Swiss presentation

1. Introduction

2. The current proposals

3. Our analysis

4. Capital resultsC i l l

5. Spread results

6. Conclusions

Background

• The Basel Committee has recently announced plans to revise the Standardised Approach (SA) to bank capital for credit risk and to employ the revised SA as a floor for bank capital based on internal models.

• The new approaches will have a major impact on the overall level of capital and its distribution across banks and asset classes.

• This paper examines the effects of the proposed changes in capital rules on the Swiss loan market.

• Using primarily public information, we estimate the effects on the capital of individual g p y p p Swiss banks broken down by asset class and infer what this is likely to imply for lending rates in the Swiss market. Findings

• We find that the proposed Basel rule changes would significantly boost capital for corporate and commercial mortgage exposures while capital for residential mortgages could actually fall for SA banks.

• This pattern of effects across asset classes is at variance with the lessons of the recent financial crisis which was triggered by the collapse of the US mortgage market.

• It is also inconsistent with current policy concerns in Switzerland where regulators have publicly expressed concern about possible overheating in the residential mortgage market.

1. Introduction

2. The current proposals

3. Our analysis

4. Capital resultsC i l l

5. Spread results

6. Conclusions

Regulatory background

• Following the crisis, the initial focus was on Basel III changes

1. Improving the quality of capital (by focussing requirements more on common equity)

2. Tightening regulation of liquidity

3. Introducing counter-cyclical buffers and higher capital for systemic institutions

4. Reviewing securitisation rules

• Now there is a second wave of changes… g Second wave of regulatory changes

• Revisions proposed for market risk and operational risk weights

• In December 2014, the Basel Committee issued two consultative documents BCBS 306 and BCBS 307 proposing changes in how banks calculate credit risk capital.

• The proposals include:

• New rules for the capital of less sophisticated banks that use Standardised Approach (SA) regulations and

• A set of capital floors for advanced banks that employ the Internal- Ratings Based Approach (IRBA).

Key features of BCBS 307

• The current credit risk SA yields capital for exposures in each asset class based on 8% x EAD x Asset-class-specific-risk-weight

• If an exposure is rated (corporate/bank/sovereign), a rating-specific risk weight is employed.

• In the revised SA, rating-based risk weights are retained for sovereigns but for other exposures, ratings are dropped and risk weights depend on financial ratios For example, corporate risk weights depend on (i) Revenue and (ii) Leverage p p g p g (assets/equity)

€5m < €50m < Revenue Revenue Revenue Revenue ≤ €5m ≤ €50m ≤ €1bn > 1bn Leverage: 1x-3x 100% 90% 80% 60% Leverage: 3x-5x 110% 100% 90% 70% Leverage: >5x 130% 120% 110% 90% Negative Equity (*) 300% 300% 300% 300%

Bank risk weights depend on (i) CET1 ratios (ii) Net non-performing assets

12% > 9.5% > 7% > 5.5% >

CET1 ratio CET1 ratio CET1 ratio CET1 ratio CET1 ratio CET1 ratio

≥ 12% ≥9.5% ≥ 7% ≥ 5.5% ≥ 4.5% < 4.5%

NNPA ratio ≤ 1% 30% 40% 60% 80% 100% 300%

1% < NNPA ratio ≤ 3% 45% 60% 80% 100% 120% 300%

NNPA ratio > 3% 60% 60% 100% 120% 140% 300%

Residential mortgage risk weights depend on Loan-to-Value and Interest Coverage Ratios

40% ≤ LTV 60% ≤ LTV 80% ≤ LTV 90% ≤ LTV

LTV < 40% < 60% < 80% < 90% < 100% 100% ≤ LTV

DSC ≤ 35% 25% 30% 40% 50% 60% 80%

DSC > 35% 30% 40% 50% 70% 80% 100%

Objectives of BCBS 306 and 307

• The proposals have multiple objectives, including:

1. Improving comparability of capital across banks

2. Reducing discrepancies between the use of capital floors by different countries, and

3. Reducing the reliance of capital on agency ratings

• Particularly important to US, as Dodd- Frank prohibits such reliance

Effects of the proposals

• Profound effects on relative distribution of capital across:

• Jurisdictions

• Banks

• Types of loan

• Likely also to boost overall capital levels

• Aspects of distributional impact result from piecemeal calibration of rules rather than fully thought-through set of changes to relative capital

Impact of the rule changes?

