By W A Wijewardena –

Asani, the bright Advanced Level student and her grandfather – Sarath Mahatthaya – a well-read retiree, are having intellectual discussions frequently on current issues of public interest. Earlier, they had two such interactive sessions, one on Millennium Challenge Corporation or MCC grant to Sri Lanka and the other implications on the economy of recent tax cuts by the present government (available here and here).

Asani had been confused by the conflicting public opinions expressed by way of slogans on the five reports on forensic audit exercise conducted by two leading international audit firms, namely, BDO India and KPMG India, at the instance of the Monetary Board of the Central Bank of Sri Lanka. She herself did not understand the forensic audit reports, nor could her teacher help her in that regard. Hence, she decided to raise the issues with her Grandpa:

Asani: Grandpa, a lot of people today speak about what is now known as forensic audit reports. What is a forensic audit? How does it differ from a normal audit?

Sarath Mahatthaya: It’s good that you have raised that issue. An audit is a study of what has happened in the past. Normally what we know is a financial audit. For instance, in the case of a company, say a company that produces polythene bags, a normal financial audit will try to establish whether that company has properly recorded its incomes, expenses and profits. A forensic audit – forensic itself means relating to judiciary – about the same company will try to establish whether there had been any irregularities in recording those transactions and if so who were responsible for them. Its findings can be used as evidence in a court of law and therefore, the forensic auditors should do their exercise carefully.

But these two audits alone are not sufficient to make any judgment about the company. Today we conduct economic audits and social audits as well. In an economic audit, we’ll seek to establish whether the resources used by company which we call economic inputs have generated sufficient economic outputs to cover the cost of inputs. A social audit is wider than that. It tries to establish whether society at large has been benefitted on a net basis. For instance, in the case of the polythene bag manufacturing company, it will seek to establish whether the costs it may have imposed on society have been sufficiently compensated by benefits that it has delivered to society. Today, in Sri Lanka, people including Parliamentarians and some scholars are highly worried about forensic audits whereas they should have asked for a social audit to assess whether the Monetary Board of the Central Bank had done its job in compliance with its obligations to society.

A: Then, what led to the conducting of the forensic audit by these two firms? Surely, they can’t do it on their own unless they had been commanded by somebody.

S: You’re correct. It’s the Monetary Board of the Central Bank which gave them this assignment. The Board was compelled to do so because the Presidential Commission on Treasure bond scams during 2015 to 2016 had in their report made a remark – I say a remark and not a recommendation – that even before 2015 there would have been irregularities in the issue of Treasury bonds and therefore, ‘it may be appropriate to conduct a forensic audit’. According to newspaper reports, the previous government immediately seized upon this remark and prevailed upon the Monetary Board to conduct a forensic audit on bond issues pertaining to the period prior to 2015.

This is embarrassing to Monetary Board because now it has to blame previous Boards for failing to act prudently.

A: There’s a general belief in the country that tenders are normally awarded without following proper procedures. Was it the case in selecting BDO India and KPMG India too?

S: No. Due process has been followed in this case and the final selection has been made by a Cabinet Appointed Tender Board, according to reports.

A: Then, what about the reports? Do you think that they’ve been done in a professional manner?

S: The five reports that’ve been presented contain a lot of important information relating to the issue of Treasury bonds, investments by the Employees Provident Fund or EPF and the supervision of primary dealers by the Monetary Board. Yet, there are certain deficiencies which have reduced the quality of the reports. Hence, the answer to your question is yes and no.

A: In that case, there’s nothing wrong in going by the findings in the five reports. Don’t you agree, Grandpa?

S: No. This is highly technical stuff and not many ordinary folk can understand it unless they’re guided by knowledgeable people. This’s true even for the so-called technocrats. You may recall, Asani, that in the case of Buddha’s Dhamma, subsequent Buddhist masters had to write elaborate commentaries known as ‘Atuvaa’ or ‘Teekaa’ explaining what exactly the Buddha meant in his discourses. A similar exercise is needed in this case too. Otherwise, it’s like the Sinhala proverb of placing a ‘sharp razor in the hands of a monkey’.

A: Oh, I see. I’m also such a monkey with a razor blade in the hands. So, I guess, before I cut myself, I need this guidance. Can you further elaborate on this?

