Decision making in a time of uncertainty
11 June 2020
INTRODUCTION: RISK AND UNCERTAINTY
We do not know now how the COVID-19 epidemic will play itself out internationally or nationally. We do not know how much of the global or our own economy will be left standing in the coming months and years.
Nor do we know the impact of the epidemic on the capacity of states around the world and the capacity of the South African state, in particular. Every evaluation of policy is subject to re-evaluation, sometimes in surprising ways. For instance, it now appears that Donald Trump was nearer to being right than Andrew Cuomo about the need for ventilators.
An important first move in dealing with a situation like this is to recognize the distinction between risk and uncertainty. Risk arises when there is a determinate list of possible outcomes, and a determinate probability of realization of each of them. Uncertainty exists when these probabilities cannot be determined, or when it is not possible even to establish a list of possible outcomes. Optimizing choices are easiest when there is certainty of outcome.
They are more complex in risky situations, but at least there are possible approaches to the problem. But as Kay and King point out, optimization is simply impossible in situations of uncertainty. Instead, the emphasis must shift towards coping, and successful coping requires judgment and coming to grips with the question: what is going on here? And, given that COVID-19 has created uncertainty rather than risk, thinking about how we cope and deepening our understanding of what is going on ought to be our priorities.
POLITICAL EMERGENCIES AND DICTATORSHIP
All politicians aspire to be dictators, and ours displayed this ambition by creating a National Coronavirus Command Council, whose proceedings are secret. Since the proclamation of the state of disaster, members of the Cabinet have reveled in new power in ways corresponding to their temperaments, from thuggery to dour puritanism and bantam cockiness.
On this view, the quality of democracy depends on the ability of society to frustrate dictatorial ambition. And, indeed, both political and legal pushback in South Africa is well under way. And there is a second constraint. One of the reasons why Italian fascism was less lethal than German Nazism was that the Italian state was less efficient than the German one. So one facet of what is going on, and what is not, is the ability of the state to deliver on command. Twenty-six years in power have not cured the ruling party from over-estimating it.
A NEW START OR GREATER CONFUSION?
Athwart dictatorship lies an ancient trope. COVID-19 is purgatory and beyond purgatory, as any Catholic will tell you, lies a new heaven, a new earth. It is remarkable how widespread the anticipation of a new beginning is, even as opinions differ about what it will be the beginning of. The anticipations are a successor to the 1990s craze for scenario construction, the attempt to shape and disseminate a narrative supporting a preferred course of action, a manoeuvre in a war of position.
They attempt to see a point which is not visible, and they skate over the hard choices about the route to be taken, and its rigours. A weary, seriously ill Roosevelt could see, at the end, the costs that the war would impose after it had ended. He could not have seen it at the beginning. Indeed, serious discussion about the dispensation after the end of the second world war was only possible once it was clear that Germany and its allies would lose it. But we have just reached Dunkirk, with a retreat from a hard lockdown, while infections continue to rise. The Blitz is yet to come.
Pathologies can be expected as interests seek to advance their positions in the fog of war, and one is developing right now. It is the disconnect between (a) the pressures on the Minister of Finance, as he prepares to present a national adjustments budget to parliament on 24 June in the light of conditions substantially worse than assumed in the February budget, and (b) the 22 May presentation by Enoch Godongwana on economic reconstruction, which presses for a state bank, a sovereign wealth fund, a state pharmaceutical company, a liquefied natural gas hub, a repurposed Central Energy Fund, subsidies for the taxis, a large infrastructure programme, and a very large expansion of the Reserve Bank’s balance sheet to provide finance to development finance institutions at ‘developmental interest rates’. How these rates are to be achieved is not explained, given the upward pressure on long rates exerted by the increasing risk associated with a rising debt to GDP ratio.
The current state of affairs creates the impression that members of the ruling party are talking past one another. This is not a situation which can endure for long. Unavoidable decisions will crystallize latent conflict and require its resolution, probably in stages. Attempts to fudge issues will occur and may succeed for a while.
But fudges will not be stable. Neither will any version of the fiscal framework last for long while the economic effects of the COVID-19 epidemic remain uncertain. Indeed, uncertainties resulting from the February budget persist.
The extent to which the February target for the reduction of the public sector wage bill will be met may not be resolved by the time that the national adjustment budget is introduced on 24 June. It has been reported that arbitration hearings would start by mid-June. An arbitrator will issue an award after the hearings are complete, with the matter potentially heading to court or resulting in a strike if the unions aren’t happy.
More recently, it has been reported that unions representing public servants have approached the labour court seeking an order declaring the government to be in breach of a resolution the parties took on salary adjustments. Moreover, the adjustment budget will not have complete second quarter data as a basis for projecting economic and fiscal conditions. We may well have a second national adjustments budget accompanying the October Medium Term Budget Policy Statement.
DO WE NEED DEBT MONETIZATION NOW?
It is possible that the considerable pressure on the National Treasury and the Reserve Bank may escalate to full-on war, which they are not guaranteed to win. In the battle, controversies about economic theory are emerging, in particular the contention between mainstream economics and modern monetary theory. As Paul Krugman has pointed out modern monetary theory is a shape shifter, but its main message, in Willem Buiter’s graphic summary is “Deficit, schmeficit. Just boost public spending or cut taxes, then monetize the resulting imbalance.”
Monetization of debt occurs when the government sells bonds directly to the Reserve Bank, which creates new bank deposits and uses them to pay for the bonds. This process creates new money and expands the money supply. Section 13(f) of the Reserve Bank Act (90 of 1989) places a limit on the extent to which the Bank can monetize debt, but the constraint is not binding at present. Monetization transfers resources to the government.
