Call and repo transactions
In comparing FY99 and FY00, an interesting yet expected development is the increasing use of call transactions in the money market. The increasing use of call transactions in FY00 can be explained by two factors: (1) with the retirement of government debt in FY99 and FY00, bank holdings of government securities (specifically T-bills) fell by almost Rs 180 billion, and (2) since May 1999, money at call and short notice is exempted from cash or liquid asset requirements (CRR and SLR).

The first point simply refers to the fact that banks were running short of T-bills and were unable or unwilling to initiate repo transactions. The second point simply says that exempting call transactions from CRR and SLR requirements has increased banks’ incentives to undertake call transactions. In effect, with lower holdings of T-bills and the higher cost of call transactions, banks had the incentive to hold more liquidity to avoid being short. This reinforced the sense of tightness in the money market.

A positive offshoot of this tight liquidity position is the increasing efficiency of the money market. As shown in Figure V.4, the spreads for both bids and offers, narrowed in FY00. For 6-month repo transactions, the range of bids/offers posted by all interbank players (for any month) narrowed in FY00 on account of increased competition and the fact that T-bill rates had fallen in FY00 relative to FY99.

To complete this Chapter of the report, the following discussion will concentrate on Pakistan’s banking system. More specifically, it will touch upon how the changes discussed above have impacted different segments of the banking system.

Pakistan's Banking Systems
As discussed earlier, the Rupee deposit base has grown very marginally over the past two years. Looking at the breakdown of bank deposits by holders, the impetus for this slowdown has come from foreign constituents. These foreign depositors include all non-resident Pakistanis and official/business/personal accounts of foreign nationals. Despite the fact that special allowances were made for diplomatic FCAs after the freeze, the loss of confidence and alternative options (not available to domestic residents) are primarily responsible for this significant shift.

Table V.7a
Growth of Scheduled Banks' Deposits by Types
(End June basis)

Looking at domestic deposit holders, government/local bodies; trust funds and non-profit organizations have not shown any significant change in behavior. The change has come in private sector deposits, where business deposits increased while growth in personal deposits has fallen considerably (see Table V.7b). The sharp increase in business deposits from FY97 till FY99 could be due to both domestic dollarization (before May 1998) and efforts to convert out of these subsequently frozen FCAs. This is consistent with the sharp shift from term to demand deposits witnessed after May 1998. Looking ahead, the acid test is whether the banking system will be able to increase the maturity of Rupee deposits, or more specifically, increase personal deposits that are longer-term in comparison with business deposits.

Table V.7b
Growth of Scheduled Banks' Deposits by Types

* Central, provincial governments and local bodies.

In terms of the breakdown of deposits by bank (see Figure V.5), despite a stagnant base in the past two years, NCBs and private domestic banks have increased market share at the direct expense of foreign banks. This is to be expected given the reliance of foreign banks on FCAs. In fact, the banking landscape had changed in the 1990s, with foreign banks taking an aggressive lead in both deposits and advances despite a limited branch network. With domestic dollarization before May 1998, a small network did not limit deposit mobilization or corporate lending. On the advances side, since foreign banks cater largely to first-tier borrowers (multinationals and reputable local corporates), their branching network was not much of a hindrance.

With the obvious shift to Rupee banking, the issue of branch network and contestable markets became increasingly important. In addition to this, foreign banks reacted more strongly to the freeze than Pakistani banks, and began taking a much more cautious approach towards the country (see section on branch network).

In terms of advances (see Figure V.6), this refers primarily to loans to the private sector. Again, foreign banks have not seen much growth. Given the dominant size of the NCBs and privatized banks (MCB and ABL), robust growth rates in FY99 and FY00 have been able to shore up the overall growth of advances.

Cautious lending by foreign banks given the sharp fall in their annual investment in T-bills may seem surprising (see Figure V.7 & Figure V.8). Staying a close second to NCBs in terms of their investment in T-bills before FY99, this has fallen sharply in the past two years.

Whether it was the fall in T-bill rates or the liquidity problem they faced with the systemic shift out of FCAs, advances by foreign banks to the private sector stagnated despite a very sharp fall in the placement of funds in government securities. However, it should be mentioned that since foreign banks took a lead in mobilizing FE 45 funds (that have been rescheduled since the freeze in end-May 1998), these banks have been given the option of using the corresponding Rupee funds at their discretion (private lending or placement in T-bills) or to place such funds with SBP at a margin above their cost of funds. This facility is available as long as the FE 45 liability still exists. The Rupee funds placed with SBP do not constitute government borrowing, which explains why funds have not been lent to the private sector or the government.

