ablender
Member of DD Central
Posts: 2,204 
Likes: 555
|
Post by ablender on Feb 28, 2016 3:10:57 GMT
Hi chris . Related to the LTV of a loan vs its security, I am curious about the Nil values attributed to some of the securities on AC, such as First Debenture, Corporate Guarantee and Personal Guarantee (even if it mentions a sum of money in the description.) As a reference, may I use loan #226. Do such Nil values effect the LTV or the overall security of a loan and how? |
|
|
|
Post by chris on Feb 28, 2016 7:00:24 GMT
Hi chris . Related to the LTV of a loan vs its security, I am curious about the Nil values attributed to some of the securities on AC, such as First Debenture, Corporate Guarantee and Personal Guarantee (even if it mentions a sum of money in the description.) As a reference, may I use loan #226. Do such Nil values effect the LTV or the overall security of a loan and how? As I understand it this is used by the team when we have taken that security but in the event of a default it can't really be counted upon by the lenders. It's a nice to have, and should have some value, but lenders shouldn't be basing their lending decisions on the full value of that asset at the time of the loan being taken out. They can also be used as an alternative route to having a claim over the primary security should the lender try and use the legal system to wriggle out of their obligations. I think a lot of PGs are entered as having no value but can be an important last line of defence.
|
|
daveb4
Member of DD Central
Posts: 220
Likes: 116
|
Post by daveb4 on Feb 28, 2016 8:05:35 GMT
I think the way AC do it is the right way. If the LTV is shown to be 70%, it ought to mean that as long as the security can be sold for net proceeds equal to 70% of its 'value' the investors ought to be able to exit without loss. Agreed relative simple calculations and ignoring all things like PG's, Debentures etc as in real world useful not necessarily worth any actual value in a liquidated company. Appreciate that this does vary across platforms and some companies are outrageous with regard to their definition of value - Again as i have mentioned in various links before I think FSA are going to get tougher over next couple of years and straight away this has got to be an area where they would like to 'assist with some guidance' to try and stop some misleading information to lenders? If this happens then I think some platforms will be in a better position than others moving forward Most people on this forum however use all sorts of their personal calculations mainly along the lines that AC use and see through most of the various 'strange' calculations. Unfortunately I think we are in the minority and the vast majority of P2P users just invest and if defaults occur don't really mind as long as overall income is over a set percentage return. My calculations are much tougher and i ideally want 100% cover of the loan based on Residential property First charge - 90% of 90 day sale price less new loan Second charge - 80% of 90 day sale price less previous debt less new loan Commercial Property First Charge - 70% of 90/180 day sale price less new loan I then have generally three amounts that I lend, small, medium and high depending on how close the new loan is to the above. I then normally tweek it slightly if it is repayment or interest only, sometimes where it is in the country and lastly an opinion on the business itself. This generally works with steady house price increases like at the moment. I have been going about a year now and getting close to being fully funded so when good loans come up I take from the smaller ones to reduce risk. Admittedly SS as one of my platforms does challenge this system a little so most investments there are on the small amount and across many businesses. Boasting time - with only one very small debt (fraud) and lending to about 200 businesses this policy seems to be working so far 
|
|
|
|
Post by reeknralf on Feb 28, 2016 10:23:32 GMT
Unfortunately, it's clear that many lenders have way too much confidence in both the valuation itself and the way that the platform defines the LTV. The issue for a platform like AC is that by being conservative they can put themselves at a competitive disadvantage. The AC methodology can overestimate the LTV in some cases. Take loan #79; this has a first charge over a hotel, and two second charges over residential property, where the free security above the first charge is quite small. The AC methodology puts the LTV at 74.5%. However, if we drop the two second charges, the LTV drops to 67.6%. So if I was a platform wanting to get maximum access to the somewhat dumb, sticky, undemanding money from cash ISAs, I'd go for the simplest methodology possible that makes LTVs as low as possible! Assuming we define LTV as the percentage value to which assets may fall without making a capital loss, there's just one right answer. In the case of loan #79, 74.5% is wrong, and 67% is right. The 74.5% is in effect assuming that AC lenders have to use 'profits' from their first charge on the hotel, to make good losses to first charge holders on the residential properties. Unless I really don't understand 1st and 2nd charges, this is a nonsense. The LTV's for the first charges on the hotel is 67.7%, while those on the residential properties are 72% and 88%. If the properties realise 74.5% of their valuations, this would yield £484k on the hotel, 100% of the first charge plus £3k for AC lenders on the first RP, and a loss of £48k for the first charge holder on the 2nd RP. AC's loan is £440k, and is amply covered by asset sales yielding £487k, but AC's calculation incorrectly assumes we have to make good the first charge holder's losses of £48k, leaving only £439K, i.e. break even (there's a rounding error). The correct LTV is 67.7%, which yields £440k on the hotel, and losses for the third parties holding first charges on the RP's (tough on them). I find it troubling that AC, or any other platform, can't do these sums right.
