DP/03/2007 (English Version)

BANK OF THAILAND DISCUSSION PAPER

Thai Commercial Banks One Decade after the Crisis: Assessment of Risk to Financial Stability
By

Don Nakornthab

July 2007
E-Mail Address: donnP@bot.or.th

The opinions expressed in this discussion paper are those of the author(s) and should not be attributed to the Bank of Thailand.

Thai Commercial Banks One Decade after the Crisis: Assessment of Risk to Financial Stability
Don Nakornthab* July 2007

1. Introduction
The Thai commercial banking sector was at a center stage of the financial crisis that hit the Thai economy in July 1997. Beyond inconsistent macroeconomic policies, aggressive lending by commercial banks was a major factor that fueled the economic bubble and massive currency mismatch in the real sector. With the collapse of the exchange rate peg, defaults skyrocketed and banks balance sheets were subject to severe distress. When the banking sector was on a verge of bankruptcy, what had started as seemingly containable exchange rate depreciation quickly became a fullblown twin crisis. A decade after Thailands worst economic calamity, Thai commercial banks have seen significant improvements in profitability, asset quality, and risk management. This paper however is not about how the public and the private sectors have together helped put the troubled sector back on track. Thailands experience with banking-sector restructuring and subsequent structural reforms has been well documented. See, for example, Santiprabhob (2003) and Vichyanond (2007). This paper is more like an input for central banks macro-prudential analysis. Its main objective is to assess Thai commercial banks current financial conditions and short-term outlook which may impact financial systems stability. The analysis in this paper draws on a number of sources. The main source of aggregate bank data is the Bank of Thailands website. The sources of individual bank data are a large database maintained by the Stock Exchange of Thailand (the SETSMART database) and banks 2006 annual reports. All Thai commercial banks except one are listed companies and therefore required to disclose to the public a wealth of financial information.

Senior Researcher, Economic Research Department, Bank of Thailand. The views expressed in this paper are my own and not necessary those of the Bank of Thailand. Paper presented at ADB-TDRI-MOF, Thailand Joint Conference, Integrating Asian Economies: Ten Yeas after the Crisis, Bangkok, 18 July, 2007. I thank Thanawat Pruksananont for able research assistant. All errors are the mine.

The rest of the paper is organized as follows. Section 2 takes stock of the state of the Thai commercial banking sector as at the end of 2006. Significant content of the section is devoted to analysis of post-crisis changes and their implications for Thai commercial banks going forward. Section 3 presents an assessment of Thai banks short-term outlook, with a focus on their credit risk exposure. Section 4 concludes. Implications for regional integration are described in Appendix A.

2. The landscape at the end of 2006

This section provides a high-level overview of Thailands commercial banking sector landscape, dividing into the players and the environment in which they operate. Where relevant, attempts are made to compare the current landscape to the one right before the 1997 crisis to highlight important changes that have taken place over the past ten years.

2.1 The players a. Industry structure Figure 1 shows the breakdown of total assets of Thailands commercial banking sector at the end of June 1997 and December 2006, respectively. In several aspects, the structure of the industry does not appear to have changed much. There are about the same number of both Thai and foreign commercial banks today as there were ten years ago (14 versus 15 and 18 versus 19, respectively). The six largest banks are the same ones (two have changed their English names). The number one foreign bank was similarly about the same size as the smallest of the medium-sized Thai banks. Taken as a whole, foreign commercial banks accounted for about 13% of the sectors total assets at the end of 2006, dropping slightly from around 15% as of the end of June 1997. Using gross loans and deposits (data not shown) in place of total assets, the corresponding shares of foreign banks at the end of 2006 were much smaller at 8.3% and 10.1%, respectively.1 Making up the differences with the total asset share is the disproportionate shares of foreign banks in interbank and money market items (on

1 It is noteworthy that foreign banks share of gross loans has not changed much from June 1997. In contrast, the deposit share of foreign banks has tripled over the period.

both the assets and the liabilities sides of the balance sheet), investments in securities (on the assets side) and borrowings (on the liabilities side). This partly reflects legal restrictions on foreign commercial banks and hence their different business models Within the group of Thai commercial banks, the industry has become slightly more concentrated, with the Herfindahl index increasing from 0.118 to 0.123 and the combined asset shares of the six largest banks increasing from 74.7% to 80.5%. This reflects post-crisis industry consolidation and a quest for scale economies by the large players. In the latter regards, it is noteworthy that large foreign players have also become larger relative to their foreign peers. The top five foreign banks now control 75% (two-thirds of which belong to Japanese banks) of foreign banks total assets, up from 42% a decade ago. Figure 1. Total assets of commercial banks and finance companies in Thailand
Jun 1997
Bangkok Bank Krungthai Bank Thai Farmer Bank (Kasikorn) Siam Commercial Bank Bank of Ayudhya Thai Military BanK (TMB) Union Bank Siam City Bank Bangkok Metropolitan Bank Bangkok Bank of Commerce Bank of Asia Thai Danu Bank Nakornthon Bank First Bangkok City Bank Laem Thong Bank All 19 foreign banks Bank of Tokyo Sakura Bank
Total assets Share of banking (Bt m) sectors assets

Dec 2006
Bangkok Bank Krungthai Bank Kasikorn Bank Siam Commercial Bank TMB Bank Bank of Ayudhya

Total assets Share of banking (Bt m) sectors assets

1,216 729 704 564 432 350 275 255 203 162 138 129 72 71 46 954 134 87

19.3% 11.6% 11.2% 9.0% 6.9% 5.6% 4.4% 4.0% 3.2% 2.6% 2.2% 2.0% 1.1% 1.1% 0.7% 15.1% 2.1% 1.4% 1.3% 0.9% 0.6%

1,496 1,206 989 942 751 665

17.3% 13.9% 11.4% 10.9% 8.7% 7.7% 4.8% 3.0% 2.5% 2.2% 2.2% 0.9% 0.5% 0.5%

Citibank 83 HSBC 56 Bank of America 40 91 finance companies

1,646

Siam City Bank 412 Thanachart Bank 257 Bank Thai 219 UOB (Thai) 189 Standard Chartered (Thai) 186 Tisco Bank 81 Kiatnakin Bank 80 ACL Bank 40 All 18 foreign banks Bank of Tokyo Mitsubishi 233 Citibank 181 Sumitomo Mitsui 178 Mizuho 140 HSBC 118 6 finance companies 81 2 retail banks 40

1,139

13.2% 2.7% 2.1% 2.1% 1.6% 1.4%

Source: C.B. 1.1 and C.B. 1.2 reports for bank data; BOT website for finance company data

The bottom of Figure 1 shows the number and the total assets of finance companies at the end of the two respective periods. Finance companies were an important fixture of the Thai financial sector and had many overlapping businesses with commercial banks. The distinguishing feature is that these companies were not allowed to raise fund in the form of deposits, but only in the form of higher-interestrate promissory notes. In return, they were given exclusive rights to operate leasing 3

and hire purchase businesses on top of almost everything that commercial banks could do.2 In 2004, the Bank of Thailand decided to phase out these companies, requiring them to return their license or upgrade to retail banks which can serve only individuals and SMEs or to commercial banks which now can operate leasing and hire purchase business as well (more detail in Section 2.2c). What is remarkable from Figure 1 is the dramatic decline in the number and the total assets of finance companies after the crisis. Right before Thailand was hit by its worst post-war economic fallout, there had been ninety-one finance companies whose combined assets exceeded one-fourth of the banking sectors total assets. By the end of 2006, there were six of these companies left, with combined assets of less than one percent of that of the banking sector. field has now become much less crowded.3 One important change that is masked by Figure 1 is the change in shareholding structures of Thai commercial banks. Although not as badly hit by the crisis as finance companies, the banking sector did suffer significant damage. From 1998 to 2000, Thai commercial banks were forced to raise around 800,000 million baht in additional capital, a substantial sum considering that the combined capital base of all Thai commercial banks at the end of June 1996 was slightly below 500,000 million baht (about 20 billion USD at the pre-crisis exchange rate). Several banks were closed down, intervened, sold, or merged into other entities. Of todays fourteen Thai commercial banks, only nine banks remain Thai privately-owned, compared to fourteen out of fifteen before the crisis. Three are majority-owned by Financial Institutions Development Fund (FIDF): Krungthai, Bank Thai, and Siam City. Two are wholly-owned by foreigners: UOB (Thai) and Standard Chartered (Thai). But even among the remaining Thai private banks, foreign ownership has also increased significantly. Most notably, Singapore-based DBS and U.S.-based GE Capital have a sizable stake in TMB Bank (18%) and Bank of Ayudhya (29%), respectively. Table 1 shows percent foreign ownership in the nine private banks at the end of 2006. Thus, to the extent that these

companies competed with commercial banks in several market segments, the playing

2 3

A notable exception is foreign exchange business. It is interesting to note that five commercial banks in operation today (Thanachart Bank, Bank Thai, Tisco Bank, Kiatnakin Bank, and ACL Bank) are reincarnation or upgrades of finance companies.