Interest rates in the banking system
The reason for the sticky nature of a downward adjustment in interest rates is primarily on account of the changing structure of large Pakistani banks and the systemic problems experienced since mid-1998. With an overhang of non-performing loans (NPLs ) and the internal restructuring of the NCBs towards their eventual privatization, these banks want to retain their operating profitability to improve their provisioning against bad loans. When T-bill rates fall, this places earning pressure on revenues (from government securities) without a corresponding reduction in costs, which explains why banks are less willing to reduce lending rates to maintain their profitability.

However, competition to attract prime borrowers has forced commercial banks to start shaving lending rates. With lower revenues from lending to both the government and private sector, banks have reduced deposit rates to lower their operating costs. With the problems banks have faced maintaining their Rupee deposit base, they have been cautious of this option and have only recently started reducing deposit rates (see Figure V.9).

As shown in Figure V.9 and Table V.8a and Table V.8b, the fall in T-bill rates paved the way for a reduction in lending and deposit rates during the past two years. However, the reduction in deposit rates has been slower, since banks have had to pay special attention to ensure that their deposit base did not erode further in the face of the systemic changes that were already underway. The importance of these developments could perhaps be gauged by the fact that despite an even distribution of the 3.2 percentage point fall in lending rates during FY99 and FY00, average deposit rates fell by 1.8 percent, with a much larger adjustment in FY00.

Table V.8a
Weighted Average Rates of Return on Deposits
Interest Based- All Banks, on end-June basis (percent per annum)

* Excluding current and other deposits.

Table V.8b
Weighted Average Rates of Return on Deposits
PLS Based- All Banks, on end-June basis (percent per annum)

* Excluding current and other deposits.

Table V8c
Ranges of PLS Deposit Rates Declared by Scheduled banks
As on June 30, 2000 - (percent per annum)

Although lending rates are an important determinant of private sector investment, it is clearly not the most binding factor. Interest rates become more influential when confidence returns and interest and exchange rate outlooks are clearer.

A brief discussion relating to the spread in lending rates is useful. As shown in Figure V.10a, the median lending rate in FY00 has fallen vis-à-vis FY98, from 16-17 percent to 13-14 percent. The first thing to note is that the bulk of private sector financing at or below 10 percent per annum is on account of export finance. The EFS rate during FY98 was 10 percent, which was brought down to 8 percent on 1st July 1998. In effect, the first two ranges (0-10 and 10-11) can be ignored in evaluating the change in the nominal cost of borrowing. Of the remaining rates, the three-percentage point reduction in the median lending rate is clearly shown by the leftward movement in the distribution curve.

Even amongst the commercial banks, foreign banks lent 71.8 percent within a range of 10-15 percent, while Pakistani commercial banks lent 72.0 percent within a range of 13-19 percent (see Figure V.10b). This confirms the view that banks have well defined market niches; foreign banks primarily focus on prime borrowers, whereas the larger Pakistani banks also cater to the middle market. Private domestic banks that began operating in the early 1990s, have well defined clients and only the larger ones have begun interacting with a broader group of borrowers.

Non-performing loans
With the change in government in October 1999, there was a clear sense that the military government intended to carry out a comprehensive accountability drive. Although the primary target of this initiative was loan defaulters; tax evaders, unauthorized usage of utilities (or gross under-reporters) and corruption charges against ex-government officials, were also investigated. However, the most publicized development was the televised one-month deadline given to loan defaulters to square up their outstanding dues or face harsh penalties. The one-month deadline ended on 16th November 1999

As shown in Table V.9, the response to this drive is clear in terms of the cash recoveries during the month of November and December 1999; the Rs 7.7 billion recovered in November and Rs 4.0 billion recovered in December are the highest volumes that were recovered since the NPL issue was first formalized by the interim government in 1993. Furthermore, this encouraging development in cash recoveries still managed to follow the seasonal increase that is witnessed during May and June of each year.

In evaluating the effectiveness of this drive, it is important to realize that given the outstanding volume of defaulted loans, concerned parties could not be expected to bring in that much liquidity without recourse to further borrowing. Since this was obviously not in the offing, commercial banks and DFIs were empowered to accept cash down payments ranging from 10 to 30 percent of the outstanding amount in consultation with the National Accountability Bureau (NAB). The remaining part of these loans was restructured, with a repayment plan that was acceptable to both the bank and the defaulter.