|
|
|
|
Post by Jack Barlow on Feb 28, 2016 11:10:36 GMT
Assuming we define LTV as the percentage value to which assets may fall without making a capital loss, there's just one right answer. In the case of loan #79, 74.5% is wrong, and 67% is right. I agree (and I flagged up this anomaly in AC's LTV calculation for this loan about a year ago: p2pindependentforum.com/post/46687/thread.) The AC calculation formula needs a simple tweak for the scenario with multiple security that includes 2nd+ charges. Fortunately, at least at the time, the anomaly only impacted the LTVs quoted on a couple of AC loans. I also agree with samford71 's earlier comments that (paraphrasing): (i) the anomaly (when it has an impact) always exaggerates the LTV, and (ii) lenders should always look beyond the quoted headline LTV value, but if AC are presenting an LTV figure to lenders I think they should really be ensuring that the underlying calculation doesn't include flawed logic.
|
|
|
|
Post by chris on Feb 28, 2016 12:18:07 GMT
The correct LTV is 67.7%, which yields £440k on the hotel, and losses for the third parties holding first charges on the RP's (tough on them). I find it troubling that AC, or any other platform, can't do these sums right. I do find that quite unfair. Your sum is correct for your definition of LTV, for which there is no standard definition. The formula we're using is that used within the banking industry, so if we can't do these sums "right" then neither can any bank. Other platforms aren't even displaying an LTV or are displaying different ones in different places with the headline LTV being displayed being wildly wrong. I do however get your point that the formula being used breaks down, to a degree, where there are more complex security arrangements than typical and can therefore be improved upon. I'll put some thought into it.
|
|
|
|
Post by reeknralf on Feb 28, 2016 15:06:48 GMT
That's at least the third time I've been taken to task on here for not understanding banking industry terminology. I used to think a sub-prime loan was a loan that was higher risk. Now I know a 3000% LTV loan to a prime borrower is still a prime loan, and a 10% ltv loan to a sub-prime borrower is still sub-prime. Now I find out that a 75% ltv loan might have more equity support than a 65% ltv loan. I guess I should know this stuff, and chris is right, but with this sort of terminology, is it any wonder the banking industry is in trouble? Perhaps p2p could use 'free equity' instead of LTV, defined as the reduction in property value that could be withstood without incurring a capital loss. After all, that's what we care about as lenders. |
|
|
|
Post by chris on Feb 28, 2016 15:29:17 GMT
That's at least the third time I've been taken to task on here for not understanding banking industry terminology. I used to think a sub-prime loan was a loan that was higher risk. Now I know a 3000% LTV loan to a prime borrower is still a prime loan, and a 10% ltv loan to a sub-prime borrower is still sub-prime. Now I find out that a 75% ltv loan might have more equity support than a 65% ltv loan. I guess I should know this stuff, and chris is right, but with this sort of terminology, is it any wonder the banking industry is in trouble? Perhaps p2p could use 'free equity' instead of LTV, defined as the reduction in property value that could be withstood without incurring a capital loss. After all, that's what we care about as lenders. That's not what I called you out on at all. You said you found it "disconcerting" that we couldn't get "these sums right", like there was some definition that we're failing to follow or a formula we couldn't calculate correctly. You came up with your own definition and called us out for not following it. There are flaws in your definition as well because it's basically ignoring the value of the second charges even though they are still providing additional security. If there is a problem with the first charge or an asymmetrical write down across the three assets then there are plenty of scenarios where the loan is stronger for having those second charges than not having them yet that is not reflected in your calculation, their presence is simply ignored. But then we're straying into the realms of trying to make LTV a rating where it is not and was never intended to be. It's a simple metric and, having reflected on this, for its flaws it's probably still better as is with a simple definition and formula than us trying to redefine the term with a much more complicated algorithm or calculation behind it.
