Structural and cyclical drivers of Japanese mezzanine debt demand and supply

25 October 213 Structural and cyclical drivers of Japanese mezzanine debt demand and supply The impact of Basel III, combined with cyclical factors, on mezzanine financing opportunities Dr Hamish Macalister Head of Research, Maiora Asset Management, Research Fellow, Singapore Management University Bank loans have traditionally represented 85-9% of Japanese construction companies and real estate asset managers interest-bearing financing. However, loan supply to these sectors has been trending lower. Implementation of the Basel III accord may aggravate this by raising the opportunity costs associated with the allocation of bank capital, and resulting in greater difficulties for construction and real estate firms to obtain bank funding with higher loan-to-value [LTV] ratios. In addition, construction sector activity has been strengthening, firms require increased funding and firms have capacity for increased funding as a result of multi-year deleveraging of balance sheets. There is consequently evidence of a growing mismatch between loan supply and loan demand for Japanese construction and real estate firms. We believe this mismatch in turn results in attractive investment opportunities in Japanese mezzanine debt, offering significant yield premiums over senior debt, relative to investment risk. Introduction A combination of structural factors (with particular emphasis on Basel III implementation) and cyclical factors raise the prospect of a mismatch between loan supply and loan demand for mezzanine financing for Japanese construction projects and real estate funding. For the purposes of this paper we define mezzanine debt as subordinated debt with a LTV ratio in the range of 65-85% (more generally, mezzanine debt refers to any subordinated debt or preference share that is senior only to common equity). Basel III requirements likely increase both the cost and complexity for banks to provide mezzanine financing. New regulatory requirements also create a strong incentive for banks to de-lever existing positions, in turn creating demand for mezzanine financing as banks partition higher LTV ratio loans into a senior component and a subordinated component. Japanese banks have already been reducing lending to the construction and real estate sectors for many years, further tightening loan supply conditions.

Conversely, the construction industry is experiencing growth in activity and therefore requires increased loan supply. In addition, the financial position of Japanese construction and real estate firms has been improving and balance sheet gearing has been decreasing, signaling capacity for firms to increase borrowing. We believe these factors combined result in a mismatch between bank willingness to lend at higher LTV ratios and construction and real estate firm demand for funding. In turn, we believe this mismatch results in an attractive return premium over senior debt yields, relative to investment risk. The Maiora Asset Management team has identified a range of projects in the 65-85% LTV ratio range with expected returns in a range of 11-14%, compared with an expected senior debt yield of 1.5%. Maiora Asset Management has launched the MAIORA JAPAN STRUCTURED FINANCE FUND to invest in and manage these opportunities. In the remainder of this paper we provide an overview of Basel III requirements and implications for Japanese Banks, followed by an introduction to a selection of cyclical factors for loan supply and demand. Structural factors: A Basel III primer The Third Basel Accord provides a set of standards designed to mitigate banking industry risks, with particular emphasis on lowering the probability of bank failure and failure contagion in the wake of the global financial crisis of the late 2s. Basel III offers a more conservative approach than Basel I and Basel II with, for example, substantial increases in bank capital requirements and the introduction of asset liquidity requirements. It should be noted that the Basel Accords are voluntary, with countries regulatory authorities implementing the Accords through local regulation and in a manner best suited to that nation. Basel III requirements have a phasein period through to 219. While there are myriad details regarding the minutiae of standards in Basel III, which are beyond the scope of this paper, the key components can be summarised as a leverage ratio, enhanced capital requirements and minimums for asset liquidity. These are briefly outlined below (and summarised in Table 1, along with phase-in stages): Leverage ratio. Basel III sets a minimum ratio of Tier 1 capital to Table 1. Basel III key components and phase-in arrangements Dates as of 1 January Phases 213 214 215 216 217 218 219 Leverage ratio Parallel run 1 Jan 213-1 Jan 217. Disclosure starts 1 Jan 215 Migration to Pillar 1 Minimum common equity capital ratio 3.5% 4.% 4.5% 4.5% 4.5% 4.5% 4.5% Capital conservation buffer.625% 1.25% 1.875% 2.5% Countercyclical buffer.625% 1.25% 1.875% 2.5% Capital surcharge for systemically important institutions Minimum common equity plus capital conservation buffer.625% 1.25% 1.875% 2.5% 3.5% 4.% 4.5% 5.125% 5.75% 6.375% 7.% Phase-in deductions from CET1 2% 4% 6% 8% 1% 1% Minimum Tier 1 capital 4.5% 5.5% 6.% 6.% 6.% 6.% 6.% Minimum total capital 8.% 8.% 8.% 8.% 8.% 8.% 8.% Minimum total capital plus conservation buffer Capital instruments that no longer qualify as non-core tier 1 and tier 2 capital 8.% 8.% 8.% 8.625% 9.25% 9.875% 1.5% Phased out over 1 year horizon beginning 213 Liquidity coverage ratio minimum 6% 7% 8% 9% 1% Net stable funding ratio Introduce minimum standard Source: Basel Committee on Banking Supervision 2