Table V.9
Cash Recovery of Non-Performing Loans
(Rs million)

As shown in Figure V.11, the outstanding stock of defaulted loans did not show much of a reduction during the recovery drive, while NPLs showed a sharp increase in October, which then declined in next two months. Three points need to be highlighted: (1) the increase in NPLs before the recovery drive is more a reflection of the enforcement of classification guidelines than an increase in the incidence of delinquent loans, (2) the recovery drive was more effective with rescheduling loans than realizing cash recoveries, and (3) the defaulted loans shown in Figure V.11, only include loans that are greater than (or equal to) Rs 1 million. One must also realize that during FY00, the public exposure given to the loan default issue placed continuous pressure on banks and DFIs to improve their reporting systems. Hence, it is safe to say that awareness of this issue and the active role of the NAB, stemmed the sanctioning of suspect loans; the increase in NPLs is simply a more accurate picture of the actual state of affairs and the technical difference between NPLs and defaulted loans.

As shown in Table V.10a, despite the recovery drive, the outstanding volume of defaulted loans increased during FY00. More specifically, private domestic banks and DFIs recorded the largest increases. As a ratio of total advances, however, there was an overall improvement in the two-year period ending June 2000. As dominant players, the NCBs and privatized banks spearheaded the improvement; both were able to reduce their outstanding defaulted loans while increasing total advances. The sharp increase posted by private domestic banks in FY00 reflects a certain coming of age; while advances increased by 10.9 percent, defaulted loans grew by 78.1 percent in the same period. Although this recent performance by private domestic banks is disappointing, there is a growing difference between individual banks that have done well and those that have not.

Table V.10b, shows that although the overall magnitude of NPLs remains stagnant (as a percentage of total loans), financial institutions that focus on directed credit fared the worst.  In fact, excluding just ADBP, the ratio of NPLs to total advances shows an impressive fall, again driven by NCBs and the privatized banks. This primarily reflects healthy growth in advances (see Figure V.6) in the two-year period, while NPLs carried by NCBs and the privatized banks actually fell from Rs 124.9 billion in end-FY98 to Rs 105.5 billion in end-FY00.

Table V.10a
Ratio of Defaulted Loans to Total Advances

What is interesting to note is that during FY99, DFIs reduced their exposure to the private sector quite sharply, while the increase in the NPL ratio in FY00 is driven by a significant increase in NPLs as advances remained stagnant. Specialized banks continued to increase lending during these two years, but their outstanding NPLs increased from Rs. 19.4 billion to Rs. 57.0 billion in this period.

Table V.10b
Ratio of Non-Performing Loans to Total Advances

Table V11
Number of Scheduled Bank Branches

In overall terms, the consolidation of NCBs and the expansion of private domestic banks is a positive sign, but the reduction in foreign banks is not as auspicious. In the first few months of FY01, the effective merger of Standard Chartered and ANZ Grindlays has increased the concentration ratio of foreign banks (in terms of market share) and also dampened sentiments in this sector. Looking ahead, with the resumption of the World Bank sponsored financial sector reforms, it is expected that NCBs will experience a similar round of changes that had been witnessed in FY98. The privatization of these banks is already behind the tentative schedule that had been agreed with the World Bank in 1997.

Credit Plan for FY01
Following the events in FY00, and anticipating stronger performance by the manufacturing sector, net credit expansion to the private sector and PSCE has been set at Rs 85.2 billion (see Table V.12). Factoring in an extension of the rescheduling period, the availability of external assistance from IFIs, a stricter fiscal deficit target, and slightly higher non-bank borrowing, the government is expected to retire Rs 2.2 billion of its debt to the banking system. Furthermore, since the Stand-By Arrangement from the IMF will look for a significant increase in Pakistan’s liquid reserves, Net Foreign Assets are projected to rise by Rs 60.0 billion in FY01. These factors should increase overall monetary growth by 10.5 percent, which is slightly higher than the target growth in nominal GDP.

In view of the sharp increase in commodity operations during FY00 and the self-liquidating nature of this facility, commodity operations are expected to have zero impact on overall growth in money supply. Also, net lending by specialized banks is expected to remain small, while SBP financing of NBFIs will also be curtailed.

Table V.12
Credit Plan FY01
(Rs billion)

* WAPDA, OGDC, PTC, SSGC, SNGPL, KESC & PAL RAILWAYS