Achieving the SDGs requires success in realizing six major transformations: quality education (SDG 4); access to good quality and affordable health care (SDG 3); renewable energy and a circular economy (SDGs 7, 12, and 13); sustainable land and marine management (SDGs 2, 14, and 15); sustainable urban infrastructure (SDGs 6, 9, and 11); and universal access to digital services (SDG 9). Each of the six transformations requires a significant scaling-up of public investments. Yet the financing needs for these SDG investments are far greater than the fiscal space available to the governments of low-income developing countries (LIDCs). To achieve the SDGs, the LIDCs will need a significant increase in fiscal space, which will require a combination of domestic and global fiscal policies.

SDG 17 (Partnerships for the Goals) explicitly recognizes the need to mobilize increased public financing for developing countries, specifying several tools to do so (including tax collection, official development assistance, other additional resources, and debt relief). The first four targets of SDG 17 are as follows:

17.1 Strengthen domestic resource mobilization, including through international support to developing countries to improve domestic capacity for tax and other revenue collection.

17.2 Developed countries to implement fully their official development assistance commitments, including the commitment by many developed countries to achieve the target of 0.7 percent of ODA/GNI to developing countries and 0.15 to 0.20 percent of ODA/GNI to least developed countries; ODA providers are encouraged to consider setting a target to provide at least 0.20 percent of ODA/GNI to least developed countries.

17.3 Mobilize additional financial resources for developing countries from multiple sources.

17.4 Assist developing countries in attaining long-term debt sustainability through coordinated policies aimed at fostering debt financing, debt relief and debt restructuring, as appropriate, and address the external debt of highly indebted poor countries (HIPC) to reduce debt distress.

Even before COVID‑19, the financing needs of the LIDCs had not yet been mobilized. In a 2019 note on financing gaps to achieve the SDGs, the International Monetary Fund (IMF), together with the UN Sustainable Development Solutions Network (SDSN), demonstrated that the incremental financial costs of achieving the SDGs in the LIDCs exceeded their potential public revenues, assuming a significant rise in the tax-GDP ratios in these countries. The IMF estimated that the LIDCs would have to increase their SDG outlays by roughly 12 percent of GDP to achieve the 2030 targets. This incremental spending was beyond the means of these countries, leading to an SDG financing gap to the order of $300 to $500 billion per year (Gaspar et al., 2019). Note that the financing gap in Gaspar's paper was based only on five sectors: health, education, roads, water and sanitation, and electrification. Including other SDG sectors would have increased the estimated financing gap.

COVID‑19 has further expanded the SDG financing gap. Given the severe economic setbacks caused by the pandemic – and the two-year delay in implementing SDG investments – the IMF estimates that incremental spending needs are now roughly 14 percent of GDP for each year to 2030: roughly 21 percent more than was estimated in 2019 (Benedek et al., 2021).

COVID‑19 has also highlighted the limited capacity of the LIDCs to tap market financing. While the high-incomecountry (HICs) governments have borrowed heavily in response to COVID‑19, low-income countries (LICs) have not been able to do so, due to their lower market creditworthiness. The HICs have taken on massive new public debts and greatly expanded the money supply (monetizing part of the new debt) without any significant rise in long-term borrowing costs, consumer price inflation, or currency depreciation.

The United States has been the biggest borrower of all during the COVID‑19 pandemic. The IMF estimates that the United States' general government deficit (covering federal, state, and local borrowing) will average 15 percent of GDP in both 2020 and 2021. According to IMF estimates, general government net debt will rise from 83 percent of GDP in 2019 to 109 percent of GDP in 2021. The US Federal Reserve has monetized a substantial proportion of public debt. The Fed's high-powered money (currency in circulation and bank reserves held at the Fed) rose by $2.4 trillion from January 1, 2020 to March 31, 2021 – or 11.4 percent of 2020 GDP.

In view of the very low borrowing costs facing the HICs, the IMF has encouraged increased public borrowing in the advanced economies to support the short-term response to COVID‑19 and the long-term investment in structural transformation, such as the rapid transition from fossil-fuel energy to renewable energy. As the IMF's Fiscal Monitor (April 2021) notes, "The response of fiscal policy has been unprecedented in speed and size. In the COVID‑19 emergency, governments used the budget promptly and decisively. In the last twelve months, countries have announced US$16 trillion in fiscal actions. Fiscal actions have enabled health systems and have provided emergency lifelines to households and firms. By doing so, fiscal policy has also mitigated the contraction in economic activity" (IMF, 2021a).

