grahamg
Member of DD Central
Posts: 220 
Likes: 62
|
Post by grahamg on Mar 29, 2016 17:22:22 GMT
The attached stats seem quite clear for a given cohort of say 100 VAT registered businesses approximately 5% of the remaining firms will fail every six months. At least 50% gone after 5 years. Seemingly regardless of experience. The ONS stats similarly indicate between 42 and 50% left after 5 years. I don’t see how those P2P companies offering SME loans as opposed to project loans can buck this long term regardless of what stats they present now. Unless crystal balls are included in the application process. Or maybe that worthless personal guarantee makes all the difference !. Maybe someone could work out the capital loss and consequently the breakeven interest rate over 5 years. Sorry a bit beyond my spreadsheet ability Are autobiddies and potential IFISA buyers who hold to term in for a bit of a shock in 5 years time? 10 year survival stats.pdf (99.43 KB) ONS failure stats.pdf (127.69 KB) |
|
pikestaff
Member of DD Central
Posts: 2,136
Likes: 1,484
|
Post by pikestaff on Mar 30, 2016 8:21:55 GMT
Yes 5 years might be too long but the data provided does not support this. Lenders need to focus on the rate of failure, not the cumulative number of failures, and on the recovery rate. It's not complicated. For example: - If you are earning 10% gross a year and a constant 5% of businesses fail each year with a 0% recovery, you will earn a net 5% pa.
- If you are earning 10% gross a year and a constant 5% of businesses fail each year with recoveries averaging 50%, you will earn a net 7.5% pa (ignoring dead time on the recoveries).
My expectation is that my long run return on SME lending will be somewhere between these two figures, hopefully toward the upper end. I have been doing considerably better than 7.5%, but the credit environment has been benign. It won't stay that way forever. The "10 year survival stats" article shows that the rate at which VAT-registered business cease to trade falls over time, from about 7% every 6 months at the outset, to about 4% every 6 months after 10 years. Double (roughly) to get the annual rate. This is scary. However, the article equates ceasing to trade with failure. Not all cessations will be failures. And, of those that are failures, not all of them will be insolvent. My guess is that the graph of all insolvent failures would be likely to show a similar slope, with a reduction in the rate of insolvent failure over time. If this were the whole story, longer loans should outperform short ones. However, I don't think it is the whole story. If the platforms' selection processes add value, their selected borrowers should be statistically better than average, when the loans are made. They may not stay that way, in which case the default rate on the platforms' selected borrowers could well rise over time despite the overall trend being downward. |
|
|
|
Post by mrclondon on Mar 30, 2016 10:58:47 GMT
And to add some empirical support to pikestaff's post, I lent extensively on FC from launch in the autumn of 2010 until the end of 2013 but have made no new loans since then. I'm still receiving recovery payments on dozens of defaulted loans each month, and at present I've received recovery payments of 38% of the total defaulted. At this stage in the credit cycle it looks as if somewhere around a 60% recovery of the FC defaults could be feasible. |
|
|
|
Post by tybalt on Mar 30, 2016 11:19:40 GMT
I also believe that there is some added value in analysis of the Loan description provided by some P2B sites.
|
|
markr
Member of DD Central
Posts: 766
Likes: 426
|
Post by markr on Mar 30, 2016 12:06:26 GMT
Any extrapolation of these figures to P2B portfolios would implicity assume that platform's loan books are a random cohort, but one would hope that this was definitely not the case. As someone else has said, equating ceasing to trade with failure is the first mistake - many businesses simply run their course and are no longer required, lifestyle businesses, one-man-bands, contractors, will probably just fizzle out. One would hope that the owners of these businesses wouldn't apply for a 5 year loan if they didn't expect to be trading in 5 years. It also assumes the platform does no better than chance when credit checking applicants. So any analysis would be unduly pessimistic in my view.
Apropos of not much, the first document fits a linear regression line through a data set that clearly isn't linear, and in fact immediately afterwards re-presents the data in a form that directly argues that the previous graph definitely isn't linear. The line is clearly nonsense (no business ever survives more than 14 years??), so why include it?
|
|
|
|
Post by propman on Mar 30, 2016 14:49:32 GMT
very interesting. Thanks to the OP for starting at Samford for his personal analysis. I agree that we can't extrapolate from the Stat paper to loan performances, but it does demonstrate the risks of long loans are not proportional to their length as the initial credit checks have decreasing relevance. I think other time periods would demonstrate this as 5-yr survival rates will vary radically between those which cross a down turn and those within the cycle. ISTR that businesses born around the trough of the cycle have a much higher survival rate, presumably because they start credit constrained and are mature (and more likely to be more robust) when the downturn hits. In addition, they probably start with less competition and are went ahead despite economic head-winds so may exclude flakier offerings that were scared off or failed to reach the VAT threshold.