It may stimulate the economy without an impact on inflation when the economy is caught in a liquidity trap. In this case, the government mobilizes idle resources. But South Africa is not currently in a liquidity trap. In this case, monetary policy remains effective and debt monetization on top of it will simply add to inflation. And inflation amounts to a regressive tax since the poor are the least able to protect themselves from it. Getting it right matters.
The Reserve Bank can influence the yield curve by operations in the secondary market, which are not restricted at all. It has recently done so, to cope with an episode of illiquidity in the bond market in March 2020, with an upward spike on yields.
This has resulted in holdings of R 8 billion in government securities on 28 February 2020 rising to R 31 billion at the end of May. The Reserve Bank has stated that stabilization is its sole aim. In that it seems to have been successful. On 4 June 2020, the yield curve beyond five years is little higher than it was six months earlier and well below its spike in late March. And the short maturity end of the yield curve is lower than six months ago.
But there is no warrant for a generally carefree approach to deficit spending. Think Germany in 1923, or contemporary Venezuela and Zimbabwe.
Return to coping and figuring out what is going on as strategies in times of uncertainty. Coping at the height of the 2008 financial crisis and its long aftermath meant that central banks had to do things they would have not dreamt of five years before the onset of the crisis. Circumstances have changed again and coping may require new actions.
On being accused of inconstancy, Keynes retorted that when the facts changed, he changed his mind. It may be that revisions to the Reserve Bank model (which is continuously updated in the light of new data) down the line indicate that some monetization of debt at some point is the best option.
That possibility cannot be ruled out in uncertain times. But it is counter-indicated now. It should also be noted that continuing (rather than once-off) monetization is needed to sustain the menu indicated in the Godongwana presentation. That would risk not only a single burst of inflation, but a progression to hyperinflation.
The greatest contributions to coping will be made by those who have the keenest insights and are able to maintain and propagate them in a noisy environment.
Charles Simkins, Head of Research, Helen Suzman Foundation.
 Those appalled by this observation may take comfort in the fact that this evaluation is subject to later re-evaluation and also from T S Eliot’s couplet in Murder in the Cathedral: The last temptation is the greatest treason/ to do the right thing for the wrong reason.
 John Jay and Mervyn King, Radical Uncertainty, The Bridge Street Press, 2020
 Bruce Bueno de Mesquita and Alistair Smith, The dictator’s handbook: why bad behaviour is almost always good politics, Public Affairs, New York 2011
 For an analysis of why long term interest rates are at historical highs relative to those in the United States and Germany, listen to the podcast on a study by Johannes Fedderke at https://www.econrsa.org/podcasts
 Reuters, South African public-sector wage dispute moves to arbitration, Daily Maverick, 25 May 2020.
 Ray Mahlaka, Public servants’ unions go to labour court over salary adjustments, Daily Maverick, 7 June 2020
 Modern monetary theory is, in fact, less modern than its name suggests. A version of it was considered and rejected by Keynes.
 Krugman’s exact words are “every time you think you’ve pinned them down on some proposition, they insist that you haven’t grasped their meaning”.Paul Krugman, Running on MMT, New York Times, 25 February 2019
 Willem H Buiter, The problem with MMT, Project Syndicate, 4 May 2020
 Specifically the Bank may not hold in stocks of the Government of the Republic which have been acquired directly from the Treasury by subscription to new issues, the conversion of existing issues or otherwise, a sum exceeding its paid-up capital and reserve fund plus one-third of its liabilities to the public in the Republic. This is a high ceiling. The share capital of the bank is R 2 million, and on 30 April the reserve fund was R 418 million, but one-third of notes and coins in circulation was R 53 billion and it may be that other liabilities can be regarded as liabilities to the public.
 A liquidity trap arises when the nominal interest rate is zero and increases in money supply are hoarded rather than spent, because the opportunity cost of holding cash is zero.
 The Reserve Bank’s model projects inflation returning to the midpoint of the current target range in 2021 and 2022, assuming no more than two further 25 basis point rate cuts in the course of 2020.
 The yield curve is a plot of yield to maturity against time to maturity of government bonds
 This contrasts with a holding of R 20 billion in government securities (up from R 8 billion on 28 February 2020 – there has been a further increase in holdings to R 31 billion at the end of May) but much of that (and all of the recent increase) consists of purchases in the secondary market, which are not restricted at all. Andrew Donaldson has called for the Reserve Bank to purchase government bonds worth R10 billion to R20 billion a week, in order to reduce the cost of government debt, stabilize the market and stimulate the economy more adequately. This would be a policy of financial repression, and would be inflationary if prolonged. By contrast, the Reserve Bank has stated that stabilization is its sole aim. In that it seems to have been successful in the past couple of months. On 4 June 2020, the yield curve beyond five years is little higher than it was six months earlier and well below its spike in late March. And the short maturity end of the yield curve is lower than six months ago.
 It is the singular merit of Zachary Carter’s book, The price of peace: money, democracy and the life of John Maynard Keynes (Random House, 2020) to encourage us to approach his thought as a continuous grappling with changing circumstances rather than a timeless systematic exposition.
There are wilder views than Godongwana’s in the political market place. The EFF is reported as opposing easing the lockdown, with Malema saying: “We should do like we did during apartheid - we must refuse to comply and stay at home. We must engage in a stay away. If this white economy collapses, let it collapse. If we are going to die of hunger, let us rather die with our boots on than dying protecting the white monopoly economy. It is not our economy. If death comes, let it come - but we must die proud that we defied protecting the white economy.” (ZingisaMvumvu, Defy the eased lockdown – even if it collapses ‘white economy’: Malema, Times Live, 28 May 2020). Apartheid tried to separate the economy into a white economy and a black economy with the relationship between them taking the form of the economic relationship between countries. The attempt failed. It couldn’t be done.