the general ledger leading to the crucial situation of final accounts not being in agreement with the books of accounts.
- When new channels like Automated Teller Machine (ATM) are added:
New channels linked recently are sometimes not tested substantially. An example of ATM itself will clear the point. In any bank, the books are closed for the day and re-opened almost immediately. Day closure is done any time from 5 pm to 10 pm. Assume that a customer has withdrawn cash on 31st March at 7 pm when the books were closed at 5 pm. This withdrawal is shown on 2nd April. If you had verified the cash at 5 pm on 31st March, even of the ATM you would find such a situation as correct. But when the posting is done on 2nd April, there is implication on revenue since the savings accounts are paid interest on daily basis. The customer gains interest of two days in this situation and more during long bank holidays. If the application has interest calculation on ‘as of basis then perhaps the interest loss will not occur.
- Borrower health classification:
Accurate classification of health of the borrowers of the branch has become the most important aspect of statutory branch audit. This has serious implications on the balance sheet of the bank in terms of provisions. Many application systems have a special routine or report or even separate software which does the work of classification of borrowers. Blind acceptance of such report will be tantamount to non-performance by the branch auditor. It is preferable for the branch auditor to test the system. What the system skips for classification of non-performing accounts (NPA) or what is incorrectly classified as NPA is of concern to the branch auditor. Branch auditor will thus have to sample test the system to assure him the accuracy of the classification as detailed below:
- Compare the previous year’s NPA list with the current list. Identify and seek reasons for why the accounts are upgraded and whether the reasons are justified as per rules permitted by the regulatory authority e. the Reserve Bank of India (RBI).
- Take the report from CBS system (as opposed to the borrower classification system which is a different application in some banks) which gives indications of nonperforming accounts in form of a few combination of reports like:
“Many application systeas have a special routine
or report or even separate software which does the
work of classification of borrowers. Blind acceptance
of such report wil be tantamount to
non-performance by tbe branch auditor.”
o List of loan accounts in arrears – if the report
permits, get those accounts where more than two installments are in arrears (since three installments in arrears is defined as NPA). This will permit you to study the borderline cases and identify some which the system may not have downgraded, o The above report does not include Cash Credit (CC) and Overdraft (OD) accounts. This is because these accounts are continuous and there is no concept of ‘installment’. What most systems will report will be CC and OD accounts which are overdrawn. Branch auditor will be able to identify the accounts which are overdrawn as on 31st March and a scrutiny of the accounts will confirm the date from which these accounts were ‘continuously’ overdrawn. If this period exceeds ninety days then the account needs to be classified as NPA. A check on classification of accounts to confirm such accounts are classified as non-performing will be adequate. A random check of earlier months will help isolate cases where accounts are continuously overdrawn but brought in order artificially for the year end. o Report of accounts not renewed/reviewed for more than a year is one of the statements to be submitted in LFAR. This list helps in confirming whether these accounts are included in the NPA list, o Report of stock statements in arrears: A report of Stock/Debtors statements not submitted to the bank for a period exceeding a user specified number of months should be generated and reviewed. If you specify the period of three months which is the trigger to downgrade the borrower’s account, this will be a ready list to validate the NPA list of the branch.
o Report on overdue Bills Purchased and Bills Discounted: This is another powerful report for verification of classification. The
importance of this report is that while the other facility of the borrower may be in order, this department may be accidentally ignored. Since one NPA account forces all facilities to the borrower to be downgraded, this oft-ignored feature should be emphasised by the branch auditor. Intervention of the Bank’s Data Centre and their professionals to make a customised query is not always needed.
