Just Share Discussion Threads

[topvest] 08/9/2018 17:41

i.e. inflating property values by more than the risk free rate.

[topvest] 08/9/2018 17:40

Legal & General do the same or very similar...maybe slightly less aggressive.

"Lifetime mortgage loans amount to £2,023m (2016: £852m). They are valued using a discounted cash flow model by projecting best‐ estimate net asset proceeds and discounting using rates inferred from current LTM pricing, thereby ensuring the value of loans at outset is consistent with the purchase price of the loan, and ensuring consistency between new and in‐force loans. Inputs to the model include property growth rates and voluntary early redemptions. The valuation at 31 December 2017 reflects a long term property growth rate assumption of RPI + 0.5%."

I take the point that it is bringing profits forward, but that's the whacky world of fair value accounting and its priced in. Valuation of JUST is very low versus the reported profits.

[eumaeus] 08/9/2018 17:26

@Topvest yes. The NNEG is incorrectly *valued*, according to the PRA (and to me). On the cheeky question, Kevin and I might be publishing something next week at Eumaeus website, but we are conscious of legal risk, not giving financial advice etc. There's a rule called 'fair comment' meaning that journalists and bloggers such as ourselves can say stuff, but all the same.

[topvest] 08/9/2018 17:22

OK, got it.... so you are saying that the NNEG risk is incorrectly valued in the valuation and that would reduce profitability and carrying values in the JUST balance sheet as a level 3 financial instrument?

[topvest] 08/9/2018 17:10

Well you are certainly confusing me. To be honest I have no idea where this will end up but its only 0.2% of my portfolio, so worse things happen!

Just to be cheeky....

What is your valuation of JUST then, as per your model?

Are you actually saying its a revenue recognition issue rather than a regulatory capital issue, in which case the accounts are misstated and KPMGs audit opinion is open to challenge?

Why are the PRA, JUST and all the analysts talking about regulatory capital then?

[eumaeus] 08/9/2018 16:55

(1) I am not long of Just, why on earth would I be?
(2) On whether Just are out on a limb, I am conflicted as ex-PRA, so can’t answer that. However I will point you a remark made by the deputy gov at Treascom, 11 July ‘I think that the more prudent end of current industry practice is not different from what we are suggesting, but inevitably there will be some people with different positions, and those with a different one are likely to give us some quite forthright evidence in our consultation.’ I can’t imagine who he was referring to.
(3) “Why is sufficient regulatory capital to withstand a c30% house price drop and c10 years of no inflation insufficient?” As I have said about 5 times, it’s not a reg cap question (i.e. capital required), it’s a valuation question. The general principle is that you (management) should not book risky profits in advance, and effectively take them from prospective shareholders, rather your shareholders should be able to benefit from the risky growth. So you must value risky assets in a way that the original shareholders do not benefit in advance, at the expense of future (=current) shareholders. It may well be, even may certainly be the case that property values produce the excess return baked into the current valuation of Just. The question is whether current shareholders will benefit. Why do you think you will?

[dasv] 08/9/2018 16:45

"they are fully pricing their HPI assumptions into the balance sheet, which overprices the deferment, and you the shareholders are buying into this"

1. The ERM's are discounted in value under SII. See presentation

2. I am not a shareholder

3. It can only be known whether ERMs are overpriced post hoc after experience.

4. And "so what?", so you stand to lose a fortune by forward selling me a flat in Mayfair at a discount to today's current price.

[topvest] 08/9/2018 16:35

That's not quite correct eumaeus - I look at the bigger picture and very much look at the financials. I don't spend my life modelling, and lets face it models are normally wrong anyway. To be honest, the more detail you get into on mathematical calculations, often the bigger mistakes you make in investing as you start missing the obvious. I will leave the modelling (and option pricing!) to those like yourselves who are better at it and enjoy the intellectual stimulus of trying to prove who has the biggest IQ. I will stick with an Annual Report and Stockopedia to analyse my investments. This company has just been hit again by regulatory change in my view, like they were when the annuity rule changes hit them a few years ago.

I'd be interested in your views then as to whether you think JUST are out on a limb or whether the rest of the industry are all doing the same thing (e.g. Legal & General, Aviva etc.)?

It's also interesting that the PRA are hitting this sector harder than the banking sector is being hit by their regulator as banks are far far more exposed to a property downturn. Keeping it simple, answer me this question please...why is sufficient regulatory capital to withstand a c30% house price drop and c10 years of no inflation thereafter insufficient? To a simple person like myself that appears to be reasonably cautious, particularly with a 31% LTV ratio.

Big picture I know, but lets try and keep it simple.

