NSW shocks by pulling $20bn rabbit out of its hat

In the AFR I write that NSW premier Dominic Perrottet and new Treasurer Matt Kean have presided over one of the most extraordinary budget turnarounds in the history of the Commonwealth.

As a result of the impact of the second COVID-19 lockdown, the NSW budget deficit is forecast to blow-out by almost $11 billion this financial year. Back in May 2021—before the Delta-variant of the virus shuttered the state—NSW announced it would issue a surprisingly chunky $35.5 billion of debt this financial year.

CBA’s analysts projected that the extreme cost of the subsequent lockdown would force this up to an astonishing $45 billion to $50 billion, or almost half what the Commonwealth government was raising for its own ever-expanding needs.

With the $11 billion fiscal deterioration revealed by Kean on Thursday, CBA’s analysts should have been vindicated: all else being equal, NSW should have announced that debt issuance this year was going to climb to $46.5 billion.

Instead, NSW shocked bond markets with the news that its debt program would be slashed by $8.1 billion to just $27.4 billion. That is about one-quarter less debt than NSW expected to issue before the second COVID-19 lockdown hit, or $20 billion less than expectations after the lockdown.

To put this in context, Victoria boosted its forecast debt issuance by $8.1 billion this year as a result of its own shorter lockdown, although those canny Mexicans had already this extra debt in the first half of the year.

So how has Kean pulled this rabbit out of his hat? Much of the credit must go to Perrottet, who in 2018 created a unique Debt Retirement Fund that was seeded with budget surpluses and asset sales, such as the WestConnex toll-road. Some time later Perrottet also committed all NSW royalties and State-owned corporation dividends to the Debt Retirement Fund, which further boosted its balances.

When the second-half of WestConnex was sold in September 2021, the total value of the Debt Retirement Fund exploded to an amazing $27 billion.

This is where the story gets more complex. When Perrottet legislated the Debt Retirement Fund in 2018, its sole purpose was to reduce the gross debt of NSW, maintain its AAA credit rating, and cut the cost of taxpayer borrowing. At that time, the budget was running a massive $4 billion surplus and the total value of NSW debt outstanding was only $35 billion.

Care of the recession induced by COVID-19 in 2020, the second lockdown in 2021, and an enormous $110 billion infrastructure program, NSW’s debt will have sky-rocketed to almost $120 billion this year.

This was precisely the shock that the Debt Retirement Fund was presciently designed to alleviate. It was intended to be a counter-cyclical fiscal policy buffer that the State would draw-down on when the budget lurched into large deficits. Contrary to some claims, it was not just a normal sovereign wealth fund that would grow forever: it was legislated as a debt repayment vehicle that could only be used to reduce, rather than increase, the liabilities of NSW taxpayers.

Yet over time, it was hijacked for other purposes. NSW’s investment arm, TCorp, is paid by NSW to manage the Debt Retirement Fund. The more money in the fund, the more fees NSW pays TCorp for both outsourcing the capital out to other fund managers, and to run some fixed-income strategies via TCorp’s internal team.

TCorp is able to pay its 180 staff remarkably well: according to its latest annual report, it spent $58.3 million on salaries and wages in 2021, or a mind-bogglingly high $324,000 per person! That is almost double the $177,000 per person the Reserve Bank of Australia spends on its staff, or one-third more than the Commonwealth Shadow Treasurer, Jim Chalmers earns. In contrast, Victoria spends $174,000 per person (half TCorp’s costs) at its debt issuance agency.

When the NSW’s budget swung into record deficit, billions of dollars of taxpayers’ royalties and dividends continued to be diverted to the Debt Retirement Fund despite the inability of the state’s revenues to cover its expenses.

As Standard & Poor’s, CBA and the AFR’s John Kehoe observed, this meant NSW was suddenly debt-funding an investment vehicle that was committing the vast majority of its money to high-risk asset-classes, dominated by equities, private equity, and junk debt, in what had become a huge levered “carry trade”. The Debt Retirement Fund was explicitly increasing NSW’s gross debt contrary to the intent of its legislation.

The gamble was that the returns from stocks exceed the cost of debt. The risk is that interest rates rise sharply and then equities and related asset-classes slump in which event taxpayers could lose tens of billions of dollars at the worst possible time.

Using the fund in this way meant, as S&P noted, that it was actually a pro-cyclical amplifier of fiscal risk: it would improve NSW’s finances during the good times, but make them worst during the bad times.

We lend billions of dollars to every major State government, including NSW, and formed the view was that there was a conflict of interest between the folks running the fund, who wanted to aggressively grow its assets under management via de-facto debt funding, and NSW taxpayers, who wanted to draw-down on the fund for its legislated purposes of retiring public debt after it had exploded.

As a lender to the State, we considered this to be a serious Environmental, Social and Governance (ESG) concern. Labor’s forensic Shadow Treasurer, Daniel Mookhey, who had led the debates over the iCare and TAHE scandals, harboured similar concerns.

The public debate that followed worked. In September Perrottet and Kean made the historically unprecedented decision to take $11 billion from the Debt Retirement Fund to pre-emptively repay taxpayer debt. It was certainly an Australian, and possibly global, first. Weeks later they announced that they would also stop diverting debt-fuelled NSW taxpayer income to the fund, saving up to $10 billion more.

This is why instead of issuing $46.5 billion of debt this year, NSW has been able to radically reduce it to just $27.4 billion. It is a remarkable achievement that Perrottet and Kean should be commended for delivering.

The latest budget says NSW will not divert royalties and dividends back to the fund until the budget returns to an “operating cash surplus” in 2023. Yet because of its huge infrastructure program, NSW’s budget will still be in an overall cash (as opposed to an “operating”) deficit in 2023.

Taxpayers should never be debt-funding equity investments to run a levered carry trade that increases fiscal risk in downturns. Kean and Perrottet should only be diverting taxpayer money to fund managers when this money is unequivocally surplus to the state’s needs. That is, when the budget is truly in an overall cash surplus after accounting for the infrastructure spending.

There is further a compelling case that the $15 billion to $17 billion left in the fund should be used to ensure NSW does not unnecessarily increase taxpayer debt to pay for its infrastructure commitments. Perrottet has previously promised to use this money for new infrastructure investments. With record levels of debt, it is a no-brainer. And if Perrottet and Kean do not see the light, I suspect their increasingly impressive opposition will.

The other big news during the week was RBA governor Phil Lowe’s speech outlining the wind-down of the central bank’s bond-buying program. While Lowe explained this could continue until May 2022, it looks highly likely the purchases will end in mid February, or shortly thereafter.

This would be ideal because this very successful program has done the job it was designed to accomplish, which is putting downward pressure on the Aussie dollar and ensuring our interest rates are not unnecessarily high vis-à-vis the rest of the world.

It is increasingly clear that further RBA bond purchases could be counter-productive as it now owns 33% of the entire Commonwealth government bond market, and more in the specific lines it has been buying.

This week the State governments have downgraded their debt issuance over the next six months by $6 billion to $8 billion (with more to come as budgets improve). Importantly, this is materially more than the $5 billion of extra buying of State bonds the RBA would do if it continued until May. In recent auctions, the RBA has had to reach to find State bonds, hinting that the program is becoming redundant. And with our analysis suggesting that the banking system will need to buy about $408 billion of government bonds over the next few years, much more than the RBA has done, there will be no shortage of demand.