sometimes the task seems overwhelming. But there are ways to succeed, and we need to remain ambitious that South Africa can be a better country.
Thank you.