|
|
ablender
Member of DD Central
Posts: 2,204
Likes: 555
|
Post by ablender on Feb 28, 2016 17:47:03 GMT
Hi chris . Related to the LTV of a loan vs its security, I am curious about the Nil values attributed to some of the securities on AC, such as First Debenture, Corporate Guarantee and Personal Guarantee (even if it mentions a sum of money in the description.) As a reference, may I use loan #226. Do such Nil values effect the LTV or the overall security of a loan and how? As I understand it this is used by the team when we have taken that security but in the event of a default it can't really be counted upon by the lenders. It's a nice to have, and should have some value, but lenders shouldn't be basing their lending decisions on the full value of that asset at the time of the loan being taken out. They can also be used as an alternative route to having a claim over the primary security should the lender try and use the legal system to wriggle out of their obligations. I think a lot of PGs are entered as having no value but can be an important last line of defence. Interesting to know this as, from the way these are presented on the platform, it is not clear.
|
|
|
|
Post by crabbyoldgit on Feb 28, 2016 21:58:49 GMT
This just gets more interesting ,so i see personal guarantees given no value some with a figure offered less than the total of the loan and therefore i guess limited to that even if the personal wealth of the borrower exceeds this sum offered in the guarantee.Some where the figure is the full value of the loan, the question is of course is if the most basic checks are made that if required,the borrower may have assets remotely likely to cover the loss if required.Heck i could offer to give a personal guarantee to the goverment of the USA to underwrite their national debt but i dont think to will mean much if they go bust.So are personal guarantees mainly hot air or do have at best marginal use.
|
|
|
|
Post by chris on Feb 28, 2016 22:48:08 GMT
This just gets more interesting ,so i see personal guarantees given no value some with a figure offered less than the total of the loan and therefore i guess limited to that even if the personal wealth of the borrower exceeds this sum offered in the guarantee.Some where the figure is the full value of the loan, the question is of course is if the most basic checks are made that if required,the borrower may have assets remotely likely to cover the loss if required.Heck i could offer to give a personal guarantee to the goverment of the USA to underwrite their national debt but i dont think to will mean much if they go bust.So are personal guarantees mainly hot air or do have at best marginal use. Not fully sure I'm following what you're saying - is this AC specific or industry wide? Surely a personal guarantee is just another tool in the box and varies massively from loan to loan, borrower to borrower, and even platform to platform depending on the processes and checks used. On AC at least there's always a credit report which will give you full details of security taken, how it's been valued, and how that security mitigates the risk of default. Where a value is given, and by implication the loan depends on it to some degree, those documents will spell out the methodology used. If a nil value has been entered then that particular asset or guarantee isn't required to justify making the loan and isn't expected to be relied upon even if the loan defaults, but is taken to provide lenders with an additional safety net should it ever be needed for whatever reason.
|
|
|
|
Post by propman on Feb 29, 2016 10:01:00 GMT
For example I did a quick sanity check on SS loan PBL052 and they don't even seem to take into account the prior charges in their calculation. It depends on where you look. For the one-line entries in the various tables, etc., PBL052 shows the LTV to be 33%. If you look at the loan Particulars, you'll find it stated as the more proper 60%. I think the way AC do it is the right way. If the LTV is shown to be 70%, it ought to mean that as long as the security can be sold for net proceeds equal to 70% of its 'value' the investors ought to be able to exit without loss. There is no one way and making the investor look closely at the calculation has merit. An appropriate methodology for a development will be different to an investment loan (or pre-let development under a fixed price "turn key" construction contract).
That aside, there are 2 factors in assessing the value of second security in addition to its recoverability in different market conditions, the drop in price achieved that still allows recovery of the loan value, and how rapidly the recovery drops if the price achieved falls below this. Considering only the security available to second charges meets the second, while the total position deals with the former.
|
|
|
|
Post by propman on Feb 29, 2016 13:10:47 GMT
My calculations are much tougher and i ideally want 100% cover of the loan based on Residential property First charge - 90% of 90 day sale price less new loan Second charge - 80% of 90 day sale price less previous debt less new loan Commercial Property First Charge - 70% of 90/180 day sale price less new loan ... Boasting time - with only one very small debt (fraud) and lending to about 200 businesses this policy seems to be working so far Any system evaluation is largely meaningless until a downturn hits. In a rising market 100% LTV should not yield any losses!
My comments on your rules of thumb are that they do assume you can rely on the valuations and that losses relative to them should be fairly uniform for each class of asset. So a second charge on a £200k asset is considered equally if £100k against £60k first charge or £20k and £140k first charge. However with a 30% reduction in value, the former would repay 2/3 of the loan and the latter nothing. Also, for resi values, if this is a development, the proportion of land value is important as it indicates the development cost that could well change, while the security of the Developer is almost as important as the security because it is going to be difficult to recover much more than land value if the developer goes bust some time before the development is completed. I would also point out that there is a world of difference having a loan to the likely date of completion (so satisfied from sales proceeds) rather than finishing during the project and so requiring refinancing.
|
|