[eumaeus] 08/9/2018 14:09

Guy is an actuary, and as you can see from his blog, is a lecturer at the University of Kent. I am now retired, worked in the regulation of capital models on and off for nearly 20 years, occasionally returning to the private sector in risk management and capital modelling roles. I have been involved in option pricing since the mid 1980s. Many of the assumptions mentioned in textbooks are sufficient but not necessary conditions. Is everyone familiar with the distinction between sufficient and necessary? On 'technical arguments', well the subject is technical. When a firm chooses to base almost its entire business model on writing large embedded put options, which are notoriously difficult to manage (tho not to value, if they are simple European options) then be cautious about investing in such firms, unless you really understand the subject. I get the impression many people here don't. Including Topvest above, who admits he looks at things without mathematics. Fair enough, but be warned!

[topvest] 08/9/2018 13:59

You guys are loving the technical arguments here aren't you? - I'm an FCA Chartered Accountant, but you are losing me. I bow to your better knowledge, but I look at things without mathematical equations. Are you actuaries / mathematicians by any chance?

[charlie] 08/9/2018 13:53

>>This really ought to get more attention and open discussion at the Institute.

Ha, something we can agree on at last!

Actually you will see in that blog that I agree with many of your arguments, in isolation and on their own terms. I just don't think they're the only arguments.

[charlie] 08/9/2018 13:52

>>This really ought to get more attention and open discussion at the Institute.

Ha, something we can agree on at last!

Actually you will see in that blog that I agree with many of your arguments. I just don't think they're the only arguments.

[eumaeus] 08/9/2018 13:23

Ah so Charlie = Guy.

Guy, I skimmed through your blog, and thanks for posting that. We already corresponded a while ago about this. Your post deserves more careful consideration than is possible on a blog that doesn't allow carriage returns or headings. I contacted Kevin and we will put together something more substantial. This really ought to get more attention and open discussion at the Institute.

[eumaeus] 08/9/2018 13:09

@dasv: Quite a number of points here. Taking one: “Conversely if you would like to forward sell me a flat in Mayfair 10 years ahead discounted at its current price by its 10 year rental income, I'd be very happy to take that bet. You, as option writer, would be betting that there is zero or negative price inflation on Mayfair property. Let me know where I can sign such a contract”. (1) A deferment is not the same as a forward. With the deferment I pay now, but with deferred possession, with the forward, it’s the same as a deferment but I pay later. This is why we have the interest rate term for the forward, but not the deferment. (2) I am not selling an option, but a forward (or a deferment). (3) Correct, in selling the deferment, or the forward, I am betting on zero or negative HPI. So what? This is not the correct comparison to ERMs, where the ERM firm is effectively selling overpriced deferments to shareholders. I.e. they are fully pricing their HPI assumptions into the balance sheet, which overprices the deferment, and you the shareholders are buying into this. Who do you think is making the money here? Think it though. You have overpaid by the amount you have paid.

[charlie] 08/9/2018 12:54

I agree dasv. I think the intellectual justification for Black-Scholes when there are no markets and no hedging is weak, in fact non-existent. This is why eumaues didn't answer your questions, and always changes the subject when certain questions are asked.

I've written up my version of the 'academic exercise' here



But hey eumaeus is the PRA, sort of (OK he's recently retired, but obviously knows better than the rest of us what the PRA is thinking). They can mandate whatever they like. If they say the moon is made of green cheese, or Black-Scholes gives a 'correct' valuation when there are no markets and no hedging, then annuities are going to be need to be priced to compensate.

[dasv] 08/9/2018 12:09

eumaeus,

Thank you for your reply. Unfortunately you didn't answer any of my questions.

I have read "Asleep at the wheel" and the articles on eumaeus.org.

My questions are academic/moot. I realise the regulator is adopting the Black 1976 approach, and views the NNEG as being an option and dismisses the use of house price inflation in the valuation of ERM's and NNEG's by life co's.

I also realise that ERM vendors apply Black 1976 incorrectly injecting the HPI future valuation for house prices - Just mentions this in their 2017 solvency report.

But my point still stands: if the aim is to value NNEG's, and Black 1976 does not not model the value of NNEG's then is it correct to apply Black 1976?

You claim this is a capitalisation question not a valuation question, but capitalisation relies on valuation. (in the blog it is also claimed that this is also not a question of risk modelling). But with respect to the valuation methodology of Black 1976, the probability of the NNEG firing is governed by both a risk model of longevity and by future HPI (the likelihood of LTV 100% being reached, as debt rolls-up via compounding).

Your answer seems to be that regardless of its applicability (it's match for actual experience), because NNEG has some option-like characteristics it is legitimate to value NNEG's using Black 1976 (even though in some respects NNEG's aren't option-like at all). For example - the lack of a market for deferred possession (and freehold/leasehold is not a market in deferred possession as claimed in the blogs), ERM holders can't decide to execute their NNEG option (other than suicide?), there is no market for the LTM's either, or the NNEG "options" themselves, nor are there potential landlords/investors who choose between rental income or deferred possession discounted on the basis of hypothetical rental income. Again to repeat - these criticisms are all academic in light of the regulator's position.

