Bank credit risk: Making sense of the current credit cycle and outlook Investec seminar Cape Town 18 March 2009 Gert Kruger, FirstRand Banking Group
Contents Features of the current credit cycle: Three analogies FRBG credit experience to date Credit outlook: Clouds and silver linings
Features of the SA retail credit cycle: Three analogies
Example 1: Strip sets training Each successive set increases heart rate at a quicker rate of change Average heart rate (AHR) bpm 180 160 140 120 100 80 60 40 20 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 AHR during warm-up and cool down AHR sprinting up hill AHR jogging down hill slowly
Example 2: Descent at the end of a race The timing of the descent matters the longer out it is the less immediate is the benefit in terms of decreased heart rate Example: Knysna half marathon 160 176 170 166 164 164 Heart Rates 160 152 164 164 166 166
Example 3: Tandem bicycle going up a hill Partner turning from help to dead weight
FRBG credit experience in the cycle: Reaction of different parts of the portfolio
NPLs and bad debts continue upward trend 1.40 1.64 4.2 1.31 3.4 0.98 2.8 1.11 2.6 0.79 2.3 0.97 0.83 0.79 0.78 Long run expected loss: 0.8 0.41 0.51 1.8 1.5 0.32 1.5 1.2 1.1 1.28 2.9 1.19 4.2 Jun '99 Jun '00 Jun '01 Jun '02 Jun '03 Jun '04 Jun '05 Jun '06 Jun '07 Dec '07 Jun '08 Dec '08 NPLs (%) Impairment charge (%) Impairment charge after credit hedge
Observation on the product line impact of successive rate increases 400 350 300 250 200 150 100 50 0 Prime rate New impairments Credit card cycle behaviour Peak early in the cycle Q2/3 2007 Impact off write-offs and deeper cycle Jul06 Sep06 Nov06 Jan07 Mar07 May07 Jul07 Sep07 Nov07 Jan08 Mar08 May08 Jul08 Sep08 Nov08 18% 16% 14% 12% 10% 8% 6% 4% 2% 0%
Observation on the product line impact of successive rate increases Wesbank impairments 400 18% 350 16% 300 250 Prime rate 14% 12% 200 150 100 New impairments 1 st Peak early in the cycle Q2/3 2007 Impact of deeper credit cycle 10% 8% 6% 4% 50 2% 0 0%
Observation on the product line impact of successive rate increases 1400 Homeloans cycle behaviour 18.0% 1200 16.0% 1000 800 Prime rate 14.0% 12.0% 10.0% 600 400 200 0 New impairments Property price slowdown impact Turning point late in the cycle Q1 2008 Jul06 Sep06 Nov06 Jan07 Mar07 May07 Jul07 Sep07 Nov07 Jan08 Mar08 May08 Jul08 Sep08 Nov08 8.0% 6.0% 4.0% 2.0% 0.0%
Outlook: Key clouds and silver linings to consider
Outlook: Key clouds and silver linings to consider Job losses House price developments Corporate credit
The macro economic backdrop for the credit cycle First impacts High inflation and interest hikes resulting in Lower disposable income Increased consumer defaults Next impacts expected Impact of the credit crunch on the real economy Lower GDP Job losses But with some relief Impact of sharp rate cuts Inflation easing
Job losses the macro data Unemployment Trends (QLFS ) 1 2005 2006 2007 2008 Services 2,192 2,319 2,560 2,661 Financial 1,296 1,309 1,482 1,636 Transport & Communications 616 611 696 774 Retail 3,024 3,055 2,935 3,164 Construction 935 1,024 1,054 1,191 Elec, Gas & Water 100 119 98 86 Manufacturing 1,706 1,737 1,757 1,944 Mining 411 398 432 321 Agriculture 925 1,088 1,041 764 Total 11,205 11,660 12,055 12,541 Source: Stats SA Source: Stats SA
How job losses manifest in a portfolio (assume 3% job losses and 0% latent job losses) Example: High default portfolio (e.g. card with 10% margin) Assume annual NPL of 12% in distressed market and 70% NPL coverage Charge (pre job losses): 12% x 70% = 8.4% Charge (incl. job losses): 15% x 70% = 10.5% Relative increase = 25% Absolute increase as % of margin: 21% Example: Low default portfolio (e.g. homeloan with 2% margin) Assume annual NPL of 7% in distressed market and 20% coverage Charge (pre job losses): 7% x 20% = 1.4% Charge (incl. job losses): 10% x 20% = 2% Relative increase = 43% Absolute increase as % of margin: 30% Low default, low margin portfolios less resilient for job losses. Need to consider wider factors though (refer next slide)
