The COVID-19 pandemic continues to weigh on the global economy. Intermittent lockdowns and protracted economic uncertainty have elevated private-sector savings rates. In response, governments have stepped up public spending to record highs, leading to a rapid increase in sovereign debt. S&P Global Ratings projects median general government debt for all 135 sovereigns we rate will rise by end-2021 to 62.6% of GDP, from 49.0% at end-2019 and 32.5% at end-2008. For the Group of Seven nations (Canada, U.S., France, Germany, Italy, Japan, and U.K.), we project average government debt at end-2021 will increase to 128.7% of GDP versus 106.3% in 2019 and 101.3% in 2008 (see table 1A).

The trend for emerging markets is similar. For end-2021, we forecast median general government debt for the largest 60 emerging-market sovereigns we rate at 65.8% of GDP, or 15.5 percentage points (ppts) above 2019 levels and 30 ppts above end-2008 figures. At the end of this year, general government debt is set to exceed 100% of GDP in 13 emerging markets, more than double the 2019 number (see table 1B).

Can Governments Steady Public Finances Over The Next Several Years?

By 2023, two-thirds of rated sovereigns will manage to at least stabilize government debt to GDP, according to our forecasts. Meanwhile, the cost of financing debt continues to be effectively zero for developed sovereigns and near all-time lows for most of those in emerging markets. That still leaves roughly one-third of rated governments unlikely to stabilize public finances by 2023, absent stronger economic performance and better fiscal outcomes than we currently expect.

Low potential growth in developed markets, such as Japan and Italy, as well as in emerging markets like South Africa, Saudi Arabia, and Brazil will make it more difficult to mend public finances. Social expenditures may stay higher for longer as some economic sectors suffer permanent damage, or should new variants of the virus prove more resistant to vaccines. Governments will likely do a lot of soul searching about whom and how they tax. Treasuries in developed economies, for example, may decide to be more aggressive about taxing the stock of wealth (accumulated savings), rather than just the flow--that is, GDP; one of the unintended consequences of negative real policy rates in developed markets is asset inflation. We expect that, sooner or later, governments could try to tax the exuberance increasingly priced into financial and nonfinancial assets. This is already evident in recent proposals in the U.S., and in the U.K. to raise the capital gains tax.

Once herd immunity has been established, financial market pressures on governments to deliver fiscal and structural reforms may intensify as policy rates gradually start to normalize. In developed Europe, now that the U.K. has left the EU, euro area governments may take steps to fully integrate the single market for services and possibly even consider introducing a pan-European labor code, though political resistance at the national level will be fierce. The alternative, the status quo on fiscal and structural economic policies, is likely to yield disappointing growth outcomes, putting further pressure on sovereign creditworthiness.

S&P Global Ratings' Key Assumptions For Sovereign Debt

Tables 1A and 1B show our projected ratios for debt to GDP out to 2023 for all developed sovereigns and the 60 largest rated emerging sovereigns, based on our current forecasts. We have sorted sovereigns according to the projected increase in general government debt during 2023.

Governments at the top of the list are projected to be unable to put debt on a downward path during 2023. Governments toward the end of the list are forecast to adjust their budgets sufficiently by 2023 so as to put debt to GDP on a downward path during that year. This does not necessarily mean that the underperformers won't be able to stabilize debt over a longer time horizon, but rather that their fiscal policy choices, nominal growth prospects, or effective funding costs up to 2023 are likely to prevent them from doing so over the next two to three years. GDP outcomes are highly sensitive to the epidemiological situation. As an example of this, when the governments of Italy and the U.K. eased lockdowns during the third quarter of last year, quarterly GDP bounced an annualized 16% in both economies, and over twice that in the U.S.

Our projections embed three key assumptions:

• Major central banks will keep policy rates at (effectively) zero until the recovery is well entrenched and will use some combination of asset purchase quantities and yield curve control to keep financial conditions loose. As a consequence, for advanced economy governments, we expect average financing costs on total debt (though not the marginal cost on new debt) to continue to decline through 2023. That would keep debt-servicing costs at about an average 1.2% of GDP or 2.9% of government revenues in advanced economies, equivalent to just over 2% of GDP (5.1% of revenues) for G-7 members and 1.4% of GDP (3.2% of revenues) in the eurozone.
• Commencing in 2022 at the latest, on the back of widespread immunization, we expect pent-up demand to propel GDP higher, while global trade volumes recover. As a consequence, the most tax-rich components of growth, namely private consumption--including import-intensive private spending--should accelerate, benefiting public revenues.
• Starting next year, governments will phase out the extraordinary increases in public spending introduced in 2020, despite potential political resistance.