Table 1. Foreign ownership of Thai private banks as of December 29, 2006.

Bangkok Bank Kasikorn Bank Siam Commercial Bank TMB Bank Bank of Ayudhya Thanachart Bank Tisco Bank Kiatnakin Bank ACL Bank Foreign ownership (%) 48.8% 49.0% 34.7% 37.9% 48.6% 0.0% 49.0% 44.0% 20.0%

Note: (1) Foreign ownership calculated as percent foreign limit less percent foreign availability. (2) Pre-crisis foreign limit was 25.0%. Source: SETSMART

Finally, it should be noted that while the total assets of the Thai banking sector grew 37% over the period (41% counting only Thai banks), the ratio of the banking sectors total assets to GDP declined from 133% to 111%. The fall in banking-sector assets relative to GDP indicates the reduced significance of the banking sector in the Thai economy.

b. Thai banks performance in 2006 On many accounts, 2006 was a solid year for Thai commercial banks as a group. Improved profitability and solvency both contributed to further strengthening of the sectors financial position. A look at Thai banks net profit in 2006 alone distorts the picture. All together, Thai commercial banks 2006 net profit dropped 40% from 2005, bringing down the sectors return on assets to 0.75, the lowest level in three years. The drop in Thai banks net profit was largely a result of IAS39-compliant loan loss provision burden. (See Section 2.2c for more detail of IAS39.) The other major factor that contributed to the reduced net profit was the expiration of crisis-related tax shield for some commercial banks. In terms of pre-provision operating profit (PPP) margin, however, 2006 was a post-crisis record-breaking year. The growth in PPP was supported by a strong increase in net interest and dividend income, which in turn benefited from improvement in net interest margin (NIM) and continued loan growth. Figure 2 traces out the evolution of Thai commercial banks net profit and the amount of loan loss provisioning expenses along with ROA and the ratio of PPP to

averaged total assets from 1990 to 2006.4 For our purpose, we use 1992-1996 as the pre-crisis base for performance comparison. 1992 was the first year that commercial bank interest rates in Thailand were fully liberalized. It was also the year that commercial banks were allowed to set up Bangkok International Banking Facility (BIBF) operation which enabled them to channel low-cost foreign funds to domestic borrowers and started the subsequent lending boom. There, we see vividly the extent of the 1997 crisis and the turnaround that followed. Provisions for massive loan losses turned an industry that had previously registered on average higher than 50,000 million baht of net profit a year into an industry with net profit losses of more than 300,000 million baht in 1998 and 1999. It took nearly five years from the beginning of the crisis to put the industry back on a stable growth path. Note that if we were to plot also retained earnings of Thai commercial banks in Figure 2, we would find that it only turned positive in 2004. Figure 2. Provisioning expenses, net profit, ROA, and PPP of commercial banks registered in Thailand, 1990-2006
Million baht Percent

400000 300000 200000 100000 0 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 -100000 -200000 -300000 -400000
Loan loss provisions Net profit ROA (RHS) PPP (RHS)
Avg. PPP (92-96) = 2.76% 2006 PPP = 1.98%

8.0% 6.0% 4.0% 2.0% 0.0% -2.0% -4.0% -6.0% -8.0%

Source: BOT website; Krungthai Bank (1997); authors calculation

While Thai banks 2006 PPP in baht amount exceeded the pre-crisis high, expressed as a percent of averaged total assets, it was still below the pre-crisis average. This is due partly to the fact that the cost-to-income ratio of Thai

commercial banks after the crisis has become higher and partly to the fact that most

To be precise, the data shown in Figure 2 to Figure 6 are for commercial banks registered in Thailand which from 2005 include retail banks and subsidiary of foreign banks in addition to Thai commercial banks. However, because of their small sizes, their inclusion makes virtually no difference with respect to the aggregate financial ratios.

Thai commercial banks still carry a sizable amount of non-income generating assets (non-performing loans plus foreclosed properties) on their balance sheets. Figure 3 plots the aggregate NIM and the aggregate cost-to-income ratio of Thai commercial banks during the same time span as Figure 2. At 3.2%, Thai banks 2006 NIM was a post-crisis high. For the second consecutive year, the increase in NIM in 2006 was propelled by a stronger increase in interest income relative to interest expenses. This stood in sharp contrast with previous years where the

primarily culprit for NIM improvement was a relative decline in interest expenses. Looking in a context of a contemporaneous interest rate cycle, this is consistent with the observation that during an interest rate downtrend, banks adjust the deposit rates first, but during an interest uptrend, banks adjust the lending rates first. Figure 3. NIM and cost-to-income ratio of commercial banks registered in Thailand, 1990-2006
4.0%
182% 174% 2.5% 2.1%

200%
3.2% 3.0% 160%

3.0%

Average 92-96 = 3.7%

120%

2.0%

1.0%
Average 92-96 = 42%

1.7% 1.8% 96% 80% 85% 1.0% 66% 59% 54% 53% 57% 0.6% 40% 0.5%

0.0% 1990 1992 1994 1996 1998 2000 2002 2004

0% 2006

Net interest income/averaged total assets

Cost-to-income ratio (RHS)

Source: BOT website; Krungthai Bank (1997); authors calculation

It should be noted that while the 2006 NIM is materially lower than the precrisis average, it is highly respectable by international standards, particularly when compared to those in developed economies. For example, the average NIM of large EU banks in 2006 was a little over one percent (ECB, 2006). Pro-market activists may argue that the high NIM reflects a low degree of competition among Thai banks. Under the same interpretation, however, the fact that the NIM is below the pre-crisis average suggests that there is more competition now than before. One negative development in 2006 was a deterioration of Thai banks cost-toincome ratio. For the first time since 1998, Thai banks cost-to-income ratio increase

over the previous year, propelled by a rise in operating expenses over operating income. The upturn suggests that the cost-to-income ratio may never go back to its pre-crisis level, i.e., there may have been a structural shift in the cost structure of Thai commercial banks. Nevertheless, the 2006 cost-to-income ratio was in line with international norm and hence not yet a cause for concern. In terms of asset quality, the ratio of non-performing loans (NPLs) to total loans of Thai commercial banks dropped further to 8.0%, the lowest level since 1997 (Figure 4). The continued improvement in the NPL ratio reflected banks effort to clean up their loan books in response to the Bank of Thailand (BOT)s more stringent provisioning rules that took effect in August 2004 as well as to meet the BOT target to bring down the ratio of net NPLs which do not count NPLs that have been fully provisioned to 2% by the end of 2007.5 Figure 4. NPL ratio of commercial banks registered in Thailand, 1997Q4-2006Q4
Percent of total loans
70 60 50 40 30 20 10 0 1997 Q4
2006 Q4 = 8.0%

Entire Banking system Thai banks only

1998

1999

2000

2001

2002

2003

2004

2005

2006

Source: BOT website (98Q2-06Q4); CEIC (97Q4 and 98Q1)

In 2006, the solvency position of Thai commercial banks improved further (Figure 5). Although one banks capital adequacy ratio fell below the regulatory minimum of 8.5%, the Tier 1 and the total capital adequacy ratios of Thai commercial banks as a whole rose from 10.0 and 13.2 at the end of 2005 to 10.9 and 13.8 at the end of 2006, respectively.

5 The ratio of net NPLs to total loans of Thai banks at the end of 2006 was 4.5%. No historical figure is available. For prudential purpose, the relevant NPLs are the net ones. This is because there is no further impairment on banks' capital from the portion of NPLs that have been fully provisioned. From a profitability point of view, however, keeping NPLs that are fully provisioned on its balance sheet entails a carrying cost for a bank.

