Race politics the last refuge of the scoundrel - Lesetja Kganyago

Governor explains how SARB has been drawn into the firing line of the corrupt

Speech by Lesetja Kganyago, Governor of the South African Reserve Bank 19th annual Stavros Niarchos Foundation lecture, Peterson Institute for International Economics, 15 April 2019

Principled agents: Reflections on central bank independence 

I am honoured to be here, but I am not sure if I should be happy. The Stavros Niarchos lectures tackle major world problems. In 2016, David Lipton discussed the backlash against globalisation. In 2017, Mervyn King explored global macroeconomic imbalances. Last year, Tharman Shanmugaratnam looked at wage stagnation and the decline of social mobility. This year I have been invited to discuss central bank independence.

There’s a pattern here. Central banks must be in trouble.

Looking around the world, there are signs of this trouble in many places – including advanced economies, where institutional stability can no longer be taken for granted. You will all be able to think of examples. The Bank of England has been criticised for warning about the costs of Brexit, which just shows that giving good advice in a polarised political environment is no way to make friends. As for the United States, everyone likes telling the Fed what to do, and as we have seen this urge is not confined to those without authority. Elsewhere, a conflict over the central bank’s reserves has led to a Governors’ resigning. And of course in South Africa, the independence of the Reserve Bank has also come under threat in recent years, which must be why you think I’m the person to discuss this subject.

Tonight, I will briefly review the textbook case for central bank independence, which centres on the well-known time-inconsistency problem. It is an elegant analysis, but it only covers a narrow part of central banking. In particular, it does not apply very well to financial sector mandates, such as financial stability or banking regulation, which are core duties for many central banks. I want to argue however that the case for central bank independence goes further. The issue is actually another classic economic concept: the principal-agent problem. How can societies get their leaders to look after the public interest instead of their own short term political interests? This is one of the fundamental difficulties of government. In my experience, it has also been the main reason the South African Reserve Bank has needed its independence.

Independence allowed us to deliver on our mandate, as set down in the constitution. Independence ensured that the tremendous powers of a central bank – such as printing money, or licensing and supervising banks – couldn’t be taken over by politically connected individuals bent on looting the state instead of serving the citizens. I’m not sure exactly how this should be modelled – how we could prove the point in a large empirical study, and get a peer-reviewed article out of it – but it has been my lived experience for some years now. There is a saying: there are no atheists in foxholes. We’ve been on the frontlines lately, the place where good and bad governance meet, and I promise you – in that situation, you really learn to believe in central bank independence.

Let’s start with the textbook. Kydland and Prescott showed that governments suffer from a time inconsistency problem. They’d like to promise low inflation in future, but when the future shows up, they discover it’s easier to tolerate higher inflation instead. As a result, lenders, firms and workers start to anticipate tighter inflation, and everyone ends up living in a world they would not choose, with a built-in bias towards higher inflation and higher interest rates. The solution to the problem is an independent central bank, with a clear mandate to control inflation. With this mandate in place, governments do not have to pay higher inflation risk premia on their debt, and society as a whole gets to enjoy a lower level of inflation. This is why central bank independence has been described as a free lunch.[1]

But this time-consistency argument does not cover the whole case for independence. In particular, it says nothing about financial stability mandate nor about narrower bank supervision duties, which occupy a lot of central bank staff.

One response to this is to concede that these things are not like price stability and do not require independence. By this logic, monetary policy should have its own safe space, but not other mandates. As Ben Bernanke has argued, for instance, and I quote:

…there should be no ‘spillover’ from monetary policy independence to independence in other spheres of activity. In practice, the Federal Reserve engages cooperatively with other agencies of the U.S. government on a wide range of financial and supervisory issues without compromising the independence of monetary policy.[2]

Unfortunately, it is not always and everywhere the case that the two can be combined so harmoniously. Serious threats to central bank independence are possible, even when the sanctity of monetary policy is intact.

For a telling example of the problem, consider Cyprus.[3] As a member of the euro area, Cyprus does not have an independent monetary policy. It has a large banking sector, however – and when this sector fell into crisis, in 2013, that also generated a serious threat to the independence of this central bank.

