This webcast (4 September 2019) by David Richardson (Just interim CEO) casts some light on how the proposed NNEG hedge will work, but also casts significant amounts of darkness. Richardson states (my transcript)
20:35 ‘the Solvency II property sensitivity is high, and here in this slide we contrast this with the expected long term cost of NNEG shortfalls. .. I am acutely aware that the solvency II treatment of the NNEG very directly impacts our business. However, what we are doing here is highlighting the difference between the regulatory treatment and expected outcomes for two reasons. First, to illustrate in a very simplified way how most of the Solvency II volatility is expected to be a time issue, and secondly to explain why NNEG risk transfer makes economic sense for counterparties who are in a position to focus on the economic view to a greater extent than us. The first number to note on the slide is £26m. This is the present value of the cost of expected NNEG shortfalls under our central property growth scenario. On the chart, you can see how far in the future these NNEG shortfalls emerge. The expected NNEG shortfall peaks 30 years from now, and overall involves relatively small numbers. If we assume a sudden one-off 10% hit to house prices, and no subsequent recovery, the cost of the NNEG shortfalls obviously goes up, but it only goes up by £27m. The contrast with the impact to the Solvency II excess own funds is stark. A one-off 10% hit to property prices, as you saw on the previous slide, results in a £312m reduction in Solvency II surplus. That is more than 10x the expected economic cost.
Now, I do accept this is a simplistic way of presenting the expected long term cost of the NNEG. In reality, house prices won’t develop in a straight line, and there will be regional and residual property variances. However, it demonstrates that most of the Solvency II volatility relates to time value rather than intrinsic value.
I cannot replicate the figures for the shortfall. My (admittedly crude) model assumes a single model point with a total loan value of £6,800m and LTV of 34%. Then using the growth assumptions given by Richardson and a loan rate of 5.5% I project in the usual way, weight by a 70 year old exit rate based on CMI, and discount at a risk free rate of 1.5%.
That is pretty crude but I am getting NNEG shortfalls well in excess of the number quoted below for the 10% fall and no subsequent recovery in the property price. The concern here is the way that relatively upbeat projections are presented in a way that makes them impossible to verify.I am not suggesting the results are wrong, simply that I cannot replicate them. And if I cannot replicate a result, the result is worthless.