Yet the LIDCs, by contrast, have not been able to engage in comparable deficit financing, as they face considerably higher borrowing costs than the advanced economies. As the Fiscal Monitor summarizes: "Average overall deficits as a share of GDP in 2020 reached 11.7 percent for advanced economies, 9.8 percent for emerging market economies, and 5.5 percent for low-income developing countries.

Countries' ability to scale up spending has diverged" (IMF, 2021a). The LIDCs have been unable to undertake the same kind of emergency response and investment-led recovery, even though they need fiscal expansion even more than the advanced economies to respond to the pandemic-induced economic crisis and the need for increased SDG investments.

The major short-term implication of the difference in fiscal space of the high-income and low-income countries is that the rich countries are likely to recover from the pandemic more quickly than the poor countries. As shown in figure 1.1, the IMF projects that sub-Saharan Africa will lag farthest behind in growth through 2023.

Note: GDP per capita, current prices, purchasing power parity, using 2019 as base 100.

Source: IMF (2021b)

The key to SDG success for the LIDCs is to enable these countries to borrow at the same scale relative to GDP and on approximately the same interest rate terms as the rich countries. This will require new forms of international policy support.

There are at least four key ways to increase the fiscal space of the LIDCs. The first is improved global monetary management, notably improved liquidity for the LIDCs. The second is improved tax collections supported by several global tax reforms. The third is increased financial intermediation by the multilateral development banks (MDBs) to support long-term development financing. The fourth is debt relief. Here we consider these four policy choices in turn.

To understand the role of improved liquidity management in raising the fiscal space of the LIDCs, it is instructive to compare the borrowing costs facing the United States versus Argentina (as of May 2021). The US 10-year Treasury bond yield is 1.6 percent, while the 10-year Argentina bond yield is 6.0 percent. The fear, of course, is that Argentina will end up in default.

Yet when we compare the fiscal fundamentals, there really is a bit of a mystery. According to IMF data (World Economic Outlook, April 2021), Argentina's general government net budget balance in 2020 was -8.9 percent of GDP, compared with -15.8 percent of GDP in the United States. The United States' gross debt as a share of GDP was 127 percent at the end of 2020, compared with 103 percent of GDP in Argentina. US general government revenues as a share of GDP were 30.3 percent in 2020, compared with 32.8 percent in Argentina. On the standard indicators, Argentina's fiscal situation is no worse than that of the United States – and is perhaps stronger. Yet its ability to borrow is obviously far lower.

The paradox seen in the disparity in borrowing terms between the United States and Argentine plays out more generally when comparing HICs and LIDCs. On average, according to the Fiscal Monitor (April 2021), the average debt of the advanced economies was 120.1 percent of GDP in 2020, compared with 64.4 percent for the emerging economies, and just 49.5 percent for the LICs (IMF, 2021e). The HICs borrow because they can; the LICs are credit-constrained even though their needs for capital investments are much larger.

On basic fiscal indicators – deficit, debt, taxes, and seigniorage relative to GDP – Argentina actually looks more fiscally sound than the United States, yet Moody's gives Argentina a sub-investment grade bond rating, while the United States is given AAA. One can argue that because Argentina has defaulted in the recent past, it is not to be trusted in the present. Yet there is another interpretation that reverses the direction of causation. Argentina lacks ready access to international capital markets and therefore pays high interest rates on its debt, which in turn makes Argentina far more likely to default. This is a self-fulfilling prophecy.

In this alternative interpretation, Argentine government risk does not reflect long-term insolvency, but rather extreme short-term illiquidity. When the US Treasury borrows, there is no liquidity risk. The Federal Reserve can always print dollars as needed to cover Treasury debts coming due. The Bank of Argentina, by contrast, cannot ensure the payment on Argentina's dollar-denominated debts. The fact that Argentina borrows in dollars rather than pesos means that the Argentine government is vulnerable to a self-fulfilling liquidity crisis, in which foreign banks and bondholders refuse to roll over Argentina's dollar-denominated debts or to extend new dollar credits to the government to service old dollar- denominated debts. If the foreign creditors fear a default by the Argentine government, they stop lending to it – and thereby trigger the very default they feared.