Cyprus’s banking crisis was the result of falling asset prices, particularly for Greek government bonds. The banks bought these in 2009 and 2010, and incurred haircuts as part of Greece’s 2011 debt restructuring. The result was banks that were probably insolvent, but which were also simultaneously too big to fail and too big to bail. Tackling this problem required several desperate measures, including emergency liquidity assistance from the European Central Bank (ECB), strict capital controls, a two-week shutdown of the entire banking system, and even bail-ins for bank bondholders and uninsured depositors.[4]

The politics of this were predictably toxic, especially in a small country where many people depended on the banks for their wealth and status. The central bank got the job of administering much of the crisis management and resolution process, not least because it had the trust of international lenders. Perhaps inevitably, it then became a scapegoat, with critics questioning its motives and patriotism.[5]

Although the independence of the central bank was secured by EU treaty, and the position of the Governor was protected, the Bank’s independence was compromised anyway. It achieved this by expanding the Board of the Central Bank of Cyprus (CBC) and giving it more power. The Board then switched reporting lines away from the Governor to its own members, so bank staff ended up answering to the Board instead. The European Central Bank issued a legal opinion questioning these measures, but they were adopted anyway, and the European Commission did not take legal action in response. Sidelined and under constant, personal attack, the Governor, Panicos Demetriades eventually resigned.[6]

I claim no great expertise about Cyprus’s economy, and there are probably people in the audience who know many interesting details of that country’s crisis that I have overlooked. I drew this account from the memoirs of the former Governor. The point I want to make is that the independence of a central bank can be gravely compromised without monetary policy coming into dispute. Indeed, even without that central bank issuing its own sovereign currency or having much control over monetary policy – which of course was being set in Frankfurt.

One answer to this problem is to say that independent agencies like central banks shouldn’t be controlling bank bailouts and resolutions. The distributional consequences are too large, the preferences of society are not clear or stable, and the scope for accountability and transparency is too limited.[7] The natural endpoint of this logic is that financial stability and bank supervision problems should stay in the political realm, with the line ministries and their political heads, instead of being delegated to independent agencies. In this way, central banks can focus narrowly on monetary policy, and have independence for that, without wading into the mire of finance and perhaps being dragged down.

But there are two problems with this argument for narrow central banking. First, not many central banks do narrow monetary policy. If the plan for safeguarding central bank independence is to hive off non-monetary mandates to other bodies, then we have lost our way. Most jurisdictions are adding new duties for central banks, not taking them away.

This is related to the second problem: in a crisis, central banks find it hard to stay on the side-lines or escape blame. To borrow an anecdote, recall that in the UK it was the Financial Services Authority and not the Bank of England which was the lead regulatory agency for the financial sector – but when Northern Rock failed in 2007, it was Mervyn King whose picture ended up in The Economist alongside the heading, ‘The bank that failed’.[8] The FSA doesn’t exist anymore, and the Bank of England now has responsibility for financial stability.

The lesson learned, one taken to heart in many places, was that central banks have unique powers as the lenders of last resort. No-one else can flood a banking system with liquidity like a central bank. As a result, a promise to stop a bank run, or to do ‘whatever it takes’, has unique power coming from a central bank.

These institutions also tend to have institutional capital, in the form of expertise, access to international networks, especially fellow central bankers, and reputations as responsible technocrats, which can be very helpful in coordinating responses amongst diverse stakeholders. For these reasons, a central bank sitting at ground zero of a financial crisis will struggle to stick to a narrow monetary mandate. And this is even before we reflect that financial crises are also likely to prompt big misses of inflation targets, as happened in many advanced economies after the global financial crisis.

The implication is that financial stability is also very important for achieving inflation targets.

So central banks probably cannot steer clear of financial stability and bank regulation. Yet that conclusion doesn’t make the space any safer for independent central banks. I can testify to this from my own experience in South Africa.

We are all used to thinking about attacks on central banks as demands to cut rates for political reasons. But I never once got a call or any other communication from the Union Buildings – the seat of the South African executive – telling me what to do with monetary policy. Similarly, my toughest public engagements haven’t been about interest rates; they have been about the financial system. Twice a year we have monetary policy forums outside financial centres, but we barely talk about monetary policy. Instead, we get questions about two things. One is a sideshow issue, which is the fact that the Reserve Bank has private shareholders. These shareholders don’t have any policy control, they get tiny dividends (about $14,000 per year collectively), and the private ownership isn’t the basis of our independence.