total assets of 3% (the US Federal Reserve has announced it will set a higher minimum level for systemically important financial institutions). Capital requirements Basel III sets a minimum ratio of common equity to risk-weighted assets of 4.5%, to which is added a Capital Conservation buffer of 2.5% (simply, additional capital which if not available results in capital distribution restrictions), a Countercyclical Capital buffer of up to 2.5% (set by a country s regulatory authorities and designed to allow for macro-prudential management of macroeconomic cycle risk) and a capital surcharge for systemically important financial institutions of up to 2.5%. The minimum ratio of Tier 1 capital to riskweighted assets is 6% and the minimum ratio of total capital to risk-weighted assets is 8%. Combined, these requirements result in a possible minimum ratio of total capital to risk-weighted assets as high as 15.5%. It should also be noted that Tier 1 capital is more narrowly defined. Liquidity requirements Basel III introduces two liquidity requirements: a liquidity coverage ratio and a net stable funding ratio. These measures are designed to require bank funding to take into consideration both short -term outflow requirements and longer term asset/liability matching. The liquidity coverage ratio is calculated as the value of high quality liquid assets divided by total net liquidity outflows over the next 3 days, and is required to be greater than or equal to 1% (i.e. short-term outflows should be easily met by liquid asset holdings). The net stable funding ratio is defined as available stable funding over required stable funding (and is required to be 1% or higher). Basel III provides details on the definitions of these components and reporting requirements. The Japanese regulatory authorities intend to implement most aspects of Basel III a year head of each required implementation date. Currently the Japanese authorities only require internationally-active banks to meet Basel III requirements, but they have also introduced new requirements for domestic banks. Japan has received broadly positive assessments from the Bank of International Settlements regarding its compliance with Basel III requirements. Basel III and implications for mezzanine debt For our evaluation of the impact of the Basel III Accord on Japanese mezzanine debt, the more crucial components of the Accord are the new capital requirements and the new liquidity requirements. Mezzanine loans are unlikely to qualify as high quality liquid assets, with the liquidity aspect being most problematic. In addition, mezzanine loans are likely included in required stable funding. Hence, banks seeking to meet Basel III liquidity requirements will need other assets to cover short-term funding requirements and will also require stable funding for longer term assets (which include mezzanine loans, and these are included with a 1% weight in the denominator of the net stable funding ratio calculation). While this secondary factor is generally true of all loans, short-term liquidity requirements may cause some banks to regard new mezzanine loans with caution. The substantial increase in bank capital requirements may be particularly challenging for mezzanine borrowers looking for new funding. The Basel III Accord makes bank capital substantially more valuable, or at least raises the opportunity costs for capital usage. 1 Figure 1 illustrates the new capital requirements, their phasing in over time, and compares these with recent figures for Japanese bank capital adequacy. Data for Japanese bank capital ratios was obtained from the Bank of Japan. It indicates that Japanese banks are, on average, already meeting the higher Basel III capital requirements (although data for Japanese domestic banks is calculated based on Basel II requirements, so should be treated with some caution). For Japanese internationally-active banks the latest available capital adequacy ratio (calculated, according to the Bank of Japan, using Basel III requirements) is in excess of not only the total ratio including the capital conservation buffer, but also in excess of the total including the maximum surcharge for systemically important institutions and the maxi- 1 There are other requirements stipulated by Basel III which further raise the value of existing capital, beyond that indicated by the discussion of capital adequacy requirements presented here. For example, measures designed to manage counterparty credit risk including higher risk-weights on large financial institutions, capital charges for mark-to-market counterparty risk and capital charges for wrong way risk (when exposure to a counterparty is strongly positively correlated with the probability of default of that counterparty). 3

