Wickremesinghe at a loss on losses to EPF

Fact Check

In response to a question from Frances Harrison, Director of the International Truth and Justice Project, regarding the Employees’ Provident Fund (EPF), the former President, in his claim, asserted that no one has lost even a cent in the ‘EPF’ due to debt restructuring.

FactCheck.lk consulted the Ministry of Finance publications related to Domestic Debt Restructuring to verify the claim.

There are three methods used to restructure debt: (i) nominal reduction of the maturity value (face value haircut), (ii) interest rate reduction (coupon haircut), and (iii) maturity-extension (reprofiling).

The former president’s claim was that the bonds held by the EPF were subject to only a maturity-extension/reprofiling (third method); hence, the EPF bond holdings did not suffer any loss in value. However, he is wrong on two counts: first, reprofiling was not the only method used, and second, the reprofiling did result in a loss of value.

First, bonds held by the EPF were not subject only to reprofiling but also to a coupon haircut. Calculations on the published data show that in the restructure the weighted average coupon on the bonds that were subject to surrender was 12.1%. In comparison, the weighted average coupon on the replacement bonds issued was 9.7%. That is, there was a coupon haircut in addition to the reprofiling.

Second, reprofiling does result in a loss if the surrendered bonds were purchased at a price below its face value (referred to as a price below par). If bought at par, the overall rate of return on the bond is the coupon rate. But when bought below par the return depends on the date the face value is repaid. Therefore, reprofiling (delaying the repayment) reduces the rate of return on the bond, when bought at a price below par.

The bonds subject to restructuring were equally accessible to all market participants, including the EPF, at the government auction. The data shows that the government sold these bonds at prices below par (see Additional Note). Therefore, reprofiling them also caused an additional loss (reduction in the rate of return) to the EPF on top of the loss caused by the coupon haircut.

The former president was wrong in asserting that there was only a reprofiling of EPF bonds and also wrong in asserting that reprofiling those bonds did not cause a loss.

Therefore, we classify the former president’s claim as BLATANTLY FALSE.

*FactCheck.lk’s verdict is based on the most recent information that is publicly accessible. As with every fact check, if new information becomes available, FactCheck.lk will revisit the assessment.

Additional Note: To verify that the bonds subject to surrender in the restructure were bought at a price below par, FactCheck.lk used the data published by the CBSL, found here. This has the face value and price of the bonds sold by the government as well as the ISIN numbers of the bonds surrendered in the restructure. FactCheck.lk calculated that the total face value of the bonds exchanged amounted to LKR 3,204 billion. And that the price at which these bonds were sold was LKR 2,907 billion, which means they were sold at 9.3% below par.

---

[Updated 25 April 2025] Right of Reply

FactCheck.lk responds to Wickremesinghe on EPF losses due to debt restructure

FactCheck.lk appreciates and welcomes the feedback from the office of former President Ranil Wickremesinghe regarding our fact-check published on 21 March 2025 in the Daily Mirror newspaper (see here). Our original analysis found his claim on the losses to the EPF following debt restructuring to be blatantly false.

This response addresses only the former president’s specific objections to our fact-check and will not engage with any tangential remarks or ad hominem attacks directed at FactCheck.lk or Verité Research.

We have taken the response of the former president seriously. We believe those we fact-check deserve the benefit of the doubt and that their responses should be given the highest consideration. The former president contests our analysis on three grounds—we will discuss each in turn.

(1) Alleges that the word “only” used in analysing Wickremesinghe’s claim results in a misrepresentation

After reviewing this claim carefully, we think that if there is any misrepresentation of his claim, it is in his response to the fact-check, and not in the fact-check itself. There are at least three reasons (linguistic, attributional and logical) for reading his statement as claiming that “the EPF was subject to only a maturity extension”. We set out these three below.

(i) The statement, which we quoted verbatim was “[…] In the case of EPF, what we dealt was the maturity extensions so not a single person has lost a cent in that.” The phrase “what we dealt” is, linguistically, a complete description of the action taken to restructure the EPF. The president did not say “one of the thingswe dealt” or “among the things we dealt”; he only said “what we dealt.”

(ii) It is an exclusive attribution. He mentions only the “maturity extension”. Nowhere in the interview does he leave any room for the listeners to hear him attributing anything other than a maturity extension to the restructure of the EPF.

(iii) Independently of the above two, the statement is logically coherent only if he was referring to the maturity extension as the only restructuring action on the EPF. The logical flow of the claim is as follows: Because X, therefore (“so”) Y did not happen. That is, the action X (maturity extension) is posited as the reason for result Y (“any EPF member losing even a cent”) not happening. This claim is logically incoherent if it were possible for there to have been other actions (other than X) that caused Y. It follows logically, therefore, that to assert as he does that Y did not happen only on the basis of action X is also to assert that in the space of actions with a bearing on Y, action X was the only action taken. That is, “the EPF was subject to only X (a maturity extension)”.

(2) Cites current market rates to argue that the rate of return on the EPF is “a significant differential in favour of the EPF’s bonds”

The former president’s response seems to argue that “In the case of EPF…not a single person has lost a cent” because “the new bonds receive a coupon rate of 12%.”  which is higher than the current market rate of 10.1%.

However, this is a spurious comparison, that does not support the fact-check contestation. The key comparison is between the EPF’s returns on the bonds that were surrendered (exchanged)—SBonds, and the replacement bonds (those received after the restructure)—RBonds. The average rate of return on SBonds was 12.1%; but the average return on RBonds was only 9.7% (because the RBonds will be stepped down to 9% from mid-2026 to maturity).

That means, the restructure reduced the interest rate (return) on the bonds held in the EPF. That reduction constitutes an economic loss—similar to having your long-term fixed deposit where your interest rate is restructured down from, say, 15% to 10%. This is a loss, even if the new fixed deposits available in the country pay only 8%. This is the comparison relevant to the fact-check, and it remains uncontested by the former president’s response.

(3) Claims that a “loss [to the EPF] cannot occur due to maturity extension of an instrument that is held to maturity and accounted in the amortized cost portfolio” and that the loss is not reported in the EPF accounts either

The former president cites the reporting of amortized value as a reason for believing that the EPF did not suffer a loss in the restructure. He references the EPF’s 2023 Annual Report, which states: “The EPF does not recognize any gain/loss on bond exchange” (page 312). This argument is a red herring.

It is because amortized value is not relevant to the calculation of the loss of returns. The loss to the EPF is from the reduction in the coupon interest rates (as discussed in part 2 above). The EPF exchanged bonds that had a higher coupon for those with a lower coupon, directly affecting the payouts to EPF beneficiaries. This loss in coupon returns on the asset is not captured in the amortized value, which is only the current capital value of the asset. To explain with the example above, you can have your fixed deposit amount unchanged but still suffer a reduction in the interest rate you get on it. Therefore, the EPF not reporting a reduction in the amortized value of the assets is irrelevant to the fact-check; it is a red-herring in the response received.

We thank the former president for his response. Our detailed response to that here is to explain why it does not warrant us updating the verdict of the original fact-check.