Risks to our projections include:

• Weaker real and nominal GDP due to a slower process of herd immunity or permanent structural damage to potential growth from the pandemic, or both.
• Premature monetary tightening by major central banks leading to an interest rate shock. None of the GIPSI sovereigns (Portugal, Ireland, Italy, Greece, and Spain) or Japan could weather a sustained rise in their funding costs. However, given increasing central bank holdings of their debt, a rate shock does not appear to us to be a realistic scenario, particularly in Japan, where inflation averaged 0.2% in 2020 and domestic residents hold the vast majority of public debt. Note that a temporary period of higher-than-expected inflation without a strong monetary response from central banks, which may disregard momentary price volatility, would lower government's ratios of debt to GDP relative to our projections.
• An inability of developed economy governments to cut back spending due to political resistance.
• Our government debt projections to end-2023 reflect limited contingent liabilities to public balance sheets from the ongoing distress that banks and nonfinancial corporations are experiencing. If the private sector requires bailouts, this would push up debt beyond our forecasts. In Europe, the average assumption of debt from contingent liabilities connected to the global financial crisis was 6%-7% of GDP. A similar cost from contingent liabilities this time around would lead to downward pressure on sovereign ratings.

Debt to GDP: Outperformers and underperformers

According to our projections (see tables 1A and 1B), nearly half (14/32) of developed sovereigns will manage to put their ratios of debt to GDP on a solid downward path by 2023, with 11 of these 14 located in the eurozone (see table 1). We project another 10 to either stabilize or slightly lower debt in 2023. For five developed sovereigns (U.S., Iceland, Finland, Czech Republic, and Israel), debt to GDP will increase slightly (by no more than 1 ppt of GDP) in 2023. And four developed sovereigns--New Zealand, Japan, Australia, and Slovakia--will see debt to GDP rise by over 1 ppt through 2023.

In emerging markets, 27 out of the 60 largest sovereigns are set to put debt to GDP on a solid downward path, according to our projections, and another 13 should either stabilize or slightly lower debt to GDP by 2023. That still leaves 20 emerging sovereigns which, according to our forecasts, won't manage to stabilize public finances over the next several years. Debt to GDP ratios in nine of those—including Brazil, Saudi Arabia, and South Africa--are forecast to continue to increase through 2023. We project that Brazilian debt to GDP will break 100% by 2023, and South Africa will nearly reach that level.

DSPB and i-g

As part of this exercise, we have calculated the simple debt stabilizing primary budgetary position (DSPB) for all rated developed sovereigns and for 60 of the largest rated emerging-market economies (see column E in tables 1A and 1B). The simple DSPB for 2023 is the budgetary position excluding interest that governments would need to operate to flatten the ratio of debt to GDP in that year, assuming no exchange rate movements and no statistical discrepancy between the fiscal balance and net financing. We have chosen 2023 because it is the furthest into the future that we currently forecast.

Since most developed economies finance themselves in their own currencies and exhibit low inflation, the DSPB is a useful estimate of where they need to get to put debt to GDP on a downward path. For emerging markets, particularly those that finance themselves in foreign currency and are subject to exchange rate volatility, the simple DSPB is a less reliable indicator but a still useful gauge for how much fiscal lifting they may need over the next three years.

As highlighted in rows A and B in tables 2A and 2B, the key determinant of future debt stability is i-g, the average cost of debt minus the growth of nominal GDP. Sovereigns where nominal GDP is growing in local currency terms at a rate identical to the average cost of debt can stabilize debt to GDP by operating a balanced primary budget (the budget balance net of interest payments on general government debt). Sovereigns where GDP is growing faster in nominal local currency terms than the average cost of debt can stabilize debt while running a primary budget deficit (the size of that debt stabilizing primary balance or DSPB is determined by the level of debt to GDP in the previous year, and the value of i-g). In contrast, sovereigns that pay more on their debt than nominal growth must operate primary surpluses to stabilize debt to GDP.

Given our expectation of a prolonged period of monetary accommodation across developed markets (except for Iceland and Israel), there is no developed sovereign for which we project i-g equal to or greater than zero. Among developed sovereigns, i-g is particularly favorable (that is, below negative 3) for nine eurozone governments (reflecting our expectation for very low average funding costs and a solid 2023 recovery), Norway, New Zealand, and Sweden (where we expect nominal GDP in 2023 to expand by 5.6%, 4.8%, and 4.6%--among the highest in the OECD). Among OECD members, i-g is the least favorable for Japan, Denmark, Israel, Italy, and Iceland, either due to weaker growth prospects (Japan and Italy) or higher funding costs (in the cases of Israel and Iceland, reflecting embedded subsidies for domestic pension funds in domestic rates).