Assumptions to make to determine impact Consensus forecasts vary But 2-3% often quoted as a possibility (250k 350k) Job losses vs banked population Probably reasonable to assume close to a 1:1 relationship Job loss demographics Data to date suggest higher incidence of job losses in lower income segments Re-employment potential Historically better for higher qualified areas Difficult to track in the credit portfolios Job losses vs latent job loss number and defaults in credit portfolios Even benign times have job losses underlying credit results Key question: overlapping or incremental? Need to evaluate impact of job losses to benefit from cash flow relief from rate cuts
Outlook: Key clouds and silver linings to consider Job losses House price developments Corporate credit
How house price movements manifest in bad debts Key parameters Volume of NPL s NPL coverage Loan to value Distress price haircut Time to recovery Level of curing Charge (Volume x coverage%) Example R1bn NPL Coverage elements LTV of 100% [R1bn] Haircut of 35% [R0.65bn] 12 months disc. factor 60% curing Coverage calc Charge: [R1bn ((R1bn x 0.65) x 1/1.14)] x 40% = R172m Coverage therefore 17.2% Evolution of normal market prices and distressed prices as % of market price are key factors to consider
Impact of a 10% drop in house prices on the provision Example (excluding fall): R1bn NPL Coverage elements LTV of 100% [R1bn of collateral] Haircut of 35% [R0.65bn] 12 months discount @ 14% 60% curing Coverage calc Charge: [R1bn ((R1bn x 0.65%) x 1/1.14)] x 40% = R172m Coverage therefore 17.2% Example (including fall): R1bn NPL Coverage elements LTV of 111% [R0.9bn value of collateral] Haircut of 35% [R0.72bn] 12 months discount @ 14% 60% curing Coverage calc Charge: [R1bn ((R0.9bn x 0.65) x 1/1.14)] x 40% = R195m Coverage therefore 19.5% Approximately 2.3% increase in coverage required for 10% fall (13% increase in relative terms)
House price development the simple view 400 350 300 +/-15% 250 200 +/-15% 150 100 50 0 Jun-94 Jun-95 Jun-96 Jun-97 Jun-98 Jun-99 Jun-00 Jun-01 Jun-02 Jun-03 Jun-04 Jun-05 Jun-06 Jun-07 Jun-08 Real House price index Nominal House price index Source: INet & BSM CPM
Nominal and real growth vs global experience Nominal growth in house prices Real growth in house prices Index 350 Index 2.5 300 SA 2.0 SA UK 250 1.5 US 200 UK 1.0 150 US Australia 0.5 100 Mar 02 Mar 03 Mar 04 Mar 05 Mar 06 Mar 07 Mar 08 0.0 Jan 02 Dec 02 Nov 03 Oct 04 Sep 05 Aug 06 Jul 07 Jun 08
House price : disposable income House prices to disposable income Index SA house price to disposable income Ratio 120 1.7 SA 1.5 UK 100 House Price / Income 1.3 US 80 Period Average 1.1 60 0.9 Jan 00 Jan 01 Jan 02 Jan 03 Jan 04 Jan 05 Jan 06 Jan 07 Jan 08 40 Dec 85 Dec 87 Dec 89 Dec 91 Dec 93 Dec 95 Dec 97 Dec 99 Dec 01 Dec 03 Dec 05 Dec 07
Outlook Residential property prices likely to remain under pressure Will put pressure on impairments through lower recoveries Rate cuts will provide some cash flow relief and reduce cash flow defaulters
Outlook: Key clouds and silver linings to consider Job losses House price developments Corporate credit
Migration risk to corporate Impairment ratio (%) 3.5 3.0 2.5 2.0 2.21 2.41 1.5 1.0 1.39 1.64 0.5 0.0 0.66 0.33 0.34 Dec '07 Jun '08* Dec '08 Total impairments Dec '08 Impairment % Retail Impairment % Wholesale * June 2008 for the 6 months ended Source: FRBG results presentation
27 How will it unfold Corporate? RMB Wholesale Credit Cycle Indicator 7.00% 6.00% 5.00% 4.00% 3.00% 2.00% 1.00% 0.00% 199501 199506 199511 199604 199609 199702 199707 199712 199805 199810 199903 199908 200001 200006 200011 200104 200109 200202 200207 200212 200305 200310 200403 200408 200501 200506 200511 200604 200609 200702 200707 200712 200805 200810 Defaults Wholesale Credit Cycle Indicator GDP
Corporates however in better health than previous cycles Key financial ratios of JSE industrial companies Source: Inet & BSM CPM
Outlook Corporate spill-over has already started Key area of pressure is smaller businesses Corporate balance sheets in reasonably good health pre-crisis Impact of global spill-overs is the main unknown
Conclusion Expect retail pressure to remain in the short run Especially for asset backed retail areas e.g. mortgages due to house prices falling Rate cuts will provide some cash flow relief But job loss uncertainties likely to mitigate benefits Corporate spill-over has started But may be mitigated by the stronger balance sheets International spill-over too early to call