S: I’ve several reservations about the reports. It appears that the forensic auditors have failed to do their analysis taking into account the mandate of the Monetary Board and the specific economic conditions the country was facing from 2005 to 2014. The Monetary Board has been given three conflicting mandates by Parliament. Their top priority is to have an inflation-free country with a stable financial system. This’s embodied in its co-objectives known as the economic and price stability and financial system stability.

Then, there’re two other conflicting mandates given to it by Parliament. One is to raise debt for the government at the lowest cost possible. The other is to give the highest return possible to members of the Employees Provident Fund. Therefore, when there’re moves to push the market interest rates by some interested groups, the Monetary Board has to intervene and stabilise the interest rates for the general good of the country.

In the first few years of the period covered by the reports, the country was at a war resulting in draining-out of all resources available to the government. The budget deficit was rising above 8%, foreign funding dwindling and inflation raising its ugly head above 20%. On top of this, the economy was faltering with growth below its potential rate. The Monetary Board had to raise funds for the government by tapping mainly the domestic market. Exploiting this situation, some of the primary dealers had attempted to push the interest rates up and it would have been fatal for economic recovery. So, the Monetary Board had to stabilise interest rates at around 20% and prevent any further interest rate hikes.

So, when the auctions of Treasury bonds began to push the rates up, the Board had to cancel the auctions and resort to an important weapon available to it to stabilise interest rates. That was to issue bonds through a system known as direct placements, a system that had been introduced in 1997 when the Board had introduced the primary dealer system and marketable Treasury bonds.

A: What you mean is that a social audit would have exonerated the Monetary Board for using direct placements more often during that period. 

S: Exactly. The actions of the Monetary Board have to be evaluated not in terms of mere financial numbers but in the context of the gravity of the economic conditions it was facing. That’s because the Board has to deliver a social good to society in terms of its mandate. When it does so, it has to necessarily incur a cost and society should come to recognise that that cost is being incurred for the greater benefit of society.

A: Good! What’re the other reservations that you have about the reports?

S: One is that there’s a fundamental error in the methodology the forensic auditors have used to calculate the losses they say have been incurred by the government when the Board had issued bonds through direct placements. That’s the assumption that the bond market is ruled by a single price throughout and if any bond has been sold below it, it’s a loss to the government.

A: That’s hilarious. Even at our AL class, we’re told that market prices don’t remain the same throughout. But, how do you apply to the bond market?

S: At the bond auctions, even for the same bond, they’re issued to different primary dealers at different prices. What each primary dealer has to pay is the price which he’s quoted at the auction and not the cut-off price of the auction which the last primary dealer has to pay. Suppose both of us bid at the auction. You quote Rs. 120 per Rs. 100 bond and I quote Rs. 100. If my price is the cut-off price, you don’t pay that. You pay the price you’ve quoted and I pay the price I’ve quoted. This’s known as the book-building auction method with multiple prices in each auction for the same bond. Hence, even in a single auction, there’s no single price.

A: But Grandpa, they say that there’s a weighted average price or WAP. What’s that?

S: That’s a price of convenience calculated by the Public Debt Department or PDD of the Central Bank to represent the results of the auction. It’s better than taking the simple average. In the above case, the simple average is just Rs. 110. But, if you bid for bonds to a value of Rs. 1,000 and I bid for a value of Rs. 500, the average price weighted by the amount which each one of us have bid is Rs. 113. This’s not an exact price but a better reflection of the results of the auctions. But it doesn’t show how far the actual prices have departed from this average. If there’s a wide range within which those actual prices have remained, it’s to be understood by reference to a statistical aggregate known as the standard deviation.

A: Then, what’s this secondary market price of bonds they talk about, Grandpa?

S: These bonds that you buy at the primary market can be sold to others before maturity and that market is called the secondary market. The actual price at which these bonds are traded in the secondary market is called the secondary market price. That can be different from the primary market price depending on the demand for and the supply of these bonds in the secondary market. Since there’s no debt exchange in Sri Lanka similar to the stock exchange for company shares, that information is available to PDD only after one day.