This self-fulfilling default risk facing governments like Argentina that borrow abroad in dollars rather than in their national currency has been called punishment for "original sin". The idea is that Argentina and countries like it are being punished for past fiscal abuses that have left them unable to borrow abroad in their own national currency, thereby leaving them highly vulnerable to a self-fulfilling liquidity crisis. The upshot is that Argentina pays far higher borrowing costs than the United States, even though Argentina's actual fiscal framework is in fact more, not less, orthodox. Argentina is punished while the United States gets away with fiscal sins.

The ease with which the US Government can borrow abroad is famously termed the "exorbitant privilege" it has due to the unique global role of the US dollar. Because its government can borrow in dollars, the United State does not face liquidity risks, and therefore faces much lower borrowing costs than countries that borrow in dollars rather than their own currency, and that are therefore vulnerable to liquidity shocks. What is interesting, and in some ways remarkable, is how the US Government is able to extend the blessing of its dollar to other selected countries through the Fed's liquidity policies, notably through Federal Reserve swap lines.

A Federal Reserve swap line allows a designated foreign central bank to receive dollars from the Fed in return for its national currency, up to a limit agreed between both central banks. In effect, the foreign central bank is enabled to "print dollars" up to the agreed swap limit. If the swap line is large enough, this forestalls the risks of a self-fulfilling liquidity crisis. The Fed has extended such swap lines to nations that are key US allies, or otherwise favored trading partners and geopolitical partners. Following the 2008 financial crisis, the Fed extended swap lines with five key central banks: the Bank of Canada, the Bank of England, the Bank of Japan, the European Central Bank, and the Swiss National Bank.

On March 19, 2020, in response to the COVID-19 crisis, the Fed added swap lines for an additional set of countries beyond the original five: Australia, Brazil, Denmark, Korea, Mexico, New Zealand, Norway, Singapore, and Sweden (Board of Governors of the Federal Reserve System, 2020). For countries receiving Fed swap lines, the March 2020 announcement had a galvanizing effect. Stock markets in emerging economies that had been plummeting because of the COVID-19 crisis suddenly turned around and began to rise, bolstered by the knowledge that the country in question was no longer as vulnerable to a self-fulfilling financial panic. Figure 1.2 shows, for example, the dramatic turnaround of the Brazilian stock exchange on March 23, soon after the Fed's announcement, pivoting from collapse to surge. In fact, US stocks also rallied sharply on the Fed's swap announcement, as it guaranteed that COVID-19 would not lead to a replay of the 2008 global financial crisis (when the Fed did not extend immediate liquidity to the market or to other central banks in the wake of the Lehman Brothers bankruptcy in September 2008).

Note: Bovespa historical data as of 10 May 2021. The U.S. Federal Reserve announced the establishment of temporary U.S. dollar liquidity arrangements (swap lines) with other central banks, including Banco Central do Brasil, on 19 March 2020.

Source: Investing.com (2021)

Countries covered by Fed swap arrangements face vastly superior borrowing terms than others. Using Ferri et al.'s (2000) method of scoring Moody's sovereign bond ratings by country, from 0 (default) to 100 (AAA), we see that countries with Fed swap arrangements have much higher Moody's ratings than those without such arrangements, controlling for the size of public debt relative to government revenues. A simple cross-country regression suggests that having the benefit of a Fed swap line raises a country's sovereign risk rating by 31 points out of 100, roughly the difference between an S&P rating of A (75) and BB (45), which can mean a difference in borrowing costs of 200 basis points or more.

This analysis, albeit only a sketch at this stage, suggests that the borrowing capacity of emerging markets could be improved markedly through greater access to liquidity. This could be accomplished in at least three ways.

1. First, the Federal Reserve could expand markedly the set of central banks with Fed swap lines. This seems like a low-cost, high-return opportunity.

2. Second, other central banks with internationally traded currencies, notably the European Central Bank, the Bank of England, the Bank of Japan, and the People's Bank of China, could establish their own swap lines with sets of developing countries.

3. Third, the IMF could be empowered to play a much more aggressive liquidity role. This is the key idea, for example, behind the issuance of Special Drawing Rights (SDRs), which offer liquidity to IMF members in a similar way to a Fed swap line – an IMF member state can exchange its SDRs for dollars, pounds, euros, or another strong global currency. The IMF plans to issue US$650 billion of SDRs to its members in the summer of 2021. This is an important step forward, yet the issuance is still too small to replicate Fed swap lines, especially since SDRs are allocated in proportion to a country's IMF quota: meaning that the bulk of the SDR allocation will go to the high-income countries.