Still, it sounds strange that there are private shareholders, and people have questions, which we answer. The other thing we get asked about is the financial sector, especially issues around transformation, financial inclusion and development. And the financial sector work is also where we have had the most difficult time, politically.

Let me just pause a moment to note that monetary policy has had its problems. The South African economy has experienced a prolonged slump, with negative per capita GDP growth since 2013. At the same time, we have had inflation mostly at or above the top end of our target range. We have had to raise interest rates in the midst of the downturn, to keep inflation expectations in check and stave off worse scenarios, like the crises that hit Turkey and Argentina last year. We are now getting inflation where we want it, in the middle of our target range, which is a rate we intend to maintain. But that hasn’t provoked more than the usual grumbling and op-eds – no major onslaughts. The attacks came for other reasons. I’ll describe three big ones.

The first involved banking services for politically connected people. As has now been widely reported, some very senior figures in the South African government developed warm relations with a family of businessmen.[9] When it became clearer that a lot of that family’s money wasn’t clean, commercial banks became unwilling to handle their accounts, for fear of violating laws against facilitating money laundering.

This made it nearly impossible for the family to run its operations and sparked a political fightback. We started feeling pressure to force banks to service these accounts, in violation of the law.[10] In addition, we came under pressure to allow this family to obtain a banking license by buying a small bank. We even faced the threat of bank licensing being taken away from the SARB altogether.[11] We were using our independence to uphold a law against dirty money flows, and that made us enemies.

The second problem was over a small mutual bank. This institution had spent most of its life quietly taking deposits from retail savers and extending mostly mortgage loans. Frankly, we didn’t realise the full extent of what was going on until the very end.

Starting in 2013, new management took over the bank and turned it into a crude Ponzi scheme. Their innovation was bribing public officials to deposit municipal funds at the mutual bank in exchange for ‘Christmas presents’. They then simply spent these deposits on themselves, accumulating fancy cars and a helicopter or two along the way.[12] When we noted the significant growth in the business, we started engaging the bank about converting to a commercial banking license.

Ultimately, we realised something was wrong when they failed to make a routine payment through the national payment system, and it turned out they had no money left. But they nearly got away with it for longer, because they almost received a huge deposit from a public sector rail operator, which would have kept the Ponzi scheme going. I should also note that one of the reasons we missed what was going on was that one of the people involved in the looting was a partner at a Big-4 accounting firm, who signed off the accounts. We accepted audited accounts as a true reflection of the business.[13]

When we put the mutual bank into curatorship, we came under attack from people who said we were targeting this business because it was a black- owned bank. We were accused of undermining black excellence and protecting the interests of white capitalists.[14] You know the saying that patriotism is the last refuge of scoundrels? Well, in South Africa, if you really need somewhere to hide, it’s not in patriotism but in race politics.[15] And these guys really needed a place to hide. They had just perpetrated a big bank robbery. Most troubling of all, they did it by stealing money sent by government for service delivery to some of South Africa’s poorest people – and that municipal money was lost. This did not stop people saying we should save the bank because it served the poor, even though it was an insolvent, corrupted institution.

The third problem was the strangest. In 2011 the Public Protector, South Africa’s ombudsman, decided to investigate a bank bailout the SARB did in 1985, for an institution called Bankorp, later taken over by ABSA, one of our big banks. There was talk that the bank had received money improperly, and it was argued that ABSA owed the government a refund. When this was investigated in the 1990s nothing came of it; the allegations were found to be baseless. Nonetheless, the Public Protector’s office decided to reopen the matter, and in June 2017, the Public Protector issued a finding that ABSA had to pay back R1 billion (roughly US$70 million).[16] Then, to our surprise, the finding included an order that parliament change the constitution, so that the Reserve Bank’s mandate stopped being to ‘protect the value of the currency in the interest of balanced and sustainable growth’. Instead, it should have a much more opaque duty to look after the socio-economic well-being of South Africans.

