[ad_1]
There are growing hopes that an expansion of the IMF’s special drawing right system could provide funding for a UN facility that aims to synthesise liquidity in the African bond market. There is logic to the plan, argue Daniela Gabor of the University of West of England Bristol and Crystal Simeoni of NAWI, but there are also under appreciated risks. The two offer an alternative design proposal to manage these issues.
Special drawing rights (SDRs) are an esoteric international token system used by the IMF to help the lender settle international debts and imbalances between countries. In the context of international sovereign finances, they are best thought of as claims on the freely available hard currency of other members.
Most of the time nobody really pays attention to SDRs, as they are merely an accounting mechanism designed to facilitate IMF lending programmes. The last time they cropped up in any significant way was when George Soros successfully argued for their emergency expansion following the Global Financial Crisis.
After the pandemic a similar argument was made for their expansion and in February the IMF was finally given approval to allocate $500bn in new rights to help ease recovery in poorer countries.
The IMF is now also contemplating a potential redistribution of high-income countries’ new SDRs to provide, as US Treasury Secretary Janet Yellen put it, “liquidity to poor countries to facilitate their much-needed health and economic recovery effortsâ€.Â
But Yellen warned that US support is conditional on “shared parameters for greater transparency and accountability in how SDRs are exchanged and usedâ€.Â
This has raised hopes that the SDR injection could be directed towards funding something called the Liquidity and Sustainability Facility (LSF). The plan by the US investment firm PIMCO and the United Nations Economic Commission for Africa (UN-ECA) is to increase the attractiveness of African sovereign bonds and their liquidity via the introduction of a third party intermediary.
First announced in the pages of the Financial Times under the heading ‘Africa needs its own repo market’, and then formally promoted by the United Nations for consideration of Ministers of Finance in September 2020, the LSF has always faced one critical stumbling block: there was limited political appetite from OECD central banks to provide the $50-100bn in senior lending to the LSF. Which is where the SDR issuance comes. Some believe it could be used as an alternative funding source for the mechanism.
Synthesising liquidity
So how does it work? The LSF at its most basic level turns the logic of debt relief on its head. Instead of persuading or coercing private creditors to share the pain, it proposes to subsidise them via repo loans collateralised by sovereign bonds.
Private bondholders would borrow from the LSF via repo loans, pledging their portfolios of sovereign bonds as collateral with the facility. The financing raised, the LSF architects hope, would be used to generate additional demand for sovereign bonds, helping to increase liquidity and reduce funding costs for the sovereign issuers.
Here’s a diagram of how the different entities would interact:
As we know from the debates around the G20’s Debt Service Suspension Initiative, indebted countries are often reluctant to apply for debt relief because they fear losing access to capital markets.
The LSF, however, promises to sidestep this trade-off, instead offering a win-win solution: bond buyers have a guaranteed mechanism to finance their bonds at better conditions than private financing markets, whereas sovereign issuers have additional demand for their bonds. It would work, in theory, because the LSF would lend to bondholders at cheaper interest rates and lower haircuts than those available in private financing markets.
Repo haircuts, however, should not be confused with haircuts found in debt restructuring deals. The former are a safety cushion that protects the lender in case the borrower defaults and the lender needs to liquidate collateral. In contrast, the latter are found in debt restructuring and inflict losses on the lender, writing off a proportion of the debt to share the burden more equally.
Repo haircuts are therefore better thought of as a risk management tool for repo lenders:Â if collateral has good credit ratings and trades in liquid markets, like US Treasury bills, haircuts would be zero because the repo lender could easily sell collateral in case the borrower defaults, and recover its cash loan. That is to say, a repo borrower gets $100 in cash for every $100 of US Treasury collateral valued at market prices. However, at a lower credit rating and lower liquidity for the collateral, the repo borrower would, for example, perhaps only get $60 for every $100 of collateral: a haircut of 40 per cent.
As it stands, data on benchmark private haircuts for African sovereign bonds is scant. Anecdotal evidence for structured repos between Egypt’s central bank and global banks points to haircuts on Egypt’s foreign currency bonds varying between above 50 per cent in 2016 – when Egypt was negotiating a bailout package with the IMF – and 25 per cent by the end of 2018. This is most likely at the low end of private haircuts, since Egypt is one of the largest issuers of Eurobonds.
