Let me explain this back-of-the-envelope analysis.
First of all, few disclaimers.
• These are personal views based on my calculations.
• Those calculations are very much simplified. Actuals might massively differ.
• Figures are based on published data as of 31.12.2023.
2/
More disclaimers:
• GFN here focuses only on T Bills + T Bonds. It excludes:
- other domestic law instruments such as SLDBs
- liabilities such as overdrawn balances in DST accounts, suppliers dues, etc.
• This is neither investment opinion nor advice. Do your own research.
3/
What is GFN?
Gross Financing Needs refer to the overall new borrowing requirement, plus debt maturing during the year.
This includes financing of fiscal deficit, other financing, plus amortisation.
Why GFN matters?
As per the approved @IMFNews program, debt restructuring should meet two separate GFN targets:
• average annual GFN below 13% (2027-2032)
• annual foreign currency GFN below 4.5%
This implies that annual domestic currency GFN is capped somewhere near 8.5%.
5/
IMF GFN Projections
After restructuring (both foreign as well as "selected" pool of domestic debt), IMF expects the GFNs to drop towards the program targets.
6/
What will be restructured?
Key point to note is that above expected drop in GFNs are post restructuring of domestic debt.
The only indication on what those assets are is below.
Interestingly, the IMF is silent on the meaning of “select pool of the remaining domestic debt”.
7/
What will be restructured?
However, Govt. recently clarified this to a certain extent.
T Bills:
Only T-Bills held by the CBSL will be restructured.
T Bonds:
Voluntary domestic debt optimization operation without coercion.
8/
Back-of-the-envelope Analysis
I arrived at an estimated GFN for T Bills, T Bonds, plus primary deficit (or surplus).
Assumptions:
• T Bills held by CBSL is restructured into 15-year T Bonds
9/
More Assumptions:
• Other T Bills are rolled-over annually, at a rate of 20% (2023) and 10% thereafter.
• T Bonds are rolled-over, for a term of 3 years, each, at a rate of 22.5% (2023) and 7.5% thereafter.
• The 10% and 7.5% rates are as per the IMF DSA assumptions.
10/
Interpretation of Results
GFN Gap = Estimated GFN - IMF GFN projection
A GFN Gap implies that servicing burden (due to domestic debt) is beyond what's estimated by IMF.
This means that such debts must be restructured, further, to align GFN with IMF levels.
11/
As the below figures show, even after restructuring of LKR 2,598 Bn T Bills held by CBSL, significant GFN gaps remain.
Note:
The favourable GFN gaps in 2024 and 2025 are due to my (conservative) assumption that all T Bond outflows in 2023 to be rolled-over for 3 years (into 2026).
This means further restructuring is needed for T Bonds.
13/
Which T Bonds are to be restructured?
Given that the GFN gaps are awful for 2029 and beyond, you might think that a significant extension of maturity of those T Bonds would resolve this.
Nope.
Let me tell you why.
14/
If you look closely, contribution to these acute GFN levels are:
• ~ 5% (of GDP) due to T Bill roll-overs
• ~ 8.5% to 16% due to T Bond roll-overs
Both are not due to maturities falling during in 2029 - 2033.
But, due to legacy maturities due from 2023 to 2028.
15/
Conclusion
This means an extension of maturity needs to be done for ALL bonds, not just medium- or long-term ones.
16/
TL;DR
• GFNs remain elevated, even after reprofiling T Bills held by CBSL, especially 2029 and beyond.
• This is due to legacy maturities and roll-overs between 2023 to 2028.
• So, An extension of maturity needs to be done for ALL bonds, not just long-term ones.
17/
That's it. Now you know the back-of-the-envelope analysis and what it means.
If you enjoyed this:
• Like this tweet to show some love
• Retweet the top tweet (below) to share with others
• Follow me @theGayan for more insights on #investing
So far #SriLanka has talked about restructuring ext debt & not domestic debt.
Our paper shows that current IMF targets for DR are unlikely to deliver a sustainable pathway for econ recovery & debt management, unless rates are brought down by over 10%.
2/10
We don’t see a path for reducing int rates adequately, given high risk in the economy & lack of confidence on financial management.
Int rates are also high because Govt has a huge borrowing appetite simply to repay what has already been borrowed.
• Nominal int rate: rate of return earned in “money” terms
• Real int rate: rate of return earned in “real” (purchasing power) terms. It's the return after adjusting for effect of #inflation (impact of rising price levels)
2/6
Why the formula is wrong?
“Fisher's Equation” gives the link between:
• Real rate (R),
• Nominal rate (N), and,
• Expected inflation rate (E).
1 + N = (1 + R) * (1 + E)
So,
R = {(1 + N)/(1 + E)} – 1
An “approximation” of this turns out to:
R = N - E
3/6