Figure 5. Total and Tier 1 capital as percent of risk-weighted assets of commercial banks registered in Thailand, 1997-2006
16 14
11.97 13.28 11.39 8.91 7.5 8.93 9.57 9.98 9.01 10.46 9.23 7.34 7.92 8.48 12.96 13.43 12.36 10.92 13.25 13.85

12 10 8 6 4 2 0 1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

Tier 1 capital

Total capital

Source: BOT website

A look at aggregate data alone often does not give a complete picture of industry-wide performance, for they are to a large extent driven by the performance of large banks. Table 2 looks at key financial ratios of Thai commercial banks by five subgroups: large private banks, medium-sized private banks, small private banks, state-owned banks, and foreign-owned Thai banks. All ratios in the table are taken or calculated from individual banks 2006 audited financial statements which are available to the general public. Table 2. 2006 key financial ratios of Thai banks by subgroup
Large private banks 3 46%
1.4% 2.5% 14.0% 3.5% 5.9% 2.5% 3.5% 52.8% 2.8% 24.7% 7.5% 3.6% 14.5%

Number of banks Share of total assets (Thai banks only) Profitability ROA PPP ROE NIM Yields on interest earning assets Funding costs Net interest spread Efficiency Cost to income Operating expenses to average assets Non-interest income to total income Asset quality and solvency NPL ratio Net NPL ratio CAR

Medium private banks 3 22%

-0.4% 0.9% -6.5% 2.5% 6.1% 3.5% 2.6% 78.1% 2.6% 14.4% 7.1% 4.3% 11.1%

Small private banks 3 3%

2.2% 2.5% 10.2% 3.9% 7.6% 4.7% 2.9% 59.0% 3.3% 20.5% 11.6% 6.0% 25.8%

Stateowned banks 3 24%

0.1% 1.6% -13.7% 3.1% 6.5% 3.0% 3.6% 62.1% 2.4% 14.3% 6.1% 4.3% 10.8%

Foreignowned banks 2 5%
1.1% 2.8% 10.8% 4.3% 7.7% 3.0% 4.7% 58.6% 3.4% 21.0% 7.0% 2.9% 15.2%

Note: (1) Simple average (2) Data from consolidated financial statements except for SCBT and TBANK (3) NPL ratios are of solo (company) basis.

(4) Large private banks: BBL, KBANK, SCB Medium private banks: BAY, TMB, TBANK Small private banks: TISCO, KK, ACL State-owned banks: KTB, SCIB, BT Foreign-owned banks: UOBT, SCBT Source: Thai banks 2006 annual reports; authors calculation

In terms of ROA, small private banks came ahead of the other four subgroups. In terms of ROE, their performance was less impressive. This is because small private banks have on their books an enormous amount of capital relative to their total assets. Their capital adequacy ratio at the end of 2006 was three times above the regulatory minimum. This raises a question whether they use their capital funds efficiently. Nevertheless, to the extent that they had the highest NPL ratios in the table, the high capital adequacy ratio may not be too overwhelming. Perhaps the most unpleasant thing for small private banks is that they had the highest funding costs (interest paid on average interest-bearing liabilities). This

reflects their small sizes and limited branch networks. The main reason that they managed to do well in 2006 were higher yields from hire purchase and leasing businesses, their traditional strength from the time they were finance companies. To the extent that other banks can now also engage in these businesses, there is a downside risk to future profitability of small private banks. Overall, the large private banks appeared to perform best in 2006 relative to the other subgroups, followed not-too-distantly by foreign-owned Thai banks. Large private banks have the lowest funding costs which reflect their market dominance. Had they not carried substantial NPLs on their books (which depressed their asset yield), their performance would have been even more impressive. Foreign-owned banks had higher NIM and net interest spread (the difference between yields on interest-earning assets and the funding costs) than large private banks, but had higher operating costs and provisioning expenses which brought down their profitability. This indicates that economies of scale may play some roles here. Recall from Figure 1 that foreign-owned banks are much smaller than the large private banks. Medium-sized private banks and state-owned private banks did not do as well as the other three subgroups in 2006. Not only did they have the lowest return on assets, but because they also had low NIM and high operating cost to income relative to the rest of the industry. One bank in each subgroup made net profit losses in 2006 which turned the average ROE of both groups into negative territory. But the others

10

also did not fare much better. Moreover, their net NPL ratios indicate that both medium-sized private banks and state-own banks may still have a lot more to do in terms of extra provisioning under IAS39. The question of interest then is whether these banks will be able to improve their performance going forward. c. The changing business model6 The 1997 crisis made it necessary for Thai commercial banks to adapt and develop in nearly every dimension of their operation. This subsection looks at

changes in Thai banks business structure, sales and services, distribution channels, and risk management. Before the crisis, Thai banks earned most of their income from lending business. Interest income accounted for nearly 90% of banks total income on

average between 1992 and 1996. Within the loan portfolios, the focus of most banks was on large corporations. Such dependency made banks profit sensitive to an economic downturn. Defaults by a few large borrowers meant huge damages to income and capital base. To achieve greater portfolio diversification, Thai banks have given more importance to consumer loans and loans to small and medium enterprises (SMEs). The proportion of consumer loans in Thai banks loan portfolio in particular rose from 13% in 1998 to 21% at the end of 2006. To reduce dependence on interest income, non-interest income, especially fees and commissions, has been emphasized. Noninterest income to total income of Thai banks averaged 22% during 2002-2006, up from 10% during 1992-1996. In terms of sales and services, cross-selling and customer segmentation was generally rare before the crisis. The need to increase the bottom line amidst increased competition after the crisis motivated Thai banks to cross-sell and move to a customer-centric architecture where customer needs come first. Several banks have adopted a customer relation management (CRM) platform to serve specific needs of different customer segments. With regards to distribution channels, the most visible change is probably branch rationalization. Even the state-owned banks have fewer people in their

branches before the crisis. At the same time, Thai banks have aggressively used
6

Most of the materials in this subsection are from Triratvorakul and Vacharachaisurapol, 2006

11

ATMs and sub-branches to increase points of service and lower cost-to-serve. Use of mobile and internet banking also has an impressive growth albeit still small in total volume. Last but perhaps most important is the change in Thai banks risk management. A good example is in the area of credit and credit risk management processes which have become more systematic. Before the crisis, Thai banks

customer acquisition and lending approval was largely decentralized with branch managers and regional managers making most of the decisions for loans of less than ten million baht. While large loans were subject to board approval, there was

typically little time to perform comprehensive credit analysis at the headquarters. The discretionary power of branch managers extended to pricing where relationship with branch managers often determined the final pricing. Figure 6. Comparison of Thai banks credit and credit risk management processes before and after the crisis
Board
Before Before crisis crisis Today Today

Board Risk Management Committee Credit Policy

Roles and Responsibilities Customer acquisition Credit review and approval Credit Policy Headquarters Branches - Analyze and review credit by themselves - Relationshipbased pricing Operation after credit authorization

Roles and Responsibilities

Customer acquisition Credit review and approval

Relationship managers Sales team/branches

Customer information

Headquarters - Analyze credit using tools such as credit rating and credit scoring - Credit reviewing committee decides on large loans

Operation after credit authorization

- Formula-based pricing with RAROC Headquarters/ Regional centers

Source: Triratvorakul and Vacharachaisurapol, 2006

After the crisis, most banks adopted a system where the main role of branches is to find and maintain customers, leaving credit review and approval to the headquarters. This creates transparency in the credit process as well as the system of check and balance and specialization among different departments. In all banks, there is now a risk management committee that sets credit risk policy and monitor banks 12

own risks. In several private banks, modern risk management tools such as credit scoring, internal rating systems, and credit portfolio models have been adopted. There is also an attempt to move towards risk-based pricing using RAROC (RiskAdjusted Return on Capital) or similar tools although success in this area has been quite limited. These changes in Thai banks credit and credit risk management

processes are summarized in Figure 6. An important caveat is that while the right panel of Figure 6 looks very much like what one would have found in the case of any international best-practice bank, the devil lies in the detail. For example, internal rating models of the large private banks currently cover only a portion of their corporate loan portfolios. Nevertheless this does not present as much a problem as the fact that certain banks lag significantly behind in the adoption of modern credit risk management tools. The difference between the good and the bad risk management will be most apparent during the downturn of the business cycle.

2.2 The environment As with the Thai commercial banking players, the environment in which they operate continues to evolve. This subsection divides Thai banks operating

environment into macroeconomic environment, industry or competitive environment, and regulatory environment. In contrast to the preceding subsection which for the most part was backward looking, this subsection focuses on the changing environment over the next 2-3 years along with their implications for Thai commercial banks. a. Macroeconomic environment After registering robust growth from 2002 to 2004, the Thai economy slowed down moderately in 2005 and 2006. 2006, in particular, was a challenging year for the Thai economy as it was buffeted by a number of negative shocks. These included political uncertainties, high oil prices, and severe flooding in many provinces. As a result, domestic demand softened markedly. The economy however managed a

respectable growth rate of 5% for the year with the help of robust exports on the back of a strong global economy.