The story is different in emerging markets, where i-g runs the gamut (see column A-B in table 2B). It is negative 15% in Ethiopia and negative 34% in Argentina (high inflation translates into high nominal GDP in both cases, but both largely finance themselves in foreign currency rendering the DSPB unreliable as an indicator of future debt dynamics). On the contrary, in seven cases, including Lebanon, Saudi Arabia, South Africa, Ecuador, Costa Rica, and Bahrain, we project nominal GDP to undershoot average interest costs in 2023, an early indicator of a potential debt trap.

Fiscal consolidation needed varies from 5 to over 10 ppts of GDP

To get to debt stabilizing primary balances (DSPBs) by 2023, advanced economy governments would need to improve their primary budgetary positions by 5.3 ppts of GDP between 2020 and 2023 (see column E-C in Table 1A); for G-7 sovereigns the average adjustment in the primary balance would have to be over twice that, at 10.6 ppts of GDP. Four developed sovereigns--Canada, U.K., Japan, and the U.S.--would need to consolidate their fiscal positions by over 10 ppts of GDP to stabilize debt. Similarly, eight emerging sovereigns would have to consolidate their fiscal position by over 8 ppts of GDP over the next three years to stabilize debt to GDP (see table 2): Oman, South Africa, Iraq, Hong Kong, Saudi Arabia, Brazil, Chile, and Bahrain.

These are large adjustments, considerably greater than the fiscal consolidation economies undertook between 2010 and 2013. Over that period, in developed economies, primary budgetary positions improved by 2.4 ppts of GDP on average and 3.2 ppts for the G-7. It was only the GIPSI members of the eurozone that, between 2010 and 2013, consolidated their budgets by higher amounts, that is, by an average 10.2 ppts of GDP at the price of almost half a decade of anemic growth and falling public and private investment. The determination not to repeat that experience lead to the European Commission's decision last year to suspend the fiscal rules written into the Stability and Growth Pact, which forms part of the Treaty on the Functioning of the European Union.

Low Rates Help A Lot

The good news for the majority of rated sovereigns is they face the current fiscal shock with historically low funding costs. Between 2015 and 2020 the effective borrowing rate (the average rate governments pay on total debt) for all developed sovereigns declined 80 bps to 2.0%, while, over the same period, the average cost of debt declined 50 bps to 2.1% for the G7. We expect developed government's borrowing costs to decline another 40 bps between now and 2023 on average. As we go to publication, the cost of issuance on new debt for nearly all developed sovereigns is well below the average cost of total debt. In the case of Italy, the current cost of 10-year funds is 0.64%, compared with the Italy's average cost of total debt (2.2% as of January 2021) and the 2023 projected average cost (1.9%). Today Japan, with gross general government debt of an estimated 239% of GDP, is financing itself at a 10-year maturity of 0.02%.

Monetary stimulus in developed markets has also pushed down funding costs in most emerging markets via rising portfolio inflows. The effective general government borrowing rate for the 60 largest rated emerging market sovereigns was 4.7% last year versus 5.1% in 2015. The noteworthy exception is EMEA, where, between 2015 and 2020, borrowing costs increased on average for some of the region´s largest sovereign borrowers including Egypt, Ghana, Turkey, and South Africa. In contrast, Brazil, which has for decades paid high real interest rates, saw a major decline in its average cost of funding from 14.7% in 2015 to 5.8% last year, and we project its average interest cost will drop to 4.5% by 2023, a trend that will provide it with some relief from otherwise mounting fiscal pressures. For emerging markets as a whole, we project that borrowing rates will decline by another 70 bps on average by 2023, across all regions except sub-Saharan Africa, due to elevated default risks, as well as South Africa's high cost of debt.

Where Does Fiscal Performance Fit Into Sovereign Ratings?

Fiscal performance is only one thing we look at when determining sovereign ratings. Some of the sovereigns highlighted in this report benefit from offsetting credit strengths: Japan is the world's largest external creditor. The U.S. controls the world's preeminent reserve currency. Between 2021 and 2027, GIPSI sovereigns are scheduled to receive an average of 7% of GDP in EU transfers from the newly created Recovery and Resilience Facility. Emerging markets such as South Africa and Brazil benefit from deep local currency capital markets and have relatively limited outstanding foreign currency debt. Higher public debt does not mechanically lead to a drop in creditworthiness, particularly when the cost of funding it is on the decline. Nevertheless, the numbers published in tables 1 and 2 should convey the size of the fiscal task that lies ahead. Future rating actions will reflect how governments proceed from here.

Table 1A

Table 1B

Table 1B

Table 2A

Table 2B

Table 2B

(Digital design by Joe Carrick-Varty)