Figure 1. Basel III capital requirements, phase-in and current Japanese bank adequacy ratios 18% 16% 14% 12% 1% 8% 6% 4% 2% % 213 214 215 216 217 218 219 Countercyclical buffer Capital surcharge for systemically important institutions Capital conservation buffer Tier 2 capital Additional Tier 1 Capital Common equity Tier 1 Capital Japanese internationally active banks capital adequacy ratio (Basel III requirements), March 213 Japanese domestic banks capital adequacy ratio (Basel II requirements), March 213 Source: Basel Committee on Banking Supervision, Bank of Japan, Maiora Asset Management mum countercyclical buffer (helped by new capital raisings in 21/11). However, some care is required evaluating this data. The IMF Financial Sector Assessment Program team has estimated Tier 1 capital ratios could be up to 2 percentage points lower and total capital adequacy ratios up to 6 percentage points lower in Japan under full and immediate implementation of Basel III capital definitions. Ratios could be pushed lower still with the application of new international standards for risk-weighting of assets. The IMF also notes that capital adequacy problems may be more acute for second tier regional banks. Aosaki (213) highlights the risk of reduced bank lending in Japan to bolster capital adequacy ratios. Aosaki also notes low Japanese bank profitability (for example, very low returns on assets and equity) suggests it is more challenging for Japanese banks to improve capital adequacy ratios via profit growth, rather than a reduction in risk-weighted assets. Similarly, a weak macroeconomic environment and overbanking in the Japanese market further reduce options for adequacy ratio improvement. Nonetheless, at face value Bank of Japan data for currency capital adequacy might suggest that there should be little concern for future bank funding of mezzanine loans. Japanese banks have indicated they see little problem meeting both capital requirements and liquidity requirements. However, higher LTV ratio loans incur large penalties in terms of risk weighting in the denominator of the capital adequacy calculation (risk-weighted assets), requiring substantial additional capital to meet adequacy needs. Unfortunately data on risk weights applied to assets is extremely difficult to come by. For illustrative purposes, Table 2 provides a summary of average risk weightings for a selection of banks surveyed by the Bank of International Settlements. Two approaches for the calculation of risk weights are compared: the advanced internal ratings-based approach [AIRB] and the foundation internal ratings-based approach [FIRB]. The average maximum risk weight for corporate assets across both methods is just under 19%. Under Basel II the Bank of International Settlement s standardised approach has a maximum risk weight for corporate assets rated B+ or lower of 15% and much higher weights for low rated securitization exposures. 4

Table 2. BIS survey of bank average risk weights All data is in percentage terms. Mean Median Min. Max. No. of banks A. Advanced internal ratings-based approach for risk [AIRB] All 31.4 32.7 9.2 77.2 67 Corporate 44.8 48.9 24.3 18.9 61 Sovereign 3.1 4.1.2 46.3 44 Bank 18. 17.2 4.5 49. 49 B. Foundation internal ratings-based approach for risk [FIRB] All 32.6 32.6 2.2 78.5 35 Corporate 55.7 57.1 5.3 93.3 3 Sovereign 2.2 4.2. 21.6 22 Bank 16.4 17.9 1.4 39.3 28 Source: Bank for International Settlements Banks employ their own proprietary models to determine default probabilities (and other risks), to which risk weights are applied. These systems are in turn vetted by regulatory authorities and are not generally publicly available. 2 Hence, we cannot be definitive across the banking sector regarding the risk weights applied to mezzanine loans (nonetheless, we are aware of banks applying risk weights of 25% and some may apply higher weights). While it is not clear that risk weightings for such loans will increase, the greater value of bank capital in a Basel III world means that bank risk teams may be much more reluctant to allow lending desks to provide such loans. Indeed, we have considerable anecdotal evidence this is precisely the case today. We are, for example, seeing high LTV loans made in 27-8 maturing and being rolled into one year loans while banks seek ways to reduce their leverage exposure. In addition, the effect is aggravated if the loans fall within the definition of a resecuritisation exposure. In this case, when cash flows from a pool of exposures are employed to service two or more risk tranches, Basel III requires substantial increases in risk weights (levels up to 1,25%). Banks will also incur higher costs for more risky assets, given a wide range of requirements for detailed and complex modeling, including stress testing, and enhanced reporting requirements. We believe these structural factors likely signal tightening lending conditions for higher LTV ratio loans, including some combination of reduced loan Figure 2. Japan Bank Loans Yen hundreds of millions, current prices, non-seasonally adjusted A. Construction B. Real Estate 35, 3, 25, 2, 15, 1, 5, 1998 21 24 27 21 213 6, 5, 4, 3, 2, 1, 1998 21 24 27 21 213 2 The subjectivity of risk weights can create additional problems. The Bank of International Settlements has in the past observed cases of banks raising their capital adequacy ratios by understating asset risk. This is an important reason behind the inclusion of the leverage ratio minimum requirement, which has total assets rather than risk-weighted assets as the denominator. See Revisiting Risk -Weighted Assets, IMF 213 for a detailed discussion of problematic issues for the calculation and comparison of risk-weighted assets. 5