In view of SDG financing needs, and therefore the need for greater market liquidity for low-income countries, the provisional conclusion is that the IMF's capacity to ensure liquidity for low-income countries needs to be further enhanced. The IMF needs even greater fire power to forestall self-fulfilling panics facing borrowing countries. This could come through a much larger SDR allocation, a lopsided allocation targeted to low-income countries, or some other mechanism through which the IMF could lend rapidly and with little or no conditionality to developing countries facing liquidity crises.

To achieve the SDGs, most countries in the world (with the exception of the already highly taxed countries of Europe) will have to increase tax revenue as a share of GDP. Greater tax revenue is needed for four incremental fiscal purposes: (1) public investments in physical infrastructure; (2) public investments in human capital (notably nutrition, health, and education); (3) public investments in R&D and technology transfer; and (4) public investments in income redistribution. The Nordic countries, top-ranked in the SDG Index (and therefore most on-track to achieve the SDGs), collect government revenue of around 50 percent of their GDPs. The United States however, far behind in SDG achievement, collects only 30 percent of its GDP in government revenue. On average, the LIDCs collect far less, only around 17.5 percent of GDP, with the emerging economies collecting only 20.5 percent of GDP. As the IMF notes, through comprehensive administrative and policy reforms, these countries should be able to increase their domestic government revenue by 3–7 percent of GDP (Benedek et al., 2021).

Yet much of the work of raising government revenues will require international tax cooperation. The rich countries, led by the United States, the United Kingdom, the Netherlands, Switzerland, Ireland, Luxembourg and some others, have created a plethora of tax havens in their own national tax jurisdictions, as well as in the Caribbean, the North Sea, and other places popularly dubbed "Treasure Islands". Such offshore tax havens are not the result of renegade small island states evading the will of the powerful countries, but of highly paid tax lawyers in New York City and London and lobbyists in Washington and European capitals who have conspired to create a truly global scam, in which hundreds of billions of dollars of corporate profits are shifted each year from the tax coffers of developing countries to these tax havens. Rich countries peform poorly on the 2021 International Spillover Index presented in this report, which includes measures related to tax havens, financial secrecy, and profit shifting.

A recent study by Garcia-Bernardo and Jansky (2021) summarizes the situation as follows:

We estimate that MNCs [(multinational corporations)] shifted US$1 trillion of profits to tax havens in 2016, which implies approximately US$200–$300 billion in tax revenue losses worldwide. MNCs headquartered in the United States and Bermuda are the most aggressive at shifting profits towards tax havens, while MNCs headquartered in India, China, Mexico and South Africa the least. We establish which countries gain and lose most from profit shifting: the Cayman Islands, Luxembourg, Bermuda, Hong Kong and the Netherlands are among the most important tax havens, whereas lowand lower-middle-income countries tend to lose more tax revenue relative to their total tax revenue. Our findings thus support the arguments of low- and lower middle-income countries that they should be represented on an equal footing during international corporate tax reform debates.

(Garcia-Bernardo and Jansky, 2021, page 3)

There are at least five kinds of global tax reforms that could significantly increase government revenues of developing countries. First, the regulatory framework that enables tax havens themselves could be eliminated through the actions of a few key countries, led by the United States, the United Kingdom, and the European Union. Second, countries could agree to reverse the recent spate of corporate tax cuts around the world, the so-called "race to the bottom" in corporate taxation, by a coordinated increase of statutory corporate tax rates. Third, countries could agree on a formula for taxing Big Tech companies (Facebook, Google, Apple, and others), whose data services are now largely or wholly untaxed. Fourth, countries could agree to collect and share a worldwide wealth tax on the world's super-rich. According to Forbes Magazine (April 2021), there are currently 2,775 billionaires worldwide, with a combined net worth of $13.1 trillion (Forbes, 2021). A 2 percent wealth tax would therefore raise as much as $260 billion per year from fewer than 3,000 taxpayers! Fifth, countries could agree on the long-discussed Financial Transactions Tax, which could also raise tens of billions of dollars per year that in turn could be directed to the SDGs.

While the HICs are taking advantage of the world's low-interest rate environment to borrow heavily for post–COVID-19 recovery, the LIDCs continue to face high borrowing costs. In addition to enhanced liquidity mechanisms discussed above, another means of increasing low-cost lending to LIDCs is through enhancing the lending capacity of the MDBs, including the World Bank and the various regional development banks. Currently, the MDBs lend slightly more than $100 billion per year, roughly half of which is from the World Bank group, with the remainder distributed by regional development banks (Nelson, 2020). There is a powerful case for a dramatic scaling-up of MDB lending in the coming decade, perhaps tripling annual lending to around US$300 billion per year, to cover about half of the SDG financing gap of the LIDCs.

