Investment aspects of CDC – Con Keating

Con 8

One of the less discussed aspects of CDC is the management of the investment fund. This differs from traditional DC where the asset management objective is to maximise asset values at all times. It is intrinsically short-term. Incidentally this conflicts with the desires of individual members, who should prefer low asset prices while they are saving and contributing, and maximal asset prices when they are decumulating their fund in retirement.

The CDC fund has an explicit target to achieve or surpass on average. The target arises from the terms on which trustees made awards to members and the contributions they had received. The averaging arises from the risk-sharing and risk pooling rules of the scheme. These rules substitute for the risk buffers and capital adequacy requirements so familiar in insurance and banking. The open question here, as well as in commercial finance elsewhere, is how large these buffers should be and how might they operate. CDC schemes have explicit target benefits which may be cut if investment performance is inadequate.

When the objective is to avoid cuts, the answer to the size question depends on the volatility of the investment portfolio. The question becomes: how long does it take to exhaust the risk-sharing capacity? A quantitative illustration is appropriate. Suppose we allow, for support to pensions in payment, an amount of up to ten percent of the value of members’ claims, and that the fund is composed of 60% active and deferreds with just 40% pensioners in payment. Then the ten percent support represents 25% of the pensioners claims. At first sight the adequacy of this is questionable. A 25% drop in the value of the portfolio in any year has a likelihood of occurring of around 5% when its portfolio volatility is 15% and over 10% when that volatility is 20%. This is a frequency of cuts that is not likely to prove attractive to members generally or pensioners in particular. It is the frequency which arises in DB scheme solvency based management, which includes Dutch CDC.

Much of the DB de-risking and LDI agenda has been driven by a desire on the part of the sponsor to avoid the calls on their guarantee associated and low volatility investment strategies have been pursued. With fund volatility at 10% calls have a frequency of 0.6% and at 5% are unlikely in many lifetimes.

However, the immediate exposure of the CDC scheme is merely to the current year’s pension payments. This is usually of the order of 3%-5% of the members claims. If the funding level is 75% then the support among members to ensure full payment is simply 1.25%, a minor fraction of the total available support, 10%. Obviously, the largest claim in any year is the full amount of the pension payment, and that is far less than to total support.

Other articles have discussed an important element of a sustainable risk-sharing scheme, the need to maintain an equitable balance among members and proposed mechanisms whereby this may be achieved.

Of course, the greatest threat of cuts comes from sequences of poor returns, rather than from extreme, catastrophic events. Let us consider sequences of losses at the conditional expected loss amounts for portfolios of different volatilities. The conditional expected losses are:

Table 1

Volatility 5% 10% 15% 20%
Expected Loss % -4 -8 -12 -16

 

The asset portfolio values resulting from repetitions of these losses are rather extreme. These are shown below for the range of portfolio volatilities.

Table 2

Year 5% 10% 15% 20%
1 96.0% 92.0% 88.0% 84.0%
2 92.2% 84.7% 77.4% 70.5%
3 88.5% 77.9% 68.1% 59.2%
4 84.9% 71.7% 59.9% 49.8%
5 81.5% 66.0% 52.7% 41.8%
6 78.3% 60.7% 46.4% 35.1%
7 75.2% 55.9% 40.8% 29.5%
8 72.1% 51.4% 35.9% 24.8%

 

The total of 10% support is exhausted after five years for the 20% volatility portfolio, after six years for the 15%, after eight years for the 10% and after ten years for the 5% volatility portfolio. No cut is likely within these periods. Put another way, even in these extreme and highly unlikely circumstances, the likelihood of any cut is 1.5% for the twenty percent volatility portfolio, 0.8% for the fifteen percent, 0.2% for the ten percent and 0.04% for the five percent volatility portfolio.

This suggests that the unutilised risk-sharing capacity and the period for which full benefits are assured by this scheme feature are useful metrics of the robustness of a CDC scheme. Certainly, they should inspire member confidence in the scheme.

There is nothing comparable to these simulations in the historic empirical record. The most extreme set of circumstances centred on the 1999 – 2002 Dotcom crisis

Diagram 1

cdc invest

Diagram one

 

 

Diagram 1 shows the call upon risk-sharing resources which might have occurred over that period had the scheme assets been fully invested in UK equity using index returns for fund performance. Annual pensions payments were a little under 5% (on average) of scheme liabilities over this period. The cumulative exposure rose over a five-year period and totalled, at its maximum, a little under 5%, but had fully recovered all of that support within two years.

The interpretation of these support periods that is relevant for the fund management objective is that these are the periods over which performance is not pressured by pension payment requirements, that full payment is essentially assured, and the scheme may be managed on the basis of its average return.

About henry tapper

Founder of the Pension PlayPen, Director of First Actuarial, partner of Stella, father of Olly . I am the Pension Plowman
This entry was posted in CDC, pensions and tagged , , , . Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

w

Connecting to %s