Figure 3. Japan Loan Survey demand by construction and real estate firms %, non-seasonally adjusted A. 1 5-5 -1-15 22 24 26 28 21 212 B. 1 5-5 -1-15 22 24 26 28 21 212 C. 1 5-5 -1-15 22 24 26 28 21 212 supply and/or higher loan costs. In the next section we introduce a range of cyclical factors which point to increased demand for such loans, just as supply may be tightening. Cyclical factors The expected impact of Basel III on the willingness of Japanese banks to provide mezzanine debt funding clashes with a range of cyclical factors that we believe contribute to a mismatch between mezzanine loan supply and mezzanine loan demand. Firstly, the total stock of Japanese bank loans to construction and real estate firms has been in decline since the mid to late 199s, falling 66% since a peak in late 1998 for construction and approximately 45% for real estate (see Figure 2). While it is debatable to what extent this phenomenon is a cyclical effect versus a structural effect, what is not debatable is it represents a massive reduction in the pool of construction loans. While loan supply has demonstrably shrunk, there is evidence of growing demand for loans by construction and real estate firms. The Bank of Japan s Loan Survey provides diffusion indices for a survey of senior loan officers at large Japanese banks by sector and firm size. They are asked how demand for loans has changed over the last three months. A positive reading indicates stronger loan demand. Over the last two to three years the indices for large, medium and small real estate and construction firms have all trended higher, recovering from post-global financial crisis levels (see Figure 3). A continuation of this trend (which wider cyclical factors would suggest is likely, including factors discussed below) would point to a growing divergence between construction loan supply and demand. Since the global financial crisis there has been a positive trend in building activity in Japan. Figure 4 provides both volumes (LHS) and total estimated costs (RHS) for building starts in Japan. Clear trends are evident in both. Further, Tankan survey data for both actual and forecast business conditions for both the real estate and construction sectors signals considerable optimism (see Figures 5 and 6). Most notably, business conditions indices for both actual and forecast activity for the construction sector turned positive earlier this year, for the first time since the late 199s. With evidence of strengthening business conditions for real estate and construction firms, expectations of further strengthening, increased construction activity 6

Figure 4. Building Construction Started, Japan A. Volume, non-seasonally adjusted B. Total estimated cost, Yen billions, current prices, season.-adj. 12, 1, 8, 6, 4, 2, 1999 21 23 25 27 29 211 213 3,5 3, 2,5 2, 1,5 1, 5 1999 21 23 25 27 29 211 213 Source: Japan Ministry of Land, Infrastructure, Transport & Tourism, Maiora Asset Maangement Figure 5. Tankan Business Conditions Real Estate Judgement of general business conditions of the responding enterprise, primarily in light of individual profits. A. Actual B. Forecast 8 6 4 2-2 -4-6 1983 1987 1991 1995 1999 23 27 211 6 5 4 3 2 1-1 -2-3 -4-5 1983 1987 1991 1995 1999 23 27 211 Figure 6. Tankan Business Conditions Construction Judgement of general business conditions of the responding enterprise, primarily in light of individual profits. A. Actual B. Forecast 8 6 4 2-2 -4-6 1983 1987 1991 1995 1999 23 27 211 8 6 4 2-2 -4-6 1983 1987 1991 1995 1999 23 27 211 7

