|
|
Post by solicitorious on Sept 18, 2015 14:25:07 GMT
|
|
locutus
Member of DD Central
Posts: 1,059 
Likes: 1,622
|
Post by locutus on Sept 18, 2015 14:24:51 GMT
What about macro economic factors? If/when an event like 2008 comes around again, those valuations won't hold and the LTVs would go over 100%. For me, this is the biggest risk.
|
|
|
|
Post by meledor on Sept 18, 2015 14:26:21 GMT
OK, taking into account the repayment of PBL32, we now have 42 loans and PF of £865.160k. Model says: Chance of any loss 7.05%, no loss 92.95% loss <0.5% 4.73% [these are cumulative, i.e. the chance of a loss between 0.5% and 1% is 2.32%-0.56% = 1.76%, etc] loss >0.5% 2.32% loss >1% 0.56% loss >2% 0.03% loss>3% 0.00% loss >5% 0.00% loss >10% 0.00% loss >20% 0.00% loss >30% 0.00% loss >40% 0.00% loss >50% 0.00% overall loss 0.03%, including times when there's no loss average loss 0.43%, if there is a loss Comparison with when there's no PF Chance of any loss 96.31%, no loss 3.69% loss <0.5% 29.67% loss >0.5% 66.64% loss >1% 37.01% loss >2% 7.05% loss>3% 0.56% loss >5% 0.00% loss >10% 0.00% loss >20% 0.00% loss >30% 0.00% loss >40% 0.00% loss >50% 0.00% overall loss 0.89%, including times when there's no loss average loss 0.93%, if there is a loss
If there really is 0% chance of a loss greater than 3% (0.56% chance with no Provision Fund) then you have to wonder why the major banks do not operate in the bridging finance sector with its attractive returns.
I would therefore suggest the loss given default implied by the model is optimistic. In a worst case scenario, thinking about some of the situations encountered in the recent credit crunch, I could well imagine a loss given default applied across the portfolio of 20%
|
|
|
|
Post by solicitorious on Sept 18, 2015 14:36:34 GMT
If there really is 0% chance of a loss greater than 3% (0.56% chance with no Provision Fund) then you have to wonder why the major banks do not operate in the bridging finance sector with its attractive returns.
I would therefore suggest the loss given default implied by the model is optimistic. In a worst case scenario, thinking about some of the situations encountered in the recent credit crunch, I could well imagine a loss given default applied across the portfolio of 20%
It's a model of 10,000 trials, so all the outputs should be treated as ≈, rather than =. In probability, there are no real zero probabilities.... OK, would you like me to try D=50%, L=40%?
|
|
Investor
Member of DD Central
Posts: 662
Likes: 590
|
Post by Investor on Sept 18, 2015 14:42:46 GMT
Go for it, though in the real world I would think that if L ever got close to 40% D would be at 100% anyway.
|
|
|
|
Post by solicitorious on Sept 18, 2015 14:57:20 GMT
Go for it, though in the real would I would think that if L ever got close to 40% D would be at 100% anyway. for D=50%, L=40%, Model says: Chance of any loss 89.39%, no loss 10.61% loss <0.5% 15.98% loss >0.5% 73.42% loss >1% 51.58% loss >2% 11.50% loss >3% 0.43% loss >5% 0.00% loss >10% 0.00% loss >20% 0.00% loss >30% 0.00% loss >40% 0.00% loss >50% 0.00% overall loss 1.06%, including times when there's no loss average loss 1.18%, if there is a loss Comparison with when there's no PF Chance of any loss 99.99%, no loss 0.01% loss <0.5% 0.01% loss >0.5% 99.99% loss >1% 99.69% loss >2% 89.40% loss >3% 51.58% loss >5% 0.43% loss >10% 0.00% loss >20% 0.00% loss >30% 0.00% loss >40% 0.00% loss >50% 0.00% overall loss 3.02%, including times when there's no loss average loss 3.02%, if there is a loss
|
|
|
|
Post by solicitorious on Sept 18, 2015 15:17:30 GMT
What about macro economic factors? If/when an event like 2008 comes around again, those valuations won't hold and the LTVs would go over 100%. For me, this is the biggest risk. There will always be a non-zero probability we could lose everything... However for D=100%, L=40%, the loss would be fixed at 4.05%. For D=100%, L=50%, the loss would be around 14%.* *I will have to tweak the model slightly for L>40% to take account of the 2nd charges, as someone said.
|
|
|
|
Post by meledor on Sept 18, 2015 15:59:17 GMT
If there really is 0% chance of a loss greater than 3% (0.56% chance with no Provision Fund) then you have to wonder why the major banks do not operate in the bridging finance sector with its attractive returns.