In other words, what’s available to PDD today is the actual price that had been there yesterday. But, by that time, today’s prices may have changed. To overcome this bottleneck, PDD collects information every morning pertaining to the maximum price at which a primary dealer is willing to sell and buy a particular bond, reduces it to a single value by using the simple average and releases to the market only for its guidance. That’s because they are not the actual prices but quotes by primary dealers.

It therefore presents some useful information about the level of the prices and the direction in which they may be moving. It is highly dangerous to use these quoted prices to calculate a loss or a profit.

A: How have the forensic auditors calculated the losses they have reported in their reports?

S: The forensic auditors have calculated the losses of issuing Treasury bonds for two different periods. One is for the period from 2005 to end-February 2015. The other is from March 2015 to April 2016. In addition, they have calculated losses to EPF too. To calculate the losses, they’ve created a fictional price called the base price. Any bond sold below that base price has been categorised as a loss to the government. By the same token, any bond sold above the base price should be reckoned as a gain for the government. However, the gain part has not been calculated by auditors since it hadn’t been a part of their mandate.

A: Have you got any reservation about the calculation of that fictional price called the base price?

S: Yes, of course. They’ve used a mixture of perceived secondary market prices to calculate the losses. Wherever the average of the secondary market price is available it has been used. But when it’s not available, it is the average of the buying and selling quotes published by PDD on a daily basis that has been used. Since that quoted price is a simple average, they’ve made an adjustment to it by reducing it by 5 cents. Obviously, the objective would have been to correct any error by going by the simple average when there’s a wide range of quoted prices. The selection of 5 cents has been done arbitrarily and not by following any statistical method like taking into account the standard deviation that measures the range of dispersion.

The use of both data sets is defective. The average secondary market price that had been used is not the one governing the market today. It pertains to the previous day and those prices are not applicable to the market today. That is because within 24 hours the market prices may have changed. The other weakness of using that price is that when you average it, you miss a whole range of prices that had prevailed on the previous day. Regarding the second type of prices they’ve used, the quotes are totally inapplicable to gauge the current market prices since they’re all perceptions of primary dealers. On top of this, the choice of 5 cents for the adjustment is arbitrary and any other researcher could have arrived at a different level of losses by assuming a different level of adjustments. For instance, if you reduce the average quoted price by 20 cents, all the losses will disappear and if you use, say 21 cents, it would result in a profit. Those prices they’ve used are useful in knowing the level of prices in the market as a guidance. They can’t be used to calculate the losses or, for that matter, gains involved in a transaction. Hence, these calculations cannot stand in a court of law as evidence.

A: What you say is that those loss calculations are wrong. Then, it must be equally applicable to losses they’ve calculated for post-2015 period as well. Isn’t it the case?

S: Yes, indeed. It applies to the loss calculation of Rs. 9.9 billion pertaining to the latter period too.

A: Media as well as politicians have gone to town on the loss numbers in the pre-2015 period. Any further elaboration on that?

S: From 2005 to end February 2015, there had been 4504 instances of placing direct placements by PDD. Of them, 2783 had been done above the so-called base price and therefore have generated a gain for the government. As I told you earlier, they’ve not calculated this gain since it was not a part of their mandate. Of the balance, 1105 direct placements have been made at prices below the so-called base price.

According to forensic auditors, it has generated a maximum loss of Rs. 10.4 billion. However, of those losses, nearly 96% has been gains by state-sector managed funds. EPF alone has gained about 61%. The rest has been earned by Bank of Ceylon, People’s Bank and the National Savings Bank. The so-called non-state sector primary dealers have got only 4%. Hence, the charge that PDD officers have unduly favoured private primary dealers isn’t confirmed by facts.

If one calculates the gains and net them against losses, there’s clearly a net gain for the government. It’s up to the Monetary Board to do this exercise and disabuse the minds of the public who have been misled by forensic auditors.

Another revelation that one can gauge from the first report is that it is only 24% of the direct placements during 2005-February 2015 that have been issued below this so-called base price. Hence, the balance 76% has been issued outside this price range.

A: It appears that the forensic auditors have taken the Monetary Board, Parliament, media and the public for a ride. 

S: Yes, this has in fact reduced the value of their report significantly. There’re many more howlers they’ve done. Let’s continue this interactive dialogue to find them out.

To be continued….

*The writer, a former Deputy Governor of the Central Bank of Sri Lanka, can be reached at waw1949@gmail.com