Figure 7. Tankan Financial Position Construction Judgement of the general cash position of the responding enterprise, on account of the level of cash and cash equivalent, lending attitude of financial institutions, and payment and repayment terms (easy, not so tight or tight). A. Construction B. Real Estate 5 4 3 2 1-1 -2-3 -4 199 1993 1996 1999 22 25 28 211 3 2 1-1 -2-3 -4-5 -6-7 199 1993 1996 1999 22 25 28 211 Figure 8. Tankan Lending Attitude of Financial Institutions Construction Judgement of financial institutions' attitude towards lending as perceived by the responding enterprise (accommodative, not so severe or severe). 6 4 2-2 -4-6 199 1993 1996 1999 22 25 28 211 Figure 9. Construction debt to debt plus equity and increased demand for loans by construction companies, but decreased loan supply, the data would indicate a need for alternative funding for construction and real estate projects. Importantly though, we need to identify the capacity of construction companies to take on board more funding. In Figure 7 we illustrate Tankan results for the construction and real estate sectors reporting of their respective financial positions calculated as easy responses less tight responses to the general cash position of the firm, payment/repayment terms and lending attitude of financial institutions. Both are now in positive territory (for the first time since the late 199s for the overall construction industry), indicating a strengthening cash flow environment. It is particularly positive for large and medium-sized construction firms. Interestingly, Figure 8 shows the Debt is estimated from Bank of Japan data for the sum of commercial paper, bonds and loans from financial institutions. Equity is estimated as the difference between total assets and total liabilities from Bank of Japan Tankan survey data. A. Construction B. Real Estate 7% 6% 5% 4% 3% 2% 1% % 24 27 21 213 1% 8% 6% 4% 2% % 24 27 21 213 8

Figure 1. Propn. of total debt represented by loans from financial institutions (construction) Debt is estimated from Bank of Japan data for the sum of commercial paper, bonds and loans from financial institutions. Figure 11. Tankan forecast change in interest rates on loans (construction sector) Judgement of changes in the interest rate on borrowings of the responding enterprise (rise, unchanged or fall). 1% 8% 6% 4% 2% % 1998 21 24 27 21 213 1 8 6 4 2-2 -4-6 -8-1 199 1993 1996 1999 22 25 28 211 construction industry s view on the lending attitude of financial institutions has also improved (the same is true for real estate firms). While this at first glance may seem at odds with previous evidence of a long term downward trend in bank lending to the sector, it is likely indicative of lending conditions for the wider financial sector, rather than bank lending. It is also consistent with robust bank willingness to lend if lending is on the basis of a senior component and a separate mezzanine tranche, the latter provided by another counterparty. With regard to capacity to borrow, we have summed Tankan data for construction and real estate sector loans, bonds and commercial paper, and then divided this sum by the difference between total assets and total liabilities (plus the prior sum) to derive an estimate of the debt to debt plus equity ratio for the sectors (by firm size). Gearing levels have been trending down for both construction and real estate firms since the data series began in late 23 (Figure 9). The same dataset can be employed to demonstrate the dominance of financial institution lending as a source of funding. Figure 1 illustrates the proportion of funding provided by financial institution lending (relative to the sum of loans, bonds and commercial paper). roughly 9% of funding for the construction sector is provided by financial sector loans (higher for medium and small firms, and under 75% for large firms), indicating the critical importance of this funding source (the profile is very similar for real estate firms). Finally note that Tankan results show an expectation of higher interest rates for construction firms, going forward (Figure 11), albeit only slightly (again, the same is true for real estate firms). This is consistent with potential for tightening of bank lending to the sector. To sum up the cyclical factors presented here, we find evidence of: i. Decreasing bank lending to construction and real estate firms; ii. Increased loan demand by construction firms; iii. Increased construction and real estate activity; iv. Strengthening cash flow position of construction and real estate firms; and, v. Decreasing construction and real estate firm gearing. When combined with the potential structural impact on bank s willingness to lend for higher LTV projects as a result of Basel III implementation, we believe a mismatch between loan supply and loan demand for the construction and real estate sectors results in very attractive opportunities for investment in mezzanine debt funding. 3 3 In this brief overview of a range of cyclical characteristics of the construction sector we have made no mention of other major drivers, including Japanese Government stimulus packages and 211/12 earthquake damage. These, and other factors are also significant contributors to our positive overall view of opportunities in the Japanese mezzanine debt market. 9

