Delivered in Parliament on 3 November 2021
Mr Speaker, I understand that the stipulations in the Government Borrowing (Miscellaneous Amendments) Bill on are largely of a technical nature, aiming to repeal various mostly spent acts, and to consolidate aggregate borrowing by the government across others. I favor such simplification, and do not have substantive concerns about specificities in the proposed amendments.
I do wish to make a few points that have more to do with a number of potentially unintended consequences of the current iteration of the bill. These have to do with the debt limit, precluding the acceptance of advance deposits by the MAS for Treasury Bills, and the lack of justification for rejecting sales.
Balancing fiscal credibility against risk of debt ceiling shenanigans
The first point worth noting is that Clause 14 of the Bill proposes to set the borrowing limit for government loans at $1.065 trillion, which represents the combined limits of $960 billion for Government securities and $105 billion Treasury Bills that currently exist under two separate acts.[1] This new limit may then be increased by Parliamentary resolution, which simplifies the, since logically, all forms of debt—whether short or long-term—should be treated as a component of the overall liability picture, and hence subject to the same form of debt ceiling discipline.
Given this observation, would it perhaps not make sense to also enfold the remaining instrument of government borrowing—SINGA bonds issued under the Significant Infrastructure Government Loan Act—to the same limit? The limit is comparatively small, at $90 billion.[2] A consolidated limit would simplify monitoring and allow fiscal policymakers to focus on one single figure.[3]
One objection to this is that the government maintains separate debt limits for debt that serve developmental objectives—such as infrastructure, which is the case for SINGA—versus those issued to either meet risk transfer and investment purposes—in the investment of CPF funds—or to fulfill market liquidity and depth functions.[4]
The Ministry of Finance has steadfastly held that these all require a separate limit account. But all money is fungible, and it is unclear why debt issued under one type of accounting constraint cannot subsequently be redeployed to other functions, so long as it is only a matter of protocol, rather than one entrenched in actual legislation. Of course, such deviations need not occur under this government, but it could easily occur under another, perhaps more profligate one. Even if we wished to maintain separate debt ceilings, perhaps it is worth considering nevertheless incorporating these distinct limits into this bill?
This is especially the case since financial market participants would view the consolidated government balance sheet as a singular whole. And, indeed, SINGA bonds do not appear to have affected the Singapore government’s overall credit rating,[5] which would be one economic reason to hive off limits for the instrument.
Concomitantly, it is also important to calibrate the frequency with which this limit is modified. While fiscal rules—such as a reasonable debt limit—have been used in many countries and can play an important role in ensuring the sustainability of fiscal affairs,[6] a debt ceiling that is too frequently binding can wreak havoc on the efficient operation of government, as the experience of the United States—where partisan brinksmanship routinely plagues efforts to raise the debt ceiling—reminds us.
To this effect, may I enquire whether the government has considered whether this limit is appropriate? What is the premise or justification behind the current debt ceiling amount of $1.065 trillion dollars, and is there is a body that routinely examines the veracity of this debt limit?
Precluding the acceptance of advance deposits by the MAS
The second point I wish to raise has to do with how Clause 20 of the Bill restricts the ability of other government entities—in particular the MAS—from accepting advance deposits in anticipation of raising loans by way of Treasury bills. While this may be a sound principle in practice—history is rife with examples of central banks that have monetized debt on behalf of a profligate sovereign, breeding inflation as a result—it is worth noting that two recent developments may yet necessitate additional flexibility for the monetary authority to act in such anticipatory manner.
The first such circumstance has to do with the zero lower bound faced by nominal interest rates. Although Singapore never quite faced this problem in the aftermath of the Global Financial Crisis of 2007/08—and the MAS, unlike most other central banks, targets the nominal exchange rate rather than the policy interest rate—it is nevertheless worth keeping in mind that one prescription to the problem of the zero lower bound is to allow for central banks to intervene directly in short-term money markets by issuing securities on their own account.[7]
The second situation has to do with the emerging central bank digital currency (CBDC) landscape. Such CBDCs are likely to become a permanent part of the future of central banking worldwide, and the MAS has even commissioned a sandbox for studying such currencies. While the detailed findings of this exercise have yet to be made public, it is reasonable to ask if the practical implementation of CBDCs may yet require the flexibility for MAS to be able to accept such advance deposits for Treasury Bills.
While I accept that these two conditions are unusual, it is nevertheless worth asking if this particular clause in the Bill may unnecessarily tie the hands of the MAS to operate in a modern global macroeconomic environment?
A justification for excluding sales
Finally, Clause 24 of the Bill explicitly states that the Authority may, “refuse any application to take up Government securities or Treasury Bills issued… without assigning any reason”.[8]
While I accept the prerogative of the Government to reserve the right to refuse the sale of its debt to any given entity, it is also surely reasonable for, at the least, Singaporean purchasers to be provided with an explanation for why they are unable to purchase the debt of their own government.
Notwithstanding these reservations and questions, Mr Speaker, I support the Bill.
[1] Specifically, the Government Securities Act and the Treasury Bills Act.
[2] Significant Infrastructure Government Loan Act (2021), Sec. 5(1).
[3] The counterargument that SINGA serves different a function—smoothing the intergenerational transfer of infrastructure spending—is also not entirely satisfactory, since longer-term government bonds serve many similar development objectives. Besides, the limit only offers a rule for all consolidated borrowing of government; monitoring and functional objectives are a separate issue and can (and should) differ.
[4] The former comprise Special Singapore Government Bonds and Singapore Saving Bonds, while the latter are Singapore Government Securities.
[5] Heng, J. (2021), “SINGA Not Expected to Hurt Singapore’s Credit Rating,” The Business Times, May 11.
[6] On the cross-country experience with fiscal rules, see Lledó, V., S. Yoon, X. Fang, S. Mbaye & Y. Kim (2017), “Fiscal Rules at a Glance,” IMF Background Paper, Washington, DC: International Monetary Fund. On general principles for selecting rules, see Eyraud, L., V. Lledó, P. Dudine & A. Peralta (2018), “How to Select Fiscal Rules: A Primer,” Fiscal Affairs How To Notes, Washington, DC: International Monetary Fund.
[7] Rule, G. (2011), “Issuing Central Bank Securities,” Centre for Central Banking Studies Handbook, London: Bank of England.
[8] Clause 24, Ln 6–8.