AMNT Green Paper Consultation March 2017

You will probably be aware that the DWP published a Green Paper last month called Sustainability and Security in Defined Benefit Pension Schemes

The Government invites comments by 14 May.

To some the document’s tone suggests that the Government is not looking to make any significant changes; but it is airing a lot of important issues, so it would be rash not to give voice to MNT views on them. It is therefore intended that AMNT will submit a response.

This should not deter members from responding individually; but you are invited in any case to let the Committee know what points you think the Association should make.

If you do so before 11 a.m. on 31 March – by e­‑mail to caroline.davies@amnt.org– your views can be taken into account in formulating questions for a survey of members that is planned to feed into a discussion at the AMNT Annual General meeting on 24 April.

The Green Paper includes an executive summary at pp. 5-8 of its printed text and a summary of consultation questions at pp. 95-99 of that text: the document is published on‑line atwww.gov.uk/government/uploads/system/uploads/attachment_data/file/595103/security-and-sustainability-in-defined-benefit-pension-schemes.pdf.

The main issues seem to be:

1.      whether DB is affordable;

2.      whether and (if so) how trustees’ investment decisions need to be systematically improved;

3.      whether and (if so) how and in what circumstances, indexation of benefits should be restricted;

4.      whether and (if so) how the consolidation of DB schemes should be incentivised or mandated;

5.      whether the Pensions Regulator should be given further powers and (if so) what;

6.      whether new legislation is needed about informing members of their schemes’ position.

Bill Trythall

 

One Comment

  1. Dennis Leech-Reply
    April 1, 2017 at 12:24 am

    Some initial points about the Green Paper.
    Dennis Leech, Professor of Economics, Warwick University.

    1. Although the Green Paper is wide ranging and acknowledges many issues on which it seeks comments, it fails to address the fundamental point of whether the methodology used to value pension schemes is suitable. Many people have argued that the mark-to-market approach to scheme funding that was enshrined in the 2005 pensions act is not working to the benefit of members because it is leading to too much scheme closure. The legislation has prescribed a methodology that has induced a bias towards deficits.

    Moreover some commentators have criticised the framework along these lines, something which the GP fails to acknowledge (although it does acknowledge some alternative approaches to valuing the liabilities – including mentioning the First Actuarial Best Estimate approach by name). The key point here is that the market prices of assets (for example, most importantly, equities) are naturally excessively volatile. That is they are more volatile than theory says should be the case. Studies by economists (notably Robert Shiller, in a long series of work published in leading economics journals and books, for example, “Irrational Exuberance”, and Robert Haugen, for example, “The Beast on Wall street”) have shown that the volatility of market prices of assets is much, much greater than what the theory of efficient markets would predict. Shiller compared asset prices with the discounted present value – at the same date – of dividends paid out later: that is the ideal values if investors behaved rationally. He found large excess volatility that he called irrational exuberance, evidence that investor behaviour is not rational. (JM Keynes called the same thing animal spirits.) One estimate was that annual volatility of stock prices was four times what would be due to the volatility of the PV of future dividends.

    This is a key point because in an open scheme with a long (or indefinite) covenant (for example the USS) the payment of pensions depends much more on dividends than share prices. Share prices are therefore secondary. There may be some selling of assets but it is not of primary importance. On the other hand, a “mature” scheme closed to new members or accrual, that is on the “flight path” to self sufficiency or whatever, will need to derive income from selling shares as well as dividends.

    The legislation seems to have been drawn up on the assumption that the latter is the normal case, that DB pension schemes should be seen as if they are on the way to closure. Hence the funding requirement. That means that using excessively volatile market prices at a point in time to value the assets – rather then valuing them using discounted expected future dividends – means their volatility has to be treated as risk. The possibility that equities underperform and their prices dip just at the moment when the scheme needs to sell them to pay pensions transforms natural volatility into risk. Likewise in attempting to place a value on the liabilities at market prices the fact that the return on assets is risky undermines its use as a discount rate. Better to use a “risk free rate” of corporate bonds or gilts even though the rate of return is lower than on equities.

    Yet in an ongoing scheme it is the dividends that matter as the main source of income for paying pensions, and they are much less volatile.

    2. There would be many advantages from using expected future dividends to value assets, in an open scheme, not least that it would reduce the need to insure against risk. The market risk to asset prices will be eliminated and hence there will be no need to use a wide margin of prudence in the valuation of liabilities.

    An even better methodology would compare projections of flows of income and outlay over time into the future. This would also obviate the need for a discount rate for the liabilities, although a discount rate would be needed to find the present value of the deficit/surplus. The rate chosen would not matter too much since it is the same for assets and liabilities.

    The source of all the problems with the 2005 pensions act methodology is that the assets and liabilities are valued in different ways. There has been a lot of discussion of the discount rate used to value the liabilities which I do not have time today to set out here.

    3. The Green Paper contains a graph (figure 6) showing the growth of assets and liabilities , and widening deficits, over time , and gives it the interpretation that it is a reflection of lower investment returns together with increased member longevity in retirement. But there is another interpretation: the period over which the increase in deficits has occurred has coincided with the period since the introduction of mark-to-market valuations. There is, therefore, a fundamental identification problem where two competing hypotheses are impossible to distinguish from the empirical evidence.

    4. Also, incidentally, the liabilities are computed using gilt rates rather than on a scheme specific basis. The latter would allow the use of the rate of return on the assets in situations where there is a strong employer covenant or where the assets are other than gilts.

    5. There is a tendency to treat gilt rates as if they are the same as investment returns (for example the graph of gilt rates in figure 1 is presented as evidence of the lower investment returns over the past 20 years. Investment rates on, for example, equities are not related to gilt rates

    6. The way that the deficit is defined makes it volatile. It is the difference between two very large numbers, the assets and liabilities, both of which are volatile. Variance of the deficit equals variance of assets plus variance of liabilities. Moreover in both cases the volatility is inessential to the valuation of pension schemes: volatility of assets is due to excess volatility of market prices and the volatility of liabilities is due primarily to a low (and variable) discount rate. They are artificial results of the valuation methodology.
    7. The methodology ignores the safety net provided by the PPF. This reduces the risk that members will not be paid their entitlement to pensions and therefore it makes sense to use it as a factor in the calculations.

    The methodology assumes the sponsor covenant to be independent of the deficit, whereas if the deficit is too large it will pose a threat to the sponsor’s solvency. Therefore the covenant and the deficit should be seen as interdependent.

Leave A Comment