You may be wondering how an ombuds office goes from investigating a thirty-year-old bank bailout to changing the Constitution. We all wondered. Ultimately the courts threw out both the findings.[17] Yet, although we were all surprised by the attack coming like that, we had not anticipated such a flagrant disregard of the law.

This brings me back to the point I raised at the start of my speech. When I reflect on what the Reserve Bank was doing during this period, I cannot say it was all about maintaining a credible commitment to sound monetary policy. The problem we were really addressing was the principal-agent problem. The people of South Africa were relying on their government to look after their interests, while some people were instead using public power to pilfer. The SARB made that more difficult, which is why the Bank was attacked.

The terrifying thing we saw was how easy it is to flip between equilibria. There is one equilibrium where the rule of law is upheld and corruption is not tolerated, and most people then do their jobs honestly. And then there’s another equilibrium where people use their power to enrich themselves, where there is impunity, and where everyone has the incentive to take what they can before there’s nothing left. Over the past 10 years we moved from the first equilibrium towards the second, slowly at first, much faster towards the end. Many good institutions were weakened.

It’s not a surprise to me that the institutions which survived best were the ones with independence, particularly the Reserve Bank and the judiciary. Independence is a powerful defence. Many times during my term we have reflected with gratitude on the foresight of our founding mothers and fathers, who saw what could happen in the future and gave us the constitutional tools to defend ourselves.

The writers of our Constitution were students of history. They reflected especially closely on the African experience, where people had suffered so many disappointments in the decades after independence, perhaps the greatest being how leaders looted and impoverished their countries instead of governing them in the public interest. When our constitution was adopted the president was Nelson Mandela, and it would have been easy to take good leadership for granted. But good laws last longer than good leaders, and it was very helpful, twenty years down the line, to have the constitutional protections, the checks and balances, the guarantees of independence.

This emphasis on how independent institutions can protect democracies from bad leadership implies a close connection between how we think about central banks and how we think about judiciaries. I appreciate that central banks have narrower tasks than judiciaries, and aren’t peer organs of government. They are not the fourth branch. But the comparison still has some utility, in two ways.

First, the example of judiciaries is relevant because they too need to honour democracy while exercising unelected power. They are also confronted with the challenge from elected officials, “I’m elected and you’re not, so shut up and listen to me.” In practice, judiciaries do not resolve this tension through a policy of maximum deference. Rather, they typically take direction from the rights and values embodied in their countries’ foundational laws, and they confront violations of these principles when they see them. They appreciate that there is more to government than having 50% plus 1 vote.

The lesson for central banks is, we can honour democracy without feeling obliged to define it in strictly majoritarian terms, in which only directly elected leaders have any legitimacy. Democracy ensures that government serves the interests of the people. It does not consist of elections only. One of the gravest threats facing any society is the ruler who is more powerful than anyone else and therefore cannot be stopped by anyone, even when he is acting against the interests of the principals, the people.

Not all leaders are bad - I have met many good leaders, and some who would have behaved better in easier circumstances. But a small portion of leaders are bad, and if you understand probability, then you know sooner or later most countries will get a bad leader. There is nothing undemocratic about buying some insurance against this eventuality, and independent central banks, like judiciaries, are useful parts of those insurance policies.

Second, like judiciaries, central banks tend to be relatively high- performing institutions. There’s a famous book about judiciaries called, ‘the least dangerous branch’.[18] The idea is that all parts of government can do terrible things, but the judiciary tends to do the least harm. I think central banks have a parallel claim. For a start, central banks generally maintain high standards of honesty and competence. Every now and then, when there’s a story about corruption in a central bank, people are shocked. By contrast, when last was anyone surprised by a corruption story in a government procurement agency or a legislative office?

Furthermore, central banks tend to be good at what they do, especially when their performances are compared with other parts of government. In emerging markets, inflation is at long-term lows. Many countries have moved beyond the problem of ‘original sin’, the inability of emerging markets to borrow long term in local currency, in large part because independent central banks mean the local money can now be trusted. Given how many financial crises have been caused by original sin, and how pervasive it was, this is important progress.

In advanced economies, central banks can make an even grander claim - to have been ‘the only game in town’ in the years after the global financial crisis. This was not a position they desired, and it is not optimal from a policy perspective. But aren’t you glad there was a game? Central bankers did as much as anyone to avert another Great Depression. We are now worried about the rise of populism and reinvigorated nationalisms, but imagine how much worse this would have been with a debt-deflation spiral and double-digit unemployment.