13

Figure 7. Thailands GDP real growth forecast

Source: Bank of Thailand Inflation Report, April 2007

Figure 7 shows the medium-term forecast of real GDP growth of the Thai economy published by the Bank of Thailand in April 2007. The figure suggests an economy on a moderate growth path. The growth is expected to come mainly from exports, given healthy global conditions, additional government spending, and lower oil prices. Meanwhile, both internal and external stability appears well under control. Low unemployment and inflation rates are expected to continue in the medium term. On the external front, the current account is expected to be in surplus at least till 2008, which is a sharp contrast with the pre-1997 situation. International reserves have risen to over 70 billion US dollars, keeping the ratio of reserve to short-term debt in a highly comfortable position. Thus, the economy looks to be in a sustainable expansion, albeit not as fast as those recorded during 2002-2004. There is virtually no sign of a recurrence of the crisis of 1997 type. Of course, there are downside risks to growth, most notably if the world economy tanks. Although unlikely, this could easily happen if the US dollar collapses or the Chinese economy goes into a hard landing. The Thai economy is particularly vulnerable to these adverse events because of its heavy reliance on exports as driver of growth. But even if the Thai economy does proceed along the projected trajectory in Figure 7, things will likely be less rosy for the commercial banks. This is because private consumption and private investment, the major drivers of commercial banks

14

revenues, are not expected to show a significant pickup until the second half of 2008. In fact, there is a possibility that 2006 could be a short-term peak in Thai banks performance. Against this backdrop, effective credit risk management will be

essential for banks to maintain a healthy bottom line going forward. b. Industry environment Jittamai, Nakornthab, and Poshyananda (2005) describe in detail four major changes in Thai commercial banks industry environment that have taken place after 1997. The four changes are disintermediation by the capital markets, the increased role of specialized financial institutions (SFIs), the rise of non-bank financial institutions, and heightened competition from foreign banks. This subsection reviews each of the changes briefly along with their implications for Thai commercial banks going forward. Figure 8. Composition of Thai financial system
26% 3% 2% 19% 6% 69% 44% 31%

1996

2006

Financial institution loans Public-sector bonds

Corporate bonds Stock market

Note: (1) Financial institution loans = loans extended by commercial banks, finance and credit foncier companies, and specialized financial institutions (2) Outstanding domestic bond value at par (3) Stock market capitalization Source: BOT website; Thai Bond Market Association

The first and perhaps also the most important element of Thai banks industry environment is the structure of the financial system. Like those in most other Asian economies, Thailands financial system is bank-based, with commercial banks being the major providers of funds for the domestic economy. Nevertheless, since the 1997 crisis there has been a gradual shift towards greater capital market financing. Figure 8 shows the structure of the Thai financial system at the end of 1996 and 2006,

15

respectively. Although not a perfect proxy7, it is quite obvious that the Thai financial structure today is becoming more balanced. This trend is expected to continue in the immediate future. The decline in the significance of financial institutions in the Thai economy seen in Figure 8 is due partly to the problems facing these institutions in the aftermath of the 1997 crisis and partly to the authorities effort to promote the development of capital markets. In the case of commercial banks, a huge pile of NPLs and balance sheet weakness had led to a contraction in bank loan supply, making capital market an attractive financing alternative, or in some cases the only choice, for certain firms. One important lesson Thailand has learned a hard way from the 1997 crisis is the need to have a developed capital market that can act as a spare tire when the banking sector fails to perform its intermediary function. It is therefore no surprise that capital market has received a high priority in the Thai authorities effort on financial system reform. Three capital market development plans were launched in 2000, 2003 and 2006. The latest, Thailands Capital Market Development Master Plan 2006-2010, targets the corporate bond market as well as equity. Perhaps one of the most aggressive measures in this plan is a liberalization of brokerage commission fees in 2010 as a preparation for the full liberalization of the entire securities industry. Several observers see the growth of the equity and bond market as a major threat for Thai commercial banks. This paper takes a different view and argues that the net effect of capital market growth is positive for Thai commercial banks. While disintermediation inevitably hurts lending, capital market development provides banks with several income-generating opportunities. The most obvious is the bond market where Thai banks are the major players. League tables published by the Thai Bond Market Association shows Thai banks consistently dominate the lists of top dealers, underwriters, and registrars of domestic bonds. In addition, Thai banks are also occasional issuers of corporate bonds. All of these mean that Thai banks stand to gain handsomely as the bond market grows.

Though often used to describe financial structure, market capitalization may overstate the importance of the stock market in the economy. A notable example is Thailand in 1993 where stock market capitalization exceeded bank loans and led Demirguc-Kunt and Levine (1999). to wrongly conclude that Thailand had a market-based financial structure.

16

Although Thai banks are prohibited to directly participate in equity securities business, many own securities and asset management companies. As a matter of fact, securities and asset management company subsidiaries of the large private banks are among leaders in their respective industries. Thus, on a consolidated perspective, disintermediation is more like a shift in revenues from interest income to fees and services income. Of course, such capital market benefits depend on how universal a particular bank is. Additionally, because all but one Thai banks are listed companies, equity market growth offers them a number of indirect financial benefits. First, as the Thai stock market grows, the cost of equity for listed firms is likely to fall. Second, a healthy stock market enables them to raise more capital easily. And third, they could in principle use their market capitalization in merger and acquisition activities. One promising area that cannot be overlooked is wealth management. Globally, wealth management for high-net worth customers is the fastest growing business segment for banks. We have not seen this happening in Thailand yet. But it is conceivable that it is just a matter of time for the wealth management industry to take off if the Thai capital market continues to deepen. Lastly, a developed capital market offers two important indirect benefits for Thai commercial banks. First, a deep capital markets promotes overall financial system development which is good for economic growth. See, among others,

Demirguc-Kunt and Levine (2001). Second, capital market development contributes towards a more stable financial system and an economy which is more resilient in the presence of shocks. Figure 9 breaks down financial institution loans at the end of 1996 and 2006 into loans from commercial banks, loans from finance companies and credit foncier companies, and loans from special financial institutions (SFIs). Credit foncier SFIs are

companies are finance companies that specialize in property lending.

government-owned development financial institutions that provide financial assistance to specific sectors of the economy. Five major SFIs are Government Housing Bank, Government Savings Bank, the Bank for Agriculture and Agricultural Cooperative, the Export-Import Bank of Thailand, and SME Bank.

17

Figure 9. Breakdown of financial institution loans

8% 21% 22% 1%

78% 70%

1996

2006

Commercial banks

Finance and credit foncier cos

SFIs

Source: BOT website

After the 1997 crisis, the role of SFIs in the Thai financial system increased significantly. The government has used SFIs to fill the gap voided by troubled banks and finance companies as well as to stimulate the battered economy, especially at the grass root level. Compared to commercial banks, SFIs are more lax in lending discipline. Part of this is due to the so-called policy-directed lending, but primitive credit risk management capabilities also play a major role. For the most part,

however, this does not present a problem to commercial banks as many of SFIs customers (mostly the poor and farmers) do not belong to their target groups. A major exception is SMEs. Weak lending discipline by SFIs effectively distorts the market and represents one barrier for commercial banks to implement credit risk management for their SME customers. Not included in Figure 9 is the amount of loans extended by non-banks or more precisely non-deposit-taking consumer finance specialists. These players

include global names such as Aeon, American Express, Cetelem, and Citiloans, as well as local ones. Non-banks operations cover primarily three businesses: hirepurchase, credit cards, and personal loans. An average customer of non-banks has lower income than that of commercial banks, but there exists a material overlap between their customer portfolios. Before 2002, non-banks were not regulated. The Bank of Thailand viewed regulation of these companies unnecessary as they do not pose systematic risk to depositors. As the Thai economy recovered however, the influence of non-banks in the Thai consumer finance market grew significantly. Concerning that their explosive

18

growth might have led to financial instability (in the case of the credit card business) and excessive buildup of household debt, the BOT issued a series of prudential guidelines on credit card loans in 2002, 2004, and 2006 and took personal loan business under its supervision in July 2005 on the ground of consumer protection. Among the regulations which apply to banks as well as non-banks are minimum income of cardholders, ceiling on interest and other charges, and a maximum credit limit for a particular borrower. Despite the increased regulatory burden, non-banks continue to expand their influence. Figure 10 shows outstanding amounts of credit card loans and personal loans extended by banks and non-banks from the first quarters that non-bank data in respective markets are available. As of 2007Q1, non-banks accounted for 46% of credit card loans outstanding (up from 40% in 2002Q4) and 53% in personal loans outstanding (up from 51% in 2005Q2). Figure 10. Outstanding values of credit card and personal loans under BOT supervision
Credit card loans
Billion baht
100 Banks 80 60 40 20 0 Non-banks

Personal loans
Billion baht
100 80 60 40 20 0 05 Q2 05 Q3 05 Q4 06 Q1 06 Q2 06 Q3 06 Q4 07 Q1 Banks Non-banks

02Q4

03Q1

03Q2

03Q3

03Q4

04Q1

04Q2

04Q3

04Q4

05Q1

05Q2

05Q3

05Q4

06Q1

06Q2

06Q3

06Q4

Source: BOT website (credit card loans); BOT internal data base (personal loans)