Banks only usually offer 5 year loans to significant businesses and will rarely exceed these terms. However amortising loans have reduced risk as much is due for repayment over a foreseeable timescale.
Re property secured loans, default rates are even more skewed to down-turns as in good times buyers are available at or above the secured debt in a fair proportion of cases. As a result you cannot meaningfully assess the credit performance on these loans until there has been a down-turn.
- PM
|
|
|
|
Post by ablrateandy on Mar 30, 2016 22:33:34 GMT
The other thing to think about is how lenders will respond in a downturn when defaults do start to add up. Will P2P see better-than-average recoveries because they are in it for the longer haul?
As a massive, sweeping generalisation let's imagine a company that owns 10 BTL houses and has borrowed up to the hilt at say an original 80% LTV. There's a market downturn which has pushed the properties to negative equity and a couple of the properties aren't let so the company is struggling :
1. Your average high street bank manager would traditionally have tried to work through this over a couple of years in order to get the maximum recovery
2. Nowadays, the Computer-Says-No attitude of banks would make many of them seek a very fast re-possession to secure the asset for themselves, rather than try to work as a partner with the borrower. The property then goes into the general pool of assets to be sold on, with no-one having a real vested interest in maximising recovery value (ie. a checklist may just say "If you can get 90% just take it"
3. Most fund managers or bondholders lack the expertise to sit through a full re-structuring so would accept a good offer for any loans extended, even if it was well below 100% recovery, purely because they can't be arsed.
So over a long term I would expect the high street bank manager to probably do best in terms of recovery, the "new" bank to come second and the fund manager to come third because they lack the patience for a full workout.
So who are P2P lenders most akin to? I would have thought the high street bank manager?
|
|
Liz
Member of DD Central
Posts: 2,426
Likes: 1,297
|
Post by Liz on Mar 30, 2016 23:09:19 GMT
Sell it take 90% and move on  |
|
|
|
Post by ablrateandy on Mar 30, 2016 23:12:47 GMT
Congratulations!! You just won a job at Blackrock  |
|
Liz
Member of DD Central
Posts: 2,426
Likes: 1,297
|
Post by Liz on Mar 30, 2016 23:26:27 GMT
Congratulations!! You just won a job at Blackrock I will take it, bet it pays more than raising 4 children.
|
|
|
|
Post by propman on Mar 31, 2016 7:58:11 GMT
Congratulations!! You just won a job at Blackrock When you are in a "higher target yield" fund, the present value of returns might well be better by taking the 90%. Add that you have to revalue your assets and so let your investors know your struggling and therefore they are subject to significant withdrawals and that they need to hire expensive "experts" to manage the assets as they don't have the expertise in house and the 90% looks like the right thing for them to do.
What interests me about P2P is how investors will react. If there is a mass panic to sell everything at discounts, then this could suck liquidity from new loans and force platforms to take large losses until the panic is over, significantly increasing platform risk.
We know that Zopas investor base is very slow to react, but we can't assume that is the same for the higher risk platforms.
|
|
|
|
Post by ablrateandy on Mar 31, 2016 8:00:24 GMT
I am continually amazed at how chilled lenders are in P2P, especially when loans get extended... in cases where banks and funds would immediately pull funding.
Maybe this will prove to be a good thing?
|
|
pom
Member of DD Central
Posts: 1,922 
Likes: 1,244
|
Post by pom on Mar 31, 2016 8:18:20 GMT
I am continually amazed at how chilled lenders are in P2P, especially when loans get extended... in cases where banks and funds would immediately pull funding. Maybe this will prove to be a good thing? Maybe cos it's our money at stake rather than a bonus?
|
|
registerme
Member of DD Central
Posts: 6,196
Likes: 6,006
|
Post by registerme on Mar 31, 2016 11:21:26 GMT
If I didn't need the cash I would be 1., if I did need the cash I would be 2.
|
|
|
|
Post by lynnanthony on Apr 1, 2016 3:24:08 GMT
..... One would hope that the owners of these businesses wouldn't apply for a 5 year loan if they didn't expect to be trading in 5 years. ..... Hmm. I'm not sure. I think it varies. But I don't have a problem with businesses taking out five year loans and for whatever reason ceasing trading provided they pay off the loan. My experience with P2P (3 years) is that a significant proportion of the loans I am in do not run to term. Sometimes they get a better offer elsewhere (i.e. a lower rate), sometimes they struggle, fail, and pay off early. And sometimes they go bad. For myself, I like to lend to businesses for expansion. Either to buy new plant or fund new premises or to cover working capital. I tend to avoid businesses that are apparently borrowing to stand still, and there seem to be rather a lot of these. Where is the light at the end of the tunnel? Is the business just hoping for better things?
|
|