This record tells us the case for independent central banks is not just theoretical. We have tried these institutions in many countries, and the experiments have generally succeeded. If anyone looks at this history and concludes independent central banks should be abolished, well, we might ask, what parts of government deserve to survive?

Of course, it’s not enough to say we have a record of success and now we must be left alone. The principal-agent problem applies to central banks too – if there are no accountability mechanisms, what is to stop central bankers privileging their own interests over those of society? Part of the solution to this so far has been cultural, in the sense that there is a community of central bankers with high standards and strong norms. This isn’t a hard or legal guarantee, but as with many professions, it is important because it works even when people aren’t watching. Even better, however, is having nothing to hide. Modern central banks, unlike their forebears, actively embrace transparency and communication. We try to be clear about what we’re doing, why, and what evidence we rely upon for our decisions. This also promotes accountability, because it is easier to assess success or failure. These values – transparency, accountability – have served us well, and I doubt central bank independence would be tenable without them.

We will also need new ideas for tackling new challenges. To conclude, I’d like to address two broad problems, one to do with financial stability mandates, and one affecting the price stability mandate.

Regarding financial stability, I opened this lecture with a problem. Financial stability is probably an inescapable responsibility for central banks, but it is a dangerous environment for an independent agency to operate - more so than the monetary policy space. How best can we navigate it? I have two suggestions.

First, we usually frame the problem in this sphere as one of independent central banks taking decisions away from the duly elected authorities. But sometimes independence helps ensure decisions are made in full view of the public, by the appropriate authorities, instead of being shirked. Let me give you an example.

In South Africa, one of the worst examples of state capture has been state owned enterprises which pose a significant fiscal risk. State-Owned Enterprises (SOEs) are a financial stability issue and should not the Reserve Bank be doing a bailout?

Absolutely not. The challenge of dealing with too-big-to-fail SOEs, of combining cash injections with conditionality measures, needs to be dealt with by the elected authorities – as it has been in the latest budget from National Treasury. The SARB’s power to say no was greatly enhanced by its independence.

Second, sometimes central banks face a tough choice between - on the one hand - stretching their mandates and raising legitimacy questions, and - on the other hand - staying in their safe spaces, but then risking bad economic outcomes. For instance, with Lehman Brothers, I understand there were legal problems, but I wish the Fed had found a way to prevent its bankruptcy. Similarly, I’m glad it was possible to give other investment banks access to Fed funds, and to prevent AIG from collapsing.

Finding the right balance here is always going to be an art. Fortunately, we have some master-class performances to draw on for lessons in how it’s done.

When ECB had to manage the euro crisis, it did with just a few words – doing whatever it takes. But the Council followed them with a major effort to convince political leaders, who ultimately backed the plan in public. They also worked to address the concerns of critics, by devising conditionality measures for bond purchases. In policy, there is always trade-off between how decisively you act and how widely you consult. In a crisis, the trade-off becomes very difficult indeed, but it doesn’t have to be impossible. I think a useful term here is one we used a lot in South Africa during the transition to democracy: ‘sufficient consensus’. It meant, given the importance of moving forward, you can’t permit filibustering, but you also can’t make decisions unilaterally, and ride roughshod over opponents. The ECB team did a masterful job of achieving sufficient consensus, and I think that’s what central banks should aim for in future financial crises.

Finally, on monetary stability, for once I am concerned about too little inflation. In emerging markets we still mostly worry about inflation being high, but some advanced economies keep on missing their targets from below. Frankly, it doesn’t seem to me so terrible to have price stability.

That was the original goal with 2% inflation targets - the idea was to get an inflation rate households and firms wouldn’t notice. The problem with winning the war on inflation, however, is that the only price movements left tend to be temporary, from supply shocks. This makes it easy to forget the responsibility of monetary policy for inflation. Indeed, the so-called Modern Monetary Theorists even think you can do away with independent central banks and tackle inflation with other tools. Well, if you build a maximum security jail and there are no escapes, does that prove you didn’t need to build a jail? I think it proves you built it right.