Therefore, as a result of the above, there is a lot of scope for the LSF to provide substantive subsidies by setting haircuts below market level.Â
Still, the LSF would not entirely abandon the risk function of haircuts: it would use a ratings-based methodology to adjust them. Haircuts would likely discriminate between sovereign issuers with more liquid bonds — say South Africa’s enjoying lower haircuts than Egypt. The haircut framework would also function as a signal to bond markets, just as the ECB’s haircuts on the collateral it accepts from Eurozone commercial banks signal its view of riskiness and liquidity.
The liquidity improvements for sovereign issuers are expected to come from the additional demand mobilised by LSF subsidies. As the UN-ECA put it “Modelled on existing market-based and commonly used facilities in Europe and the USâ€, the LSF would inject liquidity in African bond markets by attracting “a new class of investors while shaving off the higher borrowing costs that African nations face because of age-old stubbornly sticky perceptions that they are especially riskyâ€.
The increased appetite, ECA estimates, could save African issuers an estimated US$39–56 billion in interest costs over a five-year period.
But what about the vulnerabilities?
While the LSF’s ambitions are commendable, the macrofinancial risks remain under appreciated. In a new report focused on the African continent, we explore three such risks: cyclical liquidity for African sovereign issuers of collateral, perverse incentives for African countries to prioritise foreign currency debt (Eurobonds), and institutional conflicts between the commercial managers of the LSF and national central banks.
First, the LSF would rely on collateral valuation to ensure protection against its borrowers’ default. The LSF would (a) call for additional collateral (margin calls) when sovereign bond collateral falls in price, and (b) increase haircuts on sovereign bonds when credit ratings worsen. By increasing the costs of repo funding the sovereign bonds, the LSF threatens to create cyclical improvements in liquidity: that is, better liquidity for African sovereign bonds in good times – when it is needed less – that might rapidly disappear in bad times. Collateral valuation can amplify the liquidity pressures ignited by capital outflows, feeding, as Mark Carney put it, boom-bust cycles of liquidity and leverage. Such procyclical forces would shrink fiscal space, leaving citizens to carry the burden of the LSF’s “win-win†scenario.
Second, the LSF risks increasing African countries’ vulnerability to foreign currency debt. It is unclear whether the LSF would finance Eurobonds, local currency bonds, or both. The LSF’s firepower (say $100bn) would amount to roughly two-thirds of African Eurobonds outstanding mid-2020 ($150bn) and a fifth of local currency sovereign bonds ($500bn). Should the LSF accept Eurobond collateral — that is, sovereign bond collateral denominated in foreign currency (USD, EUR, CHF) — it risks creating perverse incentives for countries to shift to foreign currency debt, which is not only more difficult to service during periods of market stress, but also at odds with recent donor/ G20 initiatives to promote local currency bond markets.
Third, since the LSF would be administered by private banks, its haircut changes may undermine monetary policy autonomy in African countries and create conflicts of interest for the private administrators. Should the LSF choose to increase haircuts on some African sovereigns during bad times, it may directly hamper efforts by national central banks to preserve sovereign bond market liquidity. Furthermore, the LSF would hand over institutional power to its private commercial bank administrators, with potential conflicts of interest if these have commercial operations in the countries for which they make collateral decisions.
All of this means the LSF could perversely shrink fiscal space and poor countries’ capacity to rebuild after the COVID19 pandemic.
These risks, however, can be mitigated by ensuring the facility operates in local currency only (avoiding eurobond collateral), offers fixed rather than variable haircuts (and in so doing removes the risk its haircut policy out-signals the monetary policy decisions of national central banks) and injects countercyclical effects by waiving margin calls on private investors.
While there is a risk fixed haircuts might encourage adverse selection incentives for private investors to repo the worst-quality sovereign bonds, this too can be mitigated with quota controls. The removal of margin calls for private investors, meanwhile, would remove the risk that the LSF could trigger large portfolio exits from African sovereign bond markets rendering the funding stable and, therefore, appealing.
It is true the new institutional design would ask the LSF to assume collateral liquidity risk, but it is a risk that only generates losses for the LSF if its borrowers default and the LSF has to resort to (fire) sales of collateral.
We think it is highly unlikely that the LSF would be forced to sell collateral securities, since such action would immediately increase borrowing costs for African governments and drain liquidity. Instead, the LSF would become the outright owner of those securities, and receive coupon payments from the sovereign issuers directly.Â
One thing is certain, the LSF can be better designed to minimise the threats to fiscal and monetary autonomy in African countries if and when SDR funding heads its way.
[ad_2]
Source link