Conclusions The structural impact of Basel III implementation on bank funding of high LTV debt, combined with strengthening cyclical demand for high LTV debt by the Japanese construction and real estate firms, suggests a growing mismatch between loan supply and loan demand. International Monetary Fund, 212, Japan: Basel core principles for effective banking supervision Detailed assessment of compliance, Financial Sector Assessment Program. International Monetary Fund, 212, Japan: Financial sector stability assessment update. Le Lesle, V. & S. Avramova, 212, Revisiting risk-weighted assets, IMF Working Paper, Monetary and Capital Markets Department. There is a consequent need for mezzanine financing for these sectors, and that need is growing. In addition, banks are increasingly exploring mezzanine funding structures as an option for restructuring their exposures to higher LTV loans. At Maiora Asset Management we believe these factors have resulted in attractive investment opportunities in Japanese mezzanine debt, and we have launched the MAIORA J APAN STRUCTURED F I- NANCE FUND to invest in and manage these opportunities. We are focusing on 65-85% LTV projects with expected returns in a range of 11-14%, compared with an expected senior debt yield of 1.5% (such low senior yields being at least partially attributable to over-banking in this segment in Japan). Our portfolio management team have identified a body of pro- References Aosaki, M., 213, Implementation of Basel III: Economic impacts and policy challenges in the United States, Japan, and the European Union, The Walter H. Shorenstein Asia-Pacific Research Center, Stanford University, Bank for International Settlements, 25, An explanatory note on the Basel III IRB risk weight functions, Basel Committee on Banking Supervision. Bank for International Settlements, 211, Basel III: A global regulatory framework for more resilient banks and banking systems, Basel Committee on Banking Supervision. Bank for International Settlements, 212, Basel III regulatory consistency assessment (Level 2): Japan, Basel Committee on Banking Supervision. Bank for International Settlements, 213, Regulatory Consistency Assessment Programme (RCAP): Analysis of risk-weighted assets for credit risk in the banking book, Basel Committee on Banking Supervision. Bank for International Settlements, 213, Regulatory consistency assessment programme (RCAP) - Analysis of risk-weighted assets for market risk, Basel Committee on Banking Supervision. Bank of Japan, 213, Financial results of Japan s banks for fiscal 212. Bank of Japan, 213, Financial system report.

Notes

This document and the information contained in it, is for informational and illustrative purposes only and does not constitute an offer or solicitation to sell shares or securities in any funds managed by Maiora Asset Management Pte Ltd (Maiora), or any related or associated company. Any such offer or solicitation will only be made by means of a fund Private Placement Memorandum and in accordance with applicable laws. This material is not intended to provide, nor does it constitute, any financial or investment advice to any individual or class of persons. Prior to any investment, you should make your own appraisal of risks from legal, tax and accounting perspectives and you are encouraged to seek independent advice from qualified experts taking into account your specific investment objectives, financial situation or particular needs. The accuracy, completeness or relevance of the information provided, and any research or analysis it is derived from, is not guaranteed although it has been drawn from sources which Maiora believe to be reliable for its own use. Whilst Maiora believes the information in this material is correct at the date of production, no warranty or representation (express or implied) to this effect is given and Maiora disclaims any liability for any errors or omissions. Information on past performance is not necessarily indicative of future performance. Any projections or other predictive statements regarding future events or performance of countries, markets or companies are not necessarily indicative of, and may differ from, actual events or results. The comments contained in this material are opinions only and should not be relied upon as authoritative, or without your own independent verification, or in substitution for the exercise of your own judgement. Over time, there may be changes to the information contained in this material. Maiora reserves the right to make changes and corrections to the information in this material, including any opinions or forecasts, at any time without notice. This material is confidential and solely for the private information and use of the intended recipient. You must not copy, distribute, disclose, publish or use it in whole or in part for any purpose without the prior written consent of Maiora. Maiora Asset Management Pte. Ltd. is a Singapore registered company, number 2122739K. Maiora Asset Management Pte.Ltd. 12