The BOT internal data shows that, in addition to faster growth, non-banks as a group has a lower NPL rate in personal loans than commercial banks. The difference in NPL rates also hold in the case of credit card loans (Vanikkul, 2006). This is an interesting statistics in itself because low-income customers are the most sensitive to deteriorating economic condition. While faster growth and a lower NPL rate do not necessary correspond to superiority, there are reasons to believe that non-banks can take on more risks than commercial banks. After all, several of these firms are global players who have honed their skills and winning strategies in other markets. The incumbent Thai banks, on the other hand, are relatively new to this market. It was non-banks that first found opportunities in this market segment and sparked the 19

07Q1

subsequent lending boom. Perhaps the best example of non-banks differing strategy is their treatment of loans that miss payments. While non-banks have a higher

proportion of delinquency loans than commercial banks which reflects the greater riskiness of their portfolios, they do not allow these loans to lapse more than three months to be counted as NPLs. Their aggressive write-off and debt collection policies along with high interest rates and fees allow them to go for high-risk customers. Going forward, competition in the consumer finance market is expected to become fiercer as the softened economy slows down the now crowded market. Some banks, however, have managed to avoid direct competition with non-banks in this market by launching their own non-banks or by forming strategic alliances with nonbanks themselves. Others will have to find ways to respond to increased competition. The final element of Thai banks industry environment is heightened competition from foreign banks. Although it was a foreign bank which jumpstarted commercial banking in Thailand in 1888, foreign banks have played limited roles in Thailands modern banking era. This is in sharp contrast to the life insurance industry where American International Assurance controls roughly 45% of the markets total premium. The main reason for the limited presence of foreign banks in Thailand has to do with the Bank of Thailands regulations on foreign bank entry and branching. Today, with exceptions of Standard Chartered (Thai) and UOB (Thai) and a small foreign bank registered as a foreign subsidiary, all other foreign banks are confined to having one branch only. But even foreign wholly-owned Thai banks are a relatively new phenomenon. In an effort to stabilize the Thai banking system after the 1997 crisis, the Bank of Thailand relaxed the 25%-ceiling on foreign ownership in Thai banks. In this

process, four commercial banks were sold during 1998 and 1999 to Standard Chartered, UOB, DBS (merged their acquisition into TMB bank in 2004), and ABN Amro (sold their acquisition to UOB in 2005). More recently, the Bank of Thailand allowed GE Capital to take a majority stake in Bank of Ayudhya in 2006 and sold its FIDF holding in Bank Thai to TPG Newbridge in 2007. Another FIDF deal is also said to be forthcoming. Meanwhile, a few foreign bank branches have opted for organic growth during the past ten years. Most notable are Citibank and HSBC which have also built an

20

impressive retail operation, particularly in credit cards, despite having only one branch each. Although business models of foreign bank branches are diverse, most of them focus on corporate banking. Lending to large and medium-sized firms, structured products, and foreign exchange and money market activities are typical businesses of foreign bank branches. Figure 11 illustrates the different compositions of Thai banks and foreign bank branches loan portfolios. It is clear from the figure that the loan portfolio of foreign bank branches heavily tilts towards corporate lending. The composition

becomes even more skewed if we exclude the two foreign retail giants, Citibank and HSBC, from the picture. Figure 11. Loan portfolio compositions of Thai bank and foreign bank branches as of 2007Q1
0.2% 13% 9% 10%
Corporate loans Credit card and personal loans Housing loans (mortgages)

89% 78%

Thai banks

Foreign bank branches

Source: BOT website

Increased activities of foreign wholly-owned banks and foreign bank branches are among the major factors that have catalyzed Thai banks changing business model described in Section 2.1c. Nevertheless, because of their small sizes, foreign banks are still marginal competitors of many Thai banks, except in certain products such as credit cards and cash management. This competitive balance could change in the future however if there is a change in regulatory regime that allows them to grow at a faster rate than the current one, for which we turn to the next subsection.

21

c. Regulatory environment The 1997 crisis would not have inflicted such a deep wound on the Thai commercial banking sector if the regulatory environment that will prevail in the next two years had been in place then.8 This subsection describes the core elements of the new regulatory environment, namely, the regulatory and the supervisory frameworks that have already been or will be put in place by the end of 2008 along with the supporting infrastructure. An important milestone in Thailands post-crisis banking regulatory reform was the initiation of the first phase of the Financial Sector Master Plan (FSMP) in January 2004. Bank of Thailand (2006) provides background, rationales, and details of the plan. The FSMP is a 10-year, medium-term development plan for financial institution under the supervision of the Bank of Thailand. The most visible outcome under the first phase (2004-2007) of the FSMP is the rationalization of the structure and roles of financial institutions. A major structural weakness of the Thai financial institutions system before the 1997 crisis had been the presence of many high-risk finance (and credit foncier) companies. In some way, finance companies risks were forced on them by the regulatory system that did not allow them to raise funds in the form of deposits. Faced with higher costs, these companies naturally resorted to high risk activities in order to generate higher returns. One of the strategic objectives of the FSMP is to correct this structural weakness. This has been accomplished by the removal of the regulatory boundary line between commercial banks and finance companies businesses. Finance companies were then asked to merge with their parent bands, merge with other finance companies to become commercial banks or retail banks, or return their licenses. Another major structural weakness of the pre-crisis regime was the International Banking Facilities (IBFs), mentioned in Section 2.1a. It was IBFs that fueled the pre-crisis lending boom in Thailand. IBFs attached to commercial bank were asked to merge with their parents while stand-alone IBFs had to upgrade to foreign full branch or subsidiary or to return their license. By March 2006, there are no IBFs left in Thailand.

It would probably not be able to prevent the crisis from happening, for there were also other root causes in play.

22

The establishment of foreign subsidiary is a new concept in Thailand and represents the Bank of Thailands response to pressures for further financial liberalization. Under the FSMP, foreign banks can operate as a full branch or a subsidiary. Both full branch and subsidiary enjoy the same scope of business as Thai commercial banks, but the latter can have four branches in addition to one head office as opposed to the former which (by definition) has only one branch. The Bank of Thailand initially hoped that major foreign players such as Citibank and HSBC might have chosen to become a subsidiary. In the end, however, only Taiwans Mega International Commercial Bank (formerly ICBC) did so. Interviews with Citibank and HSBC conducted by Jittamai, Nakornthab, and Poshyananda (2005) revealed that the two banks would instead like to have 20-40 branches to be cost effective. The Bank of Thailand has now moved to work on the second phase of the FSMP which will further step up the pace of reform. The FSMP Phase II will aim at increasing efficiency of the financial institution system so as to strengthen its competitiveness and resiliency. An important element of the FSMP Phase II will be the orderly but meaningful introduction of more competition into the system between existing institutions and new entrants, domestic and/or foreign. The details of Phase II are expected to be finalized in the first quarter of 2008. Another notable development on the regulatory front is the adoption of new regulatory framework in line with the changing international standards. These

include, IAS39 (2006-2007), consolidated supervision (2007) and the Basel II Capital Accord (yearend 2008). IAS39 (International Accounting Standard 39, Financial Instruments: Recognition and Measurement) concerns several accounting areas, but the one that has affected Thai banks the most is the provisioning of NPLs. The required

provisioning amount under IAS39 is based on the difference between the outstanding loans and the present value of their collaterals as opposed to the difference between the outstanding loans and the appraisal value of their collaterals under the old BOT rule. To ensure a smooth adoption of IAS39, the Bank of Thailand has phased IAS39 compliance in three periods according to the age of the NPLs: doubtful of loss

23

loans (loans overdue more than 12 months) by December 2006, doubtful loans (loans overdue 6-12 months) by June 2007, and substandard loans (loan overdue 3-6 months) by December 1995). Nevertheless, the impact of IAS39 on banks net profit was still quite substantial in 2006, as seen in Section 2.1b. It is expected that the banking sector as a whole has to set aside 40-50 billion baht of additional provisions in 2007 compared to 62 billion last year (Kasikorn Research Center, 2007a). After 2007 however, IAS 39 will improve the transparency and provision buffer against impaired assets, significantly strengthening the banking sector. With consolidated provision, the Bank of Thailand will be able to supervise financial institutions on a consolidated basis, which has become increasingly necessary in the era of financial conglomerates. Consolidated supervision will allow the Bank of Thailand to better assess risks of banks more effectively. The adoption of Basel II slated for end-2008 will raise the standard of risk management of commercial banks. At the minimum, banks will have to hold

regulatory capital that better reflects the underlying risks of their business. The end goal is however for banks to develop risk management culture and capability that together allow them to manage their risks effectively. The Bank of Thailand has been working with the banking sector on the implementation of Basel II in Thailand for several years. It is expected that the transition to the new capital regime will be relatively smooth. On the supervisory front, the Bank of Thailand has significantly strengthened its supervisory capacity and monitoring practices. Risk-based supervision has been adopted in place of rule-based supervision. A range of database and risk-management system has been developed along with training programs for supervisors. This

includes the use of scenario analysis, an early warning system, and the publication of quarterly macro-prudential indicators. Finally, the Financial Institution Business Act (FIBA), expected to be passed this year, will strengthen the Bank of Thailands supervisory power by unifying the supervision of all types of financial institutions under its purview. The FIBA will also give the Bank legal power to enforce consolidated supervision, Basel II, and prompt corrective actions to address weakness in troubled banks.