• The effects of the rule changes remain opaque to the industry and probably also to regulators

• The Basel authorities have performed a Quantitative Impact Study (QIS) in which they ask banks to evaluate how the new rules would affect their capital calculations

Problems with the BCBS QIS

• We understand that many banks are found it difficult to assess the impact accurately because of a lack of relevant data.

• The results of the QIS remain confidential

Our response: Capital Floors, the Revised SA And the Cost of Loans in Switzerland

1. Introduction

2. The current proposals

3. Our analysis

4. Capital resultsC i l l

5. Spread results

6. Conclusions

Overview

• To shed light on what the proposals will mean, we have performed a detailed analysis of their implications for the Swiss loan market.

• We study the effects on 37 individual banks, breaking down results by different loan categories.

• We then re-aggregate the findings which we report for Large banks, Cantonal banks and Other banks

A two-step processp p

1. First, we calculate the impact on the capital that banks will have to hold against loans in different categories.

2. Second, we estimate the effect on the spreads that banks would charge for making loans in these categories.

• This involves estimating banks’ cost of funding given their regulatory capital requirements

Results

• Our analysis reveals dramatic results.

• We believe some effects of the proposed regulations are unintentional. For instance:

• Capital for corporate loans within Switzerland would be substantially boosted

• Whereas capital for residential mortgages would fall for SA banks (but increase for IRBA banks)

• IRBA banks would be required to hold substantially greater capital

• Capital for SA banks would in many cases fall

Immediate implicationsp

• Changes to capital levels matter for the rest of the economy to the extent that they affect the lending rates that banks charge.

• To reveal the implications for lending rates, we have looked at how bank loan spreads (over and above Treasury yields) would be affected.

What one needs to know

• To calculate the revised SA capital for a given bank and given asset class (e.g., corporate exposures), one needs the total exposure and the breakdown of that exposure by the risk indicator ranges specified in BCBS 307

• Such breakdowns are hard to assess from outside a bank but we have

1. Data on the credit quality of their exposures (from Pillar 3 disclosures )

2. Loan breakdowns by risk indicators and credit quality for the Swiss market (from UBS) Steps in the capital forecasting

a) For IRB banks, we infer default probabilities (PDs) from risk weights using the standard Basel formula assuming values of loss given default (LGD) and maturity (MT).

b) For SA banks, we infer ratings from RW according to the look-up tables in the current SA approach.

c) From the inferred ratings, we map the corresponding PD based on a default probability master scale table provided by UBS (see Table 8).

d) We devise two rating buckets: AAA to A- and BBB+ to default.

e) For each asset class that depends on two capital indicators, we estimate three joint distributions: one unconditional distribution and two conditional distributions for the above two rating buckets.

f) For each asset class, we associate to each of the exposure categories (broken down by credit quality) a distribution of the two capital indicators conditional on their credit quality.

g) Given the look-up table in the revised SA paper, the indicator distribution and the loan exposure at default, we calculate the risk weighted assets and capital requirement for the loan book.

What one needs to know

• To calculate the revised SA capital for a given bank and given asset class (e.g., corporate exposures), one needs the total exposure and the breakdown of that exposure by the risk indicator ranges specified in BCBS 307

• Such breakdowns are hard to assess from outside a bank but we have

1. Data on the credit quality of their exposures (from Pillar 3 disclosures )

2. Loan breakdowns by risk indicators and credit quality for the Swiss market (from UBS) Steps in the capital forecasting

a) For IRB banks, we infer default probabilities (PDs) from risk weights using the standard Basel formula assuming values of loss given default (LGD) and maturity (MT).

b) For SA banks, we infer ratings from RW according to the look-up tables in the current SA approach.

c) From the inferred ratings, we map the corresponding PD based on a default probability master scale table provided by UBS (see Table 8).

d) We devise two rating buckets: AAA to A- and BBB+ to default.

e) For each asset class that depends on two capital indicators, we estimate three joint distributions: one unconditional distribution and two conditional distributions for the above two rating buckets.

f) For each asset class, we associate to each of the exposure categories (broken down by credit quality) a distribution of the two capital indicators conditional on their credit quality.

g) Given the look-up table in the revised SA paper, the indicator distribution and the loan exposure at default, we calculate the risk weighted assets and capital requirement for the loan book.