- Anti Money Laundering (AML) application and its accurate representation:
The branch auditor is the last sentinel to protect the bank from abuse of this aspect. As the governing body – the Reserve Bank of India has amply reminded by issuance of various circulars, the statutory auditor needs to cover this aspect. Most CBS banks have resorted to software application for AML reporting. This software application may be from the CBS application provider or a third party software. Whatever be the case, the common feature of concern is that this module or application is an ‘afterthought’ as the AML guidelines came in force more than two
decades after CBS. The fundamental feature of this module/application is that it ‘reads’ data from CBS and processes it for its report. This feature of ‘reading’ is important because many banks boast NIL cases. This is usually because of ‘mapping’ error. This module looks at the wrong places and comes back with no cases to report which is misconstrued as NIL returns. Auditor can also scrutinise the ‘white listing’ of cases where the branch is better suited to remove the cases of suspicious transactions based on some rules of deposit and withdrawal which are also specified by the Reserve Bank of India. The reasons for not considering the transaction as suspicious is within the realm of the branch, by identifying retail cash business of account holder like petrol service station or grocer, etc. The classification rules are set normally at the data centre which the branch auditor need not concern with. But the reasons and replies definitely are of his concern to ensure these are logical and do not undermine the objectives of AML. Low AML training of bank staff emphasises the intervention of the branch auditor for AML audit.
Revenue accuracy assurance:
When most banks, even the co-operative banks have evolved to CBS, a topic of discussion like revenue leakages seems prima-facie redundant because the involved computer servers are the latest generation of machines having fourteen floating decimal point! Perhaps one may have to reconcile time period of this discussion being that of transitory to the destination of perfection, until which, we shall have to recognise the feature of imperfection, thus adjusting our audit plan to the inevitable verification of revenue accuracy even in the presence of the formidable machines and their magnificent chips. If one were to attempt a classification of the revenue errors, one might see a range of errors with severity to make most of us lose sleep or re-calculate every computer print. Analysis will show that operational or user error is the major cause and in very few cases, the software coding may be the cause of the error.
- Interest not levied on a particular product of the bank: For example, advance against mutual funds. All advances made under the new product will suffer the same fate. Since interest and all charges run are executed at the data centre, if one account is skipped, all accounts will be skipped at least for the branch. The source of the error is most likely at the time when new product is set up by the data centre. Mapping of the account (technical jargon) may not be done to the interest procedure. This means that the interest or charges procedure which is activated manually by the data centre does not have this product in its list.
- Non responsiveness of rate to increase in PLR/ TLPLR/ WCPLR, etc.: When the rate sanctioned for an account are linked to prime lending rates – PLR (say 1% over PLR), then the intention and expectation is that any change in the PLR will change the rates of ALL such accounts. Some accounts or group of accounts do not respond to the changes in the rates leading to under/ over recovery of interest. Two possible reasons causing such a situation may be:
- The user has not opened the account correctly in the interest section.
“Most CBS banks have resorted to software
application for AML reporting- The fundamental
feature of this application is that it ’reads’ data from
CBS and processes it for its report.”
“Where the borrower s eftmtte for EMI holiday
[for example hotsioi mo after construction or
education loan and tie eattstioo of EMI is done
manually, there is i ptssaftftmi of over or under
recovery. Under recover ■ sack coses lead to self
- Account convene-: from legacy system may not carry the PLR link feature of the new system. Idealy, this should have been detected and corrected under conversion audit.
- Installmek: holiday mz ……… s the total equal
monthly Installments (EMIk Where the borrower is eligible for EV.I f uav : r example, housing loan under constru – r e location loan) and the calculate: – : EMI – : re manually, there
is a possibility – – . – dr recovery. Under
recovery in such cases lead to self caused’ NPAs. It is also not |WWB■«**■« to note over-recovery where the loan a:: _nts reach credit balances. This is usual ■ a r-.ar._ai error caused by lackadaisicai super* s: n at the time of account opening which is the time of EMI calculation.
- Error in varying interest rate in a single advance account: OD against bank’s own fixed deposits may not always be right, for example, 12% upto
lakh & 14% between ? 2 lakh to 4 lakh. Since the manual cal: ulat ~ _ .mple procedure, one
expects it to be in the software with ease. Reality however dictates its success contingent on the caliber of the pr gram me rs. A sample audit will be in good stead . truer r over recovery is the result. Normal!the b rr wer would be the first to object in case < f . er recovery.
One hopes to reach the level of perfection until which this transitory phase shall be the irony of our audit activity where we have to re-check on our rupees two hundred worth calculator whether the computer .s rtr rupees twenty lacs has done its calculation o. rrectiv! In conclusion, though computers del: • er ate urate results, it is only because of a combination f user accuracy with that of the programmer. Auditors may be able to manage with the right report’ f the svstem itself and a bit of sample testing to perform their given task. ■