24

In terms of supporting infrastructure, three are particularly noteworthy. The first is the legal process with regards to bankruptcy cases. The Bankruptcy Act has been amended several times since 1998 to deal with NPL cases. The Central

Bankruptcy Court was established in 1999. The overall process from initial document to the end of foreclosure sale has been shortened from 42-60 months to 26-35 months (Kasikorn Research Center, 2007c). The second important supporting infrastructure is the Deposit Protection Agency (DPA) Act, expected to be passed this year, will replace the current blanket deposit guarantee in place since 1997. The new system of limited-coverage deposit insurance will be phased in over four years to allow banks and depositors to adjust to the new framework. The establishment of the DPA should reduce banks moral hazard and introduce more market discipline to the banking sector by putting pressure on banks to improve financial strengths if they want to attract and retain depositors. The third is the National Credit Bureau (NCB), formed from a merger of two credit bureaus in 2005. The NCB collects and warehouses debt and debt service records provided by its member financial institutions. The NCBs database now covers more than 10 million customers. Thus, the problem of the lack of information about loans granting to one borrower by several institutions has been significantly reduced. For example, in the area of housing loans, banks now routinely use

information from the NCB to ensure that a prospective customer does not have excessive debt burden and also to prevent occurrences of double or triple mortgages.

3. Short-term risks and outlook

The Basel II Capital Accord specifies three major risk buckets of financial institutions: credit risk, market risk, and operational risk. This section focuses primarily on credit risk, by far the largest risk exposure of all Thai commercial banks. Market risk in particular is small, as suggested by the ratio of market risk assets to total risk assets. Excluding one bank with a sizable market risk exposure, the average proportion of market risk assets to total risk assets for all other Thai commercial banks at the end of 2006 was a mere one percent. Before jumping to the credit risk issues of Thai commercial banks, it is worth noting about Thai banks foreign exchange risks, however. One of the major

misconceptions by Thailands outsiders is that the 1997 crisis resulted partly from 25

currency mismatches on banks balance sheets. For example, McKinsey & Company (2002) forcefully argues that Thai banks were vulnerable because they onlent foreign funds to their domestic borrowers without proper hedging of foreign exchange rate risks. The truth is, then as well as now, Thai banks were required by the Bank of Thailand to keep their net foreign exposure in check. Instead, what the Bank of Thailand and the banks themselves overlooked and is now a classic case study, is that the mismatches in the real sector could feed back to the banks in the form of increased credit risks. In the short-term, i.e., this year and the next, the most important factor affecting Thai banks risks and outlook is likely to be the not-so-favorable economic environment. As mentioned in Section 2.2a, the Thai economy into the end of 2008 is forecasted to be driven mainly by exports, with domestic demand particularly private investment below trend growth. In such an environment, the pressure for shareholders value creation will lead to intensifying competition among market players. The combination of the not-so-robust economy and increased competition presents an important challenge for all Thai commercial banks. Week domestic demand affects banks in two major ways. First, it exerts a downward pressure on loan growth. Second, it is likely to lead to more default occurrences. Figure 12 plots year-on-year growth rates of total loans, corporate loans, and consumer loans of Thai commercial banks against nominal GDP growth from 2004Q4 to 2007Q1. Over the period shown, Thai banks loan growth moved roughly in tandem with nominal GDP growth, falling as the economy softened since 2006Q1. Corporate loan growth, in particular, has slowed down markedly. As a matter fact, the last time that corporate loans of Thai banks registered a negative growth was in 2003.9 On the other hand, consumer loan growth appeared little affected by the economic slowdown. The robust growth in consumer loans partially offset the fall in total loan growth resulting from corporate loans.

Discontinuity in sectoral loan data in 2003Q4 prevents Figure 4 to goes back beyond the period shown. Nevertheless, that the last time corporate loan registered a negative growth was in 2003 can be inferred from the available data.

26

Figure 12. Loan growth of commercial banks registered in Thailand versus nominal GDP growth, 2004Q4-2007Q1
30% 25% 20% 15% 10% 5% 0% 04Q4 -5%

05Q1 All loans

05Q2

05Q3

05Q4

06Q1

06Q2

06Q3

06Q4

07Q1

Corporate loans

Consumer loans

Nominal GDP growth

Source: BOT website

Given the economic outlook and the contemporaneous movement between nominal GDP and total loans observed in Figure 12, Thai banks overall loan growth in 2007 will likely be lower than that in 2006 and may not pick up much in 2008. The implication from this is a downward pressure on banks interest income going forward. It should be noted that movements in bank loan growth are not entirely demand driven. Banks also watch over their credit risks. As economic uncertainty mounts, the probability that potential borrowers may not be able to repay their debt increases. So banks may keep tap on new lending when their perceived risks of borrowers increase. An important indicator of the quality of banks loan books is NPLs. After reaching a low of 8.0% at the end of 2006, Thai banks NPLs rose slightly to 8.1% in 2007Q1. Net NPLs also rose from 4.5% to 4.6%. While banks NPL rates capture their overall loan portfolio quality, they are generally not a good indicator of credit risks facing banks. In theory, one would like to use the expected default rate of new loans for that purpose. In the absence of public data, one rough proxy of bank credit risks would be the ratio of new NPLs to total loans.10,11

10

In a forthcoming study, Chantapant, Kritayakirana, and Nakornthab (2007) use a database available only to supervisors to construct historical default rates of bank corporate borrowers. 11 One may be tempted to use the change in total loans outstanding in the denominator. There are at least two problems with this. First, the amount of new loans in a given period is the change in loan outstanding plus the sum of loan repayments, write-offs, and transfers. In a period when total loans outstanding is declining (the case of

27

Figure 13 plots the ratios of new NPLs and reentry NPLs (NPLs of loans that have been previously restructured) to total loans of Thai commercial banks going back to 1999Q4 when the data became available. The sum of the two ratios captures roughly the default rates of Thai banks loan portfolios. Figure 13. Ratios of new NPLs and reentry NPLs to total loans of commercial banks registered in Thailand, 1999Q4-2007Q1
1.5% New NPLs 1.2% Reentry NPLs

0.9%

0.6%

0.3%

0.0% 99Q4

00Q4

01Q4

02Q4

03Q4

04Q4

05Q4

06Q4

Source: BOT website

Excluding the temporary increase in reentry NPLs in the first half of 2003, the credit environment facing banks appeared benign throughout the period shown. Nevertheless, since the end of 2005, there seems to be an upward shift in the level new NPL rates compared to the earlier period. The upward drift in the new NPL rate is in line with Figure 14 which shows the absolute and the relative amounts of special-mention loans of a subset of Thai commercial banks at the end of 2004, 2005, 2006, and 2007Q1. Special-mention loans are loans that are past due for more than one month, but less than three months. So they are not counted as NPLs, which by the BOTs aging criteria refers to loans that have been past due for more than three months.

Thai banks between 1998 and 2001), using changes in total loans instead of total loans as the denominator would result in negative NPL rates. Second, NPLs in the respective period may have come from loans that have been originated several periods ago.

