Since the Reserve Bank embarked on quantitative easing last March, it has been buying government bonds almost as fast as the Treasury has been issuing them.
As the bank is part of the core Crown, does that mean that net government debt has been growing a lot more slowly than it would if those bonds were held by the private sector?
Alas, no. It's complicated.
In last week's monetary policy statement, the bank reported that over the past five months the Treasury issued $29 billion of government bonds and the bank bought $23b, equivalent to nearly 80 per cent of the new issuance. It now holds about 22 per cent of the government stock on issue.
Government debt levels are a lot higher than expected at the start of the year, of course. Then the Treasury forecast the stock of government bonds on issue would be $74b by the end of the fiscal year in June. In the Covid-stricken event it was $97b.
And counting. Another $1.9b of bond issuance is scheduled this month, while since mid-June the Reserve Bank has been buying almost $1b of bonds a week. In the words of Senator Everett Dirksen, "A billion here, a billion there. Pretty soon you are talking real money."
The deficit is swollen not only by spending specifically earmarked for pandemic relief like the wage subsidy, but also by the "automatic stabilisers" — jargon for the effects of recession on tax revenues, on one side, and welfare payments, on the other.
As for how big these effects are expected to be, the resurgence of the virus means we will have to wait another month for updated pre-election forecasts.
Meanwhile, the Reserve Bank last week increased its self-imposed cap on how much bond-buying it is prepared to do on the secondary market to $100b, from $60b previously. It also said it is prepared to buy up to 60 per cent of the bonds on issue, previously 50 per cent.
Clearly, if the Reserve Bank is buying bonds at 80 per cent of the pace the Treasury is issuing them, it will eventually hit that 60 per cent limit, beyond which it reckons the liquidity and efficient functioning of the bond market would be impaired.
With nominal gross domestic product sitting around $300b if we are lucky, those caps of $100b and 60 per cent are consistent with a commitment the Treasury made back when the net debt-to-GDP ratio was falling, not to let the free float of bonds available to the market fall below 20 per cent of GDP for liquidity reasons. The arithmetic indicates that it would be unwise to assume any further increase from here.
Hence the need for the central bank and the trading banks to prepare for an inferior form of alternative monetary policy — taking the official cash rate negative. It would be accompanied by a funding-for-lending programme to mitigate the credit-crunching effects of doing that as deposit interest rates and banks' margins are squeezed.
In the meantime, though, most of the Government's fiscal deficit is being monetised, as opposed to incurring obligations on behalf of future taxpayers to private sector holders of its bonds.
Because the Reserve Bank is part of the core Crown, when the net debt position is calculated, government bonds the bank has bought are disregarded, or "eliminated on consolidation" in the jargon.
But it does not end there. The bank has bought those bonds with brand new money it has created for the purpose. That expands the level of settlement cash in the financial system, which is the sum of the balances trading banks hold on deposit at the Reserve Bank and use to square accounts with each other on behalf of their clients. Settlement cash can only be used for that purpose.
The Reserve Bank pays interest on those deposits, at the official cash rate. So settlement cash is a liability for the central bank and a corresponding asset for the banks. With the OCR at 0.25 per cent, and likely to go negative next year, it is not a very onerous liability for the Reserve Bank, and it does not do the banks' return on assets any favours either.
But settlement cash is a liability of the core Crown and counts towards net debt. It currently sits at $23b.
The net effect of all this is that when the Reserve Bank buys bonds under its QE programme, the consolidated Crown exchanges one liability (a government bond with a specified interest rate and term) for another liability with no set term and a floating interest rate (the OCR).
The benefit, for the foreseeable future, is lower interest costs. The OCR would normally be lower than the coupon interest rate payable on the bonds, and the Reserve Bank gets to set it.
Not only that. The objective of quantitative easing from the Reserve Bank's point of view is to lower longer-term interest rates, not just for the Government but for borrowers generally.
It is confident this has been working. It reckons yields on government bonds are at least 50 basis points (half a percentage point) and potentially more than 100 basis points lower than they would have been without the quantitative easing programme.
"Floating mortgage rates have declined by 75bps since the start of the year, in line with the reduction in the OCR, while longer-term mortgage rates have fallen by more — driven by the larger falls in longer-term funding costs," it said, adding that at least half of all mortgages are due to be repriced in the next 12 months.
We are still in the early, phoney-war phase of this recession. The prospects of the Reserve Bank raising interest rates because inflation is threatening the 3 per cent top of its target band and the labour market has mopped up all the spare workforce are remote indeed.
The risks are all to the downside.
The differences between funding the deficit by incurring obligations on behalf of future taxpayers to KiwiSaver providers and other private sector custodians of people's savings, on the one hand, and funding it through an enlarged obligation on the central bank to pay the OCR on banks' deposit accounts with it, on the other, favour the latter.
It is not a free lunch, but is a thrifty packed lunch.
So far the bank has used only a quarter of its QE budget, but there are limits to how long it can continue.
While it does, the Reserve Bank is able to play Robin to the Government's Batman in doing battle with the villainous Covid-19 and its economic impact.
That is not, as some commentators would have it, a cop-out or abdication of the bank's responsibilities for stabilising the cycle.
It just reflects the fact that, as governor Adrian Orr reminded us last week, the Reserve Bank cannot compel people to borrow and spend or banks to lend. It can only, through its influence on market interest rates, encourage or discourage those things.
And right now it is not the cost of credit which is holding back businesses from investing and hiring.
The Government, by contrast, can decide for itself — with Parliament's approval — how much it wants to spend and on what. Which is why right now fiscal policy has to do the heavy lifting.