Banks finance a given loan using a combination of deposit and equity funding.

• The latter is considerably more expensive to a bank, dollar for dollar.

• The volume amount of equity funding that a bank requires for a loan is determined by the regulatory capital a bank is required to hold.

Here the “capital level” is measured per Swiss franc of exposureHere the capital level is measured per Swiss franc of exposure. To estimate the return To estimate the return on equity, we use the Capital Asset Pricing Model (CAPM).

We estimate the beta using a regression:

1. Introduction

2. The current proposals

3. Our analysis

4. Capital resultsC i l l

5. Spread results

6. Conclusions

What are the effects?

• We focus on four key loan categories:

1. exposures to banks,

2. corporate loans,

3. commercial mortgages , and

4. residential mortgages.

• Corporate and commercial mortgage capital increase g g p substantially

• Capital for residential mortgages hardly changes

• this partly reflects offsetting rises and falls in increases in IRBA and SA bank capital for Note: The figure shows the percentage change in capital residential mortgages. that Swiss banks will have to hold under the new regulations against their loans to Swiss borrowers.

Weighted average SA bank RW changes IRB bank RW changes (80% asset class floor)

1. Introduction

2. The current proposals

3. Our analysis

4. Capital resultsC i l l

5. Spread results

6. Conclusions

• The table shows weighted average estimates of changes in bank lending spreads that would result from the adoption of the new bank capital rules.

• Estimates are presented for two assumptions about the relevant definition of equity capital: Common Equity Tier 1 (CET1) or Total Capital

• The increase in the cost of lending due to the spread change is CHF 1.3 billion or CHF 2.1 billion per annum (depending on which of the capital definitions is employed).

Banks Corp. CML. Res. Specialised Wtd. Mtg. g Mtg. g Lending g Avg. g

CET1 Capital Target

Wtd. Avg. of IRB Banks 3 60 57 9 70 30

Wtd. Avg. of IRB and SA Banks 7 45 14 2 70 11

Total Capital Target

Wtd. Avg. of IRB Banks 6 103 90 15 121 51

Wtd. Avg. of IRB and SA Banks 11 73 22 3 121 19

Note: The table shows weighted average estimates of changes in bank lending spreads.

Spread impact in basis points (asset class level floor for IRB banks)

Spread Impact 1 (Constant Equity Return) Spread Impact 2 (Reduced Equity Return) Speciali Speciali Cml. Res. zed Wtd. CML. Res. zed Wtd. Banks Corp. Mtg. Mtg. Lending Avg. Banks Corp. Mtg. Mtg. Lending Avg. 60% floor Wtd. Avg. of IRB Banks 3 39 38 4 49 19 3 37 36 3 46 18 Wtd. Avg. of IRB and SA Banks 8 38 8 0 49 8 7 36 7 0 46 8 CET 1 70% floor capital Wtd. Avg. of IRB Banks 4 52 49 6 62 25 3 49 47 5 59 23 target Wtd. Avg. of IRB and SA Banks 8 43 12 1 62 10 7 41 11 0 59 9 80% floor Wtd. Avg. of IRB Banks 5 65 60 11 75 33 3 60 57 9 70 30 Wtd. Avg. of IRB and SA Banks 8 47 15 2 75 13 7 45 14 2 70 11 60% floor Wtd. Avg. of IRB Banks 6 67 60 6 84 32 4 63 57 5 80 29 Wtd. Avg. of IRB and SA Banks 12 60 14 0 84 13 11 57 12 0 80 12 Total 70% floor capital Wtd. Avg. of IRB Banks 6 90 79 10 107 43 4 84 74 8 101 39 target Wtd. Avg. of IRB and SA Banks 12 68 19 2 107 17 11 65 18 1 101 15 80% floor Wtd. Avg. of IRB Banks 8 111 96 18 128 55 6 103 90 15 121 51 Wtd. Avg. of IRB and SA Banks 12 76 24 4 128 21 11 73 22 3 121 19