I would therefore suggest the loss given default implied by the model is optimistic. In a worst case scenario, thinking about some of the situations encountered in the recent credit crunch, I could well imagine a loss given default applied across the portfolio of 20%
It's a model of 10,000 trials, so all the outputs should be treated as ≈, rather than =. In probability, there are no real zero probabilities.... OK, would you like me to try D=50%, L=40%?
If we are talking worst case scenarios then I can get more pessimistic than that. Remember your L is based on valuation rather than loan amount. And in a situation where liquidity has dried up completely nobody will be buying the properties the borrowers have been developing and nobody will be interested in re-finance to enable exit. Values would plummet so an L of 60% (giving a 43% loss to the lender on a 70% LTV) would be possible especially for the non-residential loans.
|
|
Liz
Member of DD Central
Posts: 2,426
Likes: 1,297
|
Post by Liz on Sept 18, 2015 16:11:13 GMT
Can we factor in on average every loan pays 6 months interest, and that is deducted off any loss?
|
|
|
|
Post by marek63 on Sept 18, 2015 16:15:57 GMT
The issue in a crash will be how long to a saleable recovery. Say prices crash by 50%. Ie no one wants to buy anything other than at half price and that takeout finance is not available for refinancing - much of the SS exit route. So every loan is now at 50% of original value. Ie a default of 100 and lgd of 50.
That is Jan 2008 ...
|
|
|
|
Post by solicitorious on Sept 18, 2015 16:21:25 GMT
Can we factor in on average every loan pays 6 months interest, and that is deducted off any loss? Yes I suppose we could, although it might muddy the waters. Let's stick to capital losses for the time being. Besides, in a "perfect storm" they could all go down like ninepins!
|
|
|
|
Post by marek63 on Sept 18, 2015 16:33:49 GMT
PS excellent model and thread. Any chance you could look at AC loans too? ??
|
|
Investor
Member of DD Central
Posts: 662
Likes: 590
|
Post by Investor on Sept 18, 2015 16:34:10 GMT
We are now safe from the perfect storm, reading an article recently I have become confident that we will never have another issue with the trading of 'subprime mortgages'. How do I know for sure? It would appear that they have now all been renamed to 'nonprime' morgtgages.
2015
'The business of bundling riskier U.S. mortgages into bonds without government backing is gearing up for a comeback. Just don’t call it subprime.
Hedge fund Seer Capital Management, money manager Angel Oak Capital and Sydney-based bank Macquarie Group Ltd. are among firms buying up loans to borrowers who can’t qualify for conventional mortgages because of issues such as low credit scores, foreclosures or hard-to-document income. They each plan to pool the mortgages into securities of varying risk and sell some to investors this year. JPMorgan Chase & Co. analysts predict as much as $5 billion of deals could get done, while Nomura Holdings Inc. forecasts $1 billion to $2 billion.
Investment firms are looking to revive the market without repeating the mistakes that fueled the U.S. housing crisis last decade, which blew up the global economy. This time, they will retain the riskiest stakes in the deals, unlike how Wall Street banks and other issuers shifted most of the dangers before the crisis. Seer Capital and Angel Oak prefer the term “nonprime” for lending that flirts with practices that used to be employed for debt known as subprime or Alt-A.'
|
|
|
|
Post by solicitorious on Sept 18, 2015 16:39:09 GMT
PS excellent model and thread. Any chance you could look at AC loans too? ?? Gulp. Do I have to? I voted with my feet when it came to AC, without bothering to create a model... A proper scrutiny of the "LTV"s was enough for me.
|
|
|
|
Post by robberbaron on Sept 18, 2015 16:47:52 GMT
Can we factor in on average every loan pays 6 months interest, and that is deducted off any loss? Yes I suppose we could, although it might muddy the waters. Let's stick to capital losses for the time being. Besides, in a "perfect storm" they could all go down like ninepins! That's why you should probably introduce some correlation in your model. Assuming all loan default probabilities are independent is just too unrealistic. You could assume for simplicity sake the same correlation for all loans.
|
|
