Is the SA Reserve Bank doing Quantitative Easing?
Share this article:
JOHANNESBURG - The SA Reserve Bank (SARB) officials are in denial. They wrongly insist that South Africa has to hit a 0 percent nominal short-term interest rate before embarking on Quantitative Easing (QE).
This insistence has been severely criticised, correctly, by analysts such as Redge Nkosi, as having no basis in QE theory and evidence.
The time for QE is long overdue and the Covid-19 has added more urgency for the SARB and National Treasury to devise unconventional ways to fund the war against the virus, and to stem the on-going slide into a deep recession.
On March 19, the Monetary Policy Committee (MPC) of the SARB announced a one percentage point reduction in the repo rate. When Andrea Masia of Morgan Stanley correctly asked about the extreme movements in the bond markets, the SARB officials could only say that they have a range of instruments in their tool kit, and that they have not seen any stress in the financial system.
However, there was a liquidity problem in the financial system. The long rate rose by 300 basis points 17 days in the run-up to the MPC statement, more than R57 billion worth of bonds were dumped, presumably by non-residents, because the currency depreciated markedly at the same time, by more than R2.50. This should have concerned the MPC.
In response to the MPC statement, I said the 100 basis point cut will not be effective and that a QE-type intervention would have been appropriate. Covid-19 would lead to major deposit withdrawals from banks and default rates would sky-rocket as a result of cash-flow problems brought about by the lockdown and layoffs.
Judging by the SARB’s MPC statement of 19 March 2020, these concerns were not taken into account. Between 10 - 15 March, there were already reports that some banks were experiencing mounting non-performing loans across their product offerings since the beginning of 2020.
Non-performing loans rose by between 24percent and 47percent, other product impairments rose by as much as 84percent, in three months. This means that, absent Covid-19, there are serious problems in South Africa credit markets.
On the March 25, the SARB appeared to have woken up to the problem. It issued another statement. The SARB announced that it will refinance banks at the repo plus 0.3 percentage points.
The SARB went further, announcing a commencement of a programme to purchase government securities across the yield curve. This is a QE-type intervention, which was supposed to be announced in the MPC statement.
It is similar to QE2 in the US, which entailed the Fed purchase of only long-term government bonds, and it occurs with the repo rate at 5.25percent, not at the 0percent that the SARB keeps on telling us about.
Seeing this contradiction between what they say and what they do, on March 27 the SARB issued the third statement, this time disputing that it is commencing with QE. It claims that the proposed intervention seeks to better match buyers and sellers in order to reduce the effects of risk aversion. This is a form of QE.
Ben Bernanke, the former US Federal Reserve chairman known to be the architect of QE, defined QE as an asset purchase programme that is aimed at injecting liquidity in the financial system by the central bank buying illiquid assets.
It is a form of open market operation. However in the QE case, the target is largely to reduce the risk premium on longer term borrowing across the entire yield curve, as has just happened in South Africa. There is now at least an extra 2.5 percentage points that remains in the long term interest rate, which the SARB has to shave off by buying bonds, perhaps in excess of R52 billion.
In the March 27 statement, the SARB denies that it has begun QE.
It says, “Quantitative Easing is generally applied where interest rates are zero or close to zero and inflation is far below the central bank’s target or even threatening to turn negative”. The SARB then says that it “does not have interest rates at or close to zero”.
This is wrong because an emerging market does not have to hit 0percent nominal interest rates to do QE. To say so is to overlook the sovereign risk premium.
In an article I wrote on June, 12, 2019 in the Business Report, I argued that the SARB officials were wrong to say that an emerging market has to hit a 0percent interest rate in order for QE to be implemented.
I further argued that even the inflation rate of an emerging market does not have to be near 0percent for the central bank to do QE. The theoretical argument is as follows: suppose an advanced economy whose sovereign risk is zero, say the US, has hit a 0percent short term nominal interest rate. In order to stabilize the value of its currency, the emerging market will have to lower its short term interest rate to just cover its sovereign risk premium.
The sovereign risk premium is not zero in emerging markets, it is the lower bound for its short term interest rate. This is the point that the SARB officials did not get on June, 12, 2019, they still do not get it on March, 27, 2020.
When the inflation rate in the advanced economy hits its target, say of 2percent, the emerging market should hit its inflation target, say of 4.5percent. This is precisely where we are and the corresponding short term nominal interest rate is just covering the sovereign risk premium. The point in all of this is that a 0percent nominal interest rate in an advanced economy corresponds to some positive rate in an emerging market.
Therefore, the SARB officials are wrong to say we need to hit 0percent nominal interest rates before QE can be implemented.
If the SARB officials insist on going below the sovereign risk premium before they implement QE, they will have to impose exchange controls, otherwise the exchange rate would depreciate markedly. It also goes against the properties of the SARB’s own Quarterly Projection model. That model puts the nominal sovereign risk premium at 4.5percent.
Although this is under-estimated, it is far from 0percent. It is not correct to say there is still some room for the SARB to use the interest rate as an instrument of monetary policy.
Any short term interest cut now would produce a currency depreciation, absent exchange controls. Given the 300 basis point increase in the long-term interest rate, brought about
by the R52bn sell-off, the SARB will have to expand its balance sheet by at least 5percent.
It would also assist the SARB (and National Treasury) to take a step back and begin to methodically calibrate their actions within a coherent macroeconomic policy response to the crisis. QE is long overdue and it has begun. What is required is the design, size and the various forms it will take.
The SARB should also not be too quick to dismiss unconventional proposals, like money-financing of fiscal expansions at 0percent interest rate, the purchase and refinancing of private sector debt such as mortgages, and public sector debt such as that of state-owned enterprises. Otherwise, the SARB may find itself issuing one statement after another, while its actions become more and more inconsistent with what it says, all this leading to a loss of credibility.
Christopher Malikane is Associate Professor in the School of Economics and Finance at Wits University.