"What ERM firms are saying to shareholders is essentially ‘the FTSE is priced at 7,400 now, but it will clearly get back to 10,000, so pay us 10,000 now on the basis of past performance'. Would you pay me 10,000 now for my FTSE portfolio? Happy to sell it at that price, any takers?"

Conversely if you would like to forward sell me a flat in Mayfair 10 years ahead discounted at its current price by its 10 year rental income, I'd be very happy to take that bet. You, as option writer, would be betting that there is zero or negative price inflation on Mayfair property. Let me know where I can sign such a contract.


"On the logic of pricing, I shouldn’t have to be explaining it here as it has been in the public domain for more than two years"

Of course you don't have to explain it, and if you don't want to, that's fine. On the question of whether you should - that depends on the moral issue of public good and your information contributing to an "efficient market". It seems - judging by market pricing and from my own persepective, despite the regulators input spanning many years, this is news to many people.

"(1) If you choose between an ERM and a default-free loan, you choose the default-free loan, given that the NNEG may have bitten."

There is no choice between a default-free loan and an ERM. There is no default-free loan on offer.

" (2) If you choose between an ERM and deferred possession of the property you choose deferred possession, because the value of the ERM can never be higher than deferred possession (because of the NNEG)."

There is no deferred possession of a property on offer in the market either. But the ERM can of course be higher than deferred possession because deferred possession in Black 1976 is always valued less than the current price, but the LTM can appreciate in value higher than current value of the property if there is positive house price inflation.

"So ERM value < min(present value of default free loan, present value of deferred possession)." - ignoring the above point true - but a sleight of hand though:-

It's actually "PRESENT ERM value <= min(present value of default free loan, present value of deferred possession).

"This reasoning does not in any way depend on Black Scholes type reasoning." - the application of Black 1976, treatment of NNEG does however depend on the assumption that there is zero house price growth.

"I think the PRA is saying that firms should have been valuing the embedded option the correct way all along, and Solvency II therefore has nothing to do with it."

Firms have indeed been applying Black 1976 incorrectly by conflating HPI driven future price with forward price. But whether that has resulted in incorrect valuations is another matter. There is for example a wild divergence between previous experience of NNEG's by Just (13 NNEG's over 10 years?) with future estimations also immaterial vs valuations in the "Asleep at the wheel" paper which notionally estimates NNEG's as worth 52% of a typical LTM's value.

As I said earlier - this is all academic/moot. The regulator's position is that HPI cannot be used, and NNEG's are to be valued like an option. The upshot will be that life co's (those that make it through this) will have to either hold more capital (e.g. via a debt/equity raise) and/or reinsure against NNEG risk. This will obviously impact margins and reduce profitability, meaning higher ERM rates for consumers and lower availability: Just are raising age thresholds for new ERM's reducing risk of NNEG's being triggered. I guess if reinsurers think the PRA's valuation approach is not appropriate/onerous then the reinsurance route could be economic for life co's reducing the capital requirement.

[scrapheap] 08/9/2018 00:25

This shows the bond... Down a fair bit but not in distress...

hxxps://www.bourse.lu/security/XS1504958817/242337

[topvest] 07/9/2018 20:42

What are the JUST bonds trading at for anyone who has access to these?

[eumaeus] 07/9/2018 17:59

On the logic of pricing, I shouldn’t have to be explaining it here as it has been in the public domain for more than two years, but here it is. (1) If you choose between an ERM and a default-free loan, you choose the default-free loan, given that the NNEG may have bitten. (2) If you choose between an ERM and deferred possession of the property you choose deferred possession, because the value of the ERM can never be higher than deferred possession (because of the NNEG). So ERM value < min(present value of default free loan, present value of deferred possession). This reasoning does not in any way depend on Black Scholes type reasoning. I thin the PRA is saying that firms should have been valuing the embedded option the correct way all along, and Solvency II therefore has nothing to do with it. Obviously Just are putting an entirely different spin on this, but follow the logic for yourself.

[eumaeus] 07/9/2018 17:18

Regulators aren’t paid to keep changing the rules, and (as records show if anyone is bother to check there was a significantly prolonged consultation wrt ERM and NNEG valuation (DP 1/16 in March 2016 thro to CP 13/18 July 2018). No one can say there weren’t warned.

@dasv, any option pricing model works off the forward, which in the case of ERMs is the deferment curve. Google ‘Asleep at the Wheel’ by Kevin Dowd for the full explanation of how the model works. CP 13/18 and its predecessors explains the rationale but in less detail. On the reasoning that you would use past experience rather than valuation modelling, this is a question about capital requirement, not valuation. What ERM firms are saying to shareholders is essentially ‘the FTSE is priced at 7,400 now, but it will clearly get back to 10,000, so pay us 10,000 now on the basis of past performance'. Would you pay me 10,000 now for my FTSE portfolio? Happy to sell it at that price, any takers?