28

Figure 14. Special mention loans of Thai commercial banks

Million baht 150,000 Special mention loans
120,000 90,000 1.5% 60,000 1.0% 30,000 0 2004 2005 2006 2007Q1 0.5% 0.0%

Percent of total loans

3.0% 2.5% 2.0%

Source: Kasikorn Research Center (2007)

Changes in special mention loans are generally indicative of future changes in new and reentry NPLs. Thus, together with the softened economic condition, Figure 13 suggests that more new and reentry NPLs are in the pipeline this year. Despite potential increases in new and reentry NPLs, the overall NPL ratio of Thai banks is expected to fall significantly this year. As mentioned in Section 2.2c, Thai banks are required to set aside additional provisions in compliant with IAS39 this year. To reduce provisioning burden, banks are expected to transfer a substantial amount of their NPLs to asset management companies (AMCs). Estimates by

Kasikorn Research Center (2007b) put the ratios of NPLs and net NPLs to total loans between 4.0-5.5% and around 2% at the end of this year, respectively. A major caveat is that, for banks with their own AMCs (five largest private banks do), such NPL transfers merely represent a shift from a direct exposure on their loan books to an indirect exposure through their subsidiaries. To the extent that their AMCs make losses from the transferred loans, they will have to realize these losses too. An efficiency issue aside, a bank with 20 billion baht of NPLs is no different from a clean bank that owns an AMC (99.99% or 100%-owned) with 20 billion baht of NPLs. Good analysts usually look at consolidated NPL figures when assessing a banks capital vulnerabilities. To gain a little deeper understanding of Thai banks credit risk, we look separately at Thai banks corporate sector exposure and household sector exposure. Figure 15 shows time series plots of selected financial ratios of non-financial listed companies, often taken as a proxy of banks top-tier corporate customers. The figure reveals a very comfortable picture. Although the gross profit margin of non-financial 29

listed companies has been somewhat depressed recently, the other three ratios point out to exceptional financial strengths, particularly in debt service capacity. Moreover, since the crisis, many Thai private companies, including listed ones, have shied away from foreign currency funding. Accordingly, the risk of currency mismatches turning these firms into defaults like what happened during 1997-8 is remote. All these

backdrops suggest minimal credit risk coming from large corporate borrowers, at least in the immediate horizon. Figure 15. Selected financial ratios of non-financial listed companies
Percent
12 10 8 6 4 2 0
Q2 Q3 Q4 Q2 Q3 Q1 2005 Q1 2006 Q4

Return on Assets
7.1

Time
5 4 3

Debt to Equity Ratio

Average 1994-1996
Q1 2007 1999 2000 2001 2002 2003 2004

2 1 0

Average 1994-1996
1.0
Q2 Q3 Q4 Q2 Q3 Q1 2005 Q1 2006 Q4 Q1 2007 1999 2000 2001 2002 2003 2004

-2 -4 -6

10

Time
8 6 4 2 0

Interest coverage ratio

7.0

Percent
30 25 20

Gross Profit Margin

Average 1994-1996
18.4

Average 1994-1996

15 10 5 0 Q2 Q3 Q4 Q2 Q3 1999 2000 2001 2002 2003 2004 Q1 2005 Q1 2006 Q4 Q1 2007

Q2 Q3 Q4 Q2 Q3

Q1 2005

Q1 2006

Q4

Source: SET; compiled by BOTs Monetary Policy Group

The lack of data on non-listed firms makes it difficult to draw similar inferences in the case of smaller-sized corporate borrowers. Nevertheless, we know that these firms, particularly the SMEs, are in general more sensitive to adverse economic condition than their larger peers. This, together with the fact that SMEs have been a focus of Thai banks loan expansion in recent years, makes middlemarket borrowers a potential source of corporate-sector credit risks for Thai banks. At the same time, there are indications that household sector credit risks have increased. The pace of growth in consumer loans during the past four to five years has led to an increasing exposure of Thai banks to the household sector. While the strength of household lending has been a boon for banks, a spate of defaults by households would turn the situation around.

Q1 2007

1999

2000

2001

2002

2003

2004

30

Figure 16 traces the evolution of NPL ratios of Thai banks consumer loans and their subcategories over the past four years. From 2003Q4 to 2006Q4, NPL ratio of consumer loans came down significantly, reflecting an improving of banks consumer loan portfolios. The situation is different in 2007Q1 when it went up slightly on the back of increases in NPLs in the area of credit cards and housing loans.

Figure 16. NPLs of consumer loans, 2003Q4-2007Q1

% of total loans
20
Housing loans (mortgages) Consumer loans

16
Loans for personal consumption

Credit card loans

12

Personal loans

0 2003Q4 Consumer loans

2004Q4 Housing loans

2005Q4 Credit card loans

2006Q4 Personal loans

Source: BOT internal database

Although it is too early to tell that this is a start of a new trend, given the state of the economy, NPL ratios of consumer loans could rise further as the year progresses. Nevertheless, because consumer loans are among the sectors with the lowest NPL ratios, it is still a long way for their increases to pose threat to banks stability. In addition, due to the Bank of Thailands prudential regulative on loans to value (LTV) of mortgage loans, banks are more or less covered in the event of borrower default. In fact, it would require a significant financial distress in the household sector before banks potential losses from their consumer loan exposures would pose a concern for financial stability. Karnchanasai, Nakornthab, and Piamchol (2004) highlighted net properties foreclosed on banks balance sheets as a potential source of additional risks for Thai banks in the event of a property price collapse. Figure 17 plots the amount of net foreclosed assets on Thai bank balance sheets and their percent of total assets from 1997:1 to 2007:4. Since mid 1999, Thai banks properties foreclosed have ballooned. Most of these properties foreclosed were previously collateral of loans that had turned sour. While their ratio to total assets has been trending since the beginning of 2006, it

31

remains well above 2%. For comparison, the corresponding figure for foreign bank branches is 0.02%. Although a sharp decline in property prices look unlikely in the immediate horizon, having such sizeable non-income-generating and highly illiquid assets on balance sheets does entail hidden financial costs for Thai banks. Figure 17. Net properties foreclosed of commercial banks registered in Thailand, 1997:1-2007:4
200,000 180,000 160,000 140,000 120,000 100,000 80,000 60,000 40,000 20,000 0.00% 0 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 Outstanding value Percent of total assets 1.00% 1.50% 2.50% 3.00%

2.00%

0.50%

Source: BOT website

In addition to slower loan growth, increased credit risks, and carrying costs of net properties foreclosed, continued strengths in deposit growth of most Thai commercial banks is another factor weighing down banks operating performance this year. With low loan growth prospect, banks have parked their excess liquidity in the money markets. Coupled with a series of policy rate reduction in the first half of this year, this exerts a further downward pressure on both NIM and asset yields. We have seen this already happened in 2007Q1. Adding IAS39 provision burden on top of that and Thai banks net profitability in 2007 does not look very good. Consistent with this outlook are analysts expectations of banks profitability this year and the next. Table 3 shows that analysts expect earnings per share of half of the Thai commercial banks in the table to fall against their 2006 values. All commercial banks are however expected to do better in 2008 when IAS39 provision is no longer an issue.

32

Table 3. Averaged EPS forecasts of selected Thai banks as of July 6, 2007

2006A 2007F 2008F Bangkok Bank 9.35 10.20 11.25 Kasikorn Bank 5.74 5.73 6.45 Siam Commercial Bank 4.64 5.10 6.01 TMB Bank -0.86 -0.17 0.21 Bank of Ayudhya 0.58 0.59 1.45 Thanachart Bank 0.15 0.19 0.22 Tisco Bank 1.88 2.17 2.52 Kiatnakin Bank 3.98 3.36 3.66 ACL Bank 0.94 0.44 0.48 Krungthai Bank 1.26 1.20 1.51 Siam City Bank 2.02 1.45 2.18 Bank Thai -3.32 0.40 0.62 Note: Excludes UOB (Thai) which was delisted in August 2006 and Standard Chartered (Thai) which is no longer traded. Source: SAA Consensus (www.settrade.com)

4. Conclusions
Thai commercial banks have come a long way since 1997. Aided by strong economic growth, consolidation and recapitalization efforts, improved risk management, operational restructuring, and regulatory reforms, the sector has more or less returned to stability by the end of 2006. On the surface, the Thai banking sector does not appear to be much different from its pre-crisis years. A deeper look at the players and their operating environment reveals a very different industry dynamics, however. Among other things, this

implies that the risks facing Thai commercial banks have also changed from what they were facing in 1996-1997. After enjoying a smooth journey for the past several years, Thai banks have hit a bump in the road in 2007. Most important from the financial stability perspective is the thinner bottom line and increased credit risks. Nevertheless, there is no sign of sector-wide distress. All banks are also expected to do better in 2008. Of course, the outlook for Thai banks could change if the future macroeconomic and financial environment significantly diverges from its expected course. Looking forward, the major test for Thai banks will the introduction of greater competition under the second phase of the Financial Sector Master Plan. While the timing, the extent, and the speed of the planned liberalization will not be finalized until 2008, it is not difficult to see from the assessment in this paper that some banks are far more vulnerable than the others to the increased competitive pressure. These

33

banks will have to work harder than the rest to find ways to ensure their long-term place in the Thai economy.

References
Bank of Thailand (2006). Thailands Financial Master Plan Handbook, August. Bank of Thailand (2007). Inflation Report, April. Chantapant, Sukonpat, Kritayakirana, Krongkaeo, and Don Nakornthab (2007). How Vulnerable Are Thai Banks: A Structural Analysis of Bank Corporate Portfolios and Implications, forthcoming. Demirguc-Kunt, Asli and Ross E. Levine (1999). "Bank-Based and Market-Based Financial Systems: Cross-Country Comparisons," World Bank Policy Working Paper No. 2143, July. European Central Bank (2006). EU Banking Sector Stability, November. Jittamai, Pajongjit, Nakornthab, Don, and Roong Poshyananda (2005). Changing Macroeconomic Environment and Financial Landscape: Challenged for the Banking Sector, Bank of Thailands Discussion Paper No. 10/2005, November (in Thai). Karnchanasai, Chatsurang, Nakornthab, Don, and Suchot Piamchol (2004). Bank Lending, the Housing Market, and Risks: A Test for Financial Fragility, Bank of Thailands Discussion Paper No. 05/2004, November. Kasikorn Research Center (2007a). Commercial Banking Business in 2007: Facing Several Negative Factors, Current Issues, Vol. 13, No. 1943, January (in Thai). Kasikorn Research Center (2007b). Bank Loan and NPL Trends during the Rest of 2007, Current Issues, Vol. 13, No. 1974, May (in Thai). Kasikorn Research Center (2007c). Thai Economy 10 Years after Crisis: Lessons for Thai Financial Institutions, Current Issues, Vol. 13, No. 1981, July (in Thai). Krungthai Bank (1997). Key Financial Data of Thai Commercial Bank, 1988-1996. McKinsey&Company (2002). Dangerous Markets: Managing in Financial Crisis, Wiley Finance. Santiprabhob, Veerathai (2003). Lessons Learned from Thailands Experience with Financial Sector Restructuring, Thailand Development Research Institute, November.

34

Vanikkul, Krirk (2006). Bank of Thailands Policies towards Development of the Financial Institution System, paper presented at TDRI 2006 Yearend Conference, December (in Thai). Vichyanond, Pakorn (2007). Crucial Transition in Thailands Financial System, paper presented at the Brookings Institutes Asian Economic Panel, April.

Appendix A. Implications for regional integration

This appendix looks at integration in banking services in Asia from a financial stability perspective. In particular, it argues that policymakers should follow proper sequencing of first ensuring the stability of the domestic banking sector and the integrity of the regulatory and supervision framework before moving to full regional (or international) integration. Where relevant, the findings in this paper are as cases in point. Integration in banking services generally takes two primary forms. The first is cross-border provision of services, be internet banking or cross-border borrowing and lending. The second is commercial presence, either through local subsidiaries or equity stakes. Different regions follow different approaches to integration. In Latin America, the proliferation of foreign banks is the hallmark of the regions internationalized banking sector. In contrast, intraregional cross-border bank flows are the main mode of integration within the European Union. As a matter of fact, looking only at percentage of total bank assets held by foreign banks would find little integration of the EU banking sector. With some exceptions1, national banks overwhelmingly

dominate the markets in EU economies. In Asia, integration in banking services in both modes, though increasing, is still quite limited and lags significantly behind integration in trade and investment.2 It is important to note that although the shares of foreign-owned banks in the banking system in Asia are comparable to those in Europe, the underlying causes are different. In Asia, it is basically the restrictions on foreign ownership. In Europe, it is strong domestic franchises that deter foreign participation even though foreign entry has practically been fully liberalized in all EU countries.

35

In the case of Thailand, it is rather clear that the preference of policymakers before the 1997 crisis was for cross-border bank flows. The Bangkok International Banking Facilities (BIBFs) were set up with the goal of making Bangkok the third financial center in Asia after Hong Kong and Singapore. The original architects of the BIBFs actually hoped for a balance between the importing of foreign funds for domestic purposes (Out-In transactions) and the re-exporting of these funds to other countries in the region (Out-Out transactions). As it turned out, Out-Out transactions were pale in comparison with Out-In transactions and the Thai economy was flooded with foreign funds. Together with inconsistent macroeconomic policies and an

inadequate regulatory and supervisory framework, this effectively sowed the seed for the 1997 crisis that devastated the economy and the banking sector. In retrospect, the Thai crisis illustrates the peril of hasty liberalization as well as the risks associated with cross-border bank flows. It also highlights the importance of an exchange rate regime in the integration process. In particular, a fixed or tightly managed exchange rate regime could give both domestic borrowers and their lenders a false sense of security with regards to the risks involved. The risks associated with cross-border bank flows extend to the giving end as well. Foreign-currency lending exposes banks to customers default risks (pure as well as exchange-rate-induced). Korea during the Asian crisis was a case in point. As a part of Koreas outward internationalization of the banking sector, Korean banks had a sizeable exposure in Indonesia. The collapse of the Indonesian economy resulted in substantial loss for Korean bank and was one of the most visible contagion linkages between the two economies. As Asia regains its vibrant dynamism, cross-border flows within and into the region has once again been on an increasing trend. A recent study by the IMF noted that cross-border bank flows which retrenched dramatically in 1997-8 have now made a come back with the main destinations being China, India, and the newly industrialized economies.3 Nevertheless, these flows have been dominated by

interregional rather than intra-regional ones. From a regional integration perspective, little has changed since the crisis on this front. Promotion of cross-border bank flows within the region is likely to run into several limits. First, there are not many firms that can directly access the international

36

syndication market. In the past, this problem is circumvented by having banks onlend funds to domestic borrowers. After the 1997 crisis, however, many Asian firms have shunned foreign-currency borrowings which limit growth opportunity in cross-border loans. Third, the market has to be divided between intraregional and interregional banks, for which the latter seems to have an edge. Given these limitations of cross-border bank flows, local commercial presence of foreign banks seems to be the way forward for Asian regional integration in banking services. Like cross-border flows which have the receiving end and the given end, commercial presence has two dimensions: outward internationalization and liberalization of foreign entry. Japanese banks exemplify outward internationalization. As we have seen in the paper, Japanese banks dominate the foreign bank scene in Thailand. Unfortunately, as in most Asian markets, Japanese banks mainly serve Japanese MNC customers and therefore are too focused to be called truly regional banks. In contrast, Singapore, a recent comer in outward internationalization of banking services, has a high inspiration to have its banks becoming regional powerhouses. The acquisitions of a majority stake in UOB (Thai) by UOB and a strategic equity stake in TMB Bank by DBS are examples of Singapores expanding regional reach. Most Asian countries have liberalized their outward FDIs, of which outward internationalization is one of the categories allowed. However, outward

internationalization can only do so much for regional integration in Asia so long as most countries still maintain restrictions on foreign entry. In the end then, it is the openness of Asian banking sectors to foreign entry that determine the extent of regional integration in banking services. Opening up the domestic banking system by allowing foreign commercial presence entails two major benefits. First, foreign entry brings fresh capital into the system. What happened in Thailand after the crisis is the testament of this benefit. Second and perhaps more important, foreign competition in general leads to higher efficiency, lower prices, and better services. See, for example, Unite and Sullivan (2003) in the case of the Philippines4 and Claessens et al. (2001) for a cross-country study.5 Nevertheless, as with any liberalization, the stake of ill-timed or poorlymanaged banking-sector liberalization is high. Most importantly, eroded economic

37

rents may result in riskier behaviors of domestic banks. If the regulatory environment is also weak, a banking crisis could happen. Given the above risk-benefit calculus, the key policy challenge for Asian central banks is to design a liberalization program that will help the country to reap maximum benefit of foreign entry at minimum risks. The Bank of Thailands current controlled approach to liberalization of foreign entry in the banking sector represents one way to do this. The authorities recognized from the beginning that the first priority is to get the frail banking sector back to stability. The first phase of the BOT Financial Sector Mast Plan thus concentrates on beefing up Thai commercial banks so as to prepare them for greater competition. At the same time, the Bank has devoted substantial resources to the upgrading of its risk-supervisory and monitoring system so as to be able to spot the pockets of vulnerability and to address weaknesses in banks more effectively. The FSMP Phase II will build on these achievements and proceed with gradual opening of the banking sector in a way that financial stability is maintained. As a final remark, for countries contemplating opening up their bank sectors, an assessment of the sectors stability can serve as an important input for consideration on how and when to proceed. To maximize its usefulness, such a review should also be conducted on a regular basis and in a forward-looking manner.

1 2

The exceptions are Ireland, Luxembourg, United Kingdom, and the former Eastern-bloc countries. Cowen, David, Sargado, Ranil, Shah, Hemant, Teo, Leslie, and Alessandro Zanello (2006). Financial Integration in Asia: Recent Developments and Next Steps, IMF Working Paper, WP/06/196, August 3 International Monetary Fund (2007), Regional Economic Outlook: Asia and the Pacific, April 2007. 4 Unite, Angelo, and Michael Sullivan (2003), The Effect of Foreign Entry and Ownership Structure on the Philippine Domestic Banking Market, Journal of Banking & Finance, Vol. 27, pp. 2323-2345. 5 Claessens, Stijn, Asli Demirguc-Kunt, and Harry Huizinga (2001), How Does Foreign Entry Affect Domestic Banking Markets?, Journal of Banking & Finance, Vol. 25, pp. 891-911.

38