There are powerful reasons to scale up MDB lending in support of the SDGs. The MDBs borrow on highly favorable market terms (generally AAA or thereabouts) based on the borrowing capacity of their shareholder governments, which are dominated by the high-income countries. The MDBs therefore are able to borrow with long maturities and low interest rates, advantages that the banks can then pass along to the LIDC recipient countries. Moreover, the MDBs are by their very design and purpose equipped to handle complex lending for infrastructure projects that simultaneously address economic, social, and environmental considerations, and which must overcome many collective action problems for success.

MDB financing moreover offers a favorable political perspective as well for the high-income countries that provide most of the paid-in capital of the MDBs. Each $1 of shareholders' paid-in capital can support roughly $5 of lending on the balance sheet of the bank. If the project's financing is blended, with half coming from the MDB and the other half from financial markets, then each $1 of paid-in capital can support $10 in total lending. Not only does MDB financing serve to mobilize private financing, but it also "de-risks" it, since the presence of the MDB as a lead creditor lowers the operational and default risks of the project.

For a United States legislator, the choice between voting for an additional $1 of bilateral US aid versus an additional $1 of MDB paid-in capital should be clear. The $1 of bilateral aid supports $1 of spending in the recipient country. The $1 of paid-in capital, by contrast, is matched roughly by another $4 of paid-in capital by other MDB shareholders – leading to $5 of paid-in capital in total, which in turn can support $50 in total blended financing. Each $1 of MDB funding thereby supports $50 of project financing.

As mentioned earlier, in Target 17.4, the global community has committed to using debt relief and debt restructuring to help finance the SDGs, building on the success of the Highly Indebted Poor Country (HIPC) Initiative that supported the Millennium Development Goals. At the start of the COVID-19 pandemic, the G20 adopted the Debt Service Suspension Initiative (DSSI) for LICs (specifically, IDA-eligible countries), a very small step in the right direction. The DSSI provided initially for a one-year suspension of debt servicing to bilateral creditors during 2020. The deadline was twice extended by six months, to cover the period until the end of 2021. As the DSSI offers temporary relief only for LICs vis-à-vis bilateral creditors, it is of small benefit, resulting in short-term relief of around $5.7 billion of debt servicing during 2020 (IMF, 2021f).

In November 2020, the G20 introduced a new Common Framework to complement the DSSI. The Common Framework calls on all of the G20 creditor nations to work together to provide meaningful debt relief, including debt reduction as needed, on a case-by-case basis. The Common Framework is helpful for bringing the traditional bilateral donors – the so-called Paris Club – together with non-Paris Club creditors, including China, India, Turkey, and Saudi Arabia (IMF, 2021f). Still, the Common Framework lacks clear standards and metrics for debt relief and does not tackle the issue of debts owed to private-sector creditors or official institutions (such as the IMF and MDBs). The LIDCs will almost surely need a more systematic debt restructuring program along the lines of the HIPC initiative.

In fact, the developing countries as a whole have too little debt rather than too much debt. According to the World Bank International Debt Statistics 2021, low-income and middle-income countries combined have a total public and publicly guaranteed debt of $3.1 trillion, on a combined Gross National Income (GNI) of $31.1 trillion – a debt representing merely 10 percent of GNI (World Bank, 2021). Debt servicing is around $1 trillion, of which around $800 billion is principal repayments and $200 billion is interest. Total interest servicing, therefore, comes to less than one percent of GNI, a surprisingly small number, indicating a capacity to take on considerably more debt in the future, assuming that it is directed towards SDG priorities (World Bank, 2021). Of course, there are certainly individual countries with excessive debt burdens that need to be reduced.

The SDGs are first and foremost a public investment program – in core infrastructure (roads, power), digital, water and sanitation, human capital (health, education) and the environment. To achieve the SDGs, the LIDCs will have to scale up public investment outlays by another 10–15 percent of GDP per year for the coming decade. The needed financing should come through higher domestic revenues combined with significantly greater levels of international borrowing. Success will require a high level of global cooperation and solidarity: in monetary policy (for example, Special Drawing Rights), in domestic and international tax policy, in development financing through the MDBS, and in debt relief. In essence, the LIDCs will need fiscal space comparable to that enjoyed by the HICs.