The same goes for inflation. Price stability is proof of the success of independent central banks, not an argument for their abolition. But I worry there’s a new generation coming of age in the advanced economies who have never experienced inflation, and won’t appreciate that subtlety.

Ladies and gentlemen, in conclusion, central banks may be under attack, but they deserve to be defended. They have been principled agents, serving the interests of their citizens, mostly with more effectiveness than people usually get from their governments. There are any number of institutions needing reform in this world. It would be best if the energy of our critics, and our defenders, could be directed to those other causes instead. Meanwhile, for central bankers, we should always bear in mind that independence is not a birth-right, but something we need to get up every day and earn. Our insulation from day-to-day politics creates a special responsibility to do our jobs right, and to be transparent and accountable so that our principals, the citizens, can see we are living up to this privilege.

Thank you.


[1] Debelle, G. (1996) “Central bank independence: a free lunch?” IMF Working Paper 96/1; Alesina, A., & Summers, L. (1993). Central Bank Independence and Macroeconomic Performance: Some Comparative Evidence. Journal of Money, Credit and Banking 25 (2), 151-62.


[3] This account draws on the memoir of the central bank governor for the period 2012-2014. See Panicos Demetriades (2017) Diary of the Euro Crisis in Cyprus

[4] The initial bail-in arrangements included insured deposits, but these were exempted when it became clear breaking the insurance guarantee would be a major error, not least because of the spillover risks. The deposit guarantee threshold was €100,000.

[5] For details, see Demetriades, Ibid., especially chapter 15, ‘Toxic Fallout’

[6] “A blow against independence” The Economist, 10 March 2014. Available at:

[7] This discussion draws on Paul Tucker (2018) Unelected Power, Princeton, New Jersey: Princeton University Press

[8] “The bank that failed” The Economist, 20 September 2007. Available at:

[1] Kevin Granville (2017, June) “A president at war with his Fed chief, 5 decades before Trump” New York Times, Available at: war-with-his-fed-chief-5-decades-before-trump.html

[2] Burton A Abrams (2006) “How Richard Nixon pressured Arthur Burns: Evidence from the Nixon tapes” Journal of Economic Perspectives (20:4) Available at:

[9] For an overview, see Norimitsu Onishi & Selam Gebrekidan (2018, December 22) “In Gupta Brother’s Rise and Fall, the Tale of a Sullied A.N.C.” The New York Times. Available at:

[10] The banks themselves were also called in by senior government officials and accused of being anti-transformation, of perpetuating what was called ‘white monopoly capital’. Tehillah Niselow (2018, September 24) “SARB warns of ‘catastrophic’ impact if banks kept Gupta accounts open” Available at: impact-if-banks-kept-gupta-accounts-open-20180923; “Editorial: Banks’ Gupta testimony reveals ANC’s financial illiteracy” Available at: 09-21-editorial-banks-shocking-gupta-testimony/ (21 September 2018)

[11] “Zwane wanted SARB’s wings clipped after Gupta blacklisting” Available at: /news-and-analysis/zwane-wanted-sarb-role-changed-after-gupta- blackli, 31 August 2016

[12] See Jeff Wicks (2018, October 14) “Holidays, flashy cars & ‘mansions’: Lifestyle of the VBS ‘heist’ rich and infamous” Available at: lifestyles-of-the-vbs-heist-rich-and-infamous/

[13] These details are drawn from Terry Motau “The Great Bank Heist: Investigator’s Report to the Prudential Authority” Available at:

[14] “Editorial: VBS – a tale of vast risk not vile racism” not-vile-racism/ (15 March 2018)

[15] Ivo Vegter (2018, 16 October) “VBS defenders expose the intellectual poverty of the race card” Available at: intellectual-poverty-of-the-race-card/

[16] “ABSA, public protector in court over apartheid-era bailout” africa/absa-protector-in-court-over-apartheid-era-bailout (5 December 2017)

[17] Lameez Omarjee (15 August 2017) “SA Reserve Bank wins court bid against Public Protector” Available at: protector

[18] Alexander Bickel (1962) The Least Dangerous Branch, New Haven, Connecticut: Yale University Press

[19] Mohamed El-Erian (2016) The Only Game in Town, New York, NY: Random House