Monetary impact per year

Monetary cost (Constant Equity Return) Monetary cost(Reduced Equity Return) Speciali Speciali Cml. Res. zed CML. Res. zed Banks Corp. Mtg. Mtg. Lending Total Banks Corp. Mtg. Mtg. Lending Total CET 1 capital target RSA exposure level 60% floor 72 507 98 33 254 963 67 487 87 5 241 887 RSA asset class level 60% floor 68 488 96 14 250 916 64 469 86 -12 238 845 RSA bank level 60% floor 62 483 94 7 246 892 58 464 84 -19 235 822 RSA exposure level 70% floor 74 567 137 94 320 1,192 70 542 123 59 302 1,096 RSA asset class level 70% floor 69 554 136 62 319 1,140 64 530 123 30 302 1,049 RSA bank level 70% floorRSA bank level 70% floor 63 554 136 62 318 1 1331,133 59 530 123 29 301 1 0421,042 RSA exposure level 80% floor 77 623 174 181 385 1,440 72 594 157 139 362 1,323 RSA asset class level 80% floor 70 616 174 156 384 1,399 65 587 157 115 361 1,286 RSA bank level 80% floor 66 616 174 156 384 1,396 62 587 157 115 361 1,283 Total capital target RSA exposure level 60% floor 110 808 162 63 437 1,580 104 775 145 20 415 1,458 RSA asset class level 60% floor 105 775 160 33 430 1,503 98 744 144 -7 410 1,389 RSA bank level 60% floor 94 766 155 22 423 1,461 88 736 140 -17 403 1,350 RSA exposure level 70% floor 115 910 224 162 550 1,960 108 869 203 108 519 1,805 RSA asset class level 70% floor 105 888 223 111 547 1,875 99 849 203 60 518 1,728 RSA bank level 70% floor 96 888 223 110 547 1,863 89 848 202 60 517 1,717 RSA exposure level 80% floor 119 1,007 283 302 661 2,372 111 958 257 235 621 2,183 RSA asset class level 80% floor 107 994 283 261 659 2,304 100 946 258 197 620 2,121 RSA bank level 80% floor 102 994 283 261 659 2,299 95 946 258 197 620 2,116

1. Introduction

2. The current proposals

3. Our analysis

4. Capital resultsC i l l

5. Spread results

6. Conclusions

The major findings are

• Bank capital charges for Bank, Corporate, Commercial Mortgage, Residential Mortgage borrowers and Specialised Lending will increase by 18%, 160%,215% , 35% and 259% respectively.

• Total IRB bank capital will rise by 108%.

• Total SA bank capital against Residential Mortgages will fall from CHF 22 billion to CHF 21 billion.

• This reallocation of bank capital is hard to square with the fact thatThis reallocation of bank capital is hard to square with the fact that

• the recent crisis began with the collapse of the US mortgage market (albeit one located in the United States rather than Switzerland),

• many commentators have expressed concern about possible house price bubbles in Switzerland driven by readily available mortgage financing, and loans to corporates were remarkably resilient to the crisis in many countries.

The major findings are

• Substantially higher capital is likely to lead to increases in lending spreads in Swiss loan markets.

• We estimate that Corporate Loan spreads would rise by between 45 and 76 basis points assuming a full pass-through to borrowers and an 80% asset class level floor.

• If one assumes that IRB banks set prices in corporate lending, then the impact will be between 60 and 111 basis points.will be between 60 and 111 basis points.

• Multiplying loan volumes by spread changes, one obtains a transparent monetary measure of the impact of the policy changes.

• This suggests that the annual flow cost of the new rules would be CHF 1.3 billion and CHF 2.1 billion.

• A conservative measure of the present discounted cost (assuming a 3% discount rate) is between CHF 42.8 billion and CHF 70.5 billion.

Website: www.riskcontrollimited.com Telephone: +44 20 3307 0730 Address: 13-14 Dean Street, London, W1D 3RS, U.K.

William Perraudin, Director, Telephone: +44 20 3307 0731 (o) +44 7968 328 459 (m) Email: